diff options
| author | thegeorg <thegeorg@yandex-team.ru> | 2022-02-10 16:45:08 +0300 |
|---|---|---|
| committer | Daniil Cherednik <dcherednik@yandex-team.ru> | 2022-02-10 16:45:08 +0300 |
| commit | 4e839db24a3bbc9f1c610c43d6faaaa99824dcca (patch) | |
| tree | 506dac10f5df94fab310584ee51b24fc5a081c22 /contrib/libs/zstd | |
| parent | 2d37894b1b037cf24231090eda8589bbb44fb6fc (diff) | |
| download | ydb-4e839db24a3bbc9f1c610c43d6faaaa99824dcca.tar.gz | |
Restoring authorship annotation for <thegeorg@yandex-team.ru>. Commit 1 of 2.
Diffstat (limited to 'contrib/libs/zstd')
104 files changed, 17597 insertions, 17597 deletions
diff --git a/contrib/libs/zstd/CHANGELOG b/contrib/libs/zstd/CHANGELOG index 4e0045b950..34921dd46f 100644 --- a/contrib/libs/zstd/CHANGELOG +++ b/contrib/libs/zstd/CHANGELOG @@ -21,150 +21,150 @@ build: improved meson unofficial build, by @eli-schwartz cli : custom memory limit when training dictionary (#2925), by @embg cli : report advanced parameters information when compressing in very verbose mode (``-vv`), by @Svetlitski-FB -v1.5.0 (May 11, 2021) -api: Various functions promoted from experimental to stable API: (#2579-2581, @senhuang42) - `ZSTD_defaultCLevel()` - `ZSTD_getDictID_fromCDict()` -api: Several experimental functions have been deprecated and will emit a compiler warning (#2582, @senhuang42) - `ZSTD_compress_advanced()` - `ZSTD_compress_usingCDict_advanced()` - `ZSTD_compressBegin_advanced()` - `ZSTD_compressBegin_usingCDict_advanced()` - `ZSTD_initCStream_srcSize()` - `ZSTD_initCStream_usingDict()` - `ZSTD_initCStream_usingCDict()` - `ZSTD_initCStream_advanced()` - `ZSTD_initCStream_usingCDict_advanced()` - `ZSTD_resetCStream()` -api: ZSTDMT_NBWORKERS_MAX reduced to 64 for 32-bit environments (@Cyan4973) -perf: Significant speed improvements for middle compression levels (#2494, @senhuang42 @terrelln) -perf: Block splitter to improve compression ratio, enabled by default for high compression levels (#2447, @senhuang42) -perf: Decompression loop refactor, speed improvements on `clang` and for `--long` modes (#2614 #2630, @Cyan4973) -perf: Reduced stack usage during compression and decompression entropy stage (#2522 #2524, @terrelln) -bug: Improve setting permissions of created files (#2525, @felixhandte) -bug: Fix large dictionary non-determinism (#2607, @terrelln) -bug: Fix non-determinism test failures on Linux i686 (#2606, @terrelln) -bug: Fix various dedicated dictionary search bugs (#2540 #2586, @senhuang42 @felixhandte) -bug: Ensure `ZSTD_estimateCCtxSize*() `monotonically increases with compression level (#2538, @senhuang42) -bug: Fix --patch-from mode parameter bound bug with small files (#2637, @occivink) -bug: Fix UBSAN error in decompression (#2625, @terrelln) -bug: Fix superblock compression divide by zero bug (#2592, @senhuang42) -bug: Make the number of physical CPU cores detection more robust (#2517, @PaulBone) -doc: Improve `zdict.h` dictionary training API documentation (#2622, @terrelln) -doc: Note that public `ZSTD_free*()` functions accept NULL pointers (#2521, @animalize) -doc: Add style guide docs for open source contributors (#2626, @Cyan4973) -tests: Better regression test coverage for different dictionary modes (#2559, @senhuang42) -tests: Better test coverage of index reduction (#2603, @terrelln) -tests: OSS-Fuzz coverage for seekable format (#2617, @senhuang42) -tests: Test coverage for ZSTD threadpool API (#2604, @senhuang42) -build: Dynamic library built multithreaded by default (#2584, @senhuang42) -build: Move `zstd_errors.h` and `zdict.h` to `lib/` root (#2597, @terrelln) -build: Allow `ZSTDMT_JOBSIZE_MIN` to be configured at compile-time, reduce default to 512KB (#2611, @Cyan4973) -build: Single file library build script moved to `build/` directory (#2618, @felixhandte) -build: `ZBUFF_*()` is no longer built by default (#2583, @senhuang42) -build: Fixed Meson build (#2548, @SupervisedThinking @kloczek) -build: Fix excessive compiler warnings with clang-cl and CMake (#2600, @nickhutchinson) -build: Detect presence of `md5` on Darwin (#2609, @felixhandte) -build: Avoid SIGBUS on armv6 (#2633, @bmwiedmann) -cli: `--progress` flag added to always display progress bar (#2595, @senhuang42) -cli: Allow reading from block devices with `--force` (#2613, @felixhandte) -cli: Fix CLI filesize display bug (#2550, @Cyan4973) -cli: Fix windows CLI `--filelist` end-of-line bug (#2620, @Cyan4973) -contrib: Various fixes for linux kernel patch (#2539, @terrelln) -contrib: Seekable format - Decompression hanging edge case fix (#2516, @senhuang42) -contrib: Seekable format - New seek table-only API (#2113 #2518, @mdittmer @Cyan4973) -contrib: Seekable format - Fix seek table descriptor check when loading (#2534, @foxeng) -contrib: Seekable format - Decompression fix for large offsets, (#2594, @azat) -misc: Automatically published release tarballs available on Github (#2535, @felixhandte) - -v1.4.9 (Mar 1, 2021) -bug: Use `umask()` to Constrain Created File Permissions (#2495, @felixhandte) -bug: Make Simple Single-Pass Functions Ignore Advanced Parameters (#2498, @terrelln) -api: Add (De)Compression Tracing Functionality (#2482, @terrelln) -api: Support References to Multiple DDicts (#2446, @senhuang42) -api: Add Function to Generate Skippable Frame (#2439, @senhuang42) -perf: New Algorithms for the Long Distance Matcher (#2483, @mpu) -perf: Performance Improvements for Long Distance Matcher (#2464, @mpu) -perf: Don't Shrink Window Log when Streaming with a Dictionary (#2451, @terrelln) -cli: Fix `--output-dir-mirror`'s Rejection of `..`-Containing Paths (#2512, @felixhandte) -cli: Allow Input From Console When `-f`/`--force` is Passed (#2466, @felixhandte) -cli: Improve Help Message (#2500, @senhuang42) -tests: Remove Flaky Tests (#2455, #2486, #2445, @Cyan4973) -tests: Correctly Invoke md5 Utility on NetBSD (#2492, @niacat) -tests: Avoid Using `stat -c` on NetBSD (#2513, @felixhandte) -build: Zstd CLI Can Now be Linked to Dynamic `libzstd` (#2457, #2454 @Cyan4973) -build: Hide and Avoid Using Static-Only Symbols (#2501, #2504, @skitt) -build: CMake: Enable Only C for lib/ and programs/ Projects (#2498, @concatime) -build: CMake: Use `configure_file()` to Create the `.pc` File (#2462, @lazka) -build: Fix Fuzzer Compiler Detection & Update UBSAN Flags (#2503, @terrelln) -build: Add Guards for `_LARGEFILE_SOURCE` and `_LARGEFILE64_SOURCE` (#2444, @indygreg) -build: Improve `zlibwrapper` Makefile (#2437, @Cyan4973) -contrib: Add `recover_directory` Program (#2473, @terrelln) -doc: Change License Year to 2021 (#2452 & #2465, @terrelln & @senhuang42) -doc: Fix Typos (#2459, @ThomasWaldmann) - -v1.4.8 (Dec 18, 2020) -hotfix: wrong alignment of an internal buffer - -v1.4.7 (Dec 16, 2020) -perf: stronger --long mode at high compression levels, by @senhuang42 -perf: stronger --patch-from at high compression levels, thanks to --long improvements -perf: faster dictionary compression at medium compression levels, by @felixhandte -perf: small speed & memory usage improvements for ZSTD_compress2(), by @terrelln -perf: improved fast compression speeds with Visual Studio, by @animalize -cli : Set nb of threads with environment variable ZSTD_NBTHREADS, by @senhuang42 -cli : accept decompressing files with *.zstd suffix -cli : provide a condensed summary by default when processing multiple files -cli : fix : stdin input no longer confused as user prompt -cli : improve accuracy of several error messages -api : new sequence ingestion API, by @senhuang42 -api : shared thread pool: control total nb of threads used by multiple compression jobs, by @marxin -api : new ZSTD_getDictID_fromCDict(), by @LuAPi -api : zlibWrapper only uses public API, and is compatible with dynamic library, by @terrelln -api : fix : multithreaded compression has predictable output even in special cases (see #2327) (issue not accessible from cli) -api : fix : dictionary compression correctly respects dictionary compression level (see #2303) (issue not accessible from cli) -build: fix cmake script when using path with spaces, by @terrelln -build: improved compile-time detection of aarch64/neon platforms, by @bsdimp -build: Fix building on AIX 5.1, by @likema -build: compile paramgrill with cmake on Windows, requested by @mirh -doc : clarify repcode updates in format specification, by @felixhandte - -v1.4.6 -fix : Always return dstSize_tooSmall when that is the case -fix : Fix ZSTD_initCStream_advanced() with static allocation and no dictionary -perf: Improve small block decompression speed by 20%+, by @terrelln -perf: Reduce compression stack usage by 1 KB, by @terrelln -perf: Improve decompression speed by improving ZSTD_wildcopy, by @helloguo (#2252, #2256) -perf: Improve histogram construction, by @cyan4973 (#2253) -cli : Add --output-dir-mirror option, by @xxie24 (#2219) -cli : Warn when (de)compressing multiple files into a single output, by @senhuang42 (#2279) -cli : Improved progress bar and status summary when (de)compressing multiple files, by @senhuang42 (#2283) -cli : Call stat less often, by @felixhandte (#2262) -cli : Allow --patch-from XXX and --filelist XXX in addition to --patch-from=XXX and --filelist=XXX, by @cyan4973 (#2250) -cli : Allow --patch-from to compress stdin with --stream-size, by @bimbashrestha (#2206) -api : Do not install zbuff.h, since it has long been deprecated, by @cyan4973 (#2166). -api : Fix ZSTD_CCtx_setParameter() with ZSTD_c_compressionLevel to make 0 mean default level, by @i-do-cpp (#2291) -api : Rename ZSTDMT_NBTHREADS_MAX to ZSTDMT_NBWORKERS_MAX, by @marxin (#2228). -build: Install pkg-config file with CMake and MinGW, by @tonytheodore (#2183) -build: Install DLL with CMake on Windows, by @BioDataAnalysis (#2221) -build: Fix DLL install location with CMake, by @xantares and @bimbashrestha (#2186) -build: Add ZSTD_NO_UNUSED_FUNCTIONS macro to hide unused functions -build: Add ZSTD_NO_INTRINSICS macro to avoid explicit intrinsics -build: Add STATIC_BMI2 macro for compile time detection of BMI2 on MSVC, by @Niadb (#2258) -build: Fix -Wcomma warnings, by @cwoffenden -build: Remove distutils requirement for meson build, by @neheb (#2197) -build: Fix cli compilation with uclibc -build: Fix cli compilation without st_mtime, by @ffontaine (#2246) -build: Fix shadowing warnings in library -build: Fix single file library compilation with Enscripten, by @yoshihitoh (#2227) -misc: Improve single file library and include dictBuilder, by @cwoffenden -misc: Allow compression dictionaries with missing symbols -misc: Add freestanding translation script in contrib/freestanding_lib -misc: Collect all of zstd's libc dependencies into zstd_deps.h -doc : Add ZSTD_versionString() to manual, by @animalize -doc : Fix documentation for ZSTD_CCtxParams_setParameter(), by @felixhandte (#2270) - -v1.4.5 (May 22, 2020) +v1.5.0 (May 11, 2021) +api: Various functions promoted from experimental to stable API: (#2579-2581, @senhuang42) + `ZSTD_defaultCLevel()` + `ZSTD_getDictID_fromCDict()` +api: Several experimental functions have been deprecated and will emit a compiler warning (#2582, @senhuang42) + `ZSTD_compress_advanced()` + `ZSTD_compress_usingCDict_advanced()` + `ZSTD_compressBegin_advanced()` + `ZSTD_compressBegin_usingCDict_advanced()` + `ZSTD_initCStream_srcSize()` + `ZSTD_initCStream_usingDict()` + `ZSTD_initCStream_usingCDict()` + `ZSTD_initCStream_advanced()` + `ZSTD_initCStream_usingCDict_advanced()` + `ZSTD_resetCStream()` +api: ZSTDMT_NBWORKERS_MAX reduced to 64 for 32-bit environments (@Cyan4973) +perf: Significant speed improvements for middle compression levels (#2494, @senhuang42 @terrelln) +perf: Block splitter to improve compression ratio, enabled by default for high compression levels (#2447, @senhuang42) +perf: Decompression loop refactor, speed improvements on `clang` and for `--long` modes (#2614 #2630, @Cyan4973) +perf: Reduced stack usage during compression and decompression entropy stage (#2522 #2524, @terrelln) +bug: Improve setting permissions of created files (#2525, @felixhandte) +bug: Fix large dictionary non-determinism (#2607, @terrelln) +bug: Fix non-determinism test failures on Linux i686 (#2606, @terrelln) +bug: Fix various dedicated dictionary search bugs (#2540 #2586, @senhuang42 @felixhandte) +bug: Ensure `ZSTD_estimateCCtxSize*() `monotonically increases with compression level (#2538, @senhuang42) +bug: Fix --patch-from mode parameter bound bug with small files (#2637, @occivink) +bug: Fix UBSAN error in decompression (#2625, @terrelln) +bug: Fix superblock compression divide by zero bug (#2592, @senhuang42) +bug: Make the number of physical CPU cores detection more robust (#2517, @PaulBone) +doc: Improve `zdict.h` dictionary training API documentation (#2622, @terrelln) +doc: Note that public `ZSTD_free*()` functions accept NULL pointers (#2521, @animalize) +doc: Add style guide docs for open source contributors (#2626, @Cyan4973) +tests: Better regression test coverage for different dictionary modes (#2559, @senhuang42) +tests: Better test coverage of index reduction (#2603, @terrelln) +tests: OSS-Fuzz coverage for seekable format (#2617, @senhuang42) +tests: Test coverage for ZSTD threadpool API (#2604, @senhuang42) +build: Dynamic library built multithreaded by default (#2584, @senhuang42) +build: Move `zstd_errors.h` and `zdict.h` to `lib/` root (#2597, @terrelln) +build: Allow `ZSTDMT_JOBSIZE_MIN` to be configured at compile-time, reduce default to 512KB (#2611, @Cyan4973) +build: Single file library build script moved to `build/` directory (#2618, @felixhandte) +build: `ZBUFF_*()` is no longer built by default (#2583, @senhuang42) +build: Fixed Meson build (#2548, @SupervisedThinking @kloczek) +build: Fix excessive compiler warnings with clang-cl and CMake (#2600, @nickhutchinson) +build: Detect presence of `md5` on Darwin (#2609, @felixhandte) +build: Avoid SIGBUS on armv6 (#2633, @bmwiedmann) +cli: `--progress` flag added to always display progress bar (#2595, @senhuang42) +cli: Allow reading from block devices with `--force` (#2613, @felixhandte) +cli: Fix CLI filesize display bug (#2550, @Cyan4973) +cli: Fix windows CLI `--filelist` end-of-line bug (#2620, @Cyan4973) +contrib: Various fixes for linux kernel patch (#2539, @terrelln) +contrib: Seekable format - Decompression hanging edge case fix (#2516, @senhuang42) +contrib: Seekable format - New seek table-only API (#2113 #2518, @mdittmer @Cyan4973) +contrib: Seekable format - Fix seek table descriptor check when loading (#2534, @foxeng) +contrib: Seekable format - Decompression fix for large offsets, (#2594, @azat) +misc: Automatically published release tarballs available on Github (#2535, @felixhandte) + +v1.4.9 (Mar 1, 2021) +bug: Use `umask()` to Constrain Created File Permissions (#2495, @felixhandte) +bug: Make Simple Single-Pass Functions Ignore Advanced Parameters (#2498, @terrelln) +api: Add (De)Compression Tracing Functionality (#2482, @terrelln) +api: Support References to Multiple DDicts (#2446, @senhuang42) +api: Add Function to Generate Skippable Frame (#2439, @senhuang42) +perf: New Algorithms for the Long Distance Matcher (#2483, @mpu) +perf: Performance Improvements for Long Distance Matcher (#2464, @mpu) +perf: Don't Shrink Window Log when Streaming with a Dictionary (#2451, @terrelln) +cli: Fix `--output-dir-mirror`'s Rejection of `..`-Containing Paths (#2512, @felixhandte) +cli: Allow Input From Console When `-f`/`--force` is Passed (#2466, @felixhandte) +cli: Improve Help Message (#2500, @senhuang42) +tests: Remove Flaky Tests (#2455, #2486, #2445, @Cyan4973) +tests: Correctly Invoke md5 Utility on NetBSD (#2492, @niacat) +tests: Avoid Using `stat -c` on NetBSD (#2513, @felixhandte) +build: Zstd CLI Can Now be Linked to Dynamic `libzstd` (#2457, #2454 @Cyan4973) +build: Hide and Avoid Using Static-Only Symbols (#2501, #2504, @skitt) +build: CMake: Enable Only C for lib/ and programs/ Projects (#2498, @concatime) +build: CMake: Use `configure_file()` to Create the `.pc` File (#2462, @lazka) +build: Fix Fuzzer Compiler Detection & Update UBSAN Flags (#2503, @terrelln) +build: Add Guards for `_LARGEFILE_SOURCE` and `_LARGEFILE64_SOURCE` (#2444, @indygreg) +build: Improve `zlibwrapper` Makefile (#2437, @Cyan4973) +contrib: Add `recover_directory` Program (#2473, @terrelln) +doc: Change License Year to 2021 (#2452 & #2465, @terrelln & @senhuang42) +doc: Fix Typos (#2459, @ThomasWaldmann) + +v1.4.8 (Dec 18, 2020) +hotfix: wrong alignment of an internal buffer + +v1.4.7 (Dec 16, 2020) +perf: stronger --long mode at high compression levels, by @senhuang42 +perf: stronger --patch-from at high compression levels, thanks to --long improvements +perf: faster dictionary compression at medium compression levels, by @felixhandte +perf: small speed & memory usage improvements for ZSTD_compress2(), by @terrelln +perf: improved fast compression speeds with Visual Studio, by @animalize +cli : Set nb of threads with environment variable ZSTD_NBTHREADS, by @senhuang42 +cli : accept decompressing files with *.zstd suffix +cli : provide a condensed summary by default when processing multiple files +cli : fix : stdin input no longer confused as user prompt +cli : improve accuracy of several error messages +api : new sequence ingestion API, by @senhuang42 +api : shared thread pool: control total nb of threads used by multiple compression jobs, by @marxin +api : new ZSTD_getDictID_fromCDict(), by @LuAPi +api : zlibWrapper only uses public API, and is compatible with dynamic library, by @terrelln +api : fix : multithreaded compression has predictable output even in special cases (see #2327) (issue not accessible from cli) +api : fix : dictionary compression correctly respects dictionary compression level (see #2303) (issue not accessible from cli) +build: fix cmake script when using path with spaces, by @terrelln +build: improved compile-time detection of aarch64/neon platforms, by @bsdimp +build: Fix building on AIX 5.1, by @likema +build: compile paramgrill with cmake on Windows, requested by @mirh +doc : clarify repcode updates in format specification, by @felixhandte + +v1.4.6 +fix : Always return dstSize_tooSmall when that is the case +fix : Fix ZSTD_initCStream_advanced() with static allocation and no dictionary +perf: Improve small block decompression speed by 20%+, by @terrelln +perf: Reduce compression stack usage by 1 KB, by @terrelln +perf: Improve decompression speed by improving ZSTD_wildcopy, by @helloguo (#2252, #2256) +perf: Improve histogram construction, by @cyan4973 (#2253) +cli : Add --output-dir-mirror option, by @xxie24 (#2219) +cli : Warn when (de)compressing multiple files into a single output, by @senhuang42 (#2279) +cli : Improved progress bar and status summary when (de)compressing multiple files, by @senhuang42 (#2283) +cli : Call stat less often, by @felixhandte (#2262) +cli : Allow --patch-from XXX and --filelist XXX in addition to --patch-from=XXX and --filelist=XXX, by @cyan4973 (#2250) +cli : Allow --patch-from to compress stdin with --stream-size, by @bimbashrestha (#2206) +api : Do not install zbuff.h, since it has long been deprecated, by @cyan4973 (#2166). +api : Fix ZSTD_CCtx_setParameter() with ZSTD_c_compressionLevel to make 0 mean default level, by @i-do-cpp (#2291) +api : Rename ZSTDMT_NBTHREADS_MAX to ZSTDMT_NBWORKERS_MAX, by @marxin (#2228). +build: Install pkg-config file with CMake and MinGW, by @tonytheodore (#2183) +build: Install DLL with CMake on Windows, by @BioDataAnalysis (#2221) +build: Fix DLL install location with CMake, by @xantares and @bimbashrestha (#2186) +build: Add ZSTD_NO_UNUSED_FUNCTIONS macro to hide unused functions +build: Add ZSTD_NO_INTRINSICS macro to avoid explicit intrinsics +build: Add STATIC_BMI2 macro for compile time detection of BMI2 on MSVC, by @Niadb (#2258) +build: Fix -Wcomma warnings, by @cwoffenden +build: Remove distutils requirement for meson build, by @neheb (#2197) +build: Fix cli compilation with uclibc +build: Fix cli compilation without st_mtime, by @ffontaine (#2246) +build: Fix shadowing warnings in library +build: Fix single file library compilation with Enscripten, by @yoshihitoh (#2227) +misc: Improve single file library and include dictBuilder, by @cwoffenden +misc: Allow compression dictionaries with missing symbols +misc: Add freestanding translation script in contrib/freestanding_lib +misc: Collect all of zstd's libc dependencies into zstd_deps.h +doc : Add ZSTD_versionString() to manual, by @animalize +doc : Fix documentation for ZSTD_CCtxParams_setParameter(), by @felixhandte (#2270) + +v1.4.5 (May 22, 2020) fix : Compression ratio regression on huge files (> 3 GB) using high levels (--ultra) and multithreading, by @terrelln perf: Improved decompression speed: x64 : +10% (clang) / +5% (gcc); ARM : from +15% to +50%, depending on SoC, by @terrelln perf: Automatically downsizes ZSTD_DCtx when too large for too long (#2069, by @bimbashreshta) @@ -190,7 +190,7 @@ misc: Edit-distance match finder in contrib/ doc : Improved beginner CONTRIBUTING.md docs doc : New issue templates for zstd -v1.4.4 (Nov 6, 2019) +v1.4.4 (Nov 6, 2019) perf: Improved decompression speed, by > 10%, by @terrelln perf: Better compression speed when re-using a context, by @felixhandte perf: Fix compression ratio when compressing large files with small dictionary, by @senhuang42 @@ -217,18 +217,18 @@ pack: modified pkgconfig, for better integration into openwrt, requested by @neh misc: Improved documentation : ZSTD_CLEVEL, DYNAMIC_BMI2, ZSTD_CDict, function deprecation, zstd format misc: fixed educational decoder : accept larger literals section, and removed UNALIGNED() macro -v1.4.3 (Aug 20, 2019) +v1.4.3 (Aug 20, 2019) bug: Fix Dictionary Compression Ratio Regression by @cyan4973 (#1709) bug: Fix Buffer Overflow in legacy v0.3 decompression by @felixhandte (#1722) build: Add support for IAR C/C++ Compiler for Arm by @joseph0918 (#1705) -v1.4.2 (Jul 26, 2019) +v1.4.2 (Jul 26, 2019) bug: Fix bug in zstd-0.5 decoder by @terrelln (#1696) bug: Fix seekable decompression in-memory API by @iburinoc (#1695) misc: Validate blocks are smaller than size limit by @vivekmg (#1685) misc: Restructure source files by @ephiepark (#1679) -v1.4.1 (Jul 20, 2019) +v1.4.1 (Jul 20, 2019) bug: Fix data corruption in niche use cases by @terrelln (#1659) bug: Fuzz legacy modes, fix uncovered bugs by @terrelln (#1593, #1594, #1595) bug: Fix out of bounds read by @terrelln (#1590) @@ -258,7 +258,7 @@ build: Visual Studio: fix linking by @absotively (#1639) build: Fix MinGW-W64 build by @myzhang1029 (#1600) misc: Expand decodecorpus coverage by @ephiepark (#1664) -v1.4.0 (Apr 17, 2019) +v1.4.0 (Apr 17, 2019) perf: Improve level 1 compression speed in most scenarios by 6% by @gbtucker and @terrelln api: Move the advanced API, including all functions in the staging section, to the stable section api: Make ZSTD_e_flush and ZSTD_e_end block for maximum forward progress @@ -295,7 +295,7 @@ misc: Optimize dictionary memory usage in corner cases misc: Improve the dictionary builder on small or homogeneous data misc: Fix spelling across the repo by @jsoref -v1.3.8 (Dec 28, 2018) +v1.3.8 (Dec 28, 2018) perf: better decompression speed on large files (+7%) and cold dictionaries (+15%) perf: slightly better compression ratio at high compression modes api : finalized advanced API, last stage before "stable" status @@ -317,14 +317,14 @@ doc : clarified zstd_compression_format.md, by @ulikunitz misc: fixed zstdgrep, returns 1 on failure, by @lzutao misc: NEWS renamed as CHANGELOG, in accordance with fboss -v1.3.7 (Oct 20, 2018) +v1.3.7 (Oct 20, 2018) perf: slightly better decompression speed on clang (depending on hardware target) fix : performance of dictionary compression for small input < 4 KB at levels 9 and 10 build: no longer build backtrace by default in release mode; restrict further automatic mode build: control backtrace support through build macro BACKTRACE misc: added man pages for zstdless and zstdgrep, by @samrussell -v1.3.6 (Oct 6, 2018) +v1.3.6 (Oct 6, 2018) perf: much faster dictionary builder, by @jenniferliu perf: faster dictionary compression on small data when using multiple contexts, by @felixhandte perf: faster dictionary decompression when using a very large number of dictionaries simultaneously @@ -338,7 +338,7 @@ build: Read Legacy format is limited to v0.5+ by default. Can be changed at comp doc : zstd_compression_format.md updated to match wording in IETF RFC 8478 misc: tests/paramgrill, a parameter optimizer, by @GeorgeLu97 -v1.3.5 (Jun 29, 2018) +v1.3.5 (Jun 29, 2018) perf: much faster dictionary compression, by @felixhandte perf: small quality improvement for dictionary generation, by @terrelln perf: slightly improved high compression levels (notably level 19) @@ -353,7 +353,7 @@ build: make and make all are compatible with -j doc : clarify zstd_compression_format.md, updated for IETF RFC process misc: pzstd compatible with reproducible compilation, by @lamby -v1.3.4 (Mar 27, 2018) +v1.3.4 (Mar 27, 2018) perf: faster speed (especially decoding speed) on recent cpus (haswell+) perf: much better performance associating --long with multi-threading, by @terrelln perf: better compression at levels 13-15 @@ -371,7 +371,7 @@ build: VS2017 scripts, by @HaydnTrigg misc: all /contrib projects fixed misc: added /contrib/docker script by @gyscos -v1.3.3 (Dec 21, 2017) +v1.3.3 (Dec 21, 2017) perf: faster zstd_opt strategy (levels 16-19) fix : bug #944 : multithreading with shared ditionary and large data, reported by @gsliepen cli : fix : content size written in header by default @@ -383,7 +383,7 @@ api : change : when setting `pledgedSrcSize`, use `ZSTD_CONTENTSIZE_UNKNOWN` mac build: fix : compilation under rhel6 and centos6, reported by @pixelb build: added `check` target -v1.3.2 (Oct 10, 2017) +v1.3.2 (Oct 10, 2017) new : long range mode, using --long command, by Stella Lau (@stellamplau) new : ability to generate and decode magicless frames (#591) changed : maximum nb of threads reduced to 200, to avoid address space exhaustion in 32-bits mode @@ -406,7 +406,7 @@ example : added streaming_memory_usage license : changed /examples license to BSD + GPLv2 license : fix a few header files to reflect new license (#825) -v1.3.1 (Aug 21, 2017) +v1.3.1 (Aug 21, 2017) New license : BSD + GPLv2 perf: substantially decreased memory usage in Multi-threading mode, thanks to reports by Tino Reichardt (@mcmilk) perf: Multi-threading supports up to 256 threads. Cap at 256 when more are requested (#760) @@ -421,7 +421,7 @@ new : contrib/adaptive-compression, I/O driven compression strength, by Paul Cru new : contrib/long_distance_matching, statistics by Stella Lau (@stellamplau) updated : contrib/linux-kernel, by Nick Terrell (@terrelln) -v1.3.0 (Jul 6, 2017) +v1.3.0 (Jul 6, 2017) cli : new : `--list` command, by Paul Cruz cli : changed : xz/lzma support enabled by default cli : changed : `-t *` continue processing list after a decompression error @@ -436,7 +436,7 @@ tools : decodecorpus can generate random dictionary-compressed samples, by Paul new : contrib/seekable_format, demo and API, by Sean Purcell changed : contrib/linux-kernel, updated version and license, by Nick Terrell -v1.2.0 (May 5, 2017) +v1.2.0 (May 5, 2017) cli : changed : Multithreading enabled by default (use target zstd-nomt or HAVE_THREAD=0 to disable) cli : new : command -T0 means "detect and use nb of cores", by Sean Purcell cli : new : zstdmt symlink hardwired to `zstd -T0` @@ -458,7 +458,7 @@ build: enabled Multi-threading support for *BSD, by Baptiste Daroussin tools: updated Paramgrill. Command -O# provides best parameters for sample and speed target. new : contrib/linux-kernel version, by Nick Terrell -v1.1.4 (Mar 18, 2017) +v1.1.4 (Mar 18, 2017) cli : new : can compress in *.gz format, using --format=gzip command, by Przemyslaw Skibinski cli : new : advanced benchmark command --priority=rt cli : fix : write on sparse-enabled file systems in 32-bits mode, by @ds77 @@ -474,7 +474,7 @@ build : improved cmake script, by @Majlen build : added -Wformat-security flag, as recommended by Padraig Brady doc : new : educational decoder, by Sean Purcell -v1.1.3 (Feb 7, 2017) +v1.1.3 (Feb 7, 2017) cli : zstd can decompress .gz files (can be disabled with `make zstd-nogz` or `make HAVE_ZLIB=0`) cli : new : experimental target `make zstdmt`, with multi-threading support cli : new : improved dictionary builder "cover" (experimental), by Nick Terrell, based on prior work by Giuseppe Ottaviano. @@ -490,7 +490,7 @@ API : fix : all symbols properly exposed in libzstd, by Nick Terrell build : support for Solaris target, by Przemyslaw Skibinski doc : clarified specification, by Sean Purcell -v1.1.2 (Dec 15, 2016) +v1.1.2 (Dec 15, 2016) API : streaming : decompression : changed : automatic implicit reset when chain-decoding new frames without init API : experimental : added : dictID retrieval functions, and ZSTD_initCStream_srcSize() API : zbuff : changed : prototypes now generate deprecation warnings @@ -507,7 +507,7 @@ zlib_wrapper : added support for gz* functions, by Przemyslaw Skibinski install : better compatibility with FreeBSD, by Dimitry Andric source tree : changed : zbuff source files moved to lib/deprecated -v1.1.1 (Nov 2, 2016) +v1.1.1 (Nov 2, 2016) New : command -M#, --memory=, --memlimit=, --memlimit-decompress= to limit allowed memory consumption New : doc/zstd_manual.html, by Przemyslaw Skibinski Improved : slightly better compression ratio at --ultra levels (>= 20) @@ -518,7 +518,7 @@ Changed : zstd_errors.h is now installed within /include (and replaces errors_pu Updated man page Fixed : zstd-small, zstd-compress and zstd-decompress compilation targets -v1.1.0 (Sep 28, 2016) +v1.1.0 (Sep 28, 2016) New : contrib/pzstd, parallel version of zstd, by Nick Terrell added : NetBSD install target (#338) Improved : speed for batches of small files @@ -532,7 +532,7 @@ Fixed : compatibility with OpenBSD, reported by Juan Francisco Cantero Hurtado ( Fixed : compatibility with Hurd, by Przemyslaw Skibinski (#365) Fixed : zstd-pgo, reported by octoploid (#329) -v1.0.0 (Sep 1, 2016) +v1.0.0 (Sep 1, 2016) Change Licensing, all project is now BSD, Copyright Facebook Small decompression speed improvement API : Streaming API supports legacy format @@ -541,7 +541,7 @@ CLI supports legacy formats v0.4+ Fixed : compression fails on certain huge files, reported by Jesse McGrew Enhanced documentation, by Przemyslaw Skibinski -v0.8.1 (Aug 18, 2016) +v0.8.1 (Aug 18, 2016) New streaming API Changed : --ultra now enables levels beyond 19 Changed : -i# now selects benchmark time in second @@ -550,7 +550,7 @@ Fixed : speed regression on specific patterns (#272) Fixed : support for Z_SYNC_FLUSH, by Dmitry Krot (#291) Fixed : ICC compilation, by Przemyslaw Skibinski -v0.8.0 (Aug 2, 2016) +v0.8.0 (Aug 2, 2016) Improved : better speed on clang and gcc -O2, thanks to Eric Biggers New : Build on FreeBSD and DragonFly, thanks to JrMarino Changed : modified API : ZSTD_compressEnd() @@ -563,17 +563,17 @@ Modified : minor compression level adaptations Updated : compression format specification to v0.2.0 changed : zstd.h moved to /lib directory -v0.7.5 (Aug 1, 2016) +v0.7.5 (Aug 1, 2016) Transition version, supporting decoding of v0.8.x -v0.7.4 (Jul 17, 2016) +v0.7.4 (Jul 17, 2016) Added : homebrew for Mac, by Daniel Cade Added : more examples Fixed : segfault when using small dictionaries, reported by Felix Handte Modified : default compression level for CLI is now 3 Updated : specification, to v0.1.1 -v0.7.3 (Jul 9, 2016) +v0.7.3 (Jul 9, 2016) New : compression format specification New : `--` separator, stating that all following arguments are file names. Suggested by Chip Turner. New : `ZSTD_getDecompressedSize()` @@ -585,18 +585,18 @@ fixed : multi-blocks decoding with intermediate uncompressed blocks, reported by modified : removed "mem.h" and "error_public.h" dependencies from "zstd.h" (experimental section) modified : legacy functions no longer need magic number -v0.7.2 (Jul 4, 2016) +v0.7.2 (Jul 4, 2016) fixed : ZSTD_decompressBlock() using multiple consecutive blocks. Reported by Greg Slazinski. fixed : potential segfault on very large files (many gigabytes). Reported by Chip Turner. fixed : CLI displays system error message when destination file cannot be created (#231). Reported by Chip Turner. -v0.7.1 (Jun 23, 2016) +v0.7.1 (Jun 23, 2016) fixed : ZBUFF_compressEnd() called multiple times with too small `dst` buffer, reported by Christophe Chevalier fixed : dictBuilder fails if first sample is too small, reported by Руслан Ковалёв fixed : corruption issue, reported by cj modified : checksum enabled by default in command line mode -v0.7.0 (Jun 17, 2016) +v0.7.0 (Jun 17, 2016) New : Support for directory compression, using `-r`, thanks to Przemyslaw Skibinski New : Command `--rm`, to remove source file after successful de/compression New : Visual build scripts, by Christophe Chevalier @@ -609,7 +609,7 @@ API : support for custom malloc/free functions New : controllable Dictionary ID New : Support for skippable frames -v0.6.1 (May 13, 2016) +v0.6.1 (May 13, 2016) New : zlib wrapper API, thanks to Przemyslaw Skibinski New : Ability to compile compressor / decompressor separately Changed : new lib directory structure @@ -619,103 +619,103 @@ Fixed : null-string roundtrip (#176) New : benchmark mode can select directory as input Experimental : midipix support, VMS support -v0.6.0 (Apr 13, 2016) +v0.6.0 (Apr 13, 2016) Stronger high compression modes, thanks to Przemyslaw Skibinski API : ZSTD_getFrameParams() provides size of decompressed content New : highest compression modes require `--ultra` command to fully unleash their capacity Fixed : zstd cli return error code > 0 and removes dst file artifact when decompression fails, thanks to Chip Turner -v0.5.1 (Feb 18, 2016) +v0.5.1 (Feb 18, 2016) New : Optimal parsing => Very high compression modes, thanks to Przemyslaw Skibinski Changed : Dictionary builder integrated into libzstd and zstd cli Changed (!) : zstd cli now uses "multiple input files" as default mode. See `zstd -h`. Fix : high compression modes for big-endian platforms New : zstd cli : `-t` | `--test` command -v0.5.0 (Feb 5, 2016) +v0.5.0 (Feb 5, 2016) New : dictionary builder utility Changed : streaming & dictionary API Improved : better compression of small data -v0.4.7 (Jan 22, 2016) +v0.4.7 (Jan 22, 2016) Improved : small compression speed improvement in HC mode Changed : `zstd_decompress.c` has ZSTD_LEGACY_SUPPORT to 0 by default fix : bt search bug -v0.4.6 (Jan 13, 2016) +v0.4.6 (Jan 13, 2016) fix : fast compression mode on Windows New : cmake configuration file, thanks to Artyom Dymchenko Improved : high compression mode on repetitive data New : block-level API New : ZSTD_duplicateCCtx() -v0.4.5 (Dec 18, 2015) +v0.4.5 (Dec 18, 2015) new : -m/--multiple : compress/decompress multiple files -v0.4.4 (Dec 14, 2015) +v0.4.4 (Dec 14, 2015) Fixed : high compression modes for Windows 32 bits new : external dictionary API extended to buffered mode and accessible through command line new : windows DLL project, thanks to Christophe Chevalier -v0.4.3 (Dec 7, 2015) +v0.4.3 (Dec 7, 2015) new : external dictionary API new : zstd-frugal -v0.4.2 (Dec 2, 2015) +v0.4.2 (Dec 2, 2015) Generic minor improvements for small blocks Fixed : big-endian compatibility, by Peter Harris (#85) -v0.4.1 (Dec 1, 2015) +v0.4.1 (Dec 1, 2015) Fixed : ZSTD_LEGACY_SUPPORT=0 build mode (reported by Luben) removed `zstd.c` -v0.4.0 (Nov 29, 2015) +v0.4.0 (Nov 29, 2015) Command line utility compatible with high compression levels Removed zstdhc => merged into zstd Added : ZBUFF API (see zstd_buffered.h) Rolling buffer support -v0.3.6 (Nov 10, 2015) +v0.3.6 (Nov 10, 2015) small blocks params -v0.3.5 (Nov 9, 2015) +v0.3.5 (Nov 9, 2015) minor generic compression improvements -v0.3.4 (Nov 6, 2015) +v0.3.4 (Nov 6, 2015) Faster fast cLevels -v0.3.3 (Nov 5, 2015) +v0.3.3 (Nov 5, 2015) Small compression ratio improvement -v0.3.2 (Nov 2, 2015) +v0.3.2 (Nov 2, 2015) Fixed Visual Studio -v0.3.1 (Nov 2, 2015) +v0.3.1 (Nov 2, 2015) Small compression ratio improvement -v0.3 (Oct 30, 2015) +v0.3 (Oct 30, 2015) HC mode : compression levels 2-26 -v0.2.2 (Oct 28, 2015) +v0.2.2 (Oct 28, 2015) Fix : Visual Studio 2013 & 2015 release compilation, by Christophe Chevalier -v0.2.1 (Oct 24, 2015) +v0.2.1 (Oct 24, 2015) Fix : Read errors, advanced fuzzer tests, by Hanno Böck -v0.2.0 (Oct 22, 2015) +v0.2.0 (Oct 22, 2015) **Breaking format change** Faster decompression speed Can still decode v0.1 format -v0.1.3 (Oct 15, 2015) +v0.1.3 (Oct 15, 2015) fix uninitialization warning, reported by Evan Nemerson -v0.1.2 (Sep 11, 2015) +v0.1.2 (Sep 11, 2015) frame concatenation support -v0.1.1 (Aug 27, 2015) +v0.1.1 (Aug 27, 2015) fix compression bug detects write-flush errors -v0.1.0 (Aug 25, 2015) +v0.1.0 (Aug 25, 2015) first release diff --git a/contrib/libs/zstd/CONTRIBUTING.md b/contrib/libs/zstd/CONTRIBUTING.md index e7e545129e..de159064f8 100644 --- a/contrib/libs/zstd/CONTRIBUTING.md +++ b/contrib/libs/zstd/CONTRIBUTING.md @@ -5,7 +5,7 @@ possible. ## Our Development Process New versions are being developed in the "dev" branch, or in their own feature branch. -When they are deemed ready for a release, they are merged into "release". +When they are deemed ready for a release, they are merged into "release". As a consequences, all contributions must stage first through "dev" or their own feature branch. @@ -126,20 +126,20 @@ just `contrib/largeNbDicts` and nothing else, you can run: scan-build make -C contrib/largeNbDicts largeNbDicts ``` -### Pitfalls of static analysis -`scan-build` is part of our regular CI suite. Other static analyzers are not. - -It can be useful to look at additional static analyzers once in a while (and we do), but it's not a good idea to multiply the nb of analyzers run continuously at each commit and PR. The reasons are : - -- Static analyzers are full of false positive. The signal to noise ratio is actually pretty low. -- A good CI policy is "zero-warning tolerance". That means that all issues must be solved, including false positives. This quickly becomes a tedious workload. -- Multiple static analyzers will feature multiple kind of false positives, sometimes applying to the same code but in different ways leading to : - + torteous code, trying to please multiple constraints, hurting readability and therefore maintenance. Sometimes, such complexity introduce other more subtle bugs, that are just out of scope of the analyzers. - + sometimes, these constraints are mutually exclusive : if one try to solve one, the other static analyzer will complain, they can't be both happy at the same time. -- As if that was not enough, the list of false positives change with each version. It's hard enough to follow one static analyzer, but multiple ones with their own update agenda, this quickly becomes a massive velocity reducer. - -This is different from running a static analyzer once in a while, looking at the output, and __cherry picking__ a few warnings that seem helpful, either because they detected a genuine risk of bug, or because it helps expressing the code in a way which is more readable or more difficult to misuse. These kind of reports can be useful, and are accepted. - +### Pitfalls of static analysis +`scan-build` is part of our regular CI suite. Other static analyzers are not. + +It can be useful to look at additional static analyzers once in a while (and we do), but it's not a good idea to multiply the nb of analyzers run continuously at each commit and PR. The reasons are : + +- Static analyzers are full of false positive. The signal to noise ratio is actually pretty low. +- A good CI policy is "zero-warning tolerance". That means that all issues must be solved, including false positives. This quickly becomes a tedious workload. +- Multiple static analyzers will feature multiple kind of false positives, sometimes applying to the same code but in different ways leading to : + + torteous code, trying to please multiple constraints, hurting readability and therefore maintenance. Sometimes, such complexity introduce other more subtle bugs, that are just out of scope of the analyzers. + + sometimes, these constraints are mutually exclusive : if one try to solve one, the other static analyzer will complain, they can't be both happy at the same time. +- As if that was not enough, the list of false positives change with each version. It's hard enough to follow one static analyzer, but multiple ones with their own update agenda, this quickly becomes a massive velocity reducer. + +This is different from running a static analyzer once in a while, looking at the output, and __cherry picking__ a few warnings that seem helpful, either because they detected a genuine risk of bug, or because it helps expressing the code in a way which is more readable or more difficult to misuse. These kind of reports can be useful, and are accepted. + ## Continuous Integration CI tests run every time a pull request (PR) is created or updated. The exact tests that get run will depend on the destination branch you specify. Some tests take @@ -384,106 +384,106 @@ disclosure of security bugs. In those cases, please go through the process outlined on that page and do not file a public issue. ## Coding Style -It's a pretty long topic, which is difficult to summarize in a single paragraph. -As a rule of thumbs, try to imitate the coding style of -similar lines of codes around your contribution. -The following is a non-exhaustive list of rules employed in zstd code base: - -### C90 -This code base is following strict C90 standard, -with 2 extensions : 64-bit `long long` types, and variadic macros. -This rule is applied strictly to code within `lib/` and `programs/`. -Sub-project in `contrib/` are allowed to use other conventions. - -### C++ direct compatibility : symbol mangling -All public symbol declarations must be wrapped in `extern “C” { … }`, -so that this project can be compiled as C++98 code, -and linked into C++ applications. - -### Minimal Frugal -This design requirement is fundamental to preserve the portability of the code base. -#### Dependencies -- Reduce dependencies to the minimum possible level. - Any dependency should be considered “bad” by default, - and only tolerated because it provides a service in a better way than can be achieved locally. - The only external dependencies this repository tolerates are - standard C libraries, and in rare cases, system level headers. -- Within `lib/`, this policy is even more drastic. - The only external dependencies allowed are `<assert.h>`, `<stdlib.h>`, `<string.h>`, - and even then, not directly. - In particular, no function shall ever allocate on heap directly, - and must use instead `ZSTD_malloc()` and equivalent. - Other accepted non-symbol headers are `<stddef.h>` and `<limits.h>`. -- Within the project, there is a strict hierarchy of dependencies that must be respected. - `programs/` is allowed to depend on `lib/`, but only its public API. - Within `lib/`, `lib/common` doesn't depend on any other directory. - `lib/compress` and `lib/decompress` shall not depend on each other. - `lib/dictBuilder` can depend on `lib/common` and `lib/compress`, but not `lib/decompress`. -#### Resources -- Functions in `lib/` must use very little stack space, - several dozens of bytes max. - Everything larger must use the heap allocator, - or require a scratch buffer to be emplaced manually. - -### Naming -* All public symbols are prefixed with `ZSTD_` - + private symbols, with a scope limited to their own unit, are free of this restriction. - However, since `libzstd` source code can be amalgamated, - each symbol name must attempt to be (and remain) unique. - Avoid too generic names that could become ground for future collisions. - This generally implies usage of some form of prefix. -* For symbols (functions and variables), naming convention is `PREFIX_camelCase`. - + In some advanced cases, one can also find : - - `PREFIX_prefix2_camelCase` - - `PREFIX_camelCase_extendedQualifier` -* Multi-words names generally consist of an action followed by object: - - for example : `ZSTD_createCCtx()` -* Prefer positive actions - - `goBackward` rather than `notGoForward` -* Type names (`struct`, etc.) follow similar convention, - except that they are allowed and even invited to start by an Uppercase letter. - Example : `ZSTD_CCtx`, `ZSTD_CDict` -* Macro names are all Capital letters. - The same composition rules (`PREFIX_NAME_QUALIFIER`) apply. -* File names are all lowercase letters. - The convention is `snake_case`. - File names **must** be unique across the entire code base, - even when they stand in clearly separated directories. - -### Qualifiers -* This code base is `const` friendly, if not `const` fanatical. - Any variable that can be `const` (aka. read-only) **must** be `const`. - Any pointer which content will not be modified must be `const`. - This property is then controlled at compiler level. - `const` variables are an important signal to readers that this variable isn’t modified. - Conversely, non-const variables are a signal to readers to watch out for modifications later on in the function. -* If a function must be inlined, mention it explicitly, - using project's own portable macros, such as `FORCE_INLINE_ATTR`, - defined in `lib/common/compiler.h`. - -### Debugging -* **Assertions** are welcome, and should be used very liberally, - to control any condition the code expects for its correct execution. - These assertion checks will be run in debug builds, and disabled in production. -* For traces, this project provides its own debug macros, - in particular `DEBUGLOG(level, ...)`, defined in `lib/common/debug.h`. - -### Code documentation -* Avoid code documentation that merely repeats what the code is already stating. - Whenever applicable, prefer employing the code as the primary way to convey explanations. - Example 1 : `int nbTokens = n;` instead of `int i = n; /* i is a nb of tokens *./`. - Example 2 : `assert(size > 0);` instead of `/* here, size should be positive */`. -* At declaration level, the documentation explains how to use the function or variable - and when applicable why it's needed, of the scenarios where it can be useful. -* At implementation level, the documentation explains the general outline of the algorithm employed, - and when applicable why this specific choice was preferred. - -### General layout +It's a pretty long topic, which is difficult to summarize in a single paragraph. +As a rule of thumbs, try to imitate the coding style of +similar lines of codes around your contribution. +The following is a non-exhaustive list of rules employed in zstd code base: + +### C90 +This code base is following strict C90 standard, +with 2 extensions : 64-bit `long long` types, and variadic macros. +This rule is applied strictly to code within `lib/` and `programs/`. +Sub-project in `contrib/` are allowed to use other conventions. + +### C++ direct compatibility : symbol mangling +All public symbol declarations must be wrapped in `extern “C” { … }`, +so that this project can be compiled as C++98 code, +and linked into C++ applications. + +### Minimal Frugal +This design requirement is fundamental to preserve the portability of the code base. +#### Dependencies +- Reduce dependencies to the minimum possible level. + Any dependency should be considered “bad” by default, + and only tolerated because it provides a service in a better way than can be achieved locally. + The only external dependencies this repository tolerates are + standard C libraries, and in rare cases, system level headers. +- Within `lib/`, this policy is even more drastic. + The only external dependencies allowed are `<assert.h>`, `<stdlib.h>`, `<string.h>`, + and even then, not directly. + In particular, no function shall ever allocate on heap directly, + and must use instead `ZSTD_malloc()` and equivalent. + Other accepted non-symbol headers are `<stddef.h>` and `<limits.h>`. +- Within the project, there is a strict hierarchy of dependencies that must be respected. + `programs/` is allowed to depend on `lib/`, but only its public API. + Within `lib/`, `lib/common` doesn't depend on any other directory. + `lib/compress` and `lib/decompress` shall not depend on each other. + `lib/dictBuilder` can depend on `lib/common` and `lib/compress`, but not `lib/decompress`. +#### Resources +- Functions in `lib/` must use very little stack space, + several dozens of bytes max. + Everything larger must use the heap allocator, + or require a scratch buffer to be emplaced manually. + +### Naming +* All public symbols are prefixed with `ZSTD_` + + private symbols, with a scope limited to their own unit, are free of this restriction. + However, since `libzstd` source code can be amalgamated, + each symbol name must attempt to be (and remain) unique. + Avoid too generic names that could become ground for future collisions. + This generally implies usage of some form of prefix. +* For symbols (functions and variables), naming convention is `PREFIX_camelCase`. + + In some advanced cases, one can also find : + - `PREFIX_prefix2_camelCase` + - `PREFIX_camelCase_extendedQualifier` +* Multi-words names generally consist of an action followed by object: + - for example : `ZSTD_createCCtx()` +* Prefer positive actions + - `goBackward` rather than `notGoForward` +* Type names (`struct`, etc.) follow similar convention, + except that they are allowed and even invited to start by an Uppercase letter. + Example : `ZSTD_CCtx`, `ZSTD_CDict` +* Macro names are all Capital letters. + The same composition rules (`PREFIX_NAME_QUALIFIER`) apply. +* File names are all lowercase letters. + The convention is `snake_case`. + File names **must** be unique across the entire code base, + even when they stand in clearly separated directories. + +### Qualifiers +* This code base is `const` friendly, if not `const` fanatical. + Any variable that can be `const` (aka. read-only) **must** be `const`. + Any pointer which content will not be modified must be `const`. + This property is then controlled at compiler level. + `const` variables are an important signal to readers that this variable isn’t modified. + Conversely, non-const variables are a signal to readers to watch out for modifications later on in the function. +* If a function must be inlined, mention it explicitly, + using project's own portable macros, such as `FORCE_INLINE_ATTR`, + defined in `lib/common/compiler.h`. + +### Debugging +* **Assertions** are welcome, and should be used very liberally, + to control any condition the code expects for its correct execution. + These assertion checks will be run in debug builds, and disabled in production. +* For traces, this project provides its own debug macros, + in particular `DEBUGLOG(level, ...)`, defined in `lib/common/debug.h`. + +### Code documentation +* Avoid code documentation that merely repeats what the code is already stating. + Whenever applicable, prefer employing the code as the primary way to convey explanations. + Example 1 : `int nbTokens = n;` instead of `int i = n; /* i is a nb of tokens *./`. + Example 2 : `assert(size > 0);` instead of `/* here, size should be positive */`. +* At declaration level, the documentation explains how to use the function or variable + and when applicable why it's needed, of the scenarios where it can be useful. +* At implementation level, the documentation explains the general outline of the algorithm employed, + and when applicable why this specific choice was preferred. + +### General layout * 4 spaces for indentation rather than tabs -* Code documentation shall directly precede function declaration or implementation -* Function implementations and its code documentation should be preceded and followed by an empty line - +* Code documentation shall directly precede function declaration or implementation +* Function implementations and its code documentation should be preceded and followed by an empty line + ## License By contributing to Zstandard, you agree that your contributions will be licensed under both the [LICENSE](LICENSE) file and the [COPYING](COPYING) file in the root directory of this source tree. diff --git a/contrib/libs/zstd/README.md b/contrib/libs/zstd/README.md index 69720ba2cc..4329c71401 100644 --- a/contrib/libs/zstd/README.md +++ b/contrib/libs/zstd/README.md @@ -1,49 +1,49 @@ -<p align="center"><img src="https://raw.githubusercontent.com/facebook/zstd/dev/doc/images/zstd_logo86.png" alt="Zstandard"></p> +<p align="center"><img src="https://raw.githubusercontent.com/facebook/zstd/dev/doc/images/zstd_logo86.png" alt="Zstandard"></p> -__Zstandard__, or `zstd` as short version, is a fast lossless compression algorithm, -targeting real-time compression scenarios at zlib-level and better compression ratios. -It's backed by a very fast entropy stage, provided by [Huff0 and FSE library](https://github.com/Cyan4973/FiniteStateEntropy). +__Zstandard__, or `zstd` as short version, is a fast lossless compression algorithm, +targeting real-time compression scenarios at zlib-level and better compression ratios. +It's backed by a very fast entropy stage, provided by [Huff0 and FSE library](https://github.com/Cyan4973/FiniteStateEntropy). Zstandard's format is stable and documented in [RFC8878](https://datatracker.ietf.org/doc/html/rfc8878). Multiple independent implementations are already available. This repository represents the reference implementation, provided as an open-source dual [BSD](LICENSE) and [GPLv2](COPYING) licensed **C** library, -and a command line utility producing and decoding `.zst`, `.gz`, `.xz` and `.lz4` files. -Should your project require another programming language, -a list of known ports and bindings is provided on [Zstandard homepage](http://www.zstd.net/#other-languages). +and a command line utility producing and decoding `.zst`, `.gz`, `.xz` and `.lz4` files. +Should your project require another programming language, +a list of known ports and bindings is provided on [Zstandard homepage](http://www.zstd.net/#other-languages). -**Development branch status:** +**Development branch status:** -[![Build Status][travisDevBadge]][travisLink] -[![Build status][AppveyorDevBadge]][AppveyorLink] -[![Build status][CircleDevBadge]][CircleLink] -[![Build status][CirrusDevBadge]][CirrusLink] -[![Fuzzing Status][OSSFuzzBadge]][OSSFuzzLink] +[![Build Status][travisDevBadge]][travisLink] +[![Build status][AppveyorDevBadge]][AppveyorLink] +[![Build status][CircleDevBadge]][CircleLink] +[![Build status][CirrusDevBadge]][CirrusLink] +[![Fuzzing Status][OSSFuzzBadge]][OSSFuzzLink] [travisDevBadge]: https://api.travis-ci.com/facebook/zstd.svg?branch=dev "Continuous Integration test suite" [travisLink]: https://travis-ci.com/facebook/zstd -[AppveyorDevBadge]: https://ci.appveyor.com/api/projects/status/xt38wbdxjk5mrbem/branch/dev?svg=true "Windows test suite" -[AppveyorLink]: https://ci.appveyor.com/project/YannCollet/zstd-p0yf0 -[CircleDevBadge]: https://circleci.com/gh/facebook/zstd/tree/dev.svg?style=shield "Short test suite" -[CircleLink]: https://circleci.com/gh/facebook/zstd -[CirrusDevBadge]: https://api.cirrus-ci.com/github/facebook/zstd.svg?branch=dev -[CirrusLink]: https://cirrus-ci.com/github/facebook/zstd -[OSSFuzzBadge]: https://oss-fuzz-build-logs.storage.googleapis.com/badges/zstd.svg -[OSSFuzzLink]: https://bugs.chromium.org/p/oss-fuzz/issues/list?sort=-opened&can=1&q=proj:zstd - -## Benchmarks - -For reference, several fast compression algorithms were tested and compared +[AppveyorDevBadge]: https://ci.appveyor.com/api/projects/status/xt38wbdxjk5mrbem/branch/dev?svg=true "Windows test suite" +[AppveyorLink]: https://ci.appveyor.com/project/YannCollet/zstd-p0yf0 +[CircleDevBadge]: https://circleci.com/gh/facebook/zstd/tree/dev.svg?style=shield "Short test suite" +[CircleLink]: https://circleci.com/gh/facebook/zstd +[CirrusDevBadge]: https://api.cirrus-ci.com/github/facebook/zstd.svg?branch=dev +[CirrusLink]: https://cirrus-ci.com/github/facebook/zstd +[OSSFuzzBadge]: https://oss-fuzz-build-logs.storage.googleapis.com/badges/zstd.svg +[OSSFuzzLink]: https://bugs.chromium.org/p/oss-fuzz/issues/list?sort=-opened&can=1&q=proj:zstd + +## Benchmarks + +For reference, several fast compression algorithms were tested and compared on a desktop running Ubuntu 20.04 (`Linux 5.11.0-41-generic`), with a Core i7-9700K CPU @ 4.9GHz, -using [lzbench], an open-source in-memory benchmark by @inikep -compiled with [gcc] 9.3.0, -on the [Silesia compression corpus]. +using [lzbench], an open-source in-memory benchmark by @inikep +compiled with [gcc] 9.3.0, +on the [Silesia compression corpus]. -[lzbench]: https://github.com/inikep/lzbench -[Silesia compression corpus]: http://sun.aei.polsl.pl/~sdeor/index.php?page=silesia -[gcc]: https://gcc.gnu.org/ +[lzbench]: https://github.com/inikep/lzbench +[Silesia compression corpus]: http://sun.aei.polsl.pl/~sdeor/index.php?page=silesia +[gcc]: https://gcc.gnu.org/ -| Compressor name | Ratio | Compression| Decompress.| -| --------------- | ------| -----------| ---------- | +| Compressor name | Ratio | Compression| Decompress.| +| --------------- | ------| -----------| ---------- | | **zstd 1.5.1 -1** | 2.887 | 530 MB/s | 1700 MB/s | | [zlib] 1.2.11 -1 | 2.743 | 95 MB/s | 400 MB/s | | brotli 1.0.9 -0 | 2.702 | 395 MB/s | 450 MB/s | @@ -56,144 +56,144 @@ on the [Silesia compression corpus]. | lzf 3.6 -1 | 2.077 | 410 MB/s | 830 MB/s | | snappy 1.1.9 | 2.073 | 550 MB/s | 1750 MB/s | -[zlib]: http://www.zlib.net/ +[zlib]: http://www.zlib.net/ [lz4]: http://www.lz4.org/ -The negative compression levels, specified with `--fast=#`, +The negative compression levels, specified with `--fast=#`, offer faster compression and decompression speed at the cost of compression ratio (compared to level 1). -Zstd can also offer stronger compression ratios at the cost of compression speed. -Speed vs Compression trade-off is configurable by small increments. -Decompression speed is preserved and remains roughly the same at all settings, -a property shared by most LZ compression algorithms, such as [zlib] or lzma. +Zstd can also offer stronger compression ratios at the cost of compression speed. +Speed vs Compression trade-off is configurable by small increments. +Decompression speed is preserved and remains roughly the same at all settings, +a property shared by most LZ compression algorithms, such as [zlib] or lzma. -The following tests were run -on a server running Linux Debian (`Linux version 4.14.0-3-amd64`) -with a Core i7-6700K CPU @ 4.0GHz, -using [lzbench], an open-source in-memory benchmark by @inikep -compiled with [gcc] 7.3.0, -on the [Silesia compression corpus]. +The following tests were run +on a server running Linux Debian (`Linux version 4.14.0-3-amd64`) +with a Core i7-6700K CPU @ 4.0GHz, +using [lzbench], an open-source in-memory benchmark by @inikep +compiled with [gcc] 7.3.0, +on the [Silesia compression corpus]. -Compression Speed vs Ratio | Decompression Speed ----------------------------|-------------------- - |  +Compression Speed vs Ratio | Decompression Speed +---------------------------|-------------------- + |  -A few other algorithms can produce higher compression ratios at slower speeds, falling outside of the graph. -For a larger picture including slow modes, [click on this link](doc/images/DCspeed5.png). +A few other algorithms can produce higher compression ratios at slower speeds, falling outside of the graph. +For a larger picture including slow modes, [click on this link](doc/images/DCspeed5.png). -## The case for Small Data compression +## The case for Small Data compression -Previous charts provide results applicable to typical file and stream scenarios (several MB). Small data comes with different perspectives. +Previous charts provide results applicable to typical file and stream scenarios (several MB). Small data comes with different perspectives. -The smaller the amount of data to compress, the more difficult it is to compress. This problem is common to all compression algorithms, and reason is, compression algorithms learn from past data how to compress future data. But at the beginning of a new data set, there is no "past" to build upon. +The smaller the amount of data to compress, the more difficult it is to compress. This problem is common to all compression algorithms, and reason is, compression algorithms learn from past data how to compress future data. But at the beginning of a new data set, there is no "past" to build upon. -To solve this situation, Zstd offers a __training mode__, which can be used to tune the algorithm for a selected type of data. -Training Zstandard is achieved by providing it with a few samples (one file per sample). The result of this training is stored in a file called "dictionary", which must be loaded before compression and decompression. -Using this dictionary, the compression ratio achievable on small data improves dramatically. +To solve this situation, Zstd offers a __training mode__, which can be used to tune the algorithm for a selected type of data. +Training Zstandard is achieved by providing it with a few samples (one file per sample). The result of this training is stored in a file called "dictionary", which must be loaded before compression and decompression. +Using this dictionary, the compression ratio achievable on small data improves dramatically. -The following example uses the `github-users` [sample set](https://github.com/facebook/zstd/releases/tag/v1.1.3), created from [github public API](https://developer.github.com/v3/users/#get-all-users). -It consists of roughly 10K records weighing about 1KB each. +The following example uses the `github-users` [sample set](https://github.com/facebook/zstd/releases/tag/v1.1.3), created from [github public API](https://developer.github.com/v3/users/#get-all-users). +It consists of roughly 10K records weighing about 1KB each. -Compression Ratio | Compression Speed | Decompression Speed -------------------|-------------------|-------------------- - |  |  +Compression Ratio | Compression Speed | Decompression Speed +------------------|-------------------|-------------------- + |  |  -These compression gains are achieved while simultaneously providing _faster_ compression and decompression speeds. +These compression gains are achieved while simultaneously providing _faster_ compression and decompression speeds. -Training works if there is some correlation in a family of small data samples. The more data-specific a dictionary is, the more efficient it is (there is no _universal dictionary_). -Hence, deploying one dictionary per type of data will provide the greatest benefits. -Dictionary gains are mostly effective in the first few KB. Then, the compression algorithm will gradually use previously decoded content to better compress the rest of the file. +Training works if there is some correlation in a family of small data samples. The more data-specific a dictionary is, the more efficient it is (there is no _universal dictionary_). +Hence, deploying one dictionary per type of data will provide the greatest benefits. +Dictionary gains are mostly effective in the first few KB. Then, the compression algorithm will gradually use previously decoded content to better compress the rest of the file. -### Dictionary compression How To: +### Dictionary compression How To: -1. Create the dictionary +1. Create the dictionary - `zstd --train FullPathToTrainingSet/* -o dictionaryName` + `zstd --train FullPathToTrainingSet/* -o dictionaryName` -2. Compress with dictionary +2. Compress with dictionary - `zstd -D dictionaryName FILE` + `zstd -D dictionaryName FILE` -3. Decompress with dictionary +3. Decompress with dictionary - `zstd -D dictionaryName --decompress FILE.zst` + `zstd -D dictionaryName --decompress FILE.zst` -## Build instructions +## Build instructions -### Makefile +### Makefile -If your system is compatible with standard `make` (or `gmake`), -invoking `make` in root directory will generate `zstd` cli in root directory. +If your system is compatible with standard `make` (or `gmake`), +invoking `make` in root directory will generate `zstd` cli in root directory. -Other available options include: -- `make install` : create and install zstd cli, library and man pages -- `make check` : create and run `zstd`, tests its behavior on local platform +Other available options include: +- `make install` : create and install zstd cli, library and man pages +- `make check` : create and run `zstd`, tests its behavior on local platform -### cmake +### cmake -A `cmake` project generator is provided within `build/cmake`. -It can generate Makefiles or other build scripts -to create `zstd` binary, and `libzstd` dynamic and static libraries. +A `cmake` project generator is provided within `build/cmake`. +It can generate Makefiles or other build scripts +to create `zstd` binary, and `libzstd` dynamic and static libraries. -By default, `CMAKE_BUILD_TYPE` is set to `Release`. +By default, `CMAKE_BUILD_TYPE` is set to `Release`. -### Meson +### Meson -A Meson project is provided within [`build/meson`](build/meson). Follow -build instructions in that directory. +A Meson project is provided within [`build/meson`](build/meson). Follow +build instructions in that directory. -You can also take a look at [`.travis.yml`](.travis.yml) file for an -example about how Meson is used to build this project. +You can also take a look at [`.travis.yml`](.travis.yml) file for an +example about how Meson is used to build this project. -Note that default build type is **release**. +Note that default build type is **release**. -### VCPKG -You can build and install zstd [vcpkg](https://github.com/Microsoft/vcpkg/) dependency manager: +### VCPKG +You can build and install zstd [vcpkg](https://github.com/Microsoft/vcpkg/) dependency manager: - git clone https://github.com/Microsoft/vcpkg.git - cd vcpkg - ./bootstrap-vcpkg.sh - ./vcpkg integrate install - ./vcpkg install zstd + git clone https://github.com/Microsoft/vcpkg.git + cd vcpkg + ./bootstrap-vcpkg.sh + ./vcpkg integrate install + ./vcpkg install zstd -The zstd port in vcpkg is kept up to date by Microsoft team members and community contributors. -If the version is out of date, please [create an issue or pull request](https://github.com/Microsoft/vcpkg) on the vcpkg repository. +The zstd port in vcpkg is kept up to date by Microsoft team members and community contributors. +If the version is out of date, please [create an issue or pull request](https://github.com/Microsoft/vcpkg) on the vcpkg repository. -### Visual Studio (Windows) +### Visual Studio (Windows) -Going into `build` directory, you will find additional possibilities: -- Projects for Visual Studio 2005, 2008 and 2010. - + VS2010 project is compatible with VS2012, VS2013, VS2015 and VS2017. -- Automated build scripts for Visual compiler by [@KrzysFR](https://github.com/KrzysFR), in `build/VS_scripts`, - which will build `zstd` cli and `libzstd` library without any need to open Visual Studio solution. +Going into `build` directory, you will find additional possibilities: +- Projects for Visual Studio 2005, 2008 and 2010. + + VS2010 project is compatible with VS2012, VS2013, VS2015 and VS2017. +- Automated build scripts for Visual compiler by [@KrzysFR](https://github.com/KrzysFR), in `build/VS_scripts`, + which will build `zstd` cli and `libzstd` library without any need to open Visual Studio solution. -### Buck +### Buck -You can build the zstd binary via buck by executing: `buck build programs:zstd` from the root of the repo. -The output binary will be in `buck-out/gen/programs/`. +You can build the zstd binary via buck by executing: `buck build programs:zstd` from the root of the repo. +The output binary will be in `buck-out/gen/programs/`. -## Testing - -You can run quick local smoke tests by executing the `playTest.sh` script from the `src/tests` directory. -Two env variables `$ZSTD_BIN` and `$DATAGEN_BIN` are needed for the test script to locate the zstd and datagen binary. -For information on CI testing, please refer to TESTING.md - -## Status - -Zstandard is currently deployed within Facebook. It is used continuously to compress large amounts of data in multiple formats and use cases. -Zstandard is considered safe for production environments. - -## License - -Zstandard is dual-licensed under [BSD](LICENSE) and [GPLv2](COPYING). - -## Contributing - -The `dev` branch is the one where all contributions are merged before reaching `release`. -If you plan to propose a patch, please commit into the `dev` branch, or its own feature branch. -Direct commit to `release` are not permitted. -For more information, please read [CONTRIBUTING](CONTRIBUTING.md). +## Testing + +You can run quick local smoke tests by executing the `playTest.sh` script from the `src/tests` directory. +Two env variables `$ZSTD_BIN` and `$DATAGEN_BIN` are needed for the test script to locate the zstd and datagen binary. +For information on CI testing, please refer to TESTING.md + +## Status + +Zstandard is currently deployed within Facebook. It is used continuously to compress large amounts of data in multiple formats and use cases. +Zstandard is considered safe for production environments. + +## License + +Zstandard is dual-licensed under [BSD](LICENSE) and [GPLv2](COPYING). + +## Contributing + +The `dev` branch is the one where all contributions are merged before reaching `release`. +If you plan to propose a patch, please commit into the `dev` branch, or its own feature branch. +Direct commit to `release` are not permitted. +For more information, please read [CONTRIBUTING](CONTRIBUTING.md). diff --git a/contrib/libs/zstd/TESTING.md b/contrib/libs/zstd/TESTING.md index 32b133b67b..db433d0d30 100644 --- a/contrib/libs/zstd/TESTING.md +++ b/contrib/libs/zstd/TESTING.md @@ -27,7 +27,7 @@ They consist of the following tests: Long Tests ---------- -Long tests run on all commits to `release` branch, +Long tests run on all commits to `release` branch, and once a day on the current version of `dev` branch, on TravisCI. They consist of the following tests: diff --git a/contrib/libs/zstd/include/zdict.h b/contrib/libs/zstd/include/zdict.h index d85f9e1c9e..777c8c6376 100644 --- a/contrib/libs/zstd/include/zdict.h +++ b/contrib/libs/zstd/include/zdict.h @@ -1 +1 @@ -#include "../lib/zdict.h" /* inclink generated by yamaker */ +#include "../lib/zdict.h" /* inclink generated by yamaker */ diff --git a/contrib/libs/zstd/include/zstd.h b/contrib/libs/zstd/include/zstd.h index 48558d73ca..d6c5773e3e 100644 --- a/contrib/libs/zstd/include/zstd.h +++ b/contrib/libs/zstd/include/zstd.h @@ -1 +1 @@ -#include "../lib/zstd.h" /* inclink generated by yamaker */ +#include "../lib/zstd.h" /* inclink generated by yamaker */ diff --git a/contrib/libs/zstd/include/zstd_errors.h b/contrib/libs/zstd/include/zstd_errors.h index 932b4cf652..260cf32094 100644 --- a/contrib/libs/zstd/include/zstd_errors.h +++ b/contrib/libs/zstd/include/zstd_errors.h @@ -1 +1 @@ -#include "../lib/zstd_errors.h" /* inclink generated by yamaker */ +#include "../lib/zstd_errors.h" /* inclink generated by yamaker */ diff --git a/contrib/libs/zstd/lib/README.md b/contrib/libs/zstd/lib/README.md index 4c9d8f0591..52f68fdbe8 100644 --- a/contrib/libs/zstd/lib/README.md +++ b/contrib/libs/zstd/lib/README.md @@ -1,159 +1,159 @@ -Zstandard library files -================================ +Zstandard library files +================================ -The __lib__ directory is split into several sub-directories, -in order to make it easier to select or exclude features. +The __lib__ directory is split into several sub-directories, +in order to make it easier to select or exclude features. -#### Building +#### Building -`Makefile` script is provided, supporting [Makefile conventions](https://www.gnu.org/prep/standards/html_node/Makefile-Conventions.html#Makefile-Conventions), -including commands variables, staged install, directory variables and standard targets. -- `make` : generates both static and dynamic libraries -- `make install` : install libraries and headers in target system directories +`Makefile` script is provided, supporting [Makefile conventions](https://www.gnu.org/prep/standards/html_node/Makefile-Conventions.html#Makefile-Conventions), +including commands variables, staged install, directory variables and standard targets. +- `make` : generates both static and dynamic libraries +- `make install` : install libraries and headers in target system directories -`libzstd` default scope is pretty large, including compression, decompression, dictionary builder, -and support for decoding legacy formats >= v0.5.0. -The scope can be reduced on demand (see paragraph _modular build_). +`libzstd` default scope is pretty large, including compression, decompression, dictionary builder, +and support for decoding legacy formats >= v0.5.0. +The scope can be reduced on demand (see paragraph _modular build_). -#### Multithreading support +#### Multithreading support -When building with `make`, by default the dynamic library is multithreaded and static library is single-threaded (for compatibility reasons). +When building with `make`, by default the dynamic library is multithreaded and static library is single-threaded (for compatibility reasons). + +Enabling multithreading requires 2 conditions : +- set build macro `ZSTD_MULTITHREAD` (`-DZSTD_MULTITHREAD` for `gcc`) +- for POSIX systems : compile with pthread (`-pthread` compilation flag for `gcc`) -Enabling multithreading requires 2 conditions : -- set build macro `ZSTD_MULTITHREAD` (`-DZSTD_MULTITHREAD` for `gcc`) -- for POSIX systems : compile with pthread (`-pthread` compilation flag for `gcc`) +For convenience, we provide a build target to generate multi and single threaded libraries: +- Force enable multithreading on both dynamic and static libraries by appending `-mt` to the target, e.g. `make lib-mt`. +- Force disable multithreading on both dynamic and static libraries by appending `-nomt` to the target, e.g. `make lib-nomt`. +- By default, as mentioned before, dynamic library is multithreaded, and static library is single-threaded, e.g. `make lib`. -For convenience, we provide a build target to generate multi and single threaded libraries: -- Force enable multithreading on both dynamic and static libraries by appending `-mt` to the target, e.g. `make lib-mt`. -- Force disable multithreading on both dynamic and static libraries by appending `-nomt` to the target, e.g. `make lib-nomt`. -- By default, as mentioned before, dynamic library is multithreaded, and static library is single-threaded, e.g. `make lib`. +When linking a POSIX program with a multithreaded version of `libzstd`, +note that it's necessary to invoke the `-pthread` flag during link stage. -When linking a POSIX program with a multithreaded version of `libzstd`, -note that it's necessary to invoke the `-pthread` flag during link stage. +Multithreading capabilities are exposed +via the [advanced API defined in `lib/zstd.h`](https://github.com/facebook/zstd/blob/v1.4.3/lib/zstd.h#L351). -Multithreading capabilities are exposed -via the [advanced API defined in `lib/zstd.h`](https://github.com/facebook/zstd/blob/v1.4.3/lib/zstd.h#L351). +#### API -#### API +Zstandard's stable API is exposed within [lib/zstd.h](zstd.h). -Zstandard's stable API is exposed within [lib/zstd.h](zstd.h). +#### Advanced API -#### Advanced API +Optional advanced features are exposed via : -Optional advanced features are exposed via : +- `lib/zstd_errors.h` : translates `size_t` function results + into a `ZSTD_ErrorCode`, for accurate error handling. -- `lib/zstd_errors.h` : translates `size_t` function results - into a `ZSTD_ErrorCode`, for accurate error handling. +- `ZSTD_STATIC_LINKING_ONLY` : if this macro is defined _before_ including `zstd.h`, + it unlocks access to the experimental API, + exposed in the second part of `zstd.h`. + All definitions in the experimental APIs are unstable, + they may still change in the future, or even be removed. + As a consequence, experimental definitions shall ___never be used with dynamic library___ ! + Only static linking is allowed. -- `ZSTD_STATIC_LINKING_ONLY` : if this macro is defined _before_ including `zstd.h`, - it unlocks access to the experimental API, - exposed in the second part of `zstd.h`. - All definitions in the experimental APIs are unstable, - they may still change in the future, or even be removed. - As a consequence, experimental definitions shall ___never be used with dynamic library___ ! - Only static linking is allowed. +#### Modular build -#### Modular build +It's possible to compile only a limited set of features within `libzstd`. +The file structure is designed to make this selection manually achievable for any build system : -It's possible to compile only a limited set of features within `libzstd`. -The file structure is designed to make this selection manually achievable for any build system : +- Directory `lib/common` is always required, for all variants. -- Directory `lib/common` is always required, for all variants. +- Compression source code lies in `lib/compress` -- Compression source code lies in `lib/compress` +- Decompression source code lies in `lib/decompress` -- Decompression source code lies in `lib/decompress` +- It's possible to include only `compress` or only `decompress`, they don't depend on each other. -- It's possible to include only `compress` or only `decompress`, they don't depend on each other. +- `lib/dictBuilder` : makes it possible to generate dictionaries from a set of samples. + The API is exposed in `lib/dictBuilder/zdict.h`. + This module depends on both `lib/common` and `lib/compress` . -- `lib/dictBuilder` : makes it possible to generate dictionaries from a set of samples. - The API is exposed in `lib/dictBuilder/zdict.h`. - This module depends on both `lib/common` and `lib/compress` . +- `lib/legacy` : makes it possible to decompress legacy zstd formats, starting from `v0.1.0`. + This module depends on `lib/common` and `lib/decompress`. + To enable this feature, define `ZSTD_LEGACY_SUPPORT` during compilation. + Specifying a number limits versions supported to that version onward. + For example, `ZSTD_LEGACY_SUPPORT=2` means : "support legacy formats >= v0.2.0". + Conversely, `ZSTD_LEGACY_SUPPORT=0` means "do __not__ support legacy formats". + By default, this build macro is set as `ZSTD_LEGACY_SUPPORT=5`. + Decoding supported legacy format is a transparent capability triggered within decompression functions. + It's also allowed to invoke legacy API directly, exposed in `lib/legacy/zstd_legacy.h`. + Each version does also provide its own set of advanced API. + For example, advanced API for version `v0.4` is exposed in `lib/legacy/zstd_v04.h` . -- `lib/legacy` : makes it possible to decompress legacy zstd formats, starting from `v0.1.0`. - This module depends on `lib/common` and `lib/decompress`. - To enable this feature, define `ZSTD_LEGACY_SUPPORT` during compilation. - Specifying a number limits versions supported to that version onward. - For example, `ZSTD_LEGACY_SUPPORT=2` means : "support legacy formats >= v0.2.0". - Conversely, `ZSTD_LEGACY_SUPPORT=0` means "do __not__ support legacy formats". - By default, this build macro is set as `ZSTD_LEGACY_SUPPORT=5`. - Decoding supported legacy format is a transparent capability triggered within decompression functions. - It's also allowed to invoke legacy API directly, exposed in `lib/legacy/zstd_legacy.h`. - Each version does also provide its own set of advanced API. - For example, advanced API for version `v0.4` is exposed in `lib/legacy/zstd_v04.h` . +- While invoking `make libzstd`, it's possible to define build macros + `ZSTD_LIB_COMPRESSION, ZSTD_LIB_DECOMPRESSION`, `ZSTD_LIB_DICTBUILDER`, + and `ZSTD_LIB_DEPRECATED` as `0` to forgo compilation of the + corresponding features. This will also disable compilation of all + dependencies (eg. `ZSTD_LIB_COMPRESSION=0` will also disable + dictBuilder). -- While invoking `make libzstd`, it's possible to define build macros - `ZSTD_LIB_COMPRESSION, ZSTD_LIB_DECOMPRESSION`, `ZSTD_LIB_DICTBUILDER`, - and `ZSTD_LIB_DEPRECATED` as `0` to forgo compilation of the - corresponding features. This will also disable compilation of all - dependencies (eg. `ZSTD_LIB_COMPRESSION=0` will also disable - dictBuilder). +- There are a number of options that can help minimize the binary size of + `libzstd`. -- There are a number of options that can help minimize the binary size of - `libzstd`. + The first step is to select the components needed (using the above-described + `ZSTD_LIB_COMPRESSION` etc.). - The first step is to select the components needed (using the above-described - `ZSTD_LIB_COMPRESSION` etc.). + The next step is to set `ZSTD_LIB_MINIFY` to `1` when invoking `make`. This + disables various optional components and changes the compilation flags to + prioritize space-saving. - The next step is to set `ZSTD_LIB_MINIFY` to `1` when invoking `make`. This - disables various optional components and changes the compilation flags to - prioritize space-saving. + Detailed options: Zstandard's code and build environment is set up by default + to optimize above all else for performance. In pursuit of this goal, Zstandard + makes significant trade-offs in code size. For example, Zstandard often has + more than one implementation of a particular component, with each + implementation optimized for different scenarios. For example, the Huffman + decoder has complementary implementations that decode the stream one symbol at + a time or two symbols at a time. Zstd normally includes both (and dispatches + between them at runtime), but by defining `HUF_FORCE_DECOMPRESS_X1` or + `HUF_FORCE_DECOMPRESS_X2`, you can force the use of one or the other, avoiding + compilation of the other. Similarly, `ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT` + and `ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG` force the compilation and use of + only one or the other of two decompression implementations. The smallest + binary is achieved by using `HUF_FORCE_DECOMPRESS_X1` and + `ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT` (implied by `ZSTD_LIB_MINIFY`). - Detailed options: Zstandard's code and build environment is set up by default - to optimize above all else for performance. In pursuit of this goal, Zstandard - makes significant trade-offs in code size. For example, Zstandard often has - more than one implementation of a particular component, with each - implementation optimized for different scenarios. For example, the Huffman - decoder has complementary implementations that decode the stream one symbol at - a time or two symbols at a time. Zstd normally includes both (and dispatches - between them at runtime), but by defining `HUF_FORCE_DECOMPRESS_X1` or - `HUF_FORCE_DECOMPRESS_X2`, you can force the use of one or the other, avoiding - compilation of the other. Similarly, `ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT` - and `ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG` force the compilation and use of - only one or the other of two decompression implementations. The smallest - binary is achieved by using `HUF_FORCE_DECOMPRESS_X1` and - `ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT` (implied by `ZSTD_LIB_MINIFY`). + For squeezing the last ounce of size out, you can also define + `ZSTD_NO_INLINE`, which disables inlining, and `ZSTD_STRIP_ERROR_STRINGS`, + which removes the error messages that are otherwise returned by + `ZSTD_getErrorName` (implied by `ZSTD_LIB_MINIFY`). - For squeezing the last ounce of size out, you can also define - `ZSTD_NO_INLINE`, which disables inlining, and `ZSTD_STRIP_ERROR_STRINGS`, - which removes the error messages that are otherwise returned by - `ZSTD_getErrorName` (implied by `ZSTD_LIB_MINIFY`). - - Finally, when integrating into your application, make sure you're doing link- + Finally, when integrating into your application, make sure you're doing link- time optimization and unused symbol garbage collection (via some combination of, - e.g., `-flto`, `-ffat-lto-objects`, `-fuse-linker-plugin`, - `-ffunction-sections`, `-fdata-sections`, `-fmerge-all-constants`, - `-Wl,--gc-sections`, `-Wl,-z,norelro`, and an archiver that understands - the compiler's intermediate representation, e.g., `AR=gcc-ar`). Consult your - compiler's documentation. - -- While invoking `make libzstd`, the build macro `ZSTD_LEGACY_MULTITHREADED_API=1` - will expose the deprecated `ZSTDMT` API exposed by `zstdmt_compress.h` in - the shared library, which is now hidden by default. - -- The build macro `DYNAMIC_BMI2` can be set to 1 or 0 in order to generate binaries - which can detect at runtime the presence of BMI2 instructions, and use them only if present. - These instructions contribute to better performance, notably on the decoder side. - By default, this feature is automatically enabled on detecting - the right instruction set (x64) and compiler (clang or gcc >= 5). - It's obviously disabled for different cpus, - or when BMI2 instruction set is _required_ by the compiler command line - (in this case, only the BMI2 code path is generated). - Setting this macro will either force to generate the BMI2 dispatcher (1) - or prevent it (0). It overrides automatic detection. - -- The build macro `ZSTD_NO_UNUSED_FUNCTIONS` can be defined to hide the definitions of functions - that zstd does not use. Not all unused functions are hidden, but they can be if needed. - Currently, this macro will hide function definitions in FSE and HUF that use an excessive - amount of stack space. - -- The build macro `ZSTD_NO_INTRINSICS` can be defined to disable all explicit intrinsics. - Compiler builtins are still used. + e.g., `-flto`, `-ffat-lto-objects`, `-fuse-linker-plugin`, + `-ffunction-sections`, `-fdata-sections`, `-fmerge-all-constants`, + `-Wl,--gc-sections`, `-Wl,-z,norelro`, and an archiver that understands + the compiler's intermediate representation, e.g., `AR=gcc-ar`). Consult your + compiler's documentation. + +- While invoking `make libzstd`, the build macro `ZSTD_LEGACY_MULTITHREADED_API=1` + will expose the deprecated `ZSTDMT` API exposed by `zstdmt_compress.h` in + the shared library, which is now hidden by default. + +- The build macro `DYNAMIC_BMI2` can be set to 1 or 0 in order to generate binaries + which can detect at runtime the presence of BMI2 instructions, and use them only if present. + These instructions contribute to better performance, notably on the decoder side. + By default, this feature is automatically enabled on detecting + the right instruction set (x64) and compiler (clang or gcc >= 5). + It's obviously disabled for different cpus, + or when BMI2 instruction set is _required_ by the compiler command line + (in this case, only the BMI2 code path is generated). + Setting this macro will either force to generate the BMI2 dispatcher (1) + or prevent it (0). It overrides automatic detection. + +- The build macro `ZSTD_NO_UNUSED_FUNCTIONS` can be defined to hide the definitions of functions + that zstd does not use. Not all unused functions are hidden, but they can be if needed. + Currently, this macro will hide function definitions in FSE and HUF that use an excessive + amount of stack space. + +- The build macro `ZSTD_NO_INTRINSICS` can be defined to disable all explicit intrinsics. + Compiler builtins are still used. - The build macro `ZSTD_DECODER_INTERNAL_BUFFER` can be set to control the amount of extra memory used during decompression to store literals. @@ -162,56 +162,56 @@ The file structure is designed to make this selection manually achievable for an but might also result in a small decompression speed cost. -#### Windows : using MinGW+MSYS to create DLL - -DLL can be created using MinGW+MSYS with the `make libzstd` command. -This command creates `dll\libzstd.dll` and the import library `dll\libzstd.lib`. -The import library is only required with Visual C++. -The header file `zstd.h` and the dynamic library `dll\libzstd.dll` are required to -compile a project using gcc/MinGW. -The dynamic library has to be added to linking options. -It means that if a project that uses ZSTD consists of a single `test-dll.c` -file it should be linked with `dll\libzstd.dll`. For example: -``` - gcc $(CFLAGS) -Iinclude/ test-dll.c -o test-dll dll\libzstd.dll -``` -The compiled executable will require ZSTD DLL which is available at `dll\libzstd.dll`. +#### Windows : using MinGW+MSYS to create DLL +DLL can be created using MinGW+MSYS with the `make libzstd` command. +This command creates `dll\libzstd.dll` and the import library `dll\libzstd.lib`. +The import library is only required with Visual C++. +The header file `zstd.h` and the dynamic library `dll\libzstd.dll` are required to +compile a project using gcc/MinGW. +The dynamic library has to be added to linking options. +It means that if a project that uses ZSTD consists of a single `test-dll.c` +file it should be linked with `dll\libzstd.dll`. For example: +``` + gcc $(CFLAGS) -Iinclude/ test-dll.c -o test-dll dll\libzstd.dll +``` +The compiled executable will require ZSTD DLL which is available at `dll\libzstd.dll`. -#### Advanced Build options -The build system requires a hash function in order to -separate object files created with different compilation flags. -By default, it tries to use `md5sum` or equivalent. -The hash function can be manually switched by setting the `HASH` variable. -For example : `make HASH=xxhsum` -The hash function needs to generate at least 64-bit using hexadecimal format. -When no hash function is found, -the Makefile just generates all object files into the same default directory, -irrespective of compilation flags. -This functionality only matters if `libzstd` is compiled multiple times -with different build flags. +#### Advanced Build options -The build directory, where object files are stored -can also be manually controlled using variable `BUILD_DIR`, -for example `make BUILD_DIR=objectDir/v1`. -In which case, the hash function doesn't matter. +The build system requires a hash function in order to +separate object files created with different compilation flags. +By default, it tries to use `md5sum` or equivalent. +The hash function can be manually switched by setting the `HASH` variable. +For example : `make HASH=xxhsum` +The hash function needs to generate at least 64-bit using hexadecimal format. +When no hash function is found, +the Makefile just generates all object files into the same default directory, +irrespective of compilation flags. +This functionality only matters if `libzstd` is compiled multiple times +with different build flags. +The build directory, where object files are stored +can also be manually controlled using variable `BUILD_DIR`, +for example `make BUILD_DIR=objectDir/v1`. +In which case, the hash function doesn't matter. -#### Deprecated API -Obsolete API on their way out are stored in directory `lib/deprecated`. -At this stage, it contains older streaming prototypes, in `lib/deprecated/zbuff.h`. -These prototypes will be removed in some future version. -Consider migrating code towards supported streaming API exposed in `zstd.h`. +#### Deprecated API +Obsolete API on their way out are stored in directory `lib/deprecated`. +At this stage, it contains older streaming prototypes, in `lib/deprecated/zbuff.h`. +These prototypes will be removed in some future version. +Consider migrating code towards supported streaming API exposed in `zstd.h`. -#### Miscellaneous -The other files are not source code. There are : +#### Miscellaneous - - `BUCK` : support for `buck` build system (https://buckbuild.com/) - - `Makefile` : `make` script to build and install zstd library (static and dynamic) - - `README.md` : this file - - `dll/` : resources directory for Windows compilation - - `libzstd.pc.in` : script for `pkg-config` (used in `make install`) +The other files are not source code. There are : + + - `BUCK` : support for `buck` build system (https://buckbuild.com/) + - `Makefile` : `make` script to build and install zstd library (static and dynamic) + - `README.md` : this file + - `dll/` : resources directory for Windows compilation + - `libzstd.pc.in` : script for `pkg-config` (used in `make install`) diff --git a/contrib/libs/zstd/lib/common/bitstream.h b/contrib/libs/zstd/lib/common/bitstream.h index 84b6062ff3..888e1617d5 100644 --- a/contrib/libs/zstd/lib/common/bitstream.h +++ b/contrib/libs/zstd/lib/common/bitstream.h @@ -1,7 +1,7 @@ /* ****************************************************************** * bitstream * Part of FSE library - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * * You can contact the author at : * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy @@ -35,12 +35,12 @@ extern "C" { /*========================================= * Target specific =========================================*/ -#ifndef ZSTD_NO_INTRINSICS -# if defined(__BMI__) && defined(__GNUC__) -# include <immintrin.h> /* support for bextr (experimental) */ -# elif defined(__ICCARM__) -# include <intrinsics.h> -# endif +#ifndef ZSTD_NO_INTRINSICS +# if defined(__BMI__) && defined(__GNUC__) +# include <immintrin.h> /* support for bextr (experimental) */ +# elif defined(__ICCARM__) +# include <intrinsics.h> +# endif #endif #define STREAM_ACCUMULATOR_MIN_32 25 @@ -142,9 +142,9 @@ MEM_STATIC unsigned BIT_highbit32 (U32 val) assert(val != 0); { # if defined(_MSC_VER) /* Visual */ -# if STATIC_BMI2 == 1 +# if STATIC_BMI2 == 1 return _lzcnt_u32(val) ^ 31; -# else +# else if (val != 0) { unsigned long r; _BitScanReverse(&r, val); @@ -153,7 +153,7 @@ MEM_STATIC unsigned BIT_highbit32 (U32 val) /* Should not reach this code path */ __assume(0); } -# endif +# endif # elif defined(__GNUC__) && (__GNUC__ >= 3) /* Use GCC Intrinsic */ return __builtin_clz (val) ^ 31; # elif defined(__ICCARM__) /* IAR Intrinsic */ @@ -209,7 +209,7 @@ MEM_STATIC size_t BIT_initCStream(BIT_CStream_t* bitC, MEM_STATIC void BIT_addBits(BIT_CStream_t* bitC, size_t value, unsigned nbBits) { - DEBUG_STATIC_ASSERT(BIT_MASK_SIZE == 32); + DEBUG_STATIC_ASSERT(BIT_MASK_SIZE == 32); assert(nbBits < BIT_MASK_SIZE); assert(nbBits + bitC->bitPos < sizeof(bitC->bitContainer) * 8); bitC->bitContainer |= (value & BIT_mask[nbBits]) << bitC->bitPos; @@ -282,7 +282,7 @@ MEM_STATIC size_t BIT_closeCStream(BIT_CStream_t* bitC) */ MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, size_t srcSize) { - if (srcSize < 1) { ZSTD_memset(bitD, 0, sizeof(*bitD)); return ERROR(srcSize_wrong); } + if (srcSize < 1) { ZSTD_memset(bitD, 0, sizeof(*bitD)); return ERROR(srcSize_wrong); } bitD->start = (const char*)srcBuffer; bitD->limitPtr = bitD->start + sizeof(bitD->bitContainer); @@ -328,12 +328,12 @@ MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, si return srcSize; } -MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getUpperBits(size_t bitContainer, U32 const start) +MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getUpperBits(size_t bitContainer, U32 const start) { return bitContainer >> start; } -MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getMiddleBits(size_t bitContainer, U32 const start, U32 const nbBits) +MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getMiddleBits(size_t bitContainer, U32 const start, U32 const nbBits) { U32 const regMask = sizeof(bitContainer)*8 - 1; /* if start > regMask, bitstream is corrupted, and result is undefined */ @@ -350,14 +350,14 @@ MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getMiddleBits(size_t bitContainer, U32 c #endif } -MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits) +MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits) { -#if defined(STATIC_BMI2) && STATIC_BMI2 == 1 - return _bzhi_u64(bitContainer, nbBits); -#else +#if defined(STATIC_BMI2) && STATIC_BMI2 == 1 + return _bzhi_u64(bitContainer, nbBits); +#else assert(nbBits < BIT_MASK_SIZE); return bitContainer & BIT_mask[nbBits]; -#endif +#endif } /*! BIT_lookBits() : @@ -366,7 +366,7 @@ MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getLowerBits(size_t bitContainer, U32 co * On 32-bits, maxNbBits==24. * On 64-bits, maxNbBits==56. * @return : value extracted */ -MEM_STATIC FORCE_INLINE_ATTR size_t BIT_lookBits(const BIT_DStream_t* bitD, U32 nbBits) +MEM_STATIC FORCE_INLINE_ATTR size_t BIT_lookBits(const BIT_DStream_t* bitD, U32 nbBits) { /* arbitrate between double-shift and shift+mask */ #if 1 @@ -389,7 +389,7 @@ MEM_STATIC size_t BIT_lookBitsFast(const BIT_DStream_t* bitD, U32 nbBits) return (bitD->bitContainer << (bitD->bitsConsumed & regMask)) >> (((regMask+1)-nbBits) & regMask); } -MEM_STATIC FORCE_INLINE_ATTR void BIT_skipBits(BIT_DStream_t* bitD, U32 nbBits) +MEM_STATIC FORCE_INLINE_ATTR void BIT_skipBits(BIT_DStream_t* bitD, U32 nbBits) { bitD->bitsConsumed += nbBits; } @@ -398,7 +398,7 @@ MEM_STATIC FORCE_INLINE_ATTR void BIT_skipBits(BIT_DStream_t* bitD, U32 nbBits) * Read (consume) next n bits from local register and update. * Pay attention to not read more than nbBits contained into local register. * @return : extracted value. */ -MEM_STATIC FORCE_INLINE_ATTR size_t BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits) +MEM_STATIC FORCE_INLINE_ATTR size_t BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits) { size_t const value = BIT_lookBits(bitD, nbBits); BIT_skipBits(bitD, nbBits); diff --git a/contrib/libs/zstd/lib/common/compiler.h b/contrib/libs/zstd/lib/common/compiler.h index 516930c01e..fcf2eabdb4 100644 --- a/contrib/libs/zstd/lib/common/compiler.h +++ b/contrib/libs/zstd/lib/common/compiler.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -41,17 +41,17 @@ #endif /** - On MSVC qsort requires that functions passed into it use the __cdecl calling conversion(CC). + On MSVC qsort requires that functions passed into it use the __cdecl calling conversion(CC). This explicitly marks such functions as __cdecl so that the code will still compile - if a CC other than __cdecl has been made the default. -*/ -#if defined(_MSC_VER) -# define WIN_CDECL __cdecl -#else -# define WIN_CDECL -#endif - -/** + if a CC other than __cdecl has been made the default. +*/ +#if defined(_MSC_VER) +# define WIN_CDECL __cdecl +#else +# define WIN_CDECL +#endif + +/** * FORCE_INLINE_TEMPLATE is used to define C "templates", which take constant * parameters. They must be inlined for the compiler to eliminate the constant * branches. @@ -92,7 +92,7 @@ # endif #endif - + /* target attribute */ #if defined(__GNUC__) || defined(__ICCARM__) # define TARGET_ATTRIBUTE(target) __attribute__((__target__(target))) @@ -119,9 +119,9 @@ # elif defined(__GNUC__) && ( (__GNUC__ >= 4) || ( (__GNUC__ == 3) && (__GNUC_MINOR__ >= 1) ) ) # define PREFETCH_L1(ptr) __builtin_prefetch((ptr), 0 /* rw==read */, 3 /* locality */) # define PREFETCH_L2(ptr) __builtin_prefetch((ptr), 0 /* rw==read */, 2 /* locality */) -# elif defined(__aarch64__) -# define PREFETCH_L1(ptr) __asm__ __volatile__("prfm pldl1keep, %0" ::"Q"(*(ptr))) -# define PREFETCH_L2(ptr) __asm__ __volatile__("prfm pldl2keep, %0" ::"Q"(*(ptr))) +# elif defined(__aarch64__) +# define PREFETCH_L1(ptr) __asm__ __volatile__("prfm pldl1keep, %0" ::"Q"(*(ptr))) +# define PREFETCH_L2(ptr) __asm__ __volatile__("prfm pldl2keep, %0" ::"Q"(*(ptr))) # else # define PREFETCH_L1(ptr) (void)(ptr) /* disabled */ # define PREFETCH_L2(ptr) (void)(ptr) /* disabled */ @@ -175,19 +175,19 @@ # pragma warning(disable : 4324) /* disable: C4324: padded structure */ #endif -/*Like DYNAMIC_BMI2 but for compile time determination of BMI2 support*/ -#ifndef STATIC_BMI2 -# if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86)) -# ifdef __AVX2__ //MSVC does not have a BMI2 specific flag, but every CPU that supports AVX2 also supports BMI2 -# define STATIC_BMI2 1 -# endif -# endif -#endif - -#ifndef STATIC_BMI2 - #define STATIC_BMI2 0 -#endif - +/*Like DYNAMIC_BMI2 but for compile time determination of BMI2 support*/ +#ifndef STATIC_BMI2 +# if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86)) +# ifdef __AVX2__ //MSVC does not have a BMI2 specific flag, but every CPU that supports AVX2 also supports BMI2 +# define STATIC_BMI2 1 +# endif +# endif +#endif + +#ifndef STATIC_BMI2 + #define STATIC_BMI2 0 +#endif + /* compile time determination of SIMD support */ #if !defined(ZSTD_NO_INTRINSICS) # if defined(__SSE2__) || defined(_M_AMD64) || (defined (_M_IX86) && defined(_M_IX86_FP) && (_M_IX86_FP >= 2)) @@ -202,15 +202,15 @@ # elif defined(ZSTD_ARCH_ARM_NEON) # include <arm_neon.h> # endif -#endif - +#endif + /* C-language Attributes are added in C23. */ #if defined(__STDC_VERSION__) && (__STDC_VERSION__ > 201710L) && defined(__has_c_attribute) # define ZSTD_HAS_C_ATTRIBUTE(x) __has_c_attribute(x) #else # define ZSTD_HAS_C_ATTRIBUTE(x) 0 -#endif - +#endif + /* Only use C++ attributes in C++. Some compilers report support for C++ * attributes when compiling with C. */ @@ -218,8 +218,8 @@ # define ZSTD_HAS_CPP_ATTRIBUTE(x) __has_cpp_attribute(x) #else # define ZSTD_HAS_CPP_ATTRIBUTE(x) 0 -#endif - +#endif + /* Define ZSTD_FALLTHROUGH macro for annotating switch case with the 'fallthrough' attribute. * - C23: https://en.cppreference.com/w/c/language/attributes/fallthrough * - CPP17: https://en.cppreference.com/w/cpp/language/attributes/fallthrough @@ -273,63 +273,63 @@ * Sanitizer *****************************************************************/ -#if ZSTD_MEMORY_SANITIZER -/* Not all platforms that support msan provide sanitizers/msan_interface.h. - * We therefore declare the functions we need ourselves, rather than trying to - * include the header file... */ -#include <stddef.h> /* size_t */ -#define ZSTD_DEPS_NEED_STDINT -#include "zstd_deps.h" /* intptr_t */ - -/* Make memory region fully initialized (without changing its contents). */ -void __msan_unpoison(const volatile void *a, size_t size); - -/* Make memory region fully uninitialized (without changing its contents). - This is a legacy interface that does not update origin information. Use - __msan_allocated_memory() instead. */ -void __msan_poison(const volatile void *a, size_t size); - -/* Returns the offset of the first (at least partially) poisoned byte in the - memory range, or -1 if the whole range is good. */ -intptr_t __msan_test_shadow(const volatile void *x, size_t size); -#endif - -#if ZSTD_ADDRESS_SANITIZER -/* Not all platforms that support asan provide sanitizers/asan_interface.h. - * We therefore declare the functions we need ourselves, rather than trying to - * include the header file... */ -#include <stddef.h> /* size_t */ - -/** - * Marks a memory region (<c>[addr, addr+size)</c>) as unaddressable. - * - * This memory must be previously allocated by your program. Instrumented - * code is forbidden from accessing addresses in this region until it is - * unpoisoned. This function is not guaranteed to poison the entire region - - * it could poison only a subregion of <c>[addr, addr+size)</c> due to ASan - * alignment restrictions. - * - * \note This function is not thread-safe because no two threads can poison or - * unpoison memory in the same memory region simultaneously. - * - * \param addr Start of memory region. - * \param size Size of memory region. */ -void __asan_poison_memory_region(void const volatile *addr, size_t size); - -/** - * Marks a memory region (<c>[addr, addr+size)</c>) as addressable. - * - * This memory must be previously allocated by your program. Accessing - * addresses in this region is allowed until this region is poisoned again. - * This function could unpoison a super-region of <c>[addr, addr+size)</c> due - * to ASan alignment restrictions. - * - * \note This function is not thread-safe because no two threads can - * poison or unpoison memory in the same memory region simultaneously. - * - * \param addr Start of memory region. - * \param size Size of memory region. */ -void __asan_unpoison_memory_region(void const volatile *addr, size_t size); -#endif - +#if ZSTD_MEMORY_SANITIZER +/* Not all platforms that support msan provide sanitizers/msan_interface.h. + * We therefore declare the functions we need ourselves, rather than trying to + * include the header file... */ +#include <stddef.h> /* size_t */ +#define ZSTD_DEPS_NEED_STDINT +#include "zstd_deps.h" /* intptr_t */ + +/* Make memory region fully initialized (without changing its contents). */ +void __msan_unpoison(const volatile void *a, size_t size); + +/* Make memory region fully uninitialized (without changing its contents). + This is a legacy interface that does not update origin information. Use + __msan_allocated_memory() instead. */ +void __msan_poison(const volatile void *a, size_t size); + +/* Returns the offset of the first (at least partially) poisoned byte in the + memory range, or -1 if the whole range is good. */ +intptr_t __msan_test_shadow(const volatile void *x, size_t size); +#endif + +#if ZSTD_ADDRESS_SANITIZER +/* Not all platforms that support asan provide sanitizers/asan_interface.h. + * We therefore declare the functions we need ourselves, rather than trying to + * include the header file... */ +#include <stddef.h> /* size_t */ + +/** + * Marks a memory region (<c>[addr, addr+size)</c>) as unaddressable. + * + * This memory must be previously allocated by your program. Instrumented + * code is forbidden from accessing addresses in this region until it is + * unpoisoned. This function is not guaranteed to poison the entire region - + * it could poison only a subregion of <c>[addr, addr+size)</c> due to ASan + * alignment restrictions. + * + * \note This function is not thread-safe because no two threads can poison or + * unpoison memory in the same memory region simultaneously. + * + * \param addr Start of memory region. + * \param size Size of memory region. */ +void __asan_poison_memory_region(void const volatile *addr, size_t size); + +/** + * Marks a memory region (<c>[addr, addr+size)</c>) as addressable. + * + * This memory must be previously allocated by your program. Accessing + * addresses in this region is allowed until this region is poisoned again. + * This function could unpoison a super-region of <c>[addr, addr+size)</c> due + * to ASan alignment restrictions. + * + * \note This function is not thread-safe because no two threads can + * poison or unpoison memory in the same memory region simultaneously. + * + * \param addr Start of memory region. + * \param size Size of memory region. */ +void __asan_unpoison_memory_region(void const volatile *addr, size_t size); +#endif + #endif /* ZSTD_COMPILER_H */ diff --git a/contrib/libs/zstd/lib/common/cpu.h b/contrib/libs/zstd/lib/common/cpu.h index 8acd33be3c..1e669e8a59 100644 --- a/contrib/libs/zstd/lib/common/cpu.h +++ b/contrib/libs/zstd/lib/common/cpu.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Facebook, Inc. + * Copyright (c) Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the diff --git a/contrib/libs/zstd/lib/common/debug.c b/contrib/libs/zstd/lib/common/debug.c index bb863c9ea6..327651705a 100644 --- a/contrib/libs/zstd/lib/common/debug.c +++ b/contrib/libs/zstd/lib/common/debug.c @@ -1,7 +1,7 @@ /* ****************************************************************** * debug * Part of FSE library - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * * You can contact the author at : * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy diff --git a/contrib/libs/zstd/lib/common/debug.h b/contrib/libs/zstd/lib/common/debug.h index 3b2a320a18..c15ec33eda 100644 --- a/contrib/libs/zstd/lib/common/debug.h +++ b/contrib/libs/zstd/lib/common/debug.h @@ -1,7 +1,7 @@ /* ****************************************************************** * debug * Part of FSE library - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * * You can contact the author at : * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy @@ -67,8 +67,8 @@ extern "C" { */ #if (DEBUGLEVEL>=1) -# define ZSTD_DEPS_NEED_ASSERT -# include "zstd_deps.h" +# define ZSTD_DEPS_NEED_ASSERT +# include "zstd_deps.h" #else # ifndef assert /* assert may be already defined, due to prior #include <assert.h> */ # define assert(condition) ((void)0) /* disable assert (default) */ @@ -76,8 +76,8 @@ extern "C" { #endif #if (DEBUGLEVEL>=2) -# define ZSTD_DEPS_NEED_IO -# include "zstd_deps.h" +# define ZSTD_DEPS_NEED_IO +# include "zstd_deps.h" extern int g_debuglevel; /* the variable is only declared, it actually lives in debug.c, and is shared by the whole process. @@ -85,14 +85,14 @@ extern int g_debuglevel; /* the variable is only declared, It's useful when enabling very verbose levels on selective conditions (such as position in src) */ -# define RAWLOG(l, ...) { \ - if (l<=g_debuglevel) { \ - ZSTD_DEBUG_PRINT(__VA_ARGS__); \ +# define RAWLOG(l, ...) { \ + if (l<=g_debuglevel) { \ + ZSTD_DEBUG_PRINT(__VA_ARGS__); \ } } -# define DEBUGLOG(l, ...) { \ - if (l<=g_debuglevel) { \ - ZSTD_DEBUG_PRINT(__FILE__ ": " __VA_ARGS__); \ - ZSTD_DEBUG_PRINT(" \n"); \ +# define DEBUGLOG(l, ...) { \ + if (l<=g_debuglevel) { \ + ZSTD_DEBUG_PRINT(__FILE__ ": " __VA_ARGS__); \ + ZSTD_DEBUG_PRINT(" \n"); \ } } #else # define RAWLOG(l, ...) {} /* disabled */ diff --git a/contrib/libs/zstd/lib/common/entropy_common.c b/contrib/libs/zstd/lib/common/entropy_common.c index 4229b40c5e..8be420e331 100644 --- a/contrib/libs/zstd/lib/common/entropy_common.c +++ b/contrib/libs/zstd/lib/common/entropy_common.c @@ -1,6 +1,6 @@ /* ****************************************************************** * Common functions of New Generation Entropy library - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * * You can contact the author at : * - FSE+HUF source repository : https://github.com/Cyan4973/FiniteStateEntropy @@ -38,11 +38,11 @@ const char* HUF_getErrorName(size_t code) { return ERR_getErrorName(code); } /*-************************************************************** * FSE NCount encoding-decoding ****************************************************************/ -static U32 FSE_ctz(U32 val) +static U32 FSE_ctz(U32 val) { - assert(val != 0); - { -# if defined(_MSC_VER) /* Visual */ + assert(val != 0); + { +# if defined(_MSC_VER) /* Visual */ if (val != 0) { unsigned long r; _BitScanForward(&r, val); @@ -51,25 +51,25 @@ static U32 FSE_ctz(U32 val) /* Should not reach this code path */ __assume(0); } -# elif defined(__GNUC__) && (__GNUC__ >= 3) /* GCC Intrinsic */ - return __builtin_ctz(val); -# elif defined(__ICCARM__) /* IAR Intrinsic */ - return __CTZ(val); -# else /* Software version */ - U32 count = 0; - while ((val & 1) == 0) { - val >>= 1; - ++count; - } - return count; -# endif - } -} - -FORCE_INLINE_TEMPLATE -size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr, - const void* headerBuffer, size_t hbSize) -{ +# elif defined(__GNUC__) && (__GNUC__ >= 3) /* GCC Intrinsic */ + return __builtin_ctz(val); +# elif defined(__ICCARM__) /* IAR Intrinsic */ + return __CTZ(val); +# else /* Software version */ + U32 count = 0; + while ((val & 1) == 0) { + val >>= 1; + ++count; + } + return count; +# endif + } +} + +FORCE_INLINE_TEMPLATE +size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr, + const void* headerBuffer, size_t hbSize) +{ const BYTE* const istart = (const BYTE*) headerBuffer; const BYTE* const iend = istart + hbSize; const BYTE* ip = istart; @@ -79,23 +79,23 @@ size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigne U32 bitStream; int bitCount; unsigned charnum = 0; - unsigned const maxSV1 = *maxSVPtr + 1; + unsigned const maxSV1 = *maxSVPtr + 1; int previous0 = 0; - if (hbSize < 8) { - /* This function only works when hbSize >= 8 */ - char buffer[8] = {0}; - ZSTD_memcpy(buffer, headerBuffer, hbSize); + if (hbSize < 8) { + /* This function only works when hbSize >= 8 */ + char buffer[8] = {0}; + ZSTD_memcpy(buffer, headerBuffer, hbSize); { size_t const countSize = FSE_readNCount(normalizedCounter, maxSVPtr, tableLogPtr, buffer, sizeof(buffer)); if (FSE_isError(countSize)) return countSize; if (countSize > hbSize) return ERROR(corruption_detected); return countSize; } } - assert(hbSize >= 8); + assert(hbSize >= 8); /* init */ - ZSTD_memset(normalizedCounter, 0, (*maxSVPtr+1) * sizeof(normalizedCounter[0])); /* all symbols not present in NCount have a frequency of 0 */ + ZSTD_memset(normalizedCounter, 0, (*maxSVPtr+1) * sizeof(normalizedCounter[0])); /* all symbols not present in NCount have a frequency of 0 */ bitStream = MEM_readLE32(ip); nbBits = (bitStream & 0xF) + FSE_MIN_TABLELOG; /* extract tableLog */ if (nbBits > FSE_TABLELOG_ABSOLUTE_MAX) return ERROR(tableLog_tooLarge); @@ -106,58 +106,58 @@ size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigne threshold = 1<<nbBits; nbBits++; - for (;;) { + for (;;) { if (previous0) { - /* Count the number of repeats. Each time the - * 2-bit repeat code is 0b11 there is another - * repeat. - * Avoid UB by setting the high bit to 1. - */ - int repeats = FSE_ctz(~bitStream | 0x80000000) >> 1; - while (repeats >= 12) { - charnum += 3 * 12; - if (LIKELY(ip <= iend-7)) { - ip += 3; + /* Count the number of repeats. Each time the + * 2-bit repeat code is 0b11 there is another + * repeat. + * Avoid UB by setting the high bit to 1. + */ + int repeats = FSE_ctz(~bitStream | 0x80000000) >> 1; + while (repeats >= 12) { + charnum += 3 * 12; + if (LIKELY(ip <= iend-7)) { + ip += 3; } else { - bitCount -= (int)(8 * (iend - 7 - ip)); - bitCount &= 31; - ip = iend - 4; - } - bitStream = MEM_readLE32(ip) >> bitCount; - repeats = FSE_ctz(~bitStream | 0x80000000) >> 1; + bitCount -= (int)(8 * (iend - 7 - ip)); + bitCount &= 31; + ip = iend - 4; + } + bitStream = MEM_readLE32(ip) >> bitCount; + repeats = FSE_ctz(~bitStream | 0x80000000) >> 1; } - charnum += 3 * repeats; - bitStream >>= 2 * repeats; - bitCount += 2 * repeats; - - /* Add the final repeat which isn't 0b11. */ - assert((bitStream & 3) < 3); - charnum += bitStream & 3; + charnum += 3 * repeats; + bitStream >>= 2 * repeats; + bitCount += 2 * repeats; + + /* Add the final repeat which isn't 0b11. */ + assert((bitStream & 3) < 3); + charnum += bitStream & 3; bitCount += 2; - - /* This is an error, but break and return an error - * at the end, because returning out of a loop makes - * it harder for the compiler to optimize. - */ - if (charnum >= maxSV1) break; - - /* We don't need to set the normalized count to 0 - * because we already memset the whole buffer to 0. - */ - - if (LIKELY(ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) { + + /* This is an error, but break and return an error + * at the end, because returning out of a loop makes + * it harder for the compiler to optimize. + */ + if (charnum >= maxSV1) break; + + /* We don't need to set the normalized count to 0 + * because we already memset the whole buffer to 0. + */ + + if (LIKELY(ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) { assert((bitCount >> 3) <= 3); /* For first condition to work */ ip += bitCount>>3; bitCount &= 7; } else { - bitCount -= (int)(8 * (iend - 4 - ip)); - bitCount &= 31; - ip = iend - 4; - } - bitStream = MEM_readLE32(ip) >> bitCount; - } - { - int const max = (2*threshold-1) - remaining; + bitCount -= (int)(8 * (iend - 4 - ip)); + bitCount &= 31; + ip = iend - 4; + } + bitStream = MEM_readLE32(ip) >> bitCount; + } + { + int const max = (2*threshold-1) - remaining; int count; if ((bitStream & (threshold-1)) < (U32)max) { @@ -170,43 +170,43 @@ size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigne } count--; /* extra accuracy */ - /* When it matters (small blocks), this is a - * predictable branch, because we don't use -1. - */ - if (count >= 0) { - remaining -= count; - } else { - assert(count == -1); - remaining += count; - } + /* When it matters (small blocks), this is a + * predictable branch, because we don't use -1. + */ + if (count >= 0) { + remaining -= count; + } else { + assert(count == -1); + remaining += count; + } normalizedCounter[charnum++] = (short)count; previous0 = !count; - - assert(threshold > 1); - if (remaining < threshold) { - /* This branch can be folded into the - * threshold update condition because we - * know that threshold > 1. - */ - if (remaining <= 1) break; - nbBits = BIT_highbit32(remaining) + 1; - threshold = 1 << (nbBits - 1); + + assert(threshold > 1); + if (remaining < threshold) { + /* This branch can be folded into the + * threshold update condition because we + * know that threshold > 1. + */ + if (remaining <= 1) break; + nbBits = BIT_highbit32(remaining) + 1; + threshold = 1 << (nbBits - 1); } - if (charnum >= maxSV1) break; + if (charnum >= maxSV1) break; - if (LIKELY(ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) { + if (LIKELY(ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) { ip += bitCount>>3; bitCount &= 7; } else { bitCount -= (int)(8 * (iend - 4 - ip)); - bitCount &= 31; + bitCount &= 31; ip = iend - 4; } - bitStream = MEM_readLE32(ip) >> bitCount; - } } + bitStream = MEM_readLE32(ip) >> bitCount; + } } if (remaining != 1) return ERROR(corruption_detected); - /* Only possible when there are too many zeros. */ - if (charnum > maxSV1) return ERROR(maxSymbolValue_tooSmall); + /* Only possible when there are too many zeros. */ + if (charnum > maxSV1) return ERROR(maxSymbolValue_tooSmall); if (bitCount > 32) return ERROR(corruption_detected); *maxSVPtr = charnum-1; @@ -214,44 +214,44 @@ size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigne return ip-istart; } -/* Avoids the FORCE_INLINE of the _body() function. */ -static size_t FSE_readNCount_body_default( - short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr, - const void* headerBuffer, size_t hbSize) -{ - return FSE_readNCount_body(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize); -} +/* Avoids the FORCE_INLINE of the _body() function. */ +static size_t FSE_readNCount_body_default( + short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr, + const void* headerBuffer, size_t hbSize) +{ + return FSE_readNCount_body(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize); +} -#if DYNAMIC_BMI2 +#if DYNAMIC_BMI2 BMI2_TARGET_ATTRIBUTE static size_t FSE_readNCount_body_bmi2( - short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr, - const void* headerBuffer, size_t hbSize) -{ - return FSE_readNCount_body(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize); -} -#endif - -size_t FSE_readNCount_bmi2( - short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr, - const void* headerBuffer, size_t hbSize, int bmi2) -{ -#if DYNAMIC_BMI2 - if (bmi2) { - return FSE_readNCount_body_bmi2(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize); - } -#endif - (void)bmi2; - return FSE_readNCount_body_default(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize); -} - -size_t FSE_readNCount( - short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr, - const void* headerBuffer, size_t hbSize) -{ - return FSE_readNCount_bmi2(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize, /* bmi2 */ 0); -} - - + short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr, + const void* headerBuffer, size_t hbSize) +{ + return FSE_readNCount_body(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize); +} +#endif + +size_t FSE_readNCount_bmi2( + short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr, + const void* headerBuffer, size_t hbSize, int bmi2) +{ +#if DYNAMIC_BMI2 + if (bmi2) { + return FSE_readNCount_body_bmi2(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize); + } +#endif + (void)bmi2; + return FSE_readNCount_body_default(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize); +} + +size_t FSE_readNCount( + short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr, + const void* headerBuffer, size_t hbSize) +{ + return FSE_readNCount_bmi2(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize, /* bmi2 */ 0); +} + + /*! HUF_readStats() : Read compact Huffman tree, saved by HUF_writeCTable(). `huffWeight` is destination buffer. @@ -263,17 +263,17 @@ size_t HUF_readStats(BYTE* huffWeight, size_t hwSize, U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr, const void* src, size_t srcSize) { - U32 wksp[HUF_READ_STATS_WORKSPACE_SIZE_U32]; - return HUF_readStats_wksp(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, wksp, sizeof(wksp), /* bmi2 */ 0); -} - -FORCE_INLINE_TEMPLATE size_t -HUF_readStats_body(BYTE* huffWeight, size_t hwSize, U32* rankStats, - U32* nbSymbolsPtr, U32* tableLogPtr, - const void* src, size_t srcSize, - void* workSpace, size_t wkspSize, - int bmi2) -{ + U32 wksp[HUF_READ_STATS_WORKSPACE_SIZE_U32]; + return HUF_readStats_wksp(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, wksp, sizeof(wksp), /* bmi2 */ 0); +} + +FORCE_INLINE_TEMPLATE size_t +HUF_readStats_body(BYTE* huffWeight, size_t hwSize, U32* rankStats, + U32* nbSymbolsPtr, U32* tableLogPtr, + const void* src, size_t srcSize, + void* workSpace, size_t wkspSize, + int bmi2) +{ U32 weightTotal; const BYTE* ip = (const BYTE*) src; size_t iSize; @@ -281,7 +281,7 @@ HUF_readStats_body(BYTE* huffWeight, size_t hwSize, U32* rankStats, if (!srcSize) return ERROR(srcSize_wrong); iSize = ip[0]; - /* ZSTD_memset(huffWeight, 0, hwSize); *//* is not necessary, even though some analyzer complain ... */ + /* ZSTD_memset(huffWeight, 0, hwSize); *//* is not necessary, even though some analyzer complain ... */ if (iSize >= 128) { /* special header */ oSize = iSize - 127; @@ -296,13 +296,13 @@ HUF_readStats_body(BYTE* huffWeight, size_t hwSize, U32* rankStats, } } } else { /* header compressed with FSE (normal case) */ if (iSize+1 > srcSize) return ERROR(srcSize_wrong); - /* max (hwSize-1) values decoded, as last one is implied */ - oSize = FSE_decompress_wksp_bmi2(huffWeight, hwSize-1, ip+1, iSize, 6, workSpace, wkspSize, bmi2); + /* max (hwSize-1) values decoded, as last one is implied */ + oSize = FSE_decompress_wksp_bmi2(huffWeight, hwSize-1, ip+1, iSize, 6, workSpace, wkspSize, bmi2); if (FSE_isError(oSize)) return oSize; } /* collect weight stats */ - ZSTD_memset(rankStats, 0, (HUF_TABLELOG_MAX + 1) * sizeof(U32)); + ZSTD_memset(rankStats, 0, (HUF_TABLELOG_MAX + 1) * sizeof(U32)); weightTotal = 0; { U32 n; for (n=0; n<oSize; n++) { if (huffWeight[n] > HUF_TABLELOG_MAX) return ERROR(corruption_detected); @@ -332,37 +332,37 @@ HUF_readStats_body(BYTE* huffWeight, size_t hwSize, U32* rankStats, *nbSymbolsPtr = (U32)(oSize+1); return iSize+1; } - -/* Avoids the FORCE_INLINE of the _body() function. */ -static size_t HUF_readStats_body_default(BYTE* huffWeight, size_t hwSize, U32* rankStats, - U32* nbSymbolsPtr, U32* tableLogPtr, - const void* src, size_t srcSize, - void* workSpace, size_t wkspSize) -{ - return HUF_readStats_body(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize, 0); -} - -#if DYNAMIC_BMI2 + +/* Avoids the FORCE_INLINE of the _body() function. */ +static size_t HUF_readStats_body_default(BYTE* huffWeight, size_t hwSize, U32* rankStats, + U32* nbSymbolsPtr, U32* tableLogPtr, + const void* src, size_t srcSize, + void* workSpace, size_t wkspSize) +{ + return HUF_readStats_body(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize, 0); +} + +#if DYNAMIC_BMI2 static BMI2_TARGET_ATTRIBUTE size_t HUF_readStats_body_bmi2(BYTE* huffWeight, size_t hwSize, U32* rankStats, - U32* nbSymbolsPtr, U32* tableLogPtr, - const void* src, size_t srcSize, - void* workSpace, size_t wkspSize) -{ - return HUF_readStats_body(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize, 1); -} -#endif - -size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize, U32* rankStats, - U32* nbSymbolsPtr, U32* tableLogPtr, - const void* src, size_t srcSize, - void* workSpace, size_t wkspSize, - int bmi2) -{ -#if DYNAMIC_BMI2 - if (bmi2) { - return HUF_readStats_body_bmi2(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize); - } -#endif - (void)bmi2; - return HUF_readStats_body_default(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize); -} + U32* nbSymbolsPtr, U32* tableLogPtr, + const void* src, size_t srcSize, + void* workSpace, size_t wkspSize) +{ + return HUF_readStats_body(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize, 1); +} +#endif + +size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize, U32* rankStats, + U32* nbSymbolsPtr, U32* tableLogPtr, + const void* src, size_t srcSize, + void* workSpace, size_t wkspSize, + int bmi2) +{ +#if DYNAMIC_BMI2 + if (bmi2) { + return HUF_readStats_body_bmi2(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize); + } +#endif + (void)bmi2; + return HUF_readStats_body_default(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize); +} diff --git a/contrib/libs/zstd/lib/common/error_private.c b/contrib/libs/zstd/lib/common/error_private.c index 6d1135f8c3..02c74eb55b 100644 --- a/contrib/libs/zstd/lib/common/error_private.c +++ b/contrib/libs/zstd/lib/common/error_private.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -48,7 +48,7 @@ const char* ERR_getErrorString(ERR_enum code) case PREFIX(frameIndex_tooLarge): return "Frame index is too large"; case PREFIX(seekableIO): return "An I/O error occurred when reading/seeking"; case PREFIX(dstBuffer_wrong): return "Destination buffer is wrong"; - case PREFIX(srcBuffer_wrong): return "Source buffer is wrong"; + case PREFIX(srcBuffer_wrong): return "Source buffer is wrong"; case PREFIX(maxCode): default: return notErrorCode; } diff --git a/contrib/libs/zstd/lib/common/error_private.h b/contrib/libs/zstd/lib/common/error_private.h index 007d81066a..ac9cdaa475 100644 --- a/contrib/libs/zstd/lib/common/error_private.h +++ b/contrib/libs/zstd/lib/common/error_private.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -21,10 +21,10 @@ extern "C" { /* **************************************** * Dependencies ******************************************/ -#include "../zstd_errors.h" /* enum list */ +#include "../zstd_errors.h" /* enum list */ #include "compiler.h" #include "debug.h" -#include "zstd_deps.h" /* size_t */ +#include "zstd_deps.h" /* size_t */ /* **************************************** diff --git a/contrib/libs/zstd/lib/common/fse.h b/contrib/libs/zstd/lib/common/fse.h index 714bfd3e7f..ecbba3a5ae 100644 --- a/contrib/libs/zstd/lib/common/fse.h +++ b/contrib/libs/zstd/lib/common/fse.h @@ -1,7 +1,7 @@ /* ****************************************************************** * FSE : Finite State Entropy codec * Public Prototypes declaration - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * * You can contact the author at : * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy @@ -23,7 +23,7 @@ extern "C" { /*-***************************************** * Dependencies ******************************************/ -#include "zstd_deps.h" /* size_t, ptrdiff_t */ +#include "zstd_deps.h" /* size_t, ptrdiff_t */ /*-***************************************** @@ -137,16 +137,16 @@ FSE_PUBLIC_API unsigned FSE_optimalTableLog(unsigned maxTableLog, size_t srcSize /*! FSE_normalizeCount(): normalize counts so that sum(count[]) == Power_of_2 (2^tableLog) 'normalizedCounter' is a table of short, of minimum size (maxSymbolValue+1). - useLowProbCount is a boolean parameter which trades off compressed size for - faster header decoding. When it is set to 1, the compressed data will be slightly - smaller. And when it is set to 0, FSE_readNCount() and FSE_buildDTable() will be - faster. If you are compressing a small amount of data (< 2 KB) then useLowProbCount=0 - is a good default, since header deserialization makes a big speed difference. - Otherwise, useLowProbCount=1 is a good default, since the speed difference is small. + useLowProbCount is a boolean parameter which trades off compressed size for + faster header decoding. When it is set to 1, the compressed data will be slightly + smaller. And when it is set to 0, FSE_readNCount() and FSE_buildDTable() will be + faster. If you are compressing a small amount of data (< 2 KB) then useLowProbCount=0 + is a good default, since header deserialization makes a big speed difference. + Otherwise, useLowProbCount=1 is a good default, since the speed difference is small. @return : tableLog, or an errorCode, which can be tested using FSE_isError() */ FSE_PUBLIC_API size_t FSE_normalizeCount(short* normalizedCounter, unsigned tableLog, - const unsigned* count, size_t srcSize, unsigned maxSymbolValue, unsigned useLowProbCount); + const unsigned* count, size_t srcSize, unsigned maxSymbolValue, unsigned useLowProbCount); /*! FSE_NCountWriteBound(): Provides the maximum possible size of an FSE normalized table, given 'maxSymbolValue' and 'tableLog'. @@ -234,13 +234,13 @@ FSE_PUBLIC_API size_t FSE_readNCount (short* normalizedCounter, unsigned* maxSymbolValuePtr, unsigned* tableLogPtr, const void* rBuffer, size_t rBuffSize); -/*! FSE_readNCount_bmi2(): - * Same as FSE_readNCount() but pass bmi2=1 when your CPU supports BMI2 and 0 otherwise. - */ -FSE_PUBLIC_API size_t FSE_readNCount_bmi2(short* normalizedCounter, - unsigned* maxSymbolValuePtr, unsigned* tableLogPtr, - const void* rBuffer, size_t rBuffSize, int bmi2); - +/*! FSE_readNCount_bmi2(): + * Same as FSE_readNCount() but pass bmi2=1 when your CPU supports BMI2 and 0 otherwise. + */ +FSE_PUBLIC_API size_t FSE_readNCount_bmi2(short* normalizedCounter, + unsigned* maxSymbolValuePtr, unsigned* tableLogPtr, + const void* rBuffer, size_t rBuffSize, int bmi2); + /*! Constructor and Destructor of FSE_DTable. Note that its size depends on 'tableLog' */ typedef unsigned FSE_DTable; /* don't allocate that. It's just a way to be more restrictive than void* */ @@ -301,12 +301,12 @@ If there is an error, the function will return an error code, which can be teste *******************************************/ /* FSE buffer bounds */ #define FSE_NCOUNTBOUND 512 -#define FSE_BLOCKBOUND(size) ((size) + ((size)>>7) + 4 /* fse states */ + sizeof(size_t) /* bitContainer */) +#define FSE_BLOCKBOUND(size) ((size) + ((size)>>7) + 4 /* fse states */ + sizeof(size_t) /* bitContainer */) #define FSE_COMPRESSBOUND(size) (FSE_NCOUNTBOUND + FSE_BLOCKBOUND(size)) /* Macro version, useful for static allocation */ /* It is possible to statically allocate FSE CTable/DTable as a table of FSE_CTable/FSE_DTable using below macros */ -#define FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) (1 + (1<<((maxTableLog)-1)) + (((maxSymbolValue)+1)*2)) -#define FSE_DTABLE_SIZE_U32(maxTableLog) (1 + (1<<(maxTableLog))) +#define FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) (1 + (1<<((maxTableLog)-1)) + (((maxSymbolValue)+1)*2)) +#define FSE_DTABLE_SIZE_U32(maxTableLog) (1 + (1<<(maxTableLog))) /* or use the size to malloc() space directly. Pay attention to alignment restrictions though */ #define FSE_CTABLE_SIZE(maxTableLog, maxSymbolValue) (FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) * sizeof(FSE_CTable)) @@ -322,9 +322,9 @@ unsigned FSE_optimalTableLog_internal(unsigned maxTableLog, size_t srcSize, unsi /* FSE_compress_wksp() : * Same as FSE_compress2(), but using an externally allocated scratch buffer (`workSpace`). - * FSE_COMPRESS_WKSP_SIZE_U32() provides the minimum size required for `workSpace` as a table of FSE_CTable. + * FSE_COMPRESS_WKSP_SIZE_U32() provides the minimum size required for `workSpace` as a table of FSE_CTable. */ -#define FSE_COMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) ( FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) + ((maxTableLog > 12) ? (1 << (maxTableLog - 2)) : 1024) ) +#define FSE_COMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) ( FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) + ((maxTableLog > 12) ? (1 << (maxTableLog - 2)) : 1024) ) size_t FSE_compress_wksp (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize); size_t FSE_buildCTable_raw (FSE_CTable* ct, unsigned nbBits); @@ -335,32 +335,32 @@ size_t FSE_buildCTable_rle (FSE_CTable* ct, unsigned char symbolValue); /* FSE_buildCTable_wksp() : * Same as FSE_buildCTable(), but using an externally allocated scratch buffer (`workSpace`). - * `wkspSize` must be >= `FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog)` of `unsigned`. + * `wkspSize` must be >= `FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog)` of `unsigned`. * See FSE_buildCTable_wksp() for breakdown of workspace usage. */ #define FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog) (((maxSymbolValue + 2) + (1ull << (tableLog)))/2 + sizeof(U64)/sizeof(U32) /* additional 8 bytes for potential table overwrite */) -#define FSE_BUILD_CTABLE_WORKSPACE_SIZE(maxSymbolValue, tableLog) (sizeof(unsigned) * FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog)) +#define FSE_BUILD_CTABLE_WORKSPACE_SIZE(maxSymbolValue, tableLog) (sizeof(unsigned) * FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog)) size_t FSE_buildCTable_wksp(FSE_CTable* ct, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize); -#define FSE_BUILD_DTABLE_WKSP_SIZE(maxTableLog, maxSymbolValue) (sizeof(short) * (maxSymbolValue + 1) + (1ULL << maxTableLog) + 8) -#define FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) ((FSE_BUILD_DTABLE_WKSP_SIZE(maxTableLog, maxSymbolValue) + sizeof(unsigned) - 1) / sizeof(unsigned)) -FSE_PUBLIC_API size_t FSE_buildDTable_wksp(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize); -/**< Same as FSE_buildDTable(), using an externally allocated `workspace` produced with `FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxSymbolValue)` */ - +#define FSE_BUILD_DTABLE_WKSP_SIZE(maxTableLog, maxSymbolValue) (sizeof(short) * (maxSymbolValue + 1) + (1ULL << maxTableLog) + 8) +#define FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) ((FSE_BUILD_DTABLE_WKSP_SIZE(maxTableLog, maxSymbolValue) + sizeof(unsigned) - 1) / sizeof(unsigned)) +FSE_PUBLIC_API size_t FSE_buildDTable_wksp(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize); +/**< Same as FSE_buildDTable(), using an externally allocated `workspace` produced with `FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxSymbolValue)` */ + size_t FSE_buildDTable_raw (FSE_DTable* dt, unsigned nbBits); /**< build a fake FSE_DTable, designed to read a flat distribution where each symbol uses nbBits */ size_t FSE_buildDTable_rle (FSE_DTable* dt, unsigned char symbolValue); /**< build a fake FSE_DTable, designed to always generate the same symbolValue */ -#define FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) (FSE_DTABLE_SIZE_U32(maxTableLog) + FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) + (FSE_MAX_SYMBOL_VALUE + 1) / 2 + 1) -#define FSE_DECOMPRESS_WKSP_SIZE(maxTableLog, maxSymbolValue) (FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) * sizeof(unsigned)) -size_t FSE_decompress_wksp(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize); -/**< same as FSE_decompress(), using an externally allocated `workSpace` produced with `FSE_DECOMPRESS_WKSP_SIZE_U32(maxLog, maxSymbolValue)` */ - -size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize, int bmi2); -/**< Same as FSE_decompress_wksp() but with dynamic BMI2 support. Pass 1 if your CPU supports BMI2 or 0 if it doesn't. */ +#define FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) (FSE_DTABLE_SIZE_U32(maxTableLog) + FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) + (FSE_MAX_SYMBOL_VALUE + 1) / 2 + 1) +#define FSE_DECOMPRESS_WKSP_SIZE(maxTableLog, maxSymbolValue) (FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) * sizeof(unsigned)) +size_t FSE_decompress_wksp(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize); +/**< same as FSE_decompress(), using an externally allocated `workSpace` produced with `FSE_DECOMPRESS_WKSP_SIZE_U32(maxLog, maxSymbolValue)` */ +size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize, int bmi2); +/**< Same as FSE_decompress_wksp() but with dynamic BMI2 support. Pass 1 if your CPU supports BMI2 or 0 if it doesn't. */ + typedef enum { FSE_repeat_none, /**< Cannot use the previous table */ FSE_repeat_check, /**< Can use the previous table but it must be checked */ @@ -670,9 +670,9 @@ MEM_STATIC unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr) #ifndef FSE_DEFAULT_MEMORY_USAGE # define FSE_DEFAULT_MEMORY_USAGE 13 #endif -#if (FSE_DEFAULT_MEMORY_USAGE > FSE_MAX_MEMORY_USAGE) -# error "FSE_DEFAULT_MEMORY_USAGE must be <= FSE_MAX_MEMORY_USAGE" -#endif +#if (FSE_DEFAULT_MEMORY_USAGE > FSE_MAX_MEMORY_USAGE) +# error "FSE_DEFAULT_MEMORY_USAGE must be <= FSE_MAX_MEMORY_USAGE" +#endif /*!FSE_MAX_SYMBOL_VALUE : * Maximum symbol value authorized. @@ -706,7 +706,7 @@ MEM_STATIC unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr) # error "FSE_MAX_TABLELOG > FSE_TABLELOG_ABSOLUTE_MAX is not supported" #endif -#define FSE_TABLESTEP(tableSize) (((tableSize)>>1) + ((tableSize)>>3) + 3) +#define FSE_TABLESTEP(tableSize) (((tableSize)>>1) + ((tableSize)>>3) + 3) #endif /* FSE_STATIC_LINKING_ONLY */ diff --git a/contrib/libs/zstd/lib/common/fse_decompress.c b/contrib/libs/zstd/lib/common/fse_decompress.c index a5a358015f..e71676bfdb 100644 --- a/contrib/libs/zstd/lib/common/fse_decompress.c +++ b/contrib/libs/zstd/lib/common/fse_decompress.c @@ -1,6 +1,6 @@ /* ****************************************************************** * FSE : Finite State Entropy decoder - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * * You can contact the author at : * - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy @@ -16,14 +16,14 @@ /* ************************************************************** * Includes ****************************************************************/ -#include "debug.h" /* assert */ +#include "debug.h" /* assert */ #include "bitstream.h" #include "compiler.h" #define FSE_STATIC_LINKING_ONLY #include "fse.h" #include "error_private.h" -#define ZSTD_DEPS_NEED_MALLOC -#include "zstd_deps.h" +#define ZSTD_DEPS_NEED_MALLOC +#include "zstd_deps.h" /* ************************************************************** @@ -60,27 +60,27 @@ FSE_DTable* FSE_createDTable (unsigned tableLog) { if (tableLog > FSE_TABLELOG_ABSOLUTE_MAX) tableLog = FSE_TABLELOG_ABSOLUTE_MAX; - return (FSE_DTable*)ZSTD_malloc( FSE_DTABLE_SIZE_U32(tableLog) * sizeof (U32) ); + return (FSE_DTable*)ZSTD_malloc( FSE_DTABLE_SIZE_U32(tableLog) * sizeof (U32) ); } void FSE_freeDTable (FSE_DTable* dt) { - ZSTD_free(dt); + ZSTD_free(dt); } -static size_t FSE_buildDTable_internal(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize) +static size_t FSE_buildDTable_internal(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize) { void* const tdPtr = dt+1; /* because *dt is unsigned, 32-bits aligned on 32-bits */ FSE_DECODE_TYPE* const tableDecode = (FSE_DECODE_TYPE*) (tdPtr); - U16* symbolNext = (U16*)workSpace; - BYTE* spread = (BYTE*)(symbolNext + maxSymbolValue + 1); + U16* symbolNext = (U16*)workSpace; + BYTE* spread = (BYTE*)(symbolNext + maxSymbolValue + 1); U32 const maxSV1 = maxSymbolValue + 1; U32 const tableSize = 1 << tableLog; U32 highThreshold = tableSize-1; /* Sanity Checks */ - if (FSE_BUILD_DTABLE_WKSP_SIZE(tableLog, maxSymbolValue) > wkspSize) return ERROR(maxSymbolValue_tooLarge); + if (FSE_BUILD_DTABLE_WKSP_SIZE(tableLog, maxSymbolValue) > wkspSize) return ERROR(maxSymbolValue_tooLarge); if (maxSymbolValue > FSE_MAX_SYMBOL_VALUE) return ERROR(maxSymbolValue_tooLarge); if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge); @@ -98,57 +98,57 @@ static size_t FSE_buildDTable_internal(FSE_DTable* dt, const short* normalizedCo if (normalizedCounter[s] >= largeLimit) DTableH.fastMode=0; symbolNext[s] = normalizedCounter[s]; } } } - ZSTD_memcpy(dt, &DTableH, sizeof(DTableH)); + ZSTD_memcpy(dt, &DTableH, sizeof(DTableH)); } /* Spread symbols */ - if (highThreshold == tableSize - 1) { - size_t const tableMask = tableSize-1; - size_t const step = FSE_TABLESTEP(tableSize); - /* First lay down the symbols in order. - * We use a uint64_t to lay down 8 bytes at a time. This reduces branch - * misses since small blocks generally have small table logs, so nearly - * all symbols have counts <= 8. We ensure we have 8 bytes at the end of - * our buffer to handle the over-write. - */ - { - U64 const add = 0x0101010101010101ull; - size_t pos = 0; - U64 sv = 0; - U32 s; - for (s=0; s<maxSV1; ++s, sv += add) { - int i; - int const n = normalizedCounter[s]; - MEM_write64(spread + pos, sv); - for (i = 8; i < n; i += 8) { - MEM_write64(spread + pos + i, sv); - } - pos += n; - } - } - /* Now we spread those positions across the table. - * The benefit of doing it in two stages is that we avoid the the - * variable size inner loop, which caused lots of branch misses. - * Now we can run through all the positions without any branch misses. - * We unroll the loop twice, since that is what emperically worked best. - */ - { - size_t position = 0; - size_t s; - size_t const unroll = 2; - assert(tableSize % unroll == 0); /* FSE_MIN_TABLELOG is 5 */ - for (s = 0; s < (size_t)tableSize; s += unroll) { - size_t u; - for (u = 0; u < unroll; ++u) { - size_t const uPosition = (position + (u * step)) & tableMask; - tableDecode[uPosition].symbol = spread[s + u]; - } - position = (position + (unroll * step)) & tableMask; - } - assert(position == 0); - } - } else { - U32 const tableMask = tableSize-1; + if (highThreshold == tableSize - 1) { + size_t const tableMask = tableSize-1; + size_t const step = FSE_TABLESTEP(tableSize); + /* First lay down the symbols in order. + * We use a uint64_t to lay down 8 bytes at a time. This reduces branch + * misses since small blocks generally have small table logs, so nearly + * all symbols have counts <= 8. We ensure we have 8 bytes at the end of + * our buffer to handle the over-write. + */ + { + U64 const add = 0x0101010101010101ull; + size_t pos = 0; + U64 sv = 0; + U32 s; + for (s=0; s<maxSV1; ++s, sv += add) { + int i; + int const n = normalizedCounter[s]; + MEM_write64(spread + pos, sv); + for (i = 8; i < n; i += 8) { + MEM_write64(spread + pos + i, sv); + } + pos += n; + } + } + /* Now we spread those positions across the table. + * The benefit of doing it in two stages is that we avoid the the + * variable size inner loop, which caused lots of branch misses. + * Now we can run through all the positions without any branch misses. + * We unroll the loop twice, since that is what emperically worked best. + */ + { + size_t position = 0; + size_t s; + size_t const unroll = 2; + assert(tableSize % unroll == 0); /* FSE_MIN_TABLELOG is 5 */ + for (s = 0; s < (size_t)tableSize; s += unroll) { + size_t u; + for (u = 0; u < unroll; ++u) { + size_t const uPosition = (position + (u * step)) & tableMask; + tableDecode[uPosition].symbol = spread[s + u]; + } + position = (position + (unroll * step)) & tableMask; + } + assert(position == 0); + } + } else { + U32 const tableMask = tableSize-1; U32 const step = FSE_TABLESTEP(tableSize); U32 s, position = 0; for (s=0; s<maxSV1; s++) { @@ -173,12 +173,12 @@ static size_t FSE_buildDTable_internal(FSE_DTable* dt, const short* normalizedCo return 0; } -size_t FSE_buildDTable_wksp(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize) -{ - return FSE_buildDTable_internal(dt, normalizedCounter, maxSymbolValue, tableLog, workSpace, wkspSize); -} - +size_t FSE_buildDTable_wksp(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize) +{ + return FSE_buildDTable_internal(dt, normalizedCounter, maxSymbolValue, tableLog, workSpace, wkspSize); +} + #ifndef FSE_COMMONDEFS_ONLY /*-******************************************************* @@ -305,99 +305,99 @@ size_t FSE_decompress_usingDTable(void* dst, size_t originalSize, } -size_t FSE_decompress_wksp(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize) -{ - return FSE_decompress_wksp_bmi2(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, /* bmi2 */ 0); -} - -typedef struct { - short ncount[FSE_MAX_SYMBOL_VALUE + 1]; - FSE_DTable dtable[1]; /* Dynamically sized */ -} FSE_DecompressWksp; - - -FORCE_INLINE_TEMPLATE size_t FSE_decompress_wksp_body( - void* dst, size_t dstCapacity, - const void* cSrc, size_t cSrcSize, - unsigned maxLog, void* workSpace, size_t wkspSize, - int bmi2) +size_t FSE_decompress_wksp(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize) { + return FSE_decompress_wksp_bmi2(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, /* bmi2 */ 0); +} + +typedef struct { + short ncount[FSE_MAX_SYMBOL_VALUE + 1]; + FSE_DTable dtable[1]; /* Dynamically sized */ +} FSE_DecompressWksp; + + +FORCE_INLINE_TEMPLATE size_t FSE_decompress_wksp_body( + void* dst, size_t dstCapacity, + const void* cSrc, size_t cSrcSize, + unsigned maxLog, void* workSpace, size_t wkspSize, + int bmi2) +{ const BYTE* const istart = (const BYTE*)cSrc; const BYTE* ip = istart; unsigned tableLog; unsigned maxSymbolValue = FSE_MAX_SYMBOL_VALUE; - FSE_DecompressWksp* const wksp = (FSE_DecompressWksp*)workSpace; - - DEBUG_STATIC_ASSERT((FSE_MAX_SYMBOL_VALUE + 1) % 2 == 0); - if (wkspSize < sizeof(*wksp)) return ERROR(GENERIC); + FSE_DecompressWksp* const wksp = (FSE_DecompressWksp*)workSpace; + DEBUG_STATIC_ASSERT((FSE_MAX_SYMBOL_VALUE + 1) % 2 == 0); + if (wkspSize < sizeof(*wksp)) return ERROR(GENERIC); + /* normal FSE decoding mode */ - { - size_t const NCountLength = FSE_readNCount_bmi2(wksp->ncount, &maxSymbolValue, &tableLog, istart, cSrcSize, bmi2); - if (FSE_isError(NCountLength)) return NCountLength; - if (tableLog > maxLog) return ERROR(tableLog_tooLarge); - assert(NCountLength <= cSrcSize); - ip += NCountLength; - cSrcSize -= NCountLength; - } - - if (FSE_DECOMPRESS_WKSP_SIZE(tableLog, maxSymbolValue) > wkspSize) return ERROR(tableLog_tooLarge); - workSpace = wksp->dtable + FSE_DTABLE_SIZE_U32(tableLog); - wkspSize -= sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog); - - CHECK_F( FSE_buildDTable_internal(wksp->dtable, wksp->ncount, maxSymbolValue, tableLog, workSpace, wkspSize) ); - - { - const void* ptr = wksp->dtable; - const FSE_DTableHeader* DTableH = (const FSE_DTableHeader*)ptr; - const U32 fastMode = DTableH->fastMode; - - /* select fast mode (static) */ - if (fastMode) return FSE_decompress_usingDTable_generic(dst, dstCapacity, ip, cSrcSize, wksp->dtable, 1); - return FSE_decompress_usingDTable_generic(dst, dstCapacity, ip, cSrcSize, wksp->dtable, 0); - } + { + size_t const NCountLength = FSE_readNCount_bmi2(wksp->ncount, &maxSymbolValue, &tableLog, istart, cSrcSize, bmi2); + if (FSE_isError(NCountLength)) return NCountLength; + if (tableLog > maxLog) return ERROR(tableLog_tooLarge); + assert(NCountLength <= cSrcSize); + ip += NCountLength; + cSrcSize -= NCountLength; + } + + if (FSE_DECOMPRESS_WKSP_SIZE(tableLog, maxSymbolValue) > wkspSize) return ERROR(tableLog_tooLarge); + workSpace = wksp->dtable + FSE_DTABLE_SIZE_U32(tableLog); + wkspSize -= sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog); + + CHECK_F( FSE_buildDTable_internal(wksp->dtable, wksp->ncount, maxSymbolValue, tableLog, workSpace, wkspSize) ); + + { + const void* ptr = wksp->dtable; + const FSE_DTableHeader* DTableH = (const FSE_DTableHeader*)ptr; + const U32 fastMode = DTableH->fastMode; + + /* select fast mode (static) */ + if (fastMode) return FSE_decompress_usingDTable_generic(dst, dstCapacity, ip, cSrcSize, wksp->dtable, 1); + return FSE_decompress_usingDTable_generic(dst, dstCapacity, ip, cSrcSize, wksp->dtable, 0); + } } -/* Avoids the FORCE_INLINE of the _body() function. */ -static size_t FSE_decompress_wksp_body_default(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize) -{ - return FSE_decompress_wksp_body(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, 0); -} +/* Avoids the FORCE_INLINE of the _body() function. */ +static size_t FSE_decompress_wksp_body_default(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize) +{ + return FSE_decompress_wksp_body(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, 0); +} -#if DYNAMIC_BMI2 +#if DYNAMIC_BMI2 BMI2_TARGET_ATTRIBUTE static size_t FSE_decompress_wksp_body_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize) -{ - return FSE_decompress_wksp_body(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, 1); -} -#endif - -size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize, int bmi2) -{ -#if DYNAMIC_BMI2 - if (bmi2) { - return FSE_decompress_wksp_body_bmi2(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize); - } -#endif - (void)bmi2; - return FSE_decompress_wksp_body_default(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize); -} - - +{ + return FSE_decompress_wksp_body(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, 1); +} +#endif + +size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize, int bmi2) +{ +#if DYNAMIC_BMI2 + if (bmi2) { + return FSE_decompress_wksp_body_bmi2(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize); + } +#endif + (void)bmi2; + return FSE_decompress_wksp_body_default(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize); +} + + typedef FSE_DTable DTable_max_t[FSE_DTABLE_SIZE_U32(FSE_MAX_TABLELOG)]; -#ifndef ZSTD_NO_UNUSED_FUNCTIONS -size_t FSE_buildDTable(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog) { - U32 wksp[FSE_BUILD_DTABLE_WKSP_SIZE_U32(FSE_TABLELOG_ABSOLUTE_MAX, FSE_MAX_SYMBOL_VALUE)]; - return FSE_buildDTable_wksp(dt, normalizedCounter, maxSymbolValue, tableLog, wksp, sizeof(wksp)); -} - +#ifndef ZSTD_NO_UNUSED_FUNCTIONS +size_t FSE_buildDTable(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog) { + U32 wksp[FSE_BUILD_DTABLE_WKSP_SIZE_U32(FSE_TABLELOG_ABSOLUTE_MAX, FSE_MAX_SYMBOL_VALUE)]; + return FSE_buildDTable_wksp(dt, normalizedCounter, maxSymbolValue, tableLog, wksp, sizeof(wksp)); +} + size_t FSE_decompress(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize) { - /* Static analyzer seems unable to understand this table will be properly initialized later */ - U32 wksp[FSE_DECOMPRESS_WKSP_SIZE_U32(FSE_MAX_TABLELOG, FSE_MAX_SYMBOL_VALUE)]; - return FSE_decompress_wksp(dst, dstCapacity, cSrc, cSrcSize, FSE_MAX_TABLELOG, wksp, sizeof(wksp)); + /* Static analyzer seems unable to understand this table will be properly initialized later */ + U32 wksp[FSE_DECOMPRESS_WKSP_SIZE_U32(FSE_MAX_TABLELOG, FSE_MAX_SYMBOL_VALUE)]; + return FSE_decompress_wksp(dst, dstCapacity, cSrc, cSrcSize, FSE_MAX_TABLELOG, wksp, sizeof(wksp)); } -#endif +#endif #endif /* FSE_COMMONDEFS_ONLY */ diff --git a/contrib/libs/zstd/lib/common/huf.h b/contrib/libs/zstd/lib/common/huf.h index 85518481ec..7cc59e3874 100644 --- a/contrib/libs/zstd/lib/common/huf.h +++ b/contrib/libs/zstd/lib/common/huf.h @@ -1,7 +1,7 @@ /* ****************************************************************** * huff0 huffman codec, * part of Finite State Entropy library - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * * You can contact the author at : * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy @@ -20,7 +20,7 @@ extern "C" { #define HUF_H_298734234 /* *** Dependencies *** */ -#include "zstd_deps.h" /* size_t */ +#include "zstd_deps.h" /* size_t */ /* *** library symbols visibility *** */ @@ -111,8 +111,8 @@ HUF_PUBLIC_API size_t HUF_compress4X_wksp (void* dst, size_t dstCapacity, /* *** Dependencies *** */ #include "mem.h" /* U32 */ -#define FSE_STATIC_LINKING_ONLY -#include "fse.h" +#define FSE_STATIC_LINKING_ONLY +#include "fse.h" /* *** Constants *** */ @@ -135,7 +135,7 @@ HUF_PUBLIC_API size_t HUF_compress4X_wksp (void* dst, size_t dstCapacity, #define HUF_COMPRESSBOUND(size) (HUF_CTABLEBOUND + HUF_BLOCKBOUND(size)) /* Macro version, useful for static allocation */ /* static allocation of HUF's Compression Table */ -/* this is a private definition, just exposed for allocation and strict aliasing purpose. never EVER access its members directly */ +/* this is a private definition, just exposed for allocation and strict aliasing purpose. never EVER access its members directly */ typedef size_t HUF_CElt; /* consider it an incomplete type */ #define HUF_CTABLE_SIZE_ST(maxSymbolValue) ((maxSymbolValue)+2) /* Use tables of size_t, for proper alignment */ #define HUF_CTABLE_SIZE(maxSymbolValue) (HUF_CTABLE_SIZE_ST(maxSymbolValue) * sizeof(size_t)) @@ -188,7 +188,7 @@ size_t HUF_decompress4X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue); size_t HUF_buildCTable (HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue, unsigned maxNbBits); /* @return : maxNbBits; CTable and count can overlap. In which case, CTable will overwrite count content */ size_t HUF_writeCTable (void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog); -size_t HUF_writeCTable_wksp(void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog, void* workspace, size_t workspaceSize); +size_t HUF_writeCTable_wksp(void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog, void* workspace, size_t workspaceSize); size_t HUF_compress4X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable); size_t HUF_compress4X_usingCTable_bmi2(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int bmi2); size_t HUF_estimateCompressedSize(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue); @@ -230,19 +230,19 @@ size_t HUF_readStats(BYTE* huffWeight, size_t hwSize, U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr, const void* src, size_t srcSize); -/*! HUF_readStats_wksp() : - * Same as HUF_readStats() but takes an external workspace which must be - * 4-byte aligned and its size must be >= HUF_READ_STATS_WORKSPACE_SIZE. - * If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0. - */ -#define HUF_READ_STATS_WORKSPACE_SIZE_U32 FSE_DECOMPRESS_WKSP_SIZE_U32(6, HUF_TABLELOG_MAX-1) -#define HUF_READ_STATS_WORKSPACE_SIZE (HUF_READ_STATS_WORKSPACE_SIZE_U32 * sizeof(unsigned)) -size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize, - U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr, - const void* src, size_t srcSize, - void* workspace, size_t wkspSize, - int bmi2); - +/*! HUF_readStats_wksp() : + * Same as HUF_readStats() but takes an external workspace which must be + * 4-byte aligned and its size must be >= HUF_READ_STATS_WORKSPACE_SIZE. + * If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0. + */ +#define HUF_READ_STATS_WORKSPACE_SIZE_U32 FSE_DECOMPRESS_WKSP_SIZE_U32(6, HUF_TABLELOG_MAX-1) +#define HUF_READ_STATS_WORKSPACE_SIZE (HUF_READ_STATS_WORKSPACE_SIZE_U32 * sizeof(unsigned)) +size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize, + U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr, + const void* src, size_t srcSize, + void* workspace, size_t wkspSize, + int bmi2); + /** HUF_readCTable() : * Loading a CTable saved with HUF_writeCTable() */ size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize, unsigned *hasZeroWeights); @@ -276,7 +276,7 @@ U32 HUF_selectDecoder (size_t dstSize, size_t cSrcSize); * a required workspace size greater than that specified in the following * macro. */ -#define HUF_DECOMPRESS_WORKSPACE_SIZE ((2 << 10) + (1 << 9)) +#define HUF_DECOMPRESS_WORKSPACE_SIZE ((2 << 10) + (1 << 9)) #define HUF_DECOMPRESS_WORKSPACE_SIZE_U32 (HUF_DECOMPRESS_WORKSPACE_SIZE / sizeof(U32)) #ifndef HUF_FORCE_DECOMPRESS_X2 @@ -350,9 +350,9 @@ size_t HUF_decompress1X1_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstS #endif size_t HUF_decompress4X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2); size_t HUF_decompress4X_hufOnly_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2); -#ifndef HUF_FORCE_DECOMPRESS_X2 -size_t HUF_readDTableX1_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2); -#endif +#ifndef HUF_FORCE_DECOMPRESS_X2 +size_t HUF_readDTableX1_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2); +#endif #ifndef HUF_FORCE_DECOMPRESS_X1 size_t HUF_readDTableX2_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2); #endif diff --git a/contrib/libs/zstd/lib/common/mem.h b/contrib/libs/zstd/lib/common/mem.h index 85581c3847..dba1fed6b5 100644 --- a/contrib/libs/zstd/lib/common/mem.h +++ b/contrib/libs/zstd/lib/common/mem.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -18,10 +18,10 @@ extern "C" { /*-**************************************** * Dependencies ******************************************/ -#include <stddef.h> /* size_t, ptrdiff_t */ -#include "compiler.h" /* __has_builtin */ -#include "debug.h" /* DEBUG_STATIC_ASSERT */ -#include "zstd_deps.h" /* ZSTD_memcpy */ +#include <stddef.h> /* size_t, ptrdiff_t */ +#include "compiler.h" /* __has_builtin */ +#include "debug.h" /* DEBUG_STATIC_ASSERT */ +#include "zstd_deps.h" /* ZSTD_memcpy */ /*-**************************************** @@ -45,11 +45,11 @@ extern "C" { * Basic Types *****************************************************************/ #if !defined (__VMS) && (defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) ) -# if defined(_AIX) -# include <inttypes.h> -# else -# include <stdint.h> /* intptr_t */ -# endif +# if defined(_AIX) +# include <inttypes.h> +# else +# include <stdint.h> /* intptr_t */ +# endif typedef uint8_t BYTE; typedef uint8_t U8; typedef int8_t S8; @@ -85,54 +85,54 @@ extern "C" { /*-************************************************************** -* Memory I/O API -*****************************************************************/ -/*=== Static platform detection ===*/ -MEM_STATIC unsigned MEM_32bits(void); -MEM_STATIC unsigned MEM_64bits(void); -MEM_STATIC unsigned MEM_isLittleEndian(void); - -/*=== Native unaligned read/write ===*/ -MEM_STATIC U16 MEM_read16(const void* memPtr); -MEM_STATIC U32 MEM_read32(const void* memPtr); -MEM_STATIC U64 MEM_read64(const void* memPtr); -MEM_STATIC size_t MEM_readST(const void* memPtr); - -MEM_STATIC void MEM_write16(void* memPtr, U16 value); -MEM_STATIC void MEM_write32(void* memPtr, U32 value); -MEM_STATIC void MEM_write64(void* memPtr, U64 value); - -/*=== Little endian unaligned read/write ===*/ -MEM_STATIC U16 MEM_readLE16(const void* memPtr); -MEM_STATIC U32 MEM_readLE24(const void* memPtr); -MEM_STATIC U32 MEM_readLE32(const void* memPtr); -MEM_STATIC U64 MEM_readLE64(const void* memPtr); -MEM_STATIC size_t MEM_readLEST(const void* memPtr); - -MEM_STATIC void MEM_writeLE16(void* memPtr, U16 val); -MEM_STATIC void MEM_writeLE24(void* memPtr, U32 val); -MEM_STATIC void MEM_writeLE32(void* memPtr, U32 val32); -MEM_STATIC void MEM_writeLE64(void* memPtr, U64 val64); -MEM_STATIC void MEM_writeLEST(void* memPtr, size_t val); - -/*=== Big endian unaligned read/write ===*/ -MEM_STATIC U32 MEM_readBE32(const void* memPtr); -MEM_STATIC U64 MEM_readBE64(const void* memPtr); -MEM_STATIC size_t MEM_readBEST(const void* memPtr); - -MEM_STATIC void MEM_writeBE32(void* memPtr, U32 val32); -MEM_STATIC void MEM_writeBE64(void* memPtr, U64 val64); -MEM_STATIC void MEM_writeBEST(void* memPtr, size_t val); - -/*=== Byteswap ===*/ -MEM_STATIC U32 MEM_swap32(U32 in); -MEM_STATIC U64 MEM_swap64(U64 in); -MEM_STATIC size_t MEM_swapST(size_t in); - - -/*-************************************************************** -* Memory I/O Implementation +* Memory I/O API *****************************************************************/ +/*=== Static platform detection ===*/ +MEM_STATIC unsigned MEM_32bits(void); +MEM_STATIC unsigned MEM_64bits(void); +MEM_STATIC unsigned MEM_isLittleEndian(void); + +/*=== Native unaligned read/write ===*/ +MEM_STATIC U16 MEM_read16(const void* memPtr); +MEM_STATIC U32 MEM_read32(const void* memPtr); +MEM_STATIC U64 MEM_read64(const void* memPtr); +MEM_STATIC size_t MEM_readST(const void* memPtr); + +MEM_STATIC void MEM_write16(void* memPtr, U16 value); +MEM_STATIC void MEM_write32(void* memPtr, U32 value); +MEM_STATIC void MEM_write64(void* memPtr, U64 value); + +/*=== Little endian unaligned read/write ===*/ +MEM_STATIC U16 MEM_readLE16(const void* memPtr); +MEM_STATIC U32 MEM_readLE24(const void* memPtr); +MEM_STATIC U32 MEM_readLE32(const void* memPtr); +MEM_STATIC U64 MEM_readLE64(const void* memPtr); +MEM_STATIC size_t MEM_readLEST(const void* memPtr); + +MEM_STATIC void MEM_writeLE16(void* memPtr, U16 val); +MEM_STATIC void MEM_writeLE24(void* memPtr, U32 val); +MEM_STATIC void MEM_writeLE32(void* memPtr, U32 val32); +MEM_STATIC void MEM_writeLE64(void* memPtr, U64 val64); +MEM_STATIC void MEM_writeLEST(void* memPtr, size_t val); + +/*=== Big endian unaligned read/write ===*/ +MEM_STATIC U32 MEM_readBE32(const void* memPtr); +MEM_STATIC U64 MEM_readBE64(const void* memPtr); +MEM_STATIC size_t MEM_readBEST(const void* memPtr); + +MEM_STATIC void MEM_writeBE32(void* memPtr, U32 val32); +MEM_STATIC void MEM_writeBE64(void* memPtr, U64 val64); +MEM_STATIC void MEM_writeBEST(void* memPtr, size_t val); + +/*=== Byteswap ===*/ +MEM_STATIC U32 MEM_swap32(U32 in); +MEM_STATIC U64 MEM_swap64(U64 in); +MEM_STATIC size_t MEM_swapST(size_t in); + + +/*-************************************************************** +* Memory I/O Implementation +*****************************************************************/ /* MEM_FORCE_MEMORY_ACCESS : * By default, access to unaligned memory is controlled by `memcpy()`, which is safe and portable. * Unfortunately, on some target/compiler combinations, the generated assembly is sub-optimal. @@ -147,7 +147,7 @@ MEM_STATIC size_t MEM_swapST(size_t in); * Prefer these methods in priority order (0 > 1 > 2) */ #ifndef MEM_FORCE_MEMORY_ACCESS /* can be defined externally, on command line for example */ -# if defined(__INTEL_COMPILER) || defined(__GNUC__) || defined(__ICCARM__) +# if defined(__INTEL_COMPILER) || defined(__GNUC__) || defined(__ICCARM__) # define MEM_FORCE_MEMORY_ACCESS 1 # endif #endif @@ -222,37 +222,37 @@ MEM_STATIC void MEM_write64(void* memPtr, U64 value) { ((unalign64*)memPtr)->v = MEM_STATIC U16 MEM_read16(const void* memPtr) { - U16 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val; + U16 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val; } MEM_STATIC U32 MEM_read32(const void* memPtr) { - U32 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val; + U32 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val; } MEM_STATIC U64 MEM_read64(const void* memPtr) { - U64 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val; + U64 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val; } MEM_STATIC size_t MEM_readST(const void* memPtr) { - size_t val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val; + size_t val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val; } MEM_STATIC void MEM_write16(void* memPtr, U16 value) { - ZSTD_memcpy(memPtr, &value, sizeof(value)); + ZSTD_memcpy(memPtr, &value, sizeof(value)); } MEM_STATIC void MEM_write32(void* memPtr, U32 value) { - ZSTD_memcpy(memPtr, &value, sizeof(value)); + ZSTD_memcpy(memPtr, &value, sizeof(value)); } MEM_STATIC void MEM_write64(void* memPtr, U64 value) { - ZSTD_memcpy(memPtr, &value, sizeof(value)); + ZSTD_memcpy(memPtr, &value, sizeof(value)); } #endif /* MEM_FORCE_MEMORY_ACCESS */ @@ -324,7 +324,7 @@ MEM_STATIC void MEM_writeLE16(void* memPtr, U16 val) MEM_STATIC U32 MEM_readLE24(const void* memPtr) { - return (U32)MEM_readLE16(memPtr) + ((U32)(((const BYTE*)memPtr)[2]) << 16); + return (U32)MEM_readLE16(memPtr) + ((U32)(((const BYTE*)memPtr)[2]) << 16); } MEM_STATIC void MEM_writeLE24(void* memPtr, U32 val) @@ -431,10 +431,10 @@ MEM_STATIC void MEM_writeBEST(void* memPtr, size_t val) MEM_writeBE64(memPtr, (U64)val); } -/* code only tested on 32 and 64 bits systems */ -MEM_STATIC void MEM_check(void) { DEBUG_STATIC_ASSERT((sizeof(size_t)==4) || (sizeof(size_t)==8)); } - +/* code only tested on 32 and 64 bits systems */ +MEM_STATIC void MEM_check(void) { DEBUG_STATIC_ASSERT((sizeof(size_t)==4) || (sizeof(size_t)==8)); } + #if defined (__cplusplus) } #endif diff --git a/contrib/libs/zstd/lib/common/pool.c b/contrib/libs/zstd/lib/common/pool.c index 2e37cdd73c..d88f5df45e 100644 --- a/contrib/libs/zstd/lib/common/pool.c +++ b/contrib/libs/zstd/lib/common/pool.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -10,9 +10,9 @@ /* ====== Dependencies ======= */ -#include "zstd_deps.h" /* size_t */ +#include "zstd_deps.h" /* size_t */ #include "debug.h" /* assert */ -#include "zstd_internal.h" /* ZSTD_customMalloc, ZSTD_customFree */ +#include "zstd_internal.h" /* ZSTD_customMalloc, ZSTD_customFree */ #include "pool.h" /* ====== Compiler specifics ====== */ @@ -106,10 +106,10 @@ static void* POOL_thread(void* opaque) { } /* ZSTD_createThreadPool() : public access point */ -POOL_ctx* ZSTD_createThreadPool(size_t numThreads) { - return POOL_create (numThreads, 0); -} - +POOL_ctx* ZSTD_createThreadPool(size_t numThreads) { + return POOL_create (numThreads, 0); +} + POOL_ctx* POOL_create(size_t numThreads, size_t queueSize) { return POOL_create_advanced(numThreads, queueSize, ZSTD_defaultCMem); } @@ -121,14 +121,14 @@ POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize, /* Check parameters */ if (!numThreads) { return NULL; } /* Allocate the context and zero initialize */ - ctx = (POOL_ctx*)ZSTD_customCalloc(sizeof(POOL_ctx), customMem); + ctx = (POOL_ctx*)ZSTD_customCalloc(sizeof(POOL_ctx), customMem); if (!ctx) { return NULL; } /* Initialize the job queue. * It needs one extra space since one space is wasted to differentiate * empty and full queues. */ ctx->queueSize = queueSize + 1; - ctx->queue = (POOL_job*)ZSTD_customMalloc(ctx->queueSize * sizeof(POOL_job), customMem); + ctx->queue = (POOL_job*)ZSTD_customMalloc(ctx->queueSize * sizeof(POOL_job), customMem); ctx->queueHead = 0; ctx->queueTail = 0; ctx->numThreadsBusy = 0; @@ -142,7 +142,7 @@ POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize, } ctx->shutdown = 0; /* Allocate space for the thread handles */ - ctx->threads = (ZSTD_pthread_t*)ZSTD_customMalloc(numThreads * sizeof(ZSTD_pthread_t), customMem); + ctx->threads = (ZSTD_pthread_t*)ZSTD_customMalloc(numThreads * sizeof(ZSTD_pthread_t), customMem); ctx->threadCapacity = 0; ctx->customMem = customMem; /* Check for errors */ @@ -185,14 +185,14 @@ void POOL_free(POOL_ctx *ctx) { ZSTD_pthread_mutex_destroy(&ctx->queueMutex); ZSTD_pthread_cond_destroy(&ctx->queuePushCond); ZSTD_pthread_cond_destroy(&ctx->queuePopCond); - ZSTD_customFree(ctx->queue, ctx->customMem); - ZSTD_customFree(ctx->threads, ctx->customMem); - ZSTD_customFree(ctx, ctx->customMem); + ZSTD_customFree(ctx->queue, ctx->customMem); + ZSTD_customFree(ctx->threads, ctx->customMem); + ZSTD_customFree(ctx, ctx->customMem); } -void ZSTD_freeThreadPool (ZSTD_threadPool* pool) { - POOL_free (pool); -} +void ZSTD_freeThreadPool (ZSTD_threadPool* pool) { + POOL_free (pool); +} size_t POOL_sizeof(const POOL_ctx* ctx) { if (ctx==NULL) return 0; /* supports sizeof NULL */ @@ -211,11 +211,11 @@ static int POOL_resize_internal(POOL_ctx* ctx, size_t numThreads) return 0; } /* numThreads > threadCapacity */ - { ZSTD_pthread_t* const threadPool = (ZSTD_pthread_t*)ZSTD_customMalloc(numThreads * sizeof(ZSTD_pthread_t), ctx->customMem); + { ZSTD_pthread_t* const threadPool = (ZSTD_pthread_t*)ZSTD_customMalloc(numThreads * sizeof(ZSTD_pthread_t), ctx->customMem); if (!threadPool) return 1; /* replace existing thread pool */ - ZSTD_memcpy(threadPool, ctx->threads, ctx->threadCapacity * sizeof(*threadPool)); - ZSTD_customFree(ctx->threads, ctx->customMem); + ZSTD_memcpy(threadPool, ctx->threads, ctx->threadCapacity * sizeof(*threadPool)); + ZSTD_customFree(ctx->threads, ctx->customMem); ctx->threads = threadPool; /* Initialize additional threads */ { size_t threadId; @@ -310,7 +310,7 @@ int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque) struct POOL_ctx_s { int dummy; }; -static POOL_ctx g_poolCtx; +static POOL_ctx g_poolCtx; POOL_ctx* POOL_create(size_t numThreads, size_t queueSize) { return POOL_create_advanced(numThreads, queueSize, ZSTD_defaultCMem); @@ -322,11 +322,11 @@ POOL_create_advanced(size_t numThreads, size_t queueSize, ZSTD_customMem customM (void)numThreads; (void)queueSize; (void)customMem; - return &g_poolCtx; + return &g_poolCtx; } void POOL_free(POOL_ctx* ctx) { - assert(!ctx || ctx == &g_poolCtx); + assert(!ctx || ctx == &g_poolCtx); (void)ctx; } @@ -348,7 +348,7 @@ int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque) { size_t POOL_sizeof(const POOL_ctx* ctx) { if (ctx==NULL) return 0; /* supports sizeof NULL */ - assert(ctx == &g_poolCtx); + assert(ctx == &g_poolCtx); return sizeof(*ctx); } diff --git a/contrib/libs/zstd/lib/common/pool.h b/contrib/libs/zstd/lib/common/pool.h index 0ebde1805d..e48a515763 100644 --- a/contrib/libs/zstd/lib/common/pool.h +++ b/contrib/libs/zstd/lib/common/pool.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -16,7 +16,7 @@ extern "C" { #endif -#include "zstd_deps.h" +#include "zstd_deps.h" #define ZSTD_STATIC_LINKING_ONLY /* ZSTD_customMem */ #include "../zstd.h" diff --git a/contrib/libs/zstd/lib/common/threading.c b/contrib/libs/zstd/lib/common/threading.c index 92cf57c195..6729b60a35 100644 --- a/contrib/libs/zstd/lib/common/threading.c +++ b/contrib/libs/zstd/lib/common/threading.c @@ -78,12 +78,12 @@ int ZSTD_pthread_join(ZSTD_pthread_t thread, void **value_ptr) #if defined(ZSTD_MULTITHREAD) && DEBUGLEVEL >= 1 && !defined(_WIN32) -#define ZSTD_DEPS_NEED_MALLOC -#include "zstd_deps.h" +#define ZSTD_DEPS_NEED_MALLOC +#include "zstd_deps.h" int ZSTD_pthread_mutex_init(ZSTD_pthread_mutex_t* mutex, pthread_mutexattr_t const* attr) { - *mutex = (pthread_mutex_t*)ZSTD_malloc(sizeof(pthread_mutex_t)); + *mutex = (pthread_mutex_t*)ZSTD_malloc(sizeof(pthread_mutex_t)); if (!*mutex) return 1; return pthread_mutex_init(*mutex, attr); @@ -95,14 +95,14 @@ int ZSTD_pthread_mutex_destroy(ZSTD_pthread_mutex_t* mutex) return 0; { int const ret = pthread_mutex_destroy(*mutex); - ZSTD_free(*mutex); + ZSTD_free(*mutex); return ret; } } int ZSTD_pthread_cond_init(ZSTD_pthread_cond_t* cond, pthread_condattr_t const* attr) { - *cond = (pthread_cond_t*)ZSTD_malloc(sizeof(pthread_cond_t)); + *cond = (pthread_cond_t*)ZSTD_malloc(sizeof(pthread_cond_t)); if (!*cond) return 1; return pthread_cond_init(*cond, attr); @@ -114,7 +114,7 @@ int ZSTD_pthread_cond_destroy(ZSTD_pthread_cond_t* cond) return 0; { int const ret = pthread_cond_destroy(*cond); - ZSTD_free(*cond); + ZSTD_free(*cond); return ret; } } diff --git a/contrib/libs/zstd/lib/common/zstd_common.c b/contrib/libs/zstd/lib/common/zstd_common.c index 3d7e35b309..ef5ff1cbbd 100644 --- a/contrib/libs/zstd/lib/common/zstd_common.c +++ b/contrib/libs/zstd/lib/common/zstd_common.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -13,8 +13,8 @@ /*-************************************* * Dependencies ***************************************/ -#define ZSTD_DEPS_NEED_MALLOC -#include "zstd_deps.h" /* ZSTD_malloc, ZSTD_calloc, ZSTD_free, ZSTD_memset */ +#define ZSTD_DEPS_NEED_MALLOC +#include "zstd_deps.h" /* ZSTD_malloc, ZSTD_calloc, ZSTD_free, ZSTD_memset */ #include "error_private.h" #include "zstd_internal.h" @@ -53,31 +53,31 @@ const char* ZSTD_getErrorString(ZSTD_ErrorCode code) { return ERR_getErrorString /*=************************************************************** * Custom allocator ****************************************************************/ -void* ZSTD_customMalloc(size_t size, ZSTD_customMem customMem) +void* ZSTD_customMalloc(size_t size, ZSTD_customMem customMem) { if (customMem.customAlloc) return customMem.customAlloc(customMem.opaque, size); - return ZSTD_malloc(size); + return ZSTD_malloc(size); } -void* ZSTD_customCalloc(size_t size, ZSTD_customMem customMem) +void* ZSTD_customCalloc(size_t size, ZSTD_customMem customMem) { if (customMem.customAlloc) { /* calloc implemented as malloc+memset; * not as efficient as calloc, but next best guess for custom malloc */ void* const ptr = customMem.customAlloc(customMem.opaque, size); - ZSTD_memset(ptr, 0, size); + ZSTD_memset(ptr, 0, size); return ptr; } - return ZSTD_calloc(1, size); + return ZSTD_calloc(1, size); } -void ZSTD_customFree(void* ptr, ZSTD_customMem customMem) +void ZSTD_customFree(void* ptr, ZSTD_customMem customMem) { if (ptr!=NULL) { if (customMem.customFree) customMem.customFree(customMem.opaque, ptr); else - ZSTD_free(ptr); + ZSTD_free(ptr); } } diff --git a/contrib/libs/zstd/lib/common/zstd_deps.h b/contrib/libs/zstd/lib/common/zstd_deps.h index 14211344a0..ba734b770b 100644 --- a/contrib/libs/zstd/lib/common/zstd_deps.h +++ b/contrib/libs/zstd/lib/common/zstd_deps.h @@ -1,111 +1,111 @@ -/* - * Copyright (c) Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -/* This file provides common libc dependencies that zstd requires. - * The purpose is to allow replacing this file with a custom implementation - * to compile zstd without libc support. - */ - -/* Need: - * NULL - * INT_MAX - * UINT_MAX - * ZSTD_memcpy() - * ZSTD_memset() - * ZSTD_memmove() - */ -#ifndef ZSTD_DEPS_COMMON -#define ZSTD_DEPS_COMMON - -#include <limits.h> -#include <stddef.h> -#include <string.h> - -#if defined(__GNUC__) && __GNUC__ >= 4 -# define ZSTD_memcpy(d,s,l) __builtin_memcpy((d),(s),(l)) -# define ZSTD_memmove(d,s,l) __builtin_memmove((d),(s),(l)) -# define ZSTD_memset(p,v,l) __builtin_memset((p),(v),(l)) -#else -# define ZSTD_memcpy(d,s,l) memcpy((d),(s),(l)) -# define ZSTD_memmove(d,s,l) memmove((d),(s),(l)) -# define ZSTD_memset(p,v,l) memset((p),(v),(l)) -#endif - -#endif /* ZSTD_DEPS_COMMON */ - -/* Need: - * ZSTD_malloc() - * ZSTD_free() - * ZSTD_calloc() - */ -#ifdef ZSTD_DEPS_NEED_MALLOC -#ifndef ZSTD_DEPS_MALLOC -#define ZSTD_DEPS_MALLOC - -#include <stdlib.h> - -#define ZSTD_malloc(s) malloc(s) -#define ZSTD_calloc(n,s) calloc((n), (s)) -#define ZSTD_free(p) free((p)) - -#endif /* ZSTD_DEPS_MALLOC */ -#endif /* ZSTD_DEPS_NEED_MALLOC */ - -/* - * Provides 64-bit math support. - * Need: - * U64 ZSTD_div64(U64 dividend, U32 divisor) - */ -#ifdef ZSTD_DEPS_NEED_MATH64 -#ifndef ZSTD_DEPS_MATH64 -#define ZSTD_DEPS_MATH64 - -#define ZSTD_div64(dividend, divisor) ((dividend) / (divisor)) - -#endif /* ZSTD_DEPS_MATH64 */ -#endif /* ZSTD_DEPS_NEED_MATH64 */ - -/* Need: - * assert() - */ -#ifdef ZSTD_DEPS_NEED_ASSERT -#ifndef ZSTD_DEPS_ASSERT -#define ZSTD_DEPS_ASSERT - -#include <assert.h> - -#endif /* ZSTD_DEPS_ASSERT */ -#endif /* ZSTD_DEPS_NEED_ASSERT */ - -/* Need: - * ZSTD_DEBUG_PRINT() - */ -#ifdef ZSTD_DEPS_NEED_IO -#ifndef ZSTD_DEPS_IO -#define ZSTD_DEPS_IO - -#include <stdio.h> -#define ZSTD_DEBUG_PRINT(...) fprintf(stderr, __VA_ARGS__) - -#endif /* ZSTD_DEPS_IO */ -#endif /* ZSTD_DEPS_NEED_IO */ - -/* Only requested when <stdint.h> is known to be present. - * Need: - * intptr_t - */ -#ifdef ZSTD_DEPS_NEED_STDINT -#ifndef ZSTD_DEPS_STDINT -#define ZSTD_DEPS_STDINT - -#include <stdint.h> - -#endif /* ZSTD_DEPS_STDINT */ -#endif /* ZSTD_DEPS_NEED_STDINT */ +/* + * Copyright (c) Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + +/* This file provides common libc dependencies that zstd requires. + * The purpose is to allow replacing this file with a custom implementation + * to compile zstd without libc support. + */ + +/* Need: + * NULL + * INT_MAX + * UINT_MAX + * ZSTD_memcpy() + * ZSTD_memset() + * ZSTD_memmove() + */ +#ifndef ZSTD_DEPS_COMMON +#define ZSTD_DEPS_COMMON + +#include <limits.h> +#include <stddef.h> +#include <string.h> + +#if defined(__GNUC__) && __GNUC__ >= 4 +# define ZSTD_memcpy(d,s,l) __builtin_memcpy((d),(s),(l)) +# define ZSTD_memmove(d,s,l) __builtin_memmove((d),(s),(l)) +# define ZSTD_memset(p,v,l) __builtin_memset((p),(v),(l)) +#else +# define ZSTD_memcpy(d,s,l) memcpy((d),(s),(l)) +# define ZSTD_memmove(d,s,l) memmove((d),(s),(l)) +# define ZSTD_memset(p,v,l) memset((p),(v),(l)) +#endif + +#endif /* ZSTD_DEPS_COMMON */ + +/* Need: + * ZSTD_malloc() + * ZSTD_free() + * ZSTD_calloc() + */ +#ifdef ZSTD_DEPS_NEED_MALLOC +#ifndef ZSTD_DEPS_MALLOC +#define ZSTD_DEPS_MALLOC + +#include <stdlib.h> + +#define ZSTD_malloc(s) malloc(s) +#define ZSTD_calloc(n,s) calloc((n), (s)) +#define ZSTD_free(p) free((p)) + +#endif /* ZSTD_DEPS_MALLOC */ +#endif /* ZSTD_DEPS_NEED_MALLOC */ + +/* + * Provides 64-bit math support. + * Need: + * U64 ZSTD_div64(U64 dividend, U32 divisor) + */ +#ifdef ZSTD_DEPS_NEED_MATH64 +#ifndef ZSTD_DEPS_MATH64 +#define ZSTD_DEPS_MATH64 + +#define ZSTD_div64(dividend, divisor) ((dividend) / (divisor)) + +#endif /* ZSTD_DEPS_MATH64 */ +#endif /* ZSTD_DEPS_NEED_MATH64 */ + +/* Need: + * assert() + */ +#ifdef ZSTD_DEPS_NEED_ASSERT +#ifndef ZSTD_DEPS_ASSERT +#define ZSTD_DEPS_ASSERT + +#include <assert.h> + +#endif /* ZSTD_DEPS_ASSERT */ +#endif /* ZSTD_DEPS_NEED_ASSERT */ + +/* Need: + * ZSTD_DEBUG_PRINT() + */ +#ifdef ZSTD_DEPS_NEED_IO +#ifndef ZSTD_DEPS_IO +#define ZSTD_DEPS_IO + +#include <stdio.h> +#define ZSTD_DEBUG_PRINT(...) fprintf(stderr, __VA_ARGS__) + +#endif /* ZSTD_DEPS_IO */ +#endif /* ZSTD_DEPS_NEED_IO */ + +/* Only requested when <stdint.h> is known to be present. + * Need: + * intptr_t + */ +#ifdef ZSTD_DEPS_NEED_STDINT +#ifndef ZSTD_DEPS_STDINT +#define ZSTD_DEPS_STDINT + +#include <stdint.h> + +#endif /* ZSTD_DEPS_STDINT */ +#endif /* ZSTD_DEPS_NEED_STDINT */ diff --git a/contrib/libs/zstd/lib/common/zstd_internal.h b/contrib/libs/zstd/lib/common/zstd_internal.h index 1dee37cdbe..3de9b7c694 100644 --- a/contrib/libs/zstd/lib/common/zstd_internal.h +++ b/contrib/libs/zstd/lib/common/zstd_internal.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -34,11 +34,11 @@ # define XXH_STATIC_LINKING_ONLY /* XXH64_state_t */ #endif #include <contrib/libs/xxhash/xxhash.h> /* XXH_reset, update, digest */ -#ifndef ZSTD_NO_TRACE -# include "zstd_trace.h" -#else -# define ZSTD_TRACE 0 -#endif +#ifndef ZSTD_NO_TRACE +# include "zstd_trace.h" +#else +# define ZSTD_TRACE 0 +#endif #if defined (__cplusplus) extern "C" { @@ -67,7 +67,7 @@ extern "C" { #define ZSTD_OPT_NUM (1<<12) #define ZSTD_REP_NUM 3 /* number of repcodes */ -static UNUSED_ATTR const U32 repStartValue[ZSTD_REP_NUM] = { 1, 4, 8 }; +static UNUSED_ATTR const U32 repStartValue[ZSTD_REP_NUM] = { 1, 4, 8 }; #define KB *(1 <<10) #define MB *(1 <<20) @@ -81,13 +81,13 @@ static UNUSED_ATTR const U32 repStartValue[ZSTD_REP_NUM] = { 1, 4, 8 }; #define BIT0 1 #define ZSTD_WINDOWLOG_ABSOLUTEMIN 10 -static UNUSED_ATTR const size_t ZSTD_fcs_fieldSize[4] = { 0, 2, 4, 8 }; -static UNUSED_ATTR const size_t ZSTD_did_fieldSize[4] = { 0, 1, 2, 4 }; +static UNUSED_ATTR const size_t ZSTD_fcs_fieldSize[4] = { 0, 2, 4, 8 }; +static UNUSED_ATTR const size_t ZSTD_did_fieldSize[4] = { 0, 1, 2, 4 }; #define ZSTD_FRAMEIDSIZE 4 /* magic number size */ #define ZSTD_BLOCKHEADERSIZE 3 /* C standard doesn't allow `static const` variable to be init using another `static const` variable */ -static UNUSED_ATTR const size_t ZSTD_blockHeaderSize = ZSTD_BLOCKHEADERSIZE; +static UNUSED_ATTR const size_t ZSTD_blockHeaderSize = ZSTD_BLOCKHEADERSIZE; typedef enum { bt_raw, bt_rle, bt_compressed, bt_reserved } blockType_e; #define ZSTD_FRAMECHECKSUMSIZE 4 @@ -114,56 +114,56 @@ typedef enum { set_basic, set_rle, set_compressed, set_repeat } symbolEncodingTy #define OffFSELog 8 #define MaxFSELog MAX(MAX(MLFSELog, LLFSELog), OffFSELog) -#define ZSTD_MAX_HUF_HEADER_SIZE 128 /* header + <= 127 byte tree description */ -/* Each table cannot take more than #symbols * FSELog bits */ -#define ZSTD_MAX_FSE_HEADERS_SIZE (((MaxML + 1) * MLFSELog + (MaxLL + 1) * LLFSELog + (MaxOff + 1) * OffFSELog + 7) / 8) - +#define ZSTD_MAX_HUF_HEADER_SIZE 128 /* header + <= 127 byte tree description */ +/* Each table cannot take more than #symbols * FSELog bits */ +#define ZSTD_MAX_FSE_HEADERS_SIZE (((MaxML + 1) * MLFSELog + (MaxLL + 1) * LLFSELog + (MaxOff + 1) * OffFSELog + 7) / 8) + static UNUSED_ATTR const U8 LL_bits[MaxLL+1] = { - 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, - 1, 1, 1, 1, 2, 2, 3, 3, - 4, 6, 7, 8, 9,10,11,12, - 13,14,15,16 -}; -static UNUSED_ATTR const S16 LL_defaultNorm[MaxLL+1] = { - 4, 3, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 2, 2, 1, 1, 1, - 2, 2, 2, 2, 2, 2, 2, 2, - 2, 3, 2, 1, 1, 1, 1, 1, - -1,-1,-1,-1 -}; + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 1, 1, 1, 1, 2, 2, 3, 3, + 4, 6, 7, 8, 9,10,11,12, + 13,14,15,16 +}; +static UNUSED_ATTR const S16 LL_defaultNorm[MaxLL+1] = { + 4, 3, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 1, 1, 1, + 2, 2, 2, 2, 2, 2, 2, 2, + 2, 3, 2, 1, 1, 1, 1, 1, + -1,-1,-1,-1 +}; #define LL_DEFAULTNORMLOG 6 /* for static allocation */ -static UNUSED_ATTR const U32 LL_defaultNormLog = LL_DEFAULTNORMLOG; +static UNUSED_ATTR const U32 LL_defaultNormLog = LL_DEFAULTNORMLOG; static UNUSED_ATTR const U8 ML_bits[MaxML+1] = { - 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, - 1, 1, 1, 1, 2, 2, 3, 3, - 4, 4, 5, 7, 8, 9,10,11, - 12,13,14,15,16 -}; -static UNUSED_ATTR const S16 ML_defaultNorm[MaxML+1] = { - 1, 4, 3, 2, 2, 2, 2, 2, - 2, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1,-1,-1, - -1,-1,-1,-1,-1 -}; + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 1, 1, 1, 1, 2, 2, 3, 3, + 4, 4, 5, 7, 8, 9,10,11, + 12,13,14,15,16 +}; +static UNUSED_ATTR const S16 ML_defaultNorm[MaxML+1] = { + 1, 4, 3, 2, 2, 2, 2, 2, + 2, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1,-1,-1, + -1,-1,-1,-1,-1 +}; #define ML_DEFAULTNORMLOG 6 /* for static allocation */ -static UNUSED_ATTR const U32 ML_defaultNormLog = ML_DEFAULTNORMLOG; - -static UNUSED_ATTR const S16 OF_defaultNorm[DefaultMaxOff+1] = { - 1, 1, 1, 1, 1, 1, 2, 2, - 2, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, - -1,-1,-1,-1,-1 -}; +static UNUSED_ATTR const U32 ML_defaultNormLog = ML_DEFAULTNORMLOG; + +static UNUSED_ATTR const S16 OF_defaultNorm[DefaultMaxOff+1] = { + 1, 1, 1, 1, 1, 1, 2, 2, + 2, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, + -1,-1,-1,-1,-1 +}; #define OF_DEFAULTNORMLOG 5 /* for static allocation */ -static UNUSED_ATTR const U32 OF_defaultNormLog = OF_DEFAULTNORMLOG; +static UNUSED_ATTR const U32 OF_defaultNormLog = OF_DEFAULTNORMLOG; /*-******************************************* @@ -173,7 +173,7 @@ static void ZSTD_copy8(void* dst, const void* src) { #if defined(ZSTD_ARCH_ARM_NEON) vst1_u8((uint8_t*)dst, vld1_u8((const uint8_t*)src)); #else - ZSTD_memcpy(dst, src, 8); + ZSTD_memcpy(dst, src, 8); #endif } #define COPY8(d,s) { ZSTD_copy8(d,s); d+=8; s+=8; } @@ -208,13 +208,13 @@ typedef enum { } ZSTD_overlap_e; /*! ZSTD_wildcopy() : - * Custom version of ZSTD_memcpy(), can over read/write up to WILDCOPY_OVERLENGTH bytes (if length==0) + * Custom version of ZSTD_memcpy(), can over read/write up to WILDCOPY_OVERLENGTH bytes (if length==0) * @param ovtype controls the overlap detection * - ZSTD_no_overlap: The source and destination are guaranteed to be at least WILDCOPY_VECLEN bytes apart. * - ZSTD_overlap_src_before_dst: The src and dst may overlap, but they MUST be at least 8 bytes apart. * The src buffer must be before the dst buffer. */ -MEM_STATIC FORCE_INLINE_ATTR +MEM_STATIC FORCE_INLINE_ATTR void ZSTD_wildcopy(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e const ovtype) { ptrdiff_t diff = (BYTE*)dst - (const BYTE*)src; @@ -235,16 +235,16 @@ void ZSTD_wildcopy(void* dst, const void* src, ptrdiff_t length, ZSTD_overlap_e * one COPY16() in the first call. Then, do two calls per loop since * at that point it is more likely to have a high trip count. */ -#ifdef __aarch64__ +#ifdef __aarch64__ do { COPY16(op, ip); } while (op < oend); #else - ZSTD_copy16(op, ip); - if (16 >= length) return; - op += 16; - ip += 16; + ZSTD_copy16(op, ip); + if (16 >= length) return; + op += 16; + ip += 16; do { COPY16(op, ip); COPY16(op, ip); @@ -258,7 +258,7 @@ MEM_STATIC size_t ZSTD_limitCopy(void* dst, size_t dstCapacity, const void* src, { size_t const length = MIN(dstCapacity, srcSize); if (length > 0) { - ZSTD_memcpy(dst, src, length); + ZSTD_memcpy(dst, src, length); } return length; } @@ -273,13 +273,13 @@ MEM_STATIC size_t ZSTD_limitCopy(void* dst, size_t dstCapacity, const void* src, * In which case, resize it down to free some memory */ #define ZSTD_WORKSPACETOOLARGE_MAXDURATION 128 -/* Controls whether the input/output buffer is buffered or stable. */ -typedef enum { - ZSTD_bm_buffered = 0, /* Buffer the input/output */ - ZSTD_bm_stable = 1 /* ZSTD_inBuffer/ZSTD_outBuffer is stable */ -} ZSTD_bufferMode_e; - +/* Controls whether the input/output buffer is buffered or stable. */ +typedef enum { + ZSTD_bm_buffered = 0, /* Buffer the input/output */ + ZSTD_bm_stable = 1 /* ZSTD_inBuffer/ZSTD_outBuffer is stable */ +} ZSTD_bufferMode_e; + /*-******************************************* * Private declarations *********************************************/ @@ -289,30 +289,30 @@ typedef struct seqDef_s { U16 mlBase; /* mlBase == matchLength - MINMATCH */ } seqDef; -/* Controls whether seqStore has a single "long" litLength or matchLength. See seqStore_t. */ -typedef enum { - ZSTD_llt_none = 0, /* no longLengthType */ - ZSTD_llt_literalLength = 1, /* represents a long literal */ - ZSTD_llt_matchLength = 2 /* represents a long match */ -} ZSTD_longLengthType_e; - +/* Controls whether seqStore has a single "long" litLength or matchLength. See seqStore_t. */ +typedef enum { + ZSTD_llt_none = 0, /* no longLengthType */ + ZSTD_llt_literalLength = 1, /* represents a long literal */ + ZSTD_llt_matchLength = 2 /* represents a long match */ +} ZSTD_longLengthType_e; + typedef struct { seqDef* sequencesStart; - seqDef* sequences; /* ptr to end of sequences */ + seqDef* sequences; /* ptr to end of sequences */ BYTE* litStart; - BYTE* lit; /* ptr to end of literals */ + BYTE* lit; /* ptr to end of literals */ BYTE* llCode; BYTE* mlCode; BYTE* ofCode; size_t maxNbSeq; size_t maxNbLit; - - /* longLengthPos and longLengthType to allow us to represent either a single litLength or matchLength - * in the seqStore that has a value larger than U16 (if it exists). To do so, we increment - * the existing value of the litLength or matchLength by 0x10000. - */ - ZSTD_longLengthType_e longLengthType; - U32 longLengthPos; /* Index of the sequence to apply long length modification to */ + + /* longLengthPos and longLengthType to allow us to represent either a single litLength or matchLength + * in the seqStore that has a value larger than U16 (if it exists). To do so, we increment + * the existing value of the litLength or matchLength by 0x10000. + */ + ZSTD_longLengthType_e longLengthType; + U32 longLengthPos; /* Index of the sequence to apply long length modification to */ } seqStore_t; typedef struct { @@ -322,7 +322,7 @@ typedef struct { /** * Returns the ZSTD_sequenceLength for the given sequences. It handles the decoding of long sequences - * indicated by longLengthPos and longLengthType, and adds MINMATCH back to matchLength. + * indicated by longLengthPos and longLengthType, and adds MINMATCH back to matchLength. */ MEM_STATIC ZSTD_sequenceLength ZSTD_getSequenceLength(seqStore_t const* seqStore, seqDef const* seq) { @@ -330,10 +330,10 @@ MEM_STATIC ZSTD_sequenceLength ZSTD_getSequenceLength(seqStore_t const* seqStore seqLen.litLength = seq->litLength; seqLen.matchLength = seq->mlBase + MINMATCH; if (seqStore->longLengthPos == (U32)(seq - seqStore->sequencesStart)) { - if (seqStore->longLengthType == ZSTD_llt_literalLength) { + if (seqStore->longLengthType == ZSTD_llt_literalLength) { seqLen.litLength += 0xFFFF; } - if (seqStore->longLengthType == ZSTD_llt_matchLength) { + if (seqStore->longLengthType == ZSTD_llt_matchLength) { seqLen.matchLength += 0xFFFF; } } @@ -355,9 +355,9 @@ const seqStore_t* ZSTD_getSeqStore(const ZSTD_CCtx* ctx); /* compress & dictBu void ZSTD_seqToCodes(const seqStore_t* seqStorePtr); /* compress, dictBuilder, decodeCorpus (shouldn't get its definition from here) */ /* custom memory allocation functions */ -void* ZSTD_customMalloc(size_t size, ZSTD_customMem customMem); -void* ZSTD_customCalloc(size_t size, ZSTD_customMem customMem); -void ZSTD_customFree(void* ptr, ZSTD_customMem customMem); +void* ZSTD_customMalloc(size_t size, ZSTD_customMem customMem); +void* ZSTD_customCalloc(size_t size, ZSTD_customMem customMem); +void ZSTD_customFree(void* ptr, ZSTD_customMem customMem); MEM_STATIC U32 ZSTD_highbit32(U32 val) /* compress, dictBuilder, decodeCorpus */ @@ -365,9 +365,9 @@ MEM_STATIC U32 ZSTD_highbit32(U32 val) /* compress, dictBuilder, decodeCorpus assert(val != 0); { # if defined(_MSC_VER) /* Visual */ -# if STATIC_BMI2 == 1 - return _lzcnt_u32(val)^31; -# else +# if STATIC_BMI2 == 1 + return _lzcnt_u32(val)^31; +# else if (val != 0) { unsigned long r; _BitScanReverse(&r, val); @@ -376,7 +376,7 @@ MEM_STATIC U32 ZSTD_highbit32(U32 val) /* compress, dictBuilder, decodeCorpus /* Should not reach this code path */ __assume(0); } -# endif +# endif # elif defined(__GNUC__) && (__GNUC__ >= 3) /* GCC Intrinsic */ return __builtin_clz (val) ^ 31; # elif defined(__ICCARM__) /* IAR Intrinsic */ diff --git a/contrib/libs/zstd/lib/common/zstd_trace.h b/contrib/libs/zstd/lib/common/zstd_trace.h index f9121f7d8e..8b66745493 100644 --- a/contrib/libs/zstd/lib/common/zstd_trace.h +++ b/contrib/libs/zstd/lib/common/zstd_trace.h @@ -1,22 +1,22 @@ -/* - * Copyright (c) Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_TRACE_H -#define ZSTD_TRACE_H - -#if defined (__cplusplus) -extern "C" { -#endif - -#include <stddef.h> - +/* + * Copyright (c) Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + +#ifndef ZSTD_TRACE_H +#define ZSTD_TRACE_H + +#if defined (__cplusplus) +extern "C" { +#endif + +#include <stddef.h> + /* weak symbol support * For now, enable conservatively: * - Only GNUC @@ -28,136 +28,136 @@ extern "C" { #if !defined(ZSTD_HAVE_WEAK_SYMBOLS) && \ defined(__GNUC__) && defined(__ELF__) && \ (defined(__x86_64__) || defined(_M_X64) || defined(__i386__) || defined(_M_IX86)) && \ - !defined(__APPLE__) && !defined(_WIN32) && !defined(__MINGW32__) && \ + !defined(__APPLE__) && !defined(_WIN32) && !defined(__MINGW32__) && \ !defined(__CYGWIN__) && !defined(_AIX) -# define ZSTD_HAVE_WEAK_SYMBOLS 1 -#else -# define ZSTD_HAVE_WEAK_SYMBOLS 0 -#endif -#if ZSTD_HAVE_WEAK_SYMBOLS -# define ZSTD_WEAK_ATTR __attribute__((__weak__)) -#else -# define ZSTD_WEAK_ATTR -#endif - -/* Only enable tracing when weak symbols are available. */ -#ifndef ZSTD_TRACE -# define ZSTD_TRACE ZSTD_HAVE_WEAK_SYMBOLS -#endif - -#if ZSTD_TRACE - -struct ZSTD_CCtx_s; -struct ZSTD_DCtx_s; -struct ZSTD_CCtx_params_s; - -typedef struct { - /** - * ZSTD_VERSION_NUMBER - * - * This is guaranteed to be the first member of ZSTD_trace. - * Otherwise, this struct is not stable between versions. If - * the version number does not match your expectation, you - * should not interpret the rest of the struct. - */ - unsigned version; - /** - * Non-zero if streaming (de)compression is used. - */ - unsigned streaming; - /** - * The dictionary ID. - */ - unsigned dictionaryID; - /** - * Is the dictionary cold? - * Only set on decompression. - */ - unsigned dictionaryIsCold; - /** - * The dictionary size or zero if no dictionary. - */ - size_t dictionarySize; - /** - * The uncompressed size of the data. - */ - size_t uncompressedSize; - /** - * The compressed size of the data. - */ - size_t compressedSize; - /** - * The fully resolved CCtx parameters (NULL on decompression). - */ - struct ZSTD_CCtx_params_s const* params; - /** - * The ZSTD_CCtx pointer (NULL on decompression). - */ - struct ZSTD_CCtx_s const* cctx; - /** - * The ZSTD_DCtx pointer (NULL on compression). - */ - struct ZSTD_DCtx_s const* dctx; -} ZSTD_Trace; - -/** - * A tracing context. It must be 0 when tracing is disabled. - * Otherwise, any non-zero value returned by a tracing begin() - * function is presented to any subsequent calls to end(). - * - * Any non-zero value is treated as tracing is enabled and not - * interpreted by the library. - * - * Two possible uses are: - * * A timestamp for when the begin() function was called. - * * A unique key identifying the (de)compression, like the - * address of the [dc]ctx pointer if you need to track - * more information than just a timestamp. - */ -typedef unsigned long long ZSTD_TraceCtx; - -/** - * Trace the beginning of a compression call. - * @param cctx The dctx pointer for the compression. - * It can be used as a key to map begin() to end(). - * @returns Non-zero if tracing is enabled. The return value is - * passed to ZSTD_trace_compress_end(). - */ -ZSTD_WEAK_ATTR ZSTD_TraceCtx ZSTD_trace_compress_begin( - struct ZSTD_CCtx_s const* cctx); - -/** - * Trace the end of a compression call. - * @param ctx The return value of ZSTD_trace_compress_begin(). - * @param trace The zstd tracing info. - */ -ZSTD_WEAK_ATTR void ZSTD_trace_compress_end( - ZSTD_TraceCtx ctx, - ZSTD_Trace const* trace); - -/** - * Trace the beginning of a decompression call. - * @param dctx The dctx pointer for the decompression. - * It can be used as a key to map begin() to end(). - * @returns Non-zero if tracing is enabled. The return value is - * passed to ZSTD_trace_compress_end(). - */ -ZSTD_WEAK_ATTR ZSTD_TraceCtx ZSTD_trace_decompress_begin( - struct ZSTD_DCtx_s const* dctx); - -/** - * Trace the end of a decompression call. - * @param ctx The return value of ZSTD_trace_decompress_begin(). - * @param trace The zstd tracing info. - */ -ZSTD_WEAK_ATTR void ZSTD_trace_decompress_end( - ZSTD_TraceCtx ctx, - ZSTD_Trace const* trace); - -#endif /* ZSTD_TRACE */ - -#if defined (__cplusplus) -} -#endif - -#endif /* ZSTD_TRACE_H */ +# define ZSTD_HAVE_WEAK_SYMBOLS 1 +#else +# define ZSTD_HAVE_WEAK_SYMBOLS 0 +#endif +#if ZSTD_HAVE_WEAK_SYMBOLS +# define ZSTD_WEAK_ATTR __attribute__((__weak__)) +#else +# define ZSTD_WEAK_ATTR +#endif + +/* Only enable tracing when weak symbols are available. */ +#ifndef ZSTD_TRACE +# define ZSTD_TRACE ZSTD_HAVE_WEAK_SYMBOLS +#endif + +#if ZSTD_TRACE + +struct ZSTD_CCtx_s; +struct ZSTD_DCtx_s; +struct ZSTD_CCtx_params_s; + +typedef struct { + /** + * ZSTD_VERSION_NUMBER + * + * This is guaranteed to be the first member of ZSTD_trace. + * Otherwise, this struct is not stable between versions. If + * the version number does not match your expectation, you + * should not interpret the rest of the struct. + */ + unsigned version; + /** + * Non-zero if streaming (de)compression is used. + */ + unsigned streaming; + /** + * The dictionary ID. + */ + unsigned dictionaryID; + /** + * Is the dictionary cold? + * Only set on decompression. + */ + unsigned dictionaryIsCold; + /** + * The dictionary size or zero if no dictionary. + */ + size_t dictionarySize; + /** + * The uncompressed size of the data. + */ + size_t uncompressedSize; + /** + * The compressed size of the data. + */ + size_t compressedSize; + /** + * The fully resolved CCtx parameters (NULL on decompression). + */ + struct ZSTD_CCtx_params_s const* params; + /** + * The ZSTD_CCtx pointer (NULL on decompression). + */ + struct ZSTD_CCtx_s const* cctx; + /** + * The ZSTD_DCtx pointer (NULL on compression). + */ + struct ZSTD_DCtx_s const* dctx; +} ZSTD_Trace; + +/** + * A tracing context. It must be 0 when tracing is disabled. + * Otherwise, any non-zero value returned by a tracing begin() + * function is presented to any subsequent calls to end(). + * + * Any non-zero value is treated as tracing is enabled and not + * interpreted by the library. + * + * Two possible uses are: + * * A timestamp for when the begin() function was called. + * * A unique key identifying the (de)compression, like the + * address of the [dc]ctx pointer if you need to track + * more information than just a timestamp. + */ +typedef unsigned long long ZSTD_TraceCtx; + +/** + * Trace the beginning of a compression call. + * @param cctx The dctx pointer for the compression. + * It can be used as a key to map begin() to end(). + * @returns Non-zero if tracing is enabled. The return value is + * passed to ZSTD_trace_compress_end(). + */ +ZSTD_WEAK_ATTR ZSTD_TraceCtx ZSTD_trace_compress_begin( + struct ZSTD_CCtx_s const* cctx); + +/** + * Trace the end of a compression call. + * @param ctx The return value of ZSTD_trace_compress_begin(). + * @param trace The zstd tracing info. + */ +ZSTD_WEAK_ATTR void ZSTD_trace_compress_end( + ZSTD_TraceCtx ctx, + ZSTD_Trace const* trace); + +/** + * Trace the beginning of a decompression call. + * @param dctx The dctx pointer for the decompression. + * It can be used as a key to map begin() to end(). + * @returns Non-zero if tracing is enabled. The return value is + * passed to ZSTD_trace_compress_end(). + */ +ZSTD_WEAK_ATTR ZSTD_TraceCtx ZSTD_trace_decompress_begin( + struct ZSTD_DCtx_s const* dctx); + +/** + * Trace the end of a decompression call. + * @param ctx The return value of ZSTD_trace_decompress_begin(). + * @param trace The zstd tracing info. + */ +ZSTD_WEAK_ATTR void ZSTD_trace_decompress_end( + ZSTD_TraceCtx ctx, + ZSTD_Trace const* trace); + +#endif /* ZSTD_TRACE */ + +#if defined (__cplusplus) +} +#endif + +#endif /* ZSTD_TRACE_H */ diff --git a/contrib/libs/zstd/lib/compress/fse_compress.c b/contrib/libs/zstd/lib/compress/fse_compress.c index 5547b4ac09..b7c3566ff3 100644 --- a/contrib/libs/zstd/lib/compress/fse_compress.c +++ b/contrib/libs/zstd/lib/compress/fse_compress.c @@ -1,6 +1,6 @@ /* ****************************************************************** * FSE : Finite State Entropy encoder - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * * You can contact the author at : * - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy @@ -23,9 +23,9 @@ #define FSE_STATIC_LINKING_ONLY #include "../common/fse.h" #include "../common/error_private.h" -#define ZSTD_DEPS_NEED_MALLOC -#define ZSTD_DEPS_NEED_MATH64 -#include "../common/zstd_deps.h" /* ZSTD_malloc, ZSTD_free, ZSTD_memcpy, ZSTD_memset */ +#define ZSTD_DEPS_NEED_MALLOC +#define ZSTD_DEPS_NEED_MATH64 +#include "../common/zstd_deps.h" /* ZSTD_malloc, ZSTD_free, ZSTD_memcpy, ZSTD_memset */ /* ************************************************************** @@ -79,11 +79,11 @@ size_t FSE_buildCTable_wksp(FSE_CTable* ct, U16* cumul = (U16*)workSpace; /* size = maxSV1 */ FSE_FUNCTION_TYPE* const tableSymbol = (FSE_FUNCTION_TYPE*)(cumul + (maxSV1+1)); /* size = tableSize */ - + U32 highThreshold = tableSize-1; assert(((size_t)workSpace & 1) == 0); /* Must be 2 bytes-aligned */ - if (FSE_BUILD_CTABLE_WORKSPACE_SIZE(maxSymbolValue, tableLog) > wkspSize) return ERROR(tableLog_tooLarge); + if (FSE_BUILD_CTABLE_WORKSPACE_SIZE(maxSymbolValue, tableLog) > wkspSize) return ERROR(tableLog_tooLarge); /* CTable header */ tableU16[-2] = (U16) tableLog; tableU16[-1] = (U16) maxSymbolValue; @@ -93,7 +93,7 @@ size_t FSE_buildCTable_wksp(FSE_CTable* ct, * http://fastcompression.blogspot.fr/2014/02/fse-distributing-symbol-values.html */ #ifdef __clang_analyzer__ - ZSTD_memset(tableSymbol, 0, sizeof(*tableSymbol) * tableSize); /* useless initialization, just to keep scan-build happy */ + ZSTD_memset(tableSymbol, 0, sizeof(*tableSymbol) * tableSize); /* useless initialization, just to keep scan-build happy */ #endif /* symbol start positions */ @@ -347,10 +347,10 @@ FSE_CTable* FSE_createCTable (unsigned maxSymbolValue, unsigned tableLog) size_t size; if (tableLog > FSE_TABLELOG_ABSOLUTE_MAX) tableLog = FSE_TABLELOG_ABSOLUTE_MAX; size = FSE_CTABLE_SIZE_U32 (tableLog, maxSymbolValue) * sizeof(U32); - return (FSE_CTable*)ZSTD_malloc(size); + return (FSE_CTable*)ZSTD_malloc(size); } -void FSE_freeCTable (FSE_CTable* ct) { ZSTD_free(ct); } +void FSE_freeCTable (FSE_CTable* ct) { ZSTD_free(ct); } /* provides the minimum logSize to safely represent a distribution */ static unsigned FSE_minTableLog(size_t srcSize, unsigned maxSymbolValue) @@ -384,7 +384,7 @@ unsigned FSE_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxS /* Secondary normalization method. To be used when primary method fails. */ -static size_t FSE_normalizeM2(short* norm, U32 tableLog, const unsigned* count, size_t total, U32 maxSymbolValue, short lowProbCount) +static size_t FSE_normalizeM2(short* norm, U32 tableLog, const unsigned* count, size_t total, U32 maxSymbolValue, short lowProbCount) { short const NOT_YET_ASSIGNED = -2; U32 s; @@ -401,7 +401,7 @@ static size_t FSE_normalizeM2(short* norm, U32 tableLog, const unsigned* count, continue; } if (count[s] <= lowThreshold) { - norm[s] = lowProbCount; + norm[s] = lowProbCount; distributed++; total -= count[s]; continue; @@ -453,7 +453,7 @@ static size_t FSE_normalizeM2(short* norm, U32 tableLog, const unsigned* count, { U64 const vStepLog = 62 - tableLog; U64 const mid = (1ULL << (vStepLog-1)) - 1; - U64 const rStep = ZSTD_div64((((U64)1<<vStepLog) * ToDistribute) + mid, (U32)total); /* scale on remaining */ + U64 const rStep = ZSTD_div64((((U64)1<<vStepLog) * ToDistribute) + mid, (U32)total); /* scale on remaining */ U64 tmpTotal = mid; for (s=0; s<=maxSymbolValue; s++) { if (norm[s]==NOT_YET_ASSIGNED) { @@ -472,7 +472,7 @@ static size_t FSE_normalizeM2(short* norm, U32 tableLog, const unsigned* count, size_t FSE_normalizeCount (short* normalizedCounter, unsigned tableLog, const unsigned* count, size_t total, - unsigned maxSymbolValue, unsigned useLowProbCount) + unsigned maxSymbolValue, unsigned useLowProbCount) { /* Sanity checks */ if (tableLog==0) tableLog = FSE_DEFAULT_TABLELOG; @@ -481,9 +481,9 @@ size_t FSE_normalizeCount (short* normalizedCounter, unsigned tableLog, if (tableLog < FSE_minTableLog(total, maxSymbolValue)) return ERROR(GENERIC); /* Too small tableLog, compression potentially impossible */ { static U32 const rtbTable[] = { 0, 473195, 504333, 520860, 550000, 700000, 750000, 830000 }; - short const lowProbCount = useLowProbCount ? -1 : 1; + short const lowProbCount = useLowProbCount ? -1 : 1; U64 const scale = 62 - tableLog; - U64 const step = ZSTD_div64((U64)1<<62, (U32)total); /* <== here, one division ! */ + U64 const step = ZSTD_div64((U64)1<<62, (U32)total); /* <== here, one division ! */ U64 const vStep = 1ULL<<(scale-20); int stillToDistribute = 1<<tableLog; unsigned s; @@ -495,7 +495,7 @@ size_t FSE_normalizeCount (short* normalizedCounter, unsigned tableLog, if (count[s] == total) return 0; /* rle special case */ if (count[s] == 0) { normalizedCounter[s]=0; continue; } if (count[s] <= lowThreshold) { - normalizedCounter[s] = lowProbCount; + normalizedCounter[s] = lowProbCount; stillToDistribute--; } else { short proba = (short)((count[s]*step) >> scale); @@ -509,7 +509,7 @@ size_t FSE_normalizeCount (short* normalizedCounter, unsigned tableLog, } } if (-stillToDistribute >= (normalizedCounter[largest] >> 1)) { /* corner case, need another normalization method */ - size_t const errorCode = FSE_normalizeM2(normalizedCounter, tableLog, count, total, maxSymbolValue, lowProbCount); + size_t const errorCode = FSE_normalizeM2(normalizedCounter, tableLog, count, total, maxSymbolValue, lowProbCount); if (FSE_isError(errorCode)) return errorCode; } else normalizedCounter[largest] += (short)stillToDistribute; @@ -664,7 +664,7 @@ size_t FSE_compress_usingCTable (void* dst, size_t dstSize, size_t FSE_compressBound(size_t size) { return FSE_COMPRESSBOUND(size); } -#ifndef ZSTD_NO_UNUSED_FUNCTIONS +#ifndef ZSTD_NO_UNUSED_FUNCTIONS /* FSE_compress_wksp() : * Same as FSE_compress2(), but using an externally allocated scratch buffer (`workSpace`). * `wkspSize` size must be `(1<<tableLog)`. @@ -683,7 +683,7 @@ size_t FSE_compress_wksp (void* dst, size_t dstSize, const void* src, size_t src size_t const scratchBufferSize = wkspSize - (CTableSize * sizeof(FSE_CTable)); /* init conditions */ - if (wkspSize < FSE_COMPRESS_WKSP_SIZE_U32(tableLog, maxSymbolValue)) return ERROR(tableLog_tooLarge); + if (wkspSize < FSE_COMPRESS_WKSP_SIZE_U32(tableLog, maxSymbolValue)) return ERROR(tableLog_tooLarge); if (srcSize <= 1) return 0; /* Not compressible */ if (!maxSymbolValue) maxSymbolValue = FSE_MAX_SYMBOL_VALUE; if (!tableLog) tableLog = FSE_DEFAULT_TABLELOG; @@ -696,7 +696,7 @@ size_t FSE_compress_wksp (void* dst, size_t dstSize, const void* src, size_t src } tableLog = FSE_optimalTableLog(tableLog, srcSize, maxSymbolValue); - CHECK_F( FSE_normalizeCount(norm, tableLog, count, srcSize, maxSymbolValue, /* useLowProbCount */ srcSize >= 2048) ); + CHECK_F( FSE_normalizeCount(norm, tableLog, count, srcSize, maxSymbolValue, /* useLowProbCount */ srcSize >= 2048) ); /* Write table description header */ { CHECK_V_F(nc_err, FSE_writeNCount(op, oend-op, norm, maxSymbolValue, tableLog) ); @@ -718,16 +718,16 @@ size_t FSE_compress_wksp (void* dst, size_t dstSize, const void* src, size_t src typedef struct { FSE_CTable CTable_max[FSE_CTABLE_SIZE_U32(FSE_MAX_TABLELOG, FSE_MAX_SYMBOL_VALUE)]; - union { - U32 hist_wksp[HIST_WKSP_SIZE_U32]; - BYTE scratchBuffer[1 << FSE_MAX_TABLELOG]; - } workspace; + union { + U32 hist_wksp[HIST_WKSP_SIZE_U32]; + BYTE scratchBuffer[1 << FSE_MAX_TABLELOG]; + } workspace; } fseWkspMax_t; size_t FSE_compress2 (void* dst, size_t dstCapacity, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog) { fseWkspMax_t scratchBuffer; - DEBUG_STATIC_ASSERT(sizeof(scratchBuffer) >= FSE_COMPRESS_WKSP_SIZE_U32(FSE_MAX_TABLELOG, FSE_MAX_SYMBOL_VALUE)); /* compilation failures here means scratchBuffer is not large enough */ + DEBUG_STATIC_ASSERT(sizeof(scratchBuffer) >= FSE_COMPRESS_WKSP_SIZE_U32(FSE_MAX_TABLELOG, FSE_MAX_SYMBOL_VALUE)); /* compilation failures here means scratchBuffer is not large enough */ if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge); return FSE_compress_wksp(dst, dstCapacity, src, srcSize, maxSymbolValue, tableLog, &scratchBuffer, sizeof(scratchBuffer)); } @@ -736,6 +736,6 @@ size_t FSE_compress (void* dst, size_t dstCapacity, const void* src, size_t srcS { return FSE_compress2(dst, dstCapacity, src, srcSize, FSE_MAX_SYMBOL_VALUE, FSE_DEFAULT_TABLELOG); } -#endif +#endif #endif /* FSE_COMMONDEFS_ONLY */ diff --git a/contrib/libs/zstd/lib/compress/hist.c b/contrib/libs/zstd/lib/compress/hist.c index 073c57e752..c84db44368 100644 --- a/contrib/libs/zstd/lib/compress/hist.c +++ b/contrib/libs/zstd/lib/compress/hist.c @@ -1,7 +1,7 @@ /* ****************************************************************** * hist : Histogram functions * part of Finite State Entropy project - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * * You can contact the author at : * - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy @@ -34,7 +34,7 @@ unsigned HIST_count_simple(unsigned* count, unsigned* maxSymbolValuePtr, unsigned maxSymbolValue = *maxSymbolValuePtr; unsigned largestCount=0; - ZSTD_memset(count, 0, (maxSymbolValue+1) * sizeof(*count)); + ZSTD_memset(count, 0, (maxSymbolValue+1) * sizeof(*count)); if (srcSize==0) { *maxSymbolValuePtr = 0; return 0; } while (ip<end) { @@ -60,9 +60,9 @@ typedef enum { trustInput, checkMaxSymbolValue } HIST_checkInput_e; * this design makes better use of OoO cpus, * and is noticeably faster when some values are heavily repeated. * But it needs some additional workspace for intermediate tables. - * `workSpace` must be a U32 table of size >= HIST_WKSP_SIZE_U32. + * `workSpace` must be a U32 table of size >= HIST_WKSP_SIZE_U32. * @return : largest histogram frequency, - * or an error code (notably when histogram's alphabet is larger than *maxSymbolValuePtr) */ + * or an error code (notably when histogram's alphabet is larger than *maxSymbolValuePtr) */ static size_t HIST_count_parallel_wksp( unsigned* count, unsigned* maxSymbolValuePtr, const void* source, size_t sourceSize, @@ -71,7 +71,7 @@ static size_t HIST_count_parallel_wksp( { const BYTE* ip = (const BYTE*)source; const BYTE* const iend = ip+sourceSize; - size_t const countSize = (*maxSymbolValuePtr + 1) * sizeof(*count); + size_t const countSize = (*maxSymbolValuePtr + 1) * sizeof(*count); unsigned max=0; U32* const Counting1 = workSpace; U32* const Counting2 = Counting1 + 256; @@ -79,13 +79,13 @@ static size_t HIST_count_parallel_wksp( U32* const Counting4 = Counting3 + 256; /* safety checks */ - assert(*maxSymbolValuePtr <= 255); + assert(*maxSymbolValuePtr <= 255); if (!sourceSize) { - ZSTD_memset(count, 0, countSize); + ZSTD_memset(count, 0, countSize); *maxSymbolValuePtr = 0; return 0; } - ZSTD_memset(workSpace, 0, 4*256*sizeof(unsigned)); + ZSTD_memset(workSpace, 0, 4*256*sizeof(unsigned)); /* by stripes of 16 bytes */ { U32 cached = MEM_read32(ip); ip += 4; @@ -117,18 +117,18 @@ static size_t HIST_count_parallel_wksp( /* finish last symbols */ while (ip<iend) Counting1[*ip++]++; - { U32 s; - for (s=0; s<256; s++) { + { U32 s; + for (s=0; s<256; s++) { Counting1[s] += Counting2[s] + Counting3[s] + Counting4[s]; - if (Counting1[s] > max) max = Counting1[s]; + if (Counting1[s] > max) max = Counting1[s]; } } - { unsigned maxSymbolValue = 255; - while (!Counting1[maxSymbolValue]) maxSymbolValue--; - if (check && maxSymbolValue > *maxSymbolValuePtr) return ERROR(maxSymbolValue_tooSmall); - *maxSymbolValuePtr = maxSymbolValue; - ZSTD_memmove(count, Counting1, countSize); /* in case count & Counting1 are overlapping */ - } + { unsigned maxSymbolValue = 255; + while (!Counting1[maxSymbolValue]) maxSymbolValue--; + if (check && maxSymbolValue > *maxSymbolValuePtr) return ERROR(maxSymbolValue_tooSmall); + *maxSymbolValuePtr = maxSymbolValue; + ZSTD_memmove(count, Counting1, countSize); /* in case count & Counting1 are overlapping */ + } return (size_t)max; } @@ -163,19 +163,19 @@ size_t HIST_count_wksp(unsigned* count, unsigned* maxSymbolValuePtr, return HIST_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, workSpace, workSpaceSize); } -#ifndef ZSTD_NO_UNUSED_FUNCTIONS -/* fast variant (unsafe : won't check if src contains values beyond count[] limit) */ -size_t HIST_countFast(unsigned* count, unsigned* maxSymbolValuePtr, - const void* source, size_t sourceSize) -{ - unsigned tmpCounters[HIST_WKSP_SIZE_U32]; - return HIST_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, tmpCounters, sizeof(tmpCounters)); -} - +#ifndef ZSTD_NO_UNUSED_FUNCTIONS +/* fast variant (unsafe : won't check if src contains values beyond count[] limit) */ +size_t HIST_countFast(unsigned* count, unsigned* maxSymbolValuePtr, + const void* source, size_t sourceSize) +{ + unsigned tmpCounters[HIST_WKSP_SIZE_U32]; + return HIST_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, tmpCounters, sizeof(tmpCounters)); +} + size_t HIST_count(unsigned* count, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize) { unsigned tmpCounters[HIST_WKSP_SIZE_U32]; return HIST_count_wksp(count, maxSymbolValuePtr, src, srcSize, tmpCounters, sizeof(tmpCounters)); } -#endif +#endif diff --git a/contrib/libs/zstd/lib/compress/hist.h b/contrib/libs/zstd/lib/compress/hist.h index 228ed48a71..0be288463c 100644 --- a/contrib/libs/zstd/lib/compress/hist.h +++ b/contrib/libs/zstd/lib/compress/hist.h @@ -1,7 +1,7 @@ /* ****************************************************************** * hist : Histogram functions * part of Finite State Entropy project - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * * You can contact the author at : * - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy @@ -14,7 +14,7 @@ ****************************************************************** */ /* --- dependencies --- */ -#include "../common/zstd_deps.h" /* size_t */ +#include "../common/zstd_deps.h" /* size_t */ /* --- simple histogram functions --- */ diff --git a/contrib/libs/zstd/lib/compress/huf_compress.c b/contrib/libs/zstd/lib/compress/huf_compress.c index 2b3d6adc2a..85209eb02f 100644 --- a/contrib/libs/zstd/lib/compress/huf_compress.c +++ b/contrib/libs/zstd/lib/compress/huf_compress.c @@ -1,6 +1,6 @@ /* ****************************************************************** * Huffman encoder, part of New Generation Entropy library - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * * You can contact the author at : * - FSE+HUF source repository : https://github.com/Cyan4973/FiniteStateEntropy @@ -23,7 +23,7 @@ /* ************************************************************** * Includes ****************************************************************/ -#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memset */ +#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memset */ #include "../common/compiler.h" #include "../common/bitstream.h" #include "hist.h" @@ -81,15 +81,15 @@ static void* HUF_alignUpWorkspace(void* workspace, size_t* workspaceSizePtr, siz * Note : all elements within weightTable are supposed to be <= HUF_TABLELOG_MAX. */ #define MAX_FSE_TABLELOG_FOR_HUFF_HEADER 6 - -typedef struct { - FSE_CTable CTable[FSE_CTABLE_SIZE_U32(MAX_FSE_TABLELOG_FOR_HUFF_HEADER, HUF_TABLELOG_MAX)]; - U32 scratchBuffer[FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(HUF_TABLELOG_MAX, MAX_FSE_TABLELOG_FOR_HUFF_HEADER)]; - unsigned count[HUF_TABLELOG_MAX+1]; - S16 norm[HUF_TABLELOG_MAX+1]; -} HUF_CompressWeightsWksp; - -static size_t HUF_compressWeights(void* dst, size_t dstSize, const void* weightTable, size_t wtSize, void* workspace, size_t workspaceSize) + +typedef struct { + FSE_CTable CTable[FSE_CTABLE_SIZE_U32(MAX_FSE_TABLELOG_FOR_HUFF_HEADER, HUF_TABLELOG_MAX)]; + U32 scratchBuffer[FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(HUF_TABLELOG_MAX, MAX_FSE_TABLELOG_FOR_HUFF_HEADER)]; + unsigned count[HUF_TABLELOG_MAX+1]; + S16 norm[HUF_TABLELOG_MAX+1]; +} HUF_CompressWeightsWksp; + +static size_t HUF_compressWeights(void* dst, size_t dstSize, const void* weightTable, size_t wtSize, void* workspace, size_t workspaceSize) { BYTE* const ostart = (BYTE*) dst; BYTE* op = ostart; @@ -99,28 +99,28 @@ static size_t HUF_compressWeights(void* dst, size_t dstSize, const void* weightT U32 tableLog = MAX_FSE_TABLELOG_FOR_HUFF_HEADER; HUF_CompressWeightsWksp* wksp = (HUF_CompressWeightsWksp*)HUF_alignUpWorkspace(workspace, &workspaceSize, ZSTD_ALIGNOF(U32)); - if (workspaceSize < sizeof(HUF_CompressWeightsWksp)) return ERROR(GENERIC); + if (workspaceSize < sizeof(HUF_CompressWeightsWksp)) return ERROR(GENERIC); /* init conditions */ if (wtSize <= 1) return 0; /* Not compressible */ /* Scan input and build symbol stats */ - { unsigned const maxCount = HIST_count_simple(wksp->count, &maxSymbolValue, weightTable, wtSize); /* never fails */ + { unsigned const maxCount = HIST_count_simple(wksp->count, &maxSymbolValue, weightTable, wtSize); /* never fails */ if (maxCount == wtSize) return 1; /* only a single symbol in src : rle */ if (maxCount == 1) return 0; /* each symbol present maximum once => not compressible */ } tableLog = FSE_optimalTableLog(tableLog, wtSize, maxSymbolValue); - CHECK_F( FSE_normalizeCount(wksp->norm, tableLog, wksp->count, wtSize, maxSymbolValue, /* useLowProbCount */ 0) ); + CHECK_F( FSE_normalizeCount(wksp->norm, tableLog, wksp->count, wtSize, maxSymbolValue, /* useLowProbCount */ 0) ); /* Write table description header */ - { CHECK_V_F(hSize, FSE_writeNCount(op, (size_t)(oend-op), wksp->norm, maxSymbolValue, tableLog) ); + { CHECK_V_F(hSize, FSE_writeNCount(op, (size_t)(oend-op), wksp->norm, maxSymbolValue, tableLog) ); op += hSize; } /* Compress */ - CHECK_F( FSE_buildCTable_wksp(wksp->CTable, wksp->norm, maxSymbolValue, tableLog, wksp->scratchBuffer, sizeof(wksp->scratchBuffer)) ); - { CHECK_V_F(cSize, FSE_compress_usingCTable(op, (size_t)(oend - op), weightTable, wtSize, wksp->CTable) ); + CHECK_F( FSE_buildCTable_wksp(wksp->CTable, wksp->norm, maxSymbolValue, tableLog, wksp->scratchBuffer, sizeof(wksp->scratchBuffer)) ); + { CHECK_V_F(cSize, FSE_compress_usingCTable(op, (size_t)(oend - op), weightTable, wtSize, wksp->CTable) ); if (cSize == 0) return 0; /* not enough space for compressed data */ op += cSize; } @@ -163,35 +163,35 @@ static void HUF_setValue(HUF_CElt* elt, size_t value) } } -typedef struct { - HUF_CompressWeightsWksp wksp; +typedef struct { + HUF_CompressWeightsWksp wksp; BYTE bitsToWeight[HUF_TABLELOG_MAX + 1]; /* precomputed conversion table */ BYTE huffWeight[HUF_SYMBOLVALUE_MAX]; -} HUF_WriteCTableWksp; - -size_t HUF_writeCTable_wksp(void* dst, size_t maxDstSize, - const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog, - void* workspace, size_t workspaceSize) -{ +} HUF_WriteCTableWksp; + +size_t HUF_writeCTable_wksp(void* dst, size_t maxDstSize, + const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog, + void* workspace, size_t workspaceSize) +{ HUF_CElt const* const ct = CTable + 1; BYTE* op = (BYTE*)dst; U32 n; HUF_WriteCTableWksp* wksp = (HUF_WriteCTableWksp*)HUF_alignUpWorkspace(workspace, &workspaceSize, ZSTD_ALIGNOF(U32)); - /* check conditions */ - if (workspaceSize < sizeof(HUF_WriteCTableWksp)) return ERROR(GENERIC); + /* check conditions */ + if (workspaceSize < sizeof(HUF_WriteCTableWksp)) return ERROR(GENERIC); if (maxSymbolValue > HUF_SYMBOLVALUE_MAX) return ERROR(maxSymbolValue_tooLarge); /* convert to weight */ - wksp->bitsToWeight[0] = 0; + wksp->bitsToWeight[0] = 0; for (n=1; n<huffLog+1; n++) - wksp->bitsToWeight[n] = (BYTE)(huffLog + 1 - n); + wksp->bitsToWeight[n] = (BYTE)(huffLog + 1 - n); for (n=0; n<maxSymbolValue; n++) wksp->huffWeight[n] = wksp->bitsToWeight[HUF_getNbBits(ct[n])]; /* attempt weights compression by FSE */ if (maxDstSize < 1) return ERROR(dstSize_tooSmall); - { CHECK_V_F(hSize, HUF_compressWeights(op+1, maxDstSize-1, wksp->huffWeight, maxSymbolValue, &wksp->wksp, sizeof(wksp->wksp)) ); + { CHECK_V_F(hSize, HUF_compressWeights(op+1, maxDstSize-1, wksp->huffWeight, maxSymbolValue, &wksp->wksp, sizeof(wksp->wksp)) ); if ((hSize>1) & (hSize < maxSymbolValue/2)) { /* FSE compressed */ op[0] = (BYTE)hSize; return hSize+1; @@ -201,23 +201,23 @@ size_t HUF_writeCTable_wksp(void* dst, size_t maxDstSize, if (maxSymbolValue > (256-128)) return ERROR(GENERIC); /* should not happen : likely means source cannot be compressed */ if (((maxSymbolValue+1)/2) + 1 > maxDstSize) return ERROR(dstSize_tooSmall); /* not enough space within dst buffer */ op[0] = (BYTE)(128 /*special case*/ + (maxSymbolValue-1)); - wksp->huffWeight[maxSymbolValue] = 0; /* to be sure it doesn't cause msan issue in final combination */ + wksp->huffWeight[maxSymbolValue] = 0; /* to be sure it doesn't cause msan issue in final combination */ for (n=0; n<maxSymbolValue; n+=2) - op[(n/2)+1] = (BYTE)((wksp->huffWeight[n] << 4) + wksp->huffWeight[n+1]); + op[(n/2)+1] = (BYTE)((wksp->huffWeight[n] << 4) + wksp->huffWeight[n+1]); return ((maxSymbolValue+1)/2) + 1; } -/*! HUF_writeCTable() : - `CTable` : Huffman tree to save, using huf representation. - @return : size of saved CTable */ -size_t HUF_writeCTable (void* dst, size_t maxDstSize, - const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog) -{ - HUF_WriteCTableWksp wksp; - return HUF_writeCTable_wksp(dst, maxDstSize, CTable, maxSymbolValue, huffLog, &wksp, sizeof(wksp)); -} - - +/*! HUF_writeCTable() : + `CTable` : Huffman tree to save, using huf representation. + @return : size of saved CTable */ +size_t HUF_writeCTable (void* dst, size_t maxDstSize, + const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog) +{ + HUF_WriteCTableWksp wksp; + return HUF_writeCTable_wksp(dst, maxDstSize, CTable, maxSymbolValue, huffLog, &wksp, sizeof(wksp)); +} + + size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize, unsigned* hasZeroWeights) { BYTE huffWeight[HUF_SYMBOLVALUE_MAX + 1]; /* init not required, even though some static analyzer may complain */ @@ -228,7 +228,7 @@ size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void /* get symbol weights */ CHECK_V_F(readSize, HUF_readStats(huffWeight, HUF_SYMBOLVALUE_MAX+1, rankVal, &nbSymbols, &tableLog, src, srcSize)); - *hasZeroWeights = (rankVal[0] > 0); + *hasZeroWeights = (rankVal[0] > 0); /* check result */ if (tableLog > HUF_TABLELOG_MAX) return ERROR(tableLog_tooLarge); @@ -239,9 +239,9 @@ size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void /* Prepare base value per rank */ { U32 n, nextRankStart = 0; for (n=1; n<=tableLog; n++) { - U32 curr = nextRankStart; + U32 curr = nextRankStart; nextRankStart += (rankVal[n] << (n-1)); - rankVal[n] = curr; + rankVal[n] = curr; } } /* fill nbBits */ @@ -285,63 +285,63 @@ typedef struct nodeElt_s { BYTE nbBits; } nodeElt; -/** - * HUF_setMaxHeight(): - * Enforces maxNbBits on the Huffman tree described in huffNode. - * - * It sets all nodes with nbBits > maxNbBits to be maxNbBits. Then it adjusts - * the tree to so that it is a valid canonical Huffman tree. - * - * @pre The sum of the ranks of each symbol == 2^largestBits, - * where largestBits == huffNode[lastNonNull].nbBits. - * @post The sum of the ranks of each symbol == 2^largestBits, - * where largestBits is the return value <= maxNbBits. - * - * @param huffNode The Huffman tree modified in place to enforce maxNbBits. - * @param lastNonNull The symbol with the lowest count in the Huffman tree. - * @param maxNbBits The maximum allowed number of bits, which the Huffman tree - * may not respect. After this function the Huffman tree will - * respect maxNbBits. - * @return The maximum number of bits of the Huffman tree after adjustment, - * necessarily no more than maxNbBits. - */ +/** + * HUF_setMaxHeight(): + * Enforces maxNbBits on the Huffman tree described in huffNode. + * + * It sets all nodes with nbBits > maxNbBits to be maxNbBits. Then it adjusts + * the tree to so that it is a valid canonical Huffman tree. + * + * @pre The sum of the ranks of each symbol == 2^largestBits, + * where largestBits == huffNode[lastNonNull].nbBits. + * @post The sum of the ranks of each symbol == 2^largestBits, + * where largestBits is the return value <= maxNbBits. + * + * @param huffNode The Huffman tree modified in place to enforce maxNbBits. + * @param lastNonNull The symbol with the lowest count in the Huffman tree. + * @param maxNbBits The maximum allowed number of bits, which the Huffman tree + * may not respect. After this function the Huffman tree will + * respect maxNbBits. + * @return The maximum number of bits of the Huffman tree after adjustment, + * necessarily no more than maxNbBits. + */ static U32 HUF_setMaxHeight(nodeElt* huffNode, U32 lastNonNull, U32 maxNbBits) { const U32 largestBits = huffNode[lastNonNull].nbBits; - /* early exit : no elt > maxNbBits, so the tree is already valid. */ - if (largestBits <= maxNbBits) return largestBits; + /* early exit : no elt > maxNbBits, so the tree is already valid. */ + if (largestBits <= maxNbBits) return largestBits; /* there are several too large elements (at least >= 2) */ { int totalCost = 0; const U32 baseCost = 1 << (largestBits - maxNbBits); int n = (int)lastNonNull; - /* Adjust any ranks > maxNbBits to maxNbBits. - * Compute totalCost, which is how far the sum of the ranks is - * we are over 2^largestBits after adjust the offending ranks. - */ + /* Adjust any ranks > maxNbBits to maxNbBits. + * Compute totalCost, which is how far the sum of the ranks is + * we are over 2^largestBits after adjust the offending ranks. + */ while (huffNode[n].nbBits > maxNbBits) { totalCost += baseCost - (1 << (largestBits - huffNode[n].nbBits)); huffNode[n].nbBits = (BYTE)maxNbBits; - n--; - } - /* n stops at huffNode[n].nbBits <= maxNbBits */ - assert(huffNode[n].nbBits <= maxNbBits); - /* n end at index of smallest symbol using < maxNbBits */ - while (huffNode[n].nbBits == maxNbBits) --n; - - /* renorm totalCost from 2^largestBits to 2^maxNbBits - * note : totalCost is necessarily a multiple of baseCost */ - assert((totalCost & (baseCost - 1)) == 0); - totalCost >>= (largestBits - maxNbBits); - assert(totalCost > 0); + n--; + } + /* n stops at huffNode[n].nbBits <= maxNbBits */ + assert(huffNode[n].nbBits <= maxNbBits); + /* n end at index of smallest symbol using < maxNbBits */ + while (huffNode[n].nbBits == maxNbBits) --n; + + /* renorm totalCost from 2^largestBits to 2^maxNbBits + * note : totalCost is necessarily a multiple of baseCost */ + assert((totalCost & (baseCost - 1)) == 0); + totalCost >>= (largestBits - maxNbBits); + assert(totalCost > 0); /* repay normalized cost */ { U32 const noSymbol = 0xF0F0F0F0; U32 rankLast[HUF_TABLELOG_MAX+2]; - /* Get pos of last (smallest = lowest cum. count) symbol per rank */ - ZSTD_memset(rankLast, 0xF0, sizeof(rankLast)); + /* Get pos of last (smallest = lowest cum. count) symbol per rank */ + ZSTD_memset(rankLast, 0xF0, sizeof(rankLast)); { U32 currentNbBits = maxNbBits; int pos; for (pos=n ; pos >= 0; pos--) { @@ -351,65 +351,65 @@ static U32 HUF_setMaxHeight(nodeElt* huffNode, U32 lastNonNull, U32 maxNbBits) } } while (totalCost > 0) { - /* Try to reduce the next power of 2 above totalCost because we - * gain back half the rank. - */ + /* Try to reduce the next power of 2 above totalCost because we + * gain back half the rank. + */ U32 nBitsToDecrease = BIT_highbit32((U32)totalCost) + 1; for ( ; nBitsToDecrease > 1; nBitsToDecrease--) { U32 const highPos = rankLast[nBitsToDecrease]; U32 const lowPos = rankLast[nBitsToDecrease-1]; if (highPos == noSymbol) continue; - /* Decrease highPos if no symbols of lowPos or if it is - * not cheaper to remove 2 lowPos than highPos. - */ + /* Decrease highPos if no symbols of lowPos or if it is + * not cheaper to remove 2 lowPos than highPos. + */ if (lowPos == noSymbol) break; { U32 const highTotal = huffNode[highPos].count; U32 const lowTotal = 2 * huffNode[lowPos].count; if (highTotal <= lowTotal) break; } } /* only triggered when no more rank 1 symbol left => find closest one (note : there is necessarily at least one !) */ - assert(rankLast[nBitsToDecrease] != noSymbol || nBitsToDecrease == 1); + assert(rankLast[nBitsToDecrease] != noSymbol || nBitsToDecrease == 1); /* HUF_MAX_TABLELOG test just to please gcc 5+; but it should not be necessary */ while ((nBitsToDecrease<=HUF_TABLELOG_MAX) && (rankLast[nBitsToDecrease] == noSymbol)) - nBitsToDecrease++; - assert(rankLast[nBitsToDecrease] != noSymbol); - /* Increase the number of bits to gain back half the rank cost. */ + nBitsToDecrease++; + assert(rankLast[nBitsToDecrease] != noSymbol); + /* Increase the number of bits to gain back half the rank cost. */ totalCost -= 1 << (nBitsToDecrease-1); - huffNode[rankLast[nBitsToDecrease]].nbBits++; - - /* Fix up the new rank. - * If the new rank was empty, this symbol is now its smallest. - * Otherwise, this symbol will be the largest in the new rank so no adjustment. - */ + huffNode[rankLast[nBitsToDecrease]].nbBits++; + + /* Fix up the new rank. + * If the new rank was empty, this symbol is now its smallest. + * Otherwise, this symbol will be the largest in the new rank so no adjustment. + */ if (rankLast[nBitsToDecrease-1] == noSymbol) - rankLast[nBitsToDecrease-1] = rankLast[nBitsToDecrease]; - /* Fix up the old rank. - * If the symbol was at position 0, meaning it was the highest weight symbol in the tree, - * it must be the only symbol in its rank, so the old rank now has no symbols. - * Otherwise, since the Huffman nodes are sorted by count, the previous position is now - * the smallest node in the rank. If the previous position belongs to a different rank, - * then the rank is now empty. - */ + rankLast[nBitsToDecrease-1] = rankLast[nBitsToDecrease]; + /* Fix up the old rank. + * If the symbol was at position 0, meaning it was the highest weight symbol in the tree, + * it must be the only symbol in its rank, so the old rank now has no symbols. + * Otherwise, since the Huffman nodes are sorted by count, the previous position is now + * the smallest node in the rank. If the previous position belongs to a different rank, + * then the rank is now empty. + */ if (rankLast[nBitsToDecrease] == 0) /* special case, reached largest symbol */ rankLast[nBitsToDecrease] = noSymbol; else { rankLast[nBitsToDecrease]--; if (huffNode[rankLast[nBitsToDecrease]].nbBits != maxNbBits-nBitsToDecrease) rankLast[nBitsToDecrease] = noSymbol; /* this rank is now empty */ - } - } /* while (totalCost > 0) */ - - /* If we've removed too much weight, then we have to add it back. - * To avoid overshooting again, we only adjust the smallest rank. - * We take the largest nodes from the lowest rank 0 and move them - * to rank 1. There's guaranteed to be enough rank 0 symbols because - * TODO. - */ + } + } /* while (totalCost > 0) */ + + /* If we've removed too much weight, then we have to add it back. + * To avoid overshooting again, we only adjust the smallest rank. + * We take the largest nodes from the lowest rank 0 and move them + * to rank 1. There's guaranteed to be enough rank 0 symbols because + * TODO. + */ while (totalCost < 0) { /* Sometimes, cost correction overshoot */ - /* special case : no rank 1 symbol (using maxNbBits-1); - * let's create one from largest rank 0 (using maxNbBits). - */ - if (rankLast[1] == noSymbol) { + /* special case : no rank 1 symbol (using maxNbBits-1); + * let's create one from largest rank 0 (using maxNbBits). + */ + if (rankLast[1] == noSymbol) { while (huffNode[n].nbBits == maxNbBits) n--; huffNode[n+1].nbBits--; assert(n >= 0); @@ -420,9 +420,9 @@ static U32 HUF_setMaxHeight(nodeElt* huffNode, U32 lastNonNull, U32 maxNbBits) huffNode[ rankLast[1] + 1 ].nbBits--; rankLast[1]++; totalCost ++; - } - } /* repay normalized cost */ - } /* there are several too large elements (at least >= 2) */ + } + } /* repay normalized cost */ + } /* there are several too large elements (at least >= 2) */ return maxNbBits; } @@ -536,43 +536,43 @@ static void HUF_simpleQuickSort(nodeElt arr[], int low, int high) { } } -/** - * HUF_sort(): - * Sorts the symbols [0, maxSymbolValue] by count[symbol] in decreasing order. +/** + * HUF_sort(): + * Sorts the symbols [0, maxSymbolValue] by count[symbol] in decreasing order. * This is a typical bucket sorting strategy that uses either quicksort or insertion sort to sort each bucket. - * - * @param[out] huffNode Sorted symbols by decreasing count. Only members `.count` and `.byte` are filled. - * Must have (maxSymbolValue + 1) entries. - * @param[in] count Histogram of the symbols. - * @param[in] maxSymbolValue Maximum symbol value. - * @param rankPosition This is a scratch workspace. Must have RANK_POSITION_TABLE_SIZE entries. - */ + * + * @param[out] huffNode Sorted symbols by decreasing count. Only members `.count` and `.byte` are filled. + * Must have (maxSymbolValue + 1) entries. + * @param[in] count Histogram of the symbols. + * @param[in] maxSymbolValue Maximum symbol value. + * @param rankPosition This is a scratch workspace. Must have RANK_POSITION_TABLE_SIZE entries. + */ static void HUF_sort(nodeElt huffNode[], const unsigned count[], U32 const maxSymbolValue, rankPos rankPosition[]) { U32 n; U32 const maxSymbolValue1 = maxSymbolValue+1; - /* Compute base and set curr to base. + /* Compute base and set curr to base. * For symbol s let lowerRank = HUF_getIndex(count[n]) and rank = lowerRank + 1. * See HUF_getIndex to see bucketing strategy. - * We attribute each symbol to lowerRank's base value, because we want to know where - * each rank begins in the output, so for rank R we want to count ranks R+1 and above. - */ - ZSTD_memset(rankPosition, 0, sizeof(*rankPosition) * RANK_POSITION_TABLE_SIZE); - for (n = 0; n < maxSymbolValue1; ++n) { + * We attribute each symbol to lowerRank's base value, because we want to know where + * each rank begins in the output, so for rank R we want to count ranks R+1 and above. + */ + ZSTD_memset(rankPosition, 0, sizeof(*rankPosition) * RANK_POSITION_TABLE_SIZE); + for (n = 0; n < maxSymbolValue1; ++n) { U32 lowerRank = HUF_getIndex(count[n]); assert(lowerRank < RANK_POSITION_TABLE_SIZE - 1); - rankPosition[lowerRank].base++; + rankPosition[lowerRank].base++; } - assert(rankPosition[RANK_POSITION_TABLE_SIZE - 1].base == 0); + assert(rankPosition[RANK_POSITION_TABLE_SIZE - 1].base == 0); /* Set up the rankPosition table */ - for (n = RANK_POSITION_TABLE_SIZE - 1; n > 0; --n) { - rankPosition[n-1].base += rankPosition[n].base; - rankPosition[n-1].curr = rankPosition[n-1].base; - } + for (n = RANK_POSITION_TABLE_SIZE - 1; n > 0; --n) { + rankPosition[n-1].base += rankPosition[n].base; + rankPosition[n-1].curr = rankPosition[n-1].base; + } /* Insert each symbol into their appropriate bucket, setting up rankPosition table. */ - for (n = 0; n < maxSymbolValue1; ++n) { + for (n = 0; n < maxSymbolValue1; ++n) { U32 const c = count[n]; U32 const r = HUF_getIndex(c) + 1; U32 const pos = rankPosition[r].curr++; @@ -600,16 +600,16 @@ static void HUF_sort(nodeElt huffNode[], const unsigned count[], U32 const maxSy */ #define STARTNODE (HUF_SYMBOLVALUE_MAX+1) -/* HUF_buildTree(): - * Takes the huffNode array sorted by HUF_sort() and builds an unlimited-depth Huffman tree. - * - * @param huffNode The array sorted by HUF_sort(). Builds the Huffman tree in this array. - * @param maxSymbolValue The maximum symbol value. - * @return The smallest node in the Huffman tree (by count). - */ -static int HUF_buildTree(nodeElt* huffNode, U32 maxSymbolValue) +/* HUF_buildTree(): + * Takes the huffNode array sorted by HUF_sort() and builds an unlimited-depth Huffman tree. + * + * @param huffNode The array sorted by HUF_sort(). Builds the Huffman tree in this array. + * @param maxSymbolValue The maximum symbol value. + * @return The smallest node in the Huffman tree (by count). + */ +static int HUF_buildTree(nodeElt* huffNode, U32 maxSymbolValue) { - nodeElt* const huffNode0 = huffNode - 1; + nodeElt* const huffNode0 = huffNode - 1; int nonNullRank; int lowS, lowN; int nodeNb = STARTNODE; @@ -640,67 +640,67 @@ static int HUF_buildTree(nodeElt* huffNode, U32 maxSymbolValue) for (n=0; n<=nonNullRank; n++) huffNode[n].nbBits = huffNode[ huffNode[n].parent ].nbBits + 1; - return nonNullRank; -} - -/** - * HUF_buildCTableFromTree(): - * Build the CTable given the Huffman tree in huffNode. - * - * @param[out] CTable The output Huffman CTable. - * @param huffNode The Huffman tree. - * @param nonNullRank The last and smallest node in the Huffman tree. - * @param maxSymbolValue The maximum symbol value. - * @param maxNbBits The exact maximum number of bits used in the Huffman tree. - */ -static void HUF_buildCTableFromTree(HUF_CElt* CTable, nodeElt const* huffNode, int nonNullRank, U32 maxSymbolValue, U32 maxNbBits) -{ + return nonNullRank; +} + +/** + * HUF_buildCTableFromTree(): + * Build the CTable given the Huffman tree in huffNode. + * + * @param[out] CTable The output Huffman CTable. + * @param huffNode The Huffman tree. + * @param nonNullRank The last and smallest node in the Huffman tree. + * @param maxSymbolValue The maximum symbol value. + * @param maxNbBits The exact maximum number of bits used in the Huffman tree. + */ +static void HUF_buildCTableFromTree(HUF_CElt* CTable, nodeElt const* huffNode, int nonNullRank, U32 maxSymbolValue, U32 maxNbBits) +{ HUF_CElt* const ct = CTable + 1; - /* fill result into ctable (val, nbBits) */ - int n; - U16 nbPerRank[HUF_TABLELOG_MAX+1] = {0}; - U16 valPerRank[HUF_TABLELOG_MAX+1] = {0}; - int const alphabetSize = (int)(maxSymbolValue + 1); - for (n=0; n<=nonNullRank; n++) - nbPerRank[huffNode[n].nbBits]++; - /* determine starting value per rank */ - { U16 min = 0; - for (n=(int)maxNbBits; n>0; n--) { - valPerRank[n] = min; /* get starting value within each rank */ - min += nbPerRank[n]; - min >>= 1; - } } - for (n=0; n<alphabetSize; n++) + /* fill result into ctable (val, nbBits) */ + int n; + U16 nbPerRank[HUF_TABLELOG_MAX+1] = {0}; + U16 valPerRank[HUF_TABLELOG_MAX+1] = {0}; + int const alphabetSize = (int)(maxSymbolValue + 1); + for (n=0; n<=nonNullRank; n++) + nbPerRank[huffNode[n].nbBits]++; + /* determine starting value per rank */ + { U16 min = 0; + for (n=(int)maxNbBits; n>0; n--) { + valPerRank[n] = min; /* get starting value within each rank */ + min += nbPerRank[n]; + min >>= 1; + } } + for (n=0; n<alphabetSize; n++) HUF_setNbBits(ct + huffNode[n].byte, huffNode[n].nbBits); /* push nbBits per symbol, symbol order */ - for (n=0; n<alphabetSize; n++) + for (n=0; n<alphabetSize; n++) HUF_setValue(ct + n, valPerRank[HUF_getNbBits(ct[n])]++); /* assign value within rank, symbol order */ CTable[0] = maxNbBits; -} - +} + size_t HUF_buildCTable_wksp (HUF_CElt* CTable, const unsigned* count, U32 maxSymbolValue, U32 maxNbBits, void* workSpace, size_t wkspSize) -{ +{ HUF_buildCTable_wksp_tables* const wksp_tables = (HUF_buildCTable_wksp_tables*)HUF_alignUpWorkspace(workSpace, &wkspSize, ZSTD_ALIGNOF(U32)); - nodeElt* const huffNode0 = wksp_tables->huffNodeTbl; - nodeElt* const huffNode = huffNode0+1; - int nonNullRank; - - /* safety checks */ - if (wkspSize < sizeof(HUF_buildCTable_wksp_tables)) - return ERROR(workSpace_tooSmall); - if (maxNbBits == 0) maxNbBits = HUF_TABLELOG_DEFAULT; - if (maxSymbolValue > HUF_SYMBOLVALUE_MAX) - return ERROR(maxSymbolValue_tooLarge); - ZSTD_memset(huffNode0, 0, sizeof(huffNodeTable)); - - /* sort, decreasing order */ - HUF_sort(huffNode, count, maxSymbolValue, wksp_tables->rankPosition); - - /* build tree */ - nonNullRank = HUF_buildTree(huffNode, maxSymbolValue); - + nodeElt* const huffNode0 = wksp_tables->huffNodeTbl; + nodeElt* const huffNode = huffNode0+1; + int nonNullRank; + + /* safety checks */ + if (wkspSize < sizeof(HUF_buildCTable_wksp_tables)) + return ERROR(workSpace_tooSmall); + if (maxNbBits == 0) maxNbBits = HUF_TABLELOG_DEFAULT; + if (maxSymbolValue > HUF_SYMBOLVALUE_MAX) + return ERROR(maxSymbolValue_tooLarge); + ZSTD_memset(huffNode0, 0, sizeof(huffNodeTable)); + + /* sort, decreasing order */ + HUF_sort(huffNode, count, maxSymbolValue, wksp_tables->rankPosition); + + /* build tree */ + nonNullRank = HUF_buildTree(huffNode, maxSymbolValue); + /* enforce maxTableLog */ maxNbBits = HUF_setMaxHeight(huffNode, (U32)nonNullRank, maxNbBits); - if (maxNbBits > HUF_TABLELOG_MAX) return ERROR(GENERIC); /* check fit into table */ + if (maxNbBits > HUF_TABLELOG_MAX) return ERROR(GENERIC); /* check fit into table */ HUF_buildCTableFromTree(CTable, huffNode, nonNullRank, maxSymbolValue, maxNbBits); @@ -1161,18 +1161,18 @@ static size_t HUF_compressCTable_internal( typedef struct { unsigned count[HUF_SYMBOLVALUE_MAX + 1]; HUF_CElt CTable[HUF_CTABLE_SIZE_ST(HUF_SYMBOLVALUE_MAX)]; - union { - HUF_buildCTable_wksp_tables buildCTable_wksp; - HUF_WriteCTableWksp writeCTable_wksp; + union { + HUF_buildCTable_wksp_tables buildCTable_wksp; + HUF_WriteCTableWksp writeCTable_wksp; U32 hist_wksp[HIST_WKSP_SIZE_U32]; - } wksps; + } wksps; } HUF_compress_tables_t; #define SUSPECT_INCOMPRESSIBLE_SAMPLE_SIZE 4096 #define SUSPECT_INCOMPRESSIBLE_SAMPLE_RATIO 10 /* Must be >= 2 */ /* HUF_compress_internal() : - * `workSpace_align4` must be aligned on 4-bytes boundaries, + * `workSpace_align4` must be aligned on 4-bytes boundaries, * and occupies the same space as a table of HUF_WORKSPACE_SIZE_U64 unsigned */ static size_t HUF_compress_internal (void* dst, size_t dstSize, @@ -1245,7 +1245,7 @@ HUF_compress_internal (void* dst, size_t dstSize, huffLog = HUF_optimalTableLog(huffLog, srcSize, maxSymbolValue); { size_t const maxBits = HUF_buildCTable_wksp(table->CTable, table->count, maxSymbolValue, huffLog, - &table->wksps.buildCTable_wksp, sizeof(table->wksps.buildCTable_wksp)); + &table->wksps.buildCTable_wksp, sizeof(table->wksps.buildCTable_wksp)); CHECK_F(maxBits); huffLog = (U32)maxBits; } @@ -1257,8 +1257,8 @@ HUF_compress_internal (void* dst, size_t dstSize, } /* Write table description header */ - { CHECK_V_F(hSize, HUF_writeCTable_wksp(op, dstSize, table->CTable, maxSymbolValue, huffLog, - &table->wksps.writeCTable_wksp, sizeof(table->wksps.writeCTable_wksp)) ); + { CHECK_V_F(hSize, HUF_writeCTable_wksp(op, dstSize, table->CTable, maxSymbolValue, huffLog, + &table->wksps.writeCTable_wksp, sizeof(table->wksps.writeCTable_wksp)) ); /* Check if using previous huffman table is beneficial */ if (repeat && *repeat != HUF_repeat_none) { size_t const oldSize = HUF_estimateCompressedSize(oldHufTable, table->count, maxSymbolValue); @@ -1274,7 +1274,7 @@ HUF_compress_internal (void* dst, size_t dstSize, op += hSize; if (repeat) { *repeat = HUF_repeat_none; } if (oldHufTable) - ZSTD_memcpy(oldHufTable, table->CTable, sizeof(table->CTable)); /* Save new table */ + ZSTD_memcpy(oldHufTable, table->CTable, sizeof(table->CTable)); /* Save new table */ } return HUF_compressCTable_internal(ostart, op, oend, src, srcSize, @@ -1336,25 +1336,25 @@ size_t HUF_compress4X_repeat (void* dst, size_t dstSize, hufTable, repeat, preferRepeat, bmi2, suspectUncompressible); } -#ifndef ZSTD_NO_UNUSED_FUNCTIONS -/** HUF_buildCTable() : - * @return : maxNbBits - * Note : count is used before tree is written, so they can safely overlap - */ -size_t HUF_buildCTable (HUF_CElt* tree, const unsigned* count, unsigned maxSymbolValue, unsigned maxNbBits) -{ - HUF_buildCTable_wksp_tables workspace; - return HUF_buildCTable_wksp(tree, count, maxSymbolValue, maxNbBits, &workspace, sizeof(workspace)); -} - -size_t HUF_compress1X (void* dst, size_t dstSize, - const void* src, size_t srcSize, - unsigned maxSymbolValue, unsigned huffLog) -{ +#ifndef ZSTD_NO_UNUSED_FUNCTIONS +/** HUF_buildCTable() : + * @return : maxNbBits + * Note : count is used before tree is written, so they can safely overlap + */ +size_t HUF_buildCTable (HUF_CElt* tree, const unsigned* count, unsigned maxSymbolValue, unsigned maxNbBits) +{ + HUF_buildCTable_wksp_tables workspace; + return HUF_buildCTable_wksp(tree, count, maxSymbolValue, maxNbBits, &workspace, sizeof(workspace)); +} + +size_t HUF_compress1X (void* dst, size_t dstSize, + const void* src, size_t srcSize, + unsigned maxSymbolValue, unsigned huffLog) +{ U64 workSpace[HUF_WORKSPACE_SIZE_U64]; - return HUF_compress1X_wksp(dst, dstSize, src, srcSize, maxSymbolValue, huffLog, workSpace, sizeof(workSpace)); -} - + return HUF_compress1X_wksp(dst, dstSize, src, srcSize, maxSymbolValue, huffLog, workSpace, sizeof(workSpace)); +} + size_t HUF_compress2 (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned huffLog) @@ -1367,4 +1367,4 @@ size_t HUF_compress (void* dst, size_t maxDstSize, const void* src, size_t srcSi { return HUF_compress2(dst, maxDstSize, src, srcSize, 255, HUF_TABLELOG_DEFAULT); } -#endif +#endif diff --git a/contrib/libs/zstd/lib/compress/zstd_compress.c b/contrib/libs/zstd/lib/compress/zstd_compress.c index f06456af92..1c5bc26ba9 100644 --- a/contrib/libs/zstd/lib/compress/zstd_compress.c +++ b/contrib/libs/zstd/lib/compress/zstd_compress.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -11,7 +11,7 @@ /*-************************************* * Dependencies ***************************************/ -#include "../common/zstd_deps.h" /* INT_MAX, ZSTD_memset, ZSTD_memcpy */ +#include "../common/zstd_deps.h" /* INT_MAX, ZSTD_memset, ZSTD_memcpy */ #include "../common/mem.h" #include "hist.h" /* HIST_countFast_wksp */ #define FSE_STATIC_LINKING_ONLY /* FSE_encodeSymbol */ @@ -28,18 +28,18 @@ #include "zstd_ldm.h" #include "zstd_compress_superblock.h" -/* *************************************************************** -* Tuning parameters -*****************************************************************/ -/*! - * COMPRESS_HEAPMODE : - * Select how default decompression function ZSTD_compress() allocates its context, - * on stack (0, default), or into heap (1). - * Note that functions with explicit context such as ZSTD_compressCCtx() are unaffected. - */ -#ifndef ZSTD_COMPRESS_HEAPMODE -# define ZSTD_COMPRESS_HEAPMODE 0 -#endif +/* *************************************************************** +* Tuning parameters +*****************************************************************/ +/*! + * COMPRESS_HEAPMODE : + * Select how default decompression function ZSTD_compress() allocates its context, + * on stack (0, default), or into heap (1). + * Note that functions with explicit context such as ZSTD_compressCCtx() are unaffected. + */ +#ifndef ZSTD_COMPRESS_HEAPMODE +# define ZSTD_COMPRESS_HEAPMODE 0 +#endif /*! * ZSTD_HASHLOG3_MAX : @@ -53,7 +53,7 @@ #ifndef ZSTD_HASHLOG3_MAX # define ZSTD_HASHLOG3_MAX 17 #endif - + /*-************************************* * Helper functions ***************************************/ @@ -75,7 +75,7 @@ size_t ZSTD_compressBound(size_t srcSize) { struct ZSTD_CDict_s { const void* dictContent; size_t dictContentSize; - ZSTD_dictContentType_e dictContentType; /* The dictContentType the CDict was created with */ + ZSTD_dictContentType_e dictContentType; /* The dictContentType the CDict was created with */ U32* entropyWorkspace; /* entropy workspace of HUF_WORKSPACE_SIZE bytes */ ZSTD_cwksp workspace; ZSTD_matchState_t matchState; @@ -97,7 +97,7 @@ ZSTD_CCtx* ZSTD_createCCtx(void) static void ZSTD_initCCtx(ZSTD_CCtx* cctx, ZSTD_customMem memManager) { assert(cctx != NULL); - ZSTD_memset(cctx, 0, sizeof(*cctx)); + ZSTD_memset(cctx, 0, sizeof(*cctx)); cctx->customMem = memManager; cctx->bmi2 = ZSTD_cpuSupportsBmi2(); { size_t const err = ZSTD_CCtx_reset(cctx, ZSTD_reset_parameters); @@ -110,8 +110,8 @@ ZSTD_CCtx* ZSTD_createCCtx_advanced(ZSTD_customMem customMem) { ZSTD_STATIC_ASSERT(zcss_init==0); ZSTD_STATIC_ASSERT(ZSTD_CONTENTSIZE_UNKNOWN==(0ULL - 1)); - if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL; - { ZSTD_CCtx* const cctx = (ZSTD_CCtx*)ZSTD_customMalloc(sizeof(ZSTD_CCtx), customMem); + if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL; + { ZSTD_CCtx* const cctx = (ZSTD_CCtx*)ZSTD_customMalloc(sizeof(ZSTD_CCtx), customMem); if (!cctx) return NULL; ZSTD_initCCtx(cctx, customMem); return cctx; @@ -124,20 +124,20 @@ ZSTD_CCtx* ZSTD_initStaticCCtx(void* workspace, size_t workspaceSize) ZSTD_CCtx* cctx; if (workspaceSize <= sizeof(ZSTD_CCtx)) return NULL; /* minimum size */ if ((size_t)workspace & 7) return NULL; /* must be 8-aligned */ - ZSTD_cwksp_init(&ws, workspace, workspaceSize, ZSTD_cwksp_static_alloc); + ZSTD_cwksp_init(&ws, workspace, workspaceSize, ZSTD_cwksp_static_alloc); cctx = (ZSTD_CCtx*)ZSTD_cwksp_reserve_object(&ws, sizeof(ZSTD_CCtx)); if (cctx == NULL) return NULL; - ZSTD_memset(cctx, 0, sizeof(ZSTD_CCtx)); + ZSTD_memset(cctx, 0, sizeof(ZSTD_CCtx)); ZSTD_cwksp_move(&cctx->workspace, &ws); cctx->staticSize = workspaceSize; /* statically sized space. entropyWorkspace never moves (but prev/next block swap places) */ - if (!ZSTD_cwksp_check_available(&cctx->workspace, ENTROPY_WORKSPACE_SIZE + 2 * sizeof(ZSTD_compressedBlockState_t))) return NULL; + if (!ZSTD_cwksp_check_available(&cctx->workspace, ENTROPY_WORKSPACE_SIZE + 2 * sizeof(ZSTD_compressedBlockState_t))) return NULL; cctx->blockState.prevCBlock = (ZSTD_compressedBlockState_t*)ZSTD_cwksp_reserve_object(&cctx->workspace, sizeof(ZSTD_compressedBlockState_t)); cctx->blockState.nextCBlock = (ZSTD_compressedBlockState_t*)ZSTD_cwksp_reserve_object(&cctx->workspace, sizeof(ZSTD_compressedBlockState_t)); - cctx->entropyWorkspace = (U32*)ZSTD_cwksp_reserve_object(&cctx->workspace, ENTROPY_WORKSPACE_SIZE); + cctx->entropyWorkspace = (U32*)ZSTD_cwksp_reserve_object(&cctx->workspace, ENTROPY_WORKSPACE_SIZE); cctx->bmi2 = ZSTD_cpuid_bmi2(ZSTD_cpuid()); return cctx; } @@ -147,10 +147,10 @@ ZSTD_CCtx* ZSTD_initStaticCCtx(void* workspace, size_t workspaceSize) */ static void ZSTD_clearAllDicts(ZSTD_CCtx* cctx) { - ZSTD_customFree(cctx->localDict.dictBuffer, cctx->customMem); + ZSTD_customFree(cctx->localDict.dictBuffer, cctx->customMem); ZSTD_freeCDict(cctx->localDict.cdict); - ZSTD_memset(&cctx->localDict, 0, sizeof(cctx->localDict)); - ZSTD_memset(&cctx->prefixDict, 0, sizeof(cctx->prefixDict)); + ZSTD_memset(&cctx->localDict, 0, sizeof(cctx->localDict)); + ZSTD_memset(&cctx->prefixDict, 0, sizeof(cctx->prefixDict)); cctx->cdict = NULL; } @@ -181,7 +181,7 @@ size_t ZSTD_freeCCtx(ZSTD_CCtx* cctx) int cctxInWorkspace = ZSTD_cwksp_owns_buffer(&cctx->workspace, cctx); ZSTD_freeCCtxContent(cctx); if (!cctxInWorkspace) { - ZSTD_customFree(cctx, cctx->customMem); + ZSTD_customFree(cctx, cctx->customMem); } } return 0; @@ -217,38 +217,38 @@ size_t ZSTD_sizeof_CStream(const ZSTD_CStream* zcs) /* private API call, for dictBuilder only */ const seqStore_t* ZSTD_getSeqStore(const ZSTD_CCtx* ctx) { return &(ctx->seqStore); } -/* Returns true if the strategy supports using a row based matchfinder */ -static int ZSTD_rowMatchFinderSupported(const ZSTD_strategy strategy) { - return (strategy >= ZSTD_greedy && strategy <= ZSTD_lazy2); -} - -/* Returns true if the strategy and useRowMatchFinder mode indicate that we will use the row based matchfinder - * for this compression. - */ +/* Returns true if the strategy supports using a row based matchfinder */ +static int ZSTD_rowMatchFinderSupported(const ZSTD_strategy strategy) { + return (strategy >= ZSTD_greedy && strategy <= ZSTD_lazy2); +} + +/* Returns true if the strategy and useRowMatchFinder mode indicate that we will use the row based matchfinder + * for this compression. + */ static int ZSTD_rowMatchFinderUsed(const ZSTD_strategy strategy, const ZSTD_paramSwitch_e mode) { assert(mode != ZSTD_ps_auto); return ZSTD_rowMatchFinderSupported(strategy) && (mode == ZSTD_ps_enable); -} - +} + /* Returns row matchfinder usage given an initial mode and cParams */ static ZSTD_paramSwitch_e ZSTD_resolveRowMatchFinderMode(ZSTD_paramSwitch_e mode, const ZSTD_compressionParameters* const cParams) { #if defined(ZSTD_ARCH_X86_SSE2) || defined(ZSTD_ARCH_ARM_NEON) - int const kHasSIMD128 = 1; -#else - int const kHasSIMD128 = 0; -#endif + int const kHasSIMD128 = 1; +#else + int const kHasSIMD128 = 0; +#endif if (mode != ZSTD_ps_auto) return mode; /* if requested enabled, but no SIMD, we still will use row matchfinder */ mode = ZSTD_ps_disable; - if (!ZSTD_rowMatchFinderSupported(cParams->strategy)) return mode; - if (kHasSIMD128) { + if (!ZSTD_rowMatchFinderSupported(cParams->strategy)) return mode; + if (kHasSIMD128) { if (cParams->windowLog > 14) mode = ZSTD_ps_enable; - } else { + } else { if (cParams->windowLog > 17) mode = ZSTD_ps_enable; - } - return mode; -} - + } + return mode; +} + /* Returns block splitter usage (generally speaking, when using slower/stronger compression modes) */ static ZSTD_paramSwitch_e ZSTD_resolveBlockSplitterMode(ZSTD_paramSwitch_e mode, const ZSTD_compressionParameters* const cParams) { @@ -256,44 +256,44 @@ static ZSTD_paramSwitch_e ZSTD_resolveBlockSplitterMode(ZSTD_paramSwitch_e mode, return (cParams->strategy >= ZSTD_btopt && cParams->windowLog >= 17) ? ZSTD_ps_enable : ZSTD_ps_disable; } -/* Returns 1 if the arguments indicate that we should allocate a chainTable, 0 otherwise */ -static int ZSTD_allocateChainTable(const ZSTD_strategy strategy, +/* Returns 1 if the arguments indicate that we should allocate a chainTable, 0 otherwise */ +static int ZSTD_allocateChainTable(const ZSTD_strategy strategy, const ZSTD_paramSwitch_e useRowMatchFinder, - const U32 forDDSDict) { + const U32 forDDSDict) { assert(useRowMatchFinder != ZSTD_ps_auto); - /* We always should allocate a chaintable if we are allocating a matchstate for a DDS dictionary matchstate. - * We do not allocate a chaintable if we are using ZSTD_fast, or are using the row-based matchfinder. - */ - return forDDSDict || ((strategy != ZSTD_fast) && !ZSTD_rowMatchFinderUsed(strategy, useRowMatchFinder)); -} - -/* Returns 1 if compression parameters are such that we should - * enable long distance matching (wlog >= 27, strategy >= btopt). - * Returns 0 otherwise. - */ + /* We always should allocate a chaintable if we are allocating a matchstate for a DDS dictionary matchstate. + * We do not allocate a chaintable if we are using ZSTD_fast, or are using the row-based matchfinder. + */ + return forDDSDict || ((strategy != ZSTD_fast) && !ZSTD_rowMatchFinderUsed(strategy, useRowMatchFinder)); +} + +/* Returns 1 if compression parameters are such that we should + * enable long distance matching (wlog >= 27, strategy >= btopt). + * Returns 0 otherwise. + */ static ZSTD_paramSwitch_e ZSTD_resolveEnableLdm(ZSTD_paramSwitch_e mode, const ZSTD_compressionParameters* const cParams) { if (mode != ZSTD_ps_auto) return mode; return (cParams->strategy >= ZSTD_btopt && cParams->windowLog >= 27) ? ZSTD_ps_enable : ZSTD_ps_disable; -} - +} + static ZSTD_CCtx_params ZSTD_makeCCtxParamsFromCParams( ZSTD_compressionParameters cParams) { ZSTD_CCtx_params cctxParams; - /* should not matter, as all cParams are presumed properly defined */ - ZSTD_CCtxParams_init(&cctxParams, ZSTD_CLEVEL_DEFAULT); + /* should not matter, as all cParams are presumed properly defined */ + ZSTD_CCtxParams_init(&cctxParams, ZSTD_CLEVEL_DEFAULT); cctxParams.cParams = cParams; - - /* Adjust advanced params according to cParams */ + + /* Adjust advanced params according to cParams */ cctxParams.ldmParams.enableLdm = ZSTD_resolveEnableLdm(cctxParams.ldmParams.enableLdm, &cParams); if (cctxParams.ldmParams.enableLdm == ZSTD_ps_enable) { - ZSTD_ldm_adjustParameters(&cctxParams.ldmParams, &cParams); - assert(cctxParams.ldmParams.hashLog >= cctxParams.ldmParams.bucketSizeLog); - assert(cctxParams.ldmParams.hashRateLog < 32); - } + ZSTD_ldm_adjustParameters(&cctxParams.ldmParams, &cParams); + assert(cctxParams.ldmParams.hashLog >= cctxParams.ldmParams.bucketSizeLog); + assert(cctxParams.ldmParams.hashRateLog < 32); + } cctxParams.useBlockSplitter = ZSTD_resolveBlockSplitterMode(cctxParams.useBlockSplitter, &cParams); - cctxParams.useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(cctxParams.useRowMatchFinder, &cParams); + cctxParams.useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(cctxParams.useRowMatchFinder, &cParams); assert(!ZSTD_checkCParams(cParams)); return cctxParams; } @@ -302,11 +302,11 @@ static ZSTD_CCtx_params* ZSTD_createCCtxParams_advanced( ZSTD_customMem customMem) { ZSTD_CCtx_params* params; - if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL; - params = (ZSTD_CCtx_params*)ZSTD_customCalloc( + if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL; + params = (ZSTD_CCtx_params*)ZSTD_customCalloc( sizeof(ZSTD_CCtx_params), customMem); if (!params) { return NULL; } - ZSTD_CCtxParams_init(params, ZSTD_CLEVEL_DEFAULT); + ZSTD_CCtxParams_init(params, ZSTD_CLEVEL_DEFAULT); params->customMem = customMem; return params; } @@ -319,7 +319,7 @@ ZSTD_CCtx_params* ZSTD_createCCtxParams(void) size_t ZSTD_freeCCtxParams(ZSTD_CCtx_params* params) { if (params == NULL) { return 0; } - ZSTD_customFree(params, params->customMem); + ZSTD_customFree(params, params->customMem); return 0; } @@ -330,57 +330,57 @@ size_t ZSTD_CCtxParams_reset(ZSTD_CCtx_params* params) size_t ZSTD_CCtxParams_init(ZSTD_CCtx_params* cctxParams, int compressionLevel) { RETURN_ERROR_IF(!cctxParams, GENERIC, "NULL pointer!"); - ZSTD_memset(cctxParams, 0, sizeof(*cctxParams)); + ZSTD_memset(cctxParams, 0, sizeof(*cctxParams)); cctxParams->compressionLevel = compressionLevel; cctxParams->fParams.contentSizeFlag = 1; return 0; } -#define ZSTD_NO_CLEVEL 0 - -/** - * Initializes the cctxParams from params and compressionLevel. - * @param compressionLevel If params are derived from a compression level then that compression level, otherwise ZSTD_NO_CLEVEL. - */ -static void ZSTD_CCtxParams_init_internal(ZSTD_CCtx_params* cctxParams, ZSTD_parameters const* params, int compressionLevel) -{ - assert(!ZSTD_checkCParams(params->cParams)); - ZSTD_memset(cctxParams, 0, sizeof(*cctxParams)); - cctxParams->cParams = params->cParams; - cctxParams->fParams = params->fParams; - /* Should not matter, as all cParams are presumed properly defined. - * But, set it for tracing anyway. - */ - cctxParams->compressionLevel = compressionLevel; - cctxParams->useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(cctxParams->useRowMatchFinder, ¶ms->cParams); +#define ZSTD_NO_CLEVEL 0 + +/** + * Initializes the cctxParams from params and compressionLevel. + * @param compressionLevel If params are derived from a compression level then that compression level, otherwise ZSTD_NO_CLEVEL. + */ +static void ZSTD_CCtxParams_init_internal(ZSTD_CCtx_params* cctxParams, ZSTD_parameters const* params, int compressionLevel) +{ + assert(!ZSTD_checkCParams(params->cParams)); + ZSTD_memset(cctxParams, 0, sizeof(*cctxParams)); + cctxParams->cParams = params->cParams; + cctxParams->fParams = params->fParams; + /* Should not matter, as all cParams are presumed properly defined. + * But, set it for tracing anyway. + */ + cctxParams->compressionLevel = compressionLevel; + cctxParams->useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(cctxParams->useRowMatchFinder, ¶ms->cParams); cctxParams->useBlockSplitter = ZSTD_resolveBlockSplitterMode(cctxParams->useBlockSplitter, ¶ms->cParams); cctxParams->ldmParams.enableLdm = ZSTD_resolveEnableLdm(cctxParams->ldmParams.enableLdm, ¶ms->cParams); DEBUGLOG(4, "ZSTD_CCtxParams_init_internal: useRowMatchFinder=%d, useBlockSplitter=%d ldm=%d", cctxParams->useRowMatchFinder, cctxParams->useBlockSplitter, cctxParams->ldmParams.enableLdm); -} - +} + size_t ZSTD_CCtxParams_init_advanced(ZSTD_CCtx_params* cctxParams, ZSTD_parameters params) { RETURN_ERROR_IF(!cctxParams, GENERIC, "NULL pointer!"); FORWARD_IF_ERROR( ZSTD_checkCParams(params.cParams) , ""); - ZSTD_CCtxParams_init_internal(cctxParams, ¶ms, ZSTD_NO_CLEVEL); + ZSTD_CCtxParams_init_internal(cctxParams, ¶ms, ZSTD_NO_CLEVEL); return 0; } -/** - * Sets cctxParams' cParams and fParams from params, but otherwise leaves them alone. - * @param param Validated zstd parameters. - */ -static void ZSTD_CCtxParams_setZstdParams( - ZSTD_CCtx_params* cctxParams, const ZSTD_parameters* params) +/** + * Sets cctxParams' cParams and fParams from params, but otherwise leaves them alone. + * @param param Validated zstd parameters. + */ +static void ZSTD_CCtxParams_setZstdParams( + ZSTD_CCtx_params* cctxParams, const ZSTD_parameters* params) { assert(!ZSTD_checkCParams(params->cParams)); - cctxParams->cParams = params->cParams; - cctxParams->fParams = params->fParams; - /* Should not matter, as all cParams are presumed properly defined. - * But, set it for tracing anyway. - */ - cctxParams->compressionLevel = ZSTD_NO_CLEVEL; + cctxParams->cParams = params->cParams; + cctxParams->fParams = params->fParams; + /* Should not matter, as all cParams are presumed properly defined. + * But, set it for tracing anyway. + */ + cctxParams->compressionLevel = ZSTD_NO_CLEVEL; } ZSTD_bounds ZSTD_cParam_getBounds(ZSTD_cParameter param) @@ -472,11 +472,11 @@ ZSTD_bounds ZSTD_cParam_getBounds(ZSTD_cParameter param) #endif return bounds; - case ZSTD_c_enableDedicatedDictSearch: - bounds.lowerBound = 0; - bounds.upperBound = 1; - return bounds; - + case ZSTD_c_enableDedicatedDictSearch: + bounds.lowerBound = 0; + bounds.upperBound = 1; + return bounds; + case ZSTD_c_enableLongDistanceMatching: bounds.lowerBound = 0; bounds.upperBound = 1; @@ -520,7 +520,7 @@ ZSTD_bounds ZSTD_cParam_getBounds(ZSTD_cParameter param) return bounds; case ZSTD_c_forceAttachDict: - ZSTD_STATIC_ASSERT(ZSTD_dictDefaultAttach < ZSTD_dictForceLoad); + ZSTD_STATIC_ASSERT(ZSTD_dictDefaultAttach < ZSTD_dictForceLoad); bounds.lowerBound = ZSTD_dictDefaultAttach; bounds.upperBound = ZSTD_dictForceLoad; /* note : how to ensure at compile time that this is the highest value enum ? */ return bounds; @@ -541,37 +541,37 @@ ZSTD_bounds ZSTD_cParam_getBounds(ZSTD_cParameter param) bounds.upperBound = ZSTD_SRCSIZEHINT_MAX; return bounds; - case ZSTD_c_stableInBuffer: - case ZSTD_c_stableOutBuffer: - bounds.lowerBound = (int)ZSTD_bm_buffered; - bounds.upperBound = (int)ZSTD_bm_stable; - return bounds; - - case ZSTD_c_blockDelimiters: - bounds.lowerBound = (int)ZSTD_sf_noBlockDelimiters; - bounds.upperBound = (int)ZSTD_sf_explicitBlockDelimiters; - return bounds; - - case ZSTD_c_validateSequences: - bounds.lowerBound = 0; - bounds.upperBound = 1; - return bounds; - + case ZSTD_c_stableInBuffer: + case ZSTD_c_stableOutBuffer: + bounds.lowerBound = (int)ZSTD_bm_buffered; + bounds.upperBound = (int)ZSTD_bm_stable; + return bounds; + + case ZSTD_c_blockDelimiters: + bounds.lowerBound = (int)ZSTD_sf_noBlockDelimiters; + bounds.upperBound = (int)ZSTD_sf_explicitBlockDelimiters; + return bounds; + + case ZSTD_c_validateSequences: + bounds.lowerBound = 0; + bounds.upperBound = 1; + return bounds; + case ZSTD_c_useBlockSplitter: bounds.lowerBound = (int)ZSTD_ps_auto; bounds.upperBound = (int)ZSTD_ps_disable; - return bounds; - - case ZSTD_c_useRowMatchFinder: + return bounds; + + case ZSTD_c_useRowMatchFinder: bounds.lowerBound = (int)ZSTD_ps_auto; bounds.upperBound = (int)ZSTD_ps_disable; - return bounds; - - case ZSTD_c_deterministicRefPrefix: - bounds.lowerBound = 0; - bounds.upperBound = 1; - return bounds; - + return bounds; + + case ZSTD_c_deterministicRefPrefix: + bounds.lowerBound = 0; + bounds.upperBound = 1; + return bounds; + default: bounds.error = ERROR(parameter_unsupported); return bounds; @@ -619,7 +619,7 @@ static int ZSTD_isUpdateAuthorized(ZSTD_cParameter param) case ZSTD_c_jobSize: case ZSTD_c_overlapLog: case ZSTD_c_rsyncable: - case ZSTD_c_enableDedicatedDictSearch: + case ZSTD_c_enableDedicatedDictSearch: case ZSTD_c_enableLongDistanceMatching: case ZSTD_c_ldmHashLog: case ZSTD_c_ldmMinMatch: @@ -629,13 +629,13 @@ static int ZSTD_isUpdateAuthorized(ZSTD_cParameter param) case ZSTD_c_literalCompressionMode: case ZSTD_c_targetCBlockSize: case ZSTD_c_srcSizeHint: - case ZSTD_c_stableInBuffer: - case ZSTD_c_stableOutBuffer: - case ZSTD_c_blockDelimiters: - case ZSTD_c_validateSequences: + case ZSTD_c_stableInBuffer: + case ZSTD_c_stableOutBuffer: + case ZSTD_c_blockDelimiters: + case ZSTD_c_validateSequences: case ZSTD_c_useBlockSplitter: - case ZSTD_c_useRowMatchFinder: - case ZSTD_c_deterministicRefPrefix: + case ZSTD_c_useRowMatchFinder: + case ZSTD_c_deterministicRefPrefix: default: return 0; } @@ -677,20 +677,20 @@ size_t ZSTD_CCtx_setParameter(ZSTD_CCtx* cctx, ZSTD_cParameter param, int value) case ZSTD_c_jobSize: case ZSTD_c_overlapLog: case ZSTD_c_rsyncable: - case ZSTD_c_enableDedicatedDictSearch: + case ZSTD_c_enableDedicatedDictSearch: case ZSTD_c_enableLongDistanceMatching: case ZSTD_c_ldmHashLog: case ZSTD_c_ldmMinMatch: case ZSTD_c_ldmBucketSizeLog: case ZSTD_c_targetCBlockSize: case ZSTD_c_srcSizeHint: - case ZSTD_c_stableInBuffer: - case ZSTD_c_stableOutBuffer: - case ZSTD_c_blockDelimiters: - case ZSTD_c_validateSequences: + case ZSTD_c_stableInBuffer: + case ZSTD_c_stableOutBuffer: + case ZSTD_c_blockDelimiters: + case ZSTD_c_validateSequences: case ZSTD_c_useBlockSplitter: - case ZSTD_c_useRowMatchFinder: - case ZSTD_c_deterministicRefPrefix: + case ZSTD_c_useRowMatchFinder: + case ZSTD_c_deterministicRefPrefix: break; default: RETURN_ERROR(parameter_unsupported, "unknown parameter"); @@ -711,9 +711,9 @@ size_t ZSTD_CCtxParams_setParameter(ZSTD_CCtx_params* CCtxParams, case ZSTD_c_compressionLevel : { FORWARD_IF_ERROR(ZSTD_cParam_clampBounds(param, &value), ""); - if (value == 0) - CCtxParams->compressionLevel = ZSTD_CLEVEL_DEFAULT; /* 0 == default */ - else + if (value == 0) + CCtxParams->compressionLevel = ZSTD_CLEVEL_DEFAULT; /* 0 == default */ + else CCtxParams->compressionLevel = value; if (CCtxParams->compressionLevel >= 0) return (size_t)CCtxParams->compressionLevel; return 0; /* return type (size_t) cannot represent negative values */ @@ -838,10 +838,10 @@ size_t ZSTD_CCtxParams_setParameter(ZSTD_CCtx_params* CCtxParams, return CCtxParams->rsyncable; #endif - case ZSTD_c_enableDedicatedDictSearch : - CCtxParams->enableDedicatedDictSearch = (value!=0); - return CCtxParams->enableDedicatedDictSearch; - + case ZSTD_c_enableDedicatedDictSearch : + CCtxParams->enableDedicatedDictSearch = (value!=0); + return CCtxParams->enableDedicatedDictSearch; + case ZSTD_c_enableLongDistanceMatching : CCtxParams->ldmParams.enableLdm = (ZSTD_paramSwitch_e)value; return CCtxParams->ldmParams.enableLdm; @@ -882,52 +882,52 @@ size_t ZSTD_CCtxParams_setParameter(ZSTD_CCtx_params* CCtxParams, CCtxParams->srcSizeHint = value; return CCtxParams->srcSizeHint; - case ZSTD_c_stableInBuffer: - BOUNDCHECK(ZSTD_c_stableInBuffer, value); - CCtxParams->inBufferMode = (ZSTD_bufferMode_e)value; - return CCtxParams->inBufferMode; - - case ZSTD_c_stableOutBuffer: - BOUNDCHECK(ZSTD_c_stableOutBuffer, value); - CCtxParams->outBufferMode = (ZSTD_bufferMode_e)value; - return CCtxParams->outBufferMode; - - case ZSTD_c_blockDelimiters: - BOUNDCHECK(ZSTD_c_blockDelimiters, value); - CCtxParams->blockDelimiters = (ZSTD_sequenceFormat_e)value; - return CCtxParams->blockDelimiters; - - case ZSTD_c_validateSequences: - BOUNDCHECK(ZSTD_c_validateSequences, value); - CCtxParams->validateSequences = value; - return CCtxParams->validateSequences; - + case ZSTD_c_stableInBuffer: + BOUNDCHECK(ZSTD_c_stableInBuffer, value); + CCtxParams->inBufferMode = (ZSTD_bufferMode_e)value; + return CCtxParams->inBufferMode; + + case ZSTD_c_stableOutBuffer: + BOUNDCHECK(ZSTD_c_stableOutBuffer, value); + CCtxParams->outBufferMode = (ZSTD_bufferMode_e)value; + return CCtxParams->outBufferMode; + + case ZSTD_c_blockDelimiters: + BOUNDCHECK(ZSTD_c_blockDelimiters, value); + CCtxParams->blockDelimiters = (ZSTD_sequenceFormat_e)value; + return CCtxParams->blockDelimiters; + + case ZSTD_c_validateSequences: + BOUNDCHECK(ZSTD_c_validateSequences, value); + CCtxParams->validateSequences = value; + return CCtxParams->validateSequences; + case ZSTD_c_useBlockSplitter: BOUNDCHECK(ZSTD_c_useBlockSplitter, value); CCtxParams->useBlockSplitter = (ZSTD_paramSwitch_e)value; return CCtxParams->useBlockSplitter; - - case ZSTD_c_useRowMatchFinder: - BOUNDCHECK(ZSTD_c_useRowMatchFinder, value); + + case ZSTD_c_useRowMatchFinder: + BOUNDCHECK(ZSTD_c_useRowMatchFinder, value); CCtxParams->useRowMatchFinder = (ZSTD_paramSwitch_e)value; - return CCtxParams->useRowMatchFinder; - - case ZSTD_c_deterministicRefPrefix: - BOUNDCHECK(ZSTD_c_deterministicRefPrefix, value); - CCtxParams->deterministicRefPrefix = !!value; - return CCtxParams->deterministicRefPrefix; - + return CCtxParams->useRowMatchFinder; + + case ZSTD_c_deterministicRefPrefix: + BOUNDCHECK(ZSTD_c_deterministicRefPrefix, value); + CCtxParams->deterministicRefPrefix = !!value; + return CCtxParams->deterministicRefPrefix; + default: RETURN_ERROR(parameter_unsupported, "unknown parameter"); } } -size_t ZSTD_CCtx_getParameter(ZSTD_CCtx const* cctx, ZSTD_cParameter param, int* value) +size_t ZSTD_CCtx_getParameter(ZSTD_CCtx const* cctx, ZSTD_cParameter param, int* value) { return ZSTD_CCtxParams_getParameter(&cctx->requestedParams, param, value); } size_t ZSTD_CCtxParams_getParameter( - ZSTD_CCtx_params const* CCtxParams, ZSTD_cParameter param, int* value) + ZSTD_CCtx_params const* CCtxParams, ZSTD_cParameter param, int* value) { switch(param) { @@ -1004,9 +1004,9 @@ size_t ZSTD_CCtxParams_getParameter( *value = CCtxParams->rsyncable; break; #endif - case ZSTD_c_enableDedicatedDictSearch : - *value = CCtxParams->enableDedicatedDictSearch; - break; + case ZSTD_c_enableDedicatedDictSearch : + *value = CCtxParams->enableDedicatedDictSearch; + break; case ZSTD_c_enableLongDistanceMatching : *value = CCtxParams->ldmParams.enableLdm; break; @@ -1028,27 +1028,27 @@ size_t ZSTD_CCtxParams_getParameter( case ZSTD_c_srcSizeHint : *value = (int)CCtxParams->srcSizeHint; break; - case ZSTD_c_stableInBuffer : - *value = (int)CCtxParams->inBufferMode; - break; - case ZSTD_c_stableOutBuffer : - *value = (int)CCtxParams->outBufferMode; - break; - case ZSTD_c_blockDelimiters : - *value = (int)CCtxParams->blockDelimiters; - break; - case ZSTD_c_validateSequences : - *value = (int)CCtxParams->validateSequences; - break; + case ZSTD_c_stableInBuffer : + *value = (int)CCtxParams->inBufferMode; + break; + case ZSTD_c_stableOutBuffer : + *value = (int)CCtxParams->outBufferMode; + break; + case ZSTD_c_blockDelimiters : + *value = (int)CCtxParams->blockDelimiters; + break; + case ZSTD_c_validateSequences : + *value = (int)CCtxParams->validateSequences; + break; case ZSTD_c_useBlockSplitter : *value = (int)CCtxParams->useBlockSplitter; - break; - case ZSTD_c_useRowMatchFinder : - *value = (int)CCtxParams->useRowMatchFinder; - break; - case ZSTD_c_deterministicRefPrefix: - *value = (int)CCtxParams->deterministicRefPrefix; - break; + break; + case ZSTD_c_useRowMatchFinder : + *value = (int)CCtxParams->useRowMatchFinder; + break; + case ZSTD_c_deterministicRefPrefix: + *value = (int)CCtxParams->deterministicRefPrefix; + break; default: RETURN_ERROR(parameter_unsupported, "unknown parameter"); } return 0; @@ -1084,14 +1084,14 @@ size_t ZSTD_CCtx_setPledgedSrcSize(ZSTD_CCtx* cctx, unsigned long long pledgedSr return 0; } -static ZSTD_compressionParameters ZSTD_dedicatedDictSearch_getCParams( - int const compressionLevel, - size_t const dictSize); -static int ZSTD_dedicatedDictSearch_isSupported( - const ZSTD_compressionParameters* cParams); -static void ZSTD_dedicatedDictSearch_revertCParams( - ZSTD_compressionParameters* cParams); - +static ZSTD_compressionParameters ZSTD_dedicatedDictSearch_getCParams( + int const compressionLevel, + size_t const dictSize); +static int ZSTD_dedicatedDictSearch_isSupported( + const ZSTD_compressionParameters* cParams); +static void ZSTD_dedicatedDictSearch_revertCParams( + ZSTD_compressionParameters* cParams); + /** * Initializes the local dict using the requested parameters. * NOTE: This does not use the pledged src size, because it may be used for more @@ -1116,12 +1116,12 @@ static size_t ZSTD_initLocalDict(ZSTD_CCtx* cctx) assert(cctx->cdict == NULL); assert(cctx->prefixDict.dict == NULL); - dl->cdict = ZSTD_createCDict_advanced2( + dl->cdict = ZSTD_createCDict_advanced2( dl->dict, dl->dictSize, ZSTD_dlm_byRef, dl->dictContentType, - &cctx->requestedParams, + &cctx->requestedParams, cctx->customMem); RETURN_ERROR_IF(!dl->cdict, memory_allocation, "ZSTD_createCDict_advanced failed"); cctx->cdict = dl->cdict; @@ -1141,12 +1141,12 @@ size_t ZSTD_CCtx_loadDictionary_advanced( if (dictLoadMethod == ZSTD_dlm_byRef) { cctx->localDict.dict = dict; } else { - void* dictBuffer; - RETURN_ERROR_IF(cctx->staticSize, memory_allocation, - "no malloc for static CCtx"); - dictBuffer = ZSTD_customMalloc(dictSize, cctx->customMem); + void* dictBuffer; + RETURN_ERROR_IF(cctx->staticSize, memory_allocation, + "no malloc for static CCtx"); + dictBuffer = ZSTD_customMalloc(dictSize, cctx->customMem); RETURN_ERROR_IF(!dictBuffer, memory_allocation, "NULL pointer!"); - ZSTD_memcpy(dictBuffer, dict, dictSize); + ZSTD_memcpy(dictBuffer, dict, dictSize); cctx->localDict.dictBuffer = dictBuffer; cctx->localDict.dict = dictBuffer; } @@ -1179,14 +1179,14 @@ size_t ZSTD_CCtx_refCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict) return 0; } -size_t ZSTD_CCtx_refThreadPool(ZSTD_CCtx* cctx, ZSTD_threadPool* pool) -{ - RETURN_ERROR_IF(cctx->streamStage != zcss_init, stage_wrong, - "Can't ref a pool when ctx not in init stage."); - cctx->pool = pool; - return 0; -} - +size_t ZSTD_CCtx_refThreadPool(ZSTD_CCtx* cctx, ZSTD_threadPool* pool) +{ + RETURN_ERROR_IF(cctx->streamStage != zcss_init, stage_wrong, + "Can't ref a pool when ctx not in init stage."); + cctx->pool = pool; + return 0; +} + size_t ZSTD_CCtx_refPrefix(ZSTD_CCtx* cctx, const void* prefix, size_t prefixSize) { return ZSTD_CCtx_refPrefix_advanced(cctx, prefix, prefixSize, ZSTD_dct_rawContent); @@ -1271,84 +1271,84 @@ U32 ZSTD_cycleLog(U32 hashLog, ZSTD_strategy strat) return hashLog - btScale; } -/** ZSTD_dictAndWindowLog() : - * Returns an adjusted window log that is large enough to fit the source and the dictionary. - * The zstd format says that the entire dictionary is valid if one byte of the dictionary - * is within the window. So the hashLog and chainLog should be large enough to reference both - * the dictionary and the window. So we must use this adjusted dictAndWindowLog when downsizing - * the hashLog and windowLog. - * NOTE: srcSize must not be ZSTD_CONTENTSIZE_UNKNOWN. - */ -static U32 ZSTD_dictAndWindowLog(U32 windowLog, U64 srcSize, U64 dictSize) -{ - const U64 maxWindowSize = 1ULL << ZSTD_WINDOWLOG_MAX; - /* No dictionary ==> No change */ - if (dictSize == 0) { - return windowLog; - } - assert(windowLog <= ZSTD_WINDOWLOG_MAX); - assert(srcSize != ZSTD_CONTENTSIZE_UNKNOWN); /* Handled in ZSTD_adjustCParams_internal() */ - { - U64 const windowSize = 1ULL << windowLog; - U64 const dictAndWindowSize = dictSize + windowSize; - /* If the window size is already large enough to fit both the source and the dictionary - * then just use the window size. Otherwise adjust so that it fits the dictionary and - * the window. - */ - if (windowSize >= dictSize + srcSize) { - return windowLog; /* Window size large enough already */ - } else if (dictAndWindowSize >= maxWindowSize) { - return ZSTD_WINDOWLOG_MAX; /* Larger than max window log */ - } else { - return ZSTD_highbit32((U32)dictAndWindowSize - 1) + 1; - } - } -} - +/** ZSTD_dictAndWindowLog() : + * Returns an adjusted window log that is large enough to fit the source and the dictionary. + * The zstd format says that the entire dictionary is valid if one byte of the dictionary + * is within the window. So the hashLog and chainLog should be large enough to reference both + * the dictionary and the window. So we must use this adjusted dictAndWindowLog when downsizing + * the hashLog and windowLog. + * NOTE: srcSize must not be ZSTD_CONTENTSIZE_UNKNOWN. + */ +static U32 ZSTD_dictAndWindowLog(U32 windowLog, U64 srcSize, U64 dictSize) +{ + const U64 maxWindowSize = 1ULL << ZSTD_WINDOWLOG_MAX; + /* No dictionary ==> No change */ + if (dictSize == 0) { + return windowLog; + } + assert(windowLog <= ZSTD_WINDOWLOG_MAX); + assert(srcSize != ZSTD_CONTENTSIZE_UNKNOWN); /* Handled in ZSTD_adjustCParams_internal() */ + { + U64 const windowSize = 1ULL << windowLog; + U64 const dictAndWindowSize = dictSize + windowSize; + /* If the window size is already large enough to fit both the source and the dictionary + * then just use the window size. Otherwise adjust so that it fits the dictionary and + * the window. + */ + if (windowSize >= dictSize + srcSize) { + return windowLog; /* Window size large enough already */ + } else if (dictAndWindowSize >= maxWindowSize) { + return ZSTD_WINDOWLOG_MAX; /* Larger than max window log */ + } else { + return ZSTD_highbit32((U32)dictAndWindowSize - 1) + 1; + } + } +} + /** ZSTD_adjustCParams_internal() : * optimize `cPar` for a specified input (`srcSize` and `dictSize`). * mostly downsize to reduce memory consumption and initialization latency. * `srcSize` can be ZSTD_CONTENTSIZE_UNKNOWN when not known. - * `mode` is the mode for parameter adjustment. See docs for `ZSTD_cParamMode_e`. + * `mode` is the mode for parameter adjustment. See docs for `ZSTD_cParamMode_e`. * note : `srcSize==0` means 0! * condition : cPar is presumed validated (can be checked using ZSTD_checkCParams()). */ static ZSTD_compressionParameters ZSTD_adjustCParams_internal(ZSTD_compressionParameters cPar, unsigned long long srcSize, - size_t dictSize, - ZSTD_cParamMode_e mode) + size_t dictSize, + ZSTD_cParamMode_e mode) { - const U64 minSrcSize = 513; /* (1<<9) + 1 */ - const U64 maxWindowResize = 1ULL << (ZSTD_WINDOWLOG_MAX-1); + const U64 minSrcSize = 513; /* (1<<9) + 1 */ + const U64 maxWindowResize = 1ULL << (ZSTD_WINDOWLOG_MAX-1); assert(ZSTD_checkCParams(cPar)==0); - switch (mode) { - case ZSTD_cpm_unknown: - case ZSTD_cpm_noAttachDict: - /* If we don't know the source size, don't make any - * assumptions about it. We will already have selected - * smaller parameters if a dictionary is in use. - */ - break; - case ZSTD_cpm_createCDict: - /* Assume a small source size when creating a dictionary + switch (mode) { + case ZSTD_cpm_unknown: + case ZSTD_cpm_noAttachDict: + /* If we don't know the source size, don't make any + * assumptions about it. We will already have selected + * smaller parameters if a dictionary is in use. + */ + break; + case ZSTD_cpm_createCDict: + /* Assume a small source size when creating a dictionary * with an unknown source size. - */ - if (dictSize && srcSize == ZSTD_CONTENTSIZE_UNKNOWN) - srcSize = minSrcSize; - break; - case ZSTD_cpm_attachDict: - /* Dictionary has its own dedicated parameters which have - * already been selected. We are selecting parameters - * for only the source. - */ - dictSize = 0; - break; - default: - assert(0); - break; - } - + */ + if (dictSize && srcSize == ZSTD_CONTENTSIZE_UNKNOWN) + srcSize = minSrcSize; + break; + case ZSTD_cpm_attachDict: + /* Dictionary has its own dedicated parameters which have + * already been selected. We are selecting parameters + * for only the source. + */ + dictSize = 0; + break; + default: + assert(0); + break; + } + /* resize windowLog if input is small enough, to use less memory */ if ( (srcSize < maxWindowResize) && (dictSize < maxWindowResize) ) { @@ -1358,12 +1358,12 @@ ZSTD_adjustCParams_internal(ZSTD_compressionParameters cPar, ZSTD_highbit32(tSize-1) + 1; if (cPar.windowLog > srcLog) cPar.windowLog = srcLog; } - if (srcSize != ZSTD_CONTENTSIZE_UNKNOWN) { - U32 const dictAndWindowLog = ZSTD_dictAndWindowLog(cPar.windowLog, (U64)srcSize, (U64)dictSize); - U32 const cycleLog = ZSTD_cycleLog(cPar.chainLog, cPar.strategy); - if (cPar.hashLog > dictAndWindowLog+1) cPar.hashLog = dictAndWindowLog+1; - if (cycleLog > dictAndWindowLog) - cPar.chainLog -= (cycleLog - dictAndWindowLog); + if (srcSize != ZSTD_CONTENTSIZE_UNKNOWN) { + U32 const dictAndWindowLog = ZSTD_dictAndWindowLog(cPar.windowLog, (U64)srcSize, (U64)dictSize); + U32 const cycleLog = ZSTD_cycleLog(cPar.chainLog, cPar.strategy); + if (cPar.hashLog > dictAndWindowLog+1) cPar.hashLog = dictAndWindowLog+1; + if (cycleLog > dictAndWindowLog) + cPar.chainLog -= (cycleLog - dictAndWindowLog); } if (cPar.windowLog < ZSTD_WINDOWLOG_ABSOLUTEMIN) @@ -1379,50 +1379,50 @@ ZSTD_adjustCParams(ZSTD_compressionParameters cPar, { cPar = ZSTD_clampCParams(cPar); /* resulting cPar is necessarily valid (all parameters within range) */ if (srcSize == 0) srcSize = ZSTD_CONTENTSIZE_UNKNOWN; - return ZSTD_adjustCParams_internal(cPar, srcSize, dictSize, ZSTD_cpm_unknown); -} - -static ZSTD_compressionParameters ZSTD_getCParams_internal(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode); -static ZSTD_parameters ZSTD_getParams_internal(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode); - -static void ZSTD_overrideCParams( - ZSTD_compressionParameters* cParams, - const ZSTD_compressionParameters* overrides) -{ - if (overrides->windowLog) cParams->windowLog = overrides->windowLog; - if (overrides->hashLog) cParams->hashLog = overrides->hashLog; - if (overrides->chainLog) cParams->chainLog = overrides->chainLog; - if (overrides->searchLog) cParams->searchLog = overrides->searchLog; - if (overrides->minMatch) cParams->minMatch = overrides->minMatch; - if (overrides->targetLength) cParams->targetLength = overrides->targetLength; - if (overrides->strategy) cParams->strategy = overrides->strategy; -} - + return ZSTD_adjustCParams_internal(cPar, srcSize, dictSize, ZSTD_cpm_unknown); +} + +static ZSTD_compressionParameters ZSTD_getCParams_internal(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode); +static ZSTD_parameters ZSTD_getParams_internal(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode); + +static void ZSTD_overrideCParams( + ZSTD_compressionParameters* cParams, + const ZSTD_compressionParameters* overrides) +{ + if (overrides->windowLog) cParams->windowLog = overrides->windowLog; + if (overrides->hashLog) cParams->hashLog = overrides->hashLog; + if (overrides->chainLog) cParams->chainLog = overrides->chainLog; + if (overrides->searchLog) cParams->searchLog = overrides->searchLog; + if (overrides->minMatch) cParams->minMatch = overrides->minMatch; + if (overrides->targetLength) cParams->targetLength = overrides->targetLength; + if (overrides->strategy) cParams->strategy = overrides->strategy; +} + ZSTD_compressionParameters ZSTD_getCParamsFromCCtxParams( - const ZSTD_CCtx_params* CCtxParams, U64 srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode) + const ZSTD_CCtx_params* CCtxParams, U64 srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode) { ZSTD_compressionParameters cParams; if (srcSizeHint == ZSTD_CONTENTSIZE_UNKNOWN && CCtxParams->srcSizeHint > 0) { srcSizeHint = CCtxParams->srcSizeHint; } - cParams = ZSTD_getCParams_internal(CCtxParams->compressionLevel, srcSizeHint, dictSize, mode); + cParams = ZSTD_getCParams_internal(CCtxParams->compressionLevel, srcSizeHint, dictSize, mode); if (CCtxParams->ldmParams.enableLdm == ZSTD_ps_enable) cParams.windowLog = ZSTD_LDM_DEFAULT_WINDOW_LOG; - ZSTD_overrideCParams(&cParams, &CCtxParams->cParams); + ZSTD_overrideCParams(&cParams, &CCtxParams->cParams); assert(!ZSTD_checkCParams(cParams)); /* srcSizeHint == 0 means 0 */ - return ZSTD_adjustCParams_internal(cParams, srcSizeHint, dictSize, mode); + return ZSTD_adjustCParams_internal(cParams, srcSizeHint, dictSize, mode); } static size_t ZSTD_sizeof_matchState(const ZSTD_compressionParameters* const cParams, const ZSTD_paramSwitch_e useRowMatchFinder, - const U32 enableDedicatedDictSearch, + const U32 enableDedicatedDictSearch, const U32 forCCtx) { - /* chain table size should be 0 for fast or row-hash strategies */ - size_t const chainSize = ZSTD_allocateChainTable(cParams->strategy, useRowMatchFinder, enableDedicatedDictSearch && !forCCtx) - ? ((size_t)1 << cParams->chainLog) - : 0; + /* chain table size should be 0 for fast or row-hash strategies */ + size_t const chainSize = ZSTD_allocateChainTable(cParams->strategy, useRowMatchFinder, enableDedicatedDictSearch && !forCCtx) + ? ((size_t)1 << cParams->chainLog) + : 0; size_t const hSize = ((size_t)1) << cParams->hashLog; U32 const hashLog3 = (forCCtx && cParams->minMatch==3) ? MIN(ZSTD_HASHLOG3_MAX, cParams->windowLog) : 0; size_t const h3Size = hashLog3 ? ((size_t)1) << hashLog3 : 0; @@ -1432,117 +1432,117 @@ ZSTD_sizeof_matchState(const ZSTD_compressionParameters* const cParams, + hSize * sizeof(U32) + h3Size * sizeof(U32); size_t const optPotentialSpace = - ZSTD_cwksp_aligned_alloc_size((MaxML+1) * sizeof(U32)) - + ZSTD_cwksp_aligned_alloc_size((MaxLL+1) * sizeof(U32)) - + ZSTD_cwksp_aligned_alloc_size((MaxOff+1) * sizeof(U32)) - + ZSTD_cwksp_aligned_alloc_size((1<<Litbits) * sizeof(U32)) - + ZSTD_cwksp_aligned_alloc_size((ZSTD_OPT_NUM+1) * sizeof(ZSTD_match_t)) - + ZSTD_cwksp_aligned_alloc_size((ZSTD_OPT_NUM+1) * sizeof(ZSTD_optimal_t)); - size_t const lazyAdditionalSpace = ZSTD_rowMatchFinderUsed(cParams->strategy, useRowMatchFinder) - ? ZSTD_cwksp_aligned_alloc_size(hSize*sizeof(U16)) - : 0; + ZSTD_cwksp_aligned_alloc_size((MaxML+1) * sizeof(U32)) + + ZSTD_cwksp_aligned_alloc_size((MaxLL+1) * sizeof(U32)) + + ZSTD_cwksp_aligned_alloc_size((MaxOff+1) * sizeof(U32)) + + ZSTD_cwksp_aligned_alloc_size((1<<Litbits) * sizeof(U32)) + + ZSTD_cwksp_aligned_alloc_size((ZSTD_OPT_NUM+1) * sizeof(ZSTD_match_t)) + + ZSTD_cwksp_aligned_alloc_size((ZSTD_OPT_NUM+1) * sizeof(ZSTD_optimal_t)); + size_t const lazyAdditionalSpace = ZSTD_rowMatchFinderUsed(cParams->strategy, useRowMatchFinder) + ? ZSTD_cwksp_aligned_alloc_size(hSize*sizeof(U16)) + : 0; size_t const optSpace = (forCCtx && (cParams->strategy >= ZSTD_btopt)) ? optPotentialSpace : 0; - size_t const slackSpace = ZSTD_cwksp_slack_space_required(); - - /* tables are guaranteed to be sized in multiples of 64 bytes (or 16 uint32_t) */ - ZSTD_STATIC_ASSERT(ZSTD_HASHLOG_MIN >= 4 && ZSTD_WINDOWLOG_MIN >= 4 && ZSTD_CHAINLOG_MIN >= 4); + size_t const slackSpace = ZSTD_cwksp_slack_space_required(); + + /* tables are guaranteed to be sized in multiples of 64 bytes (or 16 uint32_t) */ + ZSTD_STATIC_ASSERT(ZSTD_HASHLOG_MIN >= 4 && ZSTD_WINDOWLOG_MIN >= 4 && ZSTD_CHAINLOG_MIN >= 4); assert(useRowMatchFinder != ZSTD_ps_auto); - + DEBUGLOG(4, "chainSize: %u - hSize: %u - h3Size: %u", (U32)chainSize, (U32)hSize, (U32)h3Size); - return tableSpace + optSpace + slackSpace + lazyAdditionalSpace; + return tableSpace + optSpace + slackSpace + lazyAdditionalSpace; } -static size_t ZSTD_estimateCCtxSize_usingCCtxParams_internal( - const ZSTD_compressionParameters* cParams, - const ldmParams_t* ldmParams, - const int isStatic, +static size_t ZSTD_estimateCCtxSize_usingCCtxParams_internal( + const ZSTD_compressionParameters* cParams, + const ldmParams_t* ldmParams, + const int isStatic, const ZSTD_paramSwitch_e useRowMatchFinder, - const size_t buffInSize, - const size_t buffOutSize, - const U64 pledgedSrcSize) + const size_t buffInSize, + const size_t buffOutSize, + const U64 pledgedSrcSize) { size_t const windowSize = (size_t) BOUNDED(1ULL, 1ULL << cParams->windowLog, pledgedSrcSize); - size_t const blockSize = MIN(ZSTD_BLOCKSIZE_MAX, windowSize); - U32 const divider = (cParams->minMatch==3) ? 3 : 4; - size_t const maxNbSeq = blockSize / divider; - size_t const tokenSpace = ZSTD_cwksp_alloc_size(WILDCOPY_OVERLENGTH + blockSize) - + ZSTD_cwksp_aligned_alloc_size(maxNbSeq * sizeof(seqDef)) - + 3 * ZSTD_cwksp_alloc_size(maxNbSeq * sizeof(BYTE)); - size_t const entropySpace = ZSTD_cwksp_alloc_size(ENTROPY_WORKSPACE_SIZE); - size_t const blockStateSpace = 2 * ZSTD_cwksp_alloc_size(sizeof(ZSTD_compressedBlockState_t)); - size_t const matchStateSize = ZSTD_sizeof_matchState(cParams, useRowMatchFinder, /* enableDedicatedDictSearch */ 0, /* forCCtx */ 1); - - size_t const ldmSpace = ZSTD_ldm_getTableSize(*ldmParams); - size_t const maxNbLdmSeq = ZSTD_ldm_getMaxNbSeq(*ldmParams, blockSize); + size_t const blockSize = MIN(ZSTD_BLOCKSIZE_MAX, windowSize); + U32 const divider = (cParams->minMatch==3) ? 3 : 4; + size_t const maxNbSeq = blockSize / divider; + size_t const tokenSpace = ZSTD_cwksp_alloc_size(WILDCOPY_OVERLENGTH + blockSize) + + ZSTD_cwksp_aligned_alloc_size(maxNbSeq * sizeof(seqDef)) + + 3 * ZSTD_cwksp_alloc_size(maxNbSeq * sizeof(BYTE)); + size_t const entropySpace = ZSTD_cwksp_alloc_size(ENTROPY_WORKSPACE_SIZE); + size_t const blockStateSpace = 2 * ZSTD_cwksp_alloc_size(sizeof(ZSTD_compressedBlockState_t)); + size_t const matchStateSize = ZSTD_sizeof_matchState(cParams, useRowMatchFinder, /* enableDedicatedDictSearch */ 0, /* forCCtx */ 1); + + size_t const ldmSpace = ZSTD_ldm_getTableSize(*ldmParams); + size_t const maxNbLdmSeq = ZSTD_ldm_getMaxNbSeq(*ldmParams, blockSize); size_t const ldmSeqSpace = ldmParams->enableLdm == ZSTD_ps_enable ? - ZSTD_cwksp_aligned_alloc_size(maxNbLdmSeq * sizeof(rawSeq)) : 0; - + ZSTD_cwksp_aligned_alloc_size(maxNbLdmSeq * sizeof(rawSeq)) : 0; - size_t const bufferSpace = ZSTD_cwksp_alloc_size(buffInSize) - + ZSTD_cwksp_alloc_size(buffOutSize); - size_t const cctxSpace = isStatic ? ZSTD_cwksp_alloc_size(sizeof(ZSTD_CCtx)) : 0; + size_t const bufferSpace = ZSTD_cwksp_alloc_size(buffInSize) + + ZSTD_cwksp_alloc_size(buffOutSize); - size_t const neededSpace = - cctxSpace + - entropySpace + - blockStateSpace + - ldmSpace + - ldmSeqSpace + - matchStateSize + - tokenSpace + - bufferSpace; + size_t const cctxSpace = isStatic ? ZSTD_cwksp_alloc_size(sizeof(ZSTD_CCtx)) : 0; - DEBUGLOG(5, "estimate workspace : %u", (U32)neededSpace); - return neededSpace; + size_t const neededSpace = + cctxSpace + + entropySpace + + blockStateSpace + + ldmSpace + + ldmSeqSpace + + matchStateSize + + tokenSpace + + bufferSpace; + + DEBUGLOG(5, "estimate workspace : %u", (U32)neededSpace); + return neededSpace; } -size_t ZSTD_estimateCCtxSize_usingCCtxParams(const ZSTD_CCtx_params* params) -{ - ZSTD_compressionParameters const cParams = - ZSTD_getCParamsFromCCtxParams(params, ZSTD_CONTENTSIZE_UNKNOWN, 0, ZSTD_cpm_noAttachDict); +size_t ZSTD_estimateCCtxSize_usingCCtxParams(const ZSTD_CCtx_params* params) +{ + ZSTD_compressionParameters const cParams = + ZSTD_getCParamsFromCCtxParams(params, ZSTD_CONTENTSIZE_UNKNOWN, 0, ZSTD_cpm_noAttachDict); ZSTD_paramSwitch_e const useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(params->useRowMatchFinder, &cParams); - - RETURN_ERROR_IF(params->nbWorkers > 0, GENERIC, "Estimate CCtx size is supported for single-threaded compression only."); - /* estimateCCtxSize is for one-shot compression. So no buffers should - * be needed. However, we still allocate two 0-sized buffers, which can - * take space under ASAN. */ - return ZSTD_estimateCCtxSize_usingCCtxParams_internal( - &cParams, ¶ms->ldmParams, 1, useRowMatchFinder, 0, 0, ZSTD_CONTENTSIZE_UNKNOWN); -} - + + RETURN_ERROR_IF(params->nbWorkers > 0, GENERIC, "Estimate CCtx size is supported for single-threaded compression only."); + /* estimateCCtxSize is for one-shot compression. So no buffers should + * be needed. However, we still allocate two 0-sized buffers, which can + * take space under ASAN. */ + return ZSTD_estimateCCtxSize_usingCCtxParams_internal( + &cParams, ¶ms->ldmParams, 1, useRowMatchFinder, 0, 0, ZSTD_CONTENTSIZE_UNKNOWN); +} + size_t ZSTD_estimateCCtxSize_usingCParams(ZSTD_compressionParameters cParams) { - ZSTD_CCtx_params initialParams = ZSTD_makeCCtxParamsFromCParams(cParams); - if (ZSTD_rowMatchFinderSupported(cParams.strategy)) { - /* Pick bigger of not using and using row-based matchfinder for greedy and lazy strategies */ - size_t noRowCCtxSize; - size_t rowCCtxSize; + ZSTD_CCtx_params initialParams = ZSTD_makeCCtxParamsFromCParams(cParams); + if (ZSTD_rowMatchFinderSupported(cParams.strategy)) { + /* Pick bigger of not using and using row-based matchfinder for greedy and lazy strategies */ + size_t noRowCCtxSize; + size_t rowCCtxSize; initialParams.useRowMatchFinder = ZSTD_ps_disable; - noRowCCtxSize = ZSTD_estimateCCtxSize_usingCCtxParams(&initialParams); + noRowCCtxSize = ZSTD_estimateCCtxSize_usingCCtxParams(&initialParams); initialParams.useRowMatchFinder = ZSTD_ps_enable; - rowCCtxSize = ZSTD_estimateCCtxSize_usingCCtxParams(&initialParams); - return MAX(noRowCCtxSize, rowCCtxSize); - } else { - return ZSTD_estimateCCtxSize_usingCCtxParams(&initialParams); - } + rowCCtxSize = ZSTD_estimateCCtxSize_usingCCtxParams(&initialParams); + return MAX(noRowCCtxSize, rowCCtxSize); + } else { + return ZSTD_estimateCCtxSize_usingCCtxParams(&initialParams); + } } static size_t ZSTD_estimateCCtxSize_internal(int compressionLevel) { - int tier = 0; - size_t largestSize = 0; - static const unsigned long long srcSizeTiers[4] = {16 KB, 128 KB, 256 KB, ZSTD_CONTENTSIZE_UNKNOWN}; - for (; tier < 4; ++tier) { - /* Choose the set of cParams for a given level across all srcSizes that give the largest cctxSize */ - ZSTD_compressionParameters const cParams = ZSTD_getCParams_internal(compressionLevel, srcSizeTiers[tier], 0, ZSTD_cpm_noAttachDict); - largestSize = MAX(ZSTD_estimateCCtxSize_usingCParams(cParams), largestSize); - } - return largestSize; + int tier = 0; + size_t largestSize = 0; + static const unsigned long long srcSizeTiers[4] = {16 KB, 128 KB, 256 KB, ZSTD_CONTENTSIZE_UNKNOWN}; + for (; tier < 4; ++tier) { + /* Choose the set of cParams for a given level across all srcSizes that give the largest cctxSize */ + ZSTD_compressionParameters const cParams = ZSTD_getCParams_internal(compressionLevel, srcSizeTiers[tier], 0, ZSTD_cpm_noAttachDict); + largestSize = MAX(ZSTD_estimateCCtxSize_usingCParams(cParams), largestSize); + } + return largestSize; } size_t ZSTD_estimateCCtxSize(int compressionLevel) @@ -1550,7 +1550,7 @@ size_t ZSTD_estimateCCtxSize(int compressionLevel) int level; size_t memBudget = 0; for (level=MIN(compressionLevel, 1); level<=compressionLevel; level++) { - /* Ensure monotonically increasing memory usage as compression level increases */ + /* Ensure monotonically increasing memory usage as compression level increases */ size_t const newMB = ZSTD_estimateCCtxSize_internal(level); if (newMB > memBudget) memBudget = newMB; } @@ -1561,42 +1561,42 @@ size_t ZSTD_estimateCStreamSize_usingCCtxParams(const ZSTD_CCtx_params* params) { RETURN_ERROR_IF(params->nbWorkers > 0, GENERIC, "Estimate CCtx size is supported for single-threaded compression only."); { ZSTD_compressionParameters const cParams = - ZSTD_getCParamsFromCCtxParams(params, ZSTD_CONTENTSIZE_UNKNOWN, 0, ZSTD_cpm_noAttachDict); + ZSTD_getCParamsFromCCtxParams(params, ZSTD_CONTENTSIZE_UNKNOWN, 0, ZSTD_cpm_noAttachDict); size_t const blockSize = MIN(ZSTD_BLOCKSIZE_MAX, (size_t)1 << cParams.windowLog); - size_t const inBuffSize = (params->inBufferMode == ZSTD_bm_buffered) - ? ((size_t)1 << cParams.windowLog) + blockSize - : 0; - size_t const outBuffSize = (params->outBufferMode == ZSTD_bm_buffered) - ? ZSTD_compressBound(blockSize) + 1 - : 0; + size_t const inBuffSize = (params->inBufferMode == ZSTD_bm_buffered) + ? ((size_t)1 << cParams.windowLog) + blockSize + : 0; + size_t const outBuffSize = (params->outBufferMode == ZSTD_bm_buffered) + ? ZSTD_compressBound(blockSize) + 1 + : 0; ZSTD_paramSwitch_e const useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(params->useRowMatchFinder, ¶ms->cParams); - return ZSTD_estimateCCtxSize_usingCCtxParams_internal( - &cParams, ¶ms->ldmParams, 1, useRowMatchFinder, inBuffSize, outBuffSize, - ZSTD_CONTENTSIZE_UNKNOWN); + return ZSTD_estimateCCtxSize_usingCCtxParams_internal( + &cParams, ¶ms->ldmParams, 1, useRowMatchFinder, inBuffSize, outBuffSize, + ZSTD_CONTENTSIZE_UNKNOWN); } } size_t ZSTD_estimateCStreamSize_usingCParams(ZSTD_compressionParameters cParams) { - ZSTD_CCtx_params initialParams = ZSTD_makeCCtxParamsFromCParams(cParams); - if (ZSTD_rowMatchFinderSupported(cParams.strategy)) { - /* Pick bigger of not using and using row-based matchfinder for greedy and lazy strategies */ - size_t noRowCCtxSize; - size_t rowCCtxSize; + ZSTD_CCtx_params initialParams = ZSTD_makeCCtxParamsFromCParams(cParams); + if (ZSTD_rowMatchFinderSupported(cParams.strategy)) { + /* Pick bigger of not using and using row-based matchfinder for greedy and lazy strategies */ + size_t noRowCCtxSize; + size_t rowCCtxSize; initialParams.useRowMatchFinder = ZSTD_ps_disable; - noRowCCtxSize = ZSTD_estimateCStreamSize_usingCCtxParams(&initialParams); + noRowCCtxSize = ZSTD_estimateCStreamSize_usingCCtxParams(&initialParams); initialParams.useRowMatchFinder = ZSTD_ps_enable; - rowCCtxSize = ZSTD_estimateCStreamSize_usingCCtxParams(&initialParams); - return MAX(noRowCCtxSize, rowCCtxSize); - } else { - return ZSTD_estimateCStreamSize_usingCCtxParams(&initialParams); - } + rowCCtxSize = ZSTD_estimateCStreamSize_usingCCtxParams(&initialParams); + return MAX(noRowCCtxSize, rowCCtxSize); + } else { + return ZSTD_estimateCStreamSize_usingCCtxParams(&initialParams); + } } static size_t ZSTD_estimateCStreamSize_internal(int compressionLevel) { - ZSTD_compressionParameters const cParams = ZSTD_getCParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, 0, ZSTD_cpm_noAttachDict); + ZSTD_compressionParameters const cParams = ZSTD_getCParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, 0, ZSTD_cpm_noAttachDict); return ZSTD_estimateCStreamSize_usingCParams(cParams); } @@ -1716,7 +1716,7 @@ typedef enum { ZSTD_resetTarget_CCtx } ZSTD_resetTarget_e; - + static size_t ZSTD_reset_matchState(ZSTD_matchState_t* ms, ZSTD_cwksp* ws, @@ -1726,11 +1726,11 @@ ZSTD_reset_matchState(ZSTD_matchState_t* ms, const ZSTD_indexResetPolicy_e forceResetIndex, const ZSTD_resetTarget_e forWho) { - /* disable chain table allocation for fast or row-based strategies */ - size_t const chainSize = ZSTD_allocateChainTable(cParams->strategy, useRowMatchFinder, - ms->dedicatedDictSearch && (forWho == ZSTD_resetTarget_CDict)) - ? ((size_t)1 << cParams->chainLog) - : 0; + /* disable chain table allocation for fast or row-based strategies */ + size_t const chainSize = ZSTD_allocateChainTable(cParams->strategy, useRowMatchFinder, + ms->dedicatedDictSearch && (forWho == ZSTD_resetTarget_CDict)) + ? ((size_t)1 << cParams->chainLog) + : 0; size_t const hSize = ((size_t)1) << cParams->hashLog; U32 const hashLog3 = ((forWho == ZSTD_resetTarget_CCtx) && cParams->minMatch==3) ? MIN(ZSTD_HASHLOG3_MAX, cParams->windowLog) : 0; size_t const h3Size = hashLog3 ? ((size_t)1) << hashLog3 : 0; @@ -1775,19 +1775,19 @@ ZSTD_reset_matchState(ZSTD_matchState_t* ms, ms->opt.priceTable = (ZSTD_optimal_t*)ZSTD_cwksp_reserve_aligned(ws, (ZSTD_OPT_NUM+1) * sizeof(ZSTD_optimal_t)); } - if (ZSTD_rowMatchFinderUsed(cParams->strategy, useRowMatchFinder)) { - { /* Row match finder needs an additional table of hashes ("tags") */ - size_t const tagTableSize = hSize*sizeof(U16); - ms->tagTable = (U16*)ZSTD_cwksp_reserve_aligned(ws, tagTableSize); - if (ms->tagTable) ZSTD_memset(ms->tagTable, 0, tagTableSize); - } - { /* Switch to 32-entry rows if searchLog is 5 (or more) */ + if (ZSTD_rowMatchFinderUsed(cParams->strategy, useRowMatchFinder)) { + { /* Row match finder needs an additional table of hashes ("tags") */ + size_t const tagTableSize = hSize*sizeof(U16); + ms->tagTable = (U16*)ZSTD_cwksp_reserve_aligned(ws, tagTableSize); + if (ms->tagTable) ZSTD_memset(ms->tagTable, 0, tagTableSize); + } + { /* Switch to 32-entry rows if searchLog is 5 (or more) */ U32 const rowLog = BOUNDED(4, cParams->searchLog, 6); assert(cParams->hashLog >= rowLog); - ms->rowHashLog = cParams->hashLog - rowLog; - } - } - + ms->rowHashLog = cParams->hashLog - rowLog; + } + } + ms->cParams = *cParams; RETURN_ERROR_IF(ZSTD_cwksp_reserve_failed(ws), memory_allocation, @@ -1808,87 +1808,87 @@ static int ZSTD_indexTooCloseToMax(ZSTD_window_t w) return (size_t)(w.nextSrc - w.base) > (ZSTD_CURRENT_MAX - ZSTD_INDEXOVERFLOW_MARGIN); } -/** ZSTD_dictTooBig(): - * When dictionaries are larger than ZSTD_CHUNKSIZE_MAX they can't be loaded in - * one go generically. So we ensure that in that case we reset the tables to zero, - * so that we can load as much of the dictionary as possible. - */ -static int ZSTD_dictTooBig(size_t const loadedDictSize) -{ - return loadedDictSize > ZSTD_CHUNKSIZE_MAX; -} - +/** ZSTD_dictTooBig(): + * When dictionaries are larger than ZSTD_CHUNKSIZE_MAX they can't be loaded in + * one go generically. So we ensure that in that case we reset the tables to zero, + * so that we can load as much of the dictionary as possible. + */ +static int ZSTD_dictTooBig(size_t const loadedDictSize) +{ + return loadedDictSize > ZSTD_CHUNKSIZE_MAX; +} + /*! ZSTD_resetCCtx_internal() : - * @param loadedDictSize The size of the dictionary to be loaded - * into the context, if any. If no dictionary is used, or the - * dictionary is being attached / copied, then pass 0. - * note : `params` are assumed fully validated at this stage. - */ + * @param loadedDictSize The size of the dictionary to be loaded + * into the context, if any. If no dictionary is used, or the + * dictionary is being attached / copied, then pass 0. + * note : `params` are assumed fully validated at this stage. + */ static size_t ZSTD_resetCCtx_internal(ZSTD_CCtx* zc, - ZSTD_CCtx_params const* params, + ZSTD_CCtx_params const* params, U64 const pledgedSrcSize, - size_t const loadedDictSize, + size_t const loadedDictSize, ZSTD_compResetPolicy_e const crp, ZSTD_buffered_policy_e const zbuff) { ZSTD_cwksp* const ws = &zc->workspace; DEBUGLOG(4, "ZSTD_resetCCtx_internal: pledgedSrcSize=%u, wlog=%u, useRowMatchFinder=%d useBlockSplitter=%d", (U32)pledgedSrcSize, params->cParams.windowLog, (int)params->useRowMatchFinder, (int)params->useBlockSplitter); - assert(!ZSTD_isError(ZSTD_checkCParams(params->cParams))); + assert(!ZSTD_isError(ZSTD_checkCParams(params->cParams))); zc->isFirstBlock = 1; - /* Set applied params early so we can modify them for LDM, - * and point params at the applied params. - */ - zc->appliedParams = *params; - params = &zc->appliedParams; - + /* Set applied params early so we can modify them for LDM, + * and point params at the applied params. + */ + zc->appliedParams = *params; + params = &zc->appliedParams; + assert(params->useRowMatchFinder != ZSTD_ps_auto); assert(params->useBlockSplitter != ZSTD_ps_auto); assert(params->ldmParams.enableLdm != ZSTD_ps_auto); if (params->ldmParams.enableLdm == ZSTD_ps_enable) { /* Adjust long distance matching parameters */ - ZSTD_ldm_adjustParameters(&zc->appliedParams.ldmParams, ¶ms->cParams); - assert(params->ldmParams.hashLog >= params->ldmParams.bucketSizeLog); - assert(params->ldmParams.hashRateLog < 32); + ZSTD_ldm_adjustParameters(&zc->appliedParams.ldmParams, ¶ms->cParams); + assert(params->ldmParams.hashLog >= params->ldmParams.bucketSizeLog); + assert(params->ldmParams.hashRateLog < 32); } - { size_t const windowSize = MAX(1, (size_t)MIN(((U64)1 << params->cParams.windowLog), pledgedSrcSize)); + { size_t const windowSize = MAX(1, (size_t)MIN(((U64)1 << params->cParams.windowLog), pledgedSrcSize)); size_t const blockSize = MIN(ZSTD_BLOCKSIZE_MAX, windowSize); - U32 const divider = (params->cParams.minMatch==3) ? 3 : 4; + U32 const divider = (params->cParams.minMatch==3) ? 3 : 4; size_t const maxNbSeq = blockSize / divider; - size_t const buffOutSize = (zbuff == ZSTDb_buffered && params->outBufferMode == ZSTD_bm_buffered) - ? ZSTD_compressBound(blockSize) + 1 - : 0; - size_t const buffInSize = (zbuff == ZSTDb_buffered && params->inBufferMode == ZSTD_bm_buffered) - ? windowSize + blockSize - : 0; - size_t const maxNbLdmSeq = ZSTD_ldm_getMaxNbSeq(params->ldmParams, blockSize); - - int const indexTooClose = ZSTD_indexTooCloseToMax(zc->blockState.matchState.window); - int const dictTooBig = ZSTD_dictTooBig(loadedDictSize); - ZSTD_indexResetPolicy_e needsIndexReset = - (indexTooClose || dictTooBig || !zc->initialized) ? ZSTDirp_reset : ZSTDirp_continue; - - size_t const neededSpace = - ZSTD_estimateCCtxSize_usingCCtxParams_internal( - ¶ms->cParams, ¶ms->ldmParams, zc->staticSize != 0, params->useRowMatchFinder, - buffInSize, buffOutSize, pledgedSrcSize); - int resizeWorkspace; - - FORWARD_IF_ERROR(neededSpace, "cctx size estimate failed!"); + size_t const buffOutSize = (zbuff == ZSTDb_buffered && params->outBufferMode == ZSTD_bm_buffered) + ? ZSTD_compressBound(blockSize) + 1 + : 0; + size_t const buffInSize = (zbuff == ZSTDb_buffered && params->inBufferMode == ZSTD_bm_buffered) + ? windowSize + blockSize + : 0; + size_t const maxNbLdmSeq = ZSTD_ldm_getMaxNbSeq(params->ldmParams, blockSize); + + int const indexTooClose = ZSTD_indexTooCloseToMax(zc->blockState.matchState.window); + int const dictTooBig = ZSTD_dictTooBig(loadedDictSize); + ZSTD_indexResetPolicy_e needsIndexReset = + (indexTooClose || dictTooBig || !zc->initialized) ? ZSTDirp_reset : ZSTDirp_continue; + + size_t const neededSpace = + ZSTD_estimateCCtxSize_usingCCtxParams_internal( + ¶ms->cParams, ¶ms->ldmParams, zc->staticSize != 0, params->useRowMatchFinder, + buffInSize, buffOutSize, pledgedSrcSize); + int resizeWorkspace; + + FORWARD_IF_ERROR(neededSpace, "cctx size estimate failed!"); if (!zc->staticSize) ZSTD_cwksp_bump_oversized_duration(ws, 0); - { /* Check if workspace is large enough, alloc a new one if needed */ + { /* Check if workspace is large enough, alloc a new one if needed */ int const workspaceTooSmall = ZSTD_cwksp_sizeof(ws) < neededSpace; int const workspaceWasteful = ZSTD_cwksp_check_wasteful(ws, neededSpace); - resizeWorkspace = workspaceTooSmall || workspaceWasteful; - DEBUGLOG(4, "Need %zu B workspace", neededSpace); + resizeWorkspace = workspaceTooSmall || workspaceWasteful; + DEBUGLOG(4, "Need %zu B workspace", neededSpace); DEBUGLOG(4, "windowSize: %zu - blockSize: %zu", windowSize, blockSize); - if (resizeWorkspace) { + if (resizeWorkspace) { DEBUGLOG(4, "Resize workspaceSize from %zuKB to %zuKB", ZSTD_cwksp_sizeof(ws) >> 10, neededSpace >> 10); @@ -1909,14 +1909,14 @@ static size_t ZSTD_resetCCtx_internal(ZSTD_CCtx* zc, RETURN_ERROR_IF(zc->blockState.prevCBlock == NULL, memory_allocation, "couldn't allocate prevCBlock"); zc->blockState.nextCBlock = (ZSTD_compressedBlockState_t*) ZSTD_cwksp_reserve_object(ws, sizeof(ZSTD_compressedBlockState_t)); RETURN_ERROR_IF(zc->blockState.nextCBlock == NULL, memory_allocation, "couldn't allocate nextCBlock"); - zc->entropyWorkspace = (U32*) ZSTD_cwksp_reserve_object(ws, ENTROPY_WORKSPACE_SIZE); + zc->entropyWorkspace = (U32*) ZSTD_cwksp_reserve_object(ws, ENTROPY_WORKSPACE_SIZE); RETURN_ERROR_IF(zc->entropyWorkspace == NULL, memory_allocation, "couldn't allocate entropyWorkspace"); } } ZSTD_cwksp_clear(ws); /* init params */ - zc->blockState.matchState.cParams = params->cParams; + zc->blockState.matchState.cParams = params->cParams; zc->pledgedSrcSizePlusOne = pledgedSrcSize+1; zc->consumedSrcSize = 0; zc->producedCSize = 0; @@ -1929,7 +1929,7 @@ static size_t ZSTD_resetCCtx_internal(ZSTD_CCtx* zc, XXH64_reset(&zc->xxhState, 0); zc->stage = ZSTDcs_init; zc->dictID = 0; - zc->dictContentSize = 0; + zc->dictContentSize = 0; ZSTD_reset_compressedBlockState(zc->blockState.prevCBlock); @@ -1940,7 +1940,7 @@ static size_t ZSTD_resetCCtx_internal(ZSTD_CCtx* zc, zc->seqStore.maxNbLit = blockSize; /* buffers */ - zc->bufferedPolicy = zbuff; + zc->bufferedPolicy = zbuff; zc->inBuffSize = buffInSize; zc->inBuff = (char*)ZSTD_cwksp_reserve_buffer(ws, buffInSize); zc->outBuffSize = buffOutSize; @@ -1949,11 +1949,11 @@ static size_t ZSTD_resetCCtx_internal(ZSTD_CCtx* zc, /* ldm bucketOffsets table */ if (params->ldmParams.enableLdm == ZSTD_ps_enable) { /* TODO: avoid memset? */ - size_t const numBuckets = - ((size_t)1) << (params->ldmParams.hashLog - - params->ldmParams.bucketSizeLog); - zc->ldmState.bucketOffsets = ZSTD_cwksp_reserve_buffer(ws, numBuckets); - ZSTD_memset(zc->ldmState.bucketOffsets, 0, numBuckets); + size_t const numBuckets = + ((size_t)1) << (params->ldmParams.hashLog - + params->ldmParams.bucketSizeLog); + zc->ldmState.bucketOffsets = ZSTD_cwksp_reserve_buffer(ws, numBuckets); + ZSTD_memset(zc->ldmState.bucketOffsets, 0, numBuckets); } /* sequences storage */ @@ -1967,8 +1967,8 @@ static size_t ZSTD_resetCCtx_internal(ZSTD_CCtx* zc, FORWARD_IF_ERROR(ZSTD_reset_matchState( &zc->blockState.matchState, ws, - ¶ms->cParams, - params->useRowMatchFinder, + ¶ms->cParams, + params->useRowMatchFinder, crp, needsIndexReset, ZSTD_resetTarget_CCtx), ""); @@ -1976,9 +1976,9 @@ static size_t ZSTD_resetCCtx_internal(ZSTD_CCtx* zc, /* ldm hash table */ if (params->ldmParams.enableLdm == ZSTD_ps_enable) { /* TODO: avoid memset? */ - size_t const ldmHSize = ((size_t)1) << params->ldmParams.hashLog; + size_t const ldmHSize = ((size_t)1) << params->ldmParams.hashLog; zc->ldmState.hashTable = (ldmEntry_t*)ZSTD_cwksp_reserve_aligned(ws, ldmHSize * sizeof(ldmEntry_t)); - ZSTD_memset(zc->ldmState.hashTable, 0, ldmHSize * sizeof(ldmEntry_t)); + ZSTD_memset(zc->ldmState.hashTable, 0, ldmHSize * sizeof(ldmEntry_t)); zc->ldmSequences = (rawSeq*)ZSTD_cwksp_reserve_aligned(ws, maxNbLdmSeq * sizeof(rawSeq)); zc->maxNbLdmSequences = maxNbLdmSeq; @@ -1987,8 +1987,8 @@ static size_t ZSTD_resetCCtx_internal(ZSTD_CCtx* zc, } DEBUGLOG(3, "wksp: finished allocating, %zd bytes remain available", ZSTD_cwksp_available_space(ws)); - assert(ZSTD_cwksp_estimated_space_within_bounds(ws, neededSpace, resizeWorkspace)); - + assert(ZSTD_cwksp_estimated_space_within_bounds(ws, neededSpace, resizeWorkspace)); + zc->initialized = 1; return 0; @@ -2027,14 +2027,14 @@ static int ZSTD_shouldAttachDict(const ZSTD_CDict* cdict, U64 pledgedSrcSize) { size_t cutoff = attachDictSizeCutoffs[cdict->matchState.cParams.strategy]; - int const dedicatedDictSearch = cdict->matchState.dedicatedDictSearch; - return dedicatedDictSearch - || ( ( pledgedSrcSize <= cutoff - || pledgedSrcSize == ZSTD_CONTENTSIZE_UNKNOWN - || params->attachDictPref == ZSTD_dictForceAttach ) - && params->attachDictPref != ZSTD_dictForceCopy - && !params->forceWindow ); /* dictMatchState isn't correctly - * handled in _enforceMaxDist */ + int const dedicatedDictSearch = cdict->matchState.dedicatedDictSearch; + return dedicatedDictSearch + || ( ( pledgedSrcSize <= cutoff + || pledgedSrcSize == ZSTD_CONTENTSIZE_UNKNOWN + || params->attachDictPref == ZSTD_dictForceAttach ) + && params->attachDictPref != ZSTD_dictForceCopy + && !params->forceWindow ); /* dictMatchState isn't correctly + * handled in _enforceMaxDist */ } static size_t @@ -2044,28 +2044,28 @@ ZSTD_resetCCtx_byAttachingCDict(ZSTD_CCtx* cctx, U64 pledgedSrcSize, ZSTD_buffered_policy_e zbuff) { - DEBUGLOG(4, "ZSTD_resetCCtx_byAttachingCDict() pledgedSrcSize=%llu", - (unsigned long long)pledgedSrcSize); - { - ZSTD_compressionParameters adjusted_cdict_cParams = cdict->matchState.cParams; + DEBUGLOG(4, "ZSTD_resetCCtx_byAttachingCDict() pledgedSrcSize=%llu", + (unsigned long long)pledgedSrcSize); + { + ZSTD_compressionParameters adjusted_cdict_cParams = cdict->matchState.cParams; unsigned const windowLog = params.cParams.windowLog; assert(windowLog != 0); /* Resize working context table params for input only, since the dict * has its own tables. */ - /* pledgedSrcSize == 0 means 0! */ - - if (cdict->matchState.dedicatedDictSearch) { - ZSTD_dedicatedDictSearch_revertCParams(&adjusted_cdict_cParams); - } - - params.cParams = ZSTD_adjustCParams_internal(adjusted_cdict_cParams, pledgedSrcSize, - cdict->dictContentSize, ZSTD_cpm_attachDict); + /* pledgedSrcSize == 0 means 0! */ + + if (cdict->matchState.dedicatedDictSearch) { + ZSTD_dedicatedDictSearch_revertCParams(&adjusted_cdict_cParams); + } + + params.cParams = ZSTD_adjustCParams_internal(adjusted_cdict_cParams, pledgedSrcSize, + cdict->dictContentSize, ZSTD_cpm_attachDict); params.cParams.windowLog = windowLog; - params.useRowMatchFinder = cdict->useRowMatchFinder; /* cdict overrides */ - FORWARD_IF_ERROR(ZSTD_resetCCtx_internal(cctx, ¶ms, pledgedSrcSize, - /* loadedDictSize */ 0, + params.useRowMatchFinder = cdict->useRowMatchFinder; /* cdict overrides */ + FORWARD_IF_ERROR(ZSTD_resetCCtx_internal(cctx, ¶ms, pledgedSrcSize, + /* loadedDictSize */ 0, ZSTDcrp_makeClean, zbuff), ""); - assert(cctx->appliedParams.cParams.strategy == adjusted_cdict_cParams.strategy); + assert(cctx->appliedParams.cParams.strategy == adjusted_cdict_cParams.strategy); } { const U32 cdictEnd = (U32)( cdict->matchState.window.nextSrc @@ -2090,10 +2090,10 @@ ZSTD_resetCCtx_byAttachingCDict(ZSTD_CCtx* cctx, } } cctx->dictID = cdict->dictID; - cctx->dictContentSize = cdict->dictContentSize; + cctx->dictContentSize = cdict->dictContentSize; /* copy block state */ - ZSTD_memcpy(cctx->blockState.prevCBlock, &cdict->cBlockState, sizeof(cdict->cBlockState)); + ZSTD_memcpy(cctx->blockState.prevCBlock, &cdict->cBlockState, sizeof(cdict->cBlockState)); return 0; } @@ -2106,18 +2106,18 @@ static size_t ZSTD_resetCCtx_byCopyingCDict(ZSTD_CCtx* cctx, { const ZSTD_compressionParameters *cdict_cParams = &cdict->matchState.cParams; - assert(!cdict->matchState.dedicatedDictSearch); - DEBUGLOG(4, "ZSTD_resetCCtx_byCopyingCDict() pledgedSrcSize=%llu", - (unsigned long long)pledgedSrcSize); - + assert(!cdict->matchState.dedicatedDictSearch); + DEBUGLOG(4, "ZSTD_resetCCtx_byCopyingCDict() pledgedSrcSize=%llu", + (unsigned long long)pledgedSrcSize); + { unsigned const windowLog = params.cParams.windowLog; assert(windowLog != 0); /* Copy only compression parameters related to tables. */ params.cParams = *cdict_cParams; params.cParams.windowLog = windowLog; - params.useRowMatchFinder = cdict->useRowMatchFinder; - FORWARD_IF_ERROR(ZSTD_resetCCtx_internal(cctx, ¶ms, pledgedSrcSize, - /* loadedDictSize */ 0, + params.useRowMatchFinder = cdict->useRowMatchFinder; + FORWARD_IF_ERROR(ZSTD_resetCCtx_internal(cctx, ¶ms, pledgedSrcSize, + /* loadedDictSize */ 0, ZSTDcrp_leaveDirty, zbuff), ""); assert(cctx->appliedParams.cParams.strategy == cdict_cParams->strategy); assert(cctx->appliedParams.cParams.hashLog == cdict_cParams->hashLog); @@ -2128,34 +2128,34 @@ static size_t ZSTD_resetCCtx_byCopyingCDict(ZSTD_CCtx* cctx, assert(params.useRowMatchFinder != ZSTD_ps_auto); /* copy tables */ - { size_t const chainSize = ZSTD_allocateChainTable(cdict_cParams->strategy, cdict->useRowMatchFinder, 0 /* DDS guaranteed disabled */) - ? ((size_t)1 << cdict_cParams->chainLog) - : 0; + { size_t const chainSize = ZSTD_allocateChainTable(cdict_cParams->strategy, cdict->useRowMatchFinder, 0 /* DDS guaranteed disabled */) + ? ((size_t)1 << cdict_cParams->chainLog) + : 0; size_t const hSize = (size_t)1 << cdict_cParams->hashLog; - ZSTD_memcpy(cctx->blockState.matchState.hashTable, + ZSTD_memcpy(cctx->blockState.matchState.hashTable, cdict->matchState.hashTable, hSize * sizeof(U32)); - /* Do not copy cdict's chainTable if cctx has parameters such that it would not use chainTable */ - if (ZSTD_allocateChainTable(cctx->appliedParams.cParams.strategy, cctx->appliedParams.useRowMatchFinder, 0 /* forDDSDict */)) { - ZSTD_memcpy(cctx->blockState.matchState.chainTable, + /* Do not copy cdict's chainTable if cctx has parameters such that it would not use chainTable */ + if (ZSTD_allocateChainTable(cctx->appliedParams.cParams.strategy, cctx->appliedParams.useRowMatchFinder, 0 /* forDDSDict */)) { + ZSTD_memcpy(cctx->blockState.matchState.chainTable, cdict->matchState.chainTable, chainSize * sizeof(U32)); - } - /* copy tag table */ - if (ZSTD_rowMatchFinderUsed(cdict_cParams->strategy, cdict->useRowMatchFinder)) { - size_t const tagTableSize = hSize*sizeof(U16); - ZSTD_memcpy(cctx->blockState.matchState.tagTable, - cdict->matchState.tagTable, - tagTableSize); - } + } + /* copy tag table */ + if (ZSTD_rowMatchFinderUsed(cdict_cParams->strategy, cdict->useRowMatchFinder)) { + size_t const tagTableSize = hSize*sizeof(U16); + ZSTD_memcpy(cctx->blockState.matchState.tagTable, + cdict->matchState.tagTable, + tagTableSize); + } } /* Zero the hashTable3, since the cdict never fills it */ { int const h3log = cctx->blockState.matchState.hashLog3; size_t const h3Size = h3log ? ((size_t)1 << h3log) : 0; assert(cdict->matchState.hashLog3 == 0); - ZSTD_memset(cctx->blockState.matchState.hashTable3, 0, h3Size * sizeof(U32)); + ZSTD_memset(cctx->blockState.matchState.hashTable3, 0, h3Size * sizeof(U32)); } ZSTD_cwksp_mark_tables_clean(&cctx->workspace); @@ -2169,10 +2169,10 @@ static size_t ZSTD_resetCCtx_byCopyingCDict(ZSTD_CCtx* cctx, } cctx->dictID = cdict->dictID; - cctx->dictContentSize = cdict->dictContentSize; + cctx->dictContentSize = cdict->dictContentSize; /* copy block state */ - ZSTD_memcpy(cctx->blockState.prevCBlock, &cdict->cBlockState, sizeof(cdict->cBlockState)); + ZSTD_memcpy(cctx->blockState.prevCBlock, &cdict->cBlockState, sizeof(cdict->cBlockState)); return 0; } @@ -2214,20 +2214,20 @@ static size_t ZSTD_copyCCtx_internal(ZSTD_CCtx* dstCCtx, { RETURN_ERROR_IF(srcCCtx->stage!=ZSTDcs_init, stage_wrong, "Can't copy a ctx that's not in init stage."); - DEBUGLOG(5, "ZSTD_copyCCtx_internal"); - ZSTD_memcpy(&dstCCtx->customMem, &srcCCtx->customMem, sizeof(ZSTD_customMem)); + DEBUGLOG(5, "ZSTD_copyCCtx_internal"); + ZSTD_memcpy(&dstCCtx->customMem, &srcCCtx->customMem, sizeof(ZSTD_customMem)); { ZSTD_CCtx_params params = dstCCtx->requestedParams; /* Copy only compression parameters related to tables. */ params.cParams = srcCCtx->appliedParams.cParams; assert(srcCCtx->appliedParams.useRowMatchFinder != ZSTD_ps_auto); assert(srcCCtx->appliedParams.useBlockSplitter != ZSTD_ps_auto); assert(srcCCtx->appliedParams.ldmParams.enableLdm != ZSTD_ps_auto); - params.useRowMatchFinder = srcCCtx->appliedParams.useRowMatchFinder; + params.useRowMatchFinder = srcCCtx->appliedParams.useRowMatchFinder; params.useBlockSplitter = srcCCtx->appliedParams.useBlockSplitter; params.ldmParams = srcCCtx->appliedParams.ldmParams; params.fParams = fParams; - ZSTD_resetCCtx_internal(dstCCtx, ¶ms, pledgedSrcSize, - /* loadedDictSize */ 0, + ZSTD_resetCCtx_internal(dstCCtx, ¶ms, pledgedSrcSize, + /* loadedDictSize */ 0, ZSTDcrp_leaveDirty, zbuff); assert(dstCCtx->appliedParams.cParams.windowLog == srcCCtx->appliedParams.cParams.windowLog); assert(dstCCtx->appliedParams.cParams.strategy == srcCCtx->appliedParams.cParams.strategy); @@ -2239,22 +2239,22 @@ static size_t ZSTD_copyCCtx_internal(ZSTD_CCtx* dstCCtx, ZSTD_cwksp_mark_tables_dirty(&dstCCtx->workspace); /* copy tables */ - { size_t const chainSize = ZSTD_allocateChainTable(srcCCtx->appliedParams.cParams.strategy, - srcCCtx->appliedParams.useRowMatchFinder, - 0 /* forDDSDict */) - ? ((size_t)1 << srcCCtx->appliedParams.cParams.chainLog) - : 0; + { size_t const chainSize = ZSTD_allocateChainTable(srcCCtx->appliedParams.cParams.strategy, + srcCCtx->appliedParams.useRowMatchFinder, + 0 /* forDDSDict */) + ? ((size_t)1 << srcCCtx->appliedParams.cParams.chainLog) + : 0; size_t const hSize = (size_t)1 << srcCCtx->appliedParams.cParams.hashLog; int const h3log = srcCCtx->blockState.matchState.hashLog3; size_t const h3Size = h3log ? ((size_t)1 << h3log) : 0; - ZSTD_memcpy(dstCCtx->blockState.matchState.hashTable, + ZSTD_memcpy(dstCCtx->blockState.matchState.hashTable, srcCCtx->blockState.matchState.hashTable, hSize * sizeof(U32)); - ZSTD_memcpy(dstCCtx->blockState.matchState.chainTable, + ZSTD_memcpy(dstCCtx->blockState.matchState.chainTable, srcCCtx->blockState.matchState.chainTable, chainSize * sizeof(U32)); - ZSTD_memcpy(dstCCtx->blockState.matchState.hashTable3, + ZSTD_memcpy(dstCCtx->blockState.matchState.hashTable3, srcCCtx->blockState.matchState.hashTable3, h3Size * sizeof(U32)); } @@ -2270,10 +2270,10 @@ static size_t ZSTD_copyCCtx_internal(ZSTD_CCtx* dstCCtx, dstMatchState->loadedDictEnd= srcMatchState->loadedDictEnd; } dstCCtx->dictID = srcCCtx->dictID; - dstCCtx->dictContentSize = srcCCtx->dictContentSize; + dstCCtx->dictContentSize = srcCCtx->dictContentSize; /* copy block state */ - ZSTD_memcpy(dstCCtx->blockState.prevCBlock, srcCCtx->blockState.prevCBlock, sizeof(*srcCCtx->blockState.prevCBlock)); + ZSTD_memcpy(dstCCtx->blockState.prevCBlock, srcCCtx->blockState.prevCBlock, sizeof(*srcCCtx->blockState.prevCBlock)); return 0; } @@ -2286,7 +2286,7 @@ static size_t ZSTD_copyCCtx_internal(ZSTD_CCtx* dstCCtx, size_t ZSTD_copyCCtx(ZSTD_CCtx* dstCCtx, const ZSTD_CCtx* srcCCtx, unsigned long long pledgedSrcSize) { ZSTD_frameParameters fParams = { 1 /*content*/, 0 /*checksum*/, 0 /*noDictID*/ }; - ZSTD_buffered_policy_e const zbuff = srcCCtx->bufferedPolicy; + ZSTD_buffered_policy_e const zbuff = srcCCtx->bufferedPolicy; ZSTD_STATIC_ASSERT((U32)ZSTDb_buffered==1); if (pledgedSrcSize==0) pledgedSrcSize = ZSTD_CONTENTSIZE_UNKNOWN; fParams.contentSizeFlag = (pledgedSrcSize != ZSTD_CONTENTSIZE_UNKNOWN); @@ -2315,7 +2315,7 @@ ZSTD_reduceTable_internal (U32* const table, U32 const size, U32 const reducerVa assert((size & (ZSTD_ROWSIZE-1)) == 0); /* multiple of ZSTD_ROWSIZE */ assert(size < (1U<<31)); /* can be casted to int */ -#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE) +#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE) /* To validate that the table re-use logic is sound, and that we don't * access table space that we haven't cleaned, we re-"poison" the table * space every time we mark it dirty. @@ -2364,7 +2364,7 @@ static void ZSTD_reduceIndex (ZSTD_matchState_t* ms, ZSTD_CCtx_params const* par ZSTD_reduceTable(ms->hashTable, hSize, reducerValue); } - if (ZSTD_allocateChainTable(params->cParams.strategy, params->useRowMatchFinder, (U32)ms->dedicatedDictSearch)) { + if (ZSTD_allocateChainTable(params->cParams.strategy, params->useRowMatchFinder, (U32)ms->dedicatedDictSearch)) { U32 const chainSize = (U32)1 << params->cParams.chainLog; if (params->cParams.strategy == ZSTD_btlazy2) ZSTD_reduceTable_btlazy2(ms->chainTable, chainSize, reducerValue); @@ -2401,9 +2401,9 @@ void ZSTD_seqToCodes(const seqStore_t* seqStorePtr) ofCodeTable[u] = (BYTE)ZSTD_highbit32(sequences[u].offBase); mlCodeTable[u] = (BYTE)ZSTD_MLcode(mlv); } - if (seqStorePtr->longLengthType==ZSTD_llt_literalLength) + if (seqStorePtr->longLengthType==ZSTD_llt_literalLength) llCodeTable[seqStorePtr->longLengthPos] = MaxLL; - if (seqStorePtr->longLengthType==ZSTD_llt_matchLength) + if (seqStorePtr->longLengthType==ZSTD_llt_matchLength) mlCodeTable[seqStorePtr->longLengthPos] = MaxML; } @@ -2417,161 +2417,161 @@ static int ZSTD_useTargetCBlockSize(const ZSTD_CCtx_params* cctxParams) return (cctxParams->targetCBlockSize != 0); } -/* ZSTD_blockSplitterEnabled(): - * Returns if block splitting param is being used - * If used, compression will do best effort to split a block in order to improve compression ratio. +/* ZSTD_blockSplitterEnabled(): + * Returns if block splitting param is being used + * If used, compression will do best effort to split a block in order to improve compression ratio. * At the time this function is called, the parameter must be finalized. - * Returns 1 if true, 0 otherwise. */ -static int ZSTD_blockSplitterEnabled(ZSTD_CCtx_params* cctxParams) -{ + * Returns 1 if true, 0 otherwise. */ +static int ZSTD_blockSplitterEnabled(ZSTD_CCtx_params* cctxParams) +{ DEBUGLOG(5, "ZSTD_blockSplitterEnabled (useBlockSplitter=%d)", cctxParams->useBlockSplitter); assert(cctxParams->useBlockSplitter != ZSTD_ps_auto); return (cctxParams->useBlockSplitter == ZSTD_ps_enable); -} - -/* Type returned by ZSTD_buildSequencesStatistics containing finalized symbol encoding types - * and size of the sequences statistics - */ -typedef struct { - U32 LLtype; - U32 Offtype; - U32 MLtype; - size_t size; - size_t lastCountSize; /* Accounts for bug in 1.3.4. More detail in ZSTD_entropyCompressSeqStore_internal() */ -} ZSTD_symbolEncodingTypeStats_t; - -/* ZSTD_buildSequencesStatistics(): - * Returns a ZSTD_symbolEncodingTypeStats_t, or a zstd error code in the `size` field. - * Modifies `nextEntropy` to have the appropriate values as a side effect. - * nbSeq must be greater than 0. - * - * entropyWkspSize must be of size at least ENTROPY_WORKSPACE_SIZE - (MaxSeq + 1)*sizeof(U32) - */ -static ZSTD_symbolEncodingTypeStats_t -ZSTD_buildSequencesStatistics(seqStore_t* seqStorePtr, size_t nbSeq, - const ZSTD_fseCTables_t* prevEntropy, ZSTD_fseCTables_t* nextEntropy, - BYTE* dst, const BYTE* const dstEnd, - ZSTD_strategy strategy, unsigned* countWorkspace, - void* entropyWorkspace, size_t entropyWkspSize) { - BYTE* const ostart = dst; - const BYTE* const oend = dstEnd; - BYTE* op = ostart; - FSE_CTable* CTable_LitLength = nextEntropy->litlengthCTable; - FSE_CTable* CTable_OffsetBits = nextEntropy->offcodeCTable; - FSE_CTable* CTable_MatchLength = nextEntropy->matchlengthCTable; - const BYTE* const ofCodeTable = seqStorePtr->ofCode; - const BYTE* const llCodeTable = seqStorePtr->llCode; - const BYTE* const mlCodeTable = seqStorePtr->mlCode; - ZSTD_symbolEncodingTypeStats_t stats; - - stats.lastCountSize = 0; - /* convert length/distances into codes */ - ZSTD_seqToCodes(seqStorePtr); - assert(op <= oend); - assert(nbSeq != 0); /* ZSTD_selectEncodingType() divides by nbSeq */ - /* build CTable for Literal Lengths */ - { unsigned max = MaxLL; - size_t const mostFrequent = HIST_countFast_wksp(countWorkspace, &max, llCodeTable, nbSeq, entropyWorkspace, entropyWkspSize); /* can't fail */ - DEBUGLOG(5, "Building LL table"); - nextEntropy->litlength_repeatMode = prevEntropy->litlength_repeatMode; - stats.LLtype = ZSTD_selectEncodingType(&nextEntropy->litlength_repeatMode, - countWorkspace, max, mostFrequent, nbSeq, - LLFSELog, prevEntropy->litlengthCTable, - LL_defaultNorm, LL_defaultNormLog, - ZSTD_defaultAllowed, strategy); - assert(set_basic < set_compressed && set_rle < set_compressed); - assert(!(stats.LLtype < set_compressed && nextEntropy->litlength_repeatMode != FSE_repeat_none)); /* We don't copy tables */ - { size_t const countSize = ZSTD_buildCTable( - op, (size_t)(oend - op), - CTable_LitLength, LLFSELog, (symbolEncodingType_e)stats.LLtype, - countWorkspace, max, llCodeTable, nbSeq, - LL_defaultNorm, LL_defaultNormLog, MaxLL, - prevEntropy->litlengthCTable, - sizeof(prevEntropy->litlengthCTable), - entropyWorkspace, entropyWkspSize); - if (ZSTD_isError(countSize)) { - DEBUGLOG(3, "ZSTD_buildCTable for LitLens failed"); - stats.size = countSize; - return stats; - } - if (stats.LLtype == set_compressed) - stats.lastCountSize = countSize; - op += countSize; - assert(op <= oend); - } } - /* build CTable for Offsets */ - { unsigned max = MaxOff; - size_t const mostFrequent = HIST_countFast_wksp( - countWorkspace, &max, ofCodeTable, nbSeq, entropyWorkspace, entropyWkspSize); /* can't fail */ - /* We can only use the basic table if max <= DefaultMaxOff, otherwise the offsets are too large */ - ZSTD_defaultPolicy_e const defaultPolicy = (max <= DefaultMaxOff) ? ZSTD_defaultAllowed : ZSTD_defaultDisallowed; - DEBUGLOG(5, "Building OF table"); - nextEntropy->offcode_repeatMode = prevEntropy->offcode_repeatMode; - stats.Offtype = ZSTD_selectEncodingType(&nextEntropy->offcode_repeatMode, - countWorkspace, max, mostFrequent, nbSeq, - OffFSELog, prevEntropy->offcodeCTable, - OF_defaultNorm, OF_defaultNormLog, - defaultPolicy, strategy); - assert(!(stats.Offtype < set_compressed && nextEntropy->offcode_repeatMode != FSE_repeat_none)); /* We don't copy tables */ - { size_t const countSize = ZSTD_buildCTable( - op, (size_t)(oend - op), - CTable_OffsetBits, OffFSELog, (symbolEncodingType_e)stats.Offtype, - countWorkspace, max, ofCodeTable, nbSeq, - OF_defaultNorm, OF_defaultNormLog, DefaultMaxOff, - prevEntropy->offcodeCTable, - sizeof(prevEntropy->offcodeCTable), - entropyWorkspace, entropyWkspSize); - if (ZSTD_isError(countSize)) { - DEBUGLOG(3, "ZSTD_buildCTable for Offsets failed"); - stats.size = countSize; - return stats; - } - if (stats.Offtype == set_compressed) - stats.lastCountSize = countSize; - op += countSize; - assert(op <= oend); - } } - /* build CTable for MatchLengths */ - { unsigned max = MaxML; - size_t const mostFrequent = HIST_countFast_wksp( - countWorkspace, &max, mlCodeTable, nbSeq, entropyWorkspace, entropyWkspSize); /* can't fail */ - DEBUGLOG(5, "Building ML table (remaining space : %i)", (int)(oend-op)); - nextEntropy->matchlength_repeatMode = prevEntropy->matchlength_repeatMode; - stats.MLtype = ZSTD_selectEncodingType(&nextEntropy->matchlength_repeatMode, - countWorkspace, max, mostFrequent, nbSeq, - MLFSELog, prevEntropy->matchlengthCTable, - ML_defaultNorm, ML_defaultNormLog, - ZSTD_defaultAllowed, strategy); - assert(!(stats.MLtype < set_compressed && nextEntropy->matchlength_repeatMode != FSE_repeat_none)); /* We don't copy tables */ - { size_t const countSize = ZSTD_buildCTable( - op, (size_t)(oend - op), - CTable_MatchLength, MLFSELog, (symbolEncodingType_e)stats.MLtype, - countWorkspace, max, mlCodeTable, nbSeq, - ML_defaultNorm, ML_defaultNormLog, MaxML, - prevEntropy->matchlengthCTable, - sizeof(prevEntropy->matchlengthCTable), - entropyWorkspace, entropyWkspSize); - if (ZSTD_isError(countSize)) { - DEBUGLOG(3, "ZSTD_buildCTable for MatchLengths failed"); - stats.size = countSize; - return stats; - } - if (stats.MLtype == set_compressed) - stats.lastCountSize = countSize; - op += countSize; - assert(op <= oend); - } } - stats.size = (size_t)(op-ostart); - return stats; -} - -/* ZSTD_entropyCompressSeqStore_internal(): - * compresses both literals and sequences - * Returns compressed size of block, or a zstd error. - */ +} + +/* Type returned by ZSTD_buildSequencesStatistics containing finalized symbol encoding types + * and size of the sequences statistics + */ +typedef struct { + U32 LLtype; + U32 Offtype; + U32 MLtype; + size_t size; + size_t lastCountSize; /* Accounts for bug in 1.3.4. More detail in ZSTD_entropyCompressSeqStore_internal() */ +} ZSTD_symbolEncodingTypeStats_t; + +/* ZSTD_buildSequencesStatistics(): + * Returns a ZSTD_symbolEncodingTypeStats_t, or a zstd error code in the `size` field. + * Modifies `nextEntropy` to have the appropriate values as a side effect. + * nbSeq must be greater than 0. + * + * entropyWkspSize must be of size at least ENTROPY_WORKSPACE_SIZE - (MaxSeq + 1)*sizeof(U32) + */ +static ZSTD_symbolEncodingTypeStats_t +ZSTD_buildSequencesStatistics(seqStore_t* seqStorePtr, size_t nbSeq, + const ZSTD_fseCTables_t* prevEntropy, ZSTD_fseCTables_t* nextEntropy, + BYTE* dst, const BYTE* const dstEnd, + ZSTD_strategy strategy, unsigned* countWorkspace, + void* entropyWorkspace, size_t entropyWkspSize) { + BYTE* const ostart = dst; + const BYTE* const oend = dstEnd; + BYTE* op = ostart; + FSE_CTable* CTable_LitLength = nextEntropy->litlengthCTable; + FSE_CTable* CTable_OffsetBits = nextEntropy->offcodeCTable; + FSE_CTable* CTable_MatchLength = nextEntropy->matchlengthCTable; + const BYTE* const ofCodeTable = seqStorePtr->ofCode; + const BYTE* const llCodeTable = seqStorePtr->llCode; + const BYTE* const mlCodeTable = seqStorePtr->mlCode; + ZSTD_symbolEncodingTypeStats_t stats; + + stats.lastCountSize = 0; + /* convert length/distances into codes */ + ZSTD_seqToCodes(seqStorePtr); + assert(op <= oend); + assert(nbSeq != 0); /* ZSTD_selectEncodingType() divides by nbSeq */ + /* build CTable for Literal Lengths */ + { unsigned max = MaxLL; + size_t const mostFrequent = HIST_countFast_wksp(countWorkspace, &max, llCodeTable, nbSeq, entropyWorkspace, entropyWkspSize); /* can't fail */ + DEBUGLOG(5, "Building LL table"); + nextEntropy->litlength_repeatMode = prevEntropy->litlength_repeatMode; + stats.LLtype = ZSTD_selectEncodingType(&nextEntropy->litlength_repeatMode, + countWorkspace, max, mostFrequent, nbSeq, + LLFSELog, prevEntropy->litlengthCTable, + LL_defaultNorm, LL_defaultNormLog, + ZSTD_defaultAllowed, strategy); + assert(set_basic < set_compressed && set_rle < set_compressed); + assert(!(stats.LLtype < set_compressed && nextEntropy->litlength_repeatMode != FSE_repeat_none)); /* We don't copy tables */ + { size_t const countSize = ZSTD_buildCTable( + op, (size_t)(oend - op), + CTable_LitLength, LLFSELog, (symbolEncodingType_e)stats.LLtype, + countWorkspace, max, llCodeTable, nbSeq, + LL_defaultNorm, LL_defaultNormLog, MaxLL, + prevEntropy->litlengthCTable, + sizeof(prevEntropy->litlengthCTable), + entropyWorkspace, entropyWkspSize); + if (ZSTD_isError(countSize)) { + DEBUGLOG(3, "ZSTD_buildCTable for LitLens failed"); + stats.size = countSize; + return stats; + } + if (stats.LLtype == set_compressed) + stats.lastCountSize = countSize; + op += countSize; + assert(op <= oend); + } } + /* build CTable for Offsets */ + { unsigned max = MaxOff; + size_t const mostFrequent = HIST_countFast_wksp( + countWorkspace, &max, ofCodeTable, nbSeq, entropyWorkspace, entropyWkspSize); /* can't fail */ + /* We can only use the basic table if max <= DefaultMaxOff, otherwise the offsets are too large */ + ZSTD_defaultPolicy_e const defaultPolicy = (max <= DefaultMaxOff) ? ZSTD_defaultAllowed : ZSTD_defaultDisallowed; + DEBUGLOG(5, "Building OF table"); + nextEntropy->offcode_repeatMode = prevEntropy->offcode_repeatMode; + stats.Offtype = ZSTD_selectEncodingType(&nextEntropy->offcode_repeatMode, + countWorkspace, max, mostFrequent, nbSeq, + OffFSELog, prevEntropy->offcodeCTable, + OF_defaultNorm, OF_defaultNormLog, + defaultPolicy, strategy); + assert(!(stats.Offtype < set_compressed && nextEntropy->offcode_repeatMode != FSE_repeat_none)); /* We don't copy tables */ + { size_t const countSize = ZSTD_buildCTable( + op, (size_t)(oend - op), + CTable_OffsetBits, OffFSELog, (symbolEncodingType_e)stats.Offtype, + countWorkspace, max, ofCodeTable, nbSeq, + OF_defaultNorm, OF_defaultNormLog, DefaultMaxOff, + prevEntropy->offcodeCTable, + sizeof(prevEntropy->offcodeCTable), + entropyWorkspace, entropyWkspSize); + if (ZSTD_isError(countSize)) { + DEBUGLOG(3, "ZSTD_buildCTable for Offsets failed"); + stats.size = countSize; + return stats; + } + if (stats.Offtype == set_compressed) + stats.lastCountSize = countSize; + op += countSize; + assert(op <= oend); + } } + /* build CTable for MatchLengths */ + { unsigned max = MaxML; + size_t const mostFrequent = HIST_countFast_wksp( + countWorkspace, &max, mlCodeTable, nbSeq, entropyWorkspace, entropyWkspSize); /* can't fail */ + DEBUGLOG(5, "Building ML table (remaining space : %i)", (int)(oend-op)); + nextEntropy->matchlength_repeatMode = prevEntropy->matchlength_repeatMode; + stats.MLtype = ZSTD_selectEncodingType(&nextEntropy->matchlength_repeatMode, + countWorkspace, max, mostFrequent, nbSeq, + MLFSELog, prevEntropy->matchlengthCTable, + ML_defaultNorm, ML_defaultNormLog, + ZSTD_defaultAllowed, strategy); + assert(!(stats.MLtype < set_compressed && nextEntropy->matchlength_repeatMode != FSE_repeat_none)); /* We don't copy tables */ + { size_t const countSize = ZSTD_buildCTable( + op, (size_t)(oend - op), + CTable_MatchLength, MLFSELog, (symbolEncodingType_e)stats.MLtype, + countWorkspace, max, mlCodeTable, nbSeq, + ML_defaultNorm, ML_defaultNormLog, MaxML, + prevEntropy->matchlengthCTable, + sizeof(prevEntropy->matchlengthCTable), + entropyWorkspace, entropyWkspSize); + if (ZSTD_isError(countSize)) { + DEBUGLOG(3, "ZSTD_buildCTable for MatchLengths failed"); + stats.size = countSize; + return stats; + } + if (stats.MLtype == set_compressed) + stats.lastCountSize = countSize; + op += countSize; + assert(op <= oend); + } } + stats.size = (size_t)(op-ostart); + return stats; +} + +/* ZSTD_entropyCompressSeqStore_internal(): + * compresses both literals and sequences + * Returns compressed size of block, or a zstd error. + */ #define SUSPECT_UNCOMPRESSIBLE_LITERAL_RATIO 20 MEM_STATIC size_t -ZSTD_entropyCompressSeqStore_internal(seqStore_t* seqStorePtr, +ZSTD_entropyCompressSeqStore_internal(seqStore_t* seqStorePtr, const ZSTD_entropyCTables_t* prevEntropy, ZSTD_entropyCTables_t* nextEntropy, const ZSTD_CCtx_params* cctxParams, @@ -2581,26 +2581,26 @@ ZSTD_entropyCompressSeqStore_internal(seqStore_t* seqStorePtr, { const int longOffsets = cctxParams->cParams.windowLog > STREAM_ACCUMULATOR_MIN; ZSTD_strategy const strategy = cctxParams->cParams.strategy; - unsigned* count = (unsigned*)entropyWorkspace; + unsigned* count = (unsigned*)entropyWorkspace; FSE_CTable* CTable_LitLength = nextEntropy->fse.litlengthCTable; FSE_CTable* CTable_OffsetBits = nextEntropy->fse.offcodeCTable; FSE_CTable* CTable_MatchLength = nextEntropy->fse.matchlengthCTable; const seqDef* const sequences = seqStorePtr->sequencesStart; - const size_t nbSeq = seqStorePtr->sequences - seqStorePtr->sequencesStart; + const size_t nbSeq = seqStorePtr->sequences - seqStorePtr->sequencesStart; const BYTE* const ofCodeTable = seqStorePtr->ofCode; const BYTE* const llCodeTable = seqStorePtr->llCode; const BYTE* const mlCodeTable = seqStorePtr->mlCode; BYTE* const ostart = (BYTE*)dst; BYTE* const oend = ostart + dstCapacity; BYTE* op = ostart; - size_t lastCountSize; + size_t lastCountSize; - entropyWorkspace = count + (MaxSeq + 1); - entropyWkspSize -= (MaxSeq + 1) * sizeof(*count); - - DEBUGLOG(4, "ZSTD_entropyCompressSeqStore_internal (nbSeq=%zu)", nbSeq); + entropyWorkspace = count + (MaxSeq + 1); + entropyWkspSize -= (MaxSeq + 1) * sizeof(*count); + + DEBUGLOG(4, "ZSTD_entropyCompressSeqStore_internal (nbSeq=%zu)", nbSeq); ZSTD_STATIC_ASSERT(HUF_WORKSPACE_SIZE >= (1<<MAX(MLFSELog,LLFSELog))); - assert(entropyWkspSize >= HUF_WORKSPACE_SIZE); + assert(entropyWkspSize >= HUF_WORKSPACE_SIZE); /* Compress literals */ { const BYTE* const literals = seqStorePtr->litStart; @@ -2639,23 +2639,23 @@ ZSTD_entropyCompressSeqStore_internal(seqStore_t* seqStorePtr, assert(op <= oend); if (nbSeq==0) { /* Copy the old tables over as if we repeated them */ - ZSTD_memcpy(&nextEntropy->fse, &prevEntropy->fse, sizeof(prevEntropy->fse)); + ZSTD_memcpy(&nextEntropy->fse, &prevEntropy->fse, sizeof(prevEntropy->fse)); return (size_t)(op - ostart); } - { - ZSTD_symbolEncodingTypeStats_t stats; - BYTE* seqHead = op++; - /* build stats for sequences */ - stats = ZSTD_buildSequencesStatistics(seqStorePtr, nbSeq, - &prevEntropy->fse, &nextEntropy->fse, - op, oend, - strategy, count, - entropyWorkspace, entropyWkspSize); - FORWARD_IF_ERROR(stats.size, "ZSTD_buildSequencesStatistics failed!"); - *seqHead = (BYTE)((stats.LLtype<<6) + (stats.Offtype<<4) + (stats.MLtype<<2)); - lastCountSize = stats.lastCountSize; - op += stats.size; - } + { + ZSTD_symbolEncodingTypeStats_t stats; + BYTE* seqHead = op++; + /* build stats for sequences */ + stats = ZSTD_buildSequencesStatistics(seqStorePtr, nbSeq, + &prevEntropy->fse, &nextEntropy->fse, + op, oend, + strategy, count, + entropyWorkspace, entropyWkspSize); + FORWARD_IF_ERROR(stats.size, "ZSTD_buildSequencesStatistics failed!"); + *seqHead = (BYTE)((stats.LLtype<<6) + (stats.Offtype<<4) + (stats.MLtype<<2)); + lastCountSize = stats.lastCountSize; + op += stats.size; + } { size_t const bitstreamSize = ZSTD_encodeSequences( op, (size_t)(oend - op), @@ -2675,9 +2675,9 @@ ZSTD_entropyCompressSeqStore_internal(seqStore_t* seqStorePtr, * In this exceedingly rare case, we will simply emit an uncompressed * block, since it isn't worth optimizing. */ - if (lastCountSize && (lastCountSize + bitstreamSize) < 4) { - /* lastCountSize >= 2 && bitstreamSize > 0 ==> lastCountSize == 3 */ - assert(lastCountSize + bitstreamSize == 3); + if (lastCountSize && (lastCountSize + bitstreamSize) < 4) { + /* lastCountSize >= 2 && bitstreamSize > 0 ==> lastCountSize == 3 */ + assert(lastCountSize + bitstreamSize == 3); DEBUGLOG(5, "Avoiding bug in zstd decoder in versions <= 1.3.4 by " "emitting an uncompressed block."); return 0; @@ -2689,7 +2689,7 @@ ZSTD_entropyCompressSeqStore_internal(seqStore_t* seqStorePtr, } MEM_STATIC size_t -ZSTD_entropyCompressSeqStore(seqStore_t* seqStorePtr, +ZSTD_entropyCompressSeqStore(seqStore_t* seqStorePtr, const ZSTD_entropyCTables_t* prevEntropy, ZSTD_entropyCTables_t* nextEntropy, const ZSTD_CCtx_params* cctxParams, @@ -2698,7 +2698,7 @@ ZSTD_entropyCompressSeqStore(seqStore_t* seqStorePtr, void* entropyWorkspace, size_t entropyWkspSize, int bmi2) { - size_t const cSize = ZSTD_entropyCompressSeqStore_internal( + size_t const cSize = ZSTD_entropyCompressSeqStore_internal( seqStorePtr, prevEntropy, nextEntropy, cctxParams, dst, dstCapacity, entropyWorkspace, entropyWkspSize, bmi2); @@ -2708,13 +2708,13 @@ ZSTD_entropyCompressSeqStore(seqStore_t* seqStorePtr, */ if ((cSize == ERROR(dstSize_tooSmall)) & (srcSize <= dstCapacity)) return 0; /* block not compressed */ - FORWARD_IF_ERROR(cSize, "ZSTD_entropyCompressSeqStore_internal failed"); + FORWARD_IF_ERROR(cSize, "ZSTD_entropyCompressSeqStore_internal failed"); /* Check compressibility */ { size_t const maxCSize = srcSize - ZSTD_minGain(srcSize, cctxParams->cParams.strategy); if (cSize >= maxCSize) return 0; /* block not compressed */ } - DEBUGLOG(4, "ZSTD_entropyCompressSeqStore() cSize: %zu", cSize); + DEBUGLOG(4, "ZSTD_entropyCompressSeqStore() cSize: %zu", cSize); return cSize; } @@ -2723,7 +2723,7 @@ ZSTD_entropyCompressSeqStore(seqStore_t* seqStorePtr, * assumption : strat is a valid strategy */ ZSTD_blockCompressor ZSTD_selectBlockCompressor(ZSTD_strategy strat, ZSTD_paramSwitch_e useRowMatchFinder, ZSTD_dictMode_e dictMode) { - static const ZSTD_blockCompressor blockCompressor[4][ZSTD_STRATEGY_MAX+1] = { + static const ZSTD_blockCompressor blockCompressor[4][ZSTD_STRATEGY_MAX+1] = { { ZSTD_compressBlock_fast /* default for 0 */, ZSTD_compressBlock_fast, ZSTD_compressBlock_doubleFast, @@ -2753,44 +2753,44 @@ ZSTD_blockCompressor ZSTD_selectBlockCompressor(ZSTD_strategy strat, ZSTD_paramS ZSTD_compressBlock_btlazy2_dictMatchState, ZSTD_compressBlock_btopt_dictMatchState, ZSTD_compressBlock_btultra_dictMatchState, - ZSTD_compressBlock_btultra_dictMatchState }, - { NULL /* default for 0 */, - NULL, - NULL, - ZSTD_compressBlock_greedy_dedicatedDictSearch, - ZSTD_compressBlock_lazy_dedicatedDictSearch, - ZSTD_compressBlock_lazy2_dedicatedDictSearch, - NULL, - NULL, - NULL, - NULL } + ZSTD_compressBlock_btultra_dictMatchState }, + { NULL /* default for 0 */, + NULL, + NULL, + ZSTD_compressBlock_greedy_dedicatedDictSearch, + ZSTD_compressBlock_lazy_dedicatedDictSearch, + ZSTD_compressBlock_lazy2_dedicatedDictSearch, + NULL, + NULL, + NULL, + NULL } }; ZSTD_blockCompressor selectedCompressor; ZSTD_STATIC_ASSERT((unsigned)ZSTD_fast == 1); assert(ZSTD_cParam_withinBounds(ZSTD_c_strategy, strat)); - DEBUGLOG(4, "Selected block compressor: dictMode=%d strat=%d rowMatchfinder=%d", (int)dictMode, (int)strat, (int)useRowMatchFinder); - if (ZSTD_rowMatchFinderUsed(strat, useRowMatchFinder)) { - static const ZSTD_blockCompressor rowBasedBlockCompressors[4][3] = { - { ZSTD_compressBlock_greedy_row, - ZSTD_compressBlock_lazy_row, - ZSTD_compressBlock_lazy2_row }, - { ZSTD_compressBlock_greedy_extDict_row, - ZSTD_compressBlock_lazy_extDict_row, - ZSTD_compressBlock_lazy2_extDict_row }, - { ZSTD_compressBlock_greedy_dictMatchState_row, - ZSTD_compressBlock_lazy_dictMatchState_row, - ZSTD_compressBlock_lazy2_dictMatchState_row }, - { ZSTD_compressBlock_greedy_dedicatedDictSearch_row, - ZSTD_compressBlock_lazy_dedicatedDictSearch_row, - ZSTD_compressBlock_lazy2_dedicatedDictSearch_row } - }; - DEBUGLOG(4, "Selecting a row-based matchfinder"); + DEBUGLOG(4, "Selected block compressor: dictMode=%d strat=%d rowMatchfinder=%d", (int)dictMode, (int)strat, (int)useRowMatchFinder); + if (ZSTD_rowMatchFinderUsed(strat, useRowMatchFinder)) { + static const ZSTD_blockCompressor rowBasedBlockCompressors[4][3] = { + { ZSTD_compressBlock_greedy_row, + ZSTD_compressBlock_lazy_row, + ZSTD_compressBlock_lazy2_row }, + { ZSTD_compressBlock_greedy_extDict_row, + ZSTD_compressBlock_lazy_extDict_row, + ZSTD_compressBlock_lazy2_extDict_row }, + { ZSTD_compressBlock_greedy_dictMatchState_row, + ZSTD_compressBlock_lazy_dictMatchState_row, + ZSTD_compressBlock_lazy2_dictMatchState_row }, + { ZSTD_compressBlock_greedy_dedicatedDictSearch_row, + ZSTD_compressBlock_lazy_dedicatedDictSearch_row, + ZSTD_compressBlock_lazy2_dedicatedDictSearch_row } + }; + DEBUGLOG(4, "Selecting a row-based matchfinder"); assert(useRowMatchFinder != ZSTD_ps_auto); - selectedCompressor = rowBasedBlockCompressors[(int)dictMode][(int)strat - (int)ZSTD_greedy]; - } else { - selectedCompressor = blockCompressor[(int)dictMode][(int)strat]; - } + selectedCompressor = rowBasedBlockCompressors[(int)dictMode][(int)strat - (int)ZSTD_greedy]; + } else { + selectedCompressor = blockCompressor[(int)dictMode][(int)strat]; + } assert(selectedCompressor != NULL); return selectedCompressor; } @@ -2798,7 +2798,7 @@ ZSTD_blockCompressor ZSTD_selectBlockCompressor(ZSTD_strategy strat, ZSTD_paramS static void ZSTD_storeLastLiterals(seqStore_t* seqStorePtr, const BYTE* anchor, size_t lastLLSize) { - ZSTD_memcpy(seqStorePtr->lit, anchor, lastLLSize); + ZSTD_memcpy(seqStorePtr->lit, anchor, lastLLSize); seqStorePtr->lit += lastLLSize; } @@ -2806,7 +2806,7 @@ void ZSTD_resetSeqStore(seqStore_t* ssPtr) { ssPtr->lit = ssPtr->litStart; ssPtr->sequences = ssPtr->sequencesStart; - ssPtr->longLengthType = ZSTD_llt_none; + ssPtr->longLengthType = ZSTD_llt_none; } typedef enum { ZSTDbss_compress, ZSTDbss_noCompress } ZSTD_buildSeqStore_e; @@ -2819,11 +2819,11 @@ static size_t ZSTD_buildSeqStore(ZSTD_CCtx* zc, const void* src, size_t srcSize) /* Assert that we have correctly flushed the ctx params into the ms's copy */ ZSTD_assertEqualCParams(zc->appliedParams.cParams, ms->cParams); if (srcSize < MIN_CBLOCK_SIZE+ZSTD_blockHeaderSize+1) { - if (zc->appliedParams.cParams.strategy >= ZSTD_btopt) { - ZSTD_ldm_skipRawSeqStoreBytes(&zc->externSeqStore, srcSize); - } else { - ZSTD_ldm_skipSequences(&zc->externSeqStore, srcSize, zc->appliedParams.cParams.minMatch); - } + if (zc->appliedParams.cParams.strategy >= ZSTD_btopt) { + ZSTD_ldm_skipRawSeqStoreBytes(&zc->externSeqStore, srcSize); + } else { + ZSTD_ldm_skipSequences(&zc->externSeqStore, srcSize, zc->appliedParams.cParams.minMatch); + } return ZSTDbss_noCompress; /* don't even attempt compression below a certain srcSize */ } ZSTD_resetSeqStore(&(zc->seqStore)); @@ -2839,10 +2839,10 @@ static size_t ZSTD_buildSeqStore(ZSTD_CCtx* zc, const void* src, size_t srcSize) /* limited update after a very long match */ { const BYTE* const base = ms->window.base; const BYTE* const istart = (const BYTE*)src; - const U32 curr = (U32)(istart-base); + const U32 curr = (U32)(istart-base); if (sizeof(ptrdiff_t)==8) assert(istart - base < (ptrdiff_t)(U32)(-1)); /* ensure no overflow */ - if (curr > ms->nextToUpdate + 384) - ms->nextToUpdate = curr - MIN(192, (U32)(curr - ms->nextToUpdate - 384)); + if (curr > ms->nextToUpdate + 384) + ms->nextToUpdate = curr - MIN(192, (U32)(curr - ms->nextToUpdate - 384)); } /* select and store sequences */ @@ -2859,11 +2859,11 @@ static size_t ZSTD_buildSeqStore(ZSTD_CCtx* zc, const void* src, size_t srcSize) ZSTD_ldm_blockCompress(&zc->externSeqStore, ms, &zc->seqStore, zc->blockState.nextCBlock->rep, - zc->appliedParams.useRowMatchFinder, + zc->appliedParams.useRowMatchFinder, src, srcSize); assert(zc->externSeqStore.pos <= zc->externSeqStore.size); } else if (zc->appliedParams.ldmParams.enableLdm == ZSTD_ps_enable) { - rawSeqStore_t ldmSeqStore = kNullRawSeqStore; + rawSeqStore_t ldmSeqStore = kNullRawSeqStore; ldmSeqStore.seq = zc->ldmSequences; ldmSeqStore.capacity = zc->maxNbLdmSequences; @@ -2876,14 +2876,14 @@ static size_t ZSTD_buildSeqStore(ZSTD_CCtx* zc, const void* src, size_t srcSize) ZSTD_ldm_blockCompress(&ldmSeqStore, ms, &zc->seqStore, zc->blockState.nextCBlock->rep, - zc->appliedParams.useRowMatchFinder, + zc->appliedParams.useRowMatchFinder, src, srcSize); assert(ldmSeqStore.pos == ldmSeqStore.size); } else { /* not long range mode */ - ZSTD_blockCompressor const blockCompressor = ZSTD_selectBlockCompressor(zc->appliedParams.cParams.strategy, - zc->appliedParams.useRowMatchFinder, - dictMode); - ms->ldmSeqStore = NULL; + ZSTD_blockCompressor const blockCompressor = ZSTD_selectBlockCompressor(zc->appliedParams.cParams.strategy, + zc->appliedParams.useRowMatchFinder, + dictMode); + ms->ldmSeqStore = NULL; lastLLSize = blockCompressor(ms, &zc->seqStore, zc->blockState.nextCBlock->rep, src, srcSize); } { const BYTE* const lastLiterals = (const BYTE*)src + srcSize - lastLLSize; @@ -2895,72 +2895,72 @@ static size_t ZSTD_buildSeqStore(ZSTD_CCtx* zc, const void* src, size_t srcSize) static void ZSTD_copyBlockSequences(ZSTD_CCtx* zc) { const seqStore_t* seqStore = ZSTD_getSeqStore(zc); - const seqDef* seqStoreSeqs = seqStore->sequencesStart; - size_t seqStoreSeqSize = seqStore->sequences - seqStoreSeqs; - size_t seqStoreLiteralsSize = (size_t)(seqStore->lit - seqStore->litStart); - size_t literalsRead = 0; - size_t lastLLSize; + const seqDef* seqStoreSeqs = seqStore->sequencesStart; + size_t seqStoreSeqSize = seqStore->sequences - seqStoreSeqs; + size_t seqStoreLiteralsSize = (size_t)(seqStore->lit - seqStore->litStart); + size_t literalsRead = 0; + size_t lastLLSize; ZSTD_Sequence* outSeqs = &zc->seqCollector.seqStart[zc->seqCollector.seqIndex]; - size_t i; - repcodes_t updatedRepcodes; + size_t i; + repcodes_t updatedRepcodes; assert(zc->seqCollector.seqIndex + 1 < zc->seqCollector.maxSequences); - /* Ensure we have enough space for last literals "sequence" */ - assert(zc->seqCollector.maxSequences >= seqStoreSeqSize + 1); - ZSTD_memcpy(updatedRepcodes.rep, zc->blockState.prevCBlock->rep, sizeof(repcodes_t)); - for (i = 0; i < seqStoreSeqSize; ++i) { + /* Ensure we have enough space for last literals "sequence" */ + assert(zc->seqCollector.maxSequences >= seqStoreSeqSize + 1); + ZSTD_memcpy(updatedRepcodes.rep, zc->blockState.prevCBlock->rep, sizeof(repcodes_t)); + for (i = 0; i < seqStoreSeqSize; ++i) { U32 rawOffset = seqStoreSeqs[i].offBase - ZSTD_REP_NUM; - outSeqs[i].litLength = seqStoreSeqs[i].litLength; + outSeqs[i].litLength = seqStoreSeqs[i].litLength; outSeqs[i].matchLength = seqStoreSeqs[i].mlBase + MINMATCH; - outSeqs[i].rep = 0; + outSeqs[i].rep = 0; if (i == seqStore->longLengthPos) { - if (seqStore->longLengthType == ZSTD_llt_literalLength) { + if (seqStore->longLengthType == ZSTD_llt_literalLength) { outSeqs[i].litLength += 0x10000; - } else if (seqStore->longLengthType == ZSTD_llt_matchLength) { + } else if (seqStore->longLengthType == ZSTD_llt_matchLength) { outSeqs[i].matchLength += 0x10000; } } if (seqStoreSeqs[i].offBase <= ZSTD_REP_NUM) { - /* Derive the correct offset corresponding to a repcode */ + /* Derive the correct offset corresponding to a repcode */ outSeqs[i].rep = seqStoreSeqs[i].offBase; - if (outSeqs[i].litLength != 0) { - rawOffset = updatedRepcodes.rep[outSeqs[i].rep - 1]; - } else { - if (outSeqs[i].rep == 3) { - rawOffset = updatedRepcodes.rep[0] - 1; + if (outSeqs[i].litLength != 0) { + rawOffset = updatedRepcodes.rep[outSeqs[i].rep - 1]; + } else { + if (outSeqs[i].rep == 3) { + rawOffset = updatedRepcodes.rep[0] - 1; } else { - rawOffset = updatedRepcodes.rep[outSeqs[i].rep]; + rawOffset = updatedRepcodes.rep[outSeqs[i].rep]; } } } - outSeqs[i].offset = rawOffset; - /* seqStoreSeqs[i].offset == offCode+1, and ZSTD_updateRep() expects offCode - so we provide seqStoreSeqs[i].offset - 1 */ + outSeqs[i].offset = rawOffset; + /* seqStoreSeqs[i].offset == offCode+1, and ZSTD_updateRep() expects offCode + so we provide seqStoreSeqs[i].offset - 1 */ ZSTD_updateRep(updatedRepcodes.rep, seqStoreSeqs[i].offBase - 1, seqStoreSeqs[i].litLength == 0); - literalsRead += outSeqs[i].litLength; + literalsRead += outSeqs[i].litLength; } - /* Insert last literals (if any exist) in the block as a sequence with ml == off == 0. - * If there are no last literals, then we'll emit (of: 0, ml: 0, ll: 0), which is a marker - * for the block boundary, according to the API. - */ - assert(seqStoreLiteralsSize >= literalsRead); - lastLLSize = seqStoreLiteralsSize - literalsRead; - outSeqs[i].litLength = (U32)lastLLSize; - outSeqs[i].matchLength = outSeqs[i].offset = outSeqs[i].rep = 0; - seqStoreSeqSize++; - zc->seqCollector.seqIndex += seqStoreSeqSize; + /* Insert last literals (if any exist) in the block as a sequence with ml == off == 0. + * If there are no last literals, then we'll emit (of: 0, ml: 0, ll: 0), which is a marker + * for the block boundary, according to the API. + */ + assert(seqStoreLiteralsSize >= literalsRead); + lastLLSize = seqStoreLiteralsSize - literalsRead; + outSeqs[i].litLength = (U32)lastLLSize; + outSeqs[i].matchLength = outSeqs[i].offset = outSeqs[i].rep = 0; + seqStoreSeqSize++; + zc->seqCollector.seqIndex += seqStoreSeqSize; } -size_t ZSTD_generateSequences(ZSTD_CCtx* zc, ZSTD_Sequence* outSeqs, - size_t outSeqsSize, const void* src, size_t srcSize) +size_t ZSTD_generateSequences(ZSTD_CCtx* zc, ZSTD_Sequence* outSeqs, + size_t outSeqsSize, const void* src, size_t srcSize) { const size_t dstCapacity = ZSTD_compressBound(srcSize); - void* dst = ZSTD_customMalloc(dstCapacity, ZSTD_defaultCMem); + void* dst = ZSTD_customMalloc(dstCapacity, ZSTD_defaultCMem); SeqCollector seqCollector; RETURN_ERROR_IF(dst == NULL, memory_allocation, "NULL pointer!"); @@ -2972,48 +2972,48 @@ size_t ZSTD_generateSequences(ZSTD_CCtx* zc, ZSTD_Sequence* outSeqs, zc->seqCollector = seqCollector; ZSTD_compress2(zc, dst, dstCapacity, src, srcSize); - ZSTD_customFree(dst, ZSTD_defaultCMem); + ZSTD_customFree(dst, ZSTD_defaultCMem); return zc->seqCollector.seqIndex; } -size_t ZSTD_mergeBlockDelimiters(ZSTD_Sequence* sequences, size_t seqsSize) { - size_t in = 0; - size_t out = 0; - for (; in < seqsSize; ++in) { - if (sequences[in].offset == 0 && sequences[in].matchLength == 0) { - if (in != seqsSize - 1) { - sequences[in+1].litLength += sequences[in].litLength; - } - } else { - sequences[out] = sequences[in]; - ++out; - } - } - return out; -} - -/* Unrolled loop to read four size_ts of input at a time. Returns 1 if is RLE, 0 if not. */ -static int ZSTD_isRLE(const BYTE* src, size_t length) { - const BYTE* ip = src; - const BYTE value = ip[0]; - const size_t valueST = (size_t)((U64)value * 0x0101010101010101ULL); - const size_t unrollSize = sizeof(size_t) * 4; - const size_t unrollMask = unrollSize - 1; - const size_t prefixLength = length & unrollMask; +size_t ZSTD_mergeBlockDelimiters(ZSTD_Sequence* sequences, size_t seqsSize) { + size_t in = 0; + size_t out = 0; + for (; in < seqsSize; ++in) { + if (sequences[in].offset == 0 && sequences[in].matchLength == 0) { + if (in != seqsSize - 1) { + sequences[in+1].litLength += sequences[in].litLength; + } + } else { + sequences[out] = sequences[in]; + ++out; + } + } + return out; +} + +/* Unrolled loop to read four size_ts of input at a time. Returns 1 if is RLE, 0 if not. */ +static int ZSTD_isRLE(const BYTE* src, size_t length) { + const BYTE* ip = src; + const BYTE value = ip[0]; + const size_t valueST = (size_t)((U64)value * 0x0101010101010101ULL); + const size_t unrollSize = sizeof(size_t) * 4; + const size_t unrollMask = unrollSize - 1; + const size_t prefixLength = length & unrollMask; size_t i; - size_t u; - if (length == 1) return 1; - /* Check if prefix is RLE first before using unrolled loop */ - if (prefixLength && ZSTD_count(ip+1, ip, ip+prefixLength) != prefixLength-1) { - return 0; - } - for (i = prefixLength; i != length; i += unrollSize) { - for (u = 0; u < unrollSize; u += sizeof(size_t)) { - if (MEM_readST(ip + i + u) != valueST) { - return 0; - } - } - } + size_t u; + if (length == 1) return 1; + /* Check if prefix is RLE first before using unrolled loop */ + if (prefixLength && ZSTD_count(ip+1, ip, ip+prefixLength) != prefixLength-1) { + return 0; + } + for (i = prefixLength; i != length; i += unrollSize) { + for (u = 0; u < unrollSize; u += sizeof(size_t)) { + if (MEM_readST(ip + i + u) != valueST) { + return 0; + } + } + } return 1; } @@ -3029,431 +3029,431 @@ static int ZSTD_maybeRLE(seqStore_t const* seqStore) return nbSeqs < 4 && nbLits < 10; } -static void ZSTD_blockState_confirmRepcodesAndEntropyTables(ZSTD_blockState_t* const bs) -{ - ZSTD_compressedBlockState_t* const tmp = bs->prevCBlock; - bs->prevCBlock = bs->nextCBlock; - bs->nextCBlock = tmp; -} - -/* Writes the block header */ -static void writeBlockHeader(void* op, size_t cSize, size_t blockSize, U32 lastBlock) { - U32 const cBlockHeader = cSize == 1 ? - lastBlock + (((U32)bt_rle)<<1) + (U32)(blockSize << 3) : - lastBlock + (((U32)bt_compressed)<<1) + (U32)(cSize << 3); - MEM_writeLE24(op, cBlockHeader); - DEBUGLOG(3, "writeBlockHeader: cSize: %zu blockSize: %zu lastBlock: %u", cSize, blockSize, lastBlock); -} - -/** ZSTD_buildBlockEntropyStats_literals() : - * Builds entropy for the literals. - * Stores literals block type (raw, rle, compressed, repeat) and - * huffman description table to hufMetadata. - * Requires ENTROPY_WORKSPACE_SIZE workspace - * @return : size of huffman description table or error code */ -static size_t ZSTD_buildBlockEntropyStats_literals(void* const src, size_t srcSize, - const ZSTD_hufCTables_t* prevHuf, - ZSTD_hufCTables_t* nextHuf, - ZSTD_hufCTablesMetadata_t* hufMetadata, +static void ZSTD_blockState_confirmRepcodesAndEntropyTables(ZSTD_blockState_t* const bs) +{ + ZSTD_compressedBlockState_t* const tmp = bs->prevCBlock; + bs->prevCBlock = bs->nextCBlock; + bs->nextCBlock = tmp; +} + +/* Writes the block header */ +static void writeBlockHeader(void* op, size_t cSize, size_t blockSize, U32 lastBlock) { + U32 const cBlockHeader = cSize == 1 ? + lastBlock + (((U32)bt_rle)<<1) + (U32)(blockSize << 3) : + lastBlock + (((U32)bt_compressed)<<1) + (U32)(cSize << 3); + MEM_writeLE24(op, cBlockHeader); + DEBUGLOG(3, "writeBlockHeader: cSize: %zu blockSize: %zu lastBlock: %u", cSize, blockSize, lastBlock); +} + +/** ZSTD_buildBlockEntropyStats_literals() : + * Builds entropy for the literals. + * Stores literals block type (raw, rle, compressed, repeat) and + * huffman description table to hufMetadata. + * Requires ENTROPY_WORKSPACE_SIZE workspace + * @return : size of huffman description table or error code */ +static size_t ZSTD_buildBlockEntropyStats_literals(void* const src, size_t srcSize, + const ZSTD_hufCTables_t* prevHuf, + ZSTD_hufCTables_t* nextHuf, + ZSTD_hufCTablesMetadata_t* hufMetadata, const int literalsCompressionIsDisabled, - void* workspace, size_t wkspSize) -{ - BYTE* const wkspStart = (BYTE*)workspace; - BYTE* const wkspEnd = wkspStart + wkspSize; - BYTE* const countWkspStart = wkspStart; - unsigned* const countWksp = (unsigned*)workspace; - const size_t countWkspSize = (HUF_SYMBOLVALUE_MAX + 1) * sizeof(unsigned); - BYTE* const nodeWksp = countWkspStart + countWkspSize; - const size_t nodeWkspSize = wkspEnd-nodeWksp; - unsigned maxSymbolValue = HUF_SYMBOLVALUE_MAX; - unsigned huffLog = HUF_TABLELOG_DEFAULT; - HUF_repeat repeat = prevHuf->repeatMode; - DEBUGLOG(5, "ZSTD_buildBlockEntropyStats_literals (srcSize=%zu)", srcSize); - - /* Prepare nextEntropy assuming reusing the existing table */ - ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); - + void* workspace, size_t wkspSize) +{ + BYTE* const wkspStart = (BYTE*)workspace; + BYTE* const wkspEnd = wkspStart + wkspSize; + BYTE* const countWkspStart = wkspStart; + unsigned* const countWksp = (unsigned*)workspace; + const size_t countWkspSize = (HUF_SYMBOLVALUE_MAX + 1) * sizeof(unsigned); + BYTE* const nodeWksp = countWkspStart + countWkspSize; + const size_t nodeWkspSize = wkspEnd-nodeWksp; + unsigned maxSymbolValue = HUF_SYMBOLVALUE_MAX; + unsigned huffLog = HUF_TABLELOG_DEFAULT; + HUF_repeat repeat = prevHuf->repeatMode; + DEBUGLOG(5, "ZSTD_buildBlockEntropyStats_literals (srcSize=%zu)", srcSize); + + /* Prepare nextEntropy assuming reusing the existing table */ + ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); + if (literalsCompressionIsDisabled) { - DEBUGLOG(5, "set_basic - disabled"); - hufMetadata->hType = set_basic; - return 0; - } - - /* small ? don't even attempt compression (speed opt) */ -#ifndef COMPRESS_LITERALS_SIZE_MIN -#define COMPRESS_LITERALS_SIZE_MIN 63 -#endif - { size_t const minLitSize = (prevHuf->repeatMode == HUF_repeat_valid) ? 6 : COMPRESS_LITERALS_SIZE_MIN; - if (srcSize <= minLitSize) { - DEBUGLOG(5, "set_basic - too small"); - hufMetadata->hType = set_basic; - return 0; - } - } - - /* Scan input and build symbol stats */ - { size_t const largest = HIST_count_wksp (countWksp, &maxSymbolValue, (const BYTE*)src, srcSize, workspace, wkspSize); - FORWARD_IF_ERROR(largest, "HIST_count_wksp failed"); - if (largest == srcSize) { - DEBUGLOG(5, "set_rle"); - hufMetadata->hType = set_rle; - return 0; - } - if (largest <= (srcSize >> 7)+4) { - DEBUGLOG(5, "set_basic - no gain"); - hufMetadata->hType = set_basic; - return 0; - } - } - - /* Validate the previous Huffman table */ - if (repeat == HUF_repeat_check && !HUF_validateCTable((HUF_CElt const*)prevHuf->CTable, countWksp, maxSymbolValue)) { - repeat = HUF_repeat_none; - } - - /* Build Huffman Tree */ - ZSTD_memset(nextHuf->CTable, 0, sizeof(nextHuf->CTable)); - huffLog = HUF_optimalTableLog(huffLog, srcSize, maxSymbolValue); - { size_t const maxBits = HUF_buildCTable_wksp((HUF_CElt*)nextHuf->CTable, countWksp, - maxSymbolValue, huffLog, - nodeWksp, nodeWkspSize); - FORWARD_IF_ERROR(maxBits, "HUF_buildCTable_wksp"); - huffLog = (U32)maxBits; - { /* Build and write the CTable */ - size_t const newCSize = HUF_estimateCompressedSize( - (HUF_CElt*)nextHuf->CTable, countWksp, maxSymbolValue); - size_t const hSize = HUF_writeCTable_wksp( - hufMetadata->hufDesBuffer, sizeof(hufMetadata->hufDesBuffer), - (HUF_CElt*)nextHuf->CTable, maxSymbolValue, huffLog, - nodeWksp, nodeWkspSize); - /* Check against repeating the previous CTable */ - if (repeat != HUF_repeat_none) { - size_t const oldCSize = HUF_estimateCompressedSize( - (HUF_CElt const*)prevHuf->CTable, countWksp, maxSymbolValue); - if (oldCSize < srcSize && (oldCSize <= hSize + newCSize || hSize + 12 >= srcSize)) { - DEBUGLOG(5, "set_repeat - smaller"); - ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); - hufMetadata->hType = set_repeat; - return 0; - } - } - if (newCSize + hSize >= srcSize) { - DEBUGLOG(5, "set_basic - no gains"); - ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); - hufMetadata->hType = set_basic; - return 0; - } - DEBUGLOG(5, "set_compressed (hSize=%u)", (U32)hSize); - hufMetadata->hType = set_compressed; - nextHuf->repeatMode = HUF_repeat_check; - return hSize; - } - } -} - - -/* ZSTD_buildDummySequencesStatistics(): - * Returns a ZSTD_symbolEncodingTypeStats_t with all encoding types as set_basic, - * and updates nextEntropy to the appropriate repeatMode. - */ -static ZSTD_symbolEncodingTypeStats_t -ZSTD_buildDummySequencesStatistics(ZSTD_fseCTables_t* nextEntropy) { - ZSTD_symbolEncodingTypeStats_t stats = {set_basic, set_basic, set_basic, 0, 0}; - nextEntropy->litlength_repeatMode = FSE_repeat_none; - nextEntropy->offcode_repeatMode = FSE_repeat_none; - nextEntropy->matchlength_repeatMode = FSE_repeat_none; - return stats; -} - -/** ZSTD_buildBlockEntropyStats_sequences() : - * Builds entropy for the sequences. - * Stores symbol compression modes and fse table to fseMetadata. - * Requires ENTROPY_WORKSPACE_SIZE wksp. - * @return : size of fse tables or error code */ -static size_t ZSTD_buildBlockEntropyStats_sequences(seqStore_t* seqStorePtr, - const ZSTD_fseCTables_t* prevEntropy, - ZSTD_fseCTables_t* nextEntropy, - const ZSTD_CCtx_params* cctxParams, - ZSTD_fseCTablesMetadata_t* fseMetadata, - void* workspace, size_t wkspSize) -{ - ZSTD_strategy const strategy = cctxParams->cParams.strategy; - size_t const nbSeq = seqStorePtr->sequences - seqStorePtr->sequencesStart; - BYTE* const ostart = fseMetadata->fseTablesBuffer; - BYTE* const oend = ostart + sizeof(fseMetadata->fseTablesBuffer); - BYTE* op = ostart; - unsigned* countWorkspace = (unsigned*)workspace; - unsigned* entropyWorkspace = countWorkspace + (MaxSeq + 1); - size_t entropyWorkspaceSize = wkspSize - (MaxSeq + 1) * sizeof(*countWorkspace); - ZSTD_symbolEncodingTypeStats_t stats; - - DEBUGLOG(5, "ZSTD_buildBlockEntropyStats_sequences (nbSeq=%zu)", nbSeq); - stats = nbSeq != 0 ? ZSTD_buildSequencesStatistics(seqStorePtr, nbSeq, - prevEntropy, nextEntropy, op, oend, - strategy, countWorkspace, - entropyWorkspace, entropyWorkspaceSize) - : ZSTD_buildDummySequencesStatistics(nextEntropy); - FORWARD_IF_ERROR(stats.size, "ZSTD_buildSequencesStatistics failed!"); - fseMetadata->llType = (symbolEncodingType_e) stats.LLtype; - fseMetadata->ofType = (symbolEncodingType_e) stats.Offtype; - fseMetadata->mlType = (symbolEncodingType_e) stats.MLtype; - fseMetadata->lastCountSize = stats.lastCountSize; - return stats.size; -} - - -/** ZSTD_buildBlockEntropyStats() : - * Builds entropy for the block. - * Requires workspace size ENTROPY_WORKSPACE_SIZE - * - * @return : 0 on success or error code - */ -size_t ZSTD_buildBlockEntropyStats(seqStore_t* seqStorePtr, - const ZSTD_entropyCTables_t* prevEntropy, - ZSTD_entropyCTables_t* nextEntropy, - const ZSTD_CCtx_params* cctxParams, - ZSTD_entropyCTablesMetadata_t* entropyMetadata, - void* workspace, size_t wkspSize) -{ - size_t const litSize = seqStorePtr->lit - seqStorePtr->litStart; - entropyMetadata->hufMetadata.hufDesSize = - ZSTD_buildBlockEntropyStats_literals(seqStorePtr->litStart, litSize, - &prevEntropy->huf, &nextEntropy->huf, - &entropyMetadata->hufMetadata, + DEBUGLOG(5, "set_basic - disabled"); + hufMetadata->hType = set_basic; + return 0; + } + + /* small ? don't even attempt compression (speed opt) */ +#ifndef COMPRESS_LITERALS_SIZE_MIN +#define COMPRESS_LITERALS_SIZE_MIN 63 +#endif + { size_t const minLitSize = (prevHuf->repeatMode == HUF_repeat_valid) ? 6 : COMPRESS_LITERALS_SIZE_MIN; + if (srcSize <= minLitSize) { + DEBUGLOG(5, "set_basic - too small"); + hufMetadata->hType = set_basic; + return 0; + } + } + + /* Scan input and build symbol stats */ + { size_t const largest = HIST_count_wksp (countWksp, &maxSymbolValue, (const BYTE*)src, srcSize, workspace, wkspSize); + FORWARD_IF_ERROR(largest, "HIST_count_wksp failed"); + if (largest == srcSize) { + DEBUGLOG(5, "set_rle"); + hufMetadata->hType = set_rle; + return 0; + } + if (largest <= (srcSize >> 7)+4) { + DEBUGLOG(5, "set_basic - no gain"); + hufMetadata->hType = set_basic; + return 0; + } + } + + /* Validate the previous Huffman table */ + if (repeat == HUF_repeat_check && !HUF_validateCTable((HUF_CElt const*)prevHuf->CTable, countWksp, maxSymbolValue)) { + repeat = HUF_repeat_none; + } + + /* Build Huffman Tree */ + ZSTD_memset(nextHuf->CTable, 0, sizeof(nextHuf->CTable)); + huffLog = HUF_optimalTableLog(huffLog, srcSize, maxSymbolValue); + { size_t const maxBits = HUF_buildCTable_wksp((HUF_CElt*)nextHuf->CTable, countWksp, + maxSymbolValue, huffLog, + nodeWksp, nodeWkspSize); + FORWARD_IF_ERROR(maxBits, "HUF_buildCTable_wksp"); + huffLog = (U32)maxBits; + { /* Build and write the CTable */ + size_t const newCSize = HUF_estimateCompressedSize( + (HUF_CElt*)nextHuf->CTable, countWksp, maxSymbolValue); + size_t const hSize = HUF_writeCTable_wksp( + hufMetadata->hufDesBuffer, sizeof(hufMetadata->hufDesBuffer), + (HUF_CElt*)nextHuf->CTable, maxSymbolValue, huffLog, + nodeWksp, nodeWkspSize); + /* Check against repeating the previous CTable */ + if (repeat != HUF_repeat_none) { + size_t const oldCSize = HUF_estimateCompressedSize( + (HUF_CElt const*)prevHuf->CTable, countWksp, maxSymbolValue); + if (oldCSize < srcSize && (oldCSize <= hSize + newCSize || hSize + 12 >= srcSize)) { + DEBUGLOG(5, "set_repeat - smaller"); + ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); + hufMetadata->hType = set_repeat; + return 0; + } + } + if (newCSize + hSize >= srcSize) { + DEBUGLOG(5, "set_basic - no gains"); + ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); + hufMetadata->hType = set_basic; + return 0; + } + DEBUGLOG(5, "set_compressed (hSize=%u)", (U32)hSize); + hufMetadata->hType = set_compressed; + nextHuf->repeatMode = HUF_repeat_check; + return hSize; + } + } +} + + +/* ZSTD_buildDummySequencesStatistics(): + * Returns a ZSTD_symbolEncodingTypeStats_t with all encoding types as set_basic, + * and updates nextEntropy to the appropriate repeatMode. + */ +static ZSTD_symbolEncodingTypeStats_t +ZSTD_buildDummySequencesStatistics(ZSTD_fseCTables_t* nextEntropy) { + ZSTD_symbolEncodingTypeStats_t stats = {set_basic, set_basic, set_basic, 0, 0}; + nextEntropy->litlength_repeatMode = FSE_repeat_none; + nextEntropy->offcode_repeatMode = FSE_repeat_none; + nextEntropy->matchlength_repeatMode = FSE_repeat_none; + return stats; +} + +/** ZSTD_buildBlockEntropyStats_sequences() : + * Builds entropy for the sequences. + * Stores symbol compression modes and fse table to fseMetadata. + * Requires ENTROPY_WORKSPACE_SIZE wksp. + * @return : size of fse tables or error code */ +static size_t ZSTD_buildBlockEntropyStats_sequences(seqStore_t* seqStorePtr, + const ZSTD_fseCTables_t* prevEntropy, + ZSTD_fseCTables_t* nextEntropy, + const ZSTD_CCtx_params* cctxParams, + ZSTD_fseCTablesMetadata_t* fseMetadata, + void* workspace, size_t wkspSize) +{ + ZSTD_strategy const strategy = cctxParams->cParams.strategy; + size_t const nbSeq = seqStorePtr->sequences - seqStorePtr->sequencesStart; + BYTE* const ostart = fseMetadata->fseTablesBuffer; + BYTE* const oend = ostart + sizeof(fseMetadata->fseTablesBuffer); + BYTE* op = ostart; + unsigned* countWorkspace = (unsigned*)workspace; + unsigned* entropyWorkspace = countWorkspace + (MaxSeq + 1); + size_t entropyWorkspaceSize = wkspSize - (MaxSeq + 1) * sizeof(*countWorkspace); + ZSTD_symbolEncodingTypeStats_t stats; + + DEBUGLOG(5, "ZSTD_buildBlockEntropyStats_sequences (nbSeq=%zu)", nbSeq); + stats = nbSeq != 0 ? ZSTD_buildSequencesStatistics(seqStorePtr, nbSeq, + prevEntropy, nextEntropy, op, oend, + strategy, countWorkspace, + entropyWorkspace, entropyWorkspaceSize) + : ZSTD_buildDummySequencesStatistics(nextEntropy); + FORWARD_IF_ERROR(stats.size, "ZSTD_buildSequencesStatistics failed!"); + fseMetadata->llType = (symbolEncodingType_e) stats.LLtype; + fseMetadata->ofType = (symbolEncodingType_e) stats.Offtype; + fseMetadata->mlType = (symbolEncodingType_e) stats.MLtype; + fseMetadata->lastCountSize = stats.lastCountSize; + return stats.size; +} + + +/** ZSTD_buildBlockEntropyStats() : + * Builds entropy for the block. + * Requires workspace size ENTROPY_WORKSPACE_SIZE + * + * @return : 0 on success or error code + */ +size_t ZSTD_buildBlockEntropyStats(seqStore_t* seqStorePtr, + const ZSTD_entropyCTables_t* prevEntropy, + ZSTD_entropyCTables_t* nextEntropy, + const ZSTD_CCtx_params* cctxParams, + ZSTD_entropyCTablesMetadata_t* entropyMetadata, + void* workspace, size_t wkspSize) +{ + size_t const litSize = seqStorePtr->lit - seqStorePtr->litStart; + entropyMetadata->hufMetadata.hufDesSize = + ZSTD_buildBlockEntropyStats_literals(seqStorePtr->litStart, litSize, + &prevEntropy->huf, &nextEntropy->huf, + &entropyMetadata->hufMetadata, ZSTD_literalsCompressionIsDisabled(cctxParams), - workspace, wkspSize); - FORWARD_IF_ERROR(entropyMetadata->hufMetadata.hufDesSize, "ZSTD_buildBlockEntropyStats_literals failed"); - entropyMetadata->fseMetadata.fseTablesSize = - ZSTD_buildBlockEntropyStats_sequences(seqStorePtr, - &prevEntropy->fse, &nextEntropy->fse, - cctxParams, - &entropyMetadata->fseMetadata, - workspace, wkspSize); - FORWARD_IF_ERROR(entropyMetadata->fseMetadata.fseTablesSize, "ZSTD_buildBlockEntropyStats_sequences failed"); - return 0; -} - -/* Returns the size estimate for the literals section (header + content) of a block */ -static size_t ZSTD_estimateBlockSize_literal(const BYTE* literals, size_t litSize, - const ZSTD_hufCTables_t* huf, - const ZSTD_hufCTablesMetadata_t* hufMetadata, - void* workspace, size_t wkspSize, - int writeEntropy) -{ - unsigned* const countWksp = (unsigned*)workspace; - unsigned maxSymbolValue = HUF_SYMBOLVALUE_MAX; - size_t literalSectionHeaderSize = 3 + (litSize >= 1 KB) + (litSize >= 16 KB); - U32 singleStream = litSize < 256; - - if (hufMetadata->hType == set_basic) return litSize; - else if (hufMetadata->hType == set_rle) return 1; - else if (hufMetadata->hType == set_compressed || hufMetadata->hType == set_repeat) { - size_t const largest = HIST_count_wksp (countWksp, &maxSymbolValue, (const BYTE*)literals, litSize, workspace, wkspSize); - if (ZSTD_isError(largest)) return litSize; - { size_t cLitSizeEstimate = HUF_estimateCompressedSize((const HUF_CElt*)huf->CTable, countWksp, maxSymbolValue); - if (writeEntropy) cLitSizeEstimate += hufMetadata->hufDesSize; - if (!singleStream) cLitSizeEstimate += 6; /* multi-stream huffman uses 6-byte jump table */ - return cLitSizeEstimate + literalSectionHeaderSize; - } } - assert(0); /* impossible */ - return 0; -} - -/* Returns the size estimate for the FSE-compressed symbols (of, ml, ll) of a block */ -static size_t ZSTD_estimateBlockSize_symbolType(symbolEncodingType_e type, - const BYTE* codeTable, size_t nbSeq, unsigned maxCode, - const FSE_CTable* fseCTable, + workspace, wkspSize); + FORWARD_IF_ERROR(entropyMetadata->hufMetadata.hufDesSize, "ZSTD_buildBlockEntropyStats_literals failed"); + entropyMetadata->fseMetadata.fseTablesSize = + ZSTD_buildBlockEntropyStats_sequences(seqStorePtr, + &prevEntropy->fse, &nextEntropy->fse, + cctxParams, + &entropyMetadata->fseMetadata, + workspace, wkspSize); + FORWARD_IF_ERROR(entropyMetadata->fseMetadata.fseTablesSize, "ZSTD_buildBlockEntropyStats_sequences failed"); + return 0; +} + +/* Returns the size estimate for the literals section (header + content) of a block */ +static size_t ZSTD_estimateBlockSize_literal(const BYTE* literals, size_t litSize, + const ZSTD_hufCTables_t* huf, + const ZSTD_hufCTablesMetadata_t* hufMetadata, + void* workspace, size_t wkspSize, + int writeEntropy) +{ + unsigned* const countWksp = (unsigned*)workspace; + unsigned maxSymbolValue = HUF_SYMBOLVALUE_MAX; + size_t literalSectionHeaderSize = 3 + (litSize >= 1 KB) + (litSize >= 16 KB); + U32 singleStream = litSize < 256; + + if (hufMetadata->hType == set_basic) return litSize; + else if (hufMetadata->hType == set_rle) return 1; + else if (hufMetadata->hType == set_compressed || hufMetadata->hType == set_repeat) { + size_t const largest = HIST_count_wksp (countWksp, &maxSymbolValue, (const BYTE*)literals, litSize, workspace, wkspSize); + if (ZSTD_isError(largest)) return litSize; + { size_t cLitSizeEstimate = HUF_estimateCompressedSize((const HUF_CElt*)huf->CTable, countWksp, maxSymbolValue); + if (writeEntropy) cLitSizeEstimate += hufMetadata->hufDesSize; + if (!singleStream) cLitSizeEstimate += 6; /* multi-stream huffman uses 6-byte jump table */ + return cLitSizeEstimate + literalSectionHeaderSize; + } } + assert(0); /* impossible */ + return 0; +} + +/* Returns the size estimate for the FSE-compressed symbols (of, ml, ll) of a block */ +static size_t ZSTD_estimateBlockSize_symbolType(symbolEncodingType_e type, + const BYTE* codeTable, size_t nbSeq, unsigned maxCode, + const FSE_CTable* fseCTable, const U8* additionalBits, - short const* defaultNorm, U32 defaultNormLog, U32 defaultMax, - void* workspace, size_t wkspSize) -{ - unsigned* const countWksp = (unsigned*)workspace; - const BYTE* ctp = codeTable; - const BYTE* const ctStart = ctp; - const BYTE* const ctEnd = ctStart + nbSeq; - size_t cSymbolTypeSizeEstimateInBits = 0; - unsigned max = maxCode; - - HIST_countFast_wksp(countWksp, &max, codeTable, nbSeq, workspace, wkspSize); /* can't fail */ - if (type == set_basic) { - /* We selected this encoding type, so it must be valid. */ - assert(max <= defaultMax); - (void)defaultMax; - cSymbolTypeSizeEstimateInBits = ZSTD_crossEntropyCost(defaultNorm, defaultNormLog, countWksp, max); - } else if (type == set_rle) { - cSymbolTypeSizeEstimateInBits = 0; - } else if (type == set_compressed || type == set_repeat) { - cSymbolTypeSizeEstimateInBits = ZSTD_fseBitCost(fseCTable, countWksp, max); - } - if (ZSTD_isError(cSymbolTypeSizeEstimateInBits)) { - return nbSeq * 10; - } - while (ctp < ctEnd) { - if (additionalBits) cSymbolTypeSizeEstimateInBits += additionalBits[*ctp]; - else cSymbolTypeSizeEstimateInBits += *ctp; /* for offset, offset code is also the number of additional bits */ - ctp++; - } - return cSymbolTypeSizeEstimateInBits >> 3; -} - -/* Returns the size estimate for the sequences section (header + content) of a block */ -static size_t ZSTD_estimateBlockSize_sequences(const BYTE* ofCodeTable, - const BYTE* llCodeTable, - const BYTE* mlCodeTable, - size_t nbSeq, - const ZSTD_fseCTables_t* fseTables, - const ZSTD_fseCTablesMetadata_t* fseMetadata, - void* workspace, size_t wkspSize, - int writeEntropy) -{ - size_t sequencesSectionHeaderSize = 1 /* seqHead */ + 1 /* min seqSize size */ + (nbSeq >= 128) + (nbSeq >= LONGNBSEQ); - size_t cSeqSizeEstimate = 0; - cSeqSizeEstimate += ZSTD_estimateBlockSize_symbolType(fseMetadata->ofType, ofCodeTable, nbSeq, MaxOff, - fseTables->offcodeCTable, NULL, - OF_defaultNorm, OF_defaultNormLog, DefaultMaxOff, - workspace, wkspSize); - cSeqSizeEstimate += ZSTD_estimateBlockSize_symbolType(fseMetadata->llType, llCodeTable, nbSeq, MaxLL, - fseTables->litlengthCTable, LL_bits, - LL_defaultNorm, LL_defaultNormLog, MaxLL, - workspace, wkspSize); - cSeqSizeEstimate += ZSTD_estimateBlockSize_symbolType(fseMetadata->mlType, mlCodeTable, nbSeq, MaxML, - fseTables->matchlengthCTable, ML_bits, - ML_defaultNorm, ML_defaultNormLog, MaxML, - workspace, wkspSize); - if (writeEntropy) cSeqSizeEstimate += fseMetadata->fseTablesSize; - return cSeqSizeEstimate + sequencesSectionHeaderSize; -} - -/* Returns the size estimate for a given stream of literals, of, ll, ml */ -static size_t ZSTD_estimateBlockSize(const BYTE* literals, size_t litSize, - const BYTE* ofCodeTable, - const BYTE* llCodeTable, - const BYTE* mlCodeTable, - size_t nbSeq, - const ZSTD_entropyCTables_t* entropy, - const ZSTD_entropyCTablesMetadata_t* entropyMetadata, - void* workspace, size_t wkspSize, - int writeLitEntropy, int writeSeqEntropy) { - size_t const literalsSize = ZSTD_estimateBlockSize_literal(literals, litSize, - &entropy->huf, &entropyMetadata->hufMetadata, - workspace, wkspSize, writeLitEntropy); - size_t const seqSize = ZSTD_estimateBlockSize_sequences(ofCodeTable, llCodeTable, mlCodeTable, - nbSeq, &entropy->fse, &entropyMetadata->fseMetadata, - workspace, wkspSize, writeSeqEntropy); - return seqSize + literalsSize + ZSTD_blockHeaderSize; -} - -/* Builds entropy statistics and uses them for blocksize estimation. - * - * Returns the estimated compressed size of the seqStore, or a zstd error. - */ + short const* defaultNorm, U32 defaultNormLog, U32 defaultMax, + void* workspace, size_t wkspSize) +{ + unsigned* const countWksp = (unsigned*)workspace; + const BYTE* ctp = codeTable; + const BYTE* const ctStart = ctp; + const BYTE* const ctEnd = ctStart + nbSeq; + size_t cSymbolTypeSizeEstimateInBits = 0; + unsigned max = maxCode; + + HIST_countFast_wksp(countWksp, &max, codeTable, nbSeq, workspace, wkspSize); /* can't fail */ + if (type == set_basic) { + /* We selected this encoding type, so it must be valid. */ + assert(max <= defaultMax); + (void)defaultMax; + cSymbolTypeSizeEstimateInBits = ZSTD_crossEntropyCost(defaultNorm, defaultNormLog, countWksp, max); + } else if (type == set_rle) { + cSymbolTypeSizeEstimateInBits = 0; + } else if (type == set_compressed || type == set_repeat) { + cSymbolTypeSizeEstimateInBits = ZSTD_fseBitCost(fseCTable, countWksp, max); + } + if (ZSTD_isError(cSymbolTypeSizeEstimateInBits)) { + return nbSeq * 10; + } + while (ctp < ctEnd) { + if (additionalBits) cSymbolTypeSizeEstimateInBits += additionalBits[*ctp]; + else cSymbolTypeSizeEstimateInBits += *ctp; /* for offset, offset code is also the number of additional bits */ + ctp++; + } + return cSymbolTypeSizeEstimateInBits >> 3; +} + +/* Returns the size estimate for the sequences section (header + content) of a block */ +static size_t ZSTD_estimateBlockSize_sequences(const BYTE* ofCodeTable, + const BYTE* llCodeTable, + const BYTE* mlCodeTable, + size_t nbSeq, + const ZSTD_fseCTables_t* fseTables, + const ZSTD_fseCTablesMetadata_t* fseMetadata, + void* workspace, size_t wkspSize, + int writeEntropy) +{ + size_t sequencesSectionHeaderSize = 1 /* seqHead */ + 1 /* min seqSize size */ + (nbSeq >= 128) + (nbSeq >= LONGNBSEQ); + size_t cSeqSizeEstimate = 0; + cSeqSizeEstimate += ZSTD_estimateBlockSize_symbolType(fseMetadata->ofType, ofCodeTable, nbSeq, MaxOff, + fseTables->offcodeCTable, NULL, + OF_defaultNorm, OF_defaultNormLog, DefaultMaxOff, + workspace, wkspSize); + cSeqSizeEstimate += ZSTD_estimateBlockSize_symbolType(fseMetadata->llType, llCodeTable, nbSeq, MaxLL, + fseTables->litlengthCTable, LL_bits, + LL_defaultNorm, LL_defaultNormLog, MaxLL, + workspace, wkspSize); + cSeqSizeEstimate += ZSTD_estimateBlockSize_symbolType(fseMetadata->mlType, mlCodeTable, nbSeq, MaxML, + fseTables->matchlengthCTable, ML_bits, + ML_defaultNorm, ML_defaultNormLog, MaxML, + workspace, wkspSize); + if (writeEntropy) cSeqSizeEstimate += fseMetadata->fseTablesSize; + return cSeqSizeEstimate + sequencesSectionHeaderSize; +} + +/* Returns the size estimate for a given stream of literals, of, ll, ml */ +static size_t ZSTD_estimateBlockSize(const BYTE* literals, size_t litSize, + const BYTE* ofCodeTable, + const BYTE* llCodeTable, + const BYTE* mlCodeTable, + size_t nbSeq, + const ZSTD_entropyCTables_t* entropy, + const ZSTD_entropyCTablesMetadata_t* entropyMetadata, + void* workspace, size_t wkspSize, + int writeLitEntropy, int writeSeqEntropy) { + size_t const literalsSize = ZSTD_estimateBlockSize_literal(literals, litSize, + &entropy->huf, &entropyMetadata->hufMetadata, + workspace, wkspSize, writeLitEntropy); + size_t const seqSize = ZSTD_estimateBlockSize_sequences(ofCodeTable, llCodeTable, mlCodeTable, + nbSeq, &entropy->fse, &entropyMetadata->fseMetadata, + workspace, wkspSize, writeSeqEntropy); + return seqSize + literalsSize + ZSTD_blockHeaderSize; +} + +/* Builds entropy statistics and uses them for blocksize estimation. + * + * Returns the estimated compressed size of the seqStore, or a zstd error. + */ static size_t ZSTD_buildEntropyStatisticsAndEstimateSubBlockSize(seqStore_t* seqStore, ZSTD_CCtx* zc) { ZSTD_entropyCTablesMetadata_t* entropyMetadata = &zc->blockSplitCtx.entropyMetadata; DEBUGLOG(6, "ZSTD_buildEntropyStatisticsAndEstimateSubBlockSize()"); - FORWARD_IF_ERROR(ZSTD_buildBlockEntropyStats(seqStore, - &zc->blockState.prevCBlock->entropy, - &zc->blockState.nextCBlock->entropy, - &zc->appliedParams, + FORWARD_IF_ERROR(ZSTD_buildBlockEntropyStats(seqStore, + &zc->blockState.prevCBlock->entropy, + &zc->blockState.nextCBlock->entropy, + &zc->appliedParams, entropyMetadata, - zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */), ""); - return ZSTD_estimateBlockSize(seqStore->litStart, (size_t)(seqStore->lit - seqStore->litStart), - seqStore->ofCode, seqStore->llCode, seqStore->mlCode, - (size_t)(seqStore->sequences - seqStore->sequencesStart), + zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */), ""); + return ZSTD_estimateBlockSize(seqStore->litStart, (size_t)(seqStore->lit - seqStore->litStart), + seqStore->ofCode, seqStore->llCode, seqStore->mlCode, + (size_t)(seqStore->sequences - seqStore->sequencesStart), &zc->blockState.nextCBlock->entropy, entropyMetadata, zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE, (int)(entropyMetadata->hufMetadata.hType == set_compressed), 1); -} - -/* Returns literals bytes represented in a seqStore */ -static size_t ZSTD_countSeqStoreLiteralsBytes(const seqStore_t* const seqStore) { - size_t literalsBytes = 0; - size_t const nbSeqs = seqStore->sequences - seqStore->sequencesStart; - size_t i; - for (i = 0; i < nbSeqs; ++i) { - seqDef seq = seqStore->sequencesStart[i]; - literalsBytes += seq.litLength; - if (i == seqStore->longLengthPos && seqStore->longLengthType == ZSTD_llt_literalLength) { - literalsBytes += 0x10000; - } - } - return literalsBytes; -} - -/* Returns match bytes represented in a seqStore */ -static size_t ZSTD_countSeqStoreMatchBytes(const seqStore_t* const seqStore) { - size_t matchBytes = 0; - size_t const nbSeqs = seqStore->sequences - seqStore->sequencesStart; - size_t i; - for (i = 0; i < nbSeqs; ++i) { - seqDef seq = seqStore->sequencesStart[i]; +} + +/* Returns literals bytes represented in a seqStore */ +static size_t ZSTD_countSeqStoreLiteralsBytes(const seqStore_t* const seqStore) { + size_t literalsBytes = 0; + size_t const nbSeqs = seqStore->sequences - seqStore->sequencesStart; + size_t i; + for (i = 0; i < nbSeqs; ++i) { + seqDef seq = seqStore->sequencesStart[i]; + literalsBytes += seq.litLength; + if (i == seqStore->longLengthPos && seqStore->longLengthType == ZSTD_llt_literalLength) { + literalsBytes += 0x10000; + } + } + return literalsBytes; +} + +/* Returns match bytes represented in a seqStore */ +static size_t ZSTD_countSeqStoreMatchBytes(const seqStore_t* const seqStore) { + size_t matchBytes = 0; + size_t const nbSeqs = seqStore->sequences - seqStore->sequencesStart; + size_t i; + for (i = 0; i < nbSeqs; ++i) { + seqDef seq = seqStore->sequencesStart[i]; matchBytes += seq.mlBase + MINMATCH; - if (i == seqStore->longLengthPos && seqStore->longLengthType == ZSTD_llt_matchLength) { - matchBytes += 0x10000; - } - } - return matchBytes; -} - -/* Derives the seqStore that is a chunk of the originalSeqStore from [startIdx, endIdx). - * Stores the result in resultSeqStore. - */ -static void ZSTD_deriveSeqStoreChunk(seqStore_t* resultSeqStore, - const seqStore_t* originalSeqStore, - size_t startIdx, size_t endIdx) { - BYTE* const litEnd = originalSeqStore->lit; - size_t literalsBytes; - size_t literalsBytesPreceding = 0; - - *resultSeqStore = *originalSeqStore; - if (startIdx > 0) { - resultSeqStore->sequences = originalSeqStore->sequencesStart + startIdx; - literalsBytesPreceding = ZSTD_countSeqStoreLiteralsBytes(resultSeqStore); - } - - /* Move longLengthPos into the correct position if necessary */ - if (originalSeqStore->longLengthType != ZSTD_llt_none) { - if (originalSeqStore->longLengthPos < startIdx || originalSeqStore->longLengthPos > endIdx) { - resultSeqStore->longLengthType = ZSTD_llt_none; - } else { - resultSeqStore->longLengthPos -= (U32)startIdx; - } - } - resultSeqStore->sequencesStart = originalSeqStore->sequencesStart + startIdx; - resultSeqStore->sequences = originalSeqStore->sequencesStart + endIdx; - literalsBytes = ZSTD_countSeqStoreLiteralsBytes(resultSeqStore); - resultSeqStore->litStart += literalsBytesPreceding; - if (endIdx == (size_t)(originalSeqStore->sequences - originalSeqStore->sequencesStart)) { - /* This accounts for possible last literals if the derived chunk reaches the end of the block */ - resultSeqStore->lit = litEnd; - } else { - resultSeqStore->lit = resultSeqStore->litStart+literalsBytes; - } - resultSeqStore->llCode += startIdx; - resultSeqStore->mlCode += startIdx; - resultSeqStore->ofCode += startIdx; -} - -/** - * Returns the raw offset represented by the combination of offCode, ll0, and repcode history. + if (i == seqStore->longLengthPos && seqStore->longLengthType == ZSTD_llt_matchLength) { + matchBytes += 0x10000; + } + } + return matchBytes; +} + +/* Derives the seqStore that is a chunk of the originalSeqStore from [startIdx, endIdx). + * Stores the result in resultSeqStore. + */ +static void ZSTD_deriveSeqStoreChunk(seqStore_t* resultSeqStore, + const seqStore_t* originalSeqStore, + size_t startIdx, size_t endIdx) { + BYTE* const litEnd = originalSeqStore->lit; + size_t literalsBytes; + size_t literalsBytesPreceding = 0; + + *resultSeqStore = *originalSeqStore; + if (startIdx > 0) { + resultSeqStore->sequences = originalSeqStore->sequencesStart + startIdx; + literalsBytesPreceding = ZSTD_countSeqStoreLiteralsBytes(resultSeqStore); + } + + /* Move longLengthPos into the correct position if necessary */ + if (originalSeqStore->longLengthType != ZSTD_llt_none) { + if (originalSeqStore->longLengthPos < startIdx || originalSeqStore->longLengthPos > endIdx) { + resultSeqStore->longLengthType = ZSTD_llt_none; + } else { + resultSeqStore->longLengthPos -= (U32)startIdx; + } + } + resultSeqStore->sequencesStart = originalSeqStore->sequencesStart + startIdx; + resultSeqStore->sequences = originalSeqStore->sequencesStart + endIdx; + literalsBytes = ZSTD_countSeqStoreLiteralsBytes(resultSeqStore); + resultSeqStore->litStart += literalsBytesPreceding; + if (endIdx == (size_t)(originalSeqStore->sequences - originalSeqStore->sequencesStart)) { + /* This accounts for possible last literals if the derived chunk reaches the end of the block */ + resultSeqStore->lit = litEnd; + } else { + resultSeqStore->lit = resultSeqStore->litStart+literalsBytes; + } + resultSeqStore->llCode += startIdx; + resultSeqStore->mlCode += startIdx; + resultSeqStore->ofCode += startIdx; +} + +/** + * Returns the raw offset represented by the combination of offCode, ll0, and repcode history. * offCode must represent a repcode in the numeric representation of ZSTD_storeSeq(). - */ + */ static U32 ZSTD_resolveRepcodeToRawOffset(const U32 rep[ZSTD_REP_NUM], const U32 offCode, const U32 ll0) { U32 const adjustedOffCode = STORED_REPCODE(offCode) - 1 + ll0; /* [ 0 - 3 ] */ assert(STORED_IS_REPCODE(offCode)); - if (adjustedOffCode == ZSTD_REP_NUM) { - /* litlength == 0 and offCode == 2 implies selection of first repcode - 1 */ - assert(rep[0] > 0); - return rep[0] - 1; - } - return rep[adjustedOffCode]; -} - -/** - * ZSTD_seqStore_resolveOffCodes() reconciles any possible divergences in offset history that may arise + if (adjustedOffCode == ZSTD_REP_NUM) { + /* litlength == 0 and offCode == 2 implies selection of first repcode - 1 */ + assert(rep[0] > 0); + return rep[0] - 1; + } + return rep[adjustedOffCode]; +} + +/** + * ZSTD_seqStore_resolveOffCodes() reconciles any possible divergences in offset history that may arise * due to emission of RLE/raw blocks that disturb the offset history, * and replaces any repcodes within the seqStore that may be invalid. - * + * * dRepcodes are updated as would be on the decompression side. * cRepcodes are updated exactly in accordance with the seqStore. * @@ -3461,39 +3461,39 @@ ZSTD_resolveRepcodeToRawOffset(const U32 rep[ZSTD_REP_NUM], const U32 offCode, c * 0 : invalid * 1-3 : repcode 1-3 * 4+ : real_offset+3 - */ -static void ZSTD_seqStore_resolveOffCodes(repcodes_t* const dRepcodes, repcodes_t* const cRepcodes, - seqStore_t* const seqStore, U32 const nbSeq) { - U32 idx = 0; - for (; idx < nbSeq; ++idx) { - seqDef* const seq = seqStore->sequencesStart + idx; - U32 const ll0 = (seq->litLength == 0); + */ +static void ZSTD_seqStore_resolveOffCodes(repcodes_t* const dRepcodes, repcodes_t* const cRepcodes, + seqStore_t* const seqStore, U32 const nbSeq) { + U32 idx = 0; + for (; idx < nbSeq; ++idx) { + seqDef* const seq = seqStore->sequencesStart + idx; + U32 const ll0 = (seq->litLength == 0); U32 const offCode = OFFBASE_TO_STORED(seq->offBase); assert(seq->offBase > 0); if (STORED_IS_REPCODE(offCode)) { - U32 const dRawOffset = ZSTD_resolveRepcodeToRawOffset(dRepcodes->rep, offCode, ll0); - U32 const cRawOffset = ZSTD_resolveRepcodeToRawOffset(cRepcodes->rep, offCode, ll0); - /* Adjust simulated decompression repcode history if we come across a mismatch. Replace - * the repcode with the offset it actually references, determined by the compression - * repcode history. - */ - if (dRawOffset != cRawOffset) { + U32 const dRawOffset = ZSTD_resolveRepcodeToRawOffset(dRepcodes->rep, offCode, ll0); + U32 const cRawOffset = ZSTD_resolveRepcodeToRawOffset(cRepcodes->rep, offCode, ll0); + /* Adjust simulated decompression repcode history if we come across a mismatch. Replace + * the repcode with the offset it actually references, determined by the compression + * repcode history. + */ + if (dRawOffset != cRawOffset) { seq->offBase = cRawOffset + ZSTD_REP_NUM; - } - } - /* Compression repcode history is always updated with values directly from the unmodified seqStore. - * Decompression repcode history may use modified seq->offset value taken from compression repcode history. - */ + } + } + /* Compression repcode history is always updated with values directly from the unmodified seqStore. + * Decompression repcode history may use modified seq->offset value taken from compression repcode history. + */ ZSTD_updateRep(dRepcodes->rep, OFFBASE_TO_STORED(seq->offBase), ll0); ZSTD_updateRep(cRepcodes->rep, offCode, ll0); - } -} - -/* ZSTD_compressSeqStore_singleBlock(): - * Compresses a seqStore into a block with a block header, into the buffer dst. - * - * Returns the total size of that block (including header) or a ZSTD error code. - */ + } +} + +/* ZSTD_compressSeqStore_singleBlock(): + * Compresses a seqStore into a block with a block header, into the buffer dst. + * + * Returns the total size of that block (including header) or a ZSTD error code. + */ static size_t ZSTD_compressSeqStore_singleBlock(ZSTD_CCtx* zc, seqStore_t* const seqStore, repcodes_t* const dRep, repcodes_t* const cRep, @@ -3501,87 +3501,87 @@ ZSTD_compressSeqStore_singleBlock(ZSTD_CCtx* zc, seqStore_t* const seqStore, const void* src, size_t srcSize, U32 lastBlock, U32 isPartition) { - const U32 rleMaxLength = 25; - BYTE* op = (BYTE*)dst; - const BYTE* ip = (const BYTE*)src; - size_t cSize; - size_t cSeqsSize; - - /* In case of an RLE or raw block, the simulated decompression repcode history must be reset */ - repcodes_t const dRepOriginal = *dRep; + const U32 rleMaxLength = 25; + BYTE* op = (BYTE*)dst; + const BYTE* ip = (const BYTE*)src; + size_t cSize; + size_t cSeqsSize; + + /* In case of an RLE or raw block, the simulated decompression repcode history must be reset */ + repcodes_t const dRepOriginal = *dRep; DEBUGLOG(5, "ZSTD_compressSeqStore_singleBlock"); - if (isPartition) - ZSTD_seqStore_resolveOffCodes(dRep, cRep, seqStore, (U32)(seqStore->sequences - seqStore->sequencesStart)); - + if (isPartition) + ZSTD_seqStore_resolveOffCodes(dRep, cRep, seqStore, (U32)(seqStore->sequences - seqStore->sequencesStart)); + RETURN_ERROR_IF(dstCapacity < ZSTD_blockHeaderSize, dstSize_tooSmall, "Block header doesn't fit"); - cSeqsSize = ZSTD_entropyCompressSeqStore(seqStore, - &zc->blockState.prevCBlock->entropy, &zc->blockState.nextCBlock->entropy, - &zc->appliedParams, - op + ZSTD_blockHeaderSize, dstCapacity - ZSTD_blockHeaderSize, - srcSize, - zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */, - zc->bmi2); - FORWARD_IF_ERROR(cSeqsSize, "ZSTD_entropyCompressSeqStore failed!"); - - if (!zc->isFirstBlock && - cSeqsSize < rleMaxLength && - ZSTD_isRLE((BYTE const*)src, srcSize)) { - /* We don't want to emit our first block as a RLE even if it qualifies because - * doing so will cause the decoder (cli only) to throw a "should consume all input error." - * This is only an issue for zstd <= v1.4.3 - */ - cSeqsSize = 1; - } - - if (zc->seqCollector.collectSequences) { - ZSTD_copyBlockSequences(zc); - ZSTD_blockState_confirmRepcodesAndEntropyTables(&zc->blockState); - return 0; - } - - if (cSeqsSize == 0) { - cSize = ZSTD_noCompressBlock(op, dstCapacity, ip, srcSize, lastBlock); - FORWARD_IF_ERROR(cSize, "Nocompress block failed"); - DEBUGLOG(4, "Writing out nocompress block, size: %zu", cSize); - *dRep = dRepOriginal; /* reset simulated decompression repcode history */ - } else if (cSeqsSize == 1) { - cSize = ZSTD_rleCompressBlock(op, dstCapacity, *ip, srcSize, lastBlock); - FORWARD_IF_ERROR(cSize, "RLE compress block failed"); - DEBUGLOG(4, "Writing out RLE block, size: %zu", cSize); - *dRep = dRepOriginal; /* reset simulated decompression repcode history */ - } else { - ZSTD_blockState_confirmRepcodesAndEntropyTables(&zc->blockState); - writeBlockHeader(op, cSeqsSize, srcSize, lastBlock); - cSize = ZSTD_blockHeaderSize + cSeqsSize; - DEBUGLOG(4, "Writing out compressed block, size: %zu", cSize); - } + cSeqsSize = ZSTD_entropyCompressSeqStore(seqStore, + &zc->blockState.prevCBlock->entropy, &zc->blockState.nextCBlock->entropy, + &zc->appliedParams, + op + ZSTD_blockHeaderSize, dstCapacity - ZSTD_blockHeaderSize, + srcSize, + zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */, + zc->bmi2); + FORWARD_IF_ERROR(cSeqsSize, "ZSTD_entropyCompressSeqStore failed!"); + + if (!zc->isFirstBlock && + cSeqsSize < rleMaxLength && + ZSTD_isRLE((BYTE const*)src, srcSize)) { + /* We don't want to emit our first block as a RLE even if it qualifies because + * doing so will cause the decoder (cli only) to throw a "should consume all input error." + * This is only an issue for zstd <= v1.4.3 + */ + cSeqsSize = 1; + } + + if (zc->seqCollector.collectSequences) { + ZSTD_copyBlockSequences(zc); + ZSTD_blockState_confirmRepcodesAndEntropyTables(&zc->blockState); + return 0; + } + + if (cSeqsSize == 0) { + cSize = ZSTD_noCompressBlock(op, dstCapacity, ip, srcSize, lastBlock); + FORWARD_IF_ERROR(cSize, "Nocompress block failed"); + DEBUGLOG(4, "Writing out nocompress block, size: %zu", cSize); + *dRep = dRepOriginal; /* reset simulated decompression repcode history */ + } else if (cSeqsSize == 1) { + cSize = ZSTD_rleCompressBlock(op, dstCapacity, *ip, srcSize, lastBlock); + FORWARD_IF_ERROR(cSize, "RLE compress block failed"); + DEBUGLOG(4, "Writing out RLE block, size: %zu", cSize); + *dRep = dRepOriginal; /* reset simulated decompression repcode history */ + } else { + ZSTD_blockState_confirmRepcodesAndEntropyTables(&zc->blockState); + writeBlockHeader(op, cSeqsSize, srcSize, lastBlock); + cSize = ZSTD_blockHeaderSize + cSeqsSize; + DEBUGLOG(4, "Writing out compressed block, size: %zu", cSize); + } if (zc->blockState.prevCBlock->entropy.fse.offcode_repeatMode == FSE_repeat_valid) zc->blockState.prevCBlock->entropy.fse.offcode_repeatMode = FSE_repeat_check; - return cSize; -} - -/* Struct to keep track of where we are in our recursive calls. */ -typedef struct { - U32* splitLocations; /* Array of split indices */ - size_t idx; /* The current index within splitLocations being worked on */ -} seqStoreSplits; - -#define MIN_SEQUENCES_BLOCK_SPLITTING 300 - -/* Helper function to perform the recursive search for block splits. - * Estimates the cost of seqStore prior to split, and estimates the cost of splitting the sequences in half. - * If advantageous to split, then we recurse down the two sub-blocks. If not, or if an error occurred in estimation, then - * we do not recurse. - * - * Note: The recursion depth is capped by a heuristic minimum number of sequences, defined by MIN_SEQUENCES_BLOCK_SPLITTING. - * In theory, this means the absolute largest recursion depth is 10 == log2(maxNbSeqInBlock/MIN_SEQUENCES_BLOCK_SPLITTING). - * In practice, recursion depth usually doesn't go beyond 4. - * + return cSize; +} + +/* Struct to keep track of where we are in our recursive calls. */ +typedef struct { + U32* splitLocations; /* Array of split indices */ + size_t idx; /* The current index within splitLocations being worked on */ +} seqStoreSplits; + +#define MIN_SEQUENCES_BLOCK_SPLITTING 300 + +/* Helper function to perform the recursive search for block splits. + * Estimates the cost of seqStore prior to split, and estimates the cost of splitting the sequences in half. + * If advantageous to split, then we recurse down the two sub-blocks. If not, or if an error occurred in estimation, then + * we do not recurse. + * + * Note: The recursion depth is capped by a heuristic minimum number of sequences, defined by MIN_SEQUENCES_BLOCK_SPLITTING. + * In theory, this means the absolute largest recursion depth is 10 == log2(maxNbSeqInBlock/MIN_SEQUENCES_BLOCK_SPLITTING). + * In practice, recursion depth usually doesn't go beyond 4. + * * Furthermore, the number of splits is capped by ZSTD_MAX_NB_BLOCK_SPLITS. At ZSTD_MAX_NB_BLOCK_SPLITS == 196 with the current existing blockSize - * maximum of 128 KB, this value is actually impossible to reach. - */ + * maximum of 128 KB, this value is actually impossible to reach. + */ static void ZSTD_deriveBlockSplitsHelper(seqStoreSplits* splits, size_t startIdx, size_t endIdx, ZSTD_CCtx* zc, const seqStore_t* origSeqStore) @@ -3589,15 +3589,15 @@ ZSTD_deriveBlockSplitsHelper(seqStoreSplits* splits, size_t startIdx, size_t end seqStore_t* fullSeqStoreChunk = &zc->blockSplitCtx.fullSeqStoreChunk; seqStore_t* firstHalfSeqStore = &zc->blockSplitCtx.firstHalfSeqStore; seqStore_t* secondHalfSeqStore = &zc->blockSplitCtx.secondHalfSeqStore; - size_t estimatedOriginalSize; - size_t estimatedFirstHalfSize; - size_t estimatedSecondHalfSize; - size_t midIdx = (startIdx + endIdx)/2; - + size_t estimatedOriginalSize; + size_t estimatedFirstHalfSize; + size_t estimatedSecondHalfSize; + size_t midIdx = (startIdx + endIdx)/2; + if (endIdx - startIdx < MIN_SEQUENCES_BLOCK_SPLITTING || splits->idx >= ZSTD_MAX_NB_BLOCK_SPLITS) { DEBUGLOG(6, "ZSTD_deriveBlockSplitsHelper: Too few sequences"); - return; - } + return; + } DEBUGLOG(4, "ZSTD_deriveBlockSplitsHelper: startIdx=%zu endIdx=%zu", startIdx, endIdx); ZSTD_deriveSeqStoreChunk(fullSeqStoreChunk, origSeqStore, startIdx, endIdx); ZSTD_deriveSeqStoreChunk(firstHalfSeqStore, origSeqStore, startIdx, midIdx); @@ -3606,159 +3606,159 @@ ZSTD_deriveBlockSplitsHelper(seqStoreSplits* splits, size_t startIdx, size_t end estimatedFirstHalfSize = ZSTD_buildEntropyStatisticsAndEstimateSubBlockSize(firstHalfSeqStore, zc); estimatedSecondHalfSize = ZSTD_buildEntropyStatisticsAndEstimateSubBlockSize(secondHalfSeqStore, zc); DEBUGLOG(4, "Estimated original block size: %zu -- First half split: %zu -- Second half split: %zu", - estimatedOriginalSize, estimatedFirstHalfSize, estimatedSecondHalfSize); - if (ZSTD_isError(estimatedOriginalSize) || ZSTD_isError(estimatedFirstHalfSize) || ZSTD_isError(estimatedSecondHalfSize)) { - return; - } - if (estimatedFirstHalfSize + estimatedSecondHalfSize < estimatedOriginalSize) { - ZSTD_deriveBlockSplitsHelper(splits, startIdx, midIdx, zc, origSeqStore); - splits->splitLocations[splits->idx] = (U32)midIdx; - splits->idx++; - ZSTD_deriveBlockSplitsHelper(splits, midIdx, endIdx, zc, origSeqStore); - } -} - -/* Base recursive function. Populates a table with intra-block partition indices that can improve compression ratio. - * - * Returns the number of splits made (which equals the size of the partition table - 1). - */ -static size_t ZSTD_deriveBlockSplits(ZSTD_CCtx* zc, U32 partitions[], U32 nbSeq) { - seqStoreSplits splits = {partitions, 0}; - if (nbSeq <= 4) { - DEBUGLOG(4, "ZSTD_deriveBlockSplits: Too few sequences to split"); - /* Refuse to try and split anything with less than 4 sequences */ - return 0; - } - ZSTD_deriveBlockSplitsHelper(&splits, 0, nbSeq, zc, &zc->seqStore); - splits.splitLocations[splits.idx] = nbSeq; - DEBUGLOG(5, "ZSTD_deriveBlockSplits: final nb partitions: %zu", splits.idx+1); - return splits.idx; -} - -/* ZSTD_compressBlock_splitBlock(): - * Attempts to split a given block into multiple blocks to improve compression ratio. - * - * Returns combined size of all blocks (which includes headers), or a ZSTD error code. - */ + estimatedOriginalSize, estimatedFirstHalfSize, estimatedSecondHalfSize); + if (ZSTD_isError(estimatedOriginalSize) || ZSTD_isError(estimatedFirstHalfSize) || ZSTD_isError(estimatedSecondHalfSize)) { + return; + } + if (estimatedFirstHalfSize + estimatedSecondHalfSize < estimatedOriginalSize) { + ZSTD_deriveBlockSplitsHelper(splits, startIdx, midIdx, zc, origSeqStore); + splits->splitLocations[splits->idx] = (U32)midIdx; + splits->idx++; + ZSTD_deriveBlockSplitsHelper(splits, midIdx, endIdx, zc, origSeqStore); + } +} + +/* Base recursive function. Populates a table with intra-block partition indices that can improve compression ratio. + * + * Returns the number of splits made (which equals the size of the partition table - 1). + */ +static size_t ZSTD_deriveBlockSplits(ZSTD_CCtx* zc, U32 partitions[], U32 nbSeq) { + seqStoreSplits splits = {partitions, 0}; + if (nbSeq <= 4) { + DEBUGLOG(4, "ZSTD_deriveBlockSplits: Too few sequences to split"); + /* Refuse to try and split anything with less than 4 sequences */ + return 0; + } + ZSTD_deriveBlockSplitsHelper(&splits, 0, nbSeq, zc, &zc->seqStore); + splits.splitLocations[splits.idx] = nbSeq; + DEBUGLOG(5, "ZSTD_deriveBlockSplits: final nb partitions: %zu", splits.idx+1); + return splits.idx; +} + +/* ZSTD_compressBlock_splitBlock(): + * Attempts to split a given block into multiple blocks to improve compression ratio. + * + * Returns combined size of all blocks (which includes headers), or a ZSTD error code. + */ static size_t ZSTD_compressBlock_splitBlock_internal(ZSTD_CCtx* zc, void* dst, size_t dstCapacity, const void* src, size_t blockSize, U32 lastBlock, U32 nbSeq) { - size_t cSize = 0; - const BYTE* ip = (const BYTE*)src; - BYTE* op = (BYTE*)dst; - size_t i = 0; - size_t srcBytesTotal = 0; + size_t cSize = 0; + const BYTE* ip = (const BYTE*)src; + BYTE* op = (BYTE*)dst; + size_t i = 0; + size_t srcBytesTotal = 0; U32* partitions = zc->blockSplitCtx.partitions; /* size == ZSTD_MAX_NB_BLOCK_SPLITS */ seqStore_t* nextSeqStore = &zc->blockSplitCtx.nextSeqStore; seqStore_t* currSeqStore = &zc->blockSplitCtx.currSeqStore; - size_t numSplits = ZSTD_deriveBlockSplits(zc, partitions, nbSeq); - - /* If a block is split and some partitions are emitted as RLE/uncompressed, then repcode history - * may become invalid. In order to reconcile potentially invalid repcodes, we keep track of two - * separate repcode histories that simulate repcode history on compression and decompression side, - * and use the histories to determine whether we must replace a particular repcode with its raw offset. - * - * 1) cRep gets updated for each partition, regardless of whether the block was emitted as uncompressed - * or RLE. This allows us to retrieve the offset value that an invalid repcode references within - * a nocompress/RLE block. - * 2) dRep gets updated only for compressed partitions, and when a repcode gets replaced, will use - * the replacement offset value rather than the original repcode to update the repcode history. - * dRep also will be the final repcode history sent to the next block. - * - * See ZSTD_seqStore_resolveOffCodes() for more details. - */ - repcodes_t dRep; - repcodes_t cRep; - ZSTD_memcpy(dRep.rep, zc->blockState.prevCBlock->rep, sizeof(repcodes_t)); - ZSTD_memcpy(cRep.rep, zc->blockState.prevCBlock->rep, sizeof(repcodes_t)); + size_t numSplits = ZSTD_deriveBlockSplits(zc, partitions, nbSeq); + + /* If a block is split and some partitions are emitted as RLE/uncompressed, then repcode history + * may become invalid. In order to reconcile potentially invalid repcodes, we keep track of two + * separate repcode histories that simulate repcode history on compression and decompression side, + * and use the histories to determine whether we must replace a particular repcode with its raw offset. + * + * 1) cRep gets updated for each partition, regardless of whether the block was emitted as uncompressed + * or RLE. This allows us to retrieve the offset value that an invalid repcode references within + * a nocompress/RLE block. + * 2) dRep gets updated only for compressed partitions, and when a repcode gets replaced, will use + * the replacement offset value rather than the original repcode to update the repcode history. + * dRep also will be the final repcode history sent to the next block. + * + * See ZSTD_seqStore_resolveOffCodes() for more details. + */ + repcodes_t dRep; + repcodes_t cRep; + ZSTD_memcpy(dRep.rep, zc->blockState.prevCBlock->rep, sizeof(repcodes_t)); + ZSTD_memcpy(cRep.rep, zc->blockState.prevCBlock->rep, sizeof(repcodes_t)); ZSTD_memset(nextSeqStore, 0, sizeof(seqStore_t)); - - DEBUGLOG(4, "ZSTD_compressBlock_splitBlock_internal (dstCapacity=%u, dictLimit=%u, nextToUpdate=%u)", - (unsigned)dstCapacity, (unsigned)zc->blockState.matchState.window.dictLimit, - (unsigned)zc->blockState.matchState.nextToUpdate); - - if (numSplits == 0) { - size_t cSizeSingleBlock = ZSTD_compressSeqStore_singleBlock(zc, &zc->seqStore, - &dRep, &cRep, - op, dstCapacity, - ip, blockSize, - lastBlock, 0 /* isPartition */); - FORWARD_IF_ERROR(cSizeSingleBlock, "Compressing single block from splitBlock_internal() failed!"); - DEBUGLOG(5, "ZSTD_compressBlock_splitBlock_internal: No splits"); - assert(cSizeSingleBlock <= ZSTD_BLOCKSIZE_MAX + ZSTD_blockHeaderSize); - return cSizeSingleBlock; - } - + + DEBUGLOG(4, "ZSTD_compressBlock_splitBlock_internal (dstCapacity=%u, dictLimit=%u, nextToUpdate=%u)", + (unsigned)dstCapacity, (unsigned)zc->blockState.matchState.window.dictLimit, + (unsigned)zc->blockState.matchState.nextToUpdate); + + if (numSplits == 0) { + size_t cSizeSingleBlock = ZSTD_compressSeqStore_singleBlock(zc, &zc->seqStore, + &dRep, &cRep, + op, dstCapacity, + ip, blockSize, + lastBlock, 0 /* isPartition */); + FORWARD_IF_ERROR(cSizeSingleBlock, "Compressing single block from splitBlock_internal() failed!"); + DEBUGLOG(5, "ZSTD_compressBlock_splitBlock_internal: No splits"); + assert(cSizeSingleBlock <= ZSTD_BLOCKSIZE_MAX + ZSTD_blockHeaderSize); + return cSizeSingleBlock; + } + ZSTD_deriveSeqStoreChunk(currSeqStore, &zc->seqStore, 0, partitions[0]); - for (i = 0; i <= numSplits; ++i) { - size_t srcBytes; - size_t cSizeChunk; - U32 const lastPartition = (i == numSplits); - U32 lastBlockEntireSrc = 0; - + for (i = 0; i <= numSplits; ++i) { + size_t srcBytes; + size_t cSizeChunk; + U32 const lastPartition = (i == numSplits); + U32 lastBlockEntireSrc = 0; + srcBytes = ZSTD_countSeqStoreLiteralsBytes(currSeqStore) + ZSTD_countSeqStoreMatchBytes(currSeqStore); - srcBytesTotal += srcBytes; - if (lastPartition) { - /* This is the final partition, need to account for possible last literals */ - srcBytes += blockSize - srcBytesTotal; - lastBlockEntireSrc = lastBlock; - } else { + srcBytesTotal += srcBytes; + if (lastPartition) { + /* This is the final partition, need to account for possible last literals */ + srcBytes += blockSize - srcBytesTotal; + lastBlockEntireSrc = lastBlock; + } else { ZSTD_deriveSeqStoreChunk(nextSeqStore, &zc->seqStore, partitions[i], partitions[i+1]); - } - + } + cSizeChunk = ZSTD_compressSeqStore_singleBlock(zc, currSeqStore, - &dRep, &cRep, - op, dstCapacity, - ip, srcBytes, - lastBlockEntireSrc, 1 /* isPartition */); + &dRep, &cRep, + op, dstCapacity, + ip, srcBytes, + lastBlockEntireSrc, 1 /* isPartition */); DEBUGLOG(5, "Estimated size: %zu actual size: %zu", ZSTD_buildEntropyStatisticsAndEstimateSubBlockSize(currSeqStore, zc), cSizeChunk); - FORWARD_IF_ERROR(cSizeChunk, "Compressing chunk failed!"); - - ip += srcBytes; - op += cSizeChunk; - dstCapacity -= cSizeChunk; - cSize += cSizeChunk; + FORWARD_IF_ERROR(cSizeChunk, "Compressing chunk failed!"); + + ip += srcBytes; + op += cSizeChunk; + dstCapacity -= cSizeChunk; + cSize += cSizeChunk; *currSeqStore = *nextSeqStore; - assert(cSizeChunk <= ZSTD_BLOCKSIZE_MAX + ZSTD_blockHeaderSize); - } - /* cRep and dRep may have diverged during the compression. If so, we use the dRep repcodes - * for the next block. - */ - ZSTD_memcpy(zc->blockState.prevCBlock->rep, dRep.rep, sizeof(repcodes_t)); - return cSize; -} - + assert(cSizeChunk <= ZSTD_BLOCKSIZE_MAX + ZSTD_blockHeaderSize); + } + /* cRep and dRep may have diverged during the compression. If so, we use the dRep repcodes + * for the next block. + */ + ZSTD_memcpy(zc->blockState.prevCBlock->rep, dRep.rep, sizeof(repcodes_t)); + return cSize; +} + static size_t ZSTD_compressBlock_splitBlock(ZSTD_CCtx* zc, void* dst, size_t dstCapacity, const void* src, size_t srcSize, U32 lastBlock) { - const BYTE* ip = (const BYTE*)src; - BYTE* op = (BYTE*)dst; - U32 nbSeq; - size_t cSize; - DEBUGLOG(4, "ZSTD_compressBlock_splitBlock"); + const BYTE* ip = (const BYTE*)src; + BYTE* op = (BYTE*)dst; + U32 nbSeq; + size_t cSize; + DEBUGLOG(4, "ZSTD_compressBlock_splitBlock"); assert(zc->appliedParams.useBlockSplitter == ZSTD_ps_enable); - - { const size_t bss = ZSTD_buildSeqStore(zc, src, srcSize); - FORWARD_IF_ERROR(bss, "ZSTD_buildSeqStore failed"); - if (bss == ZSTDbss_noCompress) { - if (zc->blockState.prevCBlock->entropy.fse.offcode_repeatMode == FSE_repeat_valid) - zc->blockState.prevCBlock->entropy.fse.offcode_repeatMode = FSE_repeat_check; - cSize = ZSTD_noCompressBlock(op, dstCapacity, ip, srcSize, lastBlock); - FORWARD_IF_ERROR(cSize, "ZSTD_noCompressBlock failed"); - DEBUGLOG(4, "ZSTD_compressBlock_splitBlock: Nocompress block"); - return cSize; - } - nbSeq = (U32)(zc->seqStore.sequences - zc->seqStore.sequencesStart); - } - - cSize = ZSTD_compressBlock_splitBlock_internal(zc, dst, dstCapacity, src, srcSize, lastBlock, nbSeq); - FORWARD_IF_ERROR(cSize, "Splitting blocks failed!"); - return cSize; -} - + + { const size_t bss = ZSTD_buildSeqStore(zc, src, srcSize); + FORWARD_IF_ERROR(bss, "ZSTD_buildSeqStore failed"); + if (bss == ZSTDbss_noCompress) { + if (zc->blockState.prevCBlock->entropy.fse.offcode_repeatMode == FSE_repeat_valid) + zc->blockState.prevCBlock->entropy.fse.offcode_repeatMode = FSE_repeat_check; + cSize = ZSTD_noCompressBlock(op, dstCapacity, ip, srcSize, lastBlock); + FORWARD_IF_ERROR(cSize, "ZSTD_noCompressBlock failed"); + DEBUGLOG(4, "ZSTD_compressBlock_splitBlock: Nocompress block"); + return cSize; + } + nbSeq = (U32)(zc->seqStore.sequences - zc->seqStore.sequencesStart); + } + + cSize = ZSTD_compressBlock_splitBlock_internal(zc, dst, dstCapacity, src, srcSize, lastBlock, nbSeq); + FORWARD_IF_ERROR(cSize, "Splitting blocks failed!"); + return cSize; +} + static size_t ZSTD_compressBlock_internal(ZSTD_CCtx* zc, void* dst, size_t dstCapacity, @@ -3783,17 +3783,17 @@ ZSTD_compressBlock_internal(ZSTD_CCtx* zc, if (zc->seqCollector.collectSequences) { ZSTD_copyBlockSequences(zc); - ZSTD_blockState_confirmRepcodesAndEntropyTables(&zc->blockState); + ZSTD_blockState_confirmRepcodesAndEntropyTables(&zc->blockState); return 0; } /* encode sequences and literals */ - cSize = ZSTD_entropyCompressSeqStore(&zc->seqStore, + cSize = ZSTD_entropyCompressSeqStore(&zc->seqStore, &zc->blockState.prevCBlock->entropy, &zc->blockState.nextCBlock->entropy, &zc->appliedParams, dst, dstCapacity, srcSize, - zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */, + zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */, zc->bmi2); if (frame && @@ -3811,7 +3811,7 @@ ZSTD_compressBlock_internal(ZSTD_CCtx* zc, out: if (!ZSTD_isError(cSize) && cSize > 1) { - ZSTD_blockState_confirmRepcodesAndEntropyTables(&zc->blockState); + ZSTD_blockState_confirmRepcodesAndEntropyTables(&zc->blockState); } /* We check that dictionaries have offset codes available for the first * block. After the first block, the offcode table might not have large @@ -3864,7 +3864,7 @@ static size_t ZSTD_compressBlock_targetCBlockSize_body(ZSTD_CCtx* zc, size_t const maxCSize = srcSize - ZSTD_minGain(srcSize, zc->appliedParams.cParams.strategy); FORWARD_IF_ERROR(cSize, "ZSTD_compressSuperBlock failed"); if (cSize != 0 && cSize < maxCSize + ZSTD_blockHeaderSize) { - ZSTD_blockState_confirmRepcodesAndEntropyTables(&zc->blockState); + ZSTD_blockState_confirmRepcodesAndEntropyTables(&zc->blockState); return cSize; } } @@ -3904,9 +3904,9 @@ static void ZSTD_overflowCorrectIfNeeded(ZSTD_matchState_t* ms, void const* ip, void const* iend) { - U32 const cycleLog = ZSTD_cycleLog(params->cParams.chainLog, params->cParams.strategy); - U32 const maxDist = (U32)1 << params->cParams.windowLog; - if (ZSTD_window_needOverflowCorrection(ms->window, cycleLog, maxDist, ms->loadedDictEnd, ip, iend)) { + U32 const cycleLog = ZSTD_cycleLog(params->cParams.chainLog, params->cParams.strategy); + U32 const maxDist = (U32)1 << params->cParams.windowLog; + if (ZSTD_window_needOverflowCorrection(ms->window, cycleLog, maxDist, ms->loadedDictEnd, ip, iend)) { U32 const correction = ZSTD_window_correctOverflow(&ms->window, cycleLog, maxDist, ip); ZSTD_STATIC_ASSERT(ZSTD_CHAINLOG_MAX <= 30); ZSTD_STATIC_ASSERT(ZSTD_WINDOWLOG_MAX_32 <= 30); @@ -3929,7 +3929,7 @@ static void ZSTD_overflowCorrectIfNeeded(ZSTD_matchState_t* ms, * Frame is supposed already started (header already produced) * @return : compressed size, or an error code */ -static size_t ZSTD_compress_frameChunk(ZSTD_CCtx* cctx, +static size_t ZSTD_compress_frameChunk(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize, U32 lastFrameChunk) @@ -3943,7 +3943,7 @@ static size_t ZSTD_compress_frameChunk(ZSTD_CCtx* cctx, assert(cctx->appliedParams.cParams.windowLog <= ZSTD_WINDOWLOG_MAX); - DEBUGLOG(4, "ZSTD_compress_frameChunk (blockSize=%u)", (unsigned)blockSize); + DEBUGLOG(4, "ZSTD_compress_frameChunk (blockSize=%u)", (unsigned)blockSize); if (cctx->appliedParams.fParams.checksumFlag && srcSize) XXH64_update(&cctx->xxhState, src, srcSize); @@ -3970,10 +3970,10 @@ static size_t ZSTD_compress_frameChunk(ZSTD_CCtx* cctx, FORWARD_IF_ERROR(cSize, "ZSTD_compressBlock_targetCBlockSize failed"); assert(cSize > 0); assert(cSize <= blockSize + ZSTD_blockHeaderSize); - } else if (ZSTD_blockSplitterEnabled(&cctx->appliedParams)) { - cSize = ZSTD_compressBlock_splitBlock(cctx, op, dstCapacity, ip, blockSize, lastBlock); - FORWARD_IF_ERROR(cSize, "ZSTD_compressBlock_splitBlock failed"); - assert(cSize > 0 || cctx->seqCollector.collectSequences == 1); + } else if (ZSTD_blockSplitterEnabled(&cctx->appliedParams)) { + cSize = ZSTD_compressBlock_splitBlock(cctx, op, dstCapacity, ip, blockSize, lastBlock); + FORWARD_IF_ERROR(cSize, "ZSTD_compressBlock_splitBlock failed"); + assert(cSize > 0 || cctx->seqCollector.collectSequences == 1); } else { cSize = ZSTD_compressBlock_internal(cctx, op+ZSTD_blockHeaderSize, dstCapacity-ZSTD_blockHeaderSize, @@ -4057,26 +4057,26 @@ static size_t ZSTD_writeFrameHeader(void* dst, size_t dstCapacity, return pos; } -/* ZSTD_writeSkippableFrame_advanced() : - * Writes out a skippable frame with the specified magic number variant (16 are supported), - * from ZSTD_MAGIC_SKIPPABLE_START to ZSTD_MAGIC_SKIPPABLE_START+15, and the desired source data. - * - * Returns the total number of bytes written, or a ZSTD error code. - */ -size_t ZSTD_writeSkippableFrame(void* dst, size_t dstCapacity, - const void* src, size_t srcSize, unsigned magicVariant) { - BYTE* op = (BYTE*)dst; - RETURN_ERROR_IF(dstCapacity < srcSize + ZSTD_SKIPPABLEHEADERSIZE /* Skippable frame overhead */, - dstSize_tooSmall, "Not enough room for skippable frame"); - RETURN_ERROR_IF(srcSize > (unsigned)0xFFFFFFFF, srcSize_wrong, "Src size too large for skippable frame"); - RETURN_ERROR_IF(magicVariant > 15, parameter_outOfBound, "Skippable frame magic number variant not supported"); - - MEM_writeLE32(op, (U32)(ZSTD_MAGIC_SKIPPABLE_START + magicVariant)); - MEM_writeLE32(op+4, (U32)srcSize); - ZSTD_memcpy(op+8, src, srcSize); - return srcSize + ZSTD_SKIPPABLEHEADERSIZE; -} - +/* ZSTD_writeSkippableFrame_advanced() : + * Writes out a skippable frame with the specified magic number variant (16 are supported), + * from ZSTD_MAGIC_SKIPPABLE_START to ZSTD_MAGIC_SKIPPABLE_START+15, and the desired source data. + * + * Returns the total number of bytes written, or a ZSTD error code. + */ +size_t ZSTD_writeSkippableFrame(void* dst, size_t dstCapacity, + const void* src, size_t srcSize, unsigned magicVariant) { + BYTE* op = (BYTE*)dst; + RETURN_ERROR_IF(dstCapacity < srcSize + ZSTD_SKIPPABLEHEADERSIZE /* Skippable frame overhead */, + dstSize_tooSmall, "Not enough room for skippable frame"); + RETURN_ERROR_IF(srcSize > (unsigned)0xFFFFFFFF, srcSize_wrong, "Src size too large for skippable frame"); + RETURN_ERROR_IF(magicVariant > 15, parameter_outOfBound, "Skippable frame magic number variant not supported"); + + MEM_writeLE32(op, (U32)(ZSTD_MAGIC_SKIPPABLE_START + magicVariant)); + MEM_writeLE32(op+4, (U32)srcSize); + ZSTD_memcpy(op+8, src, srcSize); + return srcSize + ZSTD_SKIPPABLEHEADERSIZE; +} + /* ZSTD_writeLastEmptyBlock() : * output an empty Block with end-of-frame mark to complete a frame * @return : size of data written into `dst` (== ZSTD_blockHeaderSize (defined in zstd_internal.h)) @@ -4103,7 +4103,7 @@ size_t ZSTD_referenceExternalSequences(ZSTD_CCtx* cctx, rawSeq* seq, size_t nbSe cctx->externSeqStore.size = nbSeq; cctx->externSeqStore.capacity = nbSeq; cctx->externSeqStore.pos = 0; - cctx->externSeqStore.posInSequence = 0; + cctx->externSeqStore.posInSequence = 0; return 0; } @@ -4133,12 +4133,12 @@ static size_t ZSTD_compressContinue_internal (ZSTD_CCtx* cctx, if (!srcSize) return fhSize; /* do not generate an empty block if no input */ - if (!ZSTD_window_update(&ms->window, src, srcSize, ms->forceNonContiguous)) { - ms->forceNonContiguous = 0; + if (!ZSTD_window_update(&ms->window, src, srcSize, ms->forceNonContiguous)) { + ms->forceNonContiguous = 0; ms->nextToUpdate = ms->window.dictLimit; } if (cctx->appliedParams.ldmParams.enableLdm == ZSTD_ps_enable) { - ZSTD_window_update(&cctx->ldmState.window, src, srcSize, /* forceNonContiguous */ 0); + ZSTD_window_update(&cctx->ldmState.window, src, srcSize, /* forceNonContiguous */ 0); } if (!frame) { @@ -4208,84 +4208,84 @@ static size_t ZSTD_loadDictionaryContent(ZSTD_matchState_t* ms, const BYTE* const iend = ip + srcSize; int const loadLdmDict = params->ldmParams.enableLdm == ZSTD_ps_enable && ls != NULL; - /* Assert that we the ms params match the params we're being given */ - ZSTD_assertEqualCParams(params->cParams, ms->cParams); - - if (srcSize > ZSTD_CHUNKSIZE_MAX) { - /* Allow the dictionary to set indices up to exactly ZSTD_CURRENT_MAX. - * Dictionaries right at the edge will immediately trigger overflow - * correction, but I don't want to insert extra constraints here. - */ - U32 const maxDictSize = ZSTD_CURRENT_MAX - 1; - /* We must have cleared our windows when our source is this large. */ - assert(ZSTD_window_isEmpty(ms->window)); - if (loadLdmDict) - assert(ZSTD_window_isEmpty(ls->window)); - /* If the dictionary is too large, only load the suffix of the dictionary. */ - if (srcSize > maxDictSize) { - ip = iend - maxDictSize; - src = ip; - srcSize = maxDictSize; - } - } - - DEBUGLOG(4, "ZSTD_loadDictionaryContent(): useRowMatchFinder=%d", (int)params->useRowMatchFinder); - ZSTD_window_update(&ms->window, src, srcSize, /* forceNonContiguous */ 0); + /* Assert that we the ms params match the params we're being given */ + ZSTD_assertEqualCParams(params->cParams, ms->cParams); + + if (srcSize > ZSTD_CHUNKSIZE_MAX) { + /* Allow the dictionary to set indices up to exactly ZSTD_CURRENT_MAX. + * Dictionaries right at the edge will immediately trigger overflow + * correction, but I don't want to insert extra constraints here. + */ + U32 const maxDictSize = ZSTD_CURRENT_MAX - 1; + /* We must have cleared our windows when our source is this large. */ + assert(ZSTD_window_isEmpty(ms->window)); + if (loadLdmDict) + assert(ZSTD_window_isEmpty(ls->window)); + /* If the dictionary is too large, only load the suffix of the dictionary. */ + if (srcSize > maxDictSize) { + ip = iend - maxDictSize; + src = ip; + srcSize = maxDictSize; + } + } + + DEBUGLOG(4, "ZSTD_loadDictionaryContent(): useRowMatchFinder=%d", (int)params->useRowMatchFinder); + ZSTD_window_update(&ms->window, src, srcSize, /* forceNonContiguous */ 0); ms->loadedDictEnd = params->forceWindow ? 0 : (U32)(iend - ms->window.base); - ms->forceNonContiguous = params->deterministicRefPrefix; + ms->forceNonContiguous = params->deterministicRefPrefix; - if (loadLdmDict) { - ZSTD_window_update(&ls->window, src, srcSize, /* forceNonContiguous */ 0); + if (loadLdmDict) { + ZSTD_window_update(&ls->window, src, srcSize, /* forceNonContiguous */ 0); ls->loadedDictEnd = params->forceWindow ? 0 : (U32)(iend - ls->window.base); } if (srcSize <= HASH_READ_SIZE) return 0; - ZSTD_overflowCorrectIfNeeded(ms, ws, params, ip, iend); - - if (loadLdmDict) - ZSTD_ldm_fillHashTable(ls, ip, iend, ¶ms->ldmParams); - - switch(params->cParams.strategy) - { - case ZSTD_fast: - ZSTD_fillHashTable(ms, iend, dtlm); - break; - case ZSTD_dfast: - ZSTD_fillDoubleHashTable(ms, iend, dtlm); - break; - - case ZSTD_greedy: - case ZSTD_lazy: - case ZSTD_lazy2: - assert(srcSize >= HASH_READ_SIZE); - if (ms->dedicatedDictSearch) { - assert(ms->chainTable != NULL); - ZSTD_dedicatedDictSearch_lazy_loadDictionary(ms, iend-HASH_READ_SIZE); - } else { + ZSTD_overflowCorrectIfNeeded(ms, ws, params, ip, iend); + + if (loadLdmDict) + ZSTD_ldm_fillHashTable(ls, ip, iend, ¶ms->ldmParams); + + switch(params->cParams.strategy) + { + case ZSTD_fast: + ZSTD_fillHashTable(ms, iend, dtlm); + break; + case ZSTD_dfast: + ZSTD_fillDoubleHashTable(ms, iend, dtlm); + break; + + case ZSTD_greedy: + case ZSTD_lazy: + case ZSTD_lazy2: + assert(srcSize >= HASH_READ_SIZE); + if (ms->dedicatedDictSearch) { + assert(ms->chainTable != NULL); + ZSTD_dedicatedDictSearch_lazy_loadDictionary(ms, iend-HASH_READ_SIZE); + } else { assert(params->useRowMatchFinder != ZSTD_ps_auto); if (params->useRowMatchFinder == ZSTD_ps_enable) { - size_t const tagTableSize = ((size_t)1 << params->cParams.hashLog) * sizeof(U16); - ZSTD_memset(ms->tagTable, 0, tagTableSize); - ZSTD_row_update(ms, iend-HASH_READ_SIZE); - DEBUGLOG(4, "Using row-based hash table for lazy dict"); - } else { - ZSTD_insertAndFindFirstIndex(ms, iend-HASH_READ_SIZE); - DEBUGLOG(4, "Using chain-based hash table for lazy dict"); - } - } - break; - - case ZSTD_btlazy2: /* we want the dictionary table fully sorted */ - case ZSTD_btopt: - case ZSTD_btultra: - case ZSTD_btultra2: - assert(srcSize >= HASH_READ_SIZE); - ZSTD_updateTree(ms, iend-HASH_READ_SIZE, iend); - break; - - default: - assert(0); /* not possible : not a valid strategy id */ + size_t const tagTableSize = ((size_t)1 << params->cParams.hashLog) * sizeof(U16); + ZSTD_memset(ms->tagTable, 0, tagTableSize); + ZSTD_row_update(ms, iend-HASH_READ_SIZE); + DEBUGLOG(4, "Using row-based hash table for lazy dict"); + } else { + ZSTD_insertAndFindFirstIndex(ms, iend-HASH_READ_SIZE); + DEBUGLOG(4, "Using chain-based hash table for lazy dict"); + } + } + break; + + case ZSTD_btlazy2: /* we want the dictionary table fully sorted */ + case ZSTD_btopt: + case ZSTD_btultra: + case ZSTD_btultra2: + assert(srcSize >= HASH_READ_SIZE); + ZSTD_updateTree(ms, iend-HASH_READ_SIZE, iend); + break; + + default: + assert(0); /* not possible : not a valid strategy id */ } ms->nextToUpdate = (U32)(iend - ms->window.base); @@ -4294,28 +4294,28 @@ static size_t ZSTD_loadDictionaryContent(ZSTD_matchState_t* ms, /* Dictionaries that assign zero probability to symbols that show up causes problems - * when FSE encoding. Mark dictionaries with zero probability symbols as FSE_repeat_check - * and only dictionaries with 100% valid symbols can be assumed valid. - */ -static FSE_repeat ZSTD_dictNCountRepeat(short* normalizedCounter, unsigned dictMaxSymbolValue, unsigned maxSymbolValue) -{ + * when FSE encoding. Mark dictionaries with zero probability symbols as FSE_repeat_check + * and only dictionaries with 100% valid symbols can be assumed valid. + */ +static FSE_repeat ZSTD_dictNCountRepeat(short* normalizedCounter, unsigned dictMaxSymbolValue, unsigned maxSymbolValue) +{ U32 s; - if (dictMaxSymbolValue < maxSymbolValue) { - return FSE_repeat_check; - } + if (dictMaxSymbolValue < maxSymbolValue) { + return FSE_repeat_check; + } for (s = 0; s <= maxSymbolValue; ++s) { - if (normalizedCounter[s] == 0) { - return FSE_repeat_check; - } + if (normalizedCounter[s] == 0) { + return FSE_repeat_check; + } } - return FSE_repeat_valid; + return FSE_repeat_valid; } size_t ZSTD_loadCEntropy(ZSTD_compressedBlockState_t* bs, void* workspace, const void* const dict, size_t dictSize) { - short offcodeNCount[MaxOff+1]; - unsigned offcodeMaxValue = MaxOff; + short offcodeNCount[MaxOff+1]; + unsigned offcodeMaxValue = MaxOff; const BYTE* dictPtr = (const BYTE*)dict; /* skip magic num and dict ID */ const BYTE* const dictEnd = dictPtr + dictSize; dictPtr += 8; @@ -4337,7 +4337,7 @@ size_t ZSTD_loadCEntropy(ZSTD_compressedBlockState_t* bs, void* workspace, } { unsigned offcodeLog; - size_t const offcodeHeaderSize = FSE_readNCount(offcodeNCount, &offcodeMaxValue, &offcodeLog, dictPtr, dictEnd-dictPtr); + size_t const offcodeHeaderSize = FSE_readNCount(offcodeNCount, &offcodeMaxValue, &offcodeLog, dictPtr, dictEnd-dictPtr); RETURN_ERROR_IF(FSE_isError(offcodeHeaderSize), dictionary_corrupted, ""); RETURN_ERROR_IF(offcodeLog > OffFSELog, dictionary_corrupted, ""); /* fill all offset symbols to avoid garbage at end of table */ @@ -4346,7 +4346,7 @@ size_t ZSTD_loadCEntropy(ZSTD_compressedBlockState_t* bs, void* workspace, offcodeNCount, MaxOff, offcodeLog, workspace, HUF_WORKSPACE_SIZE)), dictionary_corrupted, ""); - /* Defer checking offcodeMaxValue because we need to know the size of the dictionary content */ + /* Defer checking offcodeMaxValue because we need to know the size of the dictionary content */ dictPtr += offcodeHeaderSize; } @@ -4360,7 +4360,7 @@ size_t ZSTD_loadCEntropy(ZSTD_compressedBlockState_t* bs, void* workspace, matchlengthNCount, matchlengthMaxValue, matchlengthLog, workspace, HUF_WORKSPACE_SIZE)), dictionary_corrupted, ""); - bs->entropy.fse.matchlength_repeatMode = ZSTD_dictNCountRepeat(matchlengthNCount, matchlengthMaxValue, MaxML); + bs->entropy.fse.matchlength_repeatMode = ZSTD_dictNCountRepeat(matchlengthNCount, matchlengthMaxValue, MaxML); dictPtr += matchlengthHeaderSize; } @@ -4374,7 +4374,7 @@ size_t ZSTD_loadCEntropy(ZSTD_compressedBlockState_t* bs, void* workspace, litlengthNCount, litlengthMaxValue, litlengthLog, workspace, HUF_WORKSPACE_SIZE)), dictionary_corrupted, ""); - bs->entropy.fse.litlength_repeatMode = ZSTD_dictNCountRepeat(litlengthNCount, litlengthMaxValue, MaxLL); + bs->entropy.fse.litlength_repeatMode = ZSTD_dictNCountRepeat(litlengthNCount, litlengthMaxValue, MaxLL); dictPtr += litlengthHeaderSize; } @@ -4384,28 +4384,28 @@ size_t ZSTD_loadCEntropy(ZSTD_compressedBlockState_t* bs, void* workspace, bs->rep[2] = MEM_readLE32(dictPtr+8); dictPtr += 12; - { size_t const dictContentSize = (size_t)(dictEnd - dictPtr); - U32 offcodeMax = MaxOff; - if (dictContentSize <= ((U32)-1) - 128 KB) { - U32 const maxOffset = (U32)dictContentSize + 128 KB; /* The maximum offset that must be supported */ - offcodeMax = ZSTD_highbit32(maxOffset); /* Calculate minimum offset code required to represent maxOffset */ - } - /* All offset values <= dictContentSize + 128 KB must be representable for a valid table */ - bs->entropy.fse.offcode_repeatMode = ZSTD_dictNCountRepeat(offcodeNCount, offcodeMaxValue, MIN(offcodeMax, MaxOff)); - - /* All repCodes must be <= dictContentSize and != 0 */ - { U32 u; - for (u=0; u<3; u++) { - RETURN_ERROR_IF(bs->rep[u] == 0, dictionary_corrupted, ""); - RETURN_ERROR_IF(bs->rep[u] > dictContentSize, dictionary_corrupted, ""); - } } } - + { size_t const dictContentSize = (size_t)(dictEnd - dictPtr); + U32 offcodeMax = MaxOff; + if (dictContentSize <= ((U32)-1) - 128 KB) { + U32 const maxOffset = (U32)dictContentSize + 128 KB; /* The maximum offset that must be supported */ + offcodeMax = ZSTD_highbit32(maxOffset); /* Calculate minimum offset code required to represent maxOffset */ + } + /* All offset values <= dictContentSize + 128 KB must be representable for a valid table */ + bs->entropy.fse.offcode_repeatMode = ZSTD_dictNCountRepeat(offcodeNCount, offcodeMaxValue, MIN(offcodeMax, MaxOff)); + + /* All repCodes must be <= dictContentSize and != 0 */ + { U32 u; + for (u=0; u<3; u++) { + RETURN_ERROR_IF(bs->rep[u] == 0, dictionary_corrupted, ""); + RETURN_ERROR_IF(bs->rep[u] > dictContentSize, dictionary_corrupted, ""); + } } } + return dictPtr - (const BYTE*)dict; } /* Dictionary format : * See : - * https://github.com/facebook/zstd/blob/release/doc/zstd_compression_format.md#dictionary-format + * https://github.com/facebook/zstd/blob/release/doc/zstd_compression_format.md#dictionary-format */ /*! ZSTD_loadZstdDictionary() : * @return : dictID, or an error code @@ -4429,16 +4429,16 @@ static size_t ZSTD_loadZstdDictionary(ZSTD_compressedBlockState_t* bs, assert(MEM_readLE32(dictPtr) == ZSTD_MAGIC_DICTIONARY); dictID = params->fParams.noDictIDFlag ? 0 : MEM_readLE32(dictPtr + 4 /* skip magic number */ ); - eSize = ZSTD_loadCEntropy(bs, workspace, dict, dictSize); + eSize = ZSTD_loadCEntropy(bs, workspace, dict, dictSize); FORWARD_IF_ERROR(eSize, "ZSTD_loadCEntropy failed"); dictPtr += eSize; - { - size_t const dictContentSize = (size_t)(dictEnd - dictPtr); + { + size_t const dictContentSize = (size_t)(dictEnd - dictPtr); FORWARD_IF_ERROR(ZSTD_loadDictionaryContent( ms, NULL, ws, params, dictPtr, dictContentSize, dtlm), ""); } - return dictID; + return dictID; } /** ZSTD_compress_insertDictionary() : @@ -4482,7 +4482,7 @@ ZSTD_compress_insertDictionary(ZSTD_compressedBlockState_t* bs, } #define ZSTD_USE_CDICT_PARAMS_SRCSIZE_CUTOFF (128 KB) -#define ZSTD_USE_CDICT_PARAMS_DICTSIZE_MULTIPLIER (6ULL) +#define ZSTD_USE_CDICT_PARAMS_DICTSIZE_MULTIPLIER (6ULL) /*! ZSTD_compressBegin_internal() : * @return : 0, or an error code */ @@ -4494,10 +4494,10 @@ static size_t ZSTD_compressBegin_internal(ZSTD_CCtx* cctx, const ZSTD_CCtx_params* params, U64 pledgedSrcSize, ZSTD_buffered_policy_e zbuff) { - size_t const dictContentSize = cdict ? cdict->dictContentSize : dictSize; -#if ZSTD_TRACE - cctx->traceCtx = (ZSTD_trace_compress_begin != NULL) ? ZSTD_trace_compress_begin(cctx) : 0; -#endif + size_t const dictContentSize = cdict ? cdict->dictContentSize : dictSize; +#if ZSTD_TRACE + cctx->traceCtx = (ZSTD_trace_compress_begin != NULL) ? ZSTD_trace_compress_begin(cctx) : 0; +#endif DEBUGLOG(4, "ZSTD_compressBegin_internal: wlog=%u", params->cParams.windowLog); /* params are supposed to be fully validated at this point */ assert(!ZSTD_isError(ZSTD_checkCParams(params->cParams))); @@ -4512,14 +4512,14 @@ static size_t ZSTD_compressBegin_internal(ZSTD_CCtx* cctx, return ZSTD_resetCCtx_usingCDict(cctx, cdict, params, pledgedSrcSize, zbuff); } - FORWARD_IF_ERROR( ZSTD_resetCCtx_internal(cctx, params, pledgedSrcSize, - dictContentSize, + FORWARD_IF_ERROR( ZSTD_resetCCtx_internal(cctx, params, pledgedSrcSize, + dictContentSize, ZSTDcrp_makeClean, zbuff) , ""); { size_t const dictID = cdict ? ZSTD_compress_insertDictionary( cctx->blockState.prevCBlock, &cctx->blockState.matchState, &cctx->ldmState, &cctx->workspace, &cctx->appliedParams, cdict->dictContent, - cdict->dictContentSize, cdict->dictContentType, dtlm, + cdict->dictContentSize, cdict->dictContentType, dtlm, cctx->entropyWorkspace) : ZSTD_compress_insertDictionary( cctx->blockState.prevCBlock, &cctx->blockState.matchState, @@ -4528,7 +4528,7 @@ static size_t ZSTD_compressBegin_internal(ZSTD_CCtx* cctx, FORWARD_IF_ERROR(dictID, "ZSTD_compress_insertDictionary failed"); assert(dictID <= UINT_MAX); cctx->dictID = (U32)dictID; - cctx->dictContentSize = dictContentSize; + cctx->dictContentSize = dictContentSize; } return 0; } @@ -4557,8 +4557,8 @@ size_t ZSTD_compressBegin_advanced(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, ZSTD_parameters params, unsigned long long pledgedSrcSize) { - ZSTD_CCtx_params cctxParams; - ZSTD_CCtxParams_init_internal(&cctxParams, ¶ms, ZSTD_NO_CLEVEL); + ZSTD_CCtx_params cctxParams; + ZSTD_CCtxParams_init_internal(&cctxParams, ¶ms, ZSTD_NO_CLEVEL); return ZSTD_compressBegin_advanced_internal(cctx, dict, dictSize, ZSTD_dct_auto, ZSTD_dtlm_fast, NULL /*cdict*/, @@ -4567,11 +4567,11 @@ size_t ZSTD_compressBegin_advanced(ZSTD_CCtx* cctx, size_t ZSTD_compressBegin_usingDict(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, int compressionLevel) { - ZSTD_CCtx_params cctxParams; - { - ZSTD_parameters const params = ZSTD_getParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_noAttachDict); - ZSTD_CCtxParams_init_internal(&cctxParams, ¶ms, (compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : compressionLevel); - } + ZSTD_CCtx_params cctxParams; + { + ZSTD_parameters const params = ZSTD_getParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_noAttachDict); + ZSTD_CCtxParams_init_internal(&cctxParams, ¶ms, (compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : compressionLevel); + } DEBUGLOG(4, "ZSTD_compressBegin_usingDict (dictSize=%u)", (unsigned)dictSize); return ZSTD_compressBegin_internal(cctx, dict, dictSize, ZSTD_dct_auto, ZSTD_dtlm_fast, NULL, &cctxParams, ZSTD_CONTENTSIZE_UNKNOWN, ZSTDb_not_buffered); @@ -4625,30 +4625,30 @@ static size_t ZSTD_writeEpilogue(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity) return op-ostart; } -void ZSTD_CCtx_trace(ZSTD_CCtx* cctx, size_t extraCSize) -{ -#if ZSTD_TRACE - if (cctx->traceCtx && ZSTD_trace_compress_end != NULL) { - int const streaming = cctx->inBuffSize > 0 || cctx->outBuffSize > 0 || cctx->appliedParams.nbWorkers > 0; - ZSTD_Trace trace; - ZSTD_memset(&trace, 0, sizeof(trace)); - trace.version = ZSTD_VERSION_NUMBER; - trace.streaming = streaming; - trace.dictionaryID = cctx->dictID; - trace.dictionarySize = cctx->dictContentSize; - trace.uncompressedSize = cctx->consumedSrcSize; - trace.compressedSize = cctx->producedCSize + extraCSize; - trace.params = &cctx->appliedParams; - trace.cctx = cctx; - ZSTD_trace_compress_end(cctx->traceCtx, &trace); - } - cctx->traceCtx = 0; -#else - (void)cctx; - (void)extraCSize; -#endif -} - +void ZSTD_CCtx_trace(ZSTD_CCtx* cctx, size_t extraCSize) +{ +#if ZSTD_TRACE + if (cctx->traceCtx && ZSTD_trace_compress_end != NULL) { + int const streaming = cctx->inBuffSize > 0 || cctx->outBuffSize > 0 || cctx->appliedParams.nbWorkers > 0; + ZSTD_Trace trace; + ZSTD_memset(&trace, 0, sizeof(trace)); + trace.version = ZSTD_VERSION_NUMBER; + trace.streaming = streaming; + trace.dictionaryID = cctx->dictID; + trace.dictionarySize = cctx->dictContentSize; + trace.uncompressedSize = cctx->consumedSrcSize; + trace.compressedSize = cctx->producedCSize + extraCSize; + trace.params = &cctx->appliedParams; + trace.cctx = cctx; + ZSTD_trace_compress_end(cctx->traceCtx, &trace); + } + cctx->traceCtx = 0; +#else + (void)cctx; + (void)extraCSize; +#endif +} + size_t ZSTD_compressEnd (ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize) @@ -4671,7 +4671,7 @@ size_t ZSTD_compressEnd (ZSTD_CCtx* cctx, (unsigned)cctx->pledgedSrcSizePlusOne-1, (unsigned)cctx->consumedSrcSize); } - ZSTD_CCtx_trace(cctx, endResult); + ZSTD_CCtx_trace(cctx, endResult); return cSize + endResult; } @@ -4683,12 +4683,12 @@ size_t ZSTD_compress_advanced (ZSTD_CCtx* cctx, { DEBUGLOG(4, "ZSTD_compress_advanced"); FORWARD_IF_ERROR(ZSTD_checkCParams(params.cParams), ""); - ZSTD_CCtxParams_init_internal(&cctx->simpleApiParams, ¶ms, ZSTD_NO_CLEVEL); - return ZSTD_compress_advanced_internal(cctx, - dst, dstCapacity, - src, srcSize, - dict, dictSize, - &cctx->simpleApiParams); + ZSTD_CCtxParams_init_internal(&cctx->simpleApiParams, ¶ms, ZSTD_NO_CLEVEL); + return ZSTD_compress_advanced_internal(cctx, + dst, dstCapacity, + src, srcSize, + dict, dictSize, + &cctx->simpleApiParams); } /* Internal */ @@ -4712,13 +4712,13 @@ size_t ZSTD_compress_usingDict(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, int compressionLevel) { - { - ZSTD_parameters const params = ZSTD_getParams_internal(compressionLevel, srcSize, dict ? dictSize : 0, ZSTD_cpm_noAttachDict); - assert(params.fParams.contentSizeFlag == 1); - ZSTD_CCtxParams_init_internal(&cctx->simpleApiParams, ¶ms, (compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT: compressionLevel); - } + { + ZSTD_parameters const params = ZSTD_getParams_internal(compressionLevel, srcSize, dict ? dictSize : 0, ZSTD_cpm_noAttachDict); + assert(params.fParams.contentSizeFlag == 1); + ZSTD_CCtxParams_init_internal(&cctx->simpleApiParams, ¶ms, (compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT: compressionLevel); + } DEBUGLOG(4, "ZSTD_compress_usingDict (srcSize=%u)", (unsigned)srcSize); - return ZSTD_compress_advanced_internal(cctx, dst, dstCapacity, src, srcSize, dict, dictSize, &cctx->simpleApiParams); + return ZSTD_compress_advanced_internal(cctx, dst, dstCapacity, src, srcSize, dict, dictSize, &cctx->simpleApiParams); } size_t ZSTD_compressCCtx(ZSTD_CCtx* cctx, @@ -4736,17 +4736,17 @@ size_t ZSTD_compress(void* dst, size_t dstCapacity, int compressionLevel) { size_t result; -#if ZSTD_COMPRESS_HEAPMODE - ZSTD_CCtx* cctx = ZSTD_createCCtx(); - RETURN_ERROR_IF(!cctx, memory_allocation, "ZSTD_createCCtx failed"); - result = ZSTD_compressCCtx(cctx, dst, dstCapacity, src, srcSize, compressionLevel); - ZSTD_freeCCtx(cctx); -#else +#if ZSTD_COMPRESS_HEAPMODE + ZSTD_CCtx* cctx = ZSTD_createCCtx(); + RETURN_ERROR_IF(!cctx, memory_allocation, "ZSTD_createCCtx failed"); + result = ZSTD_compressCCtx(cctx, dst, dstCapacity, src, srcSize, compressionLevel); + ZSTD_freeCCtx(cctx); +#else ZSTD_CCtx ctxBody; ZSTD_initCCtx(&ctxBody, ZSTD_defaultCMem); result = ZSTD_compressCCtx(&ctxBody, dst, dstCapacity, src, srcSize, compressionLevel); ZSTD_freeCCtxContent(&ctxBody); /* can't free ctxBody itself, as it's on stack; free only heap content */ -#endif +#endif return result; } @@ -4762,17 +4762,17 @@ size_t ZSTD_estimateCDictSize_advanced( DEBUGLOG(5, "sizeof(ZSTD_CDict) : %u", (unsigned)sizeof(ZSTD_CDict)); return ZSTD_cwksp_alloc_size(sizeof(ZSTD_CDict)) + ZSTD_cwksp_alloc_size(HUF_WORKSPACE_SIZE) - /* enableDedicatedDictSearch == 1 ensures that CDict estimation will not be too small - * in case we are using DDS with row-hash. */ + /* enableDedicatedDictSearch == 1 ensures that CDict estimation will not be too small + * in case we are using DDS with row-hash. */ + ZSTD_sizeof_matchState(&cParams, ZSTD_resolveRowMatchFinderMode(ZSTD_ps_auto, &cParams), - /* enableDedicatedDictSearch */ 1, /* forCCtx */ 0) + /* enableDedicatedDictSearch */ 1, /* forCCtx */ 0) + (dictLoadMethod == ZSTD_dlm_byRef ? 0 : ZSTD_cwksp_alloc_size(ZSTD_cwksp_align(dictSize, sizeof(void *)))); } size_t ZSTD_estimateCDictSize(size_t dictSize, int compressionLevel) { - ZSTD_compressionParameters const cParams = ZSTD_getCParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_createCDict); + ZSTD_compressionParameters const cParams = ZSTD_getCParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_createCDict); return ZSTD_estimateCDictSize_advanced(dictSize, cParams, ZSTD_dlm_byCopy); } @@ -4790,22 +4790,22 @@ static size_t ZSTD_initCDict_internal( const void* dictBuffer, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType, - ZSTD_CCtx_params params) + ZSTD_CCtx_params params) { DEBUGLOG(3, "ZSTD_initCDict_internal (dictContentType:%u)", (unsigned)dictContentType); - assert(!ZSTD_checkCParams(params.cParams)); - cdict->matchState.cParams = params.cParams; - cdict->matchState.dedicatedDictSearch = params.enableDedicatedDictSearch; + assert(!ZSTD_checkCParams(params.cParams)); + cdict->matchState.cParams = params.cParams; + cdict->matchState.dedicatedDictSearch = params.enableDedicatedDictSearch; if ((dictLoadMethod == ZSTD_dlm_byRef) || (!dictBuffer) || (!dictSize)) { cdict->dictContent = dictBuffer; } else { void *internalBuffer = ZSTD_cwksp_reserve_object(&cdict->workspace, ZSTD_cwksp_align(dictSize, sizeof(void*))); RETURN_ERROR_IF(!internalBuffer, memory_allocation, "NULL pointer!"); cdict->dictContent = internalBuffer; - ZSTD_memcpy(internalBuffer, dictBuffer, dictSize); + ZSTD_memcpy(internalBuffer, dictBuffer, dictSize); } cdict->dictContentSize = dictSize; - cdict->dictContentType = dictContentType; + cdict->dictContentType = dictContentType; cdict->entropyWorkspace = (U32*)ZSTD_cwksp_reserve_object(&cdict->workspace, HUF_WORKSPACE_SIZE); @@ -4815,15 +4815,15 @@ static size_t ZSTD_initCDict_internal( FORWARD_IF_ERROR(ZSTD_reset_matchState( &cdict->matchState, &cdict->workspace, - ¶ms.cParams, - params.useRowMatchFinder, + ¶ms.cParams, + params.useRowMatchFinder, ZSTDcrp_makeClean, ZSTDirp_reset, ZSTD_resetTarget_CDict), ""); /* (Maybe) load the dictionary * Skips loading the dictionary if it is < 8 bytes. */ - { params.compressionLevel = ZSTD_CLEVEL_DEFAULT; + { params.compressionLevel = ZSTD_CLEVEL_DEFAULT; params.fParams.contentSizeFlag = 1; { size_t const dictID = ZSTD_compress_insertDictionary( &cdict->cBlockState, &cdict->matchState, NULL, &cdict->workspace, @@ -4838,129 +4838,129 @@ static size_t ZSTD_initCDict_internal( return 0; } -static ZSTD_CDict* ZSTD_createCDict_advanced_internal(size_t dictSize, +static ZSTD_CDict* ZSTD_createCDict_advanced_internal(size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_compressionParameters cParams, + ZSTD_compressionParameters cParams, ZSTD_paramSwitch_e useRowMatchFinder, - U32 enableDedicatedDictSearch, - ZSTD_customMem customMem) + U32 enableDedicatedDictSearch, + ZSTD_customMem customMem) { - if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL; + if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL; { size_t const workspaceSize = ZSTD_cwksp_alloc_size(sizeof(ZSTD_CDict)) + ZSTD_cwksp_alloc_size(HUF_WORKSPACE_SIZE) + - ZSTD_sizeof_matchState(&cParams, useRowMatchFinder, enableDedicatedDictSearch, /* forCCtx */ 0) + + ZSTD_sizeof_matchState(&cParams, useRowMatchFinder, enableDedicatedDictSearch, /* forCCtx */ 0) + (dictLoadMethod == ZSTD_dlm_byRef ? 0 : ZSTD_cwksp_alloc_size(ZSTD_cwksp_align(dictSize, sizeof(void*)))); - void* const workspace = ZSTD_customMalloc(workspaceSize, customMem); + void* const workspace = ZSTD_customMalloc(workspaceSize, customMem); ZSTD_cwksp ws; ZSTD_CDict* cdict; if (!workspace) { - ZSTD_customFree(workspace, customMem); + ZSTD_customFree(workspace, customMem); return NULL; } - ZSTD_cwksp_init(&ws, workspace, workspaceSize, ZSTD_cwksp_dynamic_alloc); + ZSTD_cwksp_init(&ws, workspace, workspaceSize, ZSTD_cwksp_dynamic_alloc); cdict = (ZSTD_CDict*)ZSTD_cwksp_reserve_object(&ws, sizeof(ZSTD_CDict)); assert(cdict != NULL); ZSTD_cwksp_move(&cdict->workspace, &ws); cdict->customMem = customMem; - cdict->compressionLevel = ZSTD_NO_CLEVEL; /* signals advanced API usage */ - cdict->useRowMatchFinder = useRowMatchFinder; + cdict->compressionLevel = ZSTD_NO_CLEVEL; /* signals advanced API usage */ + cdict->useRowMatchFinder = useRowMatchFinder; return cdict; } } -ZSTD_CDict* ZSTD_createCDict_advanced(const void* dictBuffer, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType, - ZSTD_compressionParameters cParams, - ZSTD_customMem customMem) -{ - ZSTD_CCtx_params cctxParams; - ZSTD_memset(&cctxParams, 0, sizeof(cctxParams)); - ZSTD_CCtxParams_init(&cctxParams, 0); - cctxParams.cParams = cParams; - cctxParams.customMem = customMem; - return ZSTD_createCDict_advanced2( - dictBuffer, dictSize, - dictLoadMethod, dictContentType, - &cctxParams, customMem); -} - +ZSTD_CDict* ZSTD_createCDict_advanced(const void* dictBuffer, size_t dictSize, + ZSTD_dictLoadMethod_e dictLoadMethod, + ZSTD_dictContentType_e dictContentType, + ZSTD_compressionParameters cParams, + ZSTD_customMem customMem) +{ + ZSTD_CCtx_params cctxParams; + ZSTD_memset(&cctxParams, 0, sizeof(cctxParams)); + ZSTD_CCtxParams_init(&cctxParams, 0); + cctxParams.cParams = cParams; + cctxParams.customMem = customMem; + return ZSTD_createCDict_advanced2( + dictBuffer, dictSize, + dictLoadMethod, dictContentType, + &cctxParams, customMem); +} + ZSTD_CDict* ZSTD_createCDict_advanced2( - const void* dict, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType, - const ZSTD_CCtx_params* originalCctxParams, - ZSTD_customMem customMem) -{ - ZSTD_CCtx_params cctxParams = *originalCctxParams; - ZSTD_compressionParameters cParams; - ZSTD_CDict* cdict; - - DEBUGLOG(3, "ZSTD_createCDict_advanced2, mode %u", (unsigned)dictContentType); - if (!customMem.customAlloc ^ !customMem.customFree) return NULL; - - if (cctxParams.enableDedicatedDictSearch) { - cParams = ZSTD_dedicatedDictSearch_getCParams( - cctxParams.compressionLevel, dictSize); - ZSTD_overrideCParams(&cParams, &cctxParams.cParams); - } else { - cParams = ZSTD_getCParamsFromCCtxParams( - &cctxParams, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_createCDict); - } - - if (!ZSTD_dedicatedDictSearch_isSupported(&cParams)) { - /* Fall back to non-DDSS params */ - cctxParams.enableDedicatedDictSearch = 0; - cParams = ZSTD_getCParamsFromCCtxParams( - &cctxParams, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_createCDict); - } - - DEBUGLOG(3, "ZSTD_createCDict_advanced2: DDS: %u", cctxParams.enableDedicatedDictSearch); - cctxParams.cParams = cParams; - cctxParams.useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(cctxParams.useRowMatchFinder, &cParams); - - cdict = ZSTD_createCDict_advanced_internal(dictSize, - dictLoadMethod, cctxParams.cParams, - cctxParams.useRowMatchFinder, cctxParams.enableDedicatedDictSearch, - customMem); - - if (ZSTD_isError( ZSTD_initCDict_internal(cdict, - dict, dictSize, - dictLoadMethod, dictContentType, - cctxParams) )) { - ZSTD_freeCDict(cdict); - return NULL; - } - - return cdict; -} - + const void* dict, size_t dictSize, + ZSTD_dictLoadMethod_e dictLoadMethod, + ZSTD_dictContentType_e dictContentType, + const ZSTD_CCtx_params* originalCctxParams, + ZSTD_customMem customMem) +{ + ZSTD_CCtx_params cctxParams = *originalCctxParams; + ZSTD_compressionParameters cParams; + ZSTD_CDict* cdict; + + DEBUGLOG(3, "ZSTD_createCDict_advanced2, mode %u", (unsigned)dictContentType); + if (!customMem.customAlloc ^ !customMem.customFree) return NULL; + + if (cctxParams.enableDedicatedDictSearch) { + cParams = ZSTD_dedicatedDictSearch_getCParams( + cctxParams.compressionLevel, dictSize); + ZSTD_overrideCParams(&cParams, &cctxParams.cParams); + } else { + cParams = ZSTD_getCParamsFromCCtxParams( + &cctxParams, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_createCDict); + } + + if (!ZSTD_dedicatedDictSearch_isSupported(&cParams)) { + /* Fall back to non-DDSS params */ + cctxParams.enableDedicatedDictSearch = 0; + cParams = ZSTD_getCParamsFromCCtxParams( + &cctxParams, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_createCDict); + } + + DEBUGLOG(3, "ZSTD_createCDict_advanced2: DDS: %u", cctxParams.enableDedicatedDictSearch); + cctxParams.cParams = cParams; + cctxParams.useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(cctxParams.useRowMatchFinder, &cParams); + + cdict = ZSTD_createCDict_advanced_internal(dictSize, + dictLoadMethod, cctxParams.cParams, + cctxParams.useRowMatchFinder, cctxParams.enableDedicatedDictSearch, + customMem); + + if (ZSTD_isError( ZSTD_initCDict_internal(cdict, + dict, dictSize, + dictLoadMethod, dictContentType, + cctxParams) )) { + ZSTD_freeCDict(cdict); + return NULL; + } + + return cdict; +} + ZSTD_CDict* ZSTD_createCDict(const void* dict, size_t dictSize, int compressionLevel) { - ZSTD_compressionParameters cParams = ZSTD_getCParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_createCDict); - ZSTD_CDict* const cdict = ZSTD_createCDict_advanced(dict, dictSize, + ZSTD_compressionParameters cParams = ZSTD_getCParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_createCDict); + ZSTD_CDict* const cdict = ZSTD_createCDict_advanced(dict, dictSize, ZSTD_dlm_byCopy, ZSTD_dct_auto, cParams, ZSTD_defaultCMem); if (cdict) - cdict->compressionLevel = (compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : compressionLevel; + cdict->compressionLevel = (compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : compressionLevel; return cdict; } ZSTD_CDict* ZSTD_createCDict_byReference(const void* dict, size_t dictSize, int compressionLevel) { - ZSTD_compressionParameters cParams = ZSTD_getCParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_createCDict); - ZSTD_CDict* const cdict = ZSTD_createCDict_advanced(dict, dictSize, + ZSTD_compressionParameters cParams = ZSTD_getCParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_createCDict); + ZSTD_CDict* const cdict = ZSTD_createCDict_advanced(dict, dictSize, ZSTD_dlm_byRef, ZSTD_dct_auto, cParams, ZSTD_defaultCMem); - if (cdict) - cdict->compressionLevel = (compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : compressionLevel; - return cdict; + if (cdict) + cdict->compressionLevel = (compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : compressionLevel; + return cdict; } size_t ZSTD_freeCDict(ZSTD_CDict* cdict) @@ -4970,7 +4970,7 @@ size_t ZSTD_freeCDict(ZSTD_CDict* cdict) int cdictInWorkspace = ZSTD_cwksp_owns_buffer(&cdict->workspace, cdict); ZSTD_cwksp_free(&cdict->workspace, cMem); if (!cdictInWorkspace) { - ZSTD_customFree(cdict, cMem); + ZSTD_customFree(cdict, cMem); } return 0; } @@ -4997,21 +4997,21 @@ const ZSTD_CDict* ZSTD_initStaticCDict( ZSTD_compressionParameters cParams) { ZSTD_paramSwitch_e const useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(ZSTD_ps_auto, &cParams); - /* enableDedicatedDictSearch == 1 ensures matchstate is not too small in case this CDict will be used for DDS + row hash */ - size_t const matchStateSize = ZSTD_sizeof_matchState(&cParams, useRowMatchFinder, /* enableDedicatedDictSearch */ 1, /* forCCtx */ 0); + /* enableDedicatedDictSearch == 1 ensures matchstate is not too small in case this CDict will be used for DDS + row hash */ + size_t const matchStateSize = ZSTD_sizeof_matchState(&cParams, useRowMatchFinder, /* enableDedicatedDictSearch */ 1, /* forCCtx */ 0); size_t const neededSize = ZSTD_cwksp_alloc_size(sizeof(ZSTD_CDict)) + (dictLoadMethod == ZSTD_dlm_byRef ? 0 : ZSTD_cwksp_alloc_size(ZSTD_cwksp_align(dictSize, sizeof(void*)))) + ZSTD_cwksp_alloc_size(HUF_WORKSPACE_SIZE) + matchStateSize; ZSTD_CDict* cdict; - ZSTD_CCtx_params params; + ZSTD_CCtx_params params; if ((size_t)workspace & 7) return NULL; /* 8-aligned */ { ZSTD_cwksp ws; - ZSTD_cwksp_init(&ws, workspace, workspaceSize, ZSTD_cwksp_static_alloc); + ZSTD_cwksp_init(&ws, workspace, workspaceSize, ZSTD_cwksp_static_alloc); cdict = (ZSTD_CDict*)ZSTD_cwksp_reserve_object(&ws, sizeof(ZSTD_CDict)); if (cdict == NULL) return NULL; ZSTD_cwksp_move(&cdict->workspace, &ws); @@ -5021,15 +5021,15 @@ const ZSTD_CDict* ZSTD_initStaticCDict( (unsigned)workspaceSize, (unsigned)neededSize, (unsigned)(workspaceSize < neededSize)); if (workspaceSize < neededSize) return NULL; - ZSTD_CCtxParams_init(¶ms, 0); - params.cParams = cParams; - params.useRowMatchFinder = useRowMatchFinder; - cdict->useRowMatchFinder = useRowMatchFinder; - + ZSTD_CCtxParams_init(¶ms, 0); + params.cParams = cParams; + params.useRowMatchFinder = useRowMatchFinder; + cdict->useRowMatchFinder = useRowMatchFinder; + if (ZSTD_isError( ZSTD_initCDict_internal(cdict, dict, dictSize, dictLoadMethod, dictContentType, - params) )) + params) )) return NULL; return cdict; @@ -5041,98 +5041,98 @@ ZSTD_compressionParameters ZSTD_getCParamsFromCDict(const ZSTD_CDict* cdict) return cdict->matchState.cParams; } -/*! ZSTD_getDictID_fromCDict() : - * Provides the dictID of the dictionary loaded into `cdict`. - * If @return == 0, the dictionary is not conformant to Zstandard specification, or empty. - * Non-conformant dictionaries can still be loaded, but as content-only dictionaries. */ -unsigned ZSTD_getDictID_fromCDict(const ZSTD_CDict* cdict) -{ - if (cdict==NULL) return 0; - return cdict->dictID; -} - -/* ZSTD_compressBegin_usingCDict_internal() : - * Implementation of various ZSTD_compressBegin_usingCDict* functions. - */ -static size_t ZSTD_compressBegin_usingCDict_internal( +/*! ZSTD_getDictID_fromCDict() : + * Provides the dictID of the dictionary loaded into `cdict`. + * If @return == 0, the dictionary is not conformant to Zstandard specification, or empty. + * Non-conformant dictionaries can still be loaded, but as content-only dictionaries. */ +unsigned ZSTD_getDictID_fromCDict(const ZSTD_CDict* cdict) +{ + if (cdict==NULL) return 0; + return cdict->dictID; +} + +/* ZSTD_compressBegin_usingCDict_internal() : + * Implementation of various ZSTD_compressBegin_usingCDict* functions. + */ +static size_t ZSTD_compressBegin_usingCDict_internal( ZSTD_CCtx* const cctx, const ZSTD_CDict* const cdict, ZSTD_frameParameters const fParams, unsigned long long const pledgedSrcSize) { - ZSTD_CCtx_params cctxParams; - DEBUGLOG(4, "ZSTD_compressBegin_usingCDict_internal"); + ZSTD_CCtx_params cctxParams; + DEBUGLOG(4, "ZSTD_compressBegin_usingCDict_internal"); RETURN_ERROR_IF(cdict==NULL, dictionary_wrong, "NULL pointer!"); - /* Initialize the cctxParams from the cdict */ - { - ZSTD_parameters params; - params.fParams = fParams; + /* Initialize the cctxParams from the cdict */ + { + ZSTD_parameters params; + params.fParams = fParams; params.cParams = ( pledgedSrcSize < ZSTD_USE_CDICT_PARAMS_SRCSIZE_CUTOFF || pledgedSrcSize < cdict->dictContentSize * ZSTD_USE_CDICT_PARAMS_DICTSIZE_MULTIPLIER || pledgedSrcSize == ZSTD_CONTENTSIZE_UNKNOWN - || cdict->compressionLevel == 0 ) ? + || cdict->compressionLevel == 0 ) ? ZSTD_getCParamsFromCDict(cdict) : ZSTD_getCParams(cdict->compressionLevel, pledgedSrcSize, cdict->dictContentSize); - ZSTD_CCtxParams_init_internal(&cctxParams, ¶ms, cdict->compressionLevel); - } - /* Increase window log to fit the entire dictionary and source if the - * source size is known. Limit the increase to 19, which is the - * window log for compression level 1 with the largest source size. - */ - if (pledgedSrcSize != ZSTD_CONTENTSIZE_UNKNOWN) { - U32 const limitedSrcSize = (U32)MIN(pledgedSrcSize, 1U << 19); - U32 const limitedSrcLog = limitedSrcSize > 1 ? ZSTD_highbit32(limitedSrcSize - 1) + 1 : 1; - cctxParams.cParams.windowLog = MAX(cctxParams.cParams.windowLog, limitedSrcLog); - } - return ZSTD_compressBegin_internal(cctx, - NULL, 0, ZSTD_dct_auto, ZSTD_dtlm_fast, - cdict, - &cctxParams, pledgedSrcSize, - ZSTDb_not_buffered); -} - - -/* ZSTD_compressBegin_usingCDict_advanced() : - * This function is DEPRECATED. - * cdict must be != NULL */ -size_t ZSTD_compressBegin_usingCDict_advanced( - ZSTD_CCtx* const cctx, const ZSTD_CDict* const cdict, - ZSTD_frameParameters const fParams, unsigned long long const pledgedSrcSize) -{ - return ZSTD_compressBegin_usingCDict_internal(cctx, cdict, fParams, pledgedSrcSize); -} - + ZSTD_CCtxParams_init_internal(&cctxParams, ¶ms, cdict->compressionLevel); + } + /* Increase window log to fit the entire dictionary and source if the + * source size is known. Limit the increase to 19, which is the + * window log for compression level 1 with the largest source size. + */ + if (pledgedSrcSize != ZSTD_CONTENTSIZE_UNKNOWN) { + U32 const limitedSrcSize = (U32)MIN(pledgedSrcSize, 1U << 19); + U32 const limitedSrcLog = limitedSrcSize > 1 ? ZSTD_highbit32(limitedSrcSize - 1) + 1 : 1; + cctxParams.cParams.windowLog = MAX(cctxParams.cParams.windowLog, limitedSrcLog); + } + return ZSTD_compressBegin_internal(cctx, + NULL, 0, ZSTD_dct_auto, ZSTD_dtlm_fast, + cdict, + &cctxParams, pledgedSrcSize, + ZSTDb_not_buffered); +} + + +/* ZSTD_compressBegin_usingCDict_advanced() : + * This function is DEPRECATED. + * cdict must be != NULL */ +size_t ZSTD_compressBegin_usingCDict_advanced( + ZSTD_CCtx* const cctx, const ZSTD_CDict* const cdict, + ZSTD_frameParameters const fParams, unsigned long long const pledgedSrcSize) +{ + return ZSTD_compressBegin_usingCDict_internal(cctx, cdict, fParams, pledgedSrcSize); +} + /* ZSTD_compressBegin_usingCDict() : - * cdict must be != NULL */ + * cdict must be != NULL */ size_t ZSTD_compressBegin_usingCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict) { ZSTD_frameParameters const fParams = { 0 /*content*/, 0 /*checksum*/, 0 /*noDictID*/ }; - return ZSTD_compressBegin_usingCDict_internal(cctx, cdict, fParams, ZSTD_CONTENTSIZE_UNKNOWN); + return ZSTD_compressBegin_usingCDict_internal(cctx, cdict, fParams, ZSTD_CONTENTSIZE_UNKNOWN); } -/*! ZSTD_compress_usingCDict_internal(): - * Implementation of various ZSTD_compress_usingCDict* functions. - */ -static size_t ZSTD_compress_usingCDict_internal(ZSTD_CCtx* cctx, +/*! ZSTD_compress_usingCDict_internal(): + * Implementation of various ZSTD_compress_usingCDict* functions. + */ +static size_t ZSTD_compress_usingCDict_internal(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize, const ZSTD_CDict* cdict, ZSTD_frameParameters fParams) { - FORWARD_IF_ERROR(ZSTD_compressBegin_usingCDict_internal(cctx, cdict, fParams, srcSize), ""); /* will check if cdict != NULL */ + FORWARD_IF_ERROR(ZSTD_compressBegin_usingCDict_internal(cctx, cdict, fParams, srcSize), ""); /* will check if cdict != NULL */ return ZSTD_compressEnd(cctx, dst, dstCapacity, src, srcSize); } -/*! ZSTD_compress_usingCDict_advanced(): - * This function is DEPRECATED. - */ -size_t ZSTD_compress_usingCDict_advanced(ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const ZSTD_CDict* cdict, ZSTD_frameParameters fParams) -{ - return ZSTD_compress_usingCDict_internal(cctx, dst, dstCapacity, src, srcSize, cdict, fParams); -} - +/*! ZSTD_compress_usingCDict_advanced(): + * This function is DEPRECATED. + */ +size_t ZSTD_compress_usingCDict_advanced(ZSTD_CCtx* cctx, + void* dst, size_t dstCapacity, + const void* src, size_t srcSize, + const ZSTD_CDict* cdict, ZSTD_frameParameters fParams) +{ + return ZSTD_compress_usingCDict_internal(cctx, dst, dstCapacity, src, srcSize, cdict, fParams); +} + /*! ZSTD_compress_usingCDict() : * Compression using a digested Dictionary. * Faster startup than ZSTD_compress_usingDict(), recommended when same dictionary is used multiple times. @@ -5144,7 +5144,7 @@ size_t ZSTD_compress_usingCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict) { ZSTD_frameParameters const fParams = { 1 /*content*/, 0 /*checksum*/, 0 /*noDictID*/ }; - return ZSTD_compress_usingCDict_internal(cctx, dst, dstCapacity, src, srcSize, cdict, fParams); + return ZSTD_compress_usingCDict_internal(cctx, dst, dstCapacity, src, srcSize, cdict, fParams); } @@ -5185,12 +5185,12 @@ size_t ZSTD_CStreamOutSize(void) return ZSTD_compressBound(ZSTD_BLOCKSIZE_MAX) + ZSTD_blockHeaderSize + 4 /* 32-bits hash */ ; } -static ZSTD_cParamMode_e ZSTD_getCParamMode(ZSTD_CDict const* cdict, ZSTD_CCtx_params const* params, U64 pledgedSrcSize) +static ZSTD_cParamMode_e ZSTD_getCParamMode(ZSTD_CDict const* cdict, ZSTD_CCtx_params const* params, U64 pledgedSrcSize) { - if (cdict != NULL && ZSTD_shouldAttachDict(cdict, params, pledgedSrcSize)) - return ZSTD_cpm_attachDict; - else - return ZSTD_cpm_noAttachDict; + if (cdict != NULL && ZSTD_shouldAttachDict(cdict, params, pledgedSrcSize)) + return ZSTD_cpm_attachDict; + else + return ZSTD_cpm_noAttachDict; } /* ZSTD_resetCStream(): @@ -5274,7 +5274,7 @@ size_t ZSTD_initCStream_advanced(ZSTD_CStream* zcs, FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) , ""); FORWARD_IF_ERROR( ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize) , ""); FORWARD_IF_ERROR( ZSTD_checkCParams(params.cParams) , ""); - ZSTD_CCtxParams_setZstdParams(&zcs->requestedParams, ¶ms); + ZSTD_CCtxParams_setZstdParams(&zcs->requestedParams, ¶ms); FORWARD_IF_ERROR( ZSTD_CCtx_loadDictionary(zcs, dict, dictSize) , ""); return 0; } @@ -5340,17 +5340,17 @@ static size_t ZSTD_compressStream_generic(ZSTD_CStream* zcs, /* check expectations */ DEBUGLOG(5, "ZSTD_compressStream_generic, flush=%u", (unsigned)flushMode); - if (zcs->appliedParams.inBufferMode == ZSTD_bm_buffered) { - assert(zcs->inBuff != NULL); - assert(zcs->inBuffSize > 0); - } - if (zcs->appliedParams.outBufferMode == ZSTD_bm_buffered) { - assert(zcs->outBuff != NULL); - assert(zcs->outBuffSize > 0); - } + if (zcs->appliedParams.inBufferMode == ZSTD_bm_buffered) { + assert(zcs->inBuff != NULL); + assert(zcs->inBuffSize > 0); + } + if (zcs->appliedParams.outBufferMode == ZSTD_bm_buffered) { + assert(zcs->outBuff != NULL); + assert(zcs->outBuffSize > 0); + } assert(output->pos <= output->size); assert(input->pos <= input->size); - assert((U32)flushMode <= (U32)ZSTD_e_end); + assert((U32)flushMode <= (U32)ZSTD_e_end); while (someMoreWork) { switch(zcs->streamStage) @@ -5360,8 +5360,8 @@ static size_t ZSTD_compressStream_generic(ZSTD_CStream* zcs, case zcss_load: if ( (flushMode == ZSTD_e_end) - && ( (size_t)(oend-op) >= ZSTD_compressBound(iend-ip) /* Enough output space */ - || zcs->appliedParams.outBufferMode == ZSTD_bm_stable) /* OR we are allowed to return dstSizeTooSmall */ + && ( (size_t)(oend-op) >= ZSTD_compressBound(iend-ip) /* Enough output space */ + || zcs->appliedParams.outBufferMode == ZSTD_bm_stable) /* OR we are allowed to return dstSizeTooSmall */ && (zcs->inBuffPos == 0) ) { /* shortcut to compression pass directly into output buffer */ size_t const cSize = ZSTD_compressEnd(zcs, @@ -5374,9 +5374,9 @@ static size_t ZSTD_compressStream_generic(ZSTD_CStream* zcs, ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); someMoreWork = 0; break; } - /* complete loading into inBuffer in buffered mode */ - if (zcs->appliedParams.inBufferMode == ZSTD_bm_buffered) { - size_t const toLoad = zcs->inBuffTarget - zcs->inBuffPos; + /* complete loading into inBuffer in buffered mode */ + if (zcs->appliedParams.inBufferMode == ZSTD_bm_buffered) { + size_t const toLoad = zcs->inBuffTarget - zcs->inBuffPos; size_t const loaded = ZSTD_limitCopy( zcs->inBuff + zcs->inBuffPos, toLoad, ip, iend-ip); @@ -5396,49 +5396,49 @@ static size_t ZSTD_compressStream_generic(ZSTD_CStream* zcs, } /* compress current block (note : this stage cannot be stopped in the middle) */ DEBUGLOG(5, "stream compression stage (flushMode==%u)", flushMode); - { int const inputBuffered = (zcs->appliedParams.inBufferMode == ZSTD_bm_buffered); - void* cDst; + { int const inputBuffered = (zcs->appliedParams.inBufferMode == ZSTD_bm_buffered); + void* cDst; size_t cSize; size_t oSize = oend-op; - size_t const iSize = inputBuffered - ? zcs->inBuffPos - zcs->inToCompress - : MIN((size_t)(iend - ip), zcs->blockSize); - if (oSize >= ZSTD_compressBound(iSize) || zcs->appliedParams.outBufferMode == ZSTD_bm_stable) + size_t const iSize = inputBuffered + ? zcs->inBuffPos - zcs->inToCompress + : MIN((size_t)(iend - ip), zcs->blockSize); + if (oSize >= ZSTD_compressBound(iSize) || zcs->appliedParams.outBufferMode == ZSTD_bm_stable) cDst = op; /* compress into output buffer, to skip flush stage */ else cDst = zcs->outBuff, oSize = zcs->outBuffSize; - if (inputBuffered) { - unsigned const lastBlock = (flushMode == ZSTD_e_end) && (ip==iend); - cSize = lastBlock ? - ZSTD_compressEnd(zcs, cDst, oSize, - zcs->inBuff + zcs->inToCompress, iSize) : - ZSTD_compressContinue(zcs, cDst, oSize, - zcs->inBuff + zcs->inToCompress, iSize); - FORWARD_IF_ERROR(cSize, "%s", lastBlock ? "ZSTD_compressEnd failed" : "ZSTD_compressContinue failed"); - zcs->frameEnded = lastBlock; - /* prepare next block */ - zcs->inBuffTarget = zcs->inBuffPos + zcs->blockSize; - if (zcs->inBuffTarget > zcs->inBuffSize) - zcs->inBuffPos = 0, zcs->inBuffTarget = zcs->blockSize; - DEBUGLOG(5, "inBuffTarget:%u / inBuffSize:%u", - (unsigned)zcs->inBuffTarget, (unsigned)zcs->inBuffSize); - if (!lastBlock) - assert(zcs->inBuffTarget <= zcs->inBuffSize); - zcs->inToCompress = zcs->inBuffPos; - } else { - unsigned const lastBlock = (ip + iSize == iend); - assert(flushMode == ZSTD_e_end /* Already validated */); - cSize = lastBlock ? - ZSTD_compressEnd(zcs, cDst, oSize, ip, iSize) : - ZSTD_compressContinue(zcs, cDst, oSize, ip, iSize); - /* Consume the input prior to error checking to mirror buffered mode. */ - if (iSize > 0) - ip += iSize; - FORWARD_IF_ERROR(cSize, "%s", lastBlock ? "ZSTD_compressEnd failed" : "ZSTD_compressContinue failed"); - zcs->frameEnded = lastBlock; - if (lastBlock) - assert(ip == iend); - } + if (inputBuffered) { + unsigned const lastBlock = (flushMode == ZSTD_e_end) && (ip==iend); + cSize = lastBlock ? + ZSTD_compressEnd(zcs, cDst, oSize, + zcs->inBuff + zcs->inToCompress, iSize) : + ZSTD_compressContinue(zcs, cDst, oSize, + zcs->inBuff + zcs->inToCompress, iSize); + FORWARD_IF_ERROR(cSize, "%s", lastBlock ? "ZSTD_compressEnd failed" : "ZSTD_compressContinue failed"); + zcs->frameEnded = lastBlock; + /* prepare next block */ + zcs->inBuffTarget = zcs->inBuffPos + zcs->blockSize; + if (zcs->inBuffTarget > zcs->inBuffSize) + zcs->inBuffPos = 0, zcs->inBuffTarget = zcs->blockSize; + DEBUGLOG(5, "inBuffTarget:%u / inBuffSize:%u", + (unsigned)zcs->inBuffTarget, (unsigned)zcs->inBuffSize); + if (!lastBlock) + assert(zcs->inBuffTarget <= zcs->inBuffSize); + zcs->inToCompress = zcs->inBuffPos; + } else { + unsigned const lastBlock = (ip + iSize == iend); + assert(flushMode == ZSTD_e_end /* Already validated */); + cSize = lastBlock ? + ZSTD_compressEnd(zcs, cDst, oSize, ip, iSize) : + ZSTD_compressContinue(zcs, cDst, oSize, ip, iSize); + /* Consume the input prior to error checking to mirror buffered mode. */ + if (iSize > 0) + ip += iSize; + FORWARD_IF_ERROR(cSize, "%s", lastBlock ? "ZSTD_compressEnd failed" : "ZSTD_compressContinue failed"); + zcs->frameEnded = lastBlock; + if (lastBlock) + assert(ip == iend); + } if (cDst == op) { /* no need to flush */ op += cSize; if (zcs->frameEnded) { @@ -5455,7 +5455,7 @@ static size_t ZSTD_compressStream_generic(ZSTD_CStream* zcs, ZSTD_FALLTHROUGH; case zcss_flush: DEBUGLOG(5, "flush stage"); - assert(zcs->appliedParams.outBufferMode == ZSTD_bm_buffered); + assert(zcs->appliedParams.outBufferMode == ZSTD_bm_buffered); { size_t const toFlush = zcs->outBuffContentSize - zcs->outBuffFlushedSize; size_t const flushed = ZSTD_limitCopy(op, (size_t)(oend-op), zcs->outBuff + zcs->outBuffFlushedSize, toFlush); @@ -5510,127 +5510,127 @@ size_t ZSTD_compressStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output, ZSTD_inBuf return ZSTD_nextInputSizeHint_MTorST(zcs); } -/* After a compression call set the expected input/output buffer. - * This is validated at the start of the next compression call. - */ -static void ZSTD_setBufferExpectations(ZSTD_CCtx* cctx, ZSTD_outBuffer const* output, ZSTD_inBuffer const* input) -{ - if (cctx->appliedParams.inBufferMode == ZSTD_bm_stable) { - cctx->expectedInBuffer = *input; - } - if (cctx->appliedParams.outBufferMode == ZSTD_bm_stable) { - cctx->expectedOutBufferSize = output->size - output->pos; - } -} - -/* Validate that the input/output buffers match the expectations set by - * ZSTD_setBufferExpectations. - */ -static size_t ZSTD_checkBufferStability(ZSTD_CCtx const* cctx, - ZSTD_outBuffer const* output, - ZSTD_inBuffer const* input, - ZSTD_EndDirective endOp) -{ - if (cctx->appliedParams.inBufferMode == ZSTD_bm_stable) { - ZSTD_inBuffer const expect = cctx->expectedInBuffer; - if (expect.src != input->src || expect.pos != input->pos || expect.size != input->size) - RETURN_ERROR(srcBuffer_wrong, "ZSTD_c_stableInBuffer enabled but input differs!"); - if (endOp != ZSTD_e_end) - RETURN_ERROR(srcBuffer_wrong, "ZSTD_c_stableInBuffer can only be used with ZSTD_e_end!"); - } - if (cctx->appliedParams.outBufferMode == ZSTD_bm_stable) { - size_t const outBufferSize = output->size - output->pos; - if (cctx->expectedOutBufferSize != outBufferSize) - RETURN_ERROR(dstBuffer_wrong, "ZSTD_c_stableOutBuffer enabled but output size differs!"); - } - return 0; -} - -static size_t ZSTD_CCtx_init_compressStream2(ZSTD_CCtx* cctx, - ZSTD_EndDirective endOp, - size_t inSize) { - ZSTD_CCtx_params params = cctx->requestedParams; - ZSTD_prefixDict const prefixDict = cctx->prefixDict; - FORWARD_IF_ERROR( ZSTD_initLocalDict(cctx) , ""); /* Init the local dict if present. */ - ZSTD_memset(&cctx->prefixDict, 0, sizeof(cctx->prefixDict)); /* single usage */ - assert(prefixDict.dict==NULL || cctx->cdict==NULL); /* only one can be set */ - if (cctx->cdict && !cctx->localDict.cdict) { - /* Let the cdict's compression level take priority over the requested params. - * But do not take the cdict's compression level if the "cdict" is actually a localDict - * generated from ZSTD_initLocalDict(). - */ - params.compressionLevel = cctx->cdict->compressionLevel; - } - DEBUGLOG(4, "ZSTD_compressStream2 : transparent init stage"); - if (endOp == ZSTD_e_end) cctx->pledgedSrcSizePlusOne = inSize + 1; /* auto-fix pledgedSrcSize */ - { - size_t const dictSize = prefixDict.dict - ? prefixDict.dictSize - : (cctx->cdict ? cctx->cdict->dictContentSize : 0); - ZSTD_cParamMode_e const mode = ZSTD_getCParamMode(cctx->cdict, ¶ms, cctx->pledgedSrcSizePlusOne - 1); - params.cParams = ZSTD_getCParamsFromCCtxParams( - ¶ms, cctx->pledgedSrcSizePlusOne-1, - dictSize, mode); - } - +/* After a compression call set the expected input/output buffer. + * This is validated at the start of the next compression call. + */ +static void ZSTD_setBufferExpectations(ZSTD_CCtx* cctx, ZSTD_outBuffer const* output, ZSTD_inBuffer const* input) +{ + if (cctx->appliedParams.inBufferMode == ZSTD_bm_stable) { + cctx->expectedInBuffer = *input; + } + if (cctx->appliedParams.outBufferMode == ZSTD_bm_stable) { + cctx->expectedOutBufferSize = output->size - output->pos; + } +} + +/* Validate that the input/output buffers match the expectations set by + * ZSTD_setBufferExpectations. + */ +static size_t ZSTD_checkBufferStability(ZSTD_CCtx const* cctx, + ZSTD_outBuffer const* output, + ZSTD_inBuffer const* input, + ZSTD_EndDirective endOp) +{ + if (cctx->appliedParams.inBufferMode == ZSTD_bm_stable) { + ZSTD_inBuffer const expect = cctx->expectedInBuffer; + if (expect.src != input->src || expect.pos != input->pos || expect.size != input->size) + RETURN_ERROR(srcBuffer_wrong, "ZSTD_c_stableInBuffer enabled but input differs!"); + if (endOp != ZSTD_e_end) + RETURN_ERROR(srcBuffer_wrong, "ZSTD_c_stableInBuffer can only be used with ZSTD_e_end!"); + } + if (cctx->appliedParams.outBufferMode == ZSTD_bm_stable) { + size_t const outBufferSize = output->size - output->pos; + if (cctx->expectedOutBufferSize != outBufferSize) + RETURN_ERROR(dstBuffer_wrong, "ZSTD_c_stableOutBuffer enabled but output size differs!"); + } + return 0; +} + +static size_t ZSTD_CCtx_init_compressStream2(ZSTD_CCtx* cctx, + ZSTD_EndDirective endOp, + size_t inSize) { + ZSTD_CCtx_params params = cctx->requestedParams; + ZSTD_prefixDict const prefixDict = cctx->prefixDict; + FORWARD_IF_ERROR( ZSTD_initLocalDict(cctx) , ""); /* Init the local dict if present. */ + ZSTD_memset(&cctx->prefixDict, 0, sizeof(cctx->prefixDict)); /* single usage */ + assert(prefixDict.dict==NULL || cctx->cdict==NULL); /* only one can be set */ + if (cctx->cdict && !cctx->localDict.cdict) { + /* Let the cdict's compression level take priority over the requested params. + * But do not take the cdict's compression level if the "cdict" is actually a localDict + * generated from ZSTD_initLocalDict(). + */ + params.compressionLevel = cctx->cdict->compressionLevel; + } + DEBUGLOG(4, "ZSTD_compressStream2 : transparent init stage"); + if (endOp == ZSTD_e_end) cctx->pledgedSrcSizePlusOne = inSize + 1; /* auto-fix pledgedSrcSize */ + { + size_t const dictSize = prefixDict.dict + ? prefixDict.dictSize + : (cctx->cdict ? cctx->cdict->dictContentSize : 0); + ZSTD_cParamMode_e const mode = ZSTD_getCParamMode(cctx->cdict, ¶ms, cctx->pledgedSrcSizePlusOne - 1); + params.cParams = ZSTD_getCParamsFromCCtxParams( + ¶ms, cctx->pledgedSrcSizePlusOne-1, + dictSize, mode); + } + params.useBlockSplitter = ZSTD_resolveBlockSplitterMode(params.useBlockSplitter, ¶ms.cParams); params.ldmParams.enableLdm = ZSTD_resolveEnableLdm(params.ldmParams.enableLdm, ¶ms.cParams); - params.useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(params.useRowMatchFinder, ¶ms.cParams); - -#ifdef ZSTD_MULTITHREAD - if ((cctx->pledgedSrcSizePlusOne-1) <= ZSTDMT_JOBSIZE_MIN) { - params.nbWorkers = 0; /* do not invoke multi-threading when src size is too small */ - } - if (params.nbWorkers > 0) { -#if ZSTD_TRACE - cctx->traceCtx = (ZSTD_trace_compress_begin != NULL) ? ZSTD_trace_compress_begin(cctx) : 0; -#endif - /* mt context creation */ - if (cctx->mtctx == NULL) { - DEBUGLOG(4, "ZSTD_compressStream2: creating new mtctx for nbWorkers=%u", - params.nbWorkers); - cctx->mtctx = ZSTDMT_createCCtx_advanced((U32)params.nbWorkers, cctx->customMem, cctx->pool); - RETURN_ERROR_IF(cctx->mtctx == NULL, memory_allocation, "NULL pointer!"); - } - /* mt compression */ - DEBUGLOG(4, "call ZSTDMT_initCStream_internal as nbWorkers=%u", params.nbWorkers); - FORWARD_IF_ERROR( ZSTDMT_initCStream_internal( - cctx->mtctx, - prefixDict.dict, prefixDict.dictSize, prefixDict.dictContentType, - cctx->cdict, params, cctx->pledgedSrcSizePlusOne-1) , ""); - cctx->dictID = cctx->cdict ? cctx->cdict->dictID : 0; - cctx->dictContentSize = cctx->cdict ? cctx->cdict->dictContentSize : prefixDict.dictSize; - cctx->consumedSrcSize = 0; - cctx->producedCSize = 0; - cctx->streamStage = zcss_load; - cctx->appliedParams = params; - } else -#endif - { U64 const pledgedSrcSize = cctx->pledgedSrcSizePlusOne - 1; - assert(!ZSTD_isError(ZSTD_checkCParams(params.cParams))); - FORWARD_IF_ERROR( ZSTD_compressBegin_internal(cctx, - prefixDict.dict, prefixDict.dictSize, prefixDict.dictContentType, ZSTD_dtlm_fast, - cctx->cdict, - ¶ms, pledgedSrcSize, - ZSTDb_buffered) , ""); - assert(cctx->appliedParams.nbWorkers == 0); - cctx->inToCompress = 0; - cctx->inBuffPos = 0; - if (cctx->appliedParams.inBufferMode == ZSTD_bm_buffered) { - /* for small input: avoid automatic flush on reaching end of block, since - * it would require to add a 3-bytes null block to end frame - */ - cctx->inBuffTarget = cctx->blockSize + (cctx->blockSize == pledgedSrcSize); - } else { - cctx->inBuffTarget = 0; - } - cctx->outBuffContentSize = cctx->outBuffFlushedSize = 0; - cctx->streamStage = zcss_load; - cctx->frameEnded = 0; - } - return 0; -} - + params.useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(params.useRowMatchFinder, ¶ms.cParams); + +#ifdef ZSTD_MULTITHREAD + if ((cctx->pledgedSrcSizePlusOne-1) <= ZSTDMT_JOBSIZE_MIN) { + params.nbWorkers = 0; /* do not invoke multi-threading when src size is too small */ + } + if (params.nbWorkers > 0) { +#if ZSTD_TRACE + cctx->traceCtx = (ZSTD_trace_compress_begin != NULL) ? ZSTD_trace_compress_begin(cctx) : 0; +#endif + /* mt context creation */ + if (cctx->mtctx == NULL) { + DEBUGLOG(4, "ZSTD_compressStream2: creating new mtctx for nbWorkers=%u", + params.nbWorkers); + cctx->mtctx = ZSTDMT_createCCtx_advanced((U32)params.nbWorkers, cctx->customMem, cctx->pool); + RETURN_ERROR_IF(cctx->mtctx == NULL, memory_allocation, "NULL pointer!"); + } + /* mt compression */ + DEBUGLOG(4, "call ZSTDMT_initCStream_internal as nbWorkers=%u", params.nbWorkers); + FORWARD_IF_ERROR( ZSTDMT_initCStream_internal( + cctx->mtctx, + prefixDict.dict, prefixDict.dictSize, prefixDict.dictContentType, + cctx->cdict, params, cctx->pledgedSrcSizePlusOne-1) , ""); + cctx->dictID = cctx->cdict ? cctx->cdict->dictID : 0; + cctx->dictContentSize = cctx->cdict ? cctx->cdict->dictContentSize : prefixDict.dictSize; + cctx->consumedSrcSize = 0; + cctx->producedCSize = 0; + cctx->streamStage = zcss_load; + cctx->appliedParams = params; + } else +#endif + { U64 const pledgedSrcSize = cctx->pledgedSrcSizePlusOne - 1; + assert(!ZSTD_isError(ZSTD_checkCParams(params.cParams))); + FORWARD_IF_ERROR( ZSTD_compressBegin_internal(cctx, + prefixDict.dict, prefixDict.dictSize, prefixDict.dictContentType, ZSTD_dtlm_fast, + cctx->cdict, + ¶ms, pledgedSrcSize, + ZSTDb_buffered) , ""); + assert(cctx->appliedParams.nbWorkers == 0); + cctx->inToCompress = 0; + cctx->inBuffPos = 0; + if (cctx->appliedParams.inBufferMode == ZSTD_bm_buffered) { + /* for small input: avoid automatic flush on reaching end of block, since + * it would require to add a 3-bytes null block to end frame + */ + cctx->inBuffTarget = cctx->blockSize + (cctx->blockSize == pledgedSrcSize); + } else { + cctx->inBuffTarget = 0; + } + cctx->outBuffContentSize = cctx->outBuffFlushedSize = 0; + cctx->streamStage = zcss_load; + cctx->frameEnded = 0; + } + return 0; +} + size_t ZSTD_compressStream2( ZSTD_CCtx* cctx, ZSTD_outBuffer* output, ZSTD_inBuffer* input, @@ -5638,19 +5638,19 @@ size_t ZSTD_compressStream2( ZSTD_CCtx* cctx, { DEBUGLOG(5, "ZSTD_compressStream2, endOp=%u ", (unsigned)endOp); /* check conditions */ - RETURN_ERROR_IF(output->pos > output->size, dstSize_tooSmall, "invalid output buffer"); - RETURN_ERROR_IF(input->pos > input->size, srcSize_wrong, "invalid input buffer"); - RETURN_ERROR_IF((U32)endOp > (U32)ZSTD_e_end, parameter_outOfBound, "invalid endDirective"); - assert(cctx != NULL); + RETURN_ERROR_IF(output->pos > output->size, dstSize_tooSmall, "invalid output buffer"); + RETURN_ERROR_IF(input->pos > input->size, srcSize_wrong, "invalid input buffer"); + RETURN_ERROR_IF((U32)endOp > (U32)ZSTD_e_end, parameter_outOfBound, "invalid endDirective"); + assert(cctx != NULL); /* transparent initialization stage */ if (cctx->streamStage == zcss_init) { - FORWARD_IF_ERROR(ZSTD_CCtx_init_compressStream2(cctx, endOp, input->size), "CompressStream2 initialization failed"); - ZSTD_setBufferExpectations(cctx, output, input); /* Set initial buffer expectations now that we've initialized */ - } + FORWARD_IF_ERROR(ZSTD_CCtx_init_compressStream2(cctx, endOp, input->size), "CompressStream2 initialization failed"); + ZSTD_setBufferExpectations(cctx, output, input); /* Set initial buffer expectations now that we've initialized */ + } /* end of transparent initialization stage */ - FORWARD_IF_ERROR(ZSTD_checkBufferStability(cctx, output, input, endOp), "invalid buffers"); + FORWARD_IF_ERROR(ZSTD_checkBufferStability(cctx, output, input, endOp), "invalid buffers"); /* compression stage */ #ifdef ZSTD_MULTITHREAD if (cctx->appliedParams.nbWorkers > 0) { @@ -5659,48 +5659,48 @@ size_t ZSTD_compressStream2( ZSTD_CCtx* cctx, ZSTDMT_updateCParams_whileCompressing(cctx->mtctx, &cctx->requestedParams); cctx->cParamsChanged = 0; } - for (;;) { - size_t const ipos = input->pos; - size_t const opos = output->pos; + for (;;) { + size_t const ipos = input->pos; + size_t const opos = output->pos; flushMin = ZSTDMT_compressStream_generic(cctx->mtctx, output, input, endOp); - cctx->consumedSrcSize += (U64)(input->pos - ipos); - cctx->producedCSize += (U64)(output->pos - opos); + cctx->consumedSrcSize += (U64)(input->pos - ipos); + cctx->producedCSize += (U64)(output->pos - opos); if ( ZSTD_isError(flushMin) || (endOp == ZSTD_e_end && flushMin == 0) ) { /* compression completed */ - if (flushMin == 0) - ZSTD_CCtx_trace(cctx, 0); + if (flushMin == 0) + ZSTD_CCtx_trace(cctx, 0); ZSTD_CCtx_reset(cctx, ZSTD_reset_session_only); } FORWARD_IF_ERROR(flushMin, "ZSTDMT_compressStream_generic failed"); - - if (endOp == ZSTD_e_continue) { - /* We only require some progress with ZSTD_e_continue, not maximal progress. - * We're done if we've consumed or produced any bytes, or either buffer is - * full. - */ - if (input->pos != ipos || output->pos != opos || input->pos == input->size || output->pos == output->size) - break; - } else { - assert(endOp == ZSTD_e_flush || endOp == ZSTD_e_end); - /* We require maximal progress. We're done when the flush is complete or the - * output buffer is full. - */ - if (flushMin == 0 || output->pos == output->size) - break; - } - } + + if (endOp == ZSTD_e_continue) { + /* We only require some progress with ZSTD_e_continue, not maximal progress. + * We're done if we've consumed or produced any bytes, or either buffer is + * full. + */ + if (input->pos != ipos || output->pos != opos || input->pos == input->size || output->pos == output->size) + break; + } else { + assert(endOp == ZSTD_e_flush || endOp == ZSTD_e_end); + /* We require maximal progress. We're done when the flush is complete or the + * output buffer is full. + */ + if (flushMin == 0 || output->pos == output->size) + break; + } + } DEBUGLOG(5, "completed ZSTD_compressStream2 delegating to ZSTDMT_compressStream_generic"); /* Either we don't require maximum forward progress, we've finished the * flush, or we are out of output space. */ - assert(endOp == ZSTD_e_continue || flushMin == 0 || output->pos == output->size); - ZSTD_setBufferExpectations(cctx, output, input); + assert(endOp == ZSTD_e_continue || flushMin == 0 || output->pos == output->size); + ZSTD_setBufferExpectations(cctx, output, input); return flushMin; } #endif FORWARD_IF_ERROR( ZSTD_compressStream_generic(cctx, output, input, endOp) , ""); DEBUGLOG(5, "completed ZSTD_compressStream2"); - ZSTD_setBufferExpectations(cctx, output, input); + ZSTD_setBufferExpectations(cctx, output, input); return cctx->outBuffContentSize - cctx->outBuffFlushedSize; /* remaining to flush */ } @@ -5723,22 +5723,22 @@ size_t ZSTD_compress2(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize) { - ZSTD_bufferMode_e const originalInBufferMode = cctx->requestedParams.inBufferMode; - ZSTD_bufferMode_e const originalOutBufferMode = cctx->requestedParams.outBufferMode; + ZSTD_bufferMode_e const originalInBufferMode = cctx->requestedParams.inBufferMode; + ZSTD_bufferMode_e const originalOutBufferMode = cctx->requestedParams.outBufferMode; DEBUGLOG(4, "ZSTD_compress2 (srcSize=%u)", (unsigned)srcSize); ZSTD_CCtx_reset(cctx, ZSTD_reset_session_only); - /* Enable stable input/output buffers. */ - cctx->requestedParams.inBufferMode = ZSTD_bm_stable; - cctx->requestedParams.outBufferMode = ZSTD_bm_stable; + /* Enable stable input/output buffers. */ + cctx->requestedParams.inBufferMode = ZSTD_bm_stable; + cctx->requestedParams.outBufferMode = ZSTD_bm_stable; { size_t oPos = 0; size_t iPos = 0; size_t const result = ZSTD_compressStream2_simpleArgs(cctx, dst, dstCapacity, &oPos, src, srcSize, &iPos, ZSTD_e_end); - /* Reset to the original values. */ - cctx->requestedParams.inBufferMode = originalInBufferMode; - cctx->requestedParams.outBufferMode = originalOutBufferMode; + /* Reset to the original values. */ + cctx->requestedParams.inBufferMode = originalInBufferMode; + cctx->requestedParams.outBufferMode = originalOutBufferMode; FORWARD_IF_ERROR(result, "ZSTD_compressStream2_simpleArgs failed"); if (result != 0) { /* compression not completed, due to lack of output space */ assert(oPos == dstCapacity); @@ -5749,12 +5749,12 @@ size_t ZSTD_compress2(ZSTD_CCtx* cctx, } } -typedef struct { - U32 idx; /* Index in array of ZSTD_Sequence */ - U32 posInSequence; /* Position within sequence at idx */ - size_t posInSrc; /* Number of bytes given by sequences provided so far */ -} ZSTD_sequencePosition; - +typedef struct { + U32 idx; /* Index in array of ZSTD_Sequence */ + U32 posInSequence; /* Position within sequence at idx */ + size_t posInSrc; /* Number of bytes given by sequences provided so far */ +} ZSTD_sequencePosition; + /* ZSTD_validateSequence() : * @offCode : is presumed to follow format required by ZSTD_storeSeq() * @returns a ZSTD error code if sequence is not valid @@ -5764,395 +5764,395 @@ ZSTD_validateSequence(U32 offCode, U32 matchLength, size_t posInSrc, U32 windowLog, size_t dictSize) { U32 const windowSize = 1 << windowLog; - /* posInSrc represents the amount of data the the decoder would decode up to this point. - * As long as the amount of data decoded is less than or equal to window size, offsets may be - * larger than the total length of output decoded in order to reference the dict, even larger than - * window size. After output surpasses windowSize, we're limited to windowSize offsets again. - */ + /* posInSrc represents the amount of data the the decoder would decode up to this point. + * As long as the amount of data decoded is less than or equal to window size, offsets may be + * larger than the total length of output decoded in order to reference the dict, even larger than + * window size. After output surpasses windowSize, we're limited to windowSize offsets again. + */ size_t const offsetBound = posInSrc > windowSize ? (size_t)windowSize : posInSrc + (size_t)dictSize; RETURN_ERROR_IF(offCode > STORE_OFFSET(offsetBound), corruption_detected, "Offset too large!"); RETURN_ERROR_IF(matchLength < MINMATCH, corruption_detected, "Matchlength too small"); - return 0; -} - -/* Returns an offset code, given a sequence's raw offset, the ongoing repcode array, and whether litLength == 0 */ + return 0; +} + +/* Returns an offset code, given a sequence's raw offset, the ongoing repcode array, and whether litLength == 0 */ static U32 ZSTD_finalizeOffCode(U32 rawOffset, const U32 rep[ZSTD_REP_NUM], U32 ll0) { U32 offCode = STORE_OFFSET(rawOffset); - - if (!ll0 && rawOffset == rep[0]) { + + if (!ll0 && rawOffset == rep[0]) { offCode = STORE_REPCODE_1; - } else if (rawOffset == rep[1]) { + } else if (rawOffset == rep[1]) { offCode = STORE_REPCODE(2 - ll0); - } else if (rawOffset == rep[2]) { + } else if (rawOffset == rep[2]) { offCode = STORE_REPCODE(3 - ll0); - } else if (ll0 && rawOffset == rep[0] - 1) { + } else if (ll0 && rawOffset == rep[0] - 1) { offCode = STORE_REPCODE_3; - } - return offCode; -} - -/* Returns 0 on success, and a ZSTD_error otherwise. This function scans through an array of - * ZSTD_Sequence, storing the sequences it finds, until it reaches a block delimiter. - */ + } + return offCode; +} + +/* Returns 0 on success, and a ZSTD_error otherwise. This function scans through an array of + * ZSTD_Sequence, storing the sequences it finds, until it reaches a block delimiter. + */ static size_t ZSTD_copySequencesToSeqStoreExplicitBlockDelim(ZSTD_CCtx* cctx, ZSTD_sequencePosition* seqPos, const ZSTD_Sequence* const inSeqs, size_t inSeqsSize, const void* src, size_t blockSize) { - U32 idx = seqPos->idx; - BYTE const* ip = (BYTE const*)(src); - const BYTE* const iend = ip + blockSize; - repcodes_t updatedRepcodes; - U32 dictSize; - - if (cctx->cdict) { - dictSize = (U32)cctx->cdict->dictContentSize; - } else if (cctx->prefixDict.dict) { - dictSize = (U32)cctx->prefixDict.dictSize; - } else { - dictSize = 0; - } - ZSTD_memcpy(updatedRepcodes.rep, cctx->blockState.prevCBlock->rep, sizeof(repcodes_t)); - for (; (inSeqs[idx].matchLength != 0 || inSeqs[idx].offset != 0) && idx < inSeqsSize; ++idx) { + U32 idx = seqPos->idx; + BYTE const* ip = (BYTE const*)(src); + const BYTE* const iend = ip + blockSize; + repcodes_t updatedRepcodes; + U32 dictSize; + + if (cctx->cdict) { + dictSize = (U32)cctx->cdict->dictContentSize; + } else if (cctx->prefixDict.dict) { + dictSize = (U32)cctx->prefixDict.dictSize; + } else { + dictSize = 0; + } + ZSTD_memcpy(updatedRepcodes.rep, cctx->blockState.prevCBlock->rep, sizeof(repcodes_t)); + for (; (inSeqs[idx].matchLength != 0 || inSeqs[idx].offset != 0) && idx < inSeqsSize; ++idx) { U32 const litLength = inSeqs[idx].litLength; U32 const ll0 = (litLength == 0); U32 const matchLength = inSeqs[idx].matchLength; U32 const offCode = ZSTD_finalizeOffCode(inSeqs[idx].offset, updatedRepcodes.rep, ll0); ZSTD_updateRep(updatedRepcodes.rep, offCode, ll0); - - DEBUGLOG(6, "Storing sequence: (of: %u, ml: %u, ll: %u)", offCode, matchLength, litLength); - if (cctx->appliedParams.validateSequences) { - seqPos->posInSrc += litLength + matchLength; - FORWARD_IF_ERROR(ZSTD_validateSequence(offCode, matchLength, seqPos->posInSrc, + + DEBUGLOG(6, "Storing sequence: (of: %u, ml: %u, ll: %u)", offCode, matchLength, litLength); + if (cctx->appliedParams.validateSequences) { + seqPos->posInSrc += litLength + matchLength; + FORWARD_IF_ERROR(ZSTD_validateSequence(offCode, matchLength, seqPos->posInSrc, cctx->appliedParams.cParams.windowLog, dictSize), - "Sequence validation failed"); - } - RETURN_ERROR_IF(idx - seqPos->idx > cctx->seqStore.maxNbSeq, memory_allocation, - "Not enough memory allocated. Try adjusting ZSTD_c_minMatch."); + "Sequence validation failed"); + } + RETURN_ERROR_IF(idx - seqPos->idx > cctx->seqStore.maxNbSeq, memory_allocation, + "Not enough memory allocated. Try adjusting ZSTD_c_minMatch."); ZSTD_storeSeq(&cctx->seqStore, litLength, ip, iend, offCode, matchLength); - ip += matchLength + litLength; - } - ZSTD_memcpy(cctx->blockState.nextCBlock->rep, updatedRepcodes.rep, sizeof(repcodes_t)); - - if (inSeqs[idx].litLength) { - DEBUGLOG(6, "Storing last literals of size: %u", inSeqs[idx].litLength); - ZSTD_storeLastLiterals(&cctx->seqStore, ip, inSeqs[idx].litLength); - ip += inSeqs[idx].litLength; - seqPos->posInSrc += inSeqs[idx].litLength; - } - RETURN_ERROR_IF(ip != iend, corruption_detected, "Blocksize doesn't agree with block delimiter!"); - seqPos->idx = idx+1; - return 0; -} - -/* Returns the number of bytes to move the current read position back by. Only non-zero - * if we ended up splitting a sequence. Otherwise, it may return a ZSTD error if something - * went wrong. - * - * This function will attempt to scan through blockSize bytes represented by the sequences - * in inSeqs, storing any (partial) sequences. - * - * Occasionally, we may want to change the actual number of bytes we consumed from inSeqs to - * avoid splitting a match, or to avoid splitting a match such that it would produce a match - * smaller than MINMATCH. In this case, we return the number of bytes that we didn't read from this block. - */ + ip += matchLength + litLength; + } + ZSTD_memcpy(cctx->blockState.nextCBlock->rep, updatedRepcodes.rep, sizeof(repcodes_t)); + + if (inSeqs[idx].litLength) { + DEBUGLOG(6, "Storing last literals of size: %u", inSeqs[idx].litLength); + ZSTD_storeLastLiterals(&cctx->seqStore, ip, inSeqs[idx].litLength); + ip += inSeqs[idx].litLength; + seqPos->posInSrc += inSeqs[idx].litLength; + } + RETURN_ERROR_IF(ip != iend, corruption_detected, "Blocksize doesn't agree with block delimiter!"); + seqPos->idx = idx+1; + return 0; +} + +/* Returns the number of bytes to move the current read position back by. Only non-zero + * if we ended up splitting a sequence. Otherwise, it may return a ZSTD error if something + * went wrong. + * + * This function will attempt to scan through blockSize bytes represented by the sequences + * in inSeqs, storing any (partial) sequences. + * + * Occasionally, we may want to change the actual number of bytes we consumed from inSeqs to + * avoid splitting a match, or to avoid splitting a match such that it would produce a match + * smaller than MINMATCH. In this case, we return the number of bytes that we didn't read from this block. + */ static size_t ZSTD_copySequencesToSeqStoreNoBlockDelim(ZSTD_CCtx* cctx, ZSTD_sequencePosition* seqPos, const ZSTD_Sequence* const inSeqs, size_t inSeqsSize, const void* src, size_t blockSize) { - U32 idx = seqPos->idx; - U32 startPosInSequence = seqPos->posInSequence; - U32 endPosInSequence = seqPos->posInSequence + (U32)blockSize; - size_t dictSize; - BYTE const* ip = (BYTE const*)(src); - BYTE const* iend = ip + blockSize; /* May be adjusted if we decide to process fewer than blockSize bytes */ - repcodes_t updatedRepcodes; - U32 bytesAdjustment = 0; - U32 finalMatchSplit = 0; - - if (cctx->cdict) { - dictSize = cctx->cdict->dictContentSize; - } else if (cctx->prefixDict.dict) { - dictSize = cctx->prefixDict.dictSize; - } else { - dictSize = 0; - } - DEBUGLOG(5, "ZSTD_copySequencesToSeqStore: idx: %u PIS: %u blockSize: %zu", idx, startPosInSequence, blockSize); - DEBUGLOG(5, "Start seq: idx: %u (of: %u ml: %u ll: %u)", idx, inSeqs[idx].offset, inSeqs[idx].matchLength, inSeqs[idx].litLength); - ZSTD_memcpy(updatedRepcodes.rep, cctx->blockState.prevCBlock->rep, sizeof(repcodes_t)); - while (endPosInSequence && idx < inSeqsSize && !finalMatchSplit) { - const ZSTD_Sequence currSeq = inSeqs[idx]; + U32 idx = seqPos->idx; + U32 startPosInSequence = seqPos->posInSequence; + U32 endPosInSequence = seqPos->posInSequence + (U32)blockSize; + size_t dictSize; + BYTE const* ip = (BYTE const*)(src); + BYTE const* iend = ip + blockSize; /* May be adjusted if we decide to process fewer than blockSize bytes */ + repcodes_t updatedRepcodes; + U32 bytesAdjustment = 0; + U32 finalMatchSplit = 0; + + if (cctx->cdict) { + dictSize = cctx->cdict->dictContentSize; + } else if (cctx->prefixDict.dict) { + dictSize = cctx->prefixDict.dictSize; + } else { + dictSize = 0; + } + DEBUGLOG(5, "ZSTD_copySequencesToSeqStore: idx: %u PIS: %u blockSize: %zu", idx, startPosInSequence, blockSize); + DEBUGLOG(5, "Start seq: idx: %u (of: %u ml: %u ll: %u)", idx, inSeqs[idx].offset, inSeqs[idx].matchLength, inSeqs[idx].litLength); + ZSTD_memcpy(updatedRepcodes.rep, cctx->blockState.prevCBlock->rep, sizeof(repcodes_t)); + while (endPosInSequence && idx < inSeqsSize && !finalMatchSplit) { + const ZSTD_Sequence currSeq = inSeqs[idx]; U32 litLength = currSeq.litLength; U32 matchLength = currSeq.matchLength; U32 const rawOffset = currSeq.offset; U32 offCode; - - /* Modify the sequence depending on where endPosInSequence lies */ - if (endPosInSequence >= currSeq.litLength + currSeq.matchLength) { - if (startPosInSequence >= litLength) { - startPosInSequence -= litLength; - litLength = 0; - matchLength -= startPosInSequence; - } else { - litLength -= startPosInSequence; - } - /* Move to the next sequence */ - endPosInSequence -= currSeq.litLength + currSeq.matchLength; - startPosInSequence = 0; - idx++; - } else { - /* This is the final (partial) sequence we're adding from inSeqs, and endPosInSequence - does not reach the end of the match. So, we have to split the sequence */ - DEBUGLOG(6, "Require a split: diff: %u, idx: %u PIS: %u", - currSeq.litLength + currSeq.matchLength - endPosInSequence, idx, endPosInSequence); - if (endPosInSequence > litLength) { - U32 firstHalfMatchLength; - litLength = startPosInSequence >= litLength ? 0 : litLength - startPosInSequence; - firstHalfMatchLength = endPosInSequence - startPosInSequence - litLength; - if (matchLength > blockSize && firstHalfMatchLength >= cctx->appliedParams.cParams.minMatch) { - /* Only ever split the match if it is larger than the block size */ - U32 secondHalfMatchLength = currSeq.matchLength + currSeq.litLength - endPosInSequence; - if (secondHalfMatchLength < cctx->appliedParams.cParams.minMatch) { - /* Move the endPosInSequence backward so that it creates match of minMatch length */ - endPosInSequence -= cctx->appliedParams.cParams.minMatch - secondHalfMatchLength; - bytesAdjustment = cctx->appliedParams.cParams.minMatch - secondHalfMatchLength; - firstHalfMatchLength -= bytesAdjustment; - } - matchLength = firstHalfMatchLength; - /* Flag that we split the last match - after storing the sequence, exit the loop, - but keep the value of endPosInSequence */ - finalMatchSplit = 1; - } else { - /* Move the position in sequence backwards so that we don't split match, and break to store - * the last literals. We use the original currSeq.litLength as a marker for where endPosInSequence - * should go. We prefer to do this whenever it is not necessary to split the match, or if doing so - * would cause the first half of the match to be too small - */ - bytesAdjustment = endPosInSequence - currSeq.litLength; - endPosInSequence = currSeq.litLength; - break; - } - } else { - /* This sequence ends inside the literals, break to store the last literals */ - break; - } - } - /* Check if this offset can be represented with a repcode */ + + /* Modify the sequence depending on where endPosInSequence lies */ + if (endPosInSequence >= currSeq.litLength + currSeq.matchLength) { + if (startPosInSequence >= litLength) { + startPosInSequence -= litLength; + litLength = 0; + matchLength -= startPosInSequence; + } else { + litLength -= startPosInSequence; + } + /* Move to the next sequence */ + endPosInSequence -= currSeq.litLength + currSeq.matchLength; + startPosInSequence = 0; + idx++; + } else { + /* This is the final (partial) sequence we're adding from inSeqs, and endPosInSequence + does not reach the end of the match. So, we have to split the sequence */ + DEBUGLOG(6, "Require a split: diff: %u, idx: %u PIS: %u", + currSeq.litLength + currSeq.matchLength - endPosInSequence, idx, endPosInSequence); + if (endPosInSequence > litLength) { + U32 firstHalfMatchLength; + litLength = startPosInSequence >= litLength ? 0 : litLength - startPosInSequence; + firstHalfMatchLength = endPosInSequence - startPosInSequence - litLength; + if (matchLength > blockSize && firstHalfMatchLength >= cctx->appliedParams.cParams.minMatch) { + /* Only ever split the match if it is larger than the block size */ + U32 secondHalfMatchLength = currSeq.matchLength + currSeq.litLength - endPosInSequence; + if (secondHalfMatchLength < cctx->appliedParams.cParams.minMatch) { + /* Move the endPosInSequence backward so that it creates match of minMatch length */ + endPosInSequence -= cctx->appliedParams.cParams.minMatch - secondHalfMatchLength; + bytesAdjustment = cctx->appliedParams.cParams.minMatch - secondHalfMatchLength; + firstHalfMatchLength -= bytesAdjustment; + } + matchLength = firstHalfMatchLength; + /* Flag that we split the last match - after storing the sequence, exit the loop, + but keep the value of endPosInSequence */ + finalMatchSplit = 1; + } else { + /* Move the position in sequence backwards so that we don't split match, and break to store + * the last literals. We use the original currSeq.litLength as a marker for where endPosInSequence + * should go. We prefer to do this whenever it is not necessary to split the match, or if doing so + * would cause the first half of the match to be too small + */ + bytesAdjustment = endPosInSequence - currSeq.litLength; + endPosInSequence = currSeq.litLength; + break; + } + } else { + /* This sequence ends inside the literals, break to store the last literals */ + break; + } + } + /* Check if this offset can be represented with a repcode */ { U32 const ll0 = (litLength == 0); - offCode = ZSTD_finalizeOffCode(rawOffset, updatedRepcodes.rep, ll0); + offCode = ZSTD_finalizeOffCode(rawOffset, updatedRepcodes.rep, ll0); ZSTD_updateRep(updatedRepcodes.rep, offCode, ll0); - } - - if (cctx->appliedParams.validateSequences) { - seqPos->posInSrc += litLength + matchLength; - FORWARD_IF_ERROR(ZSTD_validateSequence(offCode, matchLength, seqPos->posInSrc, + } + + if (cctx->appliedParams.validateSequences) { + seqPos->posInSrc += litLength + matchLength; + FORWARD_IF_ERROR(ZSTD_validateSequence(offCode, matchLength, seqPos->posInSrc, cctx->appliedParams.cParams.windowLog, dictSize), - "Sequence validation failed"); - } - DEBUGLOG(6, "Storing sequence: (of: %u, ml: %u, ll: %u)", offCode, matchLength, litLength); - RETURN_ERROR_IF(idx - seqPos->idx > cctx->seqStore.maxNbSeq, memory_allocation, - "Not enough memory allocated. Try adjusting ZSTD_c_minMatch."); + "Sequence validation failed"); + } + DEBUGLOG(6, "Storing sequence: (of: %u, ml: %u, ll: %u)", offCode, matchLength, litLength); + RETURN_ERROR_IF(idx - seqPos->idx > cctx->seqStore.maxNbSeq, memory_allocation, + "Not enough memory allocated. Try adjusting ZSTD_c_minMatch."); ZSTD_storeSeq(&cctx->seqStore, litLength, ip, iend, offCode, matchLength); - ip += matchLength + litLength; - } - DEBUGLOG(5, "Ending seq: idx: %u (of: %u ml: %u ll: %u)", idx, inSeqs[idx].offset, inSeqs[idx].matchLength, inSeqs[idx].litLength); - assert(idx == inSeqsSize || endPosInSequence <= inSeqs[idx].litLength + inSeqs[idx].matchLength); - seqPos->idx = idx; - seqPos->posInSequence = endPosInSequence; - ZSTD_memcpy(cctx->blockState.nextCBlock->rep, updatedRepcodes.rep, sizeof(repcodes_t)); - - iend -= bytesAdjustment; - if (ip != iend) { - /* Store any last literals */ - U32 lastLLSize = (U32)(iend - ip); - assert(ip <= iend); - DEBUGLOG(6, "Storing last literals of size: %u", lastLLSize); - ZSTD_storeLastLiterals(&cctx->seqStore, ip, lastLLSize); - seqPos->posInSrc += lastLLSize; - } - - return bytesAdjustment; -} - -typedef size_t (*ZSTD_sequenceCopier) (ZSTD_CCtx* cctx, ZSTD_sequencePosition* seqPos, - const ZSTD_Sequence* const inSeqs, size_t inSeqsSize, - const void* src, size_t blockSize); + ip += matchLength + litLength; + } + DEBUGLOG(5, "Ending seq: idx: %u (of: %u ml: %u ll: %u)", idx, inSeqs[idx].offset, inSeqs[idx].matchLength, inSeqs[idx].litLength); + assert(idx == inSeqsSize || endPosInSequence <= inSeqs[idx].litLength + inSeqs[idx].matchLength); + seqPos->idx = idx; + seqPos->posInSequence = endPosInSequence; + ZSTD_memcpy(cctx->blockState.nextCBlock->rep, updatedRepcodes.rep, sizeof(repcodes_t)); + + iend -= bytesAdjustment; + if (ip != iend) { + /* Store any last literals */ + U32 lastLLSize = (U32)(iend - ip); + assert(ip <= iend); + DEBUGLOG(6, "Storing last literals of size: %u", lastLLSize); + ZSTD_storeLastLiterals(&cctx->seqStore, ip, lastLLSize); + seqPos->posInSrc += lastLLSize; + } + + return bytesAdjustment; +} + +typedef size_t (*ZSTD_sequenceCopier) (ZSTD_CCtx* cctx, ZSTD_sequencePosition* seqPos, + const ZSTD_Sequence* const inSeqs, size_t inSeqsSize, + const void* src, size_t blockSize); static ZSTD_sequenceCopier ZSTD_selectSequenceCopier(ZSTD_sequenceFormat_e mode) { - ZSTD_sequenceCopier sequenceCopier = NULL; - assert(ZSTD_cParam_withinBounds(ZSTD_c_blockDelimiters, mode)); - if (mode == ZSTD_sf_explicitBlockDelimiters) { - return ZSTD_copySequencesToSeqStoreExplicitBlockDelim; - } else if (mode == ZSTD_sf_noBlockDelimiters) { - return ZSTD_copySequencesToSeqStoreNoBlockDelim; - } - assert(sequenceCopier != NULL); - return sequenceCopier; -} - -/* Compress, block-by-block, all of the sequences given. - * + ZSTD_sequenceCopier sequenceCopier = NULL; + assert(ZSTD_cParam_withinBounds(ZSTD_c_blockDelimiters, mode)); + if (mode == ZSTD_sf_explicitBlockDelimiters) { + return ZSTD_copySequencesToSeqStoreExplicitBlockDelim; + } else if (mode == ZSTD_sf_noBlockDelimiters) { + return ZSTD_copySequencesToSeqStoreNoBlockDelim; + } + assert(sequenceCopier != NULL); + return sequenceCopier; +} + +/* Compress, block-by-block, all of the sequences given. + * * Returns the cumulative size of all compressed blocks (including their headers), * otherwise a ZSTD error. - */ + */ static size_t ZSTD_compressSequences_internal(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const ZSTD_Sequence* inSeqs, size_t inSeqsSize, const void* src, size_t srcSize) { - size_t cSize = 0; - U32 lastBlock; - size_t blockSize; - size_t compressedSeqsSize; - size_t remaining = srcSize; - ZSTD_sequencePosition seqPos = {0, 0, 0}; - - BYTE const* ip = (BYTE const*)src; - BYTE* op = (BYTE*)dst; + size_t cSize = 0; + U32 lastBlock; + size_t blockSize; + size_t compressedSeqsSize; + size_t remaining = srcSize; + ZSTD_sequencePosition seqPos = {0, 0, 0}; + + BYTE const* ip = (BYTE const*)src; + BYTE* op = (BYTE*)dst; ZSTD_sequenceCopier const sequenceCopier = ZSTD_selectSequenceCopier(cctx->appliedParams.blockDelimiters); - - DEBUGLOG(4, "ZSTD_compressSequences_internal srcSize: %zu, inSeqsSize: %zu", srcSize, inSeqsSize); - /* Special case: empty frame */ - if (remaining == 0) { - U32 const cBlockHeader24 = 1 /* last block */ + (((U32)bt_raw)<<1); - RETURN_ERROR_IF(dstCapacity<4, dstSize_tooSmall, "No room for empty frame block header"); - MEM_writeLE32(op, cBlockHeader24); - op += ZSTD_blockHeaderSize; - dstCapacity -= ZSTD_blockHeaderSize; - cSize += ZSTD_blockHeaderSize; - } - - while (remaining) { - size_t cBlockSize; - size_t additionalByteAdjustment; - lastBlock = remaining <= cctx->blockSize; - blockSize = lastBlock ? (U32)remaining : (U32)cctx->blockSize; - ZSTD_resetSeqStore(&cctx->seqStore); - DEBUGLOG(4, "Working on new block. Blocksize: %zu", blockSize); - - additionalByteAdjustment = sequenceCopier(cctx, &seqPos, inSeqs, inSeqsSize, ip, blockSize); - FORWARD_IF_ERROR(additionalByteAdjustment, "Bad sequence copy"); - blockSize -= additionalByteAdjustment; - - /* If blocks are too small, emit as a nocompress block */ - if (blockSize < MIN_CBLOCK_SIZE+ZSTD_blockHeaderSize+1) { - cBlockSize = ZSTD_noCompressBlock(op, dstCapacity, ip, blockSize, lastBlock); - FORWARD_IF_ERROR(cBlockSize, "Nocompress block failed"); - DEBUGLOG(4, "Block too small, writing out nocompress block: cSize: %zu", cBlockSize); - cSize += cBlockSize; - ip += blockSize; - op += cBlockSize; - remaining -= blockSize; - dstCapacity -= cBlockSize; - continue; - } - - compressedSeqsSize = ZSTD_entropyCompressSeqStore(&cctx->seqStore, - &cctx->blockState.prevCBlock->entropy, &cctx->blockState.nextCBlock->entropy, - &cctx->appliedParams, - op + ZSTD_blockHeaderSize /* Leave space for block header */, dstCapacity - ZSTD_blockHeaderSize, - blockSize, - cctx->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */, - cctx->bmi2); - FORWARD_IF_ERROR(compressedSeqsSize, "Compressing sequences of block failed"); - DEBUGLOG(4, "Compressed sequences size: %zu", compressedSeqsSize); - - if (!cctx->isFirstBlock && - ZSTD_maybeRLE(&cctx->seqStore) && - ZSTD_isRLE((BYTE const*)src, srcSize)) { - /* We don't want to emit our first block as a RLE even if it qualifies because - * doing so will cause the decoder (cli only) to throw a "should consume all input error." - * This is only an issue for zstd <= v1.4.3 - */ - compressedSeqsSize = 1; - } - - if (compressedSeqsSize == 0) { - /* ZSTD_noCompressBlock writes the block header as well */ - cBlockSize = ZSTD_noCompressBlock(op, dstCapacity, ip, blockSize, lastBlock); - FORWARD_IF_ERROR(cBlockSize, "Nocompress block failed"); - DEBUGLOG(4, "Writing out nocompress block, size: %zu", cBlockSize); - } else if (compressedSeqsSize == 1) { - cBlockSize = ZSTD_rleCompressBlock(op, dstCapacity, *ip, blockSize, lastBlock); - FORWARD_IF_ERROR(cBlockSize, "RLE compress block failed"); - DEBUGLOG(4, "Writing out RLE block, size: %zu", cBlockSize); - } else { - U32 cBlockHeader; - /* Error checking and repcodes update */ - ZSTD_blockState_confirmRepcodesAndEntropyTables(&cctx->blockState); - if (cctx->blockState.prevCBlock->entropy.fse.offcode_repeatMode == FSE_repeat_valid) - cctx->blockState.prevCBlock->entropy.fse.offcode_repeatMode = FSE_repeat_check; - - /* Write block header into beginning of block*/ - cBlockHeader = lastBlock + (((U32)bt_compressed)<<1) + (U32)(compressedSeqsSize << 3); - MEM_writeLE24(op, cBlockHeader); - cBlockSize = ZSTD_blockHeaderSize + compressedSeqsSize; - DEBUGLOG(4, "Writing out compressed block, size: %zu", cBlockSize); - } - - cSize += cBlockSize; - DEBUGLOG(4, "cSize running total: %zu", cSize); - - if (lastBlock) { - break; - } else { - ip += blockSize; - op += cBlockSize; - remaining -= blockSize; - dstCapacity -= cBlockSize; - cctx->isFirstBlock = 0; - } - } - - return cSize; -} - -size_t ZSTD_compressSequences(ZSTD_CCtx* const cctx, void* dst, size_t dstCapacity, - const ZSTD_Sequence* inSeqs, size_t inSeqsSize, + + DEBUGLOG(4, "ZSTD_compressSequences_internal srcSize: %zu, inSeqsSize: %zu", srcSize, inSeqsSize); + /* Special case: empty frame */ + if (remaining == 0) { + U32 const cBlockHeader24 = 1 /* last block */ + (((U32)bt_raw)<<1); + RETURN_ERROR_IF(dstCapacity<4, dstSize_tooSmall, "No room for empty frame block header"); + MEM_writeLE32(op, cBlockHeader24); + op += ZSTD_blockHeaderSize; + dstCapacity -= ZSTD_blockHeaderSize; + cSize += ZSTD_blockHeaderSize; + } + + while (remaining) { + size_t cBlockSize; + size_t additionalByteAdjustment; + lastBlock = remaining <= cctx->blockSize; + blockSize = lastBlock ? (U32)remaining : (U32)cctx->blockSize; + ZSTD_resetSeqStore(&cctx->seqStore); + DEBUGLOG(4, "Working on new block. Blocksize: %zu", blockSize); + + additionalByteAdjustment = sequenceCopier(cctx, &seqPos, inSeqs, inSeqsSize, ip, blockSize); + FORWARD_IF_ERROR(additionalByteAdjustment, "Bad sequence copy"); + blockSize -= additionalByteAdjustment; + + /* If blocks are too small, emit as a nocompress block */ + if (blockSize < MIN_CBLOCK_SIZE+ZSTD_blockHeaderSize+1) { + cBlockSize = ZSTD_noCompressBlock(op, dstCapacity, ip, blockSize, lastBlock); + FORWARD_IF_ERROR(cBlockSize, "Nocompress block failed"); + DEBUGLOG(4, "Block too small, writing out nocompress block: cSize: %zu", cBlockSize); + cSize += cBlockSize; + ip += blockSize; + op += cBlockSize; + remaining -= blockSize; + dstCapacity -= cBlockSize; + continue; + } + + compressedSeqsSize = ZSTD_entropyCompressSeqStore(&cctx->seqStore, + &cctx->blockState.prevCBlock->entropy, &cctx->blockState.nextCBlock->entropy, + &cctx->appliedParams, + op + ZSTD_blockHeaderSize /* Leave space for block header */, dstCapacity - ZSTD_blockHeaderSize, + blockSize, + cctx->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */, + cctx->bmi2); + FORWARD_IF_ERROR(compressedSeqsSize, "Compressing sequences of block failed"); + DEBUGLOG(4, "Compressed sequences size: %zu", compressedSeqsSize); + + if (!cctx->isFirstBlock && + ZSTD_maybeRLE(&cctx->seqStore) && + ZSTD_isRLE((BYTE const*)src, srcSize)) { + /* We don't want to emit our first block as a RLE even if it qualifies because + * doing so will cause the decoder (cli only) to throw a "should consume all input error." + * This is only an issue for zstd <= v1.4.3 + */ + compressedSeqsSize = 1; + } + + if (compressedSeqsSize == 0) { + /* ZSTD_noCompressBlock writes the block header as well */ + cBlockSize = ZSTD_noCompressBlock(op, dstCapacity, ip, blockSize, lastBlock); + FORWARD_IF_ERROR(cBlockSize, "Nocompress block failed"); + DEBUGLOG(4, "Writing out nocompress block, size: %zu", cBlockSize); + } else if (compressedSeqsSize == 1) { + cBlockSize = ZSTD_rleCompressBlock(op, dstCapacity, *ip, blockSize, lastBlock); + FORWARD_IF_ERROR(cBlockSize, "RLE compress block failed"); + DEBUGLOG(4, "Writing out RLE block, size: %zu", cBlockSize); + } else { + U32 cBlockHeader; + /* Error checking and repcodes update */ + ZSTD_blockState_confirmRepcodesAndEntropyTables(&cctx->blockState); + if (cctx->blockState.prevCBlock->entropy.fse.offcode_repeatMode == FSE_repeat_valid) + cctx->blockState.prevCBlock->entropy.fse.offcode_repeatMode = FSE_repeat_check; + + /* Write block header into beginning of block*/ + cBlockHeader = lastBlock + (((U32)bt_compressed)<<1) + (U32)(compressedSeqsSize << 3); + MEM_writeLE24(op, cBlockHeader); + cBlockSize = ZSTD_blockHeaderSize + compressedSeqsSize; + DEBUGLOG(4, "Writing out compressed block, size: %zu", cBlockSize); + } + + cSize += cBlockSize; + DEBUGLOG(4, "cSize running total: %zu", cSize); + + if (lastBlock) { + break; + } else { + ip += blockSize; + op += cBlockSize; + remaining -= blockSize; + dstCapacity -= cBlockSize; + cctx->isFirstBlock = 0; + } + } + + return cSize; +} + +size_t ZSTD_compressSequences(ZSTD_CCtx* const cctx, void* dst, size_t dstCapacity, + const ZSTD_Sequence* inSeqs, size_t inSeqsSize, const void* src, size_t srcSize) { - BYTE* op = (BYTE*)dst; - size_t cSize = 0; - size_t compressedBlocksSize = 0; - size_t frameHeaderSize = 0; - - /* Transparent initialization stage, same as compressStream2() */ - DEBUGLOG(3, "ZSTD_compressSequences()"); - assert(cctx != NULL); - FORWARD_IF_ERROR(ZSTD_CCtx_init_compressStream2(cctx, ZSTD_e_end, srcSize), "CCtx initialization failed"); - /* Begin writing output, starting with frame header */ - frameHeaderSize = ZSTD_writeFrameHeader(op, dstCapacity, &cctx->appliedParams, srcSize, cctx->dictID); - op += frameHeaderSize; - dstCapacity -= frameHeaderSize; - cSize += frameHeaderSize; - if (cctx->appliedParams.fParams.checksumFlag && srcSize) { - XXH64_update(&cctx->xxhState, src, srcSize); - } - /* cSize includes block header size and compressed sequences size */ - compressedBlocksSize = ZSTD_compressSequences_internal(cctx, - op, dstCapacity, - inSeqs, inSeqsSize, - src, srcSize); - FORWARD_IF_ERROR(compressedBlocksSize, "Compressing blocks failed!"); - cSize += compressedBlocksSize; - dstCapacity -= compressedBlocksSize; - - if (cctx->appliedParams.fParams.checksumFlag) { - U32 const checksum = (U32) XXH64_digest(&cctx->xxhState); - RETURN_ERROR_IF(dstCapacity<4, dstSize_tooSmall, "no room for checksum"); - DEBUGLOG(4, "Write checksum : %08X", (unsigned)checksum); - MEM_writeLE32((char*)dst + cSize, checksum); - cSize += 4; - } - - DEBUGLOG(3, "Final compressed size: %zu", cSize); - return cSize; -} - + BYTE* op = (BYTE*)dst; + size_t cSize = 0; + size_t compressedBlocksSize = 0; + size_t frameHeaderSize = 0; + + /* Transparent initialization stage, same as compressStream2() */ + DEBUGLOG(3, "ZSTD_compressSequences()"); + assert(cctx != NULL); + FORWARD_IF_ERROR(ZSTD_CCtx_init_compressStream2(cctx, ZSTD_e_end, srcSize), "CCtx initialization failed"); + /* Begin writing output, starting with frame header */ + frameHeaderSize = ZSTD_writeFrameHeader(op, dstCapacity, &cctx->appliedParams, srcSize, cctx->dictID); + op += frameHeaderSize; + dstCapacity -= frameHeaderSize; + cSize += frameHeaderSize; + if (cctx->appliedParams.fParams.checksumFlag && srcSize) { + XXH64_update(&cctx->xxhState, src, srcSize); + } + /* cSize includes block header size and compressed sequences size */ + compressedBlocksSize = ZSTD_compressSequences_internal(cctx, + op, dstCapacity, + inSeqs, inSeqsSize, + src, srcSize); + FORWARD_IF_ERROR(compressedBlocksSize, "Compressing blocks failed!"); + cSize += compressedBlocksSize; + dstCapacity -= compressedBlocksSize; + + if (cctx->appliedParams.fParams.checksumFlag) { + U32 const checksum = (U32) XXH64_digest(&cctx->xxhState); + RETURN_ERROR_IF(dstCapacity<4, dstSize_tooSmall, "no room for checksum"); + DEBUGLOG(4, "Write checksum : %08X", (unsigned)checksum); + MEM_writeLE32((char*)dst + cSize, checksum); + cSize += 4; + } + + DEBUGLOG(3, "Final compressed size: %zu", cSize); + return cSize; +} + /*====== Finalize ======*/ /*! ZSTD_flushStream() : @@ -6185,112 +6185,112 @@ size_t ZSTD_endStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output) int ZSTD_maxCLevel(void) { return ZSTD_MAX_CLEVEL; } int ZSTD_minCLevel(void) { return (int)-ZSTD_TARGETLENGTH_MAX; } -int ZSTD_defaultCLevel(void) { return ZSTD_CLEVEL_DEFAULT; } - -static ZSTD_compressionParameters ZSTD_dedicatedDictSearch_getCParams(int const compressionLevel, size_t const dictSize) -{ - ZSTD_compressionParameters cParams = ZSTD_getCParams_internal(compressionLevel, 0, dictSize, ZSTD_cpm_createCDict); - switch (cParams.strategy) { - case ZSTD_fast: - case ZSTD_dfast: - break; - case ZSTD_greedy: - case ZSTD_lazy: - case ZSTD_lazy2: - cParams.hashLog += ZSTD_LAZY_DDSS_BUCKET_LOG; - break; - case ZSTD_btlazy2: - case ZSTD_btopt: - case ZSTD_btultra: - case ZSTD_btultra2: - break; - } - return cParams; -} - -static int ZSTD_dedicatedDictSearch_isSupported( - ZSTD_compressionParameters const* cParams) -{ - return (cParams->strategy >= ZSTD_greedy) - && (cParams->strategy <= ZSTD_lazy2) - && (cParams->hashLog > cParams->chainLog) - && (cParams->chainLog <= 24); -} - -/** - * Reverses the adjustment applied to cparams when enabling dedicated dict - * search. This is used to recover the params set to be used in the working - * context. (Otherwise, those tables would also grow.) - */ -static void ZSTD_dedicatedDictSearch_revertCParams( - ZSTD_compressionParameters* cParams) { - switch (cParams->strategy) { - case ZSTD_fast: - case ZSTD_dfast: - break; - case ZSTD_greedy: - case ZSTD_lazy: - case ZSTD_lazy2: - cParams->hashLog -= ZSTD_LAZY_DDSS_BUCKET_LOG; - if (cParams->hashLog < ZSTD_HASHLOG_MIN) { - cParams->hashLog = ZSTD_HASHLOG_MIN; - } - break; - case ZSTD_btlazy2: - case ZSTD_btopt: - case ZSTD_btultra: - case ZSTD_btultra2: - break; - } -} - -static U64 ZSTD_getCParamRowSize(U64 srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode) -{ - switch (mode) { - case ZSTD_cpm_unknown: - case ZSTD_cpm_noAttachDict: - case ZSTD_cpm_createCDict: - break; - case ZSTD_cpm_attachDict: - dictSize = 0; - break; - default: - assert(0); - break; - } - { int const unknown = srcSizeHint == ZSTD_CONTENTSIZE_UNKNOWN; - size_t const addedSize = unknown && dictSize > 0 ? 500 : 0; - return unknown && dictSize == 0 ? ZSTD_CONTENTSIZE_UNKNOWN : srcSizeHint+dictSize+addedSize; - } -} - +int ZSTD_defaultCLevel(void) { return ZSTD_CLEVEL_DEFAULT; } + +static ZSTD_compressionParameters ZSTD_dedicatedDictSearch_getCParams(int const compressionLevel, size_t const dictSize) +{ + ZSTD_compressionParameters cParams = ZSTD_getCParams_internal(compressionLevel, 0, dictSize, ZSTD_cpm_createCDict); + switch (cParams.strategy) { + case ZSTD_fast: + case ZSTD_dfast: + break; + case ZSTD_greedy: + case ZSTD_lazy: + case ZSTD_lazy2: + cParams.hashLog += ZSTD_LAZY_DDSS_BUCKET_LOG; + break; + case ZSTD_btlazy2: + case ZSTD_btopt: + case ZSTD_btultra: + case ZSTD_btultra2: + break; + } + return cParams; +} + +static int ZSTD_dedicatedDictSearch_isSupported( + ZSTD_compressionParameters const* cParams) +{ + return (cParams->strategy >= ZSTD_greedy) + && (cParams->strategy <= ZSTD_lazy2) + && (cParams->hashLog > cParams->chainLog) + && (cParams->chainLog <= 24); +} + +/** + * Reverses the adjustment applied to cparams when enabling dedicated dict + * search. This is used to recover the params set to be used in the working + * context. (Otherwise, those tables would also grow.) + */ +static void ZSTD_dedicatedDictSearch_revertCParams( + ZSTD_compressionParameters* cParams) { + switch (cParams->strategy) { + case ZSTD_fast: + case ZSTD_dfast: + break; + case ZSTD_greedy: + case ZSTD_lazy: + case ZSTD_lazy2: + cParams->hashLog -= ZSTD_LAZY_DDSS_BUCKET_LOG; + if (cParams->hashLog < ZSTD_HASHLOG_MIN) { + cParams->hashLog = ZSTD_HASHLOG_MIN; + } + break; + case ZSTD_btlazy2: + case ZSTD_btopt: + case ZSTD_btultra: + case ZSTD_btultra2: + break; + } +} + +static U64 ZSTD_getCParamRowSize(U64 srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode) +{ + switch (mode) { + case ZSTD_cpm_unknown: + case ZSTD_cpm_noAttachDict: + case ZSTD_cpm_createCDict: + break; + case ZSTD_cpm_attachDict: + dictSize = 0; + break; + default: + assert(0); + break; + } + { int const unknown = srcSizeHint == ZSTD_CONTENTSIZE_UNKNOWN; + size_t const addedSize = unknown && dictSize > 0 ? 500 : 0; + return unknown && dictSize == 0 ? ZSTD_CONTENTSIZE_UNKNOWN : srcSizeHint+dictSize+addedSize; + } +} + /*! ZSTD_getCParams_internal() : * @return ZSTD_compressionParameters structure for a selected compression level, srcSize and dictSize. * Note: srcSizeHint 0 means 0, use ZSTD_CONTENTSIZE_UNKNOWN for unknown. - * Use dictSize == 0 for unknown or unused. - * Note: `mode` controls how we treat the `dictSize`. See docs for `ZSTD_cParamMode_e`. */ -static ZSTD_compressionParameters ZSTD_getCParams_internal(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode) + * Use dictSize == 0 for unknown or unused. + * Note: `mode` controls how we treat the `dictSize`. See docs for `ZSTD_cParamMode_e`. */ +static ZSTD_compressionParameters ZSTD_getCParams_internal(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode) { - U64 const rSize = ZSTD_getCParamRowSize(srcSizeHint, dictSize, mode); + U64 const rSize = ZSTD_getCParamRowSize(srcSizeHint, dictSize, mode); U32 const tableID = (rSize <= 256 KB) + (rSize <= 128 KB) + (rSize <= 16 KB); - int row; + int row; DEBUGLOG(5, "ZSTD_getCParams_internal (cLevel=%i)", compressionLevel); - - /* row */ + + /* row */ if (compressionLevel == 0) row = ZSTD_CLEVEL_DEFAULT; /* 0 == default */ - else if (compressionLevel < 0) row = 0; /* entry 0 is baseline for fast mode */ - else if (compressionLevel > ZSTD_MAX_CLEVEL) row = ZSTD_MAX_CLEVEL; - else row = compressionLevel; - + else if (compressionLevel < 0) row = 0; /* entry 0 is baseline for fast mode */ + else if (compressionLevel > ZSTD_MAX_CLEVEL) row = ZSTD_MAX_CLEVEL; + else row = compressionLevel; + { ZSTD_compressionParameters cp = ZSTD_defaultCParameters[tableID][row]; - DEBUGLOG(5, "ZSTD_getCParams_internal selected tableID: %u row: %u strat: %u", tableID, row, (U32)cp.strategy); - /* acceleration factor */ - if (compressionLevel < 0) { - int const clampedCompressionLevel = MAX(ZSTD_minCLevel(), compressionLevel); - cp.targetLength = (unsigned)(-clampedCompressionLevel); - } + DEBUGLOG(5, "ZSTD_getCParams_internal selected tableID: %u row: %u strat: %u", tableID, row, (U32)cp.strategy); + /* acceleration factor */ + if (compressionLevel < 0) { + int const clampedCompressionLevel = MAX(ZSTD_minCLevel(), compressionLevel); + cp.targetLength = (unsigned)(-clampedCompressionLevel); + } /* refine parameters based on srcSize & dictSize */ - return ZSTD_adjustCParams_internal(cp, srcSizeHint, dictSize, mode); + return ZSTD_adjustCParams_internal(cp, srcSizeHint, dictSize, mode); } } @@ -6300,18 +6300,18 @@ static ZSTD_compressionParameters ZSTD_getCParams_internal(int compressionLevel, ZSTD_compressionParameters ZSTD_getCParams(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize) { if (srcSizeHint == 0) srcSizeHint = ZSTD_CONTENTSIZE_UNKNOWN; - return ZSTD_getCParams_internal(compressionLevel, srcSizeHint, dictSize, ZSTD_cpm_unknown); + return ZSTD_getCParams_internal(compressionLevel, srcSizeHint, dictSize, ZSTD_cpm_unknown); } /*! ZSTD_getParams() : * same idea as ZSTD_getCParams() * @return a `ZSTD_parameters` structure (instead of `ZSTD_compressionParameters`). * Fields of `ZSTD_frameParameters` are set to default values */ -static ZSTD_parameters ZSTD_getParams_internal(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode) { +static ZSTD_parameters ZSTD_getParams_internal(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode) { ZSTD_parameters params; - ZSTD_compressionParameters const cParams = ZSTD_getCParams_internal(compressionLevel, srcSizeHint, dictSize, mode); + ZSTD_compressionParameters const cParams = ZSTD_getCParams_internal(compressionLevel, srcSizeHint, dictSize, mode); DEBUGLOG(5, "ZSTD_getParams (cLevel=%i)", compressionLevel); - ZSTD_memset(¶ms, 0, sizeof(params)); + ZSTD_memset(¶ms, 0, sizeof(params)); params.cParams = cParams; params.fParams.contentSizeFlag = 1; return params; @@ -6323,5 +6323,5 @@ static ZSTD_parameters ZSTD_getParams_internal(int compressionLevel, unsigned lo * Fields of `ZSTD_frameParameters` are set to default values */ ZSTD_parameters ZSTD_getParams(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize) { if (srcSizeHint == 0) srcSizeHint = ZSTD_CONTENTSIZE_UNKNOWN; - return ZSTD_getParams_internal(compressionLevel, srcSizeHint, dictSize, ZSTD_cpm_unknown); + return ZSTD_getParams_internal(compressionLevel, srcSizeHint, dictSize, ZSTD_cpm_unknown); } diff --git a/contrib/libs/zstd/lib/compress/zstd_compress_internal.h b/contrib/libs/zstd/lib/compress/zstd_compress_internal.h index c406e794bd..e51904afa4 100644 --- a/contrib/libs/zstd/lib/compress/zstd_compress_internal.h +++ b/contrib/libs/zstd/lib/compress/zstd_compress_internal.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -81,76 +81,76 @@ typedef struct { ZSTD_fseCTables_t fse; } ZSTD_entropyCTables_t; -/*********************************************** -* Entropy buffer statistics structs and funcs * -***********************************************/ -/** ZSTD_hufCTablesMetadata_t : - * Stores Literals Block Type for a super-block in hType, and - * huffman tree description in hufDesBuffer. - * hufDesSize refers to the size of huffman tree description in bytes. - * This metadata is populated in ZSTD_buildBlockEntropyStats_literals() */ -typedef struct { - symbolEncodingType_e hType; - BYTE hufDesBuffer[ZSTD_MAX_HUF_HEADER_SIZE]; - size_t hufDesSize; -} ZSTD_hufCTablesMetadata_t; - -/** ZSTD_fseCTablesMetadata_t : - * Stores symbol compression modes for a super-block in {ll, ol, ml}Type, and - * fse tables in fseTablesBuffer. - * fseTablesSize refers to the size of fse tables in bytes. - * This metadata is populated in ZSTD_buildBlockEntropyStats_sequences() */ -typedef struct { - symbolEncodingType_e llType; - symbolEncodingType_e ofType; - symbolEncodingType_e mlType; - BYTE fseTablesBuffer[ZSTD_MAX_FSE_HEADERS_SIZE]; - size_t fseTablesSize; - size_t lastCountSize; /* This is to account for bug in 1.3.4. More detail in ZSTD_entropyCompressSeqStore_internal() */ -} ZSTD_fseCTablesMetadata_t; - -typedef struct { - ZSTD_hufCTablesMetadata_t hufMetadata; - ZSTD_fseCTablesMetadata_t fseMetadata; -} ZSTD_entropyCTablesMetadata_t; - -/** ZSTD_buildBlockEntropyStats() : - * Builds entropy for the block. - * @return : 0 on success or error code */ -size_t ZSTD_buildBlockEntropyStats(seqStore_t* seqStorePtr, - const ZSTD_entropyCTables_t* prevEntropy, - ZSTD_entropyCTables_t* nextEntropy, - const ZSTD_CCtx_params* cctxParams, - ZSTD_entropyCTablesMetadata_t* entropyMetadata, - void* workspace, size_t wkspSize); - -/********************************* -* Compression internals structs * -*********************************/ - +/*********************************************** +* Entropy buffer statistics structs and funcs * +***********************************************/ +/** ZSTD_hufCTablesMetadata_t : + * Stores Literals Block Type for a super-block in hType, and + * huffman tree description in hufDesBuffer. + * hufDesSize refers to the size of huffman tree description in bytes. + * This metadata is populated in ZSTD_buildBlockEntropyStats_literals() */ typedef struct { + symbolEncodingType_e hType; + BYTE hufDesBuffer[ZSTD_MAX_HUF_HEADER_SIZE]; + size_t hufDesSize; +} ZSTD_hufCTablesMetadata_t; + +/** ZSTD_fseCTablesMetadata_t : + * Stores symbol compression modes for a super-block in {ll, ol, ml}Type, and + * fse tables in fseTablesBuffer. + * fseTablesSize refers to the size of fse tables in bytes. + * This metadata is populated in ZSTD_buildBlockEntropyStats_sequences() */ +typedef struct { + symbolEncodingType_e llType; + symbolEncodingType_e ofType; + symbolEncodingType_e mlType; + BYTE fseTablesBuffer[ZSTD_MAX_FSE_HEADERS_SIZE]; + size_t fseTablesSize; + size_t lastCountSize; /* This is to account for bug in 1.3.4. More detail in ZSTD_entropyCompressSeqStore_internal() */ +} ZSTD_fseCTablesMetadata_t; + +typedef struct { + ZSTD_hufCTablesMetadata_t hufMetadata; + ZSTD_fseCTablesMetadata_t fseMetadata; +} ZSTD_entropyCTablesMetadata_t; + +/** ZSTD_buildBlockEntropyStats() : + * Builds entropy for the block. + * @return : 0 on success or error code */ +size_t ZSTD_buildBlockEntropyStats(seqStore_t* seqStorePtr, + const ZSTD_entropyCTables_t* prevEntropy, + ZSTD_entropyCTables_t* nextEntropy, + const ZSTD_CCtx_params* cctxParams, + ZSTD_entropyCTablesMetadata_t* entropyMetadata, + void* workspace, size_t wkspSize); + +/********************************* +* Compression internals structs * +*********************************/ + +typedef struct { U32 off; /* Offset sumtype code for the match, using ZSTD_storeSeq() format */ - U32 len; /* Raw length of match */ + U32 len; /* Raw length of match */ } ZSTD_match_t; typedef struct { - U32 offset; /* Offset of sequence */ - U32 litLength; /* Length of literals prior to match */ - U32 matchLength; /* Raw length of match */ -} rawSeq; - -typedef struct { - rawSeq* seq; /* The start of the sequences */ - size_t pos; /* The index in seq where reading stopped. pos <= size. */ - size_t posInSequence; /* The position within the sequence at seq[pos] where reading - stopped. posInSequence <= seq[pos].litLength + seq[pos].matchLength */ - size_t size; /* The number of sequences. <= capacity. */ - size_t capacity; /* The capacity starting from `seq` pointer */ -} rawSeqStore_t; - -UNUSED_ATTR static const rawSeqStore_t kNullRawSeqStore = {NULL, 0, 0, 0, 0}; - -typedef struct { + U32 offset; /* Offset of sequence */ + U32 litLength; /* Length of literals prior to match */ + U32 matchLength; /* Raw length of match */ +} rawSeq; + +typedef struct { + rawSeq* seq; /* The start of the sequences */ + size_t pos; /* The index in seq where reading stopped. pos <= size. */ + size_t posInSequence; /* The position within the sequence at seq[pos] where reading + stopped. posInSequence <= seq[pos].litLength + seq[pos].matchLength */ + size_t size; /* The number of sequences. <= capacity. */ + size_t capacity; /* The capacity starting from `seq` pointer */ +} rawSeqStore_t; + +UNUSED_ATTR static const rawSeqStore_t kNullRawSeqStore = {NULL, 0, 0, 0, 0}; + +typedef struct { int price; U32 off; U32 mlen; @@ -188,23 +188,23 @@ typedef struct { } ZSTD_compressedBlockState_t; typedef struct { - BYTE const* nextSrc; /* next block here to continue on current prefix */ - BYTE const* base; /* All regular indexes relative to this position */ - BYTE const* dictBase; /* extDict indexes relative to this position */ - U32 dictLimit; /* below that point, need extDict */ - U32 lowLimit; /* below that point, no more valid data */ - U32 nbOverflowCorrections; /* Number of times overflow correction has run since - * ZSTD_window_init(). Useful for debugging coredumps - * and for ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY. - */ + BYTE const* nextSrc; /* next block here to continue on current prefix */ + BYTE const* base; /* All regular indexes relative to this position */ + BYTE const* dictBase; /* extDict indexes relative to this position */ + U32 dictLimit; /* below that point, need extDict */ + U32 lowLimit; /* below that point, no more valid data */ + U32 nbOverflowCorrections; /* Number of times overflow correction has run since + * ZSTD_window_init(). Useful for debugging coredumps + * and for ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY. + */ } ZSTD_window_t; #define ZSTD_WINDOW_START_INDEX 2 typedef struct ZSTD_matchState_t ZSTD_matchState_t; - -#define ZSTD_ROW_HASH_CACHE_SIZE 8 /* Size of prefetching hash cache for row-based matchfinder */ - + +#define ZSTD_ROW_HASH_CACHE_SIZE 8 /* Size of prefetching hash cache for row-based matchfinder */ + struct ZSTD_matchState_t { ZSTD_window_t window; /* State for window round buffer management */ U32 loadedDictEnd; /* index of end of dictionary, within context's referential. @@ -216,24 +216,24 @@ struct ZSTD_matchState_t { */ U32 nextToUpdate; /* index from which to continue table update */ U32 hashLog3; /* dispatch table for matches of len==3 : larger == faster, more memory */ - - U32 rowHashLog; /* For row-based matchfinder: Hashlog based on nb of rows in the hashTable.*/ - U16* tagTable; /* For row-based matchFinder: A row-based table containing the hashes and head index. */ - U32 hashCache[ZSTD_ROW_HASH_CACHE_SIZE]; /* For row-based matchFinder: a cache of hashes to improve speed */ - + + U32 rowHashLog; /* For row-based matchfinder: Hashlog based on nb of rows in the hashTable.*/ + U16* tagTable; /* For row-based matchFinder: A row-based table containing the hashes and head index. */ + U32 hashCache[ZSTD_ROW_HASH_CACHE_SIZE]; /* For row-based matchFinder: a cache of hashes to improve speed */ + U32* hashTable; U32* hashTable3; U32* chainTable; - - U32 forceNonContiguous; /* Non-zero if we should force non-contiguous load for the next window update. */ - - int dedicatedDictSearch; /* Indicates whether this matchState is using the - * dedicated dictionary search structure. - */ + + U32 forceNonContiguous; /* Non-zero if we should force non-contiguous load for the next window update. */ + + int dedicatedDictSearch; /* Indicates whether this matchState is using the + * dedicated dictionary search structure. + */ optState_t opt; /* optimal parser state */ const ZSTD_matchState_t* dictMatchState; ZSTD_compressionParameters cParams; - const rawSeqStore_t* ldmSeqStore; + const rawSeqStore_t* ldmSeqStore; }; typedef struct { @@ -248,21 +248,21 @@ typedef struct { } ldmEntry_t; typedef struct { - BYTE const* split; - U32 hash; - U32 checksum; - ldmEntry_t* bucket; -} ldmMatchCandidate_t; - -#define LDM_BATCH_SIZE 64 - -typedef struct { + BYTE const* split; + U32 hash; + U32 checksum; + ldmEntry_t* bucket; +} ldmMatchCandidate_t; + +#define LDM_BATCH_SIZE 64 + +typedef struct { ZSTD_window_t window; /* State for the window round buffer management */ ldmEntry_t* hashTable; U32 loadedDictEnd; BYTE* bucketOffsets; /* Next position in bucket to insert entry */ - size_t splitIndices[LDM_BATCH_SIZE]; - ldmMatchCandidate_t matchCandidates[LDM_BATCH_SIZE]; + size_t splitIndices[LDM_BATCH_SIZE]; + ldmMatchCandidate_t matchCandidates[LDM_BATCH_SIZE]; } ldmState_t; typedef struct { @@ -308,43 +308,43 @@ struct ZSTD_CCtx_params_s { /* Long distance matching parameters */ ldmParams_t ldmParams; - /* Dedicated dict search algorithm trigger */ - int enableDedicatedDictSearch; - - /* Input/output buffer modes */ - ZSTD_bufferMode_e inBufferMode; - ZSTD_bufferMode_e outBufferMode; - - /* Sequence compression API */ - ZSTD_sequenceFormat_e blockDelimiters; - int validateSequences; - - /* Block splitting */ + /* Dedicated dict search algorithm trigger */ + int enableDedicatedDictSearch; + + /* Input/output buffer modes */ + ZSTD_bufferMode_e inBufferMode; + ZSTD_bufferMode_e outBufferMode; + + /* Sequence compression API */ + ZSTD_sequenceFormat_e blockDelimiters; + int validateSequences; + + /* Block splitting */ ZSTD_paramSwitch_e useBlockSplitter; - - /* Param for deciding whether to use row-based matchfinder */ + + /* Param for deciding whether to use row-based matchfinder */ ZSTD_paramSwitch_e useRowMatchFinder; - - /* Always load a dictionary in ext-dict mode (not prefix mode)? */ - int deterministicRefPrefix; - + + /* Always load a dictionary in ext-dict mode (not prefix mode)? */ + int deterministicRefPrefix; + /* Internal use, for createCCtxParams() and freeCCtxParams() only */ ZSTD_customMem customMem; }; /* typedef'd to ZSTD_CCtx_params within "zstd.h" */ -#define COMPRESS_SEQUENCES_WORKSPACE_SIZE (sizeof(unsigned) * (MaxSeq + 2)) -#define ENTROPY_WORKSPACE_SIZE (HUF_WORKSPACE_SIZE + COMPRESS_SEQUENCES_WORKSPACE_SIZE) - -/** - * Indicates whether this compression proceeds directly from user-provided - * source buffer to user-provided destination buffer (ZSTDb_not_buffered), or - * whether the context needs to buffer the input/output (ZSTDb_buffered). - */ -typedef enum { - ZSTDb_not_buffered, - ZSTDb_buffered -} ZSTD_buffered_policy_e; - +#define COMPRESS_SEQUENCES_WORKSPACE_SIZE (sizeof(unsigned) * (MaxSeq + 2)) +#define ENTROPY_WORKSPACE_SIZE (HUF_WORKSPACE_SIZE + COMPRESS_SEQUENCES_WORKSPACE_SIZE) + +/** + * Indicates whether this compression proceeds directly from user-provided + * source buffer to user-provided destination buffer (ZSTDb_not_buffered), or + * whether the context needs to buffer the input/output (ZSTDb_buffered). + */ +typedef enum { + ZSTDb_not_buffered, + ZSTDb_buffered +} ZSTD_buffered_policy_e; + /** * Struct that contains all elements of block splitter that should be allocated * in a wksp. @@ -367,9 +367,9 @@ struct ZSTD_CCtx_s { int bmi2; /* == 1 if the CPU supports BMI2 and 0 otherwise. CPU support is determined dynamically once per context lifetime. */ ZSTD_CCtx_params requestedParams; ZSTD_CCtx_params appliedParams; - ZSTD_CCtx_params simpleApiParams; /* Param storage used by the simple API - not sticky. Must only be used in top-level simple API functions for storage. */ + ZSTD_CCtx_params simpleApiParams; /* Param storage used by the simple API - not sticky. Must only be used in top-level simple API functions for storage. */ U32 dictID; - size_t dictContentSize; + size_t dictContentSize; ZSTD_cwksp workspace; /* manages buffer for dynamic allocations */ size_t blockSize; @@ -378,7 +378,7 @@ struct ZSTD_CCtx_s { unsigned long long producedCSize; XXH64_state_t xxhState; ZSTD_customMem customMem; - ZSTD_threadPool* pool; + ZSTD_threadPool* pool; size_t staticSize; SeqCollector seqCollector; int isFirstBlock; @@ -390,11 +390,11 @@ struct ZSTD_CCtx_s { size_t maxNbLdmSequences; rawSeqStore_t externSeqStore; /* Mutable reference to external sequences */ ZSTD_blockState_t blockState; - U32* entropyWorkspace; /* entropy workspace of ENTROPY_WORKSPACE_SIZE bytes */ + U32* entropyWorkspace; /* entropy workspace of ENTROPY_WORKSPACE_SIZE bytes */ /* Whether we are streaming or not */ - ZSTD_buffered_policy_e bufferedPolicy; - + ZSTD_buffered_policy_e bufferedPolicy; + /* streaming */ char* inBuff; size_t inBuffSize; @@ -408,10 +408,10 @@ struct ZSTD_CCtx_s { ZSTD_cStreamStage streamStage; U32 frameEnded; - /* Stable in/out buffer verification */ - ZSTD_inBuffer expectedInBuffer; - size_t expectedOutBufferSize; - + /* Stable in/out buffer verification */ + ZSTD_inBuffer expectedInBuffer; + size_t expectedOutBufferSize; + /* Dictionary */ ZSTD_localDict localDict; const ZSTD_CDict* cdict; @@ -421,11 +421,11 @@ struct ZSTD_CCtx_s { #ifdef ZSTD_MULTITHREAD ZSTDMT_CCtx* mtctx; #endif - - /* Tracing */ -#if ZSTD_TRACE - ZSTD_TraceCtx traceCtx; -#endif + + /* Tracing */ +#if ZSTD_TRACE + ZSTD_TraceCtx traceCtx; +#endif /* Workspace for block splitter */ ZSTD_blockSplitCtx blockSplitCtx; @@ -433,32 +433,32 @@ struct ZSTD_CCtx_s { typedef enum { ZSTD_dtlm_fast, ZSTD_dtlm_full } ZSTD_dictTableLoadMethod_e; -typedef enum { - ZSTD_noDict = 0, - ZSTD_extDict = 1, - ZSTD_dictMatchState = 2, - ZSTD_dedicatedDictSearch = 3 -} ZSTD_dictMode_e; - -typedef enum { - ZSTD_cpm_noAttachDict = 0, /* Compression with ZSTD_noDict or ZSTD_extDict. - * In this mode we use both the srcSize and the dictSize - * when selecting and adjusting parameters. - */ - ZSTD_cpm_attachDict = 1, /* Compression with ZSTD_dictMatchState or ZSTD_dedicatedDictSearch. - * In this mode we only take the srcSize into account when selecting - * and adjusting parameters. - */ - ZSTD_cpm_createCDict = 2, /* Creating a CDict. - * In this mode we take both the source size and the dictionary size - * into account when selecting and adjusting the parameters. - */ - ZSTD_cpm_unknown = 3, /* ZSTD_getCParams, ZSTD_getParams, ZSTD_adjustParams. - * We don't know what these parameters are for. We default to the legacy - * behavior of taking both the source size and the dict size into account - * when selecting and adjusting parameters. - */ -} ZSTD_cParamMode_e; +typedef enum { + ZSTD_noDict = 0, + ZSTD_extDict = 1, + ZSTD_dictMatchState = 2, + ZSTD_dedicatedDictSearch = 3 +} ZSTD_dictMode_e; + +typedef enum { + ZSTD_cpm_noAttachDict = 0, /* Compression with ZSTD_noDict or ZSTD_extDict. + * In this mode we use both the srcSize and the dictSize + * when selecting and adjusting parameters. + */ + ZSTD_cpm_attachDict = 1, /* Compression with ZSTD_dictMatchState or ZSTD_dedicatedDictSearch. + * In this mode we only take the srcSize into account when selecting + * and adjusting parameters. + */ + ZSTD_cpm_createCDict = 2, /* Creating a CDict. + * In this mode we take both the source size and the dictionary size + * into account when selecting and adjusting the parameters. + */ + ZSTD_cpm_unknown = 3, /* ZSTD_getCParams, ZSTD_getParams, ZSTD_adjustParams. + * We don't know what these parameters are for. We default to the legacy + * behavior of taking both the source size and the dict size into account + * when selecting and adjusting parameters. + */ +} ZSTD_cParamMode_e; typedef size_t (*ZSTD_blockCompressor) ( ZSTD_matchState_t* bs, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], @@ -518,7 +518,7 @@ MEM_STATIC size_t ZSTD_noCompressBlock (void* dst, size_t dstCapacity, const voi RETURN_ERROR_IF(srcSize + ZSTD_blockHeaderSize > dstCapacity, dstSize_tooSmall, "dst buf too small for uncompressed block"); MEM_writeLE24(dst, cBlockHeader24); - ZSTD_memcpy((BYTE*)dst + ZSTD_blockHeaderSize, src, srcSize); + ZSTD_memcpy((BYTE*)dst + ZSTD_blockHeaderSize, src, srcSize); return ZSTD_blockHeaderSize + srcSize; } @@ -633,8 +633,8 @@ ZSTD_storeSeq(seqStore_t* seqStorePtr, /* literal Length */ if (litLength>0xFFFF) { - assert(seqStorePtr->longLengthType == ZSTD_llt_none); /* there can only be a single long length */ - seqStorePtr->longLengthType = ZSTD_llt_literalLength; + assert(seqStorePtr->longLengthType == ZSTD_llt_none); /* there can only be a single long length */ + seqStorePtr->longLengthType = ZSTD_llt_literalLength; seqStorePtr->longLengthPos = (U32)(seqStorePtr->sequences - seqStorePtr->sequencesStart); } seqStorePtr->sequences[0].litLength = (U16)litLength; @@ -702,9 +702,9 @@ static unsigned ZSTD_NbCommonBytes (size_t val) if (MEM_isLittleEndian()) { if (MEM_64bits()) { # if defined(_MSC_VER) && defined(_WIN64) -# if STATIC_BMI2 - return _tzcnt_u64(val) >> 3; -# else +# if STATIC_BMI2 + return _tzcnt_u64(val) >> 3; +# else if (val != 0) { unsigned long r; _BitScanForward64(&r, (U64)val); @@ -713,7 +713,7 @@ static unsigned ZSTD_NbCommonBytes (size_t val) /* Should not reach this code path */ __assume(0); } -# endif +# endif # elif defined(__GNUC__) && (__GNUC__ >= 4) return (__builtin_ctzll((U64)val) >> 3); # else @@ -750,9 +750,9 @@ static unsigned ZSTD_NbCommonBytes (size_t val) } else { /* Big Endian CPU */ if (MEM_64bits()) { # if defined(_MSC_VER) && defined(_WIN64) -# if STATIC_BMI2 - return _lzcnt_u64(val) >> 3; -# else +# if STATIC_BMI2 + return _lzcnt_u64(val) >> 3; +# else if (val != 0) { unsigned long r; _BitScanReverse64(&r, (U64)val); @@ -761,7 +761,7 @@ static unsigned ZSTD_NbCommonBytes (size_t val) /* Should not reach this code path */ __assume(0); } -# endif +# endif # elif defined(__GNUC__) && (__GNUC__ >= 4) return (__builtin_clzll(val) >> 3); # else @@ -862,8 +862,8 @@ static const U64 prime8bytes = 0xCF1BBCDCB7A56463ULL; static size_t ZSTD_hash8(U64 u, U32 h) { return (size_t)(((u) * prime8bytes) >> (64-h)) ; } static size_t ZSTD_hash8Ptr(const void* p, U32 h) { return ZSTD_hash8(MEM_readLE64(p), h); } -MEM_STATIC FORCE_INLINE_ATTR -size_t ZSTD_hashPtr(const void* p, U32 hBits, U32 mls) +MEM_STATIC FORCE_INLINE_ATTR +size_t ZSTD_hashPtr(const void* p, U32 hBits, U32 mls) { switch(mls) { @@ -960,13 +960,13 @@ MEM_STATIC void ZSTD_window_clear(ZSTD_window_t* window) window->dictLimit = end; } -MEM_STATIC U32 ZSTD_window_isEmpty(ZSTD_window_t const window) -{ +MEM_STATIC U32 ZSTD_window_isEmpty(ZSTD_window_t const window) +{ return window.dictLimit == ZSTD_WINDOW_START_INDEX && window.lowLimit == ZSTD_WINDOW_START_INDEX && (window.nextSrc - window.base) == ZSTD_WINDOW_START_INDEX; -} - +} + /** * ZSTD_window_hasExtDict(): * Returns non-zero if the window has a non-empty extDict. @@ -986,76 +986,76 @@ MEM_STATIC ZSTD_dictMode_e ZSTD_matchState_dictMode(const ZSTD_matchState_t *ms) return ZSTD_window_hasExtDict(ms->window) ? ZSTD_extDict : ms->dictMatchState != NULL ? - (ms->dictMatchState->dedicatedDictSearch ? ZSTD_dedicatedDictSearch : ZSTD_dictMatchState) : + (ms->dictMatchState->dedicatedDictSearch ? ZSTD_dedicatedDictSearch : ZSTD_dictMatchState) : ZSTD_noDict; } -/* Defining this macro to non-zero tells zstd to run the overflow correction - * code much more frequently. This is very inefficient, and should only be - * used for tests and fuzzers. - */ -#ifndef ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY -# ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION -# define ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY 1 -# else -# define ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY 0 -# endif -#endif - +/* Defining this macro to non-zero tells zstd to run the overflow correction + * code much more frequently. This is very inefficient, and should only be + * used for tests and fuzzers. + */ +#ifndef ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY +# ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION +# define ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY 1 +# else +# define ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY 0 +# endif +#endif + /** - * ZSTD_window_canOverflowCorrect(): - * Returns non-zero if the indices are large enough for overflow correction - * to work correctly without impacting compression ratio. - */ -MEM_STATIC U32 ZSTD_window_canOverflowCorrect(ZSTD_window_t const window, - U32 cycleLog, - U32 maxDist, - U32 loadedDictEnd, - void const* src) -{ - U32 const cycleSize = 1u << cycleLog; - U32 const curr = (U32)((BYTE const*)src - window.base); + * ZSTD_window_canOverflowCorrect(): + * Returns non-zero if the indices are large enough for overflow correction + * to work correctly without impacting compression ratio. + */ +MEM_STATIC U32 ZSTD_window_canOverflowCorrect(ZSTD_window_t const window, + U32 cycleLog, + U32 maxDist, + U32 loadedDictEnd, + void const* src) +{ + U32 const cycleSize = 1u << cycleLog; + U32 const curr = (U32)((BYTE const*)src - window.base); U32 const minIndexToOverflowCorrect = cycleSize + MAX(maxDist, cycleSize) + ZSTD_WINDOW_START_INDEX; - - /* Adjust the min index to backoff the overflow correction frequency, - * so we don't waste too much CPU in overflow correction. If this - * computation overflows we don't really care, we just need to make - * sure it is at least minIndexToOverflowCorrect. - */ - U32 const adjustment = window.nbOverflowCorrections + 1; - U32 const adjustedIndex = MAX(minIndexToOverflowCorrect * adjustment, - minIndexToOverflowCorrect); - U32 const indexLargeEnough = curr > adjustedIndex; - - /* Only overflow correct early if the dictionary is invalidated already, - * so we don't hurt compression ratio. - */ - U32 const dictionaryInvalidated = curr > maxDist + loadedDictEnd; - - return indexLargeEnough && dictionaryInvalidated; -} - -/** + + /* Adjust the min index to backoff the overflow correction frequency, + * so we don't waste too much CPU in overflow correction. If this + * computation overflows we don't really care, we just need to make + * sure it is at least minIndexToOverflowCorrect. + */ + U32 const adjustment = window.nbOverflowCorrections + 1; + U32 const adjustedIndex = MAX(minIndexToOverflowCorrect * adjustment, + minIndexToOverflowCorrect); + U32 const indexLargeEnough = curr > adjustedIndex; + + /* Only overflow correct early if the dictionary is invalidated already, + * so we don't hurt compression ratio. + */ + U32 const dictionaryInvalidated = curr > maxDist + loadedDictEnd; + + return indexLargeEnough && dictionaryInvalidated; +} + +/** * ZSTD_window_needOverflowCorrection(): * Returns non-zero if the indices are getting too large and need overflow * protection. */ MEM_STATIC U32 ZSTD_window_needOverflowCorrection(ZSTD_window_t const window, - U32 cycleLog, - U32 maxDist, - U32 loadedDictEnd, - void const* src, + U32 cycleLog, + U32 maxDist, + U32 loadedDictEnd, + void const* src, void const* srcEnd) { - U32 const curr = (U32)((BYTE const*)srcEnd - window.base); - if (ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY) { - if (ZSTD_window_canOverflowCorrect(window, cycleLog, maxDist, loadedDictEnd, src)) { - return 1; - } - } - return curr > ZSTD_CURRENT_MAX; + U32 const curr = (U32)((BYTE const*)srcEnd - window.base); + if (ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY) { + if (ZSTD_window_canOverflowCorrect(window, cycleLog, maxDist, loadedDictEnd, src)) { + return 1; + } + } + return curr > ZSTD_CURRENT_MAX; } /** @@ -1089,9 +1089,9 @@ MEM_STATIC U32 ZSTD_window_correctOverflow(ZSTD_window_t* window, U32 cycleLog, * 3. (cctx->lowLimit + 1<<windowLog) < 1<<32: * windowLog <= 31 ==> 3<<29 + 1<<windowLog < 7<<29 < 1<<32. */ - U32 const cycleSize = 1u << cycleLog; - U32 const cycleMask = cycleSize - 1; - U32 const curr = (U32)((BYTE const*)src - window->base); + U32 const cycleSize = 1u << cycleLog; + U32 const cycleMask = cycleSize - 1; + U32 const curr = (U32)((BYTE const*)src - window->base); U32 const currentCycle = curr & cycleMask; /* Ensure newCurrent - maxDist >= ZSTD_WINDOW_START_INDEX. */ U32 const currentCycleCorrection = currentCycle < ZSTD_WINDOW_START_INDEX @@ -1100,18 +1100,18 @@ MEM_STATIC U32 ZSTD_window_correctOverflow(ZSTD_window_t* window, U32 cycleLog, U32 const newCurrent = currentCycle + currentCycleCorrection + MAX(maxDist, cycleSize); - U32 const correction = curr - newCurrent; - /* maxDist must be a power of two so that: - * (newCurrent & cycleMask) == (curr & cycleMask) - * This is required to not corrupt the chains / binary tree. - */ - assert((maxDist & (maxDist - 1)) == 0); - assert((curr & cycleMask) == (newCurrent & cycleMask)); - assert(curr > newCurrent); - if (!ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY) { - /* Loose bound, should be around 1<<29 (see above) */ - assert(correction > 1<<28); - } + U32 const correction = curr - newCurrent; + /* maxDist must be a power of two so that: + * (newCurrent & cycleMask) == (curr & cycleMask) + * This is required to not corrupt the chains / binary tree. + */ + assert((maxDist & (maxDist - 1)) == 0); + assert((curr & cycleMask) == (newCurrent & cycleMask)); + assert(curr > newCurrent); + if (!ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY) { + /* Loose bound, should be around 1<<29 (see above) */ + assert(correction > 1<<28); + } window->base += correction; window->dictBase += correction; @@ -1133,8 +1133,8 @@ MEM_STATIC U32 ZSTD_window_correctOverflow(ZSTD_window_t* window, U32 cycleLog, assert(window->lowLimit <= newCurrent); assert(window->dictLimit <= newCurrent); - ++window->nbOverflowCorrections; - + ++window->nbOverflowCorrections; + DEBUGLOG(4, "Correction of 0x%x bytes to lowLimit=0x%x", correction, window->lowLimit); return correction; @@ -1238,14 +1238,14 @@ ZSTD_checkDictValidity(const ZSTD_window_t* window, } MEM_STATIC void ZSTD_window_init(ZSTD_window_t* window) { - ZSTD_memset(window, 0, sizeof(*window)); + ZSTD_memset(window, 0, sizeof(*window)); window->base = (BYTE const*)" "; window->dictBase = (BYTE const*)" "; ZSTD_STATIC_ASSERT(ZSTD_DUBT_UNSORTED_MARK < ZSTD_WINDOW_START_INDEX); /* Start above ZSTD_DUBT_UNSORTED_MARK */ window->dictLimit = ZSTD_WINDOW_START_INDEX; /* start from >0, so that 1st position is valid */ window->lowLimit = ZSTD_WINDOW_START_INDEX; /* it ensures first and later CCtx usages compress the same */ window->nextSrc = window->base + ZSTD_WINDOW_START_INDEX; /* see issue #1241 */ - window->nbOverflowCorrections = 0; + window->nbOverflowCorrections = 0; } /** @@ -1256,8 +1256,8 @@ MEM_STATIC void ZSTD_window_init(ZSTD_window_t* window) { * Returns non-zero if the segment is contiguous. */ MEM_STATIC U32 ZSTD_window_update(ZSTD_window_t* window, - void const* src, size_t srcSize, - int forceNonContiguous) + void const* src, size_t srcSize, + int forceNonContiguous) { BYTE const* const ip = (BYTE const*)src; U32 contiguous = 1; @@ -1267,7 +1267,7 @@ MEM_STATIC U32 ZSTD_window_update(ZSTD_window_t* window, assert(window->base != NULL); assert(window->dictBase != NULL); /* Check if blocks follow each other */ - if (src != window->nextSrc || forceNonContiguous) { + if (src != window->nextSrc || forceNonContiguous) { /* not contiguous */ size_t const distanceFromBase = (size_t)(window->nextSrc - window->base); DEBUGLOG(5, "Non contiguous blocks, new segment starts at %u", window->dictLimit); @@ -1295,16 +1295,16 @@ MEM_STATIC U32 ZSTD_window_update(ZSTD_window_t* window, /** * Returns the lowest allowed match index. It may either be in the ext-dict or the prefix. */ -MEM_STATIC U32 ZSTD_getLowestMatchIndex(const ZSTD_matchState_t* ms, U32 curr, unsigned windowLog) +MEM_STATIC U32 ZSTD_getLowestMatchIndex(const ZSTD_matchState_t* ms, U32 curr, unsigned windowLog) { U32 const maxDistance = 1U << windowLog; U32 const lowestValid = ms->window.lowLimit; U32 const withinWindow = (curr - lowestValid > maxDistance) ? curr - maxDistance : lowestValid; U32 const isDictionary = (ms->loadedDictEnd != 0); - /* When using a dictionary the entire dictionary is valid if a single byte of the dictionary - * is within the window. We invalidate the dictionary (and set loadedDictEnd to 0) when it isn't - * valid for the entire block. So this check is sufficient to find the lowest valid match index. - */ + /* When using a dictionary the entire dictionary is valid if a single byte of the dictionary + * is within the window. We invalidate the dictionary (and set loadedDictEnd to 0) when it isn't + * valid for the entire block. So this check is sufficient to find the lowest valid match index. + */ U32 const matchLowest = isDictionary ? lowestValid : withinWindow; return matchLowest; } @@ -1312,15 +1312,15 @@ MEM_STATIC U32 ZSTD_getLowestMatchIndex(const ZSTD_matchState_t* ms, U32 curr, u /** * Returns the lowest allowed match index in the prefix. */ -MEM_STATIC U32 ZSTD_getLowestPrefixIndex(const ZSTD_matchState_t* ms, U32 curr, unsigned windowLog) +MEM_STATIC U32 ZSTD_getLowestPrefixIndex(const ZSTD_matchState_t* ms, U32 curr, unsigned windowLog) { U32 const maxDistance = 1U << windowLog; U32 const lowestValid = ms->window.dictLimit; - U32 const withinWindow = (curr - lowestValid > maxDistance) ? curr - maxDistance : lowestValid; + U32 const withinWindow = (curr - lowestValid > maxDistance) ? curr - maxDistance : lowestValid; U32 const isDictionary = (ms->loadedDictEnd != 0); - /* When computing the lowest prefix index we need to take the dictionary into account to handle - * the edge case where the dictionary and the source are contiguous in memory. - */ + /* When computing the lowest prefix index we need to take the dictionary into account to handle + * the edge case where the dictionary and the source are contiguous in memory. + */ U32 const matchLowest = isDictionary ? lowestValid : withinWindow; return matchLowest; } @@ -1389,7 +1389,7 @@ void ZSTD_reset_compressedBlockState(ZSTD_compressedBlockState_t* bs); * Note: srcSizeHint == 0 means 0! */ ZSTD_compressionParameters ZSTD_getCParamsFromCCtxParams( - const ZSTD_CCtx_params* CCtxParams, U64 srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode); + const ZSTD_CCtx_params* CCtxParams, U64 srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode); /*! ZSTD_initCStream_internal() : * Private use only. Init streaming operation. @@ -1450,9 +1450,9 @@ size_t ZSTD_referenceExternalSequences(ZSTD_CCtx* cctx, rawSeq* seq, size_t nbSe * condition for correct operation : hashLog > 1 */ U32 ZSTD_cycleLog(U32 hashLog, ZSTD_strategy strat); -/** ZSTD_CCtx_trace() : - * Trace the end of a compression call. - */ -void ZSTD_CCtx_trace(ZSTD_CCtx* cctx, size_t extraCSize); - +/** ZSTD_CCtx_trace() : + * Trace the end of a compression call. + */ +void ZSTD_CCtx_trace(ZSTD_CCtx* cctx, size_t extraCSize); + #endif /* ZSTD_COMPRESS_H */ diff --git a/contrib/libs/zstd/lib/compress/zstd_compress_literals.c b/contrib/libs/zstd/lib/compress/zstd_compress_literals.c index 52b0a8059a..325ce0e97e 100644 --- a/contrib/libs/zstd/lib/compress/zstd_compress_literals.c +++ b/contrib/libs/zstd/lib/compress/zstd_compress_literals.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -15,7 +15,7 @@ size_t ZSTD_noCompressLiterals (void* dst, size_t dstCapacity, const void* src, size_t srcSize) { - BYTE* const ostart = (BYTE*)dst; + BYTE* const ostart = (BYTE*)dst; U32 const flSize = 1 + (srcSize>31) + (srcSize>4095); RETURN_ERROR_IF(srcSize + flSize > dstCapacity, dstSize_tooSmall, ""); @@ -35,14 +35,14 @@ size_t ZSTD_noCompressLiterals (void* dst, size_t dstCapacity, const void* src, assert(0); } - ZSTD_memcpy(ostart + flSize, src, srcSize); + ZSTD_memcpy(ostart + flSize, src, srcSize); DEBUGLOG(5, "Raw literals: %u -> %u", (U32)srcSize, (U32)(srcSize + flSize)); return srcSize + flSize; } size_t ZSTD_compressRleLiteralsBlock (void* dst, size_t dstCapacity, const void* src, size_t srcSize) { - BYTE* const ostart = (BYTE*)dst; + BYTE* const ostart = (BYTE*)dst; U32 const flSize = 1 + (srcSize>31) + (srcSize>4095); (void)dstCapacity; /* dstCapacity already guaranteed to be >=4, hence large enough */ @@ -87,7 +87,7 @@ size_t ZSTD_compressLiterals (ZSTD_hufCTables_t const* prevHuf, disableLiteralCompression, (U32)srcSize); /* Prepare nextEntropy assuming reusing the existing table */ - ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); + ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); if (disableLiteralCompression) return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize); @@ -118,12 +118,12 @@ size_t ZSTD_compressLiterals (ZSTD_hufCTables_t const* prevHuf, } } - if ((cLitSize==0) || (cLitSize >= srcSize - minGain) || ERR_isError(cLitSize)) { - ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); + if ((cLitSize==0) || (cLitSize >= srcSize - minGain) || ERR_isError(cLitSize)) { + ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize); } if (cLitSize==1) { - ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); + ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); return ZSTD_compressRleLiteralsBlock(dst, dstCapacity, src, srcSize); } diff --git a/contrib/libs/zstd/lib/compress/zstd_compress_literals.h b/contrib/libs/zstd/lib/compress/zstd_compress_literals.h index 9775fb97cb..e03c47747e 100644 --- a/contrib/libs/zstd/lib/compress/zstd_compress_literals.h +++ b/contrib/libs/zstd/lib/compress/zstd_compress_literals.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the diff --git a/contrib/libs/zstd/lib/compress/zstd_compress_sequences.c b/contrib/libs/zstd/lib/compress/zstd_compress_sequences.c index f1e40af2ea..afb492fa54 100644 --- a/contrib/libs/zstd/lib/compress/zstd_compress_sequences.c +++ b/contrib/libs/zstd/lib/compress/zstd_compress_sequences.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -51,19 +51,19 @@ static unsigned ZSTD_getFSEMaxSymbolValue(FSE_CTable const* ctable) { } /** - * Returns true if we should use ncount=-1 else we should - * use ncount=1 for low probability symbols instead. - */ -static unsigned ZSTD_useLowProbCount(size_t const nbSeq) -{ - /* Heuristic: This should cover most blocks <= 16K and - * start to fade out after 16K to about 32K depending on - * comprssibility. - */ - return nbSeq >= 2048; -} - -/** + * Returns true if we should use ncount=-1 else we should + * use ncount=1 for low probability symbols instead. + */ +static unsigned ZSTD_useLowProbCount(size_t const nbSeq) +{ + /* Heuristic: This should cover most blocks <= 16K and + * start to fade out after 16K to about 32K depending on + * comprssibility. + */ + return nbSeq >= 2048; +} + +/** * Returns the cost in bytes of encoding the normalized count header. * Returns an error if any of the helper functions return an error. */ @@ -73,7 +73,7 @@ static size_t ZSTD_NCountCost(unsigned const* count, unsigned const max, BYTE wksp[FSE_NCOUNTBOUND]; S16 norm[MaxSeq + 1]; const U32 tableLog = FSE_optimalTableLog(FSELog, nbSeq, max); - FORWARD_IF_ERROR(FSE_normalizeCount(norm, tableLog, count, nbSeq, max, ZSTD_useLowProbCount(nbSeq)), ""); + FORWARD_IF_ERROR(FSE_normalizeCount(norm, tableLog, count, nbSeq, max, ZSTD_useLowProbCount(nbSeq)), ""); return FSE_writeNCount(wksp, sizeof(wksp), norm, max, tableLog); } @@ -85,8 +85,8 @@ static size_t ZSTD_entropyCost(unsigned const* count, unsigned const max, size_t { unsigned cost = 0; unsigned s; - - assert(total > 0); + + assert(total > 0); for (s = 0; s <= max; ++s) { unsigned norm = (unsigned)((256 * count[s]) / total); if (count[s] != 0 && norm == 0) @@ -234,11 +234,11 @@ ZSTD_selectEncodingType( return set_compressed; } -typedef struct { - S16 norm[MaxSeq + 1]; - U32 wksp[FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(MaxSeq, MaxFSELog)]; -} ZSTD_BuildCTableWksp; - +typedef struct { + S16 norm[MaxSeq + 1]; + U32 wksp[FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(MaxSeq, MaxFSELog)]; +} ZSTD_BuildCTableWksp; + size_t ZSTD_buildCTable(void* dst, size_t dstCapacity, FSE_CTable* nextCTable, U32 FSELog, symbolEncodingType_e type, @@ -259,13 +259,13 @@ ZSTD_buildCTable(void* dst, size_t dstCapacity, *op = codeTable[0]; return 1; case set_repeat: - ZSTD_memcpy(nextCTable, prevCTable, prevCTableSize); + ZSTD_memcpy(nextCTable, prevCTable, prevCTableSize); return 0; case set_basic: FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, defaultNorm, defaultMax, defaultNormLog, entropyWorkspace, entropyWorkspaceSize), ""); /* note : could be pre-calculated */ return 0; case set_compressed: { - ZSTD_BuildCTableWksp* wksp = (ZSTD_BuildCTableWksp*)entropyWorkspace; + ZSTD_BuildCTableWksp* wksp = (ZSTD_BuildCTableWksp*)entropyWorkspace; size_t nbSeq_1 = nbSeq; const U32 tableLog = FSE_optimalTableLog(FSELog, nbSeq, max); if (count[codeTable[nbSeq-1]] > 1) { @@ -273,8 +273,8 @@ ZSTD_buildCTable(void* dst, size_t dstCapacity, nbSeq_1--; } assert(nbSeq_1 > 1); - assert(entropyWorkspaceSize >= sizeof(ZSTD_BuildCTableWksp)); - (void)entropyWorkspaceSize; + assert(entropyWorkspaceSize >= sizeof(ZSTD_BuildCTableWksp)); + (void)entropyWorkspaceSize; FORWARD_IF_ERROR(FSE_normalizeCount(wksp->norm, tableLog, count, nbSeq_1, max, ZSTD_useLowProbCount(nbSeq_1)), "FSE_normalizeCount failed"); assert(oend >= op); { size_t const NCountSize = FSE_writeNCount(op, (size_t)(oend - op), wksp->norm, max, tableLog); /* overflow protected */ diff --git a/contrib/libs/zstd/lib/compress/zstd_compress_sequences.h b/contrib/libs/zstd/lib/compress/zstd_compress_sequences.h index 7991364c2f..857f84b541 100644 --- a/contrib/libs/zstd/lib/compress/zstd_compress_sequences.h +++ b/contrib/libs/zstd/lib/compress/zstd_compress_sequences.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the diff --git a/contrib/libs/zstd/lib/compress/zstd_compress_superblock.c b/contrib/libs/zstd/lib/compress/zstd_compress_superblock.c index 10e3378577..f2e7b6ed0e 100644 --- a/contrib/libs/zstd/lib/compress/zstd_compress_superblock.c +++ b/contrib/libs/zstd/lib/compress/zstd_compress_superblock.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -15,7 +15,7 @@ #include "../common/zstd_internal.h" /* ZSTD_getSequenceLength */ #include "hist.h" /* HIST_countFast_wksp */ -#include "zstd_compress_internal.h" /* ZSTD_[huf|fse|entropy]CTablesMetadata_t */ +#include "zstd_compress_internal.h" /* ZSTD_[huf|fse|entropy]CTablesMetadata_t */ #include "zstd_compress_sequences.h" #include "zstd_compress_literals.h" @@ -26,7 +26,7 @@ * before we know the table size + compressed size, so we have a bound on the * table size. If we guessed incorrectly, we fall back to uncompressed literals. * - * We write the header when writeEntropy=1 and set entropyWritten=1 when we succeeded + * We write the header when writeEntropy=1 and set entropyWritten=1 when we succeeded * in writing the header, otherwise it is set to 0. * * hufMetadata->hType has literals block type info. @@ -70,7 +70,7 @@ static size_t ZSTD_compressSubBlock_literal(const HUF_CElt* hufTable, assert(hufMetadata->hType == set_compressed || hufMetadata->hType == set_repeat); if (writeEntropy && hufMetadata->hType == set_compressed) { - ZSTD_memcpy(op, hufMetadata->hufDesBuffer, hufMetadata->hufDesSize); + ZSTD_memcpy(op, hufMetadata->hufDesBuffer, hufMetadata->hufDesSize); op += hufMetadata->hufDesSize; cLitSize += hufMetadata->hufDesSize; DEBUGLOG(5, "ZSTD_compressSubBlock_literal (hSize=%zu)", hufMetadata->hufDesSize); @@ -197,7 +197,7 @@ static size_t ZSTD_compressSubBlock_sequences(const ZSTD_fseCTables_t* fseTables const U32 MLtype = fseMetadata->mlType; DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (fseTablesSize=%zu)", fseMetadata->fseTablesSize); *seqHead = (BYTE)((LLtype<<6) + (Offtype<<4) + (MLtype<<2)); - ZSTD_memcpy(op, fseMetadata->fseTablesBuffer, fseMetadata->fseTablesSize); + ZSTD_memcpy(op, fseMetadata->fseTablesBuffer, fseMetadata->fseTablesSize); op += fseMetadata->fseTablesSize; } else { const U32 repeat = set_repeat; @@ -326,7 +326,7 @@ static size_t ZSTD_estimateSubBlockSize_symbolType(symbolEncodingType_e type, const BYTE* codeTable, unsigned maxCode, size_t nbSeq, const FSE_CTable* fseCTable, const U8* additionalBits, - short const* defaultNorm, U32 defaultNormLog, U32 defaultMax, + short const* defaultNorm, U32 defaultNormLog, U32 defaultMax, void* workspace, size_t wkspSize) { unsigned* const countWksp = (unsigned*)workspace; @@ -338,11 +338,11 @@ static size_t ZSTD_estimateSubBlockSize_symbolType(symbolEncodingType_e type, HIST_countFast_wksp(countWksp, &max, codeTable, nbSeq, workspace, wkspSize); /* can't fail */ if (type == set_basic) { - /* We selected this encoding type, so it must be valid. */ - assert(max <= defaultMax); - cSymbolTypeSizeEstimateInBits = max <= defaultMax - ? ZSTD_crossEntropyCost(defaultNorm, defaultNormLog, countWksp, max) - : ERROR(GENERIC); + /* We selected this encoding type, so it must be valid. */ + assert(max <= defaultMax); + cSymbolTypeSizeEstimateInBits = max <= defaultMax + ? ZSTD_crossEntropyCost(defaultNorm, defaultNormLog, countWksp, max) + : ERROR(GENERIC); } else if (type == set_rle) { cSymbolTypeSizeEstimateInBits = 0; } else if (type == set_compressed || type == set_repeat) { @@ -366,20 +366,20 @@ static size_t ZSTD_estimateSubBlockSize_sequences(const BYTE* ofCodeTable, void* workspace, size_t wkspSize, int writeEntropy) { - size_t const sequencesSectionHeaderSize = 3; /* Use hard coded size of 3 bytes */ + size_t const sequencesSectionHeaderSize = 3; /* Use hard coded size of 3 bytes */ size_t cSeqSizeEstimate = 0; - if (nbSeq == 0) return sequencesSectionHeaderSize; + if (nbSeq == 0) return sequencesSectionHeaderSize; cSeqSizeEstimate += ZSTD_estimateSubBlockSize_symbolType(fseMetadata->ofType, ofCodeTable, MaxOff, nbSeq, fseTables->offcodeCTable, NULL, - OF_defaultNorm, OF_defaultNormLog, DefaultMaxOff, + OF_defaultNorm, OF_defaultNormLog, DefaultMaxOff, workspace, wkspSize); cSeqSizeEstimate += ZSTD_estimateSubBlockSize_symbolType(fseMetadata->llType, llCodeTable, MaxLL, nbSeq, fseTables->litlengthCTable, LL_bits, - LL_defaultNorm, LL_defaultNormLog, MaxLL, + LL_defaultNorm, LL_defaultNormLog, MaxLL, workspace, wkspSize); cSeqSizeEstimate += ZSTD_estimateSubBlockSize_symbolType(fseMetadata->mlType, mlCodeTable, MaxML, nbSeq, fseTables->matchlengthCTable, ML_bits, - ML_defaultNorm, ML_defaultNormLog, MaxML, + ML_defaultNorm, ML_defaultNormLog, MaxML, workspace, wkspSize); if (writeEntropy) cSeqSizeEstimate += fseMetadata->fseTablesSize; return cSeqSizeEstimate + sequencesSectionHeaderSize; @@ -518,7 +518,7 @@ static size_t ZSTD_compressSubBlock_multi(const seqStore_t* seqStorePtr, } while (!lastSequence); if (writeLitEntropy) { DEBUGLOG(5, "ZSTD_compressSubBlock_multi has literal entropy tables unwritten"); - ZSTD_memcpy(&nextCBlock->entropy.huf, &prevCBlock->entropy.huf, sizeof(prevCBlock->entropy.huf)); + ZSTD_memcpy(&nextCBlock->entropy.huf, &prevCBlock->entropy.huf, sizeof(prevCBlock->entropy.huf)); } if (writeSeqEntropy && ZSTD_needSequenceEntropyTables(&entropyMetadata->fseMetadata)) { /* If we haven't written our entropy tables, then we've violated our contract and @@ -537,11 +537,11 @@ static size_t ZSTD_compressSubBlock_multi(const seqStore_t* seqStorePtr, if (sp < send) { seqDef const* seq; repcodes_t rep; - ZSTD_memcpy(&rep, prevCBlock->rep, sizeof(rep)); + ZSTD_memcpy(&rep, prevCBlock->rep, sizeof(rep)); for (seq = sstart; seq < sp; ++seq) { ZSTD_updateRep(rep.rep, seq->offBase - 1, ZSTD_getSequenceLength(seqStorePtr, seq).litLength == 0); } - ZSTD_memcpy(nextCBlock->rep, &rep, sizeof(rep)); + ZSTD_memcpy(nextCBlock->rep, &rep, sizeof(rep)); } } DEBUGLOG(5, "ZSTD_compressSubBlock_multi compressed"); @@ -554,12 +554,12 @@ size_t ZSTD_compressSuperBlock(ZSTD_CCtx* zc, unsigned lastBlock) { ZSTD_entropyCTablesMetadata_t entropyMetadata; - FORWARD_IF_ERROR(ZSTD_buildBlockEntropyStats(&zc->seqStore, + FORWARD_IF_ERROR(ZSTD_buildBlockEntropyStats(&zc->seqStore, &zc->blockState.prevCBlock->entropy, &zc->blockState.nextCBlock->entropy, &zc->appliedParams, &entropyMetadata, - zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */), ""); + zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */), ""); return ZSTD_compressSubBlock_multi(&zc->seqStore, zc->blockState.prevCBlock, @@ -569,5 +569,5 @@ size_t ZSTD_compressSuperBlock(ZSTD_CCtx* zc, dst, dstCapacity, src, srcSize, zc->bmi2, lastBlock, - zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */); + zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */); } diff --git a/contrib/libs/zstd/lib/compress/zstd_compress_superblock.h b/contrib/libs/zstd/lib/compress/zstd_compress_superblock.h index 176f9b106f..2d8de008ba 100644 --- a/contrib/libs/zstd/lib/compress/zstd_compress_superblock.h +++ b/contrib/libs/zstd/lib/compress/zstd_compress_superblock.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the diff --git a/contrib/libs/zstd/lib/compress/zstd_cwksp.h b/contrib/libs/zstd/lib/compress/zstd_cwksp.h index dc3f40c80c..32f8dd1156 100644 --- a/contrib/libs/zstd/lib/compress/zstd_cwksp.h +++ b/contrib/libs/zstd/lib/compress/zstd_cwksp.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -35,10 +35,10 @@ extern "C" { #define ZSTD_CWKSP_ASAN_REDZONE_SIZE 128 #endif - -/* Set our tables and aligneds to align by 64 bytes */ -#define ZSTD_CWKSP_ALIGNMENT_BYTES 64 - + +/* Set our tables and aligneds to align by 64 bytes */ +#define ZSTD_CWKSP_ALIGNMENT_BYTES 64 + /*-************************************* * Structures ***************************************/ @@ -49,16 +49,16 @@ typedef enum { } ZSTD_cwksp_alloc_phase_e; /** - * Used to describe whether the workspace is statically allocated (and will not - * necessarily ever be freed), or if it's dynamically allocated and we can - * expect a well-formed caller to free this. - */ -typedef enum { - ZSTD_cwksp_dynamic_alloc, - ZSTD_cwksp_static_alloc -} ZSTD_cwksp_static_alloc_e; - -/** + * Used to describe whether the workspace is statically allocated (and will not + * necessarily ever be freed), or if it's dynamically allocated and we can + * expect a well-formed caller to free this. + */ +typedef enum { + ZSTD_cwksp_dynamic_alloc, + ZSTD_cwksp_static_alloc +} ZSTD_cwksp_static_alloc_e; + +/** * Zstd fits all its internal datastructures into a single continuous buffer, * so that it only needs to perform a single OS allocation (or so that a buffer * can be provided to it and it can perform no allocations at all). This buffer @@ -106,7 +106,7 @@ typedef enum { * * - Static objects: this is optionally the enclosing ZSTD_CCtx or ZSTD_CDict, * so that literally everything fits in a single buffer. Note: if present, - * this must be the first object in the workspace, since ZSTD_customFree{CCtx, + * this must be the first object in the workspace, since ZSTD_customFree{CCtx, * CDict}() rely on a pointer comparison to see whether one or two frees are * required. * @@ -121,11 +121,11 @@ typedef enum { * - Tables: these are any of several different datastructures (hash tables, * chain tables, binary trees) that all respect a common format: they are * uint32_t arrays, all of whose values are between 0 and (nextSrc - base). - * Their sizes depend on the cparams. These tables are 64-byte aligned. + * Their sizes depend on the cparams. These tables are 64-byte aligned. * * - Aligned: these buffers are used for various purposes that require 4 byte - * alignment, but don't require any initialization before they're used. These - * buffers are each aligned to 64 bytes. + * alignment, but don't require any initialization before they're used. These + * buffers are each aligned to 64 bytes. * * - Buffers: these buffers are used for various purposes that don't require * any alignment or initialization before they're used. This means they can @@ -138,7 +138,7 @@ typedef enum { * * 1. Objects * 2. Buffers - * 3. Aligned/Tables + * 3. Aligned/Tables * * Attempts to reserve objects of different types out of order will fail. */ @@ -151,10 +151,10 @@ typedef struct { void* tableValidEnd; void* allocStart; - BYTE allocFailed; + BYTE allocFailed; int workspaceOversizedDuration; ZSTD_cwksp_alloc_phase_e phase; - ZSTD_cwksp_static_alloc_e isStatic; + ZSTD_cwksp_static_alloc_e isStatic; } ZSTD_cwksp; /*-************************************* @@ -191,129 +191,129 @@ MEM_STATIC size_t ZSTD_cwksp_align(size_t size, size_t const align) { * Since tables aren't currently redzoned, you don't need to call through this * to figure out how much space you need for the matchState tables. Everything * else is though. - * - * Do not use for sizing aligned buffers. Instead, use ZSTD_cwksp_aligned_alloc_size(). + * + * Do not use for sizing aligned buffers. Instead, use ZSTD_cwksp_aligned_alloc_size(). */ MEM_STATIC size_t ZSTD_cwksp_alloc_size(size_t size) { - if (size == 0) - return 0; -#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) + if (size == 0) + return 0; +#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) return size + 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE; #else return size; #endif } -/** - * Returns an adjusted alloc size that is the nearest larger multiple of 64 bytes. - * Used to determine the number of bytes required for a given "aligned". - */ -MEM_STATIC size_t ZSTD_cwksp_aligned_alloc_size(size_t size) { - return ZSTD_cwksp_alloc_size(ZSTD_cwksp_align(size, ZSTD_CWKSP_ALIGNMENT_BYTES)); -} - -/** - * Returns the amount of additional space the cwksp must allocate - * for internal purposes (currently only alignment). - */ -MEM_STATIC size_t ZSTD_cwksp_slack_space_required(void) { - /* For alignment, the wksp will always allocate an additional n_1=[1, 64] bytes - * to align the beginning of tables section, as well as another n_2=[0, 63] bytes +/** + * Returns an adjusted alloc size that is the nearest larger multiple of 64 bytes. + * Used to determine the number of bytes required for a given "aligned". + */ +MEM_STATIC size_t ZSTD_cwksp_aligned_alloc_size(size_t size) { + return ZSTD_cwksp_alloc_size(ZSTD_cwksp_align(size, ZSTD_CWKSP_ALIGNMENT_BYTES)); +} + +/** + * Returns the amount of additional space the cwksp must allocate + * for internal purposes (currently only alignment). + */ +MEM_STATIC size_t ZSTD_cwksp_slack_space_required(void) { + /* For alignment, the wksp will always allocate an additional n_1=[1, 64] bytes + * to align the beginning of tables section, as well as another n_2=[0, 63] bytes * to align the beginning of the aligned section. - * - * n_1 + n_2 == 64 bytes if the cwksp is freshly allocated, due to tables and - * aligneds being sized in multiples of 64 bytes. - */ - size_t const slackSpace = ZSTD_CWKSP_ALIGNMENT_BYTES; - return slackSpace; -} - - -/** - * Return the number of additional bytes required to align a pointer to the given number of bytes. - * alignBytes must be a power of two. - */ -MEM_STATIC size_t ZSTD_cwksp_bytes_to_align_ptr(void* ptr, const size_t alignBytes) { - size_t const alignBytesMask = alignBytes - 1; - size_t const bytes = (alignBytes - ((size_t)ptr & (alignBytesMask))) & alignBytesMask; - assert((alignBytes & alignBytesMask) == 0); - assert(bytes != ZSTD_CWKSP_ALIGNMENT_BYTES); - return bytes; -} - -/** - * Internal function. Do not use directly. + * + * n_1 + n_2 == 64 bytes if the cwksp is freshly allocated, due to tables and + * aligneds being sized in multiples of 64 bytes. + */ + size_t const slackSpace = ZSTD_CWKSP_ALIGNMENT_BYTES; + return slackSpace; +} + + +/** + * Return the number of additional bytes required to align a pointer to the given number of bytes. + * alignBytes must be a power of two. + */ +MEM_STATIC size_t ZSTD_cwksp_bytes_to_align_ptr(void* ptr, const size_t alignBytes) { + size_t const alignBytesMask = alignBytes - 1; + size_t const bytes = (alignBytes - ((size_t)ptr & (alignBytesMask))) & alignBytesMask; + assert((alignBytes & alignBytesMask) == 0); + assert(bytes != ZSTD_CWKSP_ALIGNMENT_BYTES); + return bytes; +} + +/** + * Internal function. Do not use directly. * Reserves the given number of bytes within the aligned/buffer segment of the wksp, * which counts from the end of the wksp (as opposed to the object/table segment). - * - * Returns a pointer to the beginning of that space. - */ + * + * Returns a pointer to the beginning of that space. + */ MEM_STATIC void* ZSTD_cwksp_reserve_internal_buffer_space(ZSTD_cwksp* ws, size_t const bytes) { - void* const alloc = (BYTE*)ws->allocStart - bytes; - void* const bottom = ws->tableEnd; - DEBUGLOG(5, "cwksp: reserving %p %zd bytes, %zd bytes remaining", - alloc, bytes, ZSTD_cwksp_available_space(ws) - bytes); - ZSTD_cwksp_assert_internal_consistency(ws); - assert(alloc >= bottom); - if (alloc < bottom) { - DEBUGLOG(4, "cwksp: alloc failed!"); - ws->allocFailed = 1; - return NULL; - } + void* const alloc = (BYTE*)ws->allocStart - bytes; + void* const bottom = ws->tableEnd; + DEBUGLOG(5, "cwksp: reserving %p %zd bytes, %zd bytes remaining", + alloc, bytes, ZSTD_cwksp_available_space(ws) - bytes); + ZSTD_cwksp_assert_internal_consistency(ws); + assert(alloc >= bottom); + if (alloc < bottom) { + DEBUGLOG(4, "cwksp: alloc failed!"); + ws->allocFailed = 1; + return NULL; + } /* the area is reserved from the end of wksp. * If it overlaps with tableValidEnd, it voids guarantees on values' range */ - if (alloc < ws->tableValidEnd) { - ws->tableValidEnd = alloc; - } - ws->allocStart = alloc; - return alloc; -} - -/** - * Moves the cwksp to the next phase, and does any necessary allocations. + if (alloc < ws->tableValidEnd) { + ws->tableValidEnd = alloc; + } + ws->allocStart = alloc; + return alloc; +} + +/** + * Moves the cwksp to the next phase, and does any necessary allocations. * cwksp initialization must necessarily go through each phase in order. - * Returns a 0 on success, or zstd error - */ + * Returns a 0 on success, or zstd error + */ MEM_STATIC size_t ZSTD_cwksp_internal_advance_phase(ZSTD_cwksp* ws, ZSTD_cwksp_alloc_phase_e phase) { assert(phase >= ws->phase); if (phase > ws->phase) { - /* Going from allocating objects to allocating buffers */ + /* Going from allocating objects to allocating buffers */ if (ws->phase < ZSTD_cwksp_alloc_buffers && phase >= ZSTD_cwksp_alloc_buffers) { ws->tableValidEnd = ws->objectEnd; } - - /* Going from allocating buffers to allocating aligneds/tables */ + + /* Going from allocating buffers to allocating aligneds/tables */ if (ws->phase < ZSTD_cwksp_alloc_aligned && phase >= ZSTD_cwksp_alloc_aligned) { - { /* Align the start of the "aligned" to 64 bytes. Use [1, 64] bytes. */ - size_t const bytesToAlign = - ZSTD_CWKSP_ALIGNMENT_BYTES - ZSTD_cwksp_bytes_to_align_ptr(ws->allocStart, ZSTD_CWKSP_ALIGNMENT_BYTES); - DEBUGLOG(5, "reserving aligned alignment addtl space: %zu", bytesToAlign); - ZSTD_STATIC_ASSERT((ZSTD_CWKSP_ALIGNMENT_BYTES & (ZSTD_CWKSP_ALIGNMENT_BYTES - 1)) == 0); /* power of 2 */ - RETURN_ERROR_IF(!ZSTD_cwksp_reserve_internal_buffer_space(ws, bytesToAlign), - memory_allocation, "aligned phase - alignment initial allocation failed!"); + { /* Align the start of the "aligned" to 64 bytes. Use [1, 64] bytes. */ + size_t const bytesToAlign = + ZSTD_CWKSP_ALIGNMENT_BYTES - ZSTD_cwksp_bytes_to_align_ptr(ws->allocStart, ZSTD_CWKSP_ALIGNMENT_BYTES); + DEBUGLOG(5, "reserving aligned alignment addtl space: %zu", bytesToAlign); + ZSTD_STATIC_ASSERT((ZSTD_CWKSP_ALIGNMENT_BYTES & (ZSTD_CWKSP_ALIGNMENT_BYTES - 1)) == 0); /* power of 2 */ + RETURN_ERROR_IF(!ZSTD_cwksp_reserve_internal_buffer_space(ws, bytesToAlign), + memory_allocation, "aligned phase - alignment initial allocation failed!"); } - { /* Align the start of the tables to 64 bytes. Use [0, 63] bytes */ - void* const alloc = ws->objectEnd; - size_t const bytesToAlign = ZSTD_cwksp_bytes_to_align_ptr(alloc, ZSTD_CWKSP_ALIGNMENT_BYTES); + { /* Align the start of the tables to 64 bytes. Use [0, 63] bytes */ + void* const alloc = ws->objectEnd; + size_t const bytesToAlign = ZSTD_cwksp_bytes_to_align_ptr(alloc, ZSTD_CWKSP_ALIGNMENT_BYTES); void* const objectEnd = (BYTE*)alloc + bytesToAlign; - DEBUGLOG(5, "reserving table alignment addtl space: %zu", bytesToAlign); + DEBUGLOG(5, "reserving table alignment addtl space: %zu", bytesToAlign); RETURN_ERROR_IF(objectEnd > ws->workspaceEnd, memory_allocation, - "table phase - alignment initial allocation failed!"); + "table phase - alignment initial allocation failed!"); ws->objectEnd = objectEnd; ws->tableEnd = objectEnd; /* table area starts being empty */ if (ws->tableValidEnd < ws->tableEnd) { ws->tableValidEnd = ws->tableEnd; } } } ws->phase = phase; - ZSTD_cwksp_assert_internal_consistency(ws); + ZSTD_cwksp_assert_internal_consistency(ws); } - return 0; + return 0; } /** @@ -331,26 +331,26 @@ MEM_STATIC void* ZSTD_cwksp_reserve_internal(ZSTD_cwksp* ws, size_t bytes, ZSTD_cwksp_alloc_phase_e phase) { void* alloc; - if (ZSTD_isError(ZSTD_cwksp_internal_advance_phase(ws, phase)) || bytes == 0) { - return NULL; - } - -#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) + if (ZSTD_isError(ZSTD_cwksp_internal_advance_phase(ws, phase)) || bytes == 0) { + return NULL; + } + +#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) /* over-reserve space */ - bytes += 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE; + bytes += 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE; #endif - alloc = ZSTD_cwksp_reserve_internal_buffer_space(ws, bytes); + alloc = ZSTD_cwksp_reserve_internal_buffer_space(ws, bytes); -#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) +#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) /* Move alloc so there's ZSTD_CWKSP_ASAN_REDZONE_SIZE unused space on * either size. */ - if (alloc) { - alloc = (BYTE *)alloc + ZSTD_CWKSP_ASAN_REDZONE_SIZE; - if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) { - __asan_unpoison_memory_region(alloc, bytes); - } - } + if (alloc) { + alloc = (BYTE *)alloc + ZSTD_CWKSP_ASAN_REDZONE_SIZE; + if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) { + __asan_unpoison_memory_region(alloc, bytes); + } + } #endif return alloc; @@ -365,35 +365,35 @@ MEM_STATIC BYTE* ZSTD_cwksp_reserve_buffer(ZSTD_cwksp* ws, size_t bytes) } /** - * Reserves and returns memory sized on and aligned on ZSTD_CWKSP_ALIGNMENT_BYTES (64 bytes). + * Reserves and returns memory sized on and aligned on ZSTD_CWKSP_ALIGNMENT_BYTES (64 bytes). */ MEM_STATIC void* ZSTD_cwksp_reserve_aligned(ZSTD_cwksp* ws, size_t bytes) { - void* ptr = ZSTD_cwksp_reserve_internal(ws, ZSTD_cwksp_align(bytes, ZSTD_CWKSP_ALIGNMENT_BYTES), - ZSTD_cwksp_alloc_aligned); - assert(((size_t)ptr & (ZSTD_CWKSP_ALIGNMENT_BYTES-1))== 0); - return ptr; + void* ptr = ZSTD_cwksp_reserve_internal(ws, ZSTD_cwksp_align(bytes, ZSTD_CWKSP_ALIGNMENT_BYTES), + ZSTD_cwksp_alloc_aligned); + assert(((size_t)ptr & (ZSTD_CWKSP_ALIGNMENT_BYTES-1))== 0); + return ptr; } /** - * Aligned on 64 bytes. These buffers have the special property that + * Aligned on 64 bytes. These buffers have the special property that * their values remain constrained, allowing us to re-use them without * memset()-ing them. */ MEM_STATIC void* ZSTD_cwksp_reserve_table(ZSTD_cwksp* ws, size_t bytes) { const ZSTD_cwksp_alloc_phase_e phase = ZSTD_cwksp_alloc_aligned; - void* alloc; - void* end; - void* top; - - if (ZSTD_isError(ZSTD_cwksp_internal_advance_phase(ws, phase))) { - return NULL; - } - alloc = ws->tableEnd; - end = (BYTE *)alloc + bytes; - top = ws->allocStart; - + void* alloc; + void* end; + void* top; + + if (ZSTD_isError(ZSTD_cwksp_internal_advance_phase(ws, phase))) { + return NULL; + } + alloc = ws->tableEnd; + end = (BYTE *)alloc + bytes; + top = ws->allocStart; + DEBUGLOG(5, "cwksp: reserving %p table %zd bytes, %zd bytes remaining", alloc, bytes, ZSTD_cwksp_available_space(ws) - bytes); assert((bytes & (sizeof(U32)-1)) == 0); @@ -406,14 +406,14 @@ MEM_STATIC void* ZSTD_cwksp_reserve_table(ZSTD_cwksp* ws, size_t bytes) } ws->tableEnd = end; -#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) - if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) { - __asan_unpoison_memory_region(alloc, bytes); - } +#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) + if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) { + __asan_unpoison_memory_region(alloc, bytes); + } #endif - assert((bytes & (ZSTD_CWKSP_ALIGNMENT_BYTES-1)) == 0); - assert(((size_t)alloc & (ZSTD_CWKSP_ALIGNMENT_BYTES-1))== 0); + assert((bytes & (ZSTD_CWKSP_ALIGNMENT_BYTES-1)) == 0); + assert(((size_t)alloc & (ZSTD_CWKSP_ALIGNMENT_BYTES-1))== 0); return alloc; } @@ -427,7 +427,7 @@ MEM_STATIC void* ZSTD_cwksp_reserve_object(ZSTD_cwksp* ws, size_t bytes) void* alloc = ws->objectEnd; void* end = (BYTE*)alloc + roundedBytes; -#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) +#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) /* over-reserve space */ end = (BYTE *)end + 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE; #endif @@ -448,13 +448,13 @@ MEM_STATIC void* ZSTD_cwksp_reserve_object(ZSTD_cwksp* ws, size_t bytes) ws->tableEnd = end; ws->tableValidEnd = end; -#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) +#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) /* Move alloc so there's ZSTD_CWKSP_ASAN_REDZONE_SIZE unused space on * either size. */ alloc = (BYTE*)alloc + ZSTD_CWKSP_ASAN_REDZONE_SIZE; - if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) { - __asan_unpoison_memory_region(alloc, bytes); - } + if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) { + __asan_unpoison_memory_region(alloc, bytes); + } #endif return alloc; @@ -464,7 +464,7 @@ MEM_STATIC void ZSTD_cwksp_mark_tables_dirty(ZSTD_cwksp* ws) { DEBUGLOG(4, "cwksp: ZSTD_cwksp_mark_tables_dirty"); -#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE) +#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE) /* To validate that the table re-use logic is sound, and that we don't * access table space that we haven't cleaned, we re-"poison" the table * space every time we mark it dirty. */ @@ -499,7 +499,7 @@ MEM_STATIC void ZSTD_cwksp_clean_tables(ZSTD_cwksp* ws) { assert(ws->tableValidEnd >= ws->objectEnd); assert(ws->tableValidEnd <= ws->allocStart); if (ws->tableValidEnd < ws->tableEnd) { - ZSTD_memset(ws->tableValidEnd, 0, (BYTE*)ws->tableEnd - (BYTE*)ws->tableValidEnd); + ZSTD_memset(ws->tableValidEnd, 0, (BYTE*)ws->tableEnd - (BYTE*)ws->tableValidEnd); } ZSTD_cwksp_mark_tables_clean(ws); } @@ -511,12 +511,12 @@ MEM_STATIC void ZSTD_cwksp_clean_tables(ZSTD_cwksp* ws) { MEM_STATIC void ZSTD_cwksp_clear_tables(ZSTD_cwksp* ws) { DEBUGLOG(4, "cwksp: clearing tables!"); -#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) - /* We don't do this when the workspace is statically allocated, because - * when that is the case, we have no capability to hook into the end of the - * workspace's lifecycle to unpoison the memory. - */ - if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) { +#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) + /* We don't do this when the workspace is statically allocated, because + * when that is the case, we have no capability to hook into the end of the + * workspace's lifecycle to unpoison the memory. + */ + if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) { size_t size = (BYTE*)ws->tableValidEnd - (BYTE*)ws->objectEnd; __asan_poison_memory_region(ws->objectEnd, size); } @@ -533,7 +533,7 @@ MEM_STATIC void ZSTD_cwksp_clear_tables(ZSTD_cwksp* ws) { MEM_STATIC void ZSTD_cwksp_clear(ZSTD_cwksp* ws) { DEBUGLOG(4, "cwksp: clearing!"); -#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE) +#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE) /* To validate that the context re-use logic is sound, and that we don't * access stuff that this compression hasn't initialized, we re-"poison" * the workspace (or at least the non-static, non-table parts of it) @@ -544,12 +544,12 @@ MEM_STATIC void ZSTD_cwksp_clear(ZSTD_cwksp* ws) { } #endif -#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) - /* We don't do this when the workspace is statically allocated, because - * when that is the case, we have no capability to hook into the end of the - * workspace's lifecycle to unpoison the memory. - */ - if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) { +#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) + /* We don't do this when the workspace is statically allocated, because + * when that is the case, we have no capability to hook into the end of the + * workspace's lifecycle to unpoison the memory. + */ + if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) { size_t size = (BYTE*)ws->workspaceEnd - (BYTE*)ws->objectEnd; __asan_poison_memory_region(ws->objectEnd, size); } @@ -569,7 +569,7 @@ MEM_STATIC void ZSTD_cwksp_clear(ZSTD_cwksp* ws) { * Any existing values in the workspace are ignored (the previously managed * buffer, if present, must be separately freed). */ -MEM_STATIC void ZSTD_cwksp_init(ZSTD_cwksp* ws, void* start, size_t size, ZSTD_cwksp_static_alloc_e isStatic) { +MEM_STATIC void ZSTD_cwksp_init(ZSTD_cwksp* ws, void* start, size_t size, ZSTD_cwksp_static_alloc_e isStatic) { DEBUGLOG(4, "cwksp: init'ing workspace with %zd bytes", size); assert(((size_t)start & (sizeof(void*)-1)) == 0); /* ensure correct alignment */ ws->workspace = start; @@ -577,45 +577,45 @@ MEM_STATIC void ZSTD_cwksp_init(ZSTD_cwksp* ws, void* start, size_t size, ZSTD_c ws->objectEnd = ws->workspace; ws->tableValidEnd = ws->objectEnd; ws->phase = ZSTD_cwksp_alloc_objects; - ws->isStatic = isStatic; + ws->isStatic = isStatic; ZSTD_cwksp_clear(ws); ws->workspaceOversizedDuration = 0; ZSTD_cwksp_assert_internal_consistency(ws); } MEM_STATIC size_t ZSTD_cwksp_create(ZSTD_cwksp* ws, size_t size, ZSTD_customMem customMem) { - void* workspace = ZSTD_customMalloc(size, customMem); + void* workspace = ZSTD_customMalloc(size, customMem); DEBUGLOG(4, "cwksp: creating new workspace with %zd bytes", size); RETURN_ERROR_IF(workspace == NULL, memory_allocation, "NULL pointer!"); - ZSTD_cwksp_init(ws, workspace, size, ZSTD_cwksp_dynamic_alloc); + ZSTD_cwksp_init(ws, workspace, size, ZSTD_cwksp_dynamic_alloc); return 0; } MEM_STATIC void ZSTD_cwksp_free(ZSTD_cwksp* ws, ZSTD_customMem customMem) { void *ptr = ws->workspace; DEBUGLOG(4, "cwksp: freeing workspace"); - ZSTD_memset(ws, 0, sizeof(ZSTD_cwksp)); - ZSTD_customFree(ptr, customMem); + ZSTD_memset(ws, 0, sizeof(ZSTD_cwksp)); + ZSTD_customFree(ptr, customMem); } /** * Moves the management of a workspace from one cwksp to another. The src cwksp - * is left in an invalid state (src must be re-init()'ed before it's used again). + * is left in an invalid state (src must be re-init()'ed before it's used again). */ MEM_STATIC void ZSTD_cwksp_move(ZSTD_cwksp* dst, ZSTD_cwksp* src) { *dst = *src; - ZSTD_memset(src, 0, sizeof(ZSTD_cwksp)); + ZSTD_memset(src, 0, sizeof(ZSTD_cwksp)); } MEM_STATIC size_t ZSTD_cwksp_sizeof(const ZSTD_cwksp* ws) { return (size_t)((BYTE*)ws->workspaceEnd - (BYTE*)ws->workspace); } -MEM_STATIC size_t ZSTD_cwksp_used(const ZSTD_cwksp* ws) { - return (size_t)((BYTE*)ws->tableEnd - (BYTE*)ws->workspace) - + (size_t)((BYTE*)ws->workspaceEnd - (BYTE*)ws->allocStart); -} - +MEM_STATIC size_t ZSTD_cwksp_used(const ZSTD_cwksp* ws) { + return (size_t)((BYTE*)ws->tableEnd - (BYTE*)ws->workspace) + + (size_t)((BYTE*)ws->workspaceEnd - (BYTE*)ws->allocStart); +} + MEM_STATIC int ZSTD_cwksp_reserve_failed(const ZSTD_cwksp* ws) { return ws->allocFailed; } @@ -624,24 +624,24 @@ MEM_STATIC int ZSTD_cwksp_reserve_failed(const ZSTD_cwksp* ws) { * Functions Checking Free Space ***************************************/ -/* ZSTD_alignmentSpaceWithinBounds() : - * Returns if the estimated space needed for a wksp is within an acceptable limit of the - * actual amount of space used. - */ -MEM_STATIC int ZSTD_cwksp_estimated_space_within_bounds(const ZSTD_cwksp* const ws, - size_t const estimatedSpace, int resizedWorkspace) { - if (resizedWorkspace) { - /* Resized/newly allocated wksp should have exact bounds */ - return ZSTD_cwksp_used(ws) == estimatedSpace; - } else { - /* Due to alignment, when reusing a workspace, we can actually consume 63 fewer or more bytes - * than estimatedSpace. See the comments in zstd_cwksp.h for details. - */ - return (ZSTD_cwksp_used(ws) >= estimatedSpace - 63) && (ZSTD_cwksp_used(ws) <= estimatedSpace + 63); - } -} - - +/* ZSTD_alignmentSpaceWithinBounds() : + * Returns if the estimated space needed for a wksp is within an acceptable limit of the + * actual amount of space used. + */ +MEM_STATIC int ZSTD_cwksp_estimated_space_within_bounds(const ZSTD_cwksp* const ws, + size_t const estimatedSpace, int resizedWorkspace) { + if (resizedWorkspace) { + /* Resized/newly allocated wksp should have exact bounds */ + return ZSTD_cwksp_used(ws) == estimatedSpace; + } else { + /* Due to alignment, when reusing a workspace, we can actually consume 63 fewer or more bytes + * than estimatedSpace. See the comments in zstd_cwksp.h for details. + */ + return (ZSTD_cwksp_used(ws) >= estimatedSpace - 63) && (ZSTD_cwksp_used(ws) <= estimatedSpace + 63); + } +} + + MEM_STATIC size_t ZSTD_cwksp_available_space(ZSTD_cwksp* ws) { return (size_t)((BYTE*)ws->allocStart - (BYTE*)ws->tableEnd); } diff --git a/contrib/libs/zstd/lib/compress/zstd_double_fast.c b/contrib/libs/zstd/lib/compress/zstd_double_fast.c index 76933dea26..1b8abb8841 100644 --- a/contrib/libs/zstd/lib/compress/zstd_double_fast.c +++ b/contrib/libs/zstd/lib/compress/zstd_double_fast.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -31,15 +31,15 @@ void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms, * is empty. */ for (; ip + fastHashFillStep - 1 <= iend; ip += fastHashFillStep) { - U32 const curr = (U32)(ip - base); + U32 const curr = (U32)(ip - base); U32 i; for (i = 0; i < fastHashFillStep; ++i) { size_t const smHash = ZSTD_hashPtr(ip + i, hBitsS, mls); size_t const lgHash = ZSTD_hashPtr(ip + i, hBitsL, 8); if (i == 0) - hashSmall[smHash] = curr + i; + hashSmall[smHash] = curr + i; if (i == 0 || hashLarge[lgHash] == 0) - hashLarge[lgHash] = curr + i; + hashLarge[lgHash] = curr + i; /* Only load extra positions for ZSTD_dtlm_full */ if (dtlm == ZSTD_dtlm_fast) break; @@ -311,16 +311,16 @@ size_t ZSTD_compressBlock_doubleFast_dictMatchState_generic( size_t const h = ZSTD_hashPtr(ip, hBitsS, mls); size_t const dictHL = ZSTD_hashPtr(ip, dictHBitsL, 8); size_t const dictHS = ZSTD_hashPtr(ip, dictHBitsS, mls); - U32 const curr = (U32)(ip-base); + U32 const curr = (U32)(ip-base); U32 const matchIndexL = hashLong[h2]; U32 matchIndexS = hashSmall[h]; const BYTE* matchLong = base + matchIndexL; const BYTE* match = base + matchIndexS; - const U32 repIndex = curr + 1 - offset_1; + const U32 repIndex = curr + 1 - offset_1; const BYTE* repMatch = (repIndex < prefixLowestIndex) ? dictBase + (repIndex - dictIndexDelta) : base + repIndex; - hashLong[h2] = hashSmall[h] = curr; /* update hash tables */ + hashLong[h2] = hashSmall[h] = curr; /* update hash tables */ /* check repcode */ if (((U32)((prefixLowestIndex-1) - repIndex) >= 3 /* intentional underflow */) @@ -348,7 +348,7 @@ size_t ZSTD_compressBlock_doubleFast_dictMatchState_generic( if (dictMatchL > dictStart && MEM_read64(dictMatchL) == MEM_read64(ip)) { mLength = ZSTD_count_2segments(ip+8, dictMatchL+8, iend, dictEnd, prefixLowest) + 8; - offset = (U32)(curr - dictMatchIndexL - dictIndexDelta); + offset = (U32)(curr - dictMatchIndexL - dictIndexDelta); while (((ip>anchor) & (dictMatchL>dictStart)) && (ip[-1] == dictMatchL[-1])) { ip--; dictMatchL--; mLength++; } /* catch up */ goto _match_found; } } @@ -380,7 +380,7 @@ _search_next_long: size_t const dictHLNext = ZSTD_hashPtr(ip+1, dictHBitsL, 8); U32 const matchIndexL3 = hashLong[hl3]; const BYTE* matchL3 = base + matchIndexL3; - hashLong[hl3] = curr + 1; + hashLong[hl3] = curr + 1; /* check prefix long +1 match */ if (matchIndexL3 > prefixLowestIndex) { @@ -399,7 +399,7 @@ _search_next_long: if (dictMatchL3 > dictStart && MEM_read64(dictMatchL3) == MEM_read64(ip+1)) { mLength = ZSTD_count_2segments(ip+1+8, dictMatchL3+8, iend, dictEnd, prefixLowest) + 8; ip++; - offset = (U32)(curr + 1 - dictMatchIndexL3 - dictIndexDelta); + offset = (U32)(curr + 1 - dictMatchIndexL3 - dictIndexDelta); while (((ip>anchor) & (dictMatchL3>dictStart)) && (ip[-1] == dictMatchL3[-1])) { ip--; dictMatchL3--; mLength++; } /* catch up */ goto _match_found; } } } @@ -407,7 +407,7 @@ _search_next_long: /* if no long +1 match, explore the short match we found */ if (matchIndexS < prefixLowestIndex) { mLength = ZSTD_count_2segments(ip+4, match+4, iend, dictEnd, prefixLowest) + 4; - offset = (U32)(curr - matchIndexS); + offset = (U32)(curr - matchIndexS); while (((ip>anchor) & (match>dictStart)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */ } else { mLength = ZSTD_count(ip+4, match+4, iend) + 4; @@ -429,7 +429,7 @@ _match_stored: if (ip <= ilimit) { /* Complementary insertion */ /* done after iLimit test, as candidates could be > iend-8 */ - { U32 const indexToInsert = curr+2; + { U32 const indexToInsert = curr+2; hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert; hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base); hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert; @@ -572,12 +572,12 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic( const BYTE* const matchLongBase = matchLongIndex < prefixStartIndex ? dictBase : base; const BYTE* matchLong = matchLongBase + matchLongIndex; - const U32 curr = (U32)(ip-base); - const U32 repIndex = curr + 1 - offset_1; /* offset_1 expected <= curr +1 */ + const U32 curr = (U32)(ip-base); + const U32 repIndex = curr + 1 - offset_1; /* offset_1 expected <= curr +1 */ const BYTE* const repBase = repIndex < prefixStartIndex ? dictBase : base; const BYTE* const repMatch = repBase + repIndex; size_t mLength; - hashSmall[hSmall] = hashLong[hLong] = curr; /* update hash table */ + hashSmall[hSmall] = hashLong[hLong] = curr; /* update hash table */ if ((((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow : ensure repIndex doesn't overlap dict + prefix */ & (offset_1 <= curr+1 - dictStartIndex)) /* note: we are searching at curr+1 */ @@ -592,7 +592,7 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic( const BYTE* const lowMatchPtr = matchLongIndex < prefixStartIndex ? dictStart : prefixStart; U32 offset; mLength = ZSTD_count_2segments(ip+8, matchLong+8, iend, matchEnd, prefixStart) + 8; - offset = curr - matchLongIndex; + offset = curr - matchLongIndex; while (((ip>anchor) & (matchLong>lowMatchPtr)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; } /* catch up */ offset_2 = offset_1; offset_1 = offset; @@ -604,19 +604,19 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic( const BYTE* const match3Base = matchIndex3 < prefixStartIndex ? dictBase : base; const BYTE* match3 = match3Base + matchIndex3; U32 offset; - hashLong[h3] = curr + 1; + hashLong[h3] = curr + 1; if ( (matchIndex3 > dictStartIndex) && (MEM_read64(match3) == MEM_read64(ip+1)) ) { const BYTE* const matchEnd = matchIndex3 < prefixStartIndex ? dictEnd : iend; const BYTE* const lowMatchPtr = matchIndex3 < prefixStartIndex ? dictStart : prefixStart; mLength = ZSTD_count_2segments(ip+9, match3+8, iend, matchEnd, prefixStart) + 8; ip++; - offset = curr+1 - matchIndex3; + offset = curr+1 - matchIndex3; while (((ip>anchor) & (match3>lowMatchPtr)) && (ip[-1] == match3[-1])) { ip--; match3--; mLength++; } /* catch up */ } else { const BYTE* const matchEnd = matchIndex < prefixStartIndex ? dictEnd : iend; const BYTE* const lowMatchPtr = matchIndex < prefixStartIndex ? dictStart : prefixStart; mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, prefixStart) + 4; - offset = curr - matchIndex; + offset = curr - matchIndex; while (((ip>anchor) & (match>lowMatchPtr)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */ } offset_2 = offset_1; @@ -635,7 +635,7 @@ static size_t ZSTD_compressBlock_doubleFast_extDict_generic( if (ip <= ilimit) { /* Complementary insertion */ /* done after iLimit test, as candidates could be > iend-8 */ - { U32 const indexToInsert = curr+2; + { U32 const indexToInsert = curr+2; hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert; hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base); hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert; diff --git a/contrib/libs/zstd/lib/compress/zstd_double_fast.h b/contrib/libs/zstd/lib/compress/zstd_double_fast.h index e16b7b03a3..a80779c82e 100644 --- a/contrib/libs/zstd/lib/compress/zstd_double_fast.h +++ b/contrib/libs/zstd/lib/compress/zstd_double_fast.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the diff --git a/contrib/libs/zstd/lib/compress/zstd_fast.c b/contrib/libs/zstd/lib/compress/zstd_fast.c index 802fc31579..3110551326 100644 --- a/contrib/libs/zstd/lib/compress/zstd_fast.c +++ b/contrib/libs/zstd/lib/compress/zstd_fast.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -29,16 +29,16 @@ void ZSTD_fillHashTable(ZSTD_matchState_t* ms, * Insert the other positions if their hash entry is empty. */ for ( ; ip + fastHashFillStep < iend + 2; ip += fastHashFillStep) { - U32 const curr = (U32)(ip - base); + U32 const curr = (U32)(ip - base); size_t const hash0 = ZSTD_hashPtr(ip, hBits, mls); - hashTable[hash0] = curr; + hashTable[hash0] = curr; if (dtlm == ZSTD_dtlm_fast) continue; /* Only load extra positions for ZSTD_dtlm_full */ { U32 p; for (p = 1; p < fastHashFillStep; ++p) { size_t const hash = ZSTD_hashPtr(ip + p, hBits, mls); if (hashTable[hash] == 0) { /* not yet filled */ - hashTable[hash] = curr + p; + hashTable[hash] = curr + p; } } } } } @@ -138,9 +138,9 @@ ZSTD_compressBlock_fast_noDict_generic( DEBUGLOG(5, "ZSTD_compressBlock_fast_generic"); ip0 += (ip0 == prefixStart); - { U32 const curr = (U32)(ip0 - base); - U32 const windowLow = ZSTD_getLowestPrefixIndex(ms, curr, cParams->windowLog); - U32 const maxRep = curr - windowLow; + { U32 const curr = (U32)(ip0 - base); + U32 const windowLow = ZSTD_getLowestPrefixIndex(ms, curr, cParams->windowLog); + U32 const maxRep = curr - windowLow; if (rep_offset2 > maxRep) offsetSaved = rep_offset2, rep_offset2 = 0; if (rep_offset1 > maxRep) offsetSaved = rep_offset1, rep_offset1 = 0; } @@ -409,7 +409,7 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic( (void)hasStep; /* not currently specialized on whether it's accelerated */ - /* ensure there will be no underflow + /* ensure there will be no underflow * when translating a dict index into a local index */ assert(prefixStartIndex >= (U32)(dictEnd - dictBase)); @@ -425,14 +425,14 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic( while (ip < ilimit) { /* < instead of <=, because repcode check at (ip+1) */ size_t mLength; size_t const h = ZSTD_hashPtr(ip, hlog, mls); - U32 const curr = (U32)(ip-base); + U32 const curr = (U32)(ip-base); U32 const matchIndex = hashTable[h]; const BYTE* match = base + matchIndex; - const U32 repIndex = curr + 1 - offset_1; + const U32 repIndex = curr + 1 - offset_1; const BYTE* repMatch = (repIndex < prefixStartIndex) ? dictBase + (repIndex - dictIndexDelta) : base + repIndex; - hashTable[h] = curr; /* update hash table */ + hashTable[h] = curr; /* update hash table */ if ( ((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow : ensure repIndex isn't overlapping dict + prefix */ && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) { @@ -451,7 +451,7 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic( continue; } else { /* found a dict match */ - U32 const offset = (U32)(curr-dictMatchIndex-dictIndexDelta); + U32 const offset = (U32)(curr-dictMatchIndex-dictIndexDelta); mLength = ZSTD_count_2segments(ip+4, dictMatch+4, iend, dictEnd, prefixStart) + 4; while (((ip>anchor) & (dictMatch>dictStart)) && (ip[-1] == dictMatch[-1])) { @@ -483,8 +483,8 @@ size_t ZSTD_compressBlock_fast_dictMatchState_generic( if (ip <= ilimit) { /* Fill Table */ - assert(base+curr+2 > istart); /* check base overflow */ - hashTable[ZSTD_hashPtr(base+curr+2, hlog, mls)] = curr+2; /* here because curr+2 could be > iend-8 */ + assert(base+curr+2 > istart); /* check base overflow */ + hashTable[ZSTD_hashPtr(base+curr+2, hlog, mls)] = curr+2; /* here because curr+2 could be > iend-8 */ hashTable[ZSTD_hashPtr(ip-2, hlog, mls)] = (U32)(ip-2-base); /* check immediate repcode */ @@ -585,14 +585,14 @@ static size_t ZSTD_compressBlock_fast_extDict_generic( const U32 matchIndex = hashTable[h]; const BYTE* const matchBase = matchIndex < prefixStartIndex ? dictBase : base; const BYTE* match = matchBase + matchIndex; - const U32 curr = (U32)(ip-base); - const U32 repIndex = curr + 1 - offset_1; + const U32 curr = (U32)(ip-base); + const U32 repIndex = curr + 1 - offset_1; const BYTE* const repBase = repIndex < prefixStartIndex ? dictBase : base; const BYTE* const repMatch = repBase + repIndex; - hashTable[h] = curr; /* update hash table */ - DEBUGLOG(7, "offset_1 = %u , curr = %u", offset_1, curr); + hashTable[h] = curr; /* update hash table */ + DEBUGLOG(7, "offset_1 = %u , curr = %u", offset_1, curr); - if ( ( ((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow */ + if ( ( ((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow */ & (offset_1 <= curr+1 - dictStartIndex) ) /* note: we are searching at curr+1 */ && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) { const BYTE* const repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend; @@ -610,7 +610,7 @@ static size_t ZSTD_compressBlock_fast_extDict_generic( } { const BYTE* const matchEnd = matchIndex < prefixStartIndex ? dictEnd : iend; const BYTE* const lowMatchPtr = matchIndex < prefixStartIndex ? dictStart : prefixStart; - U32 const offset = curr - matchIndex; + U32 const offset = curr - matchIndex; size_t mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, prefixStart) + 4; while (((ip>anchor) & (match>lowMatchPtr)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */ offset_2 = offset_1; offset_1 = offset; /* update offset history */ @@ -621,7 +621,7 @@ static size_t ZSTD_compressBlock_fast_extDict_generic( if (ip <= ilimit) { /* Fill Table */ - hashTable[ZSTD_hashPtr(base+curr+2, hlog, mls)] = curr+2; + hashTable[ZSTD_hashPtr(base+curr+2, hlog, mls)] = curr+2; hashTable[ZSTD_hashPtr(ip-2, hlog, mls)] = (U32)(ip-2-base); /* check immediate repcode */ while (ip <= ilimit) { diff --git a/contrib/libs/zstd/lib/compress/zstd_fast.h b/contrib/libs/zstd/lib/compress/zstd_fast.h index 0d4a0c1090..4440bec22b 100644 --- a/contrib/libs/zstd/lib/compress/zstd_fast.h +++ b/contrib/libs/zstd/lib/compress/zstd_fast.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the diff --git a/contrib/libs/zstd/lib/compress/zstd_lazy.c b/contrib/libs/zstd/lib/compress/zstd_lazy.c index 2e38dcb46d..44399a6e0c 100644 --- a/contrib/libs/zstd/lib/compress/zstd_lazy.c +++ b/contrib/libs/zstd/lib/compress/zstd_lazy.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -58,11 +58,11 @@ ZSTD_updateDUBT(ZSTD_matchState_t* ms, /** ZSTD_insertDUBT1() : * sort one already inserted but unsorted position - * assumption : curr >= btlow == (curr - btmask) + * assumption : curr >= btlow == (curr - btmask) * doesn't fail */ static void ZSTD_insertDUBT1(const ZSTD_matchState_t* ms, - U32 curr, const BYTE* inputEnd, + U32 curr, const BYTE* inputEnd, U32 nbCompares, U32 btLow, const ZSTD_dictMode_e dictMode) { @@ -74,41 +74,41 @@ ZSTD_insertDUBT1(const ZSTD_matchState_t* ms, const BYTE* const base = ms->window.base; const BYTE* const dictBase = ms->window.dictBase; const U32 dictLimit = ms->window.dictLimit; - const BYTE* const ip = (curr>=dictLimit) ? base + curr : dictBase + curr; - const BYTE* const iend = (curr>=dictLimit) ? inputEnd : dictBase + dictLimit; + const BYTE* const ip = (curr>=dictLimit) ? base + curr : dictBase + curr; + const BYTE* const iend = (curr>=dictLimit) ? inputEnd : dictBase + dictLimit; const BYTE* const dictEnd = dictBase + dictLimit; const BYTE* const prefixStart = base + dictLimit; const BYTE* match; - U32* smallerPtr = bt + 2*(curr&btMask); + U32* smallerPtr = bt + 2*(curr&btMask); U32* largerPtr = smallerPtr + 1; U32 matchIndex = *smallerPtr; /* this candidate is unsorted : next sorted candidate is reached through *smallerPtr, while *largerPtr contains previous unsorted candidate (which is already saved and can be overwritten) */ U32 dummy32; /* to be nullified at the end */ U32 const windowValid = ms->window.lowLimit; U32 const maxDistance = 1U << cParams->windowLog; - U32 const windowLow = (curr - windowValid > maxDistance) ? curr - maxDistance : windowValid; + U32 const windowLow = (curr - windowValid > maxDistance) ? curr - maxDistance : windowValid; DEBUGLOG(8, "ZSTD_insertDUBT1(%u) (dictLimit=%u, lowLimit=%u)", - curr, dictLimit, windowLow); - assert(curr >= btLow); + curr, dictLimit, windowLow); + assert(curr >= btLow); assert(ip < iend); /* condition for ZSTD_count */ for (; nbCompares && (matchIndex > windowLow); --nbCompares) { U32* const nextPtr = bt + 2*(matchIndex & btMask); size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */ - assert(matchIndex < curr); + assert(matchIndex < curr); /* note : all candidates are now supposed sorted, * but it's still possible to have nextPtr[1] == ZSTD_DUBT_UNSORTED_MARK * when a real index has the same value as ZSTD_DUBT_UNSORTED_MARK */ if ( (dictMode != ZSTD_extDict) || (matchIndex+matchLength >= dictLimit) /* both in current segment*/ - || (curr < dictLimit) /* both in extDict */) { + || (curr < dictLimit) /* both in extDict */) { const BYTE* const mBase = ( (dictMode != ZSTD_extDict) || (matchIndex+matchLength >= dictLimit)) ? base : dictBase; assert( (matchIndex+matchLength >= dictLimit) /* might be wrong if extDict is incorrectly set to 0 */ - || (curr < dictLimit) ); + || (curr < dictLimit) ); match = mBase + matchIndex; matchLength += ZSTD_count(ip+matchLength, match+matchLength, iend); } else { @@ -119,7 +119,7 @@ ZSTD_insertDUBT1(const ZSTD_matchState_t* ms, } DEBUGLOG(8, "ZSTD_insertDUBT1: comparing %u with %u : found %u common bytes ", - curr, matchIndex, (U32)matchLength); + curr, matchIndex, (U32)matchLength); if (ip+matchLength == iend) { /* equal : no way to know if inf or sup */ break; /* drop , to guarantee consistency ; miss a bit of compression, but other solutions can corrupt tree */ @@ -168,7 +168,7 @@ ZSTD_DUBT_findBetterDictMatch ( const BYTE* const base = ms->window.base; const BYTE* const prefixStart = base + ms->window.dictLimit; - U32 const curr = (U32)(ip-base); + U32 const curr = (U32)(ip-base); const BYTE* const dictBase = dms->window.base; const BYTE* const dictEnd = dms->window.nextSrc; U32 const dictHighLimit = (U32)(dms->window.nextSrc - dms->window.base); @@ -195,7 +195,7 @@ ZSTD_DUBT_findBetterDictMatch ( if (matchLength > bestLength) { U32 matchIndex = dictMatchIndex + dictIndexDelta; - if ( (4*(int)(matchLength-bestLength)) > (int)(ZSTD_highbit32(curr-matchIndex+1) - ZSTD_highbit32((U32)offsetPtr[0]+1)) ) { + if ( (4*(int)(matchLength-bestLength)) > (int)(ZSTD_highbit32(curr-matchIndex+1) - ZSTD_highbit32((U32)offsetPtr[0]+1)) ) { DEBUGLOG(9, "ZSTD_DUBT_findBetterDictMatch(%u) : found better match length %u -> %u and offsetCode %u -> %u (dictMatchIndex %u, matchIndex %u)", curr, (U32)bestLength, (U32)matchLength, (U32)*offsetPtr, STORE_OFFSET(curr - matchIndex), dictMatchIndex, matchIndex); bestLength = matchLength, *offsetPtr = STORE_OFFSET(curr - matchIndex); @@ -220,7 +220,7 @@ ZSTD_DUBT_findBetterDictMatch ( if (bestLength >= MINMATCH) { U32 const mIndex = curr - (U32)STORED_OFFSET(*offsetPtr); (void)mIndex; DEBUGLOG(8, "ZSTD_DUBT_findBetterDictMatch(%u) : found match of length %u and offsetCode %u (pos %u)", - curr, (U32)bestLength, (U32)*offsetPtr, mIndex); + curr, (U32)bestLength, (U32)*offsetPtr, mIndex); } return bestLength; @@ -241,13 +241,13 @@ ZSTD_DUBT_findBestMatch(ZSTD_matchState_t* ms, U32 matchIndex = hashTable[h]; const BYTE* const base = ms->window.base; - U32 const curr = (U32)(ip-base); - U32 const windowLow = ZSTD_getLowestMatchIndex(ms, curr, cParams->windowLog); + U32 const curr = (U32)(ip-base); + U32 const windowLow = ZSTD_getLowestMatchIndex(ms, curr, cParams->windowLog); U32* const bt = ms->chainTable; U32 const btLog = cParams->chainLog - 1; U32 const btMask = (1 << btLog) - 1; - U32 const btLow = (btMask >= curr) ? 0 : curr - btMask; + U32 const btLow = (btMask >= curr) ? 0 : curr - btMask; U32 const unsortLimit = MAX(btLow, windowLow); U32* nextCandidate = bt + 2*(matchIndex&btMask); @@ -256,9 +256,9 @@ ZSTD_DUBT_findBestMatch(ZSTD_matchState_t* ms, U32 nbCandidates = nbCompares; U32 previousCandidate = 0; - DEBUGLOG(7, "ZSTD_DUBT_findBestMatch (%u) ", curr); + DEBUGLOG(7, "ZSTD_DUBT_findBestMatch (%u) ", curr); assert(ip <= iend-8); /* required for h calculation */ - assert(dictMode != ZSTD_dedicatedDictSearch); + assert(dictMode != ZSTD_dedicatedDictSearch); /* reach end of unsorted candidates list */ while ( (matchIndex > unsortLimit) @@ -300,14 +300,14 @@ ZSTD_DUBT_findBestMatch(ZSTD_matchState_t* ms, const U32 dictLimit = ms->window.dictLimit; const BYTE* const dictEnd = dictBase + dictLimit; const BYTE* const prefixStart = base + dictLimit; - U32* smallerPtr = bt + 2*(curr&btMask); - U32* largerPtr = bt + 2*(curr&btMask) + 1; - U32 matchEndIdx = curr + 8 + 1; + U32* smallerPtr = bt + 2*(curr&btMask); + U32* largerPtr = bt + 2*(curr&btMask) + 1; + U32 matchEndIdx = curr + 8 + 1; U32 dummy32; /* to be nullified at the end */ size_t bestLength = 0; matchIndex = hashTable[h]; - hashTable[h] = curr; /* Update Hash Table */ + hashTable[h] = curr; /* Update Hash Table */ for (; nbCompares && (matchIndex > windowLow); --nbCompares) { U32* const nextPtr = bt + 2*(matchIndex & btMask); @@ -327,7 +327,7 @@ ZSTD_DUBT_findBestMatch(ZSTD_matchState_t* ms, if (matchLength > bestLength) { if (matchLength > matchEndIdx - matchIndex) matchEndIdx = matchIndex + (U32)matchLength; - if ( (4*(int)(matchLength-bestLength)) > (int)(ZSTD_highbit32(curr-matchIndex+1) - ZSTD_highbit32((U32)offsetPtr[0]+1)) ) + if ( (4*(int)(matchLength-bestLength)) > (int)(ZSTD_highbit32(curr-matchIndex+1) - ZSTD_highbit32((U32)offsetPtr[0]+1)) ) bestLength = matchLength, *offsetPtr = STORE_OFFSET(curr - matchIndex); if (ip+matchLength == iend) { /* equal : no way to know if inf or sup */ if (dictMode == ZSTD_dictMatchState) { @@ -365,12 +365,12 @@ ZSTD_DUBT_findBestMatch(ZSTD_matchState_t* ms, mls, dictMode); } - assert(matchEndIdx > curr+8); /* ensure nextToUpdate is increased */ + assert(matchEndIdx > curr+8); /* ensure nextToUpdate is increased */ ms->nextToUpdate = matchEndIdx - 8; /* skip repetitive patterns */ if (bestLength >= MINMATCH) { U32 const mIndex = curr - (U32)STORED_OFFSET(*offsetPtr); (void)mIndex; DEBUGLOG(8, "ZSTD_DUBT_findBestMatch(%u) : found match of length %u and offsetCode %u (pos %u)", - curr, (U32)bestLength, (U32)*offsetPtr, mIndex); + curr, (U32)bestLength, (U32)*offsetPtr, mIndex); } return bestLength; } @@ -391,258 +391,258 @@ ZSTD_BtFindBestMatch( ZSTD_matchState_t* ms, return ZSTD_DUBT_findBestMatch(ms, ip, iLimit, offsetPtr, mls, dictMode); } -/*********************************** -* Dedicated dict search +/*********************************** +* Dedicated dict search ***********************************/ -void ZSTD_dedicatedDictSearch_lazy_loadDictionary(ZSTD_matchState_t* ms, const BYTE* const ip) -{ - const BYTE* const base = ms->window.base; - U32 const target = (U32)(ip - base); - U32* const hashTable = ms->hashTable; - U32* const chainTable = ms->chainTable; - U32 const chainSize = 1 << ms->cParams.chainLog; - U32 idx = ms->nextToUpdate; +void ZSTD_dedicatedDictSearch_lazy_loadDictionary(ZSTD_matchState_t* ms, const BYTE* const ip) +{ + const BYTE* const base = ms->window.base; + U32 const target = (U32)(ip - base); + U32* const hashTable = ms->hashTable; + U32* const chainTable = ms->chainTable; + U32 const chainSize = 1 << ms->cParams.chainLog; + U32 idx = ms->nextToUpdate; U32 const minChain = chainSize < target - idx ? target - chainSize : idx; - U32 const bucketSize = 1 << ZSTD_LAZY_DDSS_BUCKET_LOG; - U32 const cacheSize = bucketSize - 1; - U32 const chainAttempts = (1 << ms->cParams.searchLog) - cacheSize; - U32 const chainLimit = chainAttempts > 255 ? 255 : chainAttempts; - - /* We know the hashtable is oversized by a factor of `bucketSize`. - * We are going to temporarily pretend `bucketSize == 1`, keeping only a - * single entry. We will use the rest of the space to construct a temporary - * chaintable. - */ - U32 const hashLog = ms->cParams.hashLog - ZSTD_LAZY_DDSS_BUCKET_LOG; - U32* const tmpHashTable = hashTable; - U32* const tmpChainTable = hashTable + ((size_t)1 << hashLog); + U32 const bucketSize = 1 << ZSTD_LAZY_DDSS_BUCKET_LOG; + U32 const cacheSize = bucketSize - 1; + U32 const chainAttempts = (1 << ms->cParams.searchLog) - cacheSize; + U32 const chainLimit = chainAttempts > 255 ? 255 : chainAttempts; + + /* We know the hashtable is oversized by a factor of `bucketSize`. + * We are going to temporarily pretend `bucketSize == 1`, keeping only a + * single entry. We will use the rest of the space to construct a temporary + * chaintable. + */ + U32 const hashLog = ms->cParams.hashLog - ZSTD_LAZY_DDSS_BUCKET_LOG; + U32* const tmpHashTable = hashTable; + U32* const tmpChainTable = hashTable + ((size_t)1 << hashLog); U32 const tmpChainSize = (U32)((1 << ZSTD_LAZY_DDSS_BUCKET_LOG) - 1) << hashLog; - U32 const tmpMinChain = tmpChainSize < target ? target - tmpChainSize : idx; - U32 hashIdx; - - assert(ms->cParams.chainLog <= 24); - assert(ms->cParams.hashLog > ms->cParams.chainLog); - assert(idx != 0); - assert(tmpMinChain <= minChain); - - /* fill conventional hash table and conventional chain table */ - for ( ; idx < target; idx++) { - U32 const h = (U32)ZSTD_hashPtr(base + idx, hashLog, ms->cParams.minMatch); - if (idx >= tmpMinChain) { - tmpChainTable[idx - tmpMinChain] = hashTable[h]; - } - tmpHashTable[h] = idx; - } - - /* sort chains into ddss chain table */ - { - U32 chainPos = 0; - for (hashIdx = 0; hashIdx < (1U << hashLog); hashIdx++) { - U32 count; - U32 countBeyondMinChain = 0; - U32 i = tmpHashTable[hashIdx]; - for (count = 0; i >= tmpMinChain && count < cacheSize; count++) { - /* skip through the chain to the first position that won't be - * in the hash cache bucket */ - if (i < minChain) { - countBeyondMinChain++; - } - i = tmpChainTable[i - tmpMinChain]; - } - if (count == cacheSize) { - for (count = 0; count < chainLimit;) { - if (i < minChain) { - if (!i || ++countBeyondMinChain > cacheSize) { - /* only allow pulling `cacheSize` number of entries - * into the cache or chainTable beyond `minChain`, - * to replace the entries pulled out of the - * chainTable into the cache. This lets us reach - * back further without increasing the total number - * of entries in the chainTable, guaranteeing the - * DDSS chain table will fit into the space - * allocated for the regular one. */ - break; - } - } - chainTable[chainPos++] = i; - count++; - if (i < tmpMinChain) { - break; - } - i = tmpChainTable[i - tmpMinChain]; - } - } else { - count = 0; - } - if (count) { - tmpHashTable[hashIdx] = ((chainPos - count) << 8) + count; - } else { - tmpHashTable[hashIdx] = 0; - } - } - assert(chainPos <= chainSize); /* I believe this is guaranteed... */ - } - - /* move chain pointers into the last entry of each hash bucket */ - for (hashIdx = (1 << hashLog); hashIdx; ) { - U32 const bucketIdx = --hashIdx << ZSTD_LAZY_DDSS_BUCKET_LOG; - U32 const chainPackedPointer = tmpHashTable[hashIdx]; - U32 i; - for (i = 0; i < cacheSize; i++) { - hashTable[bucketIdx + i] = 0; - } - hashTable[bucketIdx + bucketSize - 1] = chainPackedPointer; - } - - /* fill the buckets of the hash table */ - for (idx = ms->nextToUpdate; idx < target; idx++) { - U32 const h = (U32)ZSTD_hashPtr(base + idx, hashLog, ms->cParams.minMatch) - << ZSTD_LAZY_DDSS_BUCKET_LOG; - U32 i; - /* Shift hash cache down 1. */ - for (i = cacheSize - 1; i; i--) - hashTable[h + i] = hashTable[h + i - 1]; - hashTable[h] = idx; - } - - ms->nextToUpdate = target; -} - -/* Returns the longest match length found in the dedicated dict search structure. - * If none are longer than the argument ml, then ml will be returned. - */ -FORCE_INLINE_TEMPLATE -size_t ZSTD_dedicatedDictSearch_lazy_search(size_t* offsetPtr, size_t ml, U32 nbAttempts, - const ZSTD_matchState_t* const dms, - const BYTE* const ip, const BYTE* const iLimit, - const BYTE* const prefixStart, const U32 curr, - const U32 dictLimit, const size_t ddsIdx) { - const U32 ddsLowestIndex = dms->window.dictLimit; - const BYTE* const ddsBase = dms->window.base; - const BYTE* const ddsEnd = dms->window.nextSrc; - const U32 ddsSize = (U32)(ddsEnd - ddsBase); - const U32 ddsIndexDelta = dictLimit - ddsSize; - const U32 bucketSize = (1 << ZSTD_LAZY_DDSS_BUCKET_LOG); - const U32 bucketLimit = nbAttempts < bucketSize - 1 ? nbAttempts : bucketSize - 1; - U32 ddsAttempt; - U32 matchIndex; - - for (ddsAttempt = 0; ddsAttempt < bucketSize - 1; ddsAttempt++) { - PREFETCH_L1(ddsBase + dms->hashTable[ddsIdx + ddsAttempt]); - } - - { - U32 const chainPackedPointer = dms->hashTable[ddsIdx + bucketSize - 1]; - U32 const chainIndex = chainPackedPointer >> 8; - - PREFETCH_L1(&dms->chainTable[chainIndex]); - } - - for (ddsAttempt = 0; ddsAttempt < bucketLimit; ddsAttempt++) { - size_t currentMl=0; - const BYTE* match; - matchIndex = dms->hashTable[ddsIdx + ddsAttempt]; - match = ddsBase + matchIndex; - - if (!matchIndex) { - return ml; - } - - /* guaranteed by table construction */ - (void)ddsLowestIndex; - assert(matchIndex >= ddsLowestIndex); - assert(match+4 <= ddsEnd); - if (MEM_read32(match) == MEM_read32(ip)) { - /* assumption : matchIndex <= dictLimit-4 (by table construction) */ - currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, ddsEnd, prefixStart) + 4; - } - - /* save best solution */ - if (currentMl > ml) { - ml = currentMl; + U32 const tmpMinChain = tmpChainSize < target ? target - tmpChainSize : idx; + U32 hashIdx; + + assert(ms->cParams.chainLog <= 24); + assert(ms->cParams.hashLog > ms->cParams.chainLog); + assert(idx != 0); + assert(tmpMinChain <= minChain); + + /* fill conventional hash table and conventional chain table */ + for ( ; idx < target; idx++) { + U32 const h = (U32)ZSTD_hashPtr(base + idx, hashLog, ms->cParams.minMatch); + if (idx >= tmpMinChain) { + tmpChainTable[idx - tmpMinChain] = hashTable[h]; + } + tmpHashTable[h] = idx; + } + + /* sort chains into ddss chain table */ + { + U32 chainPos = 0; + for (hashIdx = 0; hashIdx < (1U << hashLog); hashIdx++) { + U32 count; + U32 countBeyondMinChain = 0; + U32 i = tmpHashTable[hashIdx]; + for (count = 0; i >= tmpMinChain && count < cacheSize; count++) { + /* skip through the chain to the first position that won't be + * in the hash cache bucket */ + if (i < minChain) { + countBeyondMinChain++; + } + i = tmpChainTable[i - tmpMinChain]; + } + if (count == cacheSize) { + for (count = 0; count < chainLimit;) { + if (i < minChain) { + if (!i || ++countBeyondMinChain > cacheSize) { + /* only allow pulling `cacheSize` number of entries + * into the cache or chainTable beyond `minChain`, + * to replace the entries pulled out of the + * chainTable into the cache. This lets us reach + * back further without increasing the total number + * of entries in the chainTable, guaranteeing the + * DDSS chain table will fit into the space + * allocated for the regular one. */ + break; + } + } + chainTable[chainPos++] = i; + count++; + if (i < tmpMinChain) { + break; + } + i = tmpChainTable[i - tmpMinChain]; + } + } else { + count = 0; + } + if (count) { + tmpHashTable[hashIdx] = ((chainPos - count) << 8) + count; + } else { + tmpHashTable[hashIdx] = 0; + } + } + assert(chainPos <= chainSize); /* I believe this is guaranteed... */ + } + + /* move chain pointers into the last entry of each hash bucket */ + for (hashIdx = (1 << hashLog); hashIdx; ) { + U32 const bucketIdx = --hashIdx << ZSTD_LAZY_DDSS_BUCKET_LOG; + U32 const chainPackedPointer = tmpHashTable[hashIdx]; + U32 i; + for (i = 0; i < cacheSize; i++) { + hashTable[bucketIdx + i] = 0; + } + hashTable[bucketIdx + bucketSize - 1] = chainPackedPointer; + } + + /* fill the buckets of the hash table */ + for (idx = ms->nextToUpdate; idx < target; idx++) { + U32 const h = (U32)ZSTD_hashPtr(base + idx, hashLog, ms->cParams.minMatch) + << ZSTD_LAZY_DDSS_BUCKET_LOG; + U32 i; + /* Shift hash cache down 1. */ + for (i = cacheSize - 1; i; i--) + hashTable[h + i] = hashTable[h + i - 1]; + hashTable[h] = idx; + } + + ms->nextToUpdate = target; +} + +/* Returns the longest match length found in the dedicated dict search structure. + * If none are longer than the argument ml, then ml will be returned. + */ +FORCE_INLINE_TEMPLATE +size_t ZSTD_dedicatedDictSearch_lazy_search(size_t* offsetPtr, size_t ml, U32 nbAttempts, + const ZSTD_matchState_t* const dms, + const BYTE* const ip, const BYTE* const iLimit, + const BYTE* const prefixStart, const U32 curr, + const U32 dictLimit, const size_t ddsIdx) { + const U32 ddsLowestIndex = dms->window.dictLimit; + const BYTE* const ddsBase = dms->window.base; + const BYTE* const ddsEnd = dms->window.nextSrc; + const U32 ddsSize = (U32)(ddsEnd - ddsBase); + const U32 ddsIndexDelta = dictLimit - ddsSize; + const U32 bucketSize = (1 << ZSTD_LAZY_DDSS_BUCKET_LOG); + const U32 bucketLimit = nbAttempts < bucketSize - 1 ? nbAttempts : bucketSize - 1; + U32 ddsAttempt; + U32 matchIndex; + + for (ddsAttempt = 0; ddsAttempt < bucketSize - 1; ddsAttempt++) { + PREFETCH_L1(ddsBase + dms->hashTable[ddsIdx + ddsAttempt]); + } + + { + U32 const chainPackedPointer = dms->hashTable[ddsIdx + bucketSize - 1]; + U32 const chainIndex = chainPackedPointer >> 8; + + PREFETCH_L1(&dms->chainTable[chainIndex]); + } + + for (ddsAttempt = 0; ddsAttempt < bucketLimit; ddsAttempt++) { + size_t currentMl=0; + const BYTE* match; + matchIndex = dms->hashTable[ddsIdx + ddsAttempt]; + match = ddsBase + matchIndex; + + if (!matchIndex) { + return ml; + } + + /* guaranteed by table construction */ + (void)ddsLowestIndex; + assert(matchIndex >= ddsLowestIndex); + assert(match+4 <= ddsEnd); + if (MEM_read32(match) == MEM_read32(ip)) { + /* assumption : matchIndex <= dictLimit-4 (by table construction) */ + currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, ddsEnd, prefixStart) + 4; + } + + /* save best solution */ + if (currentMl > ml) { + ml = currentMl; *offsetPtr = STORE_OFFSET(curr - (matchIndex + ddsIndexDelta)); - if (ip+currentMl == iLimit) { - /* best possible, avoids read overflow on next attempt */ - return ml; - } - } - } - - { - U32 const chainPackedPointer = dms->hashTable[ddsIdx + bucketSize - 1]; - U32 chainIndex = chainPackedPointer >> 8; - U32 const chainLength = chainPackedPointer & 0xFF; - U32 const chainAttempts = nbAttempts - ddsAttempt; - U32 const chainLimit = chainAttempts > chainLength ? chainLength : chainAttempts; - U32 chainAttempt; - - for (chainAttempt = 0 ; chainAttempt < chainLimit; chainAttempt++) { - PREFETCH_L1(ddsBase + dms->chainTable[chainIndex + chainAttempt]); - } - - for (chainAttempt = 0 ; chainAttempt < chainLimit; chainAttempt++, chainIndex++) { - size_t currentMl=0; - const BYTE* match; - matchIndex = dms->chainTable[chainIndex]; - match = ddsBase + matchIndex; - - /* guaranteed by table construction */ - assert(matchIndex >= ddsLowestIndex); - assert(match+4 <= ddsEnd); - if (MEM_read32(match) == MEM_read32(ip)) { - /* assumption : matchIndex <= dictLimit-4 (by table construction) */ - currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, ddsEnd, prefixStart) + 4; - } - - /* save best solution */ - if (currentMl > ml) { - ml = currentMl; + if (ip+currentMl == iLimit) { + /* best possible, avoids read overflow on next attempt */ + return ml; + } + } + } + + { + U32 const chainPackedPointer = dms->hashTable[ddsIdx + bucketSize - 1]; + U32 chainIndex = chainPackedPointer >> 8; + U32 const chainLength = chainPackedPointer & 0xFF; + U32 const chainAttempts = nbAttempts - ddsAttempt; + U32 const chainLimit = chainAttempts > chainLength ? chainLength : chainAttempts; + U32 chainAttempt; + + for (chainAttempt = 0 ; chainAttempt < chainLimit; chainAttempt++) { + PREFETCH_L1(ddsBase + dms->chainTable[chainIndex + chainAttempt]); + } + + for (chainAttempt = 0 ; chainAttempt < chainLimit; chainAttempt++, chainIndex++) { + size_t currentMl=0; + const BYTE* match; + matchIndex = dms->chainTable[chainIndex]; + match = ddsBase + matchIndex; + + /* guaranteed by table construction */ + assert(matchIndex >= ddsLowestIndex); + assert(match+4 <= ddsEnd); + if (MEM_read32(match) == MEM_read32(ip)) { + /* assumption : matchIndex <= dictLimit-4 (by table construction) */ + currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, ddsEnd, prefixStart) + 4; + } + + /* save best solution */ + if (currentMl > ml) { + ml = currentMl; *offsetPtr = STORE_OFFSET(curr - (matchIndex + ddsIndexDelta)); - if (ip+currentMl == iLimit) break; /* best possible, avoids read overflow on next attempt */ - } - } - } - return ml; -} - - -/* ********************************* -* Hash Chain -***********************************/ -#define NEXT_IN_CHAIN(d, mask) chainTable[(d) & (mask)] - -/* Update chains up to ip (excluded) - Assumption : always within prefix (i.e. not within extDict) */ -FORCE_INLINE_TEMPLATE U32 ZSTD_insertAndFindFirstIndex_internal( - ZSTD_matchState_t* ms, - const ZSTD_compressionParameters* const cParams, - const BYTE* ip, U32 const mls) -{ - U32* const hashTable = ms->hashTable; - const U32 hashLog = cParams->hashLog; - U32* const chainTable = ms->chainTable; - const U32 chainMask = (1 << cParams->chainLog) - 1; - const BYTE* const base = ms->window.base; - const U32 target = (U32)(ip - base); - U32 idx = ms->nextToUpdate; - - while(idx < target) { /* catch up */ - size_t const h = ZSTD_hashPtr(base+idx, hashLog, mls); - NEXT_IN_CHAIN(idx, chainMask) = hashTable[h]; - hashTable[h] = idx; - idx++; - } - - ms->nextToUpdate = target; - return hashTable[ZSTD_hashPtr(ip, hashLog, mls)]; -} - -U32 ZSTD_insertAndFindFirstIndex(ZSTD_matchState_t* ms, const BYTE* ip) { - const ZSTD_compressionParameters* const cParams = &ms->cParams; - return ZSTD_insertAndFindFirstIndex_internal(ms, cParams, ip, ms->cParams.minMatch); -} - + if (ip+currentMl == iLimit) break; /* best possible, avoids read overflow on next attempt */ + } + } + } + return ml; +} + + +/* ********************************* +* Hash Chain +***********************************/ +#define NEXT_IN_CHAIN(d, mask) chainTable[(d) & (mask)] + +/* Update chains up to ip (excluded) + Assumption : always within prefix (i.e. not within extDict) */ +FORCE_INLINE_TEMPLATE U32 ZSTD_insertAndFindFirstIndex_internal( + ZSTD_matchState_t* ms, + const ZSTD_compressionParameters* const cParams, + const BYTE* ip, U32 const mls) +{ + U32* const hashTable = ms->hashTable; + const U32 hashLog = cParams->hashLog; + U32* const chainTable = ms->chainTable; + const U32 chainMask = (1 << cParams->chainLog) - 1; + const BYTE* const base = ms->window.base; + const U32 target = (U32)(ip - base); + U32 idx = ms->nextToUpdate; + + while(idx < target) { /* catch up */ + size_t const h = ZSTD_hashPtr(base+idx, hashLog, mls); + NEXT_IN_CHAIN(idx, chainMask) = hashTable[h]; + hashTable[h] = idx; + idx++; + } + + ms->nextToUpdate = target; + return hashTable[ZSTD_hashPtr(ip, hashLog, mls)]; +} + +U32 ZSTD_insertAndFindFirstIndex(ZSTD_matchState_t* ms, const BYTE* ip) { + const ZSTD_compressionParameters* const cParams = &ms->cParams; + return ZSTD_insertAndFindFirstIndex_internal(ms, cParams, ip, ms->cParams.minMatch); +} + /* inlining is important to hardwire a hot branch (template emulation) */ FORCE_INLINE_TEMPLATE size_t ZSTD_HcFindBestMatch( @@ -660,33 +660,33 @@ size_t ZSTD_HcFindBestMatch( const U32 dictLimit = ms->window.dictLimit; const BYTE* const prefixStart = base + dictLimit; const BYTE* const dictEnd = dictBase + dictLimit; - const U32 curr = (U32)(ip-base); + const U32 curr = (U32)(ip-base); const U32 maxDistance = 1U << cParams->windowLog; const U32 lowestValid = ms->window.lowLimit; - const U32 withinMaxDistance = (curr - lowestValid > maxDistance) ? curr - maxDistance : lowestValid; + const U32 withinMaxDistance = (curr - lowestValid > maxDistance) ? curr - maxDistance : lowestValid; const U32 isDictionary = (ms->loadedDictEnd != 0); const U32 lowLimit = isDictionary ? lowestValid : withinMaxDistance; - const U32 minChain = curr > chainSize ? curr - chainSize : 0; + const U32 minChain = curr > chainSize ? curr - chainSize : 0; U32 nbAttempts = 1U << cParams->searchLog; size_t ml=4-1; - const ZSTD_matchState_t* const dms = ms->dictMatchState; - const U32 ddsHashLog = dictMode == ZSTD_dedicatedDictSearch - ? dms->cParams.hashLog - ZSTD_LAZY_DDSS_BUCKET_LOG : 0; - const size_t ddsIdx = dictMode == ZSTD_dedicatedDictSearch - ? ZSTD_hashPtr(ip, ddsHashLog, mls) << ZSTD_LAZY_DDSS_BUCKET_LOG : 0; - - U32 matchIndex; - - if (dictMode == ZSTD_dedicatedDictSearch) { - const U32* entry = &dms->hashTable[ddsIdx]; - PREFETCH_L1(entry); - } - + const ZSTD_matchState_t* const dms = ms->dictMatchState; + const U32 ddsHashLog = dictMode == ZSTD_dedicatedDictSearch + ? dms->cParams.hashLog - ZSTD_LAZY_DDSS_BUCKET_LOG : 0; + const size_t ddsIdx = dictMode == ZSTD_dedicatedDictSearch + ? ZSTD_hashPtr(ip, ddsHashLog, mls) << ZSTD_LAZY_DDSS_BUCKET_LOG : 0; + + U32 matchIndex; + + if (dictMode == ZSTD_dedicatedDictSearch) { + const U32* entry = &dms->hashTable[ddsIdx]; + PREFETCH_L1(entry); + } + /* HC4 match finder */ - matchIndex = ZSTD_insertAndFindFirstIndex_internal(ms, cParams, ip, mls); + matchIndex = ZSTD_insertAndFindFirstIndex_internal(ms, cParams, ip, mls); - for ( ; (matchIndex>=lowLimit) & (nbAttempts>0) ; nbAttempts--) { + for ( ; (matchIndex>=lowLimit) & (nbAttempts>0) ; nbAttempts--) { size_t currentMl=0; if ((dictMode != ZSTD_extDict) || matchIndex >= dictLimit) { const BYTE* const match = base + matchIndex; @@ -712,10 +712,10 @@ size_t ZSTD_HcFindBestMatch( } assert(nbAttempts <= (1U << ZSTD_SEARCHLOG_MAX)); /* Check we haven't underflowed. */ - if (dictMode == ZSTD_dedicatedDictSearch) { - ml = ZSTD_dedicatedDictSearch_lazy_search(offsetPtr, ml, nbAttempts, dms, - ip, iLimit, prefixStart, curr, dictLimit, ddsIdx); - } else if (dictMode == ZSTD_dictMatchState) { + if (dictMode == ZSTD_dedicatedDictSearch) { + ml = ZSTD_dedicatedDictSearch_lazy_search(offsetPtr, ml, nbAttempts, dms, + ip, iLimit, prefixStart, curr, dictLimit, ddsIdx); + } else if (dictMode == ZSTD_dictMatchState) { const U32* const dmsChainTable = dms->chainTable; const U32 dmsChainSize = (1 << dms->cParams.chainLog); const U32 dmsChainMask = dmsChainSize - 1; @@ -728,7 +728,7 @@ size_t ZSTD_HcFindBestMatch( matchIndex = dms->hashTable[ZSTD_hashPtr(ip, dms->cParams.hashLog, mls)]; - for ( ; (matchIndex>=dmsLowestIndex) & (nbAttempts>0) ; nbAttempts--) { + for ( ; (matchIndex>=dmsLowestIndex) & (nbAttempts>0) ; nbAttempts--) { size_t currentMl=0; const BYTE* const match = dmsBase + matchIndex; assert(match+4 <= dmsEnd); @@ -744,7 +744,7 @@ size_t ZSTD_HcFindBestMatch( } if (matchIndex <= dmsMinChain) break; - + matchIndex = dmsChainTable[matchIndex & dmsChainMask]; } } @@ -752,24 +752,24 @@ size_t ZSTD_HcFindBestMatch( return ml; } -/* ********************************* -* (SIMD) Row-based matchfinder -***********************************/ -/* Constants for row-based hash */ +/* ********************************* +* (SIMD) Row-based matchfinder +***********************************/ +/* Constants for row-based hash */ #define ZSTD_ROW_HASH_TAG_OFFSET 16 /* byte offset of hashes in the match state's tagTable from the beginning of a row */ #define ZSTD_ROW_HASH_TAG_BITS 8 /* nb bits to use for the tag */ -#define ZSTD_ROW_HASH_TAG_MASK ((1u << ZSTD_ROW_HASH_TAG_BITS) - 1) +#define ZSTD_ROW_HASH_TAG_MASK ((1u << ZSTD_ROW_HASH_TAG_BITS) - 1) #define ZSTD_ROW_HASH_MAX_ENTRIES 64 /* absolute maximum number of entries per row, for all configurations */ -#define ZSTD_ROW_HASH_CACHE_MASK (ZSTD_ROW_HASH_CACHE_SIZE - 1) - +#define ZSTD_ROW_HASH_CACHE_MASK (ZSTD_ROW_HASH_CACHE_SIZE - 1) + typedef U64 ZSTD_VecMask; /* Clarifies when we are interacting with a U64 representing a mask of matches */ - -/* ZSTD_VecMask_next(): - * Starting from the LSB, returns the idx of the next non-zero bit. - * Basically counting the nb of trailing zeroes. - */ -static U32 ZSTD_VecMask_next(ZSTD_VecMask val) { + +/* ZSTD_VecMask_next(): + * Starting from the LSB, returns the idx of the next non-zero bit. + * Basically counting the nb of trailing zeroes. + */ +static U32 ZSTD_VecMask_next(ZSTD_VecMask val) { assert(val != 0); # if defined(_MSC_VER) && defined(_WIN64) if (val != 0) { @@ -792,7 +792,7 @@ static U32 ZSTD_VecMask_next(ZSTD_VecMask val) { } else { return (U32)__builtin_ctzll(val); } -# else +# else /* Software ctz version: http://aggregate.org/MAGIC/#Trailing%20Zero%20Count * and: https://stackoverflow.com/questions/2709430/count-number-of-bits-in-a-64-bit-long-big-integer */ @@ -800,20 +800,20 @@ static U32 ZSTD_VecMask_next(ZSTD_VecMask val) { val = val - ((val >> 1) & 0x5555555555555555); val = (val & 0x3333333333333333ULL) + ((val >> 2) & 0x3333333333333333ULL); return (U32)((((val + (val >> 4)) & 0xF0F0F0F0F0F0F0FULL) * 0x101010101010101ULL) >> 56); -# endif -} - +# endif +} + /* ZSTD_rotateRight_*(): * Rotates a bitfield to the right by "count" bits. * https://en.wikipedia.org/w/index.php?title=Circular_shift&oldid=991635599#Implementing_circular_shifts - */ + */ FORCE_INLINE_TEMPLATE U64 ZSTD_rotateRight_U64(U64 const value, U32 count) { assert(count < 64); count &= 0x3F; /* for fickle pattern recognition */ return (value >> count) | (U64)(value << ((0U - count) & 0x3F)); -} - +} + FORCE_INLINE_TEMPLATE U32 ZSTD_rotateRight_U32(U32 const value, U32 count) { assert(count < 32); @@ -828,113 +828,113 @@ U16 ZSTD_rotateRight_U16(U16 const value, U32 count) { return (value >> count) | (U16)(value << ((0U - count) & 0x0F)); } -/* ZSTD_row_nextIndex(): - * Returns the next index to insert at within a tagTable row, and updates the "head" - * value to reflect the update. Essentially cycles backwards from [0, {entries per row}) - */ -FORCE_INLINE_TEMPLATE U32 ZSTD_row_nextIndex(BYTE* const tagRow, U32 const rowMask) { - U32 const next = (*tagRow - 1) & rowMask; - *tagRow = (BYTE)next; - return next; -} - -/* ZSTD_isAligned(): - * Checks that a pointer is aligned to "align" bytes which must be a power of 2. - */ -MEM_STATIC int ZSTD_isAligned(void const* ptr, size_t align) { - assert((align & (align - 1)) == 0); - return (((size_t)ptr) & (align - 1)) == 0; -} - -/* ZSTD_row_prefetch(): - * Performs prefetching for the hashTable and tagTable at a given row. - */ -FORCE_INLINE_TEMPLATE void ZSTD_row_prefetch(U32 const* hashTable, U16 const* tagTable, U32 const relRow, U32 const rowLog) { - PREFETCH_L1(hashTable + relRow); +/* ZSTD_row_nextIndex(): + * Returns the next index to insert at within a tagTable row, and updates the "head" + * value to reflect the update. Essentially cycles backwards from [0, {entries per row}) + */ +FORCE_INLINE_TEMPLATE U32 ZSTD_row_nextIndex(BYTE* const tagRow, U32 const rowMask) { + U32 const next = (*tagRow - 1) & rowMask; + *tagRow = (BYTE)next; + return next; +} + +/* ZSTD_isAligned(): + * Checks that a pointer is aligned to "align" bytes which must be a power of 2. + */ +MEM_STATIC int ZSTD_isAligned(void const* ptr, size_t align) { + assert((align & (align - 1)) == 0); + return (((size_t)ptr) & (align - 1)) == 0; +} + +/* ZSTD_row_prefetch(): + * Performs prefetching for the hashTable and tagTable at a given row. + */ +FORCE_INLINE_TEMPLATE void ZSTD_row_prefetch(U32 const* hashTable, U16 const* tagTable, U32 const relRow, U32 const rowLog) { + PREFETCH_L1(hashTable + relRow); if (rowLog >= 5) { - PREFETCH_L1(hashTable + relRow + 16); + PREFETCH_L1(hashTable + relRow + 16); /* Note: prefetching more of the hash table does not appear to be beneficial for 128-entry rows */ - } - PREFETCH_L1(tagTable + relRow); + } + PREFETCH_L1(tagTable + relRow); if (rowLog == 6) { PREFETCH_L1(tagTable + relRow + 32); } assert(rowLog == 4 || rowLog == 5 || rowLog == 6); - assert(ZSTD_isAligned(hashTable + relRow, 64)); /* prefetched hash row always 64-byte aligned */ + assert(ZSTD_isAligned(hashTable + relRow, 64)); /* prefetched hash row always 64-byte aligned */ assert(ZSTD_isAligned(tagTable + relRow, (size_t)1 << rowLog)); /* prefetched tagRow sits on correct multiple of bytes (32,64,128) */ -} - -/* ZSTD_row_fillHashCache(): - * Fill up the hash cache starting at idx, prefetching up to ZSTD_ROW_HASH_CACHE_SIZE entries, - * but not beyond iLimit. - */ +} + +/* ZSTD_row_fillHashCache(): + * Fill up the hash cache starting at idx, prefetching up to ZSTD_ROW_HASH_CACHE_SIZE entries, + * but not beyond iLimit. + */ FORCE_INLINE_TEMPLATE void ZSTD_row_fillHashCache(ZSTD_matchState_t* ms, const BYTE* base, - U32 const rowLog, U32 const mls, - U32 idx, const BYTE* const iLimit) -{ - U32 const* const hashTable = ms->hashTable; - U16 const* const tagTable = ms->tagTable; - U32 const hashLog = ms->rowHashLog; - U32 const maxElemsToPrefetch = (base + idx) > iLimit ? 0 : (U32)(iLimit - (base + idx) + 1); - U32 const lim = idx + MIN(ZSTD_ROW_HASH_CACHE_SIZE, maxElemsToPrefetch); - - for (; idx < lim; ++idx) { - U32 const hash = (U32)ZSTD_hashPtr(base + idx, hashLog + ZSTD_ROW_HASH_TAG_BITS, mls); - U32 const row = (hash >> ZSTD_ROW_HASH_TAG_BITS) << rowLog; - ZSTD_row_prefetch(hashTable, tagTable, row, rowLog); - ms->hashCache[idx & ZSTD_ROW_HASH_CACHE_MASK] = hash; - } - - DEBUGLOG(6, "ZSTD_row_fillHashCache(): [%u %u %u %u %u %u %u %u]", ms->hashCache[0], ms->hashCache[1], - ms->hashCache[2], ms->hashCache[3], ms->hashCache[4], - ms->hashCache[5], ms->hashCache[6], ms->hashCache[7]); -} - -/* ZSTD_row_nextCachedHash(): - * Returns the hash of base + idx, and replaces the hash in the hash cache with the byte at - * base + idx + ZSTD_ROW_HASH_CACHE_SIZE. Also prefetches the appropriate rows from hashTable and tagTable. - */ -FORCE_INLINE_TEMPLATE U32 ZSTD_row_nextCachedHash(U32* cache, U32 const* hashTable, - U16 const* tagTable, BYTE const* base, - U32 idx, U32 const hashLog, - U32 const rowLog, U32 const mls) -{ - U32 const newHash = (U32)ZSTD_hashPtr(base+idx+ZSTD_ROW_HASH_CACHE_SIZE, hashLog + ZSTD_ROW_HASH_TAG_BITS, mls); - U32 const row = (newHash >> ZSTD_ROW_HASH_TAG_BITS) << rowLog; - ZSTD_row_prefetch(hashTable, tagTable, row, rowLog); - { U32 const hash = cache[idx & ZSTD_ROW_HASH_CACHE_MASK]; - cache[idx & ZSTD_ROW_HASH_CACHE_MASK] = newHash; - return hash; - } -} - + U32 const rowLog, U32 const mls, + U32 idx, const BYTE* const iLimit) +{ + U32 const* const hashTable = ms->hashTable; + U16 const* const tagTable = ms->tagTable; + U32 const hashLog = ms->rowHashLog; + U32 const maxElemsToPrefetch = (base + idx) > iLimit ? 0 : (U32)(iLimit - (base + idx) + 1); + U32 const lim = idx + MIN(ZSTD_ROW_HASH_CACHE_SIZE, maxElemsToPrefetch); + + for (; idx < lim; ++idx) { + U32 const hash = (U32)ZSTD_hashPtr(base + idx, hashLog + ZSTD_ROW_HASH_TAG_BITS, mls); + U32 const row = (hash >> ZSTD_ROW_HASH_TAG_BITS) << rowLog; + ZSTD_row_prefetch(hashTable, tagTable, row, rowLog); + ms->hashCache[idx & ZSTD_ROW_HASH_CACHE_MASK] = hash; + } + + DEBUGLOG(6, "ZSTD_row_fillHashCache(): [%u %u %u %u %u %u %u %u]", ms->hashCache[0], ms->hashCache[1], + ms->hashCache[2], ms->hashCache[3], ms->hashCache[4], + ms->hashCache[5], ms->hashCache[6], ms->hashCache[7]); +} + +/* ZSTD_row_nextCachedHash(): + * Returns the hash of base + idx, and replaces the hash in the hash cache with the byte at + * base + idx + ZSTD_ROW_HASH_CACHE_SIZE. Also prefetches the appropriate rows from hashTable and tagTable. + */ +FORCE_INLINE_TEMPLATE U32 ZSTD_row_nextCachedHash(U32* cache, U32 const* hashTable, + U16 const* tagTable, BYTE const* base, + U32 idx, U32 const hashLog, + U32 const rowLog, U32 const mls) +{ + U32 const newHash = (U32)ZSTD_hashPtr(base+idx+ZSTD_ROW_HASH_CACHE_SIZE, hashLog + ZSTD_ROW_HASH_TAG_BITS, mls); + U32 const row = (newHash >> ZSTD_ROW_HASH_TAG_BITS) << rowLog; + ZSTD_row_prefetch(hashTable, tagTable, row, rowLog); + { U32 const hash = cache[idx & ZSTD_ROW_HASH_CACHE_MASK]; + cache[idx & ZSTD_ROW_HASH_CACHE_MASK] = newHash; + return hash; + } +} + /* ZSTD_row_update_internalImpl(): * Updates the hash table with positions starting from updateStartIdx until updateEndIdx. - */ + */ FORCE_INLINE_TEMPLATE void ZSTD_row_update_internalImpl(ZSTD_matchState_t* ms, U32 updateStartIdx, U32 const updateEndIdx, U32 const mls, U32 const rowLog, U32 const rowMask, U32 const useCache) -{ - U32* const hashTable = ms->hashTable; - U16* const tagTable = ms->tagTable; - U32 const hashLog = ms->rowHashLog; - const BYTE* const base = ms->window.base; - +{ + U32* const hashTable = ms->hashTable; + U16* const tagTable = ms->tagTable; + U32 const hashLog = ms->rowHashLog; + const BYTE* const base = ms->window.base; + DEBUGLOG(6, "ZSTD_row_update_internalImpl(): updateStartIdx=%u, updateEndIdx=%u", updateStartIdx, updateEndIdx); for (; updateStartIdx < updateEndIdx; ++updateStartIdx) { U32 const hash = useCache ? ZSTD_row_nextCachedHash(ms->hashCache, hashTable, tagTable, base, updateStartIdx, hashLog, rowLog, mls) : (U32)ZSTD_hashPtr(base + updateStartIdx, hashLog + ZSTD_ROW_HASH_TAG_BITS, mls); - U32 const relRow = (hash >> ZSTD_ROW_HASH_TAG_BITS) << rowLog; - U32* const row = hashTable + relRow; - BYTE* tagRow = (BYTE*)(tagTable + relRow); /* Though tagTable is laid out as a table of U16, each tag is only 1 byte. - Explicit cast allows us to get exact desired position within each row */ - U32 const pos = ZSTD_row_nextIndex(tagRow, rowMask); - + U32 const relRow = (hash >> ZSTD_ROW_HASH_TAG_BITS) << rowLog; + U32* const row = hashTable + relRow; + BYTE* tagRow = (BYTE*)(tagTable + relRow); /* Though tagTable is laid out as a table of U16, each tag is only 1 byte. + Explicit cast allows us to get exact desired position within each row */ + U32 const pos = ZSTD_row_nextIndex(tagRow, rowMask); + assert(hash == ZSTD_hashPtr(base + updateStartIdx, hashLog + ZSTD_ROW_HASH_TAG_BITS, mls)); - ((BYTE*)tagRow)[pos + ZSTD_ROW_HASH_TAG_OFFSET] = hash & ZSTD_ROW_HASH_TAG_MASK; + ((BYTE*)tagRow)[pos + ZSTD_ROW_HASH_TAG_OFFSET] = hash & ZSTD_ROW_HASH_TAG_MASK; row[pos] = updateStartIdx; - } + } } /* ZSTD_row_update_internal(): @@ -967,22 +967,22 @@ FORCE_INLINE_TEMPLATE void ZSTD_row_update_internal(ZSTD_matchState_t* ms, const } assert(target >= idx); ZSTD_row_update_internalImpl(ms, idx, target, mls, rowLog, rowMask, useCache); - ms->nextToUpdate = target; -} - -/* ZSTD_row_update(): - * External wrapper for ZSTD_row_update_internal(). Used for filling the hashtable during dictionary - * processing. - */ -void ZSTD_row_update(ZSTD_matchState_t* const ms, const BYTE* ip) { + ms->nextToUpdate = target; +} + +/* ZSTD_row_update(): + * External wrapper for ZSTD_row_update_internal(). Used for filling the hashtable during dictionary + * processing. + */ +void ZSTD_row_update(ZSTD_matchState_t* const ms, const BYTE* ip) { const U32 rowLog = BOUNDED(4, ms->cParams.searchLog, 6); - const U32 rowMask = (1u << rowLog) - 1; - const U32 mls = MIN(ms->cParams.minMatch, 6 /* mls caps out at 6 */); - - DEBUGLOG(5, "ZSTD_row_update(), rowLog=%u", rowLog); - ZSTD_row_update_internal(ms, ip, mls, rowLog, rowMask, 0 /* dont use cache */); -} - + const U32 rowMask = (1u << rowLog) - 1; + const U32 mls = MIN(ms->cParams.minMatch, 6 /* mls caps out at 6 */); + + DEBUGLOG(5, "ZSTD_row_update(), rowLog=%u", rowLog); + ZSTD_row_update_internal(ms, ip, mls, rowLog, rowMask, 0 /* dont use cache */); +} + #if defined(ZSTD_ARCH_X86_SSE2) FORCE_INLINE_TEMPLATE ZSTD_VecMask ZSTD_row_getSSEMask(int nbChunks, const BYTE* const src, const BYTE tag, const U32 head) @@ -1003,8 +1003,8 @@ ZSTD_row_getSSEMask(int nbChunks, const BYTE* const src, const BYTE tag, const U } #endif -/* Returns a ZSTD_VecMask (U32) that has the nth bit set to 1 if the newly-computed "tag" matches - * the hash at the nth position in a row of the tagTable. +/* Returns a ZSTD_VecMask (U32) that has the nth bit set to 1 if the newly-computed "tag" matches + * the hash at the nth position in a row of the tagTable. * Each row is a circular buffer beginning at the value of "head". So we must rotate the "matches" bitfield * to match up with the actual layout of the entries within the hashTable */ FORCE_INLINE_TEMPLATE ZSTD_VecMask @@ -1064,7 +1064,7 @@ ZSTD_row_getMatchMask(const BYTE* const tagRow, const BYTE tag, const U32 head, const U64 matches = vget_lane_u64(vreinterpret_u64_u8(t4), 0); return ZSTD_rotateRight_U64(matches, head); } - } + } # endif /* ZSTD_ARCH_ARM_NEON */ /* SWAR */ { const size_t chunkSize = sizeof(size_t); @@ -1108,56 +1108,56 @@ ZSTD_row_getMatchMask(const BYTE* const tagRow, const BYTE tag, const U32 head, } } #endif -} - -/* The high-level approach of the SIMD row based match finder is as follows: - * - Figure out where to insert the new entry: - * - Generate a hash from a byte along with an additional 1-byte "short hash". The additional byte is our "tag" - * - The hashTable is effectively split into groups or "rows" of 16 or 32 entries of U32, and the hash determines - * which row to insert into. - * - Determine the correct position within the row to insert the entry into. Each row of 16 or 32 can - * be considered as a circular buffer with a "head" index that resides in the tagTable. - * - Also insert the "tag" into the equivalent row and position in the tagTable. - * - Note: The tagTable has 17 or 33 1-byte entries per row, due to 16 or 32 tags, and 1 "head" entry. - * The 17 or 33 entry rows are spaced out to occur every 32 or 64 bytes, respectively, - * for alignment/performance reasons, leaving some bytes unused. - * - Use SIMD to efficiently compare the tags in the tagTable to the 1-byte "short hash" and - * generate a bitfield that we can cycle through to check the collisions in the hash table. - * - Pick the longest match. - */ -FORCE_INLINE_TEMPLATE +} + +/* The high-level approach of the SIMD row based match finder is as follows: + * - Figure out where to insert the new entry: + * - Generate a hash from a byte along with an additional 1-byte "short hash". The additional byte is our "tag" + * - The hashTable is effectively split into groups or "rows" of 16 or 32 entries of U32, and the hash determines + * which row to insert into. + * - Determine the correct position within the row to insert the entry into. Each row of 16 or 32 can + * be considered as a circular buffer with a "head" index that resides in the tagTable. + * - Also insert the "tag" into the equivalent row and position in the tagTable. + * - Note: The tagTable has 17 or 33 1-byte entries per row, due to 16 or 32 tags, and 1 "head" entry. + * The 17 or 33 entry rows are spaced out to occur every 32 or 64 bytes, respectively, + * for alignment/performance reasons, leaving some bytes unused. + * - Use SIMD to efficiently compare the tags in the tagTable to the 1-byte "short hash" and + * generate a bitfield that we can cycle through to check the collisions in the hash table. + * - Pick the longest match. + */ +FORCE_INLINE_TEMPLATE size_t ZSTD_RowFindBestMatch( - ZSTD_matchState_t* ms, - const BYTE* const ip, const BYTE* const iLimit, - size_t* offsetPtr, - const U32 mls, const ZSTD_dictMode_e dictMode, - const U32 rowLog) -{ - U32* const hashTable = ms->hashTable; - U16* const tagTable = ms->tagTable; - U32* const hashCache = ms->hashCache; - const U32 hashLog = ms->rowHashLog; - const ZSTD_compressionParameters* const cParams = &ms->cParams; - const BYTE* const base = ms->window.base; - const BYTE* const dictBase = ms->window.dictBase; - const U32 dictLimit = ms->window.dictLimit; - const BYTE* const prefixStart = base + dictLimit; - const BYTE* const dictEnd = dictBase + dictLimit; - const U32 curr = (U32)(ip-base); - const U32 maxDistance = 1U << cParams->windowLog; - const U32 lowestValid = ms->window.lowLimit; - const U32 withinMaxDistance = (curr - lowestValid > maxDistance) ? curr - maxDistance : lowestValid; - const U32 isDictionary = (ms->loadedDictEnd != 0); - const U32 lowLimit = isDictionary ? lowestValid : withinMaxDistance; - const U32 rowEntries = (1U << rowLog); - const U32 rowMask = rowEntries - 1; - const U32 cappedSearchLog = MIN(cParams->searchLog, rowLog); /* nb of searches is capped at nb entries per row */ - U32 nbAttempts = 1U << cappedSearchLog; - size_t ml=4-1; - - /* DMS/DDS variables that may be referenced laster */ - const ZSTD_matchState_t* const dms = ms->dictMatchState; - + ZSTD_matchState_t* ms, + const BYTE* const ip, const BYTE* const iLimit, + size_t* offsetPtr, + const U32 mls, const ZSTD_dictMode_e dictMode, + const U32 rowLog) +{ + U32* const hashTable = ms->hashTable; + U16* const tagTable = ms->tagTable; + U32* const hashCache = ms->hashCache; + const U32 hashLog = ms->rowHashLog; + const ZSTD_compressionParameters* const cParams = &ms->cParams; + const BYTE* const base = ms->window.base; + const BYTE* const dictBase = ms->window.dictBase; + const U32 dictLimit = ms->window.dictLimit; + const BYTE* const prefixStart = base + dictLimit; + const BYTE* const dictEnd = dictBase + dictLimit; + const U32 curr = (U32)(ip-base); + const U32 maxDistance = 1U << cParams->windowLog; + const U32 lowestValid = ms->window.lowLimit; + const U32 withinMaxDistance = (curr - lowestValid > maxDistance) ? curr - maxDistance : lowestValid; + const U32 isDictionary = (ms->loadedDictEnd != 0); + const U32 lowLimit = isDictionary ? lowestValid : withinMaxDistance; + const U32 rowEntries = (1U << rowLog); + const U32 rowMask = rowEntries - 1; + const U32 cappedSearchLog = MIN(cParams->searchLog, rowLog); /* nb of searches is capped at nb entries per row */ + U32 nbAttempts = 1U << cappedSearchLog; + size_t ml=4-1; + + /* DMS/DDS variables that may be referenced laster */ + const ZSTD_matchState_t* const dms = ms->dictMatchState; + /* Initialize the following variables to satisfy static analyzer */ size_t ddsIdx = 0; U32 ddsExtraAttempts = 0; /* cctx hash tables are limited in searches, but allow extra searches into DDS */ @@ -1165,144 +1165,144 @@ size_t ZSTD_RowFindBestMatch( U32* dmsRow = NULL; BYTE* dmsTagRow = NULL; - if (dictMode == ZSTD_dedicatedDictSearch) { - const U32 ddsHashLog = dms->cParams.hashLog - ZSTD_LAZY_DDSS_BUCKET_LOG; - { /* Prefetch DDS hashtable entry */ - ddsIdx = ZSTD_hashPtr(ip, ddsHashLog, mls) << ZSTD_LAZY_DDSS_BUCKET_LOG; - PREFETCH_L1(&dms->hashTable[ddsIdx]); - } - ddsExtraAttempts = cParams->searchLog > rowLog ? 1U << (cParams->searchLog - rowLog) : 0; - } - - if (dictMode == ZSTD_dictMatchState) { - /* Prefetch DMS rows */ - U32* const dmsHashTable = dms->hashTable; - U16* const dmsTagTable = dms->tagTable; - U32 const dmsHash = (U32)ZSTD_hashPtr(ip, dms->rowHashLog + ZSTD_ROW_HASH_TAG_BITS, mls); - U32 const dmsRelRow = (dmsHash >> ZSTD_ROW_HASH_TAG_BITS) << rowLog; - dmsTag = dmsHash & ZSTD_ROW_HASH_TAG_MASK; - dmsTagRow = (BYTE*)(dmsTagTable + dmsRelRow); - dmsRow = dmsHashTable + dmsRelRow; - ZSTD_row_prefetch(dmsHashTable, dmsTagTable, dmsRelRow, rowLog); - } - - /* Update the hashTable and tagTable up to (but not including) ip */ - ZSTD_row_update_internal(ms, ip, mls, rowLog, rowMask, 1 /* useCache */); - { /* Get the hash for ip, compute the appropriate row */ - U32 const hash = ZSTD_row_nextCachedHash(hashCache, hashTable, tagTable, base, curr, hashLog, rowLog, mls); - U32 const relRow = (hash >> ZSTD_ROW_HASH_TAG_BITS) << rowLog; - U32 const tag = hash & ZSTD_ROW_HASH_TAG_MASK; - U32* const row = hashTable + relRow; - BYTE* tagRow = (BYTE*)(tagTable + relRow); - U32 const head = *tagRow & rowMask; + if (dictMode == ZSTD_dedicatedDictSearch) { + const U32 ddsHashLog = dms->cParams.hashLog - ZSTD_LAZY_DDSS_BUCKET_LOG; + { /* Prefetch DDS hashtable entry */ + ddsIdx = ZSTD_hashPtr(ip, ddsHashLog, mls) << ZSTD_LAZY_DDSS_BUCKET_LOG; + PREFETCH_L1(&dms->hashTable[ddsIdx]); + } + ddsExtraAttempts = cParams->searchLog > rowLog ? 1U << (cParams->searchLog - rowLog) : 0; + } + + if (dictMode == ZSTD_dictMatchState) { + /* Prefetch DMS rows */ + U32* const dmsHashTable = dms->hashTable; + U16* const dmsTagTable = dms->tagTable; + U32 const dmsHash = (U32)ZSTD_hashPtr(ip, dms->rowHashLog + ZSTD_ROW_HASH_TAG_BITS, mls); + U32 const dmsRelRow = (dmsHash >> ZSTD_ROW_HASH_TAG_BITS) << rowLog; + dmsTag = dmsHash & ZSTD_ROW_HASH_TAG_MASK; + dmsTagRow = (BYTE*)(dmsTagTable + dmsRelRow); + dmsRow = dmsHashTable + dmsRelRow; + ZSTD_row_prefetch(dmsHashTable, dmsTagTable, dmsRelRow, rowLog); + } + + /* Update the hashTable and tagTable up to (but not including) ip */ + ZSTD_row_update_internal(ms, ip, mls, rowLog, rowMask, 1 /* useCache */); + { /* Get the hash for ip, compute the appropriate row */ + U32 const hash = ZSTD_row_nextCachedHash(hashCache, hashTable, tagTable, base, curr, hashLog, rowLog, mls); + U32 const relRow = (hash >> ZSTD_ROW_HASH_TAG_BITS) << rowLog; + U32 const tag = hash & ZSTD_ROW_HASH_TAG_MASK; + U32* const row = hashTable + relRow; + BYTE* tagRow = (BYTE*)(tagTable + relRow); + U32 const head = *tagRow & rowMask; U32 matchBuffer[ZSTD_ROW_HASH_MAX_ENTRIES]; - size_t numMatches = 0; - size_t currMatch = 0; - ZSTD_VecMask matches = ZSTD_row_getMatchMask(tagRow, (BYTE)tag, head, rowEntries); - - /* Cycle through the matches and prefetch */ - for (; (matches > 0) && (nbAttempts > 0); --nbAttempts, matches &= (matches - 1)) { - U32 const matchPos = (head + ZSTD_VecMask_next(matches)) & rowMask; - U32 const matchIndex = row[matchPos]; - assert(numMatches < rowEntries); - if (matchIndex < lowLimit) - break; - if ((dictMode != ZSTD_extDict) || matchIndex >= dictLimit) { - PREFETCH_L1(base + matchIndex); - } else { - PREFETCH_L1(dictBase + matchIndex); - } - matchBuffer[numMatches++] = matchIndex; - } - - /* Speed opt: insert current byte into hashtable too. This allows us to avoid one iteration of the loop - in ZSTD_row_update_internal() at the next search. */ - { - U32 const pos = ZSTD_row_nextIndex(tagRow, rowMask); - tagRow[pos + ZSTD_ROW_HASH_TAG_OFFSET] = (BYTE)tag; - row[pos] = ms->nextToUpdate++; - } - - /* Return the longest match */ - for (; currMatch < numMatches; ++currMatch) { - U32 const matchIndex = matchBuffer[currMatch]; - size_t currentMl=0; - assert(matchIndex < curr); - assert(matchIndex >= lowLimit); - - if ((dictMode != ZSTD_extDict) || matchIndex >= dictLimit) { - const BYTE* const match = base + matchIndex; - assert(matchIndex >= dictLimit); /* ensures this is true if dictMode != ZSTD_extDict */ - if (match[ml] == ip[ml]) /* potentially better */ - currentMl = ZSTD_count(ip, match, iLimit); - } else { - const BYTE* const match = dictBase + matchIndex; - assert(match+4 <= dictEnd); - if (MEM_read32(match) == MEM_read32(ip)) /* assumption : matchIndex <= dictLimit-4 (by table construction) */ - currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, dictEnd, prefixStart) + 4; - } - - /* Save best solution */ - if (currentMl > ml) { - ml = currentMl; + size_t numMatches = 0; + size_t currMatch = 0; + ZSTD_VecMask matches = ZSTD_row_getMatchMask(tagRow, (BYTE)tag, head, rowEntries); + + /* Cycle through the matches and prefetch */ + for (; (matches > 0) && (nbAttempts > 0); --nbAttempts, matches &= (matches - 1)) { + U32 const matchPos = (head + ZSTD_VecMask_next(matches)) & rowMask; + U32 const matchIndex = row[matchPos]; + assert(numMatches < rowEntries); + if (matchIndex < lowLimit) + break; + if ((dictMode != ZSTD_extDict) || matchIndex >= dictLimit) { + PREFETCH_L1(base + matchIndex); + } else { + PREFETCH_L1(dictBase + matchIndex); + } + matchBuffer[numMatches++] = matchIndex; + } + + /* Speed opt: insert current byte into hashtable too. This allows us to avoid one iteration of the loop + in ZSTD_row_update_internal() at the next search. */ + { + U32 const pos = ZSTD_row_nextIndex(tagRow, rowMask); + tagRow[pos + ZSTD_ROW_HASH_TAG_OFFSET] = (BYTE)tag; + row[pos] = ms->nextToUpdate++; + } + + /* Return the longest match */ + for (; currMatch < numMatches; ++currMatch) { + U32 const matchIndex = matchBuffer[currMatch]; + size_t currentMl=0; + assert(matchIndex < curr); + assert(matchIndex >= lowLimit); + + if ((dictMode != ZSTD_extDict) || matchIndex >= dictLimit) { + const BYTE* const match = base + matchIndex; + assert(matchIndex >= dictLimit); /* ensures this is true if dictMode != ZSTD_extDict */ + if (match[ml] == ip[ml]) /* potentially better */ + currentMl = ZSTD_count(ip, match, iLimit); + } else { + const BYTE* const match = dictBase + matchIndex; + assert(match+4 <= dictEnd); + if (MEM_read32(match) == MEM_read32(ip)) /* assumption : matchIndex <= dictLimit-4 (by table construction) */ + currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, dictEnd, prefixStart) + 4; + } + + /* Save best solution */ + if (currentMl > ml) { + ml = currentMl; *offsetPtr = STORE_OFFSET(curr - matchIndex); - if (ip+currentMl == iLimit) break; /* best possible, avoids read overflow on next attempt */ - } - } - } - + if (ip+currentMl == iLimit) break; /* best possible, avoids read overflow on next attempt */ + } + } + } + assert(nbAttempts <= (1U << ZSTD_SEARCHLOG_MAX)); /* Check we haven't underflowed. */ - if (dictMode == ZSTD_dedicatedDictSearch) { - ml = ZSTD_dedicatedDictSearch_lazy_search(offsetPtr, ml, nbAttempts + ddsExtraAttempts, dms, - ip, iLimit, prefixStart, curr, dictLimit, ddsIdx); - } else if (dictMode == ZSTD_dictMatchState) { - /* TODO: Measure and potentially add prefetching to DMS */ - const U32 dmsLowestIndex = dms->window.dictLimit; - const BYTE* const dmsBase = dms->window.base; - const BYTE* const dmsEnd = dms->window.nextSrc; - const U32 dmsSize = (U32)(dmsEnd - dmsBase); - const U32 dmsIndexDelta = dictLimit - dmsSize; - - { U32 const head = *dmsTagRow & rowMask; + if (dictMode == ZSTD_dedicatedDictSearch) { + ml = ZSTD_dedicatedDictSearch_lazy_search(offsetPtr, ml, nbAttempts + ddsExtraAttempts, dms, + ip, iLimit, prefixStart, curr, dictLimit, ddsIdx); + } else if (dictMode == ZSTD_dictMatchState) { + /* TODO: Measure and potentially add prefetching to DMS */ + const U32 dmsLowestIndex = dms->window.dictLimit; + const BYTE* const dmsBase = dms->window.base; + const BYTE* const dmsEnd = dms->window.nextSrc; + const U32 dmsSize = (U32)(dmsEnd - dmsBase); + const U32 dmsIndexDelta = dictLimit - dmsSize; + + { U32 const head = *dmsTagRow & rowMask; U32 matchBuffer[ZSTD_ROW_HASH_MAX_ENTRIES]; - size_t numMatches = 0; - size_t currMatch = 0; - ZSTD_VecMask matches = ZSTD_row_getMatchMask(dmsTagRow, (BYTE)dmsTag, head, rowEntries); - - for (; (matches > 0) && (nbAttempts > 0); --nbAttempts, matches &= (matches - 1)) { - U32 const matchPos = (head + ZSTD_VecMask_next(matches)) & rowMask; - U32 const matchIndex = dmsRow[matchPos]; - if (matchIndex < dmsLowestIndex) - break; - PREFETCH_L1(dmsBase + matchIndex); - matchBuffer[numMatches++] = matchIndex; - } - - /* Return the longest match */ - for (; currMatch < numMatches; ++currMatch) { - U32 const matchIndex = matchBuffer[currMatch]; - size_t currentMl=0; - assert(matchIndex >= dmsLowestIndex); - assert(matchIndex < curr); - - { const BYTE* const match = dmsBase + matchIndex; - assert(match+4 <= dmsEnd); - if (MEM_read32(match) == MEM_read32(ip)) - currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, dmsEnd, prefixStart) + 4; - } - - if (currentMl > ml) { - ml = currentMl; + size_t numMatches = 0; + size_t currMatch = 0; + ZSTD_VecMask matches = ZSTD_row_getMatchMask(dmsTagRow, (BYTE)dmsTag, head, rowEntries); + + for (; (matches > 0) && (nbAttempts > 0); --nbAttempts, matches &= (matches - 1)) { + U32 const matchPos = (head + ZSTD_VecMask_next(matches)) & rowMask; + U32 const matchIndex = dmsRow[matchPos]; + if (matchIndex < dmsLowestIndex) + break; + PREFETCH_L1(dmsBase + matchIndex); + matchBuffer[numMatches++] = matchIndex; + } + + /* Return the longest match */ + for (; currMatch < numMatches; ++currMatch) { + U32 const matchIndex = matchBuffer[currMatch]; + size_t currentMl=0; + assert(matchIndex >= dmsLowestIndex); + assert(matchIndex < curr); + + { const BYTE* const match = dmsBase + matchIndex; + assert(match+4 <= dmsEnd); + if (MEM_read32(match) == MEM_read32(ip)) + currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, dmsEnd, prefixStart) + 4; + } + + if (currentMl > ml) { + ml = currentMl; assert(curr > matchIndex + dmsIndexDelta); *offsetPtr = STORE_OFFSET(curr - (matchIndex + dmsIndexDelta)); - if (ip+currentMl == iLimit) break; - } - } - } - } - return ml; -} - + if (ip+currentMl == iLimit) break; + } + } + } + } + return ml; +} + typedef size_t (*searchMax_f)( ZSTD_matchState_t* ms, @@ -1406,41 +1406,41 @@ ZSTD_FOR_EACH_DICT_MODE(ZSTD_FOR_EACH_MLS, GEN_ZSTD_HC_VTABLE) &ZSTD_BtVTable_##dictMode##_4, \ &ZSTD_BtVTable_##dictMode##_5, \ &ZSTD_BtVTable_##dictMode##_6 \ - } - + } + #define GEN_ZSTD_HC_VTABLE_ARRAY(dictMode) \ { \ &ZSTD_HcVTable_##dictMode##_4, \ &ZSTD_HcVTable_##dictMode##_5, \ &ZSTD_HcVTable_##dictMode##_6 \ - } - + } + #define GEN_ZSTD_ROW_VTABLE_ARRAY_(dictMode, mls) \ { \ &ZSTD_RowVTable_##dictMode##_##mls##_4, \ &ZSTD_RowVTable_##dictMode##_##mls##_5, \ &ZSTD_RowVTable_##dictMode##_##mls##_6 \ - } - + } + #define GEN_ZSTD_ROW_VTABLE_ARRAY(dictMode) \ { \ GEN_ZSTD_ROW_VTABLE_ARRAY_(dictMode, 4), \ GEN_ZSTD_ROW_VTABLE_ARRAY_(dictMode, 5), \ GEN_ZSTD_ROW_VTABLE_ARRAY_(dictMode, 6) \ - } - + } + #define GEN_ZSTD_VTABLE_ARRAY(X) \ { \ X(noDict), \ X(extDict), \ X(dictMatchState), \ X(dedicatedDictSearch) \ - } - + } + /* ******************************* * Common parser - lazy strategy *********************************/ -typedef enum { search_hashChain=0, search_binaryTree=1, search_rowHash=2 } searchMethod_e; +typedef enum { search_hashChain=0, search_binaryTree=1, search_rowHash=2 } searchMethod_e; /** * This table is indexed first by the four ZSTD_dictMode_e values, and then @@ -1492,44 +1492,44 @@ ZSTD_compressBlock_lazy_generic( searchMax_f const searchMax = ZSTD_selectLazyVTable(ms, searchMethod, dictMode)->searchMax; U32 offset_1 = rep[0], offset_2 = rep[1], savedOffset=0; - const int isDMS = dictMode == ZSTD_dictMatchState; - const int isDDS = dictMode == ZSTD_dedicatedDictSearch; - const int isDxS = isDMS || isDDS; + const int isDMS = dictMode == ZSTD_dictMatchState; + const int isDDS = dictMode == ZSTD_dedicatedDictSearch; + const int isDxS = isDMS || isDDS; const ZSTD_matchState_t* const dms = ms->dictMatchState; - const U32 dictLowestIndex = isDxS ? dms->window.dictLimit : 0; - const BYTE* const dictBase = isDxS ? dms->window.base : NULL; - const BYTE* const dictLowest = isDxS ? dictBase + dictLowestIndex : NULL; - const BYTE* const dictEnd = isDxS ? dms->window.nextSrc : NULL; - const U32 dictIndexDelta = isDxS ? + const U32 dictLowestIndex = isDxS ? dms->window.dictLimit : 0; + const BYTE* const dictBase = isDxS ? dms->window.base : NULL; + const BYTE* const dictLowest = isDxS ? dictBase + dictLowestIndex : NULL; + const BYTE* const dictEnd = isDxS ? dms->window.nextSrc : NULL; + const U32 dictIndexDelta = isDxS ? prefixLowestIndex - (U32)(dictEnd - dictBase) : 0; const U32 dictAndPrefixLength = (U32)((ip - prefixLowest) + (dictEnd - dictLowest)); - assert(searchMax != NULL); - - DEBUGLOG(5, "ZSTD_compressBlock_lazy_generic (dictMode=%u) (searchFunc=%u)", (U32)dictMode, (U32)searchMethod); + assert(searchMax != NULL); + + DEBUGLOG(5, "ZSTD_compressBlock_lazy_generic (dictMode=%u) (searchFunc=%u)", (U32)dictMode, (U32)searchMethod); ip += (dictAndPrefixLength == 0); if (dictMode == ZSTD_noDict) { - U32 const curr = (U32)(ip - base); - U32 const windowLow = ZSTD_getLowestPrefixIndex(ms, curr, ms->cParams.windowLog); - U32 const maxRep = curr - windowLow; + U32 const curr = (U32)(ip - base); + U32 const windowLow = ZSTD_getLowestPrefixIndex(ms, curr, ms->cParams.windowLog); + U32 const maxRep = curr - windowLow; if (offset_2 > maxRep) savedOffset = offset_2, offset_2 = 0; if (offset_1 > maxRep) savedOffset = offset_1, offset_1 = 0; } - if (isDxS) { + if (isDxS) { /* dictMatchState repCode checks don't currently handle repCode == 0 * disabling. */ assert(offset_1 <= dictAndPrefixLength); assert(offset_2 <= dictAndPrefixLength); } - if (searchMethod == search_rowHash) { + if (searchMethod == search_rowHash) { const U32 rowLog = MAX(4, MIN(6, ms->cParams.searchLog)); - ZSTD_row_fillHashCache(ms, base, rowLog, - MIN(ms->cParams.minMatch, 6 /* mls caps out at 6 */), - ms->nextToUpdate, ilimit); - } - + ZSTD_row_fillHashCache(ms, base, rowLog, + MIN(ms->cParams.minMatch, 6 /* mls caps out at 6 */), + ms->nextToUpdate, ilimit); + } + /* Match Loop */ #if defined(__GNUC__) && defined(__x86_64__) /* I've measured random a 5% speed loss on levels 5 & 6 (greedy) when the @@ -1544,9 +1544,9 @@ ZSTD_compressBlock_lazy_generic( DEBUGLOG(7, "search baseline (depth 0)"); /* check repCode */ - if (isDxS) { + if (isDxS) { const U32 repIndex = (U32)(ip - base) + 1 - offset_1; - const BYTE* repMatch = ((dictMode == ZSTD_dictMatchState || dictMode == ZSTD_dedicatedDictSearch) + const BYTE* repMatch = ((dictMode == ZSTD_dictMatchState || dictMode == ZSTD_dedicatedDictSearch) && repIndex < prefixLowestIndex) ? dictBase + (repIndex - dictIndexDelta) : base + repIndex; @@ -1588,7 +1588,7 @@ ZSTD_compressBlock_lazy_generic( if ((mlRep >= 4) && (gain2 > gain1)) matchLength = mlRep, offcode = STORE_REPCODE_1, start = ip; } - if (isDxS) { + if (isDxS) { const U32 repIndex = (U32)(ip - base) - offset_1; const BYTE* repMatch = repIndex < prefixLowestIndex ? dictBase + (repIndex - dictIndexDelta) : @@ -1624,7 +1624,7 @@ ZSTD_compressBlock_lazy_generic( if ((mlRep >= 4) && (gain2 > gain1)) matchLength = mlRep, offcode = STORE_REPCODE_1, start = ip; } - if (isDxS) { + if (isDxS) { const U32 repIndex = (U32)(ip - base) - offset_1; const BYTE* repMatch = repIndex < prefixLowestIndex ? dictBase + (repIndex - dictIndexDelta) : @@ -1661,7 +1661,7 @@ ZSTD_compressBlock_lazy_generic( && (start[-1] == (start-STORED_OFFSET(offcode))[-1]) ) /* only search for offset within prefix */ { start--; matchLength++; } } - if (isDxS) { + if (isDxS) { U32 const matchIndex = (U32)((size_t)(start-base) - STORED_OFFSET(offcode)); const BYTE* match = (matchIndex < prefixLowestIndex) ? dictBase + matchIndex - dictIndexDelta : base + matchIndex; const BYTE* const mStart = (matchIndex < prefixLowestIndex) ? dictLowest : prefixLowest; @@ -1677,11 +1677,11 @@ _storeSequence: } /* check immediate repcode */ - if (isDxS) { + if (isDxS) { while (ip <= ilimit) { U32 const current2 = (U32)(ip-base); U32 const repIndex = current2 - offset_2; - const BYTE* repMatch = repIndex < prefixLowestIndex ? + const BYTE* repMatch = repIndex < prefixLowestIndex ? dictBase - dictIndexDelta + repIndex : base + repIndex; if ( ((U32)((prefixLowestIndex-1) - (U32)repIndex) >= 3 /* intentional overflow */) @@ -1776,92 +1776,92 @@ size_t ZSTD_compressBlock_greedy_dictMatchState( } -size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 2, ZSTD_dedicatedDictSearch); -} - -size_t ZSTD_compressBlock_lazy_dedicatedDictSearch( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 1, ZSTD_dedicatedDictSearch); -} - -size_t ZSTD_compressBlock_greedy_dedicatedDictSearch( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 0, ZSTD_dedicatedDictSearch); -} - -/* Row-based matchfinder */ -size_t ZSTD_compressBlock_lazy2_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 2, ZSTD_noDict); -} - -size_t ZSTD_compressBlock_lazy_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 1, ZSTD_noDict); -} - -size_t ZSTD_compressBlock_greedy_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 0, ZSTD_noDict); -} - -size_t ZSTD_compressBlock_lazy2_dictMatchState_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 2, ZSTD_dictMatchState); -} - -size_t ZSTD_compressBlock_lazy_dictMatchState_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 1, ZSTD_dictMatchState); -} - -size_t ZSTD_compressBlock_greedy_dictMatchState_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 0, ZSTD_dictMatchState); -} - - -size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 2, ZSTD_dedicatedDictSearch); -} - -size_t ZSTD_compressBlock_lazy_dedicatedDictSearch_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 1, ZSTD_dedicatedDictSearch); -} - -size_t ZSTD_compressBlock_greedy_dedicatedDictSearch_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 0, ZSTD_dedicatedDictSearch); -} - +size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize) +{ + return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 2, ZSTD_dedicatedDictSearch); +} + +size_t ZSTD_compressBlock_lazy_dedicatedDictSearch( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize) +{ + return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 1, ZSTD_dedicatedDictSearch); +} + +size_t ZSTD_compressBlock_greedy_dedicatedDictSearch( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize) +{ + return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 0, ZSTD_dedicatedDictSearch); +} + +/* Row-based matchfinder */ +size_t ZSTD_compressBlock_lazy2_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize) +{ + return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 2, ZSTD_noDict); +} + +size_t ZSTD_compressBlock_lazy_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize) +{ + return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 1, ZSTD_noDict); +} + +size_t ZSTD_compressBlock_greedy_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize) +{ + return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 0, ZSTD_noDict); +} + +size_t ZSTD_compressBlock_lazy2_dictMatchState_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize) +{ + return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 2, ZSTD_dictMatchState); +} + +size_t ZSTD_compressBlock_lazy_dictMatchState_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize) +{ + return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 1, ZSTD_dictMatchState); +} + +size_t ZSTD_compressBlock_greedy_dictMatchState_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize) +{ + return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 0, ZSTD_dictMatchState); +} + + +size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize) +{ + return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 2, ZSTD_dedicatedDictSearch); +} + +size_t ZSTD_compressBlock_lazy_dedicatedDictSearch_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize) +{ + return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 1, ZSTD_dedicatedDictSearch); +} + +size_t ZSTD_compressBlock_greedy_dedicatedDictSearch_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize) +{ + return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 0, ZSTD_dedicatedDictSearch); +} + FORCE_INLINE_TEMPLATE size_t ZSTD_compressBlock_lazy_extDict_generic( ZSTD_matchState_t* ms, seqStore_t* seqStore, @@ -1873,7 +1873,7 @@ size_t ZSTD_compressBlock_lazy_extDict_generic( const BYTE* ip = istart; const BYTE* anchor = istart; const BYTE* const iend = istart + srcSize; - const BYTE* const ilimit = searchMethod == search_rowHash ? iend - 8 - ZSTD_ROW_HASH_CACHE_SIZE : iend - 8; + const BYTE* const ilimit = searchMethod == search_rowHash ? iend - 8 - ZSTD_ROW_HASH_CACHE_SIZE : iend - 8; const BYTE* const base = ms->window.base; const U32 dictLimit = ms->window.dictLimit; const BYTE* const prefixStart = base + dictLimit; @@ -1881,20 +1881,20 @@ size_t ZSTD_compressBlock_lazy_extDict_generic( const BYTE* const dictEnd = dictBase + dictLimit; const BYTE* const dictStart = dictBase + ms->window.lowLimit; const U32 windowLog = ms->cParams.windowLog; - const U32 rowLog = ms->cParams.searchLog < 5 ? 4 : 5; + const U32 rowLog = ms->cParams.searchLog < 5 ? 4 : 5; searchMax_f const searchMax = ZSTD_selectLazyVTable(ms, searchMethod, ZSTD_extDict)->searchMax; U32 offset_1 = rep[0], offset_2 = rep[1]; - DEBUGLOG(5, "ZSTD_compressBlock_lazy_extDict_generic (searchFunc=%u)", (U32)searchMethod); + DEBUGLOG(5, "ZSTD_compressBlock_lazy_extDict_generic (searchFunc=%u)", (U32)searchMethod); /* init */ ip += (ip == prefixStart); - if (searchMethod == search_rowHash) { - ZSTD_row_fillHashCache(ms, base, rowLog, - MIN(ms->cParams.minMatch, 6 /* mls caps out at 6 */), - ms->nextToUpdate, ilimit); - } + if (searchMethod == search_rowHash) { + ZSTD_row_fillHashCache(ms, base, rowLog, + MIN(ms->cParams.minMatch, 6 /* mls caps out at 6 */), + ms->nextToUpdate, ilimit); + } /* Match Loop */ #if defined(__GNUC__) && defined(__x86_64__) @@ -1907,14 +1907,14 @@ size_t ZSTD_compressBlock_lazy_extDict_generic( size_t matchLength=0; size_t offcode=STORE_REPCODE_1; const BYTE* start=ip+1; - U32 curr = (U32)(ip-base); + U32 curr = (U32)(ip-base); /* check repCode */ - { const U32 windowLow = ZSTD_getLowestMatchIndex(ms, curr+1, windowLog); - const U32 repIndex = (U32)(curr+1 - offset_1); + { const U32 windowLow = ZSTD_getLowestMatchIndex(ms, curr+1, windowLog); + const U32 repIndex = (U32)(curr+1 - offset_1); const BYTE* const repBase = repIndex < dictLimit ? dictBase : base; const BYTE* const repMatch = repBase + repIndex; - if ( ((U32)((dictLimit-1) - repIndex) >= 3) /* intentional overflow */ + if ( ((U32)((dictLimit-1) - repIndex) >= 3) /* intentional overflow */ & (offset_1 <= curr+1 - windowLow) ) /* note: we are searching at curr+1 */ if (MEM_read32(ip+1) == MEM_read32(repMatch)) { /* repcode detected we should take it */ @@ -1939,14 +1939,14 @@ size_t ZSTD_compressBlock_lazy_extDict_generic( if (depth>=1) while (ip<ilimit) { ip ++; - curr++; + curr++; /* check repCode */ if (offcode) { - const U32 windowLow = ZSTD_getLowestMatchIndex(ms, curr, windowLog); - const U32 repIndex = (U32)(curr - offset_1); + const U32 windowLow = ZSTD_getLowestMatchIndex(ms, curr, windowLog); + const U32 repIndex = (U32)(curr - offset_1); const BYTE* const repBase = repIndex < dictLimit ? dictBase : base; const BYTE* const repMatch = repBase + repIndex; - if ( ((U32)((dictLimit-1) - repIndex) >= 3) /* intentional overflow : do not test positions overlapping 2 memory segments */ + if ( ((U32)((dictLimit-1) - repIndex) >= 3) /* intentional overflow : do not test positions overlapping 2 memory segments */ & (offset_1 <= curr - windowLow) ) /* equivalent to `curr > repIndex >= windowLow` */ if (MEM_read32(ip) == MEM_read32(repMatch)) { /* repcode detected */ @@ -1971,14 +1971,14 @@ size_t ZSTD_compressBlock_lazy_extDict_generic( /* let's find an even better one */ if ((depth==2) && (ip<ilimit)) { ip ++; - curr++; + curr++; /* check repCode */ if (offcode) { - const U32 windowLow = ZSTD_getLowestMatchIndex(ms, curr, windowLog); - const U32 repIndex = (U32)(curr - offset_1); + const U32 windowLow = ZSTD_getLowestMatchIndex(ms, curr, windowLog); + const U32 repIndex = (U32)(curr - offset_1); const BYTE* const repBase = repIndex < dictLimit ? dictBase : base; const BYTE* const repMatch = repBase + repIndex; - if ( ((U32)((dictLimit-1) - repIndex) >= 3) /* intentional overflow : do not test positions overlapping 2 memory segments */ + if ( ((U32)((dictLimit-1) - repIndex) >= 3) /* intentional overflow : do not test positions overlapping 2 memory segments */ & (offset_1 <= curr - windowLow) ) /* equivalent to `curr > repIndex >= windowLow` */ if (MEM_read32(ip) == MEM_read32(repMatch)) { /* repcode detected */ @@ -2025,7 +2025,7 @@ _storeSequence: const U32 repIndex = repCurrent - offset_2; const BYTE* const repBase = repIndex < dictLimit ? dictBase : base; const BYTE* const repMatch = repBase + repIndex; - if ( ((U32)((dictLimit-1) - repIndex) >= 3) /* intentional overflow : do not test positions overlapping 2 memory segments */ + if ( ((U32)((dictLimit-1) - repIndex) >= 3) /* intentional overflow : do not test positions overlapping 2 memory segments */ & (offset_2 <= repCurrent - windowLow) ) /* equivalent to `curr > repIndex >= windowLow` */ if (MEM_read32(ip) == MEM_read32(repMatch)) { /* repcode detected we should take it */ @@ -2079,26 +2079,26 @@ size_t ZSTD_compressBlock_btlazy2_extDict( { return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_binaryTree, 2); } - -size_t ZSTD_compressBlock_greedy_extDict_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 0); -} - -size_t ZSTD_compressBlock_lazy_extDict_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) - -{ - return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 1); -} - -size_t ZSTD_compressBlock_lazy2_extDict_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) - -{ - return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 2); -} + +size_t ZSTD_compressBlock_greedy_extDict_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize) +{ + return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 0); +} + +size_t ZSTD_compressBlock_lazy_extDict_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize) + +{ + return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 1); +} + +size_t ZSTD_compressBlock_lazy2_extDict_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize) + +{ + return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 2); +} diff --git a/contrib/libs/zstd/lib/compress/zstd_lazy.h b/contrib/libs/zstd/lib/compress/zstd_lazy.h index 150f7b390b..bb24a2c61d 100644 --- a/contrib/libs/zstd/lib/compress/zstd_lazy.h +++ b/contrib/libs/zstd/lib/compress/zstd_lazy.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -17,19 +17,19 @@ extern "C" { #include "zstd_compress_internal.h" -/** - * Dedicated Dictionary Search Structure bucket log. In the - * ZSTD_dedicatedDictSearch mode, the hashTable has - * 2 ** ZSTD_LAZY_DDSS_BUCKET_LOG entries in each bucket, rather than just - * one. - */ -#define ZSTD_LAZY_DDSS_BUCKET_LOG 2 - +/** + * Dedicated Dictionary Search Structure bucket log. In the + * ZSTD_dedicatedDictSearch mode, the hashTable has + * 2 ** ZSTD_LAZY_DDSS_BUCKET_LOG entries in each bucket, rather than just + * one. + */ +#define ZSTD_LAZY_DDSS_BUCKET_LOG 2 + U32 ZSTD_insertAndFindFirstIndex(ZSTD_matchState_t* ms, const BYTE* ip); -void ZSTD_row_update(ZSTD_matchState_t* const ms, const BYTE* ip); - -void ZSTD_dedicatedDictSearch_lazy_loadDictionary(ZSTD_matchState_t* ms, const BYTE* const ip); +void ZSTD_row_update(ZSTD_matchState_t* const ms, const BYTE* ip); +void ZSTD_dedicatedDictSearch_lazy_loadDictionary(ZSTD_matchState_t* ms, const BYTE* const ip); + void ZSTD_preserveUnsortedMark (U32* const table, U32 const size, U32 const reducerValue); /*! used in ZSTD_reduceIndex(). preemptively increase value of ZSTD_DUBT_UNSORTED_MARK */ size_t ZSTD_compressBlock_btlazy2( @@ -44,15 +44,15 @@ size_t ZSTD_compressBlock_lazy( size_t ZSTD_compressBlock_greedy( ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy2_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_greedy_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); +size_t ZSTD_compressBlock_lazy2_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize); +size_t ZSTD_compressBlock_lazy_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize); +size_t ZSTD_compressBlock_greedy_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize); size_t ZSTD_compressBlock_btlazy2_dictMatchState( ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], @@ -66,35 +66,35 @@ size_t ZSTD_compressBlock_lazy_dictMatchState( size_t ZSTD_compressBlock_greedy_dictMatchState( ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy2_dictMatchState_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy_dictMatchState_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_greedy_dictMatchState_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); - -size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy_dedicatedDictSearch( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_greedy_dedicatedDictSearch( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy_dedicatedDictSearch_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_greedy_dedicatedDictSearch_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); +size_t ZSTD_compressBlock_lazy2_dictMatchState_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize); +size_t ZSTD_compressBlock_lazy_dictMatchState_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize); +size_t ZSTD_compressBlock_greedy_dictMatchState_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize); +size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize); +size_t ZSTD_compressBlock_lazy_dedicatedDictSearch( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize); +size_t ZSTD_compressBlock_greedy_dedicatedDictSearch( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize); +size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize); +size_t ZSTD_compressBlock_lazy_dedicatedDictSearch_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize); +size_t ZSTD_compressBlock_greedy_dedicatedDictSearch_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize); + size_t ZSTD_compressBlock_greedy_extDict( ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], void const* src, size_t srcSize); @@ -104,19 +104,19 @@ size_t ZSTD_compressBlock_lazy_extDict( size_t ZSTD_compressBlock_lazy2_extDict( ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], void const* src, size_t srcSize); -size_t ZSTD_compressBlock_greedy_extDict_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy_extDict_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy2_extDict_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); +size_t ZSTD_compressBlock_greedy_extDict_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize); +size_t ZSTD_compressBlock_lazy_extDict_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize); +size_t ZSTD_compressBlock_lazy2_extDict_row( + ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], + void const* src, size_t srcSize); size_t ZSTD_compressBlock_btlazy2_extDict( ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], void const* src, size_t srcSize); - + #if defined (__cplusplus) } diff --git a/contrib/libs/zstd/lib/compress/zstd_ldm.c b/contrib/libs/zstd/lib/compress/zstd_ldm.c index 476b45746e..9eba8e7b79 100644 --- a/contrib/libs/zstd/lib/compress/zstd_ldm.c +++ b/contrib/libs/zstd/lib/compress/zstd_ldm.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -11,127 +11,127 @@ #include "zstd_ldm.h" #include "../common/debug.h" -#include <contrib/libs/xxhash/xxhash.h> +#include <contrib/libs/xxhash/xxhash.h> #include "zstd_fast.h" /* ZSTD_fillHashTable() */ #include "zstd_double_fast.h" /* ZSTD_fillDoubleHashTable() */ -#include "zstd_ldm_geartab.h" +#include "zstd_ldm_geartab.h" #define LDM_BUCKET_SIZE_LOG 3 #define LDM_MIN_MATCH_LENGTH 64 #define LDM_HASH_RLOG 7 -typedef struct { - U64 rolling; - U64 stopMask; -} ldmRollingHashState_t; - -/** ZSTD_ldm_gear_init(): - * - * Initializes the rolling hash state such that it will honor the - * settings in params. */ -static void ZSTD_ldm_gear_init(ldmRollingHashState_t* state, ldmParams_t const* params) -{ - unsigned maxBitsInMask = MIN(params->minMatchLength, 64); - unsigned hashRateLog = params->hashRateLog; - - state->rolling = ~(U32)0; - - /* The choice of the splitting criterion is subject to two conditions: - * 1. it has to trigger on average every 2^(hashRateLog) bytes; - * 2. ideally, it has to depend on a window of minMatchLength bytes. - * - * In the gear hash algorithm, bit n depends on the last n bytes; - * so in order to obtain a good quality splitting criterion it is - * preferable to use bits with high weight. - * - * To match condition 1 we use a mask with hashRateLog bits set - * and, because of the previous remark, we make sure these bits - * have the highest possible weight while still respecting - * condition 2. - */ - if (hashRateLog > 0 && hashRateLog <= maxBitsInMask) { - state->stopMask = (((U64)1 << hashRateLog) - 1) << (maxBitsInMask - hashRateLog); - } else { - /* In this degenerate case we simply honor the hash rate. */ - state->stopMask = ((U64)1 << hashRateLog) - 1; - } -} - -/** ZSTD_ldm_gear_reset() - * Feeds [data, data + minMatchLength) into the hash without registering any - * splits. This effectively resets the hash state. This is used when skipping - * over data, either at the beginning of a block, or skipping sections. - */ -static void ZSTD_ldm_gear_reset(ldmRollingHashState_t* state, - BYTE const* data, size_t minMatchLength) -{ - U64 hash = state->rolling; - size_t n = 0; - -#define GEAR_ITER_ONCE() do { \ - hash = (hash << 1) + ZSTD_ldm_gearTab[data[n] & 0xff]; \ - n += 1; \ - } while (0) - while (n + 3 < minMatchLength) { - GEAR_ITER_ONCE(); - GEAR_ITER_ONCE(); - GEAR_ITER_ONCE(); - GEAR_ITER_ONCE(); - } - while (n < minMatchLength) { - GEAR_ITER_ONCE(); - } -#undef GEAR_ITER_ONCE -} - -/** ZSTD_ldm_gear_feed(): - * - * Registers in the splits array all the split points found in the first - * size bytes following the data pointer. This function terminates when - * either all the data has been processed or LDM_BATCH_SIZE splits are - * present in the splits array. - * - * Precondition: The splits array must not be full. - * Returns: The number of bytes processed. */ -static size_t ZSTD_ldm_gear_feed(ldmRollingHashState_t* state, - BYTE const* data, size_t size, - size_t* splits, unsigned* numSplits) -{ - size_t n; - U64 hash, mask; - - hash = state->rolling; - mask = state->stopMask; - n = 0; - -#define GEAR_ITER_ONCE() do { \ - hash = (hash << 1) + ZSTD_ldm_gearTab[data[n] & 0xff]; \ - n += 1; \ - if (UNLIKELY((hash & mask) == 0)) { \ - splits[*numSplits] = n; \ - *numSplits += 1; \ - if (*numSplits == LDM_BATCH_SIZE) \ - goto done; \ - } \ - } while (0) - - while (n + 3 < size) { - GEAR_ITER_ONCE(); - GEAR_ITER_ONCE(); - GEAR_ITER_ONCE(); - GEAR_ITER_ONCE(); - } - while (n < size) { - GEAR_ITER_ONCE(); - } - -#undef GEAR_ITER_ONCE - -done: - state->rolling = hash; - return n; -} - +typedef struct { + U64 rolling; + U64 stopMask; +} ldmRollingHashState_t; + +/** ZSTD_ldm_gear_init(): + * + * Initializes the rolling hash state such that it will honor the + * settings in params. */ +static void ZSTD_ldm_gear_init(ldmRollingHashState_t* state, ldmParams_t const* params) +{ + unsigned maxBitsInMask = MIN(params->minMatchLength, 64); + unsigned hashRateLog = params->hashRateLog; + + state->rolling = ~(U32)0; + + /* The choice of the splitting criterion is subject to two conditions: + * 1. it has to trigger on average every 2^(hashRateLog) bytes; + * 2. ideally, it has to depend on a window of minMatchLength bytes. + * + * In the gear hash algorithm, bit n depends on the last n bytes; + * so in order to obtain a good quality splitting criterion it is + * preferable to use bits with high weight. + * + * To match condition 1 we use a mask with hashRateLog bits set + * and, because of the previous remark, we make sure these bits + * have the highest possible weight while still respecting + * condition 2. + */ + if (hashRateLog > 0 && hashRateLog <= maxBitsInMask) { + state->stopMask = (((U64)1 << hashRateLog) - 1) << (maxBitsInMask - hashRateLog); + } else { + /* In this degenerate case we simply honor the hash rate. */ + state->stopMask = ((U64)1 << hashRateLog) - 1; + } +} + +/** ZSTD_ldm_gear_reset() + * Feeds [data, data + minMatchLength) into the hash without registering any + * splits. This effectively resets the hash state. This is used when skipping + * over data, either at the beginning of a block, or skipping sections. + */ +static void ZSTD_ldm_gear_reset(ldmRollingHashState_t* state, + BYTE const* data, size_t minMatchLength) +{ + U64 hash = state->rolling; + size_t n = 0; + +#define GEAR_ITER_ONCE() do { \ + hash = (hash << 1) + ZSTD_ldm_gearTab[data[n] & 0xff]; \ + n += 1; \ + } while (0) + while (n + 3 < minMatchLength) { + GEAR_ITER_ONCE(); + GEAR_ITER_ONCE(); + GEAR_ITER_ONCE(); + GEAR_ITER_ONCE(); + } + while (n < minMatchLength) { + GEAR_ITER_ONCE(); + } +#undef GEAR_ITER_ONCE +} + +/** ZSTD_ldm_gear_feed(): + * + * Registers in the splits array all the split points found in the first + * size bytes following the data pointer. This function terminates when + * either all the data has been processed or LDM_BATCH_SIZE splits are + * present in the splits array. + * + * Precondition: The splits array must not be full. + * Returns: The number of bytes processed. */ +static size_t ZSTD_ldm_gear_feed(ldmRollingHashState_t* state, + BYTE const* data, size_t size, + size_t* splits, unsigned* numSplits) +{ + size_t n; + U64 hash, mask; + + hash = state->rolling; + mask = state->stopMask; + n = 0; + +#define GEAR_ITER_ONCE() do { \ + hash = (hash << 1) + ZSTD_ldm_gearTab[data[n] & 0xff]; \ + n += 1; \ + if (UNLIKELY((hash & mask) == 0)) { \ + splits[*numSplits] = n; \ + *numSplits += 1; \ + if (*numSplits == LDM_BATCH_SIZE) \ + goto done; \ + } \ + } while (0) + + while (n + 3 < size) { + GEAR_ITER_ONCE(); + GEAR_ITER_ONCE(); + GEAR_ITER_ONCE(); + GEAR_ITER_ONCE(); + } + while (n < size) { + GEAR_ITER_ONCE(); + } + +#undef GEAR_ITER_ONCE + +done: + state->rolling = hash; + return n; +} + void ZSTD_ldm_adjustParameters(ldmParams_t* params, ZSTD_compressionParameters const* cParams) { @@ -181,12 +181,12 @@ static void ZSTD_ldm_insertEntry(ldmState_t* ldmState, size_t const hash, const ldmEntry_t entry, ldmParams_t const ldmParams) { - BYTE* const pOffset = ldmState->bucketOffsets + hash; - unsigned const offset = *pOffset; - - *(ZSTD_ldm_getBucket(ldmState, hash, ldmParams) + offset) = entry; - *pOffset = (BYTE)((offset + 1) & ((1u << ldmParams.bucketSizeLog) - 1)); + BYTE* const pOffset = ldmState->bucketOffsets + hash; + unsigned const offset = *pOffset; + *(ZSTD_ldm_getBucket(ldmState, hash, ldmParams) + offset) = entry; + *pOffset = (BYTE)((offset + 1) & ((1u << ldmParams.bucketSizeLog) - 1)); + } /** ZSTD_ldm_countBackwardsMatch() : @@ -195,10 +195,10 @@ static void ZSTD_ldm_insertEntry(ldmState_t* ldmState, * We count only bytes where pMatch >= pBase and pIn >= pAnchor. */ static size_t ZSTD_ldm_countBackwardsMatch( const BYTE* pIn, const BYTE* pAnchor, - const BYTE* pMatch, const BYTE* pMatchBase) + const BYTE* pMatch, const BYTE* pMatchBase) { size_t matchLength = 0; - while (pIn > pAnchor && pMatch > pMatchBase && pIn[-1] == pMatch[-1]) { + while (pIn > pAnchor && pMatch > pMatchBase && pIn[-1] == pMatch[-1]) { pIn--; pMatch--; matchLength++; @@ -206,27 +206,27 @@ static size_t ZSTD_ldm_countBackwardsMatch( return matchLength; } -/** ZSTD_ldm_countBackwardsMatch_2segments() : - * Returns the number of bytes that match backwards from pMatch, - * even with the backwards match spanning 2 different segments. - * - * On reaching `pMatchBase`, start counting from mEnd */ -static size_t ZSTD_ldm_countBackwardsMatch_2segments( - const BYTE* pIn, const BYTE* pAnchor, - const BYTE* pMatch, const BYTE* pMatchBase, - const BYTE* pExtDictStart, const BYTE* pExtDictEnd) -{ - size_t matchLength = ZSTD_ldm_countBackwardsMatch(pIn, pAnchor, pMatch, pMatchBase); - if (pMatch - matchLength != pMatchBase || pMatchBase == pExtDictStart) { - /* If backwards match is entirely in the extDict or prefix, immediately return */ - return matchLength; - } - DEBUGLOG(7, "ZSTD_ldm_countBackwardsMatch_2segments: found 2-parts backwards match (length in prefix==%zu)", matchLength); - matchLength += ZSTD_ldm_countBackwardsMatch(pIn - matchLength, pAnchor, pExtDictEnd, pExtDictStart); - DEBUGLOG(7, "final backwards match length = %zu", matchLength); - return matchLength; -} - +/** ZSTD_ldm_countBackwardsMatch_2segments() : + * Returns the number of bytes that match backwards from pMatch, + * even with the backwards match spanning 2 different segments. + * + * On reaching `pMatchBase`, start counting from mEnd */ +static size_t ZSTD_ldm_countBackwardsMatch_2segments( + const BYTE* pIn, const BYTE* pAnchor, + const BYTE* pMatch, const BYTE* pMatchBase, + const BYTE* pExtDictStart, const BYTE* pExtDictEnd) +{ + size_t matchLength = ZSTD_ldm_countBackwardsMatch(pIn, pAnchor, pMatch, pMatchBase); + if (pMatch - matchLength != pMatchBase || pMatchBase == pExtDictStart) { + /* If backwards match is entirely in the extDict or prefix, immediately return */ + return matchLength; + } + DEBUGLOG(7, "ZSTD_ldm_countBackwardsMatch_2segments: found 2-parts backwards match (length in prefix==%zu)", matchLength); + matchLength += ZSTD_ldm_countBackwardsMatch(pIn - matchLength, pAnchor, pExtDictEnd, pExtDictStart); + DEBUGLOG(7, "final backwards match length = %zu", matchLength); + return matchLength; +} + /** ZSTD_ldm_fillFastTables() : * * Fills the relevant tables for the ZSTD_fast and ZSTD_dfast strategies. @@ -265,41 +265,41 @@ static size_t ZSTD_ldm_fillFastTables(ZSTD_matchState_t* ms, } void ZSTD_ldm_fillHashTable( - ldmState_t* ldmState, const BYTE* ip, + ldmState_t* ldmState, const BYTE* ip, const BYTE* iend, ldmParams_t const* params) { - U32 const minMatchLength = params->minMatchLength; - U32 const hBits = params->hashLog - params->bucketSizeLog; - BYTE const* const base = ldmState->window.base; - BYTE const* const istart = ip; - ldmRollingHashState_t hashState; - size_t* const splits = ldmState->splitIndices; - unsigned numSplits; - + U32 const minMatchLength = params->minMatchLength; + U32 const hBits = params->hashLog - params->bucketSizeLog; + BYTE const* const base = ldmState->window.base; + BYTE const* const istart = ip; + ldmRollingHashState_t hashState; + size_t* const splits = ldmState->splitIndices; + unsigned numSplits; + DEBUGLOG(5, "ZSTD_ldm_fillHashTable"); - - ZSTD_ldm_gear_init(&hashState, params); - while (ip < iend) { - size_t hashed; - unsigned n; - - numSplits = 0; - hashed = ZSTD_ldm_gear_feed(&hashState, ip, iend - ip, splits, &numSplits); - - for (n = 0; n < numSplits; n++) { - if (ip + splits[n] >= istart + minMatchLength) { - BYTE const* const split = ip + splits[n] - minMatchLength; - U64 const xxhash = XXH64(split, minMatchLength, 0); - U32 const hash = (U32)(xxhash & (((U32)1 << hBits) - 1)); - ldmEntry_t entry; - - entry.offset = (U32)(split - base); - entry.checksum = (U32)(xxhash >> 32); - ZSTD_ldm_insertEntry(ldmState, hash, entry, *params); - } - } - - ip += hashed; + + ZSTD_ldm_gear_init(&hashState, params); + while (ip < iend) { + size_t hashed; + unsigned n; + + numSplits = 0; + hashed = ZSTD_ldm_gear_feed(&hashState, ip, iend - ip, splits, &numSplits); + + for (n = 0; n < numSplits; n++) { + if (ip + splits[n] >= istart + minMatchLength) { + BYTE const* const split = ip + splits[n] - minMatchLength; + U64 const xxhash = XXH64(split, minMatchLength, 0); + U32 const hash = (U32)(xxhash & (((U32)1 << hBits) - 1)); + ldmEntry_t entry; + + entry.offset = (U32)(split - base); + entry.checksum = (U32)(xxhash >> 32); + ZSTD_ldm_insertEntry(ldmState, hash, entry, *params); + } + } + + ip += hashed; } } @@ -311,10 +311,10 @@ void ZSTD_ldm_fillHashTable( * (after a long match, only update tables a limited amount). */ static void ZSTD_ldm_limitTableUpdate(ZSTD_matchState_t* ms, const BYTE* anchor) { - U32 const curr = (U32)(anchor - ms->window.base); - if (curr > ms->nextToUpdate + 1024) { + U32 const curr = (U32)(anchor - ms->window.base); + if (curr > ms->nextToUpdate + 1024) { ms->nextToUpdate = - curr - MIN(512, curr - ms->nextToUpdate - 1024); + curr - MIN(512, curr - ms->nextToUpdate - 1024); } } @@ -325,7 +325,7 @@ static size_t ZSTD_ldm_generateSequences_internal( /* LDM parameters */ int const extDict = ZSTD_window_hasExtDict(ldmState->window); U32 const minMatchLength = params->minMatchLength; - U32 const entsPerBucket = 1U << params->bucketSizeLog; + U32 const entsPerBucket = 1U << params->bucketSizeLog; U32 const hBits = params->hashLog - params->bucketSizeLog; /* Prefix and extDict parameters */ U32 const dictLimit = ldmState->window.dictLimit; @@ -338,69 +338,69 @@ static size_t ZSTD_ldm_generateSequences_internal( /* Input bounds */ BYTE const* const istart = (BYTE const*)src; BYTE const* const iend = istart + srcSize; - BYTE const* const ilimit = iend - HASH_READ_SIZE; + BYTE const* const ilimit = iend - HASH_READ_SIZE; /* Input positions */ BYTE const* anchor = istart; BYTE const* ip = istart; - /* Rolling hash state */ - ldmRollingHashState_t hashState; - /* Arrays for staged-processing */ - size_t* const splits = ldmState->splitIndices; - ldmMatchCandidate_t* const candidates = ldmState->matchCandidates; - unsigned numSplits; - - if (srcSize < minMatchLength) - return iend - anchor; - - /* Initialize the rolling hash state with the first minMatchLength bytes */ - ZSTD_ldm_gear_init(&hashState, params); - ZSTD_ldm_gear_reset(&hashState, ip, minMatchLength); - ip += minMatchLength; - - while (ip < ilimit) { - size_t hashed; - unsigned n; - - numSplits = 0; - hashed = ZSTD_ldm_gear_feed(&hashState, ip, ilimit - ip, - splits, &numSplits); - - for (n = 0; n < numSplits; n++) { - BYTE const* const split = ip + splits[n] - minMatchLength; - U64 const xxhash = XXH64(split, minMatchLength, 0); - U32 const hash = (U32)(xxhash & (((U32)1 << hBits) - 1)); - - candidates[n].split = split; - candidates[n].hash = hash; - candidates[n].checksum = (U32)(xxhash >> 32); - candidates[n].bucket = ZSTD_ldm_getBucket(ldmState, hash, *params); - PREFETCH_L1(candidates[n].bucket); + /* Rolling hash state */ + ldmRollingHashState_t hashState; + /* Arrays for staged-processing */ + size_t* const splits = ldmState->splitIndices; + ldmMatchCandidate_t* const candidates = ldmState->matchCandidates; + unsigned numSplits; + + if (srcSize < minMatchLength) + return iend - anchor; + + /* Initialize the rolling hash state with the first minMatchLength bytes */ + ZSTD_ldm_gear_init(&hashState, params); + ZSTD_ldm_gear_reset(&hashState, ip, minMatchLength); + ip += minMatchLength; + + while (ip < ilimit) { + size_t hashed; + unsigned n; + + numSplits = 0; + hashed = ZSTD_ldm_gear_feed(&hashState, ip, ilimit - ip, + splits, &numSplits); + + for (n = 0; n < numSplits; n++) { + BYTE const* const split = ip + splits[n] - minMatchLength; + U64 const xxhash = XXH64(split, minMatchLength, 0); + U32 const hash = (U32)(xxhash & (((U32)1 << hBits) - 1)); + + candidates[n].split = split; + candidates[n].hash = hash; + candidates[n].checksum = (U32)(xxhash >> 32); + candidates[n].bucket = ZSTD_ldm_getBucket(ldmState, hash, *params); + PREFETCH_L1(candidates[n].bucket); } - for (n = 0; n < numSplits; n++) { - size_t forwardMatchLength = 0, backwardMatchLength = 0, - bestMatchLength = 0, mLength; - U32 offset; - BYTE const* const split = candidates[n].split; - U32 const checksum = candidates[n].checksum; - U32 const hash = candidates[n].hash; - ldmEntry_t* const bucket = candidates[n].bucket; - ldmEntry_t const* cur; - ldmEntry_t const* bestEntry = NULL; - ldmEntry_t newEntry; - - newEntry.offset = (U32)(split - base); - newEntry.checksum = checksum; - - /* If a split point would generate a sequence overlapping with - * the previous one, we merely register it in the hash table and - * move on */ - if (split < anchor) { - ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params); - continue; - } - - for (cur = bucket; cur < bucket + entsPerBucket; cur++) { + for (n = 0; n < numSplits; n++) { + size_t forwardMatchLength = 0, backwardMatchLength = 0, + bestMatchLength = 0, mLength; + U32 offset; + BYTE const* const split = candidates[n].split; + U32 const checksum = candidates[n].checksum; + U32 const hash = candidates[n].hash; + ldmEntry_t* const bucket = candidates[n].bucket; + ldmEntry_t const* cur; + ldmEntry_t const* bestEntry = NULL; + ldmEntry_t newEntry; + + newEntry.offset = (U32)(split - base); + newEntry.checksum = checksum; + + /* If a split point would generate a sequence overlapping with + * the previous one, we merely register it in the hash table and + * move on */ + if (split < anchor) { + ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params); + continue; + } + + for (cur = bucket; cur < bucket + entsPerBucket; cur++) { size_t curForwardMatchLength, curBackwardMatchLength, curTotalMatchLength; if (cur->checksum != checksum || cur->offset <= lowestIndex) { @@ -414,23 +414,23 @@ static size_t ZSTD_ldm_generateSequences_internal( cur->offset < dictLimit ? dictEnd : iend; BYTE const* const lowMatchPtr = cur->offset < dictLimit ? dictStart : lowPrefixPtr; - curForwardMatchLength = - ZSTD_count_2segments(split, pMatch, iend, matchEnd, lowPrefixPtr); + curForwardMatchLength = + ZSTD_count_2segments(split, pMatch, iend, matchEnd, lowPrefixPtr); if (curForwardMatchLength < minMatchLength) { continue; } - curBackwardMatchLength = ZSTD_ldm_countBackwardsMatch_2segments( - split, anchor, pMatch, lowMatchPtr, dictStart, dictEnd); + curBackwardMatchLength = ZSTD_ldm_countBackwardsMatch_2segments( + split, anchor, pMatch, lowMatchPtr, dictStart, dictEnd); } else { /* !extDict */ BYTE const* const pMatch = base + cur->offset; - curForwardMatchLength = ZSTD_count(split, pMatch, iend); + curForwardMatchLength = ZSTD_count(split, pMatch, iend); if (curForwardMatchLength < minMatchLength) { continue; } curBackwardMatchLength = - ZSTD_ldm_countBackwardsMatch(split, anchor, pMatch, lowPrefixPtr); + ZSTD_ldm_countBackwardsMatch(split, anchor, pMatch, lowPrefixPtr); } - curTotalMatchLength = curForwardMatchLength + curBackwardMatchLength; + curTotalMatchLength = curForwardMatchLength + curBackwardMatchLength; if (curTotalMatchLength > bestMatchLength) { bestMatchLength = curTotalMatchLength; @@ -440,52 +440,52 @@ static size_t ZSTD_ldm_generateSequences_internal( } } - /* No match found -- insert an entry into the hash table - * and process the next candidate match */ - if (bestEntry == NULL) { - ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params); - continue; - } - - /* Match found */ - offset = (U32)(split - base) - bestEntry->offset; - mLength = forwardMatchLength + backwardMatchLength; - { - rawSeq* const seq = rawSeqStore->seq + rawSeqStore->size; - - /* Out of sequence storage */ - if (rawSeqStore->size == rawSeqStore->capacity) - return ERROR(dstSize_tooSmall); - seq->litLength = (U32)(split - backwardMatchLength - anchor); - seq->matchLength = (U32)mLength; - seq->offset = offset; - rawSeqStore->size++; - } - - /* Insert the current entry into the hash table --- it must be - * done after the previous block to avoid clobbering bestEntry */ - ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params); - - anchor = split + forwardMatchLength; - - /* If we find a match that ends after the data that we've hashed - * then we have a repeating, overlapping, pattern. E.g. all zeros. - * If one repetition of the pattern matches our `stopMask` then all - * repetitions will. We don't need to insert them all into out table, - * only the first one. So skip over overlapping matches. - * This is a major speed boost (20x) for compressing a single byte - * repeated, when that byte ends up in the table. - */ - if (anchor > ip + hashed) { - ZSTD_ldm_gear_reset(&hashState, anchor - minMatchLength, minMatchLength); + /* No match found -- insert an entry into the hash table + * and process the next candidate match */ + if (bestEntry == NULL) { + ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params); + continue; + } + + /* Match found */ + offset = (U32)(split - base) - bestEntry->offset; + mLength = forwardMatchLength + backwardMatchLength; + { + rawSeq* const seq = rawSeqStore->seq + rawSeqStore->size; + + /* Out of sequence storage */ + if (rawSeqStore->size == rawSeqStore->capacity) + return ERROR(dstSize_tooSmall); + seq->litLength = (U32)(split - backwardMatchLength - anchor); + seq->matchLength = (U32)mLength; + seq->offset = offset; + rawSeqStore->size++; + } + + /* Insert the current entry into the hash table --- it must be + * done after the previous block to avoid clobbering bestEntry */ + ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params); + + anchor = split + forwardMatchLength; + + /* If we find a match that ends after the data that we've hashed + * then we have a repeating, overlapping, pattern. E.g. all zeros. + * If one repetition of the pattern matches our `stopMask` then all + * repetitions will. We don't need to insert them all into out table, + * only the first one. So skip over overlapping matches. + * This is a major speed boost (20x) for compressing a single byte + * repeated, when that byte ends up in the table. + */ + if (anchor > ip + hashed) { + ZSTD_ldm_gear_reset(&hashState, anchor - minMatchLength, minMatchLength); /* Continue the outer loop at anchor (ip + hashed == anchor). */ - ip = anchor - hashed; - break; - } + ip = anchor - hashed; + break; + } } - ip += hashed; - } + ip += hashed; + } return iend - anchor; } @@ -535,7 +535,7 @@ size_t ZSTD_ldm_generateSequences( assert(chunkStart < iend); /* 1. Perform overflow correction if necessary. */ - if (ZSTD_window_needOverflowCorrection(ldmState->window, 0, maxDist, ldmState->loadedDictEnd, chunkStart, chunkEnd)) { + if (ZSTD_window_needOverflowCorrection(ldmState->window, 0, maxDist, ldmState->loadedDictEnd, chunkStart, chunkEnd)) { U32 const ldmHSize = 1U << params->hashLog; U32 const correction = ZSTD_window_correctOverflow( &ldmState->window, /* cycleLog */ 0, maxDist, chunkStart); @@ -640,23 +640,23 @@ static rawSeq maybeSplitSequence(rawSeqStore_t* rawSeqStore, return sequence; } -void ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes) { - U32 currPos = (U32)(rawSeqStore->posInSequence + nbBytes); - while (currPos && rawSeqStore->pos < rawSeqStore->size) { - rawSeq currSeq = rawSeqStore->seq[rawSeqStore->pos]; - if (currPos >= currSeq.litLength + currSeq.matchLength) { - currPos -= currSeq.litLength + currSeq.matchLength; - rawSeqStore->pos++; - } else { - rawSeqStore->posInSequence = currPos; - break; - } - } - if (currPos == 0 || rawSeqStore->pos == rawSeqStore->size) { - rawSeqStore->posInSequence = 0; - } -} - +void ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes) { + U32 currPos = (U32)(rawSeqStore->posInSequence + nbBytes); + while (currPos && rawSeqStore->pos < rawSeqStore->size) { + rawSeq currSeq = rawSeqStore->seq[rawSeqStore->pos]; + if (currPos >= currSeq.litLength + currSeq.matchLength) { + currPos -= currSeq.litLength + currSeq.matchLength; + rawSeqStore->pos++; + } else { + rawSeqStore->posInSequence = currPos; + break; + } + } + if (currPos == 0 || rawSeqStore->pos == rawSeqStore->size) { + rawSeqStore->posInSequence = 0; + } +} + size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore, ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], ZSTD_paramSwitch_e useRowMatchFinder, @@ -665,7 +665,7 @@ size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore, const ZSTD_compressionParameters* const cParams = &ms->cParams; unsigned const minMatch = cParams->minMatch; ZSTD_blockCompressor const blockCompressor = - ZSTD_selectBlockCompressor(cParams->strategy, useRowMatchFinder, ZSTD_matchState_dictMode(ms)); + ZSTD_selectBlockCompressor(cParams->strategy, useRowMatchFinder, ZSTD_matchState_dictMode(ms)); /* Input bounds */ BYTE const* const istart = (BYTE const*)src; BYTE const* const iend = istart + srcSize; @@ -673,18 +673,18 @@ size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore, BYTE const* ip = istart; DEBUGLOG(5, "ZSTD_ldm_blockCompress: srcSize=%zu", srcSize); - /* If using opt parser, use LDMs only as candidates rather than always accepting them */ - if (cParams->strategy >= ZSTD_btopt) { - size_t lastLLSize; - ms->ldmSeqStore = rawSeqStore; - lastLLSize = blockCompressor(ms, seqStore, rep, src, srcSize); - ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore, srcSize); - return lastLLSize; - } - + /* If using opt parser, use LDMs only as candidates rather than always accepting them */ + if (cParams->strategy >= ZSTD_btopt) { + size_t lastLLSize; + ms->ldmSeqStore = rawSeqStore; + lastLLSize = blockCompressor(ms, seqStore, rep, src, srcSize); + ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore, srcSize); + return lastLLSize; + } + assert(rawSeqStore->pos <= rawSeqStore->size); assert(rawSeqStore->size <= rawSeqStore->capacity); - /* Loop through each sequence and apply the block compressor to the literals */ + /* Loop through each sequence and apply the block compressor to the literals */ while (rawSeqStore->pos < rawSeqStore->size && ip < iend) { /* maybeSplitSequence updates rawSeqStore->pos */ rawSeq const sequence = maybeSplitSequence(rawSeqStore, diff --git a/contrib/libs/zstd/lib/compress/zstd_ldm.h b/contrib/libs/zstd/lib/compress/zstd_ldm.h index 4e68dbf52e..36b4a84e66 100644 --- a/contrib/libs/zstd/lib/compress/zstd_ldm.h +++ b/contrib/libs/zstd/lib/compress/zstd_ldm.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -74,17 +74,17 @@ size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore, * * Skip past `srcSize` bytes worth of sequences in `rawSeqStore`. * Avoids emitting matches less than `minMatch` bytes. - * Must be called for data that is not passed to ZSTD_ldm_blockCompress(). + * Must be called for data that is not passed to ZSTD_ldm_blockCompress(). */ void ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize, U32 const minMatch); -/* ZSTD_ldm_skipRawSeqStoreBytes(): - * Moves forward in rawSeqStore by nbBytes, updating fields 'pos' and 'posInSequence'. - * Not to be used in conjunction with ZSTD_ldm_skipSequences(). - * Must be called for data with is not passed to ZSTD_ldm_blockCompress(). - */ -void ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes); +/* ZSTD_ldm_skipRawSeqStoreBytes(): + * Moves forward in rawSeqStore by nbBytes, updating fields 'pos' and 'posInSequence'. + * Not to be used in conjunction with ZSTD_ldm_skipSequences(). + * Must be called for data with is not passed to ZSTD_ldm_blockCompress(). + */ +void ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes); /** ZSTD_ldm_getTableSize() : * Estimate the space needed for long distance matching tables or 0 if LDM is diff --git a/contrib/libs/zstd/lib/compress/zstd_ldm_geartab.h b/contrib/libs/zstd/lib/compress/zstd_ldm_geartab.h index 647f865be2..5196e46238 100644 --- a/contrib/libs/zstd/lib/compress/zstd_ldm_geartab.h +++ b/contrib/libs/zstd/lib/compress/zstd_ldm_geartab.h @@ -1,106 +1,106 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_LDM_GEARTAB_H -#define ZSTD_LDM_GEARTAB_H - +/* + * Copyright (c) Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + +#ifndef ZSTD_LDM_GEARTAB_H +#define ZSTD_LDM_GEARTAB_H + #include "../common/compiler.h" /* UNUSED_ATTR */ #include "../common/mem.h" /* U64 */ static UNUSED_ATTR const U64 ZSTD_ldm_gearTab[256] = { - 0xf5b8f72c5f77775c, 0x84935f266b7ac412, 0xb647ada9ca730ccc, - 0xb065bb4b114fb1de, 0x34584e7e8c3a9fd0, 0x4e97e17c6ae26b05, - 0x3a03d743bc99a604, 0xcecd042422c4044f, 0x76de76c58524259e, - 0x9c8528f65badeaca, 0x86563706e2097529, 0x2902475fa375d889, - 0xafb32a9739a5ebe6, 0xce2714da3883e639, 0x21eaf821722e69e, - 0x37b628620b628, 0x49a8d455d88caf5, 0x8556d711e6958140, - 0x4f7ae74fc605c1f, 0x829f0c3468bd3a20, 0x4ffdc885c625179e, - 0x8473de048a3daf1b, 0x51008822b05646b2, 0x69d75d12b2d1cc5f, - 0x8c9d4a19159154bc, 0xc3cc10f4abbd4003, 0xd06ddc1cecb97391, - 0xbe48e6e7ed80302e, 0x3481db31cee03547, 0xacc3f67cdaa1d210, - 0x65cb771d8c7f96cc, 0x8eb27177055723dd, 0xc789950d44cd94be, - 0x934feadc3700b12b, 0x5e485f11edbdf182, 0x1e2e2a46fd64767a, - 0x2969ca71d82efa7c, 0x9d46e9935ebbba2e, 0xe056b67e05e6822b, - 0x94d73f55739d03a0, 0xcd7010bdb69b5a03, 0x455ef9fcd79b82f4, - 0x869cb54a8749c161, 0x38d1a4fa6185d225, 0xb475166f94bbe9bb, - 0xa4143548720959f1, 0x7aed4780ba6b26ba, 0xd0ce264439e02312, - 0x84366d746078d508, 0xa8ce973c72ed17be, 0x21c323a29a430b01, - 0x9962d617e3af80ee, 0xab0ce91d9c8cf75b, 0x530e8ee6d19a4dbc, - 0x2ef68c0cf53f5d72, 0xc03a681640a85506, 0x496e4e9f9c310967, - 0x78580472b59b14a0, 0x273824c23b388577, 0x66bf923ad45cb553, - 0x47ae1a5a2492ba86, 0x35e304569e229659, 0x4765182a46870b6f, - 0x6cbab625e9099412, 0xddac9a2e598522c1, 0x7172086e666624f2, - 0xdf5003ca503b7837, 0x88c0c1db78563d09, 0x58d51865acfc289d, - 0x177671aec65224f1, 0xfb79d8a241e967d7, 0x2be1e101cad9a49a, - 0x6625682f6e29186b, 0x399553457ac06e50, 0x35dffb4c23abb74, - 0x429db2591f54aade, 0xc52802a8037d1009, 0x6acb27381f0b25f3, - 0xf45e2551ee4f823b, 0x8b0ea2d99580c2f7, 0x3bed519cbcb4e1e1, - 0xff452823dbb010a, 0x9d42ed614f3dd267, 0x5b9313c06257c57b, - 0xa114b8008b5e1442, 0xc1fe311c11c13d4b, 0x66e8763ea34c5568, - 0x8b982af1c262f05d, 0xee8876faaa75fbb7, 0x8a62a4d0d172bb2a, - 0xc13d94a3b7449a97, 0x6dbbba9dc15d037c, 0xc786101f1d92e0f1, - 0xd78681a907a0b79b, 0xf61aaf2962c9abb9, 0x2cfd16fcd3cb7ad9, - 0x868c5b6744624d21, 0x25e650899c74ddd7, 0xba042af4a7c37463, - 0x4eb1a539465a3eca, 0xbe09dbf03b05d5ca, 0x774e5a362b5472ba, - 0x47a1221229d183cd, 0x504b0ca18ef5a2df, 0xdffbdfbde2456eb9, - 0x46cd2b2fbee34634, 0xf2aef8fe819d98c3, 0x357f5276d4599d61, - 0x24a5483879c453e3, 0x88026889192b4b9, 0x28da96671782dbec, - 0x4ef37c40588e9aaa, 0x8837b90651bc9fb3, 0xc164f741d3f0e5d6, - 0xbc135a0a704b70ba, 0x69cd868f7622ada, 0xbc37ba89e0b9c0ab, - 0x47c14a01323552f6, 0x4f00794bacee98bb, 0x7107de7d637a69d5, - 0x88af793bb6f2255e, 0xf3c6466b8799b598, 0xc288c616aa7f3b59, - 0x81ca63cf42fca3fd, 0x88d85ace36a2674b, 0xd056bd3792389e7, - 0xe55c396c4e9dd32d, 0xbefb504571e6c0a6, 0x96ab32115e91e8cc, - 0xbf8acb18de8f38d1, 0x66dae58801672606, 0x833b6017872317fb, - 0xb87c16f2d1c92864, 0xdb766a74e58b669c, 0x89659f85c61417be, - 0xc8daad856011ea0c, 0x76a4b565b6fe7eae, 0xa469d085f6237312, - 0xaaf0365683a3e96c, 0x4dbb746f8424f7b8, 0x638755af4e4acc1, - 0x3d7807f5bde64486, 0x17be6d8f5bbb7639, 0x903f0cd44dc35dc, - 0x67b672eafdf1196c, 0xa676ff93ed4c82f1, 0x521d1004c5053d9d, - 0x37ba9ad09ccc9202, 0x84e54d297aacfb51, 0xa0b4b776a143445, - 0x820d471e20b348e, 0x1874383cb83d46dc, 0x97edeec7a1efe11c, - 0xb330e50b1bdc42aa, 0x1dd91955ce70e032, 0xa514cdb88f2939d5, - 0x2791233fd90db9d3, 0x7b670a4cc50f7a9b, 0x77c07d2a05c6dfa5, - 0xe3778b6646d0a6fa, 0xb39c8eda47b56749, 0x933ed448addbef28, - 0xaf846af6ab7d0bf4, 0xe5af208eb666e49, 0x5e6622f73534cd6a, - 0x297daeca42ef5b6e, 0x862daef3d35539a6, 0xe68722498f8e1ea9, - 0x981c53093dc0d572, 0xfa09b0bfbf86fbf5, 0x30b1e96166219f15, - 0x70e7d466bdc4fb83, 0x5a66736e35f2a8e9, 0xcddb59d2b7c1baef, - 0xd6c7d247d26d8996, 0xea4e39eac8de1ba3, 0x539c8bb19fa3aff2, - 0x9f90e4c5fd508d8, 0xa34e5956fbaf3385, 0x2e2f8e151d3ef375, - 0x173691e9b83faec1, 0xb85a8d56bf016379, 0x8382381267408ae3, - 0xb90f901bbdc0096d, 0x7c6ad32933bcec65, 0x76bb5e2f2c8ad595, - 0x390f851a6cf46d28, 0xc3e6064da1c2da72, 0xc52a0c101cfa5389, - 0xd78eaf84a3fbc530, 0x3781b9e2288b997e, 0x73c2f6dea83d05c4, - 0x4228e364c5b5ed7, 0x9d7a3edf0da43911, 0x8edcfeda24686756, - 0x5e7667a7b7a9b3a1, 0x4c4f389fa143791d, 0xb08bc1023da7cddc, - 0x7ab4be3ae529b1cc, 0x754e6132dbe74ff9, 0x71635442a839df45, - 0x2f6fb1643fbe52de, 0x961e0a42cf7a8177, 0xf3b45d83d89ef2ea, - 0xee3de4cf4a6e3e9b, 0xcd6848542c3295e7, 0xe4cee1664c78662f, - 0x9947548b474c68c4, 0x25d73777a5ed8b0b, 0xc915b1d636b7fc, - 0x21c2ba75d9b0d2da, 0x5f6b5dcf608a64a1, 0xdcf333255ff9570c, - 0x633b922418ced4ee, 0xc136dde0b004b34a, 0x58cc83b05d4b2f5a, - 0x5eb424dda28e42d2, 0x62df47369739cd98, 0xb4e0b42485e4ce17, - 0x16e1f0c1f9a8d1e7, 0x8ec3916707560ebf, 0x62ba6e2df2cc9db3, - 0xcbf9f4ff77d83a16, 0x78d9d7d07d2bbcc4, 0xef554ce1e02c41f4, - 0x8d7581127eccf94d, 0xa9b53336cb3c8a05, 0x38c42c0bf45c4f91, - 0x640893cdf4488863, 0x80ec34bc575ea568, 0x39f324f5b48eaa40, - 0xe9d9ed1f8eff527f, 0x9224fc058cc5a214, 0xbaba00b04cfe7741, - 0x309a9f120fcf52af, 0xa558f3ec65626212, 0x424bec8b7adabe2f, - 0x41622513a6aea433, 0xb88da2d5324ca798, 0xd287733b245528a4, - 0x9a44697e6d68aec3, 0x7b1093be2f49bb28, 0x50bbec632e3d8aad, - 0x6cd90723e1ea8283, 0x897b9e7431b02bf3, 0x219efdcb338a7047, - 0x3b0311f0a27c0656, 0xdb17bf91c0db96e7, 0x8cd4fd6b4e85a5b2, - 0xfab071054ba6409d, 0x40d6fe831fa9dfd9, 0xaf358debad7d791e, - 0xeb8d0e25a65e3e58, 0xbbcbd3df14e08580, 0xcf751f27ecdab2b, - 0x2b4da14f2613d8f4 -}; - -#endif /* ZSTD_LDM_GEARTAB_H */ + 0xf5b8f72c5f77775c, 0x84935f266b7ac412, 0xb647ada9ca730ccc, + 0xb065bb4b114fb1de, 0x34584e7e8c3a9fd0, 0x4e97e17c6ae26b05, + 0x3a03d743bc99a604, 0xcecd042422c4044f, 0x76de76c58524259e, + 0x9c8528f65badeaca, 0x86563706e2097529, 0x2902475fa375d889, + 0xafb32a9739a5ebe6, 0xce2714da3883e639, 0x21eaf821722e69e, + 0x37b628620b628, 0x49a8d455d88caf5, 0x8556d711e6958140, + 0x4f7ae74fc605c1f, 0x829f0c3468bd3a20, 0x4ffdc885c625179e, + 0x8473de048a3daf1b, 0x51008822b05646b2, 0x69d75d12b2d1cc5f, + 0x8c9d4a19159154bc, 0xc3cc10f4abbd4003, 0xd06ddc1cecb97391, + 0xbe48e6e7ed80302e, 0x3481db31cee03547, 0xacc3f67cdaa1d210, + 0x65cb771d8c7f96cc, 0x8eb27177055723dd, 0xc789950d44cd94be, + 0x934feadc3700b12b, 0x5e485f11edbdf182, 0x1e2e2a46fd64767a, + 0x2969ca71d82efa7c, 0x9d46e9935ebbba2e, 0xe056b67e05e6822b, + 0x94d73f55739d03a0, 0xcd7010bdb69b5a03, 0x455ef9fcd79b82f4, + 0x869cb54a8749c161, 0x38d1a4fa6185d225, 0xb475166f94bbe9bb, + 0xa4143548720959f1, 0x7aed4780ba6b26ba, 0xd0ce264439e02312, + 0x84366d746078d508, 0xa8ce973c72ed17be, 0x21c323a29a430b01, + 0x9962d617e3af80ee, 0xab0ce91d9c8cf75b, 0x530e8ee6d19a4dbc, + 0x2ef68c0cf53f5d72, 0xc03a681640a85506, 0x496e4e9f9c310967, + 0x78580472b59b14a0, 0x273824c23b388577, 0x66bf923ad45cb553, + 0x47ae1a5a2492ba86, 0x35e304569e229659, 0x4765182a46870b6f, + 0x6cbab625e9099412, 0xddac9a2e598522c1, 0x7172086e666624f2, + 0xdf5003ca503b7837, 0x88c0c1db78563d09, 0x58d51865acfc289d, + 0x177671aec65224f1, 0xfb79d8a241e967d7, 0x2be1e101cad9a49a, + 0x6625682f6e29186b, 0x399553457ac06e50, 0x35dffb4c23abb74, + 0x429db2591f54aade, 0xc52802a8037d1009, 0x6acb27381f0b25f3, + 0xf45e2551ee4f823b, 0x8b0ea2d99580c2f7, 0x3bed519cbcb4e1e1, + 0xff452823dbb010a, 0x9d42ed614f3dd267, 0x5b9313c06257c57b, + 0xa114b8008b5e1442, 0xc1fe311c11c13d4b, 0x66e8763ea34c5568, + 0x8b982af1c262f05d, 0xee8876faaa75fbb7, 0x8a62a4d0d172bb2a, + 0xc13d94a3b7449a97, 0x6dbbba9dc15d037c, 0xc786101f1d92e0f1, + 0xd78681a907a0b79b, 0xf61aaf2962c9abb9, 0x2cfd16fcd3cb7ad9, + 0x868c5b6744624d21, 0x25e650899c74ddd7, 0xba042af4a7c37463, + 0x4eb1a539465a3eca, 0xbe09dbf03b05d5ca, 0x774e5a362b5472ba, + 0x47a1221229d183cd, 0x504b0ca18ef5a2df, 0xdffbdfbde2456eb9, + 0x46cd2b2fbee34634, 0xf2aef8fe819d98c3, 0x357f5276d4599d61, + 0x24a5483879c453e3, 0x88026889192b4b9, 0x28da96671782dbec, + 0x4ef37c40588e9aaa, 0x8837b90651bc9fb3, 0xc164f741d3f0e5d6, + 0xbc135a0a704b70ba, 0x69cd868f7622ada, 0xbc37ba89e0b9c0ab, + 0x47c14a01323552f6, 0x4f00794bacee98bb, 0x7107de7d637a69d5, + 0x88af793bb6f2255e, 0xf3c6466b8799b598, 0xc288c616aa7f3b59, + 0x81ca63cf42fca3fd, 0x88d85ace36a2674b, 0xd056bd3792389e7, + 0xe55c396c4e9dd32d, 0xbefb504571e6c0a6, 0x96ab32115e91e8cc, + 0xbf8acb18de8f38d1, 0x66dae58801672606, 0x833b6017872317fb, + 0xb87c16f2d1c92864, 0xdb766a74e58b669c, 0x89659f85c61417be, + 0xc8daad856011ea0c, 0x76a4b565b6fe7eae, 0xa469d085f6237312, + 0xaaf0365683a3e96c, 0x4dbb746f8424f7b8, 0x638755af4e4acc1, + 0x3d7807f5bde64486, 0x17be6d8f5bbb7639, 0x903f0cd44dc35dc, + 0x67b672eafdf1196c, 0xa676ff93ed4c82f1, 0x521d1004c5053d9d, + 0x37ba9ad09ccc9202, 0x84e54d297aacfb51, 0xa0b4b776a143445, + 0x820d471e20b348e, 0x1874383cb83d46dc, 0x97edeec7a1efe11c, + 0xb330e50b1bdc42aa, 0x1dd91955ce70e032, 0xa514cdb88f2939d5, + 0x2791233fd90db9d3, 0x7b670a4cc50f7a9b, 0x77c07d2a05c6dfa5, + 0xe3778b6646d0a6fa, 0xb39c8eda47b56749, 0x933ed448addbef28, + 0xaf846af6ab7d0bf4, 0xe5af208eb666e49, 0x5e6622f73534cd6a, + 0x297daeca42ef5b6e, 0x862daef3d35539a6, 0xe68722498f8e1ea9, + 0x981c53093dc0d572, 0xfa09b0bfbf86fbf5, 0x30b1e96166219f15, + 0x70e7d466bdc4fb83, 0x5a66736e35f2a8e9, 0xcddb59d2b7c1baef, + 0xd6c7d247d26d8996, 0xea4e39eac8de1ba3, 0x539c8bb19fa3aff2, + 0x9f90e4c5fd508d8, 0xa34e5956fbaf3385, 0x2e2f8e151d3ef375, + 0x173691e9b83faec1, 0xb85a8d56bf016379, 0x8382381267408ae3, + 0xb90f901bbdc0096d, 0x7c6ad32933bcec65, 0x76bb5e2f2c8ad595, + 0x390f851a6cf46d28, 0xc3e6064da1c2da72, 0xc52a0c101cfa5389, + 0xd78eaf84a3fbc530, 0x3781b9e2288b997e, 0x73c2f6dea83d05c4, + 0x4228e364c5b5ed7, 0x9d7a3edf0da43911, 0x8edcfeda24686756, + 0x5e7667a7b7a9b3a1, 0x4c4f389fa143791d, 0xb08bc1023da7cddc, + 0x7ab4be3ae529b1cc, 0x754e6132dbe74ff9, 0x71635442a839df45, + 0x2f6fb1643fbe52de, 0x961e0a42cf7a8177, 0xf3b45d83d89ef2ea, + 0xee3de4cf4a6e3e9b, 0xcd6848542c3295e7, 0xe4cee1664c78662f, + 0x9947548b474c68c4, 0x25d73777a5ed8b0b, 0xc915b1d636b7fc, + 0x21c2ba75d9b0d2da, 0x5f6b5dcf608a64a1, 0xdcf333255ff9570c, + 0x633b922418ced4ee, 0xc136dde0b004b34a, 0x58cc83b05d4b2f5a, + 0x5eb424dda28e42d2, 0x62df47369739cd98, 0xb4e0b42485e4ce17, + 0x16e1f0c1f9a8d1e7, 0x8ec3916707560ebf, 0x62ba6e2df2cc9db3, + 0xcbf9f4ff77d83a16, 0x78d9d7d07d2bbcc4, 0xef554ce1e02c41f4, + 0x8d7581127eccf94d, 0xa9b53336cb3c8a05, 0x38c42c0bf45c4f91, + 0x640893cdf4488863, 0x80ec34bc575ea568, 0x39f324f5b48eaa40, + 0xe9d9ed1f8eff527f, 0x9224fc058cc5a214, 0xbaba00b04cfe7741, + 0x309a9f120fcf52af, 0xa558f3ec65626212, 0x424bec8b7adabe2f, + 0x41622513a6aea433, 0xb88da2d5324ca798, 0xd287733b245528a4, + 0x9a44697e6d68aec3, 0x7b1093be2f49bb28, 0x50bbec632e3d8aad, + 0x6cd90723e1ea8283, 0x897b9e7431b02bf3, 0x219efdcb338a7047, + 0x3b0311f0a27c0656, 0xdb17bf91c0db96e7, 0x8cd4fd6b4e85a5b2, + 0xfab071054ba6409d, 0x40d6fe831fa9dfd9, 0xaf358debad7d791e, + 0xeb8d0e25a65e3e58, 0xbbcbd3df14e08580, 0xcf751f27ecdab2b, + 0x2b4da14f2613d8f4 +}; + +#endif /* ZSTD_LDM_GEARTAB_H */ diff --git a/contrib/libs/zstd/lib/compress/zstd_opt.c b/contrib/libs/zstd/lib/compress/zstd_opt.c index 1b1ddad428..2d5e0935ee 100644 --- a/contrib/libs/zstd/lib/compress/zstd_opt.c +++ b/contrib/libs/zstd/lib/compress/zstd_opt.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Przemyslaw Skibinski, Yann Collet, Facebook, Inc. + * Copyright (c) Przemyslaw Skibinski, Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -429,36 +429,36 @@ static U32 ZSTD_insertBt1( const BYTE* const dictEnd = dictBase + dictLimit; const BYTE* const prefixStart = base + dictLimit; const BYTE* match; - const U32 curr = (U32)(ip-base); - const U32 btLow = btMask >= curr ? 0 : curr - btMask; - U32* smallerPtr = bt + 2*(curr&btMask); + const U32 curr = (U32)(ip-base); + const U32 btLow = btMask >= curr ? 0 : curr - btMask; + U32* smallerPtr = bt + 2*(curr&btMask); U32* largerPtr = smallerPtr + 1; U32 dummy32; /* to be nullified at the end */ /* windowLow is based on target because * we only need positions that will be in the window at the end of the tree update. */ U32 const windowLow = ZSTD_getLowestMatchIndex(ms, target, cParams->windowLog); - U32 matchEndIdx = curr+8+1; + U32 matchEndIdx = curr+8+1; size_t bestLength = 8; U32 nbCompares = 1U << cParams->searchLog; #ifdef ZSTD_C_PREDICT - U32 predictedSmall = *(bt + 2*((curr-1)&btMask) + 0); - U32 predictedLarge = *(bt + 2*((curr-1)&btMask) + 1); + U32 predictedSmall = *(bt + 2*((curr-1)&btMask) + 0); + U32 predictedLarge = *(bt + 2*((curr-1)&btMask) + 1); predictedSmall += (predictedSmall>0); predictedLarge += (predictedLarge>0); #endif /* ZSTD_C_PREDICT */ - DEBUGLOG(8, "ZSTD_insertBt1 (%u)", curr); + DEBUGLOG(8, "ZSTD_insertBt1 (%u)", curr); assert(curr <= target); assert(ip <= iend-8); /* required for h calculation */ - hashTable[h] = curr; /* Update Hash Table */ + hashTable[h] = curr; /* Update Hash Table */ assert(windowLow > 0); for (; nbCompares && (matchIndex >= windowLow); --nbCompares) { U32* const nextPtr = bt + 2*(matchIndex & btMask); size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */ - assert(matchIndex < curr); + assert(matchIndex < curr); #ifdef ZSTD_C_PREDICT /* note : can create issues when hlog small <= 11 */ const U32* predictPtr = bt + 2*((matchIndex-1) & btMask); /* written this way, as bt is a roll buffer */ @@ -521,8 +521,8 @@ static U32 ZSTD_insertBt1( *smallerPtr = *largerPtr = 0; { U32 positions = 0; if (bestLength > 384) positions = MIN(192, (U32)(bestLength - 384)); /* speed optimization */ - assert(matchEndIdx > curr + 8); - return MAX(positions, matchEndIdx - (curr + 8)); + assert(matchEndIdx > curr + 8); + return MAX(positions, matchEndIdx - (curr + 8)); } } @@ -566,7 +566,7 @@ U32 ZSTD_insertBtAndGetAllMatches ( const ZSTD_compressionParameters* const cParams = &ms->cParams; U32 const sufficient_len = MIN(cParams->targetLength, ZSTD_OPT_NUM -1); const BYTE* const base = ms->window.base; - U32 const curr = (U32)(ip-base); + U32 const curr = (U32)(ip-base); U32 const hashLog = cParams->hashLog; U32 const minMatch = (mls==3) ? 3 : 4; U32* const hashTable = ms->hashTable; @@ -580,12 +580,12 @@ U32 ZSTD_insertBtAndGetAllMatches ( U32 const dictLimit = ms->window.dictLimit; const BYTE* const dictEnd = dictBase + dictLimit; const BYTE* const prefixStart = base + dictLimit; - U32 const btLow = (btMask >= curr) ? 0 : curr - btMask; - U32 const windowLow = ZSTD_getLowestMatchIndex(ms, curr, cParams->windowLog); + U32 const btLow = (btMask >= curr) ? 0 : curr - btMask; + U32 const windowLow = ZSTD_getLowestMatchIndex(ms, curr, cParams->windowLog); U32 const matchLow = windowLow ? windowLow : 1; - U32* smallerPtr = bt + 2*(curr&btMask); - U32* largerPtr = bt + 2*(curr&btMask) + 1; - U32 matchEndIdx = curr+8+1; /* farthest referenced position of any match => detects repetitive patterns */ + U32* smallerPtr = bt + 2*(curr&btMask); + U32* largerPtr = bt + 2*(curr&btMask) + 1; + U32 matchEndIdx = curr+8+1; /* farthest referenced position of any match => detects repetitive patterns */ U32 dummy32; /* to be nullified at the end */ U32 mnum = 0; U32 nbCompares = 1U << cParams->searchLog; @@ -604,7 +604,7 @@ U32 ZSTD_insertBtAndGetAllMatches ( U32 const dmsBtLow = dictMode == ZSTD_dictMatchState && dmsBtMask < dmsHighLimit - dmsLowLimit ? dmsHighLimit - dmsBtMask : dmsLowLimit; size_t bestLength = lengthToBeat-1; - DEBUGLOG(8, "ZSTD_insertBtAndGetAllMatches: current=%u", curr); + DEBUGLOG(8, "ZSTD_insertBtAndGetAllMatches: current=%u", curr); /* check repCode */ assert(ll0 <= 1); /* necessarily 1 or 0 */ @@ -612,29 +612,29 @@ U32 ZSTD_insertBtAndGetAllMatches ( U32 repCode; for (repCode = ll0; repCode < lastR; repCode++) { U32 const repOffset = (repCode==ZSTD_REP_NUM) ? (rep[0] - 1) : rep[repCode]; - U32 const repIndex = curr - repOffset; + U32 const repIndex = curr - repOffset; U32 repLen = 0; - assert(curr >= dictLimit); - if (repOffset-1 /* intentional overflow, discards 0 and -1 */ < curr-dictLimit) { /* equivalent to `curr > repIndex >= dictLimit` */ + assert(curr >= dictLimit); + if (repOffset-1 /* intentional overflow, discards 0 and -1 */ < curr-dictLimit) { /* equivalent to `curr > repIndex >= dictLimit` */ /* We must validate the repcode offset because when we're using a dictionary the * valid offset range shrinks when the dictionary goes out of bounds. */ if ((repIndex >= windowLow) & (ZSTD_readMINMATCH(ip, minMatch) == ZSTD_readMINMATCH(ip - repOffset, minMatch))) { repLen = (U32)ZSTD_count(ip+minMatch, ip+minMatch-repOffset, iLimit) + minMatch; } - } else { /* repIndex < dictLimit || repIndex >= curr */ + } else { /* repIndex < dictLimit || repIndex >= curr */ const BYTE* const repMatch = dictMode == ZSTD_dictMatchState ? dmsBase + repIndex - dmsIndexDelta : dictBase + repIndex; - assert(curr >= windowLow); + assert(curr >= windowLow); if ( dictMode == ZSTD_extDict - && ( ((repOffset-1) /*intentional overflow*/ < curr - windowLow) /* equivalent to `curr > repIndex >= windowLow` */ + && ( ((repOffset-1) /*intentional overflow*/ < curr - windowLow) /* equivalent to `curr > repIndex >= windowLow` */ & (((U32)((dictLimit-1) - repIndex) >= 3) ) /* intentional overflow : do not test positions overlapping 2 memory segments */) && (ZSTD_readMINMATCH(ip, minMatch) == ZSTD_readMINMATCH(repMatch, minMatch)) ) { repLen = (U32)ZSTD_count_2segments(ip+minMatch, repMatch+minMatch, iLimit, dictEnd, prefixStart) + minMatch; } if (dictMode == ZSTD_dictMatchState - && ( ((repOffset-1) /*intentional overflow*/ < curr - (dmsLowLimit + dmsIndexDelta)) /* equivalent to `curr > repIndex >= dmsLowLimit` */ + && ( ((repOffset-1) /*intentional overflow*/ < curr - (dmsLowLimit + dmsIndexDelta)) /* equivalent to `curr > repIndex >= dmsLowLimit` */ & ((U32)((dictLimit-1) - repIndex) >= 3) ) /* intentional overflow : do not test positions overlapping 2 memory segments */ && (ZSTD_readMINMATCH(ip, minMatch) == ZSTD_readMINMATCH(repMatch, minMatch)) ) { repLen = (U32)ZSTD_count_2segments(ip+minMatch, repMatch+minMatch, iLimit, dmsEnd, prefixStart) + minMatch; @@ -656,7 +656,7 @@ U32 ZSTD_insertBtAndGetAllMatches ( if ((mls == 3) /*static*/ && (bestLength < mls)) { U32 const matchIndex3 = ZSTD_insertAndFindFirstIndexHash3(ms, nextToUpdate3, ip); if ((matchIndex3 >= matchLow) - & (curr - matchIndex3 < (1<<18)) /*heuristic : longer distance likely too expensive*/ ) { + & (curr - matchIndex3 < (1<<18)) /*heuristic : longer distance likely too expensive*/ ) { size_t mlen; if ((dictMode == ZSTD_noDict) /*static*/ || (dictMode == ZSTD_dictMatchState) /*static*/ || (matchIndex3 >= dictLimit)) { const BYTE* const match = base + matchIndex3; @@ -671,26 +671,26 @@ U32 ZSTD_insertBtAndGetAllMatches ( DEBUGLOG(8, "found small match with hlog3, of length %u", (U32)mlen); bestLength = mlen; - assert(curr > matchIndex3); + assert(curr > matchIndex3); assert(mnum==0); /* no prior solution */ matches[0].off = STORE_OFFSET(curr - matchIndex3); matches[0].len = (U32)mlen; mnum = 1; if ( (mlen > sufficient_len) | (ip+mlen == iLimit) ) { /* best possible length */ - ms->nextToUpdate = curr+1; /* skip insertion */ + ms->nextToUpdate = curr+1; /* skip insertion */ return 1; } } } /* no dictMatchState lookup: dicts don't have a populated HC3 table */ } /* if (mls == 3) */ - hashTable[h] = curr; /* Update Hash Table */ + hashTable[h] = curr; /* Update Hash Table */ for (; nbCompares && (matchIndex >= matchLow); --nbCompares) { U32* const nextPtr = bt + 2*(matchIndex & btMask); const BYTE* match; size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */ - assert(curr > matchIndex); + assert(curr > matchIndex); if ((dictMode == ZSTD_noDict) || (dictMode == ZSTD_dictMatchState) || (matchIndex+matchLength >= dictLimit)) { assert(matchIndex+matchLength >= dictLimit); /* ensure the condition is correct when !extDict */ @@ -777,7 +777,7 @@ U32 ZSTD_insertBtAndGetAllMatches ( dictMatchIndex = nextPtr[0]; } } } /* if (dictMode == ZSTD_dictMatchState) */ - assert(matchEndIdx > curr+8); + assert(matchEndIdx > curr+8); ms->nextToUpdate = matchEndIdx - 8; /* skip repetitive patterns */ return mnum; } @@ -862,150 +862,150 @@ ZSTD_selectBtGetAllMatches(ZSTD_matchState_t const* ms, ZSTD_dictMode_e const di return getAllMatchesFns[(int)dictMode][mls - 3]; } -/************************* -* LDM helper functions * -*************************/ +/************************* +* LDM helper functions * +*************************/ -/* Struct containing info needed to make decision about ldm inclusion */ -typedef struct { +/* Struct containing info needed to make decision about ldm inclusion */ +typedef struct { rawSeqStore_t seqStore; /* External match candidates store for this block */ U32 startPosInBlock; /* Start position of the current match candidate */ U32 endPosInBlock; /* End position of the current match candidate */ U32 offset; /* Offset of the match candidate */ -} ZSTD_optLdm_t; - -/* ZSTD_optLdm_skipRawSeqStoreBytes(): +} ZSTD_optLdm_t; + +/* ZSTD_optLdm_skipRawSeqStoreBytes(): * Moves forward in @rawSeqStore by @nbBytes, * which will update the fields 'pos' and 'posInSequence'. - */ + */ static void ZSTD_optLdm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes) { - U32 currPos = (U32)(rawSeqStore->posInSequence + nbBytes); - while (currPos && rawSeqStore->pos < rawSeqStore->size) { - rawSeq currSeq = rawSeqStore->seq[rawSeqStore->pos]; - if (currPos >= currSeq.litLength + currSeq.matchLength) { - currPos -= currSeq.litLength + currSeq.matchLength; - rawSeqStore->pos++; - } else { - rawSeqStore->posInSequence = currPos; - break; - } - } - if (currPos == 0 || rawSeqStore->pos == rawSeqStore->size) { - rawSeqStore->posInSequence = 0; - } -} - -/* ZSTD_opt_getNextMatchAndUpdateSeqStore(): - * Calculates the beginning and end of the next match in the current block. - * Updates 'pos' and 'posInSequence' of the ldmSeqStore. - */ + U32 currPos = (U32)(rawSeqStore->posInSequence + nbBytes); + while (currPos && rawSeqStore->pos < rawSeqStore->size) { + rawSeq currSeq = rawSeqStore->seq[rawSeqStore->pos]; + if (currPos >= currSeq.litLength + currSeq.matchLength) { + currPos -= currSeq.litLength + currSeq.matchLength; + rawSeqStore->pos++; + } else { + rawSeqStore->posInSequence = currPos; + break; + } + } + if (currPos == 0 || rawSeqStore->pos == rawSeqStore->size) { + rawSeqStore->posInSequence = 0; + } +} + +/* ZSTD_opt_getNextMatchAndUpdateSeqStore(): + * Calculates the beginning and end of the next match in the current block. + * Updates 'pos' and 'posInSequence' of the ldmSeqStore. + */ static void ZSTD_opt_getNextMatchAndUpdateSeqStore(ZSTD_optLdm_t* optLdm, U32 currPosInBlock, U32 blockBytesRemaining) { - rawSeq currSeq; - U32 currBlockEndPos; - U32 literalsBytesRemaining; - U32 matchBytesRemaining; - - /* Setting match end position to MAX to ensure we never use an LDM during this block */ - if (optLdm->seqStore.size == 0 || optLdm->seqStore.pos >= optLdm->seqStore.size) { - optLdm->startPosInBlock = UINT_MAX; - optLdm->endPosInBlock = UINT_MAX; - return; - } + rawSeq currSeq; + U32 currBlockEndPos; + U32 literalsBytesRemaining; + U32 matchBytesRemaining; + + /* Setting match end position to MAX to ensure we never use an LDM during this block */ + if (optLdm->seqStore.size == 0 || optLdm->seqStore.pos >= optLdm->seqStore.size) { + optLdm->startPosInBlock = UINT_MAX; + optLdm->endPosInBlock = UINT_MAX; + return; + } /* Calculate appropriate bytes left in matchLength and litLength * after adjusting based on ldmSeqStore->posInSequence */ - currSeq = optLdm->seqStore.seq[optLdm->seqStore.pos]; - assert(optLdm->seqStore.posInSequence <= currSeq.litLength + currSeq.matchLength); - currBlockEndPos = currPosInBlock + blockBytesRemaining; - literalsBytesRemaining = (optLdm->seqStore.posInSequence < currSeq.litLength) ? - currSeq.litLength - (U32)optLdm->seqStore.posInSequence : - 0; - matchBytesRemaining = (literalsBytesRemaining == 0) ? - currSeq.matchLength - ((U32)optLdm->seqStore.posInSequence - currSeq.litLength) : - currSeq.matchLength; - - /* If there are more literal bytes than bytes remaining in block, no ldm is possible */ - if (literalsBytesRemaining >= blockBytesRemaining) { - optLdm->startPosInBlock = UINT_MAX; - optLdm->endPosInBlock = UINT_MAX; - ZSTD_optLdm_skipRawSeqStoreBytes(&optLdm->seqStore, blockBytesRemaining); - return; - } - - /* Matches may be < MINMATCH by this process. In that case, we will reject them - when we are deciding whether or not to add the ldm */ - optLdm->startPosInBlock = currPosInBlock + literalsBytesRemaining; - optLdm->endPosInBlock = optLdm->startPosInBlock + matchBytesRemaining; - optLdm->offset = currSeq.offset; - - if (optLdm->endPosInBlock > currBlockEndPos) { - /* Match ends after the block ends, we can't use the whole match */ - optLdm->endPosInBlock = currBlockEndPos; - ZSTD_optLdm_skipRawSeqStoreBytes(&optLdm->seqStore, currBlockEndPos - currPosInBlock); - } else { - /* Consume nb of bytes equal to size of sequence left */ - ZSTD_optLdm_skipRawSeqStoreBytes(&optLdm->seqStore, literalsBytesRemaining + matchBytesRemaining); - } -} - -/* ZSTD_optLdm_maybeAddMatch(): + currSeq = optLdm->seqStore.seq[optLdm->seqStore.pos]; + assert(optLdm->seqStore.posInSequence <= currSeq.litLength + currSeq.matchLength); + currBlockEndPos = currPosInBlock + blockBytesRemaining; + literalsBytesRemaining = (optLdm->seqStore.posInSequence < currSeq.litLength) ? + currSeq.litLength - (U32)optLdm->seqStore.posInSequence : + 0; + matchBytesRemaining = (literalsBytesRemaining == 0) ? + currSeq.matchLength - ((U32)optLdm->seqStore.posInSequence - currSeq.litLength) : + currSeq.matchLength; + + /* If there are more literal bytes than bytes remaining in block, no ldm is possible */ + if (literalsBytesRemaining >= blockBytesRemaining) { + optLdm->startPosInBlock = UINT_MAX; + optLdm->endPosInBlock = UINT_MAX; + ZSTD_optLdm_skipRawSeqStoreBytes(&optLdm->seqStore, blockBytesRemaining); + return; + } + + /* Matches may be < MINMATCH by this process. In that case, we will reject them + when we are deciding whether or not to add the ldm */ + optLdm->startPosInBlock = currPosInBlock + literalsBytesRemaining; + optLdm->endPosInBlock = optLdm->startPosInBlock + matchBytesRemaining; + optLdm->offset = currSeq.offset; + + if (optLdm->endPosInBlock > currBlockEndPos) { + /* Match ends after the block ends, we can't use the whole match */ + optLdm->endPosInBlock = currBlockEndPos; + ZSTD_optLdm_skipRawSeqStoreBytes(&optLdm->seqStore, currBlockEndPos - currPosInBlock); + } else { + /* Consume nb of bytes equal to size of sequence left */ + ZSTD_optLdm_skipRawSeqStoreBytes(&optLdm->seqStore, literalsBytesRemaining + matchBytesRemaining); + } +} + +/* ZSTD_optLdm_maybeAddMatch(): * Adds a match if it's long enough, * based on it's 'matchStartPosInBlock' and 'matchEndPosInBlock', * into 'matches'. Maintains the correct ordering of 'matches'. - */ -static void ZSTD_optLdm_maybeAddMatch(ZSTD_match_t* matches, U32* nbMatches, + */ +static void ZSTD_optLdm_maybeAddMatch(ZSTD_match_t* matches, U32* nbMatches, const ZSTD_optLdm_t* optLdm, U32 currPosInBlock) { U32 const posDiff = currPosInBlock - optLdm->startPosInBlock; - /* Note: ZSTD_match_t actually contains offCode and matchLength (before subtracting MINMATCH) */ + /* Note: ZSTD_match_t actually contains offCode and matchLength (before subtracting MINMATCH) */ U32 const candidateMatchLength = optLdm->endPosInBlock - optLdm->startPosInBlock - posDiff; - - /* Ensure that current block position is not outside of the match */ - if (currPosInBlock < optLdm->startPosInBlock - || currPosInBlock >= optLdm->endPosInBlock - || candidateMatchLength < MINMATCH) { - return; - } - - if (*nbMatches == 0 || ((candidateMatchLength > matches[*nbMatches-1].len) && *nbMatches < ZSTD_OPT_NUM)) { + + /* Ensure that current block position is not outside of the match */ + if (currPosInBlock < optLdm->startPosInBlock + || currPosInBlock >= optLdm->endPosInBlock + || candidateMatchLength < MINMATCH) { + return; + } + + if (*nbMatches == 0 || ((candidateMatchLength > matches[*nbMatches-1].len) && *nbMatches < ZSTD_OPT_NUM)) { U32 const candidateOffCode = STORE_OFFSET(optLdm->offset); - DEBUGLOG(6, "ZSTD_optLdm_maybeAddMatch(): Adding ldm candidate match (offCode: %u matchLength %u) at block position=%u", - candidateOffCode, candidateMatchLength, currPosInBlock); - matches[*nbMatches].len = candidateMatchLength; - matches[*nbMatches].off = candidateOffCode; - (*nbMatches)++; - } -} - -/* ZSTD_optLdm_processMatchCandidate(): - * Wrapper function to update ldm seq store and call ldm functions as necessary. - */ + DEBUGLOG(6, "ZSTD_optLdm_maybeAddMatch(): Adding ldm candidate match (offCode: %u matchLength %u) at block position=%u", + candidateOffCode, candidateMatchLength, currPosInBlock); + matches[*nbMatches].len = candidateMatchLength; + matches[*nbMatches].off = candidateOffCode; + (*nbMatches)++; + } +} + +/* ZSTD_optLdm_processMatchCandidate(): + * Wrapper function to update ldm seq store and call ldm functions as necessary. + */ static void ZSTD_optLdm_processMatchCandidate(ZSTD_optLdm_t* optLdm, ZSTD_match_t* matches, U32* nbMatches, U32 currPosInBlock, U32 remainingBytes) { - if (optLdm->seqStore.size == 0 || optLdm->seqStore.pos >= optLdm->seqStore.size) { - return; - } - - if (currPosInBlock >= optLdm->endPosInBlock) { - if (currPosInBlock > optLdm->endPosInBlock) { - /* The position at which ZSTD_optLdm_processMatchCandidate() is called is not necessarily - * at the end of a match from the ldm seq store, and will often be some bytes - * over beyond matchEndPosInBlock. As such, we need to correct for these "overshoots" - */ + if (optLdm->seqStore.size == 0 || optLdm->seqStore.pos >= optLdm->seqStore.size) { + return; + } + + if (currPosInBlock >= optLdm->endPosInBlock) { + if (currPosInBlock > optLdm->endPosInBlock) { + /* The position at which ZSTD_optLdm_processMatchCandidate() is called is not necessarily + * at the end of a match from the ldm seq store, and will often be some bytes + * over beyond matchEndPosInBlock. As such, we need to correct for these "overshoots" + */ U32 const posOvershoot = currPosInBlock - optLdm->endPosInBlock; - ZSTD_optLdm_skipRawSeqStoreBytes(&optLdm->seqStore, posOvershoot); + ZSTD_optLdm_skipRawSeqStoreBytes(&optLdm->seqStore, posOvershoot); } - ZSTD_opt_getNextMatchAndUpdateSeqStore(optLdm, currPosInBlock, remainingBytes); - } - ZSTD_optLdm_maybeAddMatch(matches, nbMatches, optLdm, currPosInBlock); -} - + ZSTD_opt_getNextMatchAndUpdateSeqStore(optLdm, currPosInBlock, remainingBytes); + } + ZSTD_optLdm_maybeAddMatch(matches, nbMatches, optLdm, currPosInBlock); +} + /*-******************************* * Optimal parser @@ -1060,12 +1060,12 @@ ZSTD_compressBlock_opt_generic(ZSTD_matchState_t* ms, ZSTD_optimal_t* const opt = optStatePtr->priceTable; ZSTD_match_t* const matches = optStatePtr->matchTable; ZSTD_optimal_t lastSequence; - ZSTD_optLdm_t optLdm; - - optLdm.seqStore = ms->ldmSeqStore ? *ms->ldmSeqStore : kNullRawSeqStore; - optLdm.endPosInBlock = optLdm.startPosInBlock = optLdm.offset = 0; - ZSTD_opt_getNextMatchAndUpdateSeqStore(&optLdm, (U32)(ip-istart), (U32)(iend-ip)); + ZSTD_optLdm_t optLdm; + optLdm.seqStore = ms->ldmSeqStore ? *ms->ldmSeqStore : kNullRawSeqStore; + optLdm.endPosInBlock = optLdm.startPosInBlock = optLdm.offset = 0; + ZSTD_opt_getNextMatchAndUpdateSeqStore(&optLdm, (U32)(ip-istart), (U32)(iend-ip)); + /* init */ DEBUGLOG(5, "ZSTD_compressBlock_opt_generic: current=%u, prefix=%u, nextToUpdate=%u", (U32)(ip - base), ms->window.dictLimit, ms->nextToUpdate); @@ -1081,8 +1081,8 @@ ZSTD_compressBlock_opt_generic(ZSTD_matchState_t* ms, { U32 const litlen = (U32)(ip - anchor); U32 const ll0 = !litlen; U32 nbMatches = getAllMatches(matches, ms, &nextToUpdate3, ip, iend, rep, ll0, minMatch); - ZSTD_optLdm_processMatchCandidate(&optLdm, matches, &nbMatches, - (U32)(ip-istart), (U32)(iend - ip)); + ZSTD_optLdm_processMatchCandidate(&optLdm, matches, &nbMatches, + (U32)(ip-istart), (U32)(iend - ip)); if (!nbMatches) { ip++; continue; } /* initialize opt[0] */ @@ -1176,9 +1176,9 @@ ZSTD_compressBlock_opt_generic(ZSTD_matchState_t* ms, if (opt[cur].mlen != 0) { U32 const prev = cur - opt[cur].mlen; repcodes_t const newReps = ZSTD_newRep(opt[prev].rep, opt[cur].off, opt[cur].litlen==0); - ZSTD_memcpy(opt[cur].rep, &newReps, sizeof(repcodes_t)); + ZSTD_memcpy(opt[cur].rep, &newReps, sizeof(repcodes_t)); } else { - ZSTD_memcpy(opt[cur].rep, opt[cur - 1].rep, sizeof(repcodes_t)); + ZSTD_memcpy(opt[cur].rep, opt[cur - 1].rep, sizeof(repcodes_t)); } /* last match must start at a minimum distance of 8 from oend */ @@ -1199,10 +1199,10 @@ ZSTD_compressBlock_opt_generic(ZSTD_matchState_t* ms, U32 const basePrice = previousPrice + ZSTD_litLengthPrice(0, optStatePtr, optLevel); U32 nbMatches = getAllMatches(matches, ms, &nextToUpdate3, inr, iend, opt[cur].rep, ll0, minMatch); U32 matchNb; - - ZSTD_optLdm_processMatchCandidate(&optLdm, matches, &nbMatches, - (U32)(inr-istart), (U32)(iend-inr)); - + + ZSTD_optLdm_processMatchCandidate(&optLdm, matches, &nbMatches, + (U32)(inr-istart), (U32)(iend-inr)); + if (!nbMatches) { DEBUGLOG(7, "rPos:%u : no match found", cur); continue; @@ -1266,9 +1266,9 @@ _shortestPath: /* cur, last_pos, best_mlen, best_off have to be set */ */ if (lastSequence.mlen != 0) { repcodes_t const reps = ZSTD_newRep(opt[cur].rep, lastSequence.off, lastSequence.litlen==0); - ZSTD_memcpy(rep, &reps, sizeof(reps)); + ZSTD_memcpy(rep, &reps, sizeof(reps)); } else { - ZSTD_memcpy(rep, opt[cur].rep, sizeof(repcodes_t)); + ZSTD_memcpy(rep, opt[cur].rep, sizeof(repcodes_t)); } { U32 const storeEnd = cur + 1; @@ -1358,7 +1358,7 @@ ZSTD_initStats_ultra(ZSTD_matchState_t* ms, const void* src, size_t srcSize) { U32 tmpRep[ZSTD_REP_NUM]; /* updated rep codes will sink here */ - ZSTD_memcpy(tmpRep, rep, sizeof(tmpRep)); + ZSTD_memcpy(tmpRep, rep, sizeof(tmpRep)); DEBUGLOG(4, "ZSTD_initStats_ultra (srcSize=%zu)", srcSize); assert(ms->opt.litLengthSum == 0); /* first block */ @@ -1389,7 +1389,7 @@ size_t ZSTD_compressBlock_btultra2( ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], const void* src, size_t srcSize) { - U32 const curr = (U32)((const BYTE*)src - ms->window.base); + U32 const curr = (U32)((const BYTE*)src - ms->window.base); DEBUGLOG(5, "ZSTD_compressBlock_btultra2 (srcSize=%zu)", srcSize); /* 2-pass strategy: @@ -1404,7 +1404,7 @@ size_t ZSTD_compressBlock_btultra2( if ( (ms->opt.litLengthSum==0) /* first block */ && (seqStore->sequences == seqStore->sequencesStart) /* no ldm */ && (ms->window.dictLimit == ms->window.lowLimit) /* no dictionary */ - && (curr == ms->window.dictLimit) /* start of frame, nothing already loaded nor skipped */ + && (curr == ms->window.dictLimit) /* start of frame, nothing already loaded nor skipped */ && (srcSize > ZSTD_PREDEF_THRESHOLD) ) { ZSTD_initStats_ultra(ms, seqStore, rep, src, srcSize); diff --git a/contrib/libs/zstd/lib/compress/zstd_opt.h b/contrib/libs/zstd/lib/compress/zstd_opt.h index 627255f53d..4dcb6bf79b 100644 --- a/contrib/libs/zstd/lib/compress/zstd_opt.h +++ b/contrib/libs/zstd/lib/compress/zstd_opt.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the diff --git a/contrib/libs/zstd/lib/compress/zstdmt_compress.c b/contrib/libs/zstd/lib/compress/zstdmt_compress.c index 6bc14b035e..869fbf37cb 100644 --- a/contrib/libs/zstd/lib/compress/zstdmt_compress.c +++ b/contrib/libs/zstd/lib/compress/zstdmt_compress.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -20,7 +20,7 @@ /* ====== Dependencies ====== */ -#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memset, INT_MAX, UINT_MAX */ +#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memset, INT_MAX, UINT_MAX */ #include "../common/mem.h" /* MEM_STATIC */ #include "../common/pool.h" /* threadpool */ #include "../common/threading.h" /* mutex */ @@ -104,11 +104,11 @@ typedef struct ZSTDMT_bufferPool_s { static ZSTDMT_bufferPool* ZSTDMT_createBufferPool(unsigned maxNbBuffers, ZSTD_customMem cMem) { - ZSTDMT_bufferPool* const bufPool = (ZSTDMT_bufferPool*)ZSTD_customCalloc( + ZSTDMT_bufferPool* const bufPool = (ZSTDMT_bufferPool*)ZSTD_customCalloc( sizeof(ZSTDMT_bufferPool) + (maxNbBuffers-1) * sizeof(buffer_t), cMem); if (bufPool==NULL) return NULL; if (ZSTD_pthread_mutex_init(&bufPool->poolMutex, NULL)) { - ZSTD_customFree(bufPool, cMem); + ZSTD_customFree(bufPool, cMem); return NULL; } bufPool->bufferSize = 64 KB; @@ -125,10 +125,10 @@ static void ZSTDMT_freeBufferPool(ZSTDMT_bufferPool* bufPool) if (!bufPool) return; /* compatibility with free on NULL */ for (u=0; u<bufPool->totalBuffers; u++) { DEBUGLOG(4, "free buffer %2u (address:%08X)", u, (U32)(size_t)bufPool->bTable[u].start); - ZSTD_customFree(bufPool->bTable[u].start, bufPool->cMem); + ZSTD_customFree(bufPool->bTable[u].start, bufPool->cMem); } ZSTD_pthread_mutex_destroy(&bufPool->poolMutex); - ZSTD_customFree(bufPool, bufPool->cMem); + ZSTD_customFree(bufPool, bufPool->cMem); } /* only works at initialization, not during compression */ @@ -198,13 +198,13 @@ static buffer_t ZSTDMT_getBuffer(ZSTDMT_bufferPool* bufPool) } /* size conditions not respected : scratch this buffer, create new one */ DEBUGLOG(5, "ZSTDMT_getBuffer: existing buffer does not meet size conditions => freeing"); - ZSTD_customFree(buf.start, bufPool->cMem); + ZSTD_customFree(buf.start, bufPool->cMem); } ZSTD_pthread_mutex_unlock(&bufPool->poolMutex); /* create new buffer */ DEBUGLOG(5, "ZSTDMT_getBuffer: create a new buffer"); { buffer_t buffer; - void* const start = ZSTD_customMalloc(bSize, bufPool->cMem); + void* const start = ZSTD_customMalloc(bSize, bufPool->cMem); buffer.start = start; /* note : start can be NULL if malloc fails ! */ buffer.capacity = (start==NULL) ? 0 : bSize; if (start==NULL) { @@ -226,13 +226,13 @@ static buffer_t ZSTDMT_resizeBuffer(ZSTDMT_bufferPool* bufPool, buffer_t buffer) { size_t const bSize = bufPool->bufferSize; if (buffer.capacity < bSize) { - void* const start = ZSTD_customMalloc(bSize, bufPool->cMem); + void* const start = ZSTD_customMalloc(bSize, bufPool->cMem); buffer_t newBuffer; newBuffer.start = start; newBuffer.capacity = start == NULL ? 0 : bSize; if (start != NULL) { assert(newBuffer.capacity >= buffer.capacity); - ZSTD_memcpy(newBuffer.start, buffer.start, buffer.capacity); + ZSTD_memcpy(newBuffer.start, buffer.start, buffer.capacity); DEBUGLOG(5, "ZSTDMT_resizeBuffer: created buffer of size %u", (U32)bSize); return newBuffer; } @@ -258,7 +258,7 @@ static void ZSTDMT_releaseBuffer(ZSTDMT_bufferPool* bufPool, buffer_t buf) ZSTD_pthread_mutex_unlock(&bufPool->poolMutex); /* Reached bufferPool capacity (should not happen) */ DEBUGLOG(5, "ZSTDMT_releaseBuffer: pool capacity reached => freeing "); - ZSTD_customFree(buf.start, bufPool->cMem); + ZSTD_customFree(buf.start, bufPool->cMem); } /* We need 2 output buffers per worker since each dstBuff must be flushed after it is released. @@ -283,7 +283,7 @@ static size_t ZSTDMT_sizeof_seqPool(ZSTDMT_seqPool* seqPool) static rawSeqStore_t bufferToSeq(buffer_t buffer) { - rawSeqStore_t seq = kNullRawSeqStore; + rawSeqStore_t seq = kNullRawSeqStore; seq.seq = (rawSeq*)buffer.start; seq.capacity = buffer.capacity / sizeof(rawSeq); return seq; @@ -359,7 +359,7 @@ static void ZSTDMT_freeCCtxPool(ZSTDMT_CCtxPool* pool) for (cid=0; cid<pool->totalCCtx; cid++) ZSTD_freeCCtx(pool->cctx[cid]); /* note : compatible with free on NULL */ ZSTD_pthread_mutex_destroy(&pool->poolMutex); - ZSTD_customFree(pool, pool->cMem); + ZSTD_customFree(pool, pool->cMem); } /* ZSTDMT_createCCtxPool() : @@ -367,12 +367,12 @@ static void ZSTDMT_freeCCtxPool(ZSTDMT_CCtxPool* pool) static ZSTDMT_CCtxPool* ZSTDMT_createCCtxPool(int nbWorkers, ZSTD_customMem cMem) { - ZSTDMT_CCtxPool* const cctxPool = (ZSTDMT_CCtxPool*) ZSTD_customCalloc( + ZSTDMT_CCtxPool* const cctxPool = (ZSTDMT_CCtxPool*) ZSTD_customCalloc( sizeof(ZSTDMT_CCtxPool) + (nbWorkers-1)*sizeof(ZSTD_CCtx*), cMem); assert(nbWorkers > 0); if (!cctxPool) return NULL; if (ZSTD_pthread_mutex_init(&cctxPool->poolMutex, NULL)) { - ZSTD_customFree(cctxPool, cMem); + ZSTD_customFree(cctxPool, cMem); return NULL; } cctxPool->cMem = cMem; @@ -481,7 +481,7 @@ ZSTDMT_serialState_reset(serialState_t* serialState, assert(params.ldmParams.hashLog >= params.ldmParams.bucketSizeLog); assert(params.ldmParams.hashRateLog < 32); } else { - ZSTD_memset(¶ms.ldmParams, 0, sizeof(params.ldmParams)); + ZSTD_memset(¶ms.ldmParams, 0, sizeof(params.ldmParams)); } serialState->nextJobID = 0; if (params.fParams.checksumFlag) @@ -495,32 +495,32 @@ ZSTDMT_serialState_reset(serialState_t* serialState, unsigned const prevBucketLog = serialState->params.ldmParams.hashLog - serialState->params.ldmParams.bucketSizeLog; - size_t const numBuckets = (size_t)1 << bucketLog; + size_t const numBuckets = (size_t)1 << bucketLog; /* Size the seq pool tables */ ZSTDMT_setNbSeq(seqPool, ZSTD_ldm_getMaxNbSeq(params.ldmParams, jobSize)); /* Reset the window */ ZSTD_window_init(&serialState->ldmState.window); /* Resize tables and output space if necessary. */ if (serialState->ldmState.hashTable == NULL || serialState->params.ldmParams.hashLog < hashLog) { - ZSTD_customFree(serialState->ldmState.hashTable, cMem); - serialState->ldmState.hashTable = (ldmEntry_t*)ZSTD_customMalloc(hashSize, cMem); + ZSTD_customFree(serialState->ldmState.hashTable, cMem); + serialState->ldmState.hashTable = (ldmEntry_t*)ZSTD_customMalloc(hashSize, cMem); } if (serialState->ldmState.bucketOffsets == NULL || prevBucketLog < bucketLog) { - ZSTD_customFree(serialState->ldmState.bucketOffsets, cMem); - serialState->ldmState.bucketOffsets = (BYTE*)ZSTD_customMalloc(numBuckets, cMem); + ZSTD_customFree(serialState->ldmState.bucketOffsets, cMem); + serialState->ldmState.bucketOffsets = (BYTE*)ZSTD_customMalloc(numBuckets, cMem); } if (!serialState->ldmState.hashTable || !serialState->ldmState.bucketOffsets) return 1; /* Zero the tables */ - ZSTD_memset(serialState->ldmState.hashTable, 0, hashSize); - ZSTD_memset(serialState->ldmState.bucketOffsets, 0, numBuckets); + ZSTD_memset(serialState->ldmState.hashTable, 0, hashSize); + ZSTD_memset(serialState->ldmState.bucketOffsets, 0, numBuckets); /* Update window state and fill hash table with dict */ serialState->ldmState.loadedDictEnd = 0; if (dictSize > 0) { if (dictContentType == ZSTD_dct_rawContent) { BYTE const* const dictEnd = (const BYTE*)dict + dictSize; - ZSTD_window_update(&serialState->ldmState.window, dict, dictSize, /* forceNonContiguous */ 0); + ZSTD_window_update(&serialState->ldmState.window, dict, dictSize, /* forceNonContiguous */ 0); ZSTD_ldm_fillHashTable(&serialState->ldmState, (const BYTE*)dict, dictEnd, ¶ms.ldmParams); serialState->ldmState.loadedDictEnd = params.forceWindow ? 0 : (U32)(dictEnd - serialState->ldmState.window.base); } else { @@ -540,7 +540,7 @@ ZSTDMT_serialState_reset(serialState_t* serialState, static int ZSTDMT_serialState_init(serialState_t* serialState) { int initError = 0; - ZSTD_memset(serialState, 0, sizeof(*serialState)); + ZSTD_memset(serialState, 0, sizeof(*serialState)); initError |= ZSTD_pthread_mutex_init(&serialState->mutex, NULL); initError |= ZSTD_pthread_cond_init(&serialState->cond, NULL); initError |= ZSTD_pthread_mutex_init(&serialState->ldmWindowMutex, NULL); @@ -555,8 +555,8 @@ static void ZSTDMT_serialState_free(serialState_t* serialState) ZSTD_pthread_cond_destroy(&serialState->cond); ZSTD_pthread_mutex_destroy(&serialState->ldmWindowMutex); ZSTD_pthread_cond_destroy(&serialState->ldmWindowCond); - ZSTD_customFree(serialState->ldmState.hashTable, cMem); - ZSTD_customFree(serialState->ldmState.bucketOffsets, cMem); + ZSTD_customFree(serialState->ldmState.hashTable, cMem); + ZSTD_customFree(serialState->ldmState.bucketOffsets, cMem); } static void ZSTDMT_serialState_update(serialState_t* serialState, @@ -577,7 +577,7 @@ static void ZSTDMT_serialState_update(serialState_t* serialState, assert(seqStore.seq != NULL && seqStore.pos == 0 && seqStore.size == 0 && seqStore.capacity > 0); assert(src.size <= serialState->params.jobSize); - ZSTD_window_update(&serialState->ldmState.window, src.start, src.size, /* forceNonContiguous */ 0); + ZSTD_window_update(&serialState->ldmState.window, src.start, src.size, /* forceNonContiguous */ 0); error = ZSTD_ldm_generateSequences( &serialState->ldmState, &seqStore, &serialState->params.ldmParams, src.start, src.size); @@ -689,8 +689,8 @@ static void ZSTDMT_compressionJob(void* jobDescription) if (job->jobID != 0) jobParams.fParams.checksumFlag = 0; /* Don't run LDM for the chunks, since we handle it externally */ jobParams.ldmParams.enableLdm = ZSTD_ps_disable; - /* Correct nbWorkers to 0. */ - jobParams.nbWorkers = 0; + /* Correct nbWorkers to 0. */ + jobParams.nbWorkers = 0; /* init */ @@ -703,10 +703,10 @@ static void ZSTDMT_compressionJob(void* jobDescription) { size_t const forceWindowError = ZSTD_CCtxParams_setParameter(&jobParams, ZSTD_c_forceMaxWindow, !job->firstJob); if (ZSTD_isError(forceWindowError)) JOB_ERROR(forceWindowError); } - if (!job->firstJob) { - size_t const err = ZSTD_CCtxParams_setParameter(&jobParams, ZSTD_c_deterministicRefPrefix, 0); - if (ZSTD_isError(err)) JOB_ERROR(err); - } + if (!job->firstJob) { + size_t const err = ZSTD_CCtxParams_setParameter(&jobParams, ZSTD_c_deterministicRefPrefix, 0); + if (ZSTD_isError(err)) JOB_ERROR(err); + } { size_t const initError = ZSTD_compressBegin_advanced_internal(cctx, job->prefix.start, job->prefix.size, ZSTD_dct_rawContent, /* load dictionary in "content-only" mode (no header analysis) */ ZSTD_dtlm_fast, @@ -762,13 +762,13 @@ static void ZSTDMT_compressionJob(void* jobDescription) if (ZSTD_isError(cSize)) JOB_ERROR(cSize); lastCBlockSize = cSize; } } - if (!job->firstJob) { - /* Double check that we don't have an ext-dict, because then our - * repcode invalidation doesn't work. - */ - assert(!ZSTD_window_hasExtDict(cctx->blockState.matchState.window)); - } - ZSTD_CCtx_trace(cctx, 0); + if (!job->firstJob) { + /* Double check that we don't have an ext-dict, because then our + * repcode invalidation doesn't work. + */ + assert(!ZSTD_window_hasExtDict(cctx->blockState.matchState.window)); + } + ZSTD_CCtx_trace(cctx, 0); _endJob: ZSTDMT_serialState_ensureFinished(job->serial, job->jobID, job->cSize); @@ -856,7 +856,7 @@ struct ZSTDMT_CCtx_s { ZSTD_customMem cMem; ZSTD_CDict* cdictLocal; const ZSTD_CDict* cdict; - unsigned providedFactory: 1; + unsigned providedFactory: 1; }; static void ZSTDMT_freeJobsTable(ZSTDMT_jobDescription* jobTable, U32 nbJobs, ZSTD_customMem cMem) @@ -867,7 +867,7 @@ static void ZSTDMT_freeJobsTable(ZSTDMT_jobDescription* jobTable, U32 nbJobs, ZS ZSTD_pthread_mutex_destroy(&jobTable[jobNb].job_mutex); ZSTD_pthread_cond_destroy(&jobTable[jobNb].job_cond); } - ZSTD_customFree(jobTable, cMem); + ZSTD_customFree(jobTable, cMem); } /* ZSTDMT_allocJobsTable() @@ -879,7 +879,7 @@ static ZSTDMT_jobDescription* ZSTDMT_createJobsTable(U32* nbJobsPtr, ZSTD_custom U32 const nbJobs = 1 << nbJobsLog2; U32 jobNb; ZSTDMT_jobDescription* const jobTable = (ZSTDMT_jobDescription*) - ZSTD_customCalloc(nbJobs * sizeof(ZSTDMT_jobDescription), cMem); + ZSTD_customCalloc(nbJobs * sizeof(ZSTDMT_jobDescription), cMem); int initError = 0; if (jobTable==NULL) return NULL; *nbJobsPtr = nbJobs; @@ -910,12 +910,12 @@ static size_t ZSTDMT_expandJobsTable (ZSTDMT_CCtx* mtctx, U32 nbWorkers) { /* ZSTDMT_CCtxParam_setNbWorkers(): * Internal use only */ -static size_t ZSTDMT_CCtxParam_setNbWorkers(ZSTD_CCtx_params* params, unsigned nbWorkers) +static size_t ZSTDMT_CCtxParam_setNbWorkers(ZSTD_CCtx_params* params, unsigned nbWorkers) { return ZSTD_CCtxParams_setParameter(params, ZSTD_c_nbWorkers, (int)nbWorkers); } -MEM_STATIC ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced_internal(unsigned nbWorkers, ZSTD_customMem cMem, ZSTD_threadPool* pool) +MEM_STATIC ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced_internal(unsigned nbWorkers, ZSTD_customMem cMem, ZSTD_threadPool* pool) { ZSTDMT_CCtx* mtctx; U32 nbJobs = nbWorkers + 2; @@ -928,19 +928,19 @@ MEM_STATIC ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced_internal(unsigned nbWorkers, /* invalid custom allocator */ return NULL; - mtctx = (ZSTDMT_CCtx*) ZSTD_customCalloc(sizeof(ZSTDMT_CCtx), cMem); + mtctx = (ZSTDMT_CCtx*) ZSTD_customCalloc(sizeof(ZSTDMT_CCtx), cMem); if (!mtctx) return NULL; ZSTDMT_CCtxParam_setNbWorkers(&mtctx->params, nbWorkers); mtctx->cMem = cMem; mtctx->allJobsCompleted = 1; - if (pool != NULL) { - mtctx->factory = pool; - mtctx->providedFactory = 1; - } - else { - mtctx->factory = POOL_create_advanced(nbWorkers, 0, cMem); - mtctx->providedFactory = 0; - } + if (pool != NULL) { + mtctx->factory = pool; + mtctx->providedFactory = 1; + } + else { + mtctx->factory = POOL_create_advanced(nbWorkers, 0, cMem); + mtctx->providedFactory = 0; + } mtctx->jobs = ZSTDMT_createJobsTable(&nbJobs, cMem); assert(nbJobs > 0); assert((nbJobs & (nbJobs - 1)) == 0); /* ensure nbJobs is a power of 2 */ mtctx->jobIDMask = nbJobs - 1; @@ -957,14 +957,14 @@ MEM_STATIC ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced_internal(unsigned nbWorkers, return mtctx; } -ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced(unsigned nbWorkers, ZSTD_customMem cMem, ZSTD_threadPool* pool) +ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced(unsigned nbWorkers, ZSTD_customMem cMem, ZSTD_threadPool* pool) { #ifdef ZSTD_MULTITHREAD - return ZSTDMT_createCCtx_advanced_internal(nbWorkers, cMem, pool); + return ZSTDMT_createCCtx_advanced_internal(nbWorkers, cMem, pool); #else (void)nbWorkers; (void)cMem; - (void)pool; + (void)pool; return NULL; #endif } @@ -985,7 +985,7 @@ static void ZSTDMT_releaseAllJobResources(ZSTDMT_CCtx* mtctx) ZSTDMT_releaseBuffer(mtctx->bufPool, mtctx->jobs[jobID].dstBuff); /* Clear the job description, but keep the mutex/cond */ - ZSTD_memset(&mtctx->jobs[jobID], 0, sizeof(mtctx->jobs[jobID])); + ZSTD_memset(&mtctx->jobs[jobID], 0, sizeof(mtctx->jobs[jobID])); mtctx->jobs[jobID].job_mutex = mutex; mtctx->jobs[jobID].job_cond = cond; } @@ -1012,8 +1012,8 @@ static void ZSTDMT_waitForAllJobsCompleted(ZSTDMT_CCtx* mtctx) size_t ZSTDMT_freeCCtx(ZSTDMT_CCtx* mtctx) { if (mtctx==NULL) return 0; /* compatible with free on NULL */ - if (!mtctx->providedFactory) - POOL_free(mtctx->factory); /* stop and free worker threads */ + if (!mtctx->providedFactory) + POOL_free(mtctx->factory); /* stop and free worker threads */ ZSTDMT_releaseAllJobResources(mtctx); /* release job resources into pools first */ ZSTDMT_freeJobsTable(mtctx->jobs, mtctx->jobIDMask+1, mtctx->cMem); ZSTDMT_freeBufferPool(mtctx->bufPool); @@ -1022,8 +1022,8 @@ size_t ZSTDMT_freeCCtx(ZSTDMT_CCtx* mtctx) ZSTDMT_serialState_free(&mtctx->serial); ZSTD_freeCDict(mtctx->cdictLocal); if (mtctx->roundBuff.buffer) - ZSTD_customFree(mtctx->roundBuff.buffer, mtctx->cMem); - ZSTD_customFree(mtctx, mtctx->cMem); + ZSTD_customFree(mtctx->roundBuff.buffer, mtctx->cMem); + ZSTD_customFree(mtctx, mtctx->cMem); return 0; } @@ -1068,7 +1068,7 @@ void ZSTDMT_updateCParams_whileCompressing(ZSTDMT_CCtx* mtctx, const ZSTD_CCtx_p DEBUGLOG(5, "ZSTDMT_updateCParams_whileCompressing (level:%i)", compressionLevel); mtctx->params.compressionLevel = compressionLevel; - { ZSTD_compressionParameters cParams = ZSTD_getCParamsFromCCtxParams(cctxParams, ZSTD_CONTENTSIZE_UNKNOWN, 0, ZSTD_cpm_noAttachDict); + { ZSTD_compressionParameters cParams = ZSTD_getCParamsFromCCtxParams(cctxParams, ZSTD_CONTENTSIZE_UNKNOWN, 0, ZSTD_cpm_noAttachDict); cParams.windowLog = saved_wlog; mtctx->params.cParams = cParams; } @@ -1155,8 +1155,8 @@ static unsigned ZSTDMT_computeTargetJobLog(const ZSTD_CCtx_params* params) if (params->ldmParams.enableLdm == ZSTD_ps_enable) { /* In Long Range Mode, the windowLog is typically oversized. * In which case, it's preferable to determine the jobSize - * based on cycleLog instead. */ - jobLog = MAX(21, ZSTD_cycleLog(params->cParams.chainLog, params->cParams.strategy) + 3); + * based on cycleLog instead. */ + jobLog = MAX(21, ZSTD_cycleLog(params->cParams.chainLog, params->cParams.strategy) + 3); } else { jobLog = MAX(20, params->cParams.windowLog + 2); } @@ -1267,8 +1267,8 @@ size_t ZSTDMT_initCStream_internal( if (params.rsyncable) { /* Aim for the targetsectionSize as the average job size. */ - U32 const jobSizeKB = (U32)(mtctx->targetSectionSize >> 10); - U32 const rsyncBits = (assert(jobSizeKB >= 1), ZSTD_highbit32(jobSizeKB) + 10); + U32 const jobSizeKB = (U32)(mtctx->targetSectionSize >> 10); + U32 const rsyncBits = (assert(jobSizeKB >= 1), ZSTD_highbit32(jobSizeKB) + 10); /* We refuse to create jobs < RSYNC_MIN_BLOCK_SIZE bytes, so make sure our * expected job size is at least 4x larger. */ assert(rsyncBits >= RSYNC_MIN_BLOCK_LOG + 2); @@ -1298,8 +1298,8 @@ size_t ZSTDMT_initCStream_internal( size_t const capacity = MAX(windowSize, sectionsSize) + slackSize; if (mtctx->roundBuff.capacity < capacity) { if (mtctx->roundBuff.buffer) - ZSTD_customFree(mtctx->roundBuff.buffer, mtctx->cMem); - mtctx->roundBuff.buffer = (BYTE*)ZSTD_customMalloc(capacity, mtctx->cMem); + ZSTD_customFree(mtctx->roundBuff.buffer, mtctx->cMem); + mtctx->roundBuff.buffer = (BYTE*)ZSTD_customMalloc(capacity, mtctx->cMem); if (mtctx->roundBuff.buffer == NULL) { mtctx->roundBuff.capacity = 0; return ERROR(memory_allocation); @@ -1487,7 +1487,7 @@ static size_t ZSTDMT_flushProduced(ZSTDMT_CCtx* mtctx, ZSTD_outBuffer* output, u assert(cSize >= mtctx->jobs[wJobID].dstFlushed); assert(mtctx->jobs[wJobID].dstBuff.start != NULL); if (toFlush > 0) { - ZSTD_memcpy((char*)output->dst + output->pos, + ZSTD_memcpy((char*)output->dst + output->pos, (const char*)mtctx->jobs[wJobID].dstBuff.start + mtctx->jobs[wJobID].dstFlushed, toFlush); } @@ -1645,7 +1645,7 @@ static int ZSTDMT_tryGetInputRange(ZSTDMT_CCtx* mtctx) return 0; } ZSTDMT_waitForLdmComplete(mtctx, buffer); - ZSTD_memmove(start, mtctx->inBuff.prefix.start, prefixSize); + ZSTD_memmove(start, mtctx->inBuff.prefix.start, prefixSize); mtctx->inBuff.prefix.start = start; mtctx->roundBuff.pos = prefixSize; } @@ -1742,16 +1742,16 @@ findSynchronizationPoint(ZSTDMT_CCtx const* mtctx, ZSTD_inBuffer const input) pos = 0; prev = (BYTE const*)mtctx->inBuff.buffer.start + mtctx->inBuff.filled - RSYNC_LENGTH; hash = ZSTD_rollingHash_compute(prev, RSYNC_LENGTH); - if ((hash & hitMask) == hitMask) { - /* We're already at a sync point so don't load any more until - * we're able to flush this sync point. - * This likely happened because the job table was full so we - * couldn't add our job. - */ - syncPoint.toLoad = 0; - syncPoint.flush = 1; - return syncPoint; - } + if ((hash & hitMask) == hitMask) { + /* We're already at a sync point so don't load any more until + * we're able to flush this sync point. + * This likely happened because the job table was full so we + * couldn't add our job. + */ + syncPoint.toLoad = 0; + syncPoint.flush = 1; + return syncPoint; + } } /* Starting with the hash of the previous RSYNC_LENGTH bytes, roll * through the input. If we hit a synchronization point, then cut the @@ -1824,22 +1824,22 @@ size_t ZSTDMT_compressStream_generic(ZSTDMT_CCtx* mtctx, assert(mtctx->inBuff.buffer.capacity >= mtctx->targetSectionSize); DEBUGLOG(5, "ZSTDMT_compressStream_generic: adding %u bytes on top of %u to buffer of size %u", (U32)syncPoint.toLoad, (U32)mtctx->inBuff.filled, (U32)mtctx->targetSectionSize); - ZSTD_memcpy((char*)mtctx->inBuff.buffer.start + mtctx->inBuff.filled, (const char*)input->src + input->pos, syncPoint.toLoad); + ZSTD_memcpy((char*)mtctx->inBuff.buffer.start + mtctx->inBuff.filled, (const char*)input->src + input->pos, syncPoint.toLoad); input->pos += syncPoint.toLoad; mtctx->inBuff.filled += syncPoint.toLoad; forwardInputProgress = syncPoint.toLoad>0; } } - if ((input->pos < input->size) && (endOp == ZSTD_e_end)) { - /* Can't end yet because the input is not fully consumed. - * We are in one of these cases: - * - mtctx->inBuff is NULL & empty: we couldn't get an input buffer so don't create a new job. - * - We filled the input buffer: flush this job but don't end the frame. - * - We hit a synchronization point: flush this job but don't end the frame. - */ - assert(mtctx->inBuff.filled == 0 || mtctx->inBuff.filled == mtctx->targetSectionSize || mtctx->params.rsyncable); - endOp = ZSTD_e_flush; - } + if ((input->pos < input->size) && (endOp == ZSTD_e_end)) { + /* Can't end yet because the input is not fully consumed. + * We are in one of these cases: + * - mtctx->inBuff is NULL & empty: we couldn't get an input buffer so don't create a new job. + * - We filled the input buffer: flush this job but don't end the frame. + * - We hit a synchronization point: flush this job but don't end the frame. + */ + assert(mtctx->inBuff.filled == 0 || mtctx->inBuff.filled == mtctx->targetSectionSize || mtctx->params.rsyncable); + endOp = ZSTD_e_flush; + } if ( (mtctx->jobReady) || (mtctx->inBuff.filled >= mtctx->targetSectionSize) /* filled enough : let's compress */ diff --git a/contrib/libs/zstd/lib/compress/zstdmt_compress.h b/contrib/libs/zstd/lib/compress/zstdmt_compress.h index 271eb1ac71..808a04613e 100644 --- a/contrib/libs/zstd/lib/compress/zstdmt_compress.h +++ b/contrib/libs/zstd/lib/compress/zstdmt_compress.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -19,60 +19,60 @@ /* Note : This is an internal API. * These APIs used to be exposed with ZSTDLIB_API, * because it used to be the only way to invoke MT compression. - * Now, you must use ZSTD_compress2 and ZSTD_compressStream2() instead. + * Now, you must use ZSTD_compress2 and ZSTD_compressStream2() instead. * * This API requires ZSTD_MULTITHREAD to be defined during compilation, * otherwise ZSTDMT_createCCtx*() will fail. */ /* === Dependencies === */ -#include "../common/zstd_deps.h" /* size_t */ +#include "../common/zstd_deps.h" /* size_t */ #define ZSTD_STATIC_LINKING_ONLY /* ZSTD_parameters */ #include "../zstd.h" /* ZSTD_inBuffer, ZSTD_outBuffer, ZSTDLIB_API */ /* === Constants === */ -#ifndef ZSTDMT_NBWORKERS_MAX /* a different value can be selected at compile time */ -# define ZSTDMT_NBWORKERS_MAX ((sizeof(void*)==4) /*32-bit*/ ? 64 : 256) +#ifndef ZSTDMT_NBWORKERS_MAX /* a different value can be selected at compile time */ +# define ZSTDMT_NBWORKERS_MAX ((sizeof(void*)==4) /*32-bit*/ ? 64 : 256) #endif -#ifndef ZSTDMT_JOBSIZE_MIN /* a different value can be selected at compile time */ -# define ZSTDMT_JOBSIZE_MIN (512 KB) +#ifndef ZSTDMT_JOBSIZE_MIN /* a different value can be selected at compile time */ +# define ZSTDMT_JOBSIZE_MIN (512 KB) #endif #define ZSTDMT_JOBLOG_MAX (MEM_32bits() ? 29 : 30) #define ZSTDMT_JOBSIZE_MAX (MEM_32bits() ? (512 MB) : (1024 MB)) -/* ======================================================== - * === Private interface, for use by ZSTD_compress.c === - * === Not exposed in libzstd. Never invoke directly === - * ======================================================== */ - +/* ======================================================== + * === Private interface, for use by ZSTD_compress.c === + * === Not exposed in libzstd. Never invoke directly === + * ======================================================== */ + /* === Memory management === */ typedef struct ZSTDMT_CCtx_s ZSTDMT_CCtx; /* Requires ZSTD_MULTITHREAD to be defined during compilation, otherwise it will return NULL. */ -ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced(unsigned nbWorkers, - ZSTD_customMem cMem, - ZSTD_threadPool *pool); -size_t ZSTDMT_freeCCtx(ZSTDMT_CCtx* mtctx); +ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced(unsigned nbWorkers, + ZSTD_customMem cMem, + ZSTD_threadPool *pool); +size_t ZSTDMT_freeCCtx(ZSTDMT_CCtx* mtctx); -size_t ZSTDMT_sizeof_CCtx(ZSTDMT_CCtx* mtctx); +size_t ZSTDMT_sizeof_CCtx(ZSTDMT_CCtx* mtctx); /* === Streaming functions === */ -size_t ZSTDMT_nextInputSizeHint(const ZSTDMT_CCtx* mtctx); +size_t ZSTDMT_nextInputSizeHint(const ZSTDMT_CCtx* mtctx); -/*! ZSTDMT_initCStream_internal() : - * Private use only. Init streaming operation. - * expects params to be valid. - * must receive dict, or cdict, or none, but not both. +/*! ZSTDMT_initCStream_internal() : + * Private use only. Init streaming operation. + * expects params to be valid. + * must receive dict, or cdict, or none, but not both. * mtctx can be freshly constructed or reused from a prior compression. * If mtctx is reused, memory allocations from the prior compression may not be freed, * even if they are not needed for the current compression. - * @return : 0, or an error code */ + * @return : 0, or an error code */ size_t ZSTDMT_initCStream_internal(ZSTDMT_CCtx* mtctx, - const void* dict, size_t dictSize, ZSTD_dictContentType_e dictContentType, - const ZSTD_CDict* cdict, - ZSTD_CCtx_params params, unsigned long long pledgedSrcSize); + const void* dict, size_t dictSize, ZSTD_dictContentType_e dictContentType, + const ZSTD_CDict* cdict, + ZSTD_CCtx_params params, unsigned long long pledgedSrcSize); /*! ZSTDMT_compressStream_generic() : * Combines ZSTDMT_compressStream() with optional ZSTDMT_flushStream() or ZSTDMT_endStream() @@ -81,10 +81,10 @@ size_t ZSTDMT_initCStream_internal(ZSTDMT_CCtx* mtctx, * 0 if fully flushed * or an error code * note : needs to be init using any ZSTD_initCStream*() variant */ -size_t ZSTDMT_compressStream_generic(ZSTDMT_CCtx* mtctx, - ZSTD_outBuffer* output, - ZSTD_inBuffer* input, - ZSTD_EndDirective endOp); +size_t ZSTDMT_compressStream_generic(ZSTDMT_CCtx* mtctx, + ZSTD_outBuffer* output, + ZSTD_inBuffer* input, + ZSTD_EndDirective endOp); /*! ZSTDMT_toFlushNow() * Tell how many bytes are ready to be flushed immediately. diff --git a/contrib/libs/zstd/lib/decompress/huf_decompress.c b/contrib/libs/zstd/lib/decompress/huf_decompress.c index 2027188255..5d96fb1fa5 100644 --- a/contrib/libs/zstd/lib/decompress/huf_decompress.c +++ b/contrib/libs/zstd/lib/decompress/huf_decompress.c @@ -1,7 +1,7 @@ /* ****************************************************************** * huff0 huffman decoder, * part of Finite State Entropy library - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * * You can contact the author at : * - FSE+HUF source repository : https://github.com/Cyan4973/FiniteStateEntropy @@ -15,7 +15,7 @@ /* ************************************************************** * Dependencies ****************************************************************/ -#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memset */ +#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memset */ #include "../common/compiler.h" #include "../common/bitstream.h" /* BIT_* */ #include "../common/fse.h" /* to compress headers */ @@ -134,7 +134,7 @@ typedef struct { BYTE maxTableLog; BYTE tableType; BYTE tableLog; BYTE reserved; static DTableDesc HUF_getDTableDesc(const HUF_DTable* table) { DTableDesc dtd; - ZSTD_memcpy(&dtd, table, sizeof(dtd)); + ZSTD_memcpy(&dtd, table, sizeof(dtd)); return dtd; } @@ -280,21 +280,21 @@ static size_t HUF_initRemainingDStream(BIT_DStream_t* bit, HUF_DecompressAsmArgs /*-***************************/ typedef struct { BYTE nbBits; BYTE byte; } HUF_DEltX1; /* single-symbol decoding */ -/** - * Packs 4 HUF_DEltX1 structs into a U64. This is used to lay down 4 entries at - * a time. - */ -static U64 HUF_DEltX1_set4(BYTE symbol, BYTE nbBits) { - U64 D4; - if (MEM_isLittleEndian()) { +/** + * Packs 4 HUF_DEltX1 structs into a U64. This is used to lay down 4 entries at + * a time. + */ +static U64 HUF_DEltX1_set4(BYTE symbol, BYTE nbBits) { + U64 D4; + if (MEM_isLittleEndian()) { D4 = (symbol << 8) + nbBits; } else { - D4 = symbol + (nbBits << 8); - } - D4 *= 0x0001000100010001ULL; - return D4; -} - + D4 = symbol + (nbBits << 8); + } + D4 *= 0x0001000100010001ULL; + return D4; +} + /** * Increase the tableLog to targetTableLog and rescales the stats. * If tableLog > targetTableLog this is a no-op. @@ -325,36 +325,36 @@ static U32 HUF_rescaleStats(BYTE* huffWeight, U32* rankVal, U32 nbSymbols, U32 t return targetTableLog; } -typedef struct { - U32 rankVal[HUF_TABLELOG_ABSOLUTEMAX + 1]; - U32 rankStart[HUF_TABLELOG_ABSOLUTEMAX + 1]; - U32 statsWksp[HUF_READ_STATS_WORKSPACE_SIZE_U32]; - BYTE symbols[HUF_SYMBOLVALUE_MAX + 1]; - BYTE huffWeight[HUF_SYMBOLVALUE_MAX + 1]; -} HUF_ReadDTableX1_Workspace; - - +typedef struct { + U32 rankVal[HUF_TABLELOG_ABSOLUTEMAX + 1]; + U32 rankStart[HUF_TABLELOG_ABSOLUTEMAX + 1]; + U32 statsWksp[HUF_READ_STATS_WORKSPACE_SIZE_U32]; + BYTE symbols[HUF_SYMBOLVALUE_MAX + 1]; + BYTE huffWeight[HUF_SYMBOLVALUE_MAX + 1]; +} HUF_ReadDTableX1_Workspace; + + size_t HUF_readDTableX1_wksp(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize) { - return HUF_readDTableX1_wksp_bmi2(DTable, src, srcSize, workSpace, wkspSize, /* bmi2 */ 0); -} - -size_t HUF_readDTableX1_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2) -{ + return HUF_readDTableX1_wksp_bmi2(DTable, src, srcSize, workSpace, wkspSize, /* bmi2 */ 0); +} + +size_t HUF_readDTableX1_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2) +{ U32 tableLog = 0; U32 nbSymbols = 0; size_t iSize; void* const dtPtr = DTable + 1; HUF_DEltX1* const dt = (HUF_DEltX1*)dtPtr; - HUF_ReadDTableX1_Workspace* wksp = (HUF_ReadDTableX1_Workspace*)workSpace; + HUF_ReadDTableX1_Workspace* wksp = (HUF_ReadDTableX1_Workspace*)workSpace; - DEBUG_STATIC_ASSERT(HUF_DECOMPRESS_WORKSPACE_SIZE >= sizeof(*wksp)); - if (sizeof(*wksp) > wkspSize) return ERROR(tableLog_tooLarge); + DEBUG_STATIC_ASSERT(HUF_DECOMPRESS_WORKSPACE_SIZE >= sizeof(*wksp)); + if (sizeof(*wksp) > wkspSize) return ERROR(tableLog_tooLarge); DEBUG_STATIC_ASSERT(sizeof(DTableDesc) == sizeof(HUF_DTable)); - /* ZSTD_memset(huffWeight, 0, sizeof(huffWeight)); */ /* is not necessary, even though some analyzer complain ... */ + /* ZSTD_memset(huffWeight, 0, sizeof(huffWeight)); */ /* is not necessary, even though some analyzer complain ... */ - iSize = HUF_readStats_wksp(wksp->huffWeight, HUF_SYMBOLVALUE_MAX + 1, wksp->rankVal, &nbSymbols, &tableLog, src, srcSize, wksp->statsWksp, sizeof(wksp->statsWksp), bmi2); + iSize = HUF_readStats_wksp(wksp->huffWeight, HUF_SYMBOLVALUE_MAX + 1, wksp->rankVal, &nbSymbols, &tableLog, src, srcSize, wksp->statsWksp, sizeof(wksp->statsWksp), bmi2); if (HUF_isError(iSize)) return iSize; @@ -366,114 +366,114 @@ size_t HUF_readDTableX1_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t sr if (tableLog > (U32)(dtd.maxTableLog+1)) return ERROR(tableLog_tooLarge); /* DTable too small, Huffman tree cannot fit in */ dtd.tableType = 0; dtd.tableLog = (BYTE)tableLog; - ZSTD_memcpy(DTable, &dtd, sizeof(dtd)); + ZSTD_memcpy(DTable, &dtd, sizeof(dtd)); } - /* Compute symbols and rankStart given rankVal: - * - * rankVal already contains the number of values of each weight. - * - * symbols contains the symbols ordered by weight. First are the rankVal[0] - * weight 0 symbols, followed by the rankVal[1] weight 1 symbols, and so on. - * symbols[0] is filled (but unused) to avoid a branch. - * - * rankStart contains the offset where each rank belongs in the DTable. - * rankStart[0] is not filled because there are no entries in the table for - * weight 0. - */ - { - int n; - int nextRankStart = 0; - int const unroll = 4; - int const nLimit = (int)nbSymbols - unroll + 1; - for (n=0; n<(int)tableLog+1; n++) { - U32 const curr = nextRankStart; - nextRankStart += wksp->rankVal[n]; - wksp->rankStart[n] = curr; - } - for (n=0; n < nLimit; n += unroll) { - int u; - for (u=0; u < unroll; ++u) { - size_t const w = wksp->huffWeight[n+u]; - wksp->symbols[wksp->rankStart[w]++] = (BYTE)(n+u); - } - } - for (; n < (int)nbSymbols; ++n) { - size_t const w = wksp->huffWeight[n]; - wksp->symbols[wksp->rankStart[w]++] = (BYTE)n; - } - } - - /* fill DTable - * We fill all entries of each weight in order. + /* Compute symbols and rankStart given rankVal: + * + * rankVal already contains the number of values of each weight. + * + * symbols contains the symbols ordered by weight. First are the rankVal[0] + * weight 0 symbols, followed by the rankVal[1] weight 1 symbols, and so on. + * symbols[0] is filled (but unused) to avoid a branch. + * + * rankStart contains the offset where each rank belongs in the DTable. + * rankStart[0] is not filled because there are no entries in the table for + * weight 0. + */ + { + int n; + int nextRankStart = 0; + int const unroll = 4; + int const nLimit = (int)nbSymbols - unroll + 1; + for (n=0; n<(int)tableLog+1; n++) { + U32 const curr = nextRankStart; + nextRankStart += wksp->rankVal[n]; + wksp->rankStart[n] = curr; + } + for (n=0; n < nLimit; n += unroll) { + int u; + for (u=0; u < unroll; ++u) { + size_t const w = wksp->huffWeight[n+u]; + wksp->symbols[wksp->rankStart[w]++] = (BYTE)(n+u); + } + } + for (; n < (int)nbSymbols; ++n) { + size_t const w = wksp->huffWeight[n]; + wksp->symbols[wksp->rankStart[w]++] = (BYTE)n; + } + } + + /* fill DTable + * We fill all entries of each weight in order. * That way length is a constant for each iteration of the outer loop. - * We can switch based on the length to a different inner loop which is - * optimized for that particular case. - */ - { - U32 w; - int symbol=wksp->rankVal[0]; - int rankStart=0; - for (w=1; w<tableLog+1; ++w) { - int const symbolCount = wksp->rankVal[w]; - int const length = (1 << w) >> 1; - int uStart = rankStart; - BYTE const nbBits = (BYTE)(tableLog + 1 - w); - int s; - int u; - switch (length) { - case 1: - for (s=0; s<symbolCount; ++s) { - HUF_DEltX1 D; - D.byte = wksp->symbols[symbol + s]; - D.nbBits = nbBits; - dt[uStart] = D; - uStart += 1; - } - break; - case 2: - for (s=0; s<symbolCount; ++s) { - HUF_DEltX1 D; - D.byte = wksp->symbols[symbol + s]; - D.nbBits = nbBits; - dt[uStart+0] = D; - dt[uStart+1] = D; - uStart += 2; - } - break; - case 4: - for (s=0; s<symbolCount; ++s) { - U64 const D4 = HUF_DEltX1_set4(wksp->symbols[symbol + s], nbBits); - MEM_write64(dt + uStart, D4); - uStart += 4; - } - break; - case 8: - for (s=0; s<symbolCount; ++s) { - U64 const D4 = HUF_DEltX1_set4(wksp->symbols[symbol + s], nbBits); - MEM_write64(dt + uStart, D4); - MEM_write64(dt + uStart + 4, D4); - uStart += 8; - } - break; - default: - for (s=0; s<symbolCount; ++s) { - U64 const D4 = HUF_DEltX1_set4(wksp->symbols[symbol + s], nbBits); - for (u=0; u < length; u += 16) { - MEM_write64(dt + uStart + u + 0, D4); - MEM_write64(dt + uStart + u + 4, D4); - MEM_write64(dt + uStart + u + 8, D4); - MEM_write64(dt + uStart + u + 12, D4); - } - assert(u == length); - uStart += length; - } - break; - } - symbol += symbolCount; - rankStart += symbolCount * length; - } - } + * We can switch based on the length to a different inner loop which is + * optimized for that particular case. + */ + { + U32 w; + int symbol=wksp->rankVal[0]; + int rankStart=0; + for (w=1; w<tableLog+1; ++w) { + int const symbolCount = wksp->rankVal[w]; + int const length = (1 << w) >> 1; + int uStart = rankStart; + BYTE const nbBits = (BYTE)(tableLog + 1 - w); + int s; + int u; + switch (length) { + case 1: + for (s=0; s<symbolCount; ++s) { + HUF_DEltX1 D; + D.byte = wksp->symbols[symbol + s]; + D.nbBits = nbBits; + dt[uStart] = D; + uStart += 1; + } + break; + case 2: + for (s=0; s<symbolCount; ++s) { + HUF_DEltX1 D; + D.byte = wksp->symbols[symbol + s]; + D.nbBits = nbBits; + dt[uStart+0] = D; + dt[uStart+1] = D; + uStart += 2; + } + break; + case 4: + for (s=0; s<symbolCount; ++s) { + U64 const D4 = HUF_DEltX1_set4(wksp->symbols[symbol + s], nbBits); + MEM_write64(dt + uStart, D4); + uStart += 4; + } + break; + case 8: + for (s=0; s<symbolCount; ++s) { + U64 const D4 = HUF_DEltX1_set4(wksp->symbols[symbol + s], nbBits); + MEM_write64(dt + uStart, D4); + MEM_write64(dt + uStart + 4, D4); + uStart += 8; + } + break; + default: + for (s=0; s<symbolCount; ++s) { + U64 const D4 = HUF_DEltX1_set4(wksp->symbols[symbol + s], nbBits); + for (u=0; u < length; u += 16) { + MEM_write64(dt + uStart + u + 0, D4); + MEM_write64(dt + uStart + u + 4, D4); + MEM_write64(dt + uStart + u + 8, D4); + MEM_write64(dt + uStart + u + 12, D4); + } + assert(u == length); + uStart += length; + } + break; + } + symbol += symbolCount; + rankStart += symbolCount * length; + } + } return iSize; } @@ -791,7 +791,7 @@ static size_t HUF_decompress4X1_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size { const BYTE* ip = (const BYTE*) cSrc; - size_t const hSize = HUF_readDTableX1_wksp_bmi2(dctx, cSrc, cSrcSize, workSpace, wkspSize, bmi2); + size_t const hSize = HUF_readDTableX1_wksp_bmi2(dctx, cSrc, cSrcSize, workSpace, wkspSize, bmi2); if (HUF_isError(hSize)) return hSize; if (hSize >= cSrcSize) return ERROR(srcSize_wrong); ip += hSize; cSrcSize -= hSize; @@ -1033,15 +1033,15 @@ static void HUF_fillDTableX2(HUF_DEltX2* DTable, const U32 targetLog, } } -typedef struct { - rankValCol_t rankVal[HUF_TABLELOG_MAX]; - U32 rankStats[HUF_TABLELOG_MAX + 1]; +typedef struct { + rankValCol_t rankVal[HUF_TABLELOG_MAX]; + U32 rankStats[HUF_TABLELOG_MAX + 1]; U32 rankStart0[HUF_TABLELOG_MAX + 3]; - sortedSymbol_t sortedSymbol[HUF_SYMBOLVALUE_MAX + 1]; - BYTE weightList[HUF_SYMBOLVALUE_MAX + 1]; - U32 calleeWksp[HUF_READ_STATS_WORKSPACE_SIZE_U32]; -} HUF_ReadDTableX2_Workspace; - + sortedSymbol_t sortedSymbol[HUF_SYMBOLVALUE_MAX + 1]; + BYTE weightList[HUF_SYMBOLVALUE_MAX + 1]; + U32 calleeWksp[HUF_READ_STATS_WORKSPACE_SIZE_U32]; +} HUF_ReadDTableX2_Workspace; + size_t HUF_readDTableX2_wksp(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize) @@ -1061,17 +1061,17 @@ size_t HUF_readDTableX2_wksp_bmi2(HUF_DTable* DTable, HUF_DEltX2* const dt = (HUF_DEltX2*)dtPtr; U32 *rankStart; - HUF_ReadDTableX2_Workspace* const wksp = (HUF_ReadDTableX2_Workspace*)workSpace; + HUF_ReadDTableX2_Workspace* const wksp = (HUF_ReadDTableX2_Workspace*)workSpace; - if (sizeof(*wksp) > wkspSize) return ERROR(GENERIC); + if (sizeof(*wksp) > wkspSize) return ERROR(GENERIC); - rankStart = wksp->rankStart0 + 1; - ZSTD_memset(wksp->rankStats, 0, sizeof(wksp->rankStats)); - ZSTD_memset(wksp->rankStart0, 0, sizeof(wksp->rankStart0)); + rankStart = wksp->rankStart0 + 1; + ZSTD_memset(wksp->rankStats, 0, sizeof(wksp->rankStats)); + ZSTD_memset(wksp->rankStart0, 0, sizeof(wksp->rankStart0)); DEBUG_STATIC_ASSERT(sizeof(HUF_DEltX2) == sizeof(HUF_DTable)); /* if compiler fails here, assertion is wrong */ if (maxTableLog > HUF_TABLELOG_MAX) return ERROR(tableLog_tooLarge); - /* ZSTD_memset(weightList, 0, sizeof(weightList)); */ /* is not necessary, even though some analyzer complain ... */ + /* ZSTD_memset(weightList, 0, sizeof(weightList)); */ /* is not necessary, even though some analyzer complain ... */ iSize = HUF_readStats_wksp(wksp->weightList, HUF_SYMBOLVALUE_MAX + 1, wksp->rankStats, &nbSymbols, &tableLog, src, srcSize, wksp->calleeWksp, sizeof(wksp->calleeWksp), bmi2); if (HUF_isError(iSize)) return iSize; @@ -1081,14 +1081,14 @@ size_t HUF_readDTableX2_wksp_bmi2(HUF_DTable* DTable, if (tableLog <= HUF_DECODER_FAST_TABLELOG && maxTableLog > HUF_DECODER_FAST_TABLELOG) maxTableLog = HUF_DECODER_FAST_TABLELOG; /* find maxWeight */ - for (maxW = tableLog; wksp->rankStats[maxW]==0; maxW--) {} /* necessarily finds a solution before 0 */ + for (maxW = tableLog; wksp->rankStats[maxW]==0; maxW--) {} /* necessarily finds a solution before 0 */ /* Get start index of each weight */ { U32 w, nextRankStart = 0; for (w=1; w<maxW+1; w++) { - U32 curr = nextRankStart; - nextRankStart += wksp->rankStats[w]; - rankStart[w] = curr; + U32 curr = nextRankStart; + nextRankStart += wksp->rankStats[w]; + rankStart[w] = curr; } rankStart[0] = nextRankStart; /* put all 0w symbols at the end of sorted list*/ rankStart[maxW+1] = nextRankStart; @@ -1097,27 +1097,27 @@ size_t HUF_readDTableX2_wksp_bmi2(HUF_DTable* DTable, /* sort symbols by weight */ { U32 s; for (s=0; s<nbSymbols; s++) { - U32 const w = wksp->weightList[s]; + U32 const w = wksp->weightList[s]; U32 const r = rankStart[w]++; - wksp->sortedSymbol[r].symbol = (BYTE)s; + wksp->sortedSymbol[r].symbol = (BYTE)s; } rankStart[0] = 0; /* forget 0w symbols; this is beginning of weight(1) */ } /* Build rankVal */ - { U32* const rankVal0 = wksp->rankVal[0]; + { U32* const rankVal0 = wksp->rankVal[0]; { int const rescale = (maxTableLog-tableLog) - 1; /* tableLog <= maxTableLog */ U32 nextRankVal = 0; U32 w; for (w=1; w<maxW+1; w++) { - U32 curr = nextRankVal; - nextRankVal += wksp->rankStats[w] << (w+rescale); - rankVal0[w] = curr; + U32 curr = nextRankVal; + nextRankVal += wksp->rankStats[w] << (w+rescale); + rankVal0[w] = curr; } } { U32 const minBits = tableLog+1 - maxW; U32 consumed; for (consumed = minBits; consumed < maxTableLog - minBits + 1; consumed++) { - U32* const rankValPtr = wksp->rankVal[consumed]; + U32* const rankValPtr = wksp->rankVal[consumed]; U32 w; for (w = 1; w < maxW+1; w++) { rankValPtr[w] = rankVal0[w] >> consumed; @@ -1125,12 +1125,12 @@ size_t HUF_readDTableX2_wksp_bmi2(HUF_DTable* DTable, HUF_fillDTableX2(dt, maxTableLog, wksp->sortedSymbol, - wksp->rankStart0, wksp->rankVal, maxW, + wksp->rankStart0, wksp->rankVal, maxW, tableLog+1); dtd.tableLog = (BYTE)maxTableLog; dtd.tableType = 1; - ZSTD_memcpy(DTable, &dtd, sizeof(dtd)); + ZSTD_memcpy(DTable, &dtd, sizeof(dtd)); return iSize; } @@ -1648,8 +1648,8 @@ size_t HUF_decompress1X_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, /* validation checks */ if (dstSize == 0) return ERROR(dstSize_tooSmall); if (cSrcSize > dstSize) return ERROR(corruption_detected); /* invalid */ - if (cSrcSize == dstSize) { ZSTD_memcpy(dst, cSrc, dstSize); return dstSize; } /* not compressed */ - if (cSrcSize == 1) { ZSTD_memset(dst, *(const BYTE*)cSrc, dstSize); return dstSize; } /* RLE */ + if (cSrcSize == dstSize) { ZSTD_memcpy(dst, cSrc, dstSize); return dstSize; } /* not compressed */ + if (cSrcSize == 1) { ZSTD_memset(dst, *(const BYTE*)cSrc, dstSize); return dstSize; } /* RLE */ { U32 const algoNb = HUF_selectDecoder(dstSize, cSrcSize); #if defined(HUF_FORCE_DECOMPRESS_X1) @@ -1694,7 +1694,7 @@ size_t HUF_decompress1X1_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstS { const BYTE* ip = (const BYTE*) cSrc; - size_t const hSize = HUF_readDTableX1_wksp_bmi2(dctx, cSrc, cSrcSize, workSpace, wkspSize, bmi2); + size_t const hSize = HUF_readDTableX1_wksp_bmi2(dctx, cSrc, cSrcSize, workSpace, wkspSize, bmi2); if (HUF_isError(hSize)) return hSize; if (hSize >= cSrcSize) return ERROR(srcSize_wrong); ip += hSize; cSrcSize -= hSize; @@ -1741,149 +1741,149 @@ size_t HUF_decompress4X_hufOnly_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t ds #endif } } - -#ifndef ZSTD_NO_UNUSED_FUNCTIONS -#ifndef HUF_FORCE_DECOMPRESS_X2 -size_t HUF_readDTableX1(HUF_DTable* DTable, const void* src, size_t srcSize) -{ - U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; - return HUF_readDTableX1_wksp(DTable, src, srcSize, - workSpace, sizeof(workSpace)); -} - -size_t HUF_decompress1X1_DCtx(HUF_DTable* DCtx, void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize) -{ - U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; - return HUF_decompress1X1_DCtx_wksp(DCtx, dst, dstSize, cSrc, cSrcSize, - workSpace, sizeof(workSpace)); -} - -size_t HUF_decompress1X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) -{ - HUF_CREATE_STATIC_DTABLEX1(DTable, HUF_TABLELOG_MAX); - return HUF_decompress1X1_DCtx (DTable, dst, dstSize, cSrc, cSrcSize); -} -#endif - -#ifndef HUF_FORCE_DECOMPRESS_X1 -size_t HUF_readDTableX2(HUF_DTable* DTable, const void* src, size_t srcSize) -{ - U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; - return HUF_readDTableX2_wksp(DTable, src, srcSize, - workSpace, sizeof(workSpace)); -} - -size_t HUF_decompress1X2_DCtx(HUF_DTable* DCtx, void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize) -{ - U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; - return HUF_decompress1X2_DCtx_wksp(DCtx, dst, dstSize, cSrc, cSrcSize, - workSpace, sizeof(workSpace)); -} - -size_t HUF_decompress1X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) -{ - HUF_CREATE_STATIC_DTABLEX2(DTable, HUF_TABLELOG_MAX); - return HUF_decompress1X2_DCtx(DTable, dst, dstSize, cSrc, cSrcSize); -} -#endif - -#ifndef HUF_FORCE_DECOMPRESS_X2 -size_t HUF_decompress4X1_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) -{ - U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; - return HUF_decompress4X1_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, - workSpace, sizeof(workSpace)); -} -size_t HUF_decompress4X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) -{ - HUF_CREATE_STATIC_DTABLEX1(DTable, HUF_TABLELOG_MAX); - return HUF_decompress4X1_DCtx(DTable, dst, dstSize, cSrc, cSrcSize); -} -#endif - -#ifndef HUF_FORCE_DECOMPRESS_X1 -size_t HUF_decompress4X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize) -{ - U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; - return HUF_decompress4X2_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, - workSpace, sizeof(workSpace)); -} - -size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) -{ - HUF_CREATE_STATIC_DTABLEX2(DTable, HUF_TABLELOG_MAX); - return HUF_decompress4X2_DCtx(DTable, dst, dstSize, cSrc, cSrcSize); -} -#endif - -typedef size_t (*decompressionAlgo)(void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); - -size_t HUF_decompress (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) -{ -#if !defined(HUF_FORCE_DECOMPRESS_X1) && !defined(HUF_FORCE_DECOMPRESS_X2) - static const decompressionAlgo decompress[2] = { HUF_decompress4X1, HUF_decompress4X2 }; -#endif - - /* validation checks */ - if (dstSize == 0) return ERROR(dstSize_tooSmall); - if (cSrcSize > dstSize) return ERROR(corruption_detected); /* invalid */ - if (cSrcSize == dstSize) { ZSTD_memcpy(dst, cSrc, dstSize); return dstSize; } /* not compressed */ - if (cSrcSize == 1) { ZSTD_memset(dst, *(const BYTE*)cSrc, dstSize); return dstSize; } /* RLE */ - - { U32 const algoNb = HUF_selectDecoder(dstSize, cSrcSize); -#if defined(HUF_FORCE_DECOMPRESS_X1) - (void)algoNb; - assert(algoNb == 0); - return HUF_decompress4X1(dst, dstSize, cSrc, cSrcSize); -#elif defined(HUF_FORCE_DECOMPRESS_X2) - (void)algoNb; - assert(algoNb == 1); - return HUF_decompress4X2(dst, dstSize, cSrc, cSrcSize); -#else - return decompress[algoNb](dst, dstSize, cSrc, cSrcSize); -#endif - } -} - -size_t HUF_decompress4X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) -{ - /* validation checks */ - if (dstSize == 0) return ERROR(dstSize_tooSmall); - if (cSrcSize > dstSize) return ERROR(corruption_detected); /* invalid */ - if (cSrcSize == dstSize) { ZSTD_memcpy(dst, cSrc, dstSize); return dstSize; } /* not compressed */ - if (cSrcSize == 1) { ZSTD_memset(dst, *(const BYTE*)cSrc, dstSize); return dstSize; } /* RLE */ - - { U32 const algoNb = HUF_selectDecoder(dstSize, cSrcSize); -#if defined(HUF_FORCE_DECOMPRESS_X1) - (void)algoNb; - assert(algoNb == 0); - return HUF_decompress4X1_DCtx(dctx, dst, dstSize, cSrc, cSrcSize); -#elif defined(HUF_FORCE_DECOMPRESS_X2) - (void)algoNb; - assert(algoNb == 1); - return HUF_decompress4X2_DCtx(dctx, dst, dstSize, cSrc, cSrcSize); -#else - return algoNb ? HUF_decompress4X2_DCtx(dctx, dst, dstSize, cSrc, cSrcSize) : - HUF_decompress4X1_DCtx(dctx, dst, dstSize, cSrc, cSrcSize) ; -#endif - } -} - -size_t HUF_decompress4X_hufOnly(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) -{ - U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; - return HUF_decompress4X_hufOnly_wksp(dctx, dst, dstSize, cSrc, cSrcSize, - workSpace, sizeof(workSpace)); -} - -size_t HUF_decompress1X_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize) -{ - U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; - return HUF_decompress1X_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, - workSpace, sizeof(workSpace)); -} -#endif + +#ifndef ZSTD_NO_UNUSED_FUNCTIONS +#ifndef HUF_FORCE_DECOMPRESS_X2 +size_t HUF_readDTableX1(HUF_DTable* DTable, const void* src, size_t srcSize) +{ + U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; + return HUF_readDTableX1_wksp(DTable, src, srcSize, + workSpace, sizeof(workSpace)); +} + +size_t HUF_decompress1X1_DCtx(HUF_DTable* DCtx, void* dst, size_t dstSize, + const void* cSrc, size_t cSrcSize) +{ + U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; + return HUF_decompress1X1_DCtx_wksp(DCtx, dst, dstSize, cSrc, cSrcSize, + workSpace, sizeof(workSpace)); +} + +size_t HUF_decompress1X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) +{ + HUF_CREATE_STATIC_DTABLEX1(DTable, HUF_TABLELOG_MAX); + return HUF_decompress1X1_DCtx (DTable, dst, dstSize, cSrc, cSrcSize); +} +#endif + +#ifndef HUF_FORCE_DECOMPRESS_X1 +size_t HUF_readDTableX2(HUF_DTable* DTable, const void* src, size_t srcSize) +{ + U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; + return HUF_readDTableX2_wksp(DTable, src, srcSize, + workSpace, sizeof(workSpace)); +} + +size_t HUF_decompress1X2_DCtx(HUF_DTable* DCtx, void* dst, size_t dstSize, + const void* cSrc, size_t cSrcSize) +{ + U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; + return HUF_decompress1X2_DCtx_wksp(DCtx, dst, dstSize, cSrc, cSrcSize, + workSpace, sizeof(workSpace)); +} + +size_t HUF_decompress1X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) +{ + HUF_CREATE_STATIC_DTABLEX2(DTable, HUF_TABLELOG_MAX); + return HUF_decompress1X2_DCtx(DTable, dst, dstSize, cSrc, cSrcSize); +} +#endif + +#ifndef HUF_FORCE_DECOMPRESS_X2 +size_t HUF_decompress4X1_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) +{ + U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; + return HUF_decompress4X1_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, + workSpace, sizeof(workSpace)); +} +size_t HUF_decompress4X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) +{ + HUF_CREATE_STATIC_DTABLEX1(DTable, HUF_TABLELOG_MAX); + return HUF_decompress4X1_DCtx(DTable, dst, dstSize, cSrc, cSrcSize); +} +#endif + +#ifndef HUF_FORCE_DECOMPRESS_X1 +size_t HUF_decompress4X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, + const void* cSrc, size_t cSrcSize) +{ + U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; + return HUF_decompress4X2_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, + workSpace, sizeof(workSpace)); +} + +size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) +{ + HUF_CREATE_STATIC_DTABLEX2(DTable, HUF_TABLELOG_MAX); + return HUF_decompress4X2_DCtx(DTable, dst, dstSize, cSrc, cSrcSize); +} +#endif + +typedef size_t (*decompressionAlgo)(void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); + +size_t HUF_decompress (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) +{ +#if !defined(HUF_FORCE_DECOMPRESS_X1) && !defined(HUF_FORCE_DECOMPRESS_X2) + static const decompressionAlgo decompress[2] = { HUF_decompress4X1, HUF_decompress4X2 }; +#endif + + /* validation checks */ + if (dstSize == 0) return ERROR(dstSize_tooSmall); + if (cSrcSize > dstSize) return ERROR(corruption_detected); /* invalid */ + if (cSrcSize == dstSize) { ZSTD_memcpy(dst, cSrc, dstSize); return dstSize; } /* not compressed */ + if (cSrcSize == 1) { ZSTD_memset(dst, *(const BYTE*)cSrc, dstSize); return dstSize; } /* RLE */ + + { U32 const algoNb = HUF_selectDecoder(dstSize, cSrcSize); +#if defined(HUF_FORCE_DECOMPRESS_X1) + (void)algoNb; + assert(algoNb == 0); + return HUF_decompress4X1(dst, dstSize, cSrc, cSrcSize); +#elif defined(HUF_FORCE_DECOMPRESS_X2) + (void)algoNb; + assert(algoNb == 1); + return HUF_decompress4X2(dst, dstSize, cSrc, cSrcSize); +#else + return decompress[algoNb](dst, dstSize, cSrc, cSrcSize); +#endif + } +} + +size_t HUF_decompress4X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) +{ + /* validation checks */ + if (dstSize == 0) return ERROR(dstSize_tooSmall); + if (cSrcSize > dstSize) return ERROR(corruption_detected); /* invalid */ + if (cSrcSize == dstSize) { ZSTD_memcpy(dst, cSrc, dstSize); return dstSize; } /* not compressed */ + if (cSrcSize == 1) { ZSTD_memset(dst, *(const BYTE*)cSrc, dstSize); return dstSize; } /* RLE */ + + { U32 const algoNb = HUF_selectDecoder(dstSize, cSrcSize); +#if defined(HUF_FORCE_DECOMPRESS_X1) + (void)algoNb; + assert(algoNb == 0); + return HUF_decompress4X1_DCtx(dctx, dst, dstSize, cSrc, cSrcSize); +#elif defined(HUF_FORCE_DECOMPRESS_X2) + (void)algoNb; + assert(algoNb == 1); + return HUF_decompress4X2_DCtx(dctx, dst, dstSize, cSrc, cSrcSize); +#else + return algoNb ? HUF_decompress4X2_DCtx(dctx, dst, dstSize, cSrc, cSrcSize) : + HUF_decompress4X1_DCtx(dctx, dst, dstSize, cSrc, cSrcSize) ; +#endif + } +} + +size_t HUF_decompress4X_hufOnly(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) +{ + U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; + return HUF_decompress4X_hufOnly_wksp(dctx, dst, dstSize, cSrc, cSrcSize, + workSpace, sizeof(workSpace)); +} + +size_t HUF_decompress1X_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, + const void* cSrc, size_t cSrcSize) +{ + U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; + return HUF_decompress1X_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, + workSpace, sizeof(workSpace)); +} +#endif diff --git a/contrib/libs/zstd/lib/decompress/zstd_ddict.c b/contrib/libs/zstd/lib/decompress/zstd_ddict.c index ce335477b3..821f1e0213 100644 --- a/contrib/libs/zstd/lib/decompress/zstd_ddict.c +++ b/contrib/libs/zstd/lib/decompress/zstd_ddict.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -14,7 +14,7 @@ /*-******************************************************* * Dependencies *********************************************************/ -#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memmove, ZSTD_memset */ +#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memmove, ZSTD_memset */ #include "../common/cpu.h" /* bmi2 */ #include "../common/mem.h" /* low level memory routines */ #define FSE_STATIC_LINKING_ONLY @@ -127,11 +127,11 @@ static size_t ZSTD_initDDict_internal(ZSTD_DDict* ddict, ddict->dictContent = dict; if (!dict) dictSize = 0; } else { - void* const internalBuffer = ZSTD_customMalloc(dictSize, ddict->cMem); + void* const internalBuffer = ZSTD_customMalloc(dictSize, ddict->cMem); ddict->dictBuffer = internalBuffer; ddict->dictContent = internalBuffer; if (!internalBuffer) return ERROR(memory_allocation); - ZSTD_memcpy(internalBuffer, dict, dictSize); + ZSTD_memcpy(internalBuffer, dict, dictSize); } ddict->dictSize = dictSize; ddict->entropy.hufTable[0] = (HUF_DTable)((HufLog)*0x1000001); /* cover both little and big endian */ @@ -147,9 +147,9 @@ ZSTD_DDict* ZSTD_createDDict_advanced(const void* dict, size_t dictSize, ZSTD_dictContentType_e dictContentType, ZSTD_customMem customMem) { - if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL; + if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL; - { ZSTD_DDict* const ddict = (ZSTD_DDict*) ZSTD_customMalloc(sizeof(ZSTD_DDict), customMem); + { ZSTD_DDict* const ddict = (ZSTD_DDict*) ZSTD_customMalloc(sizeof(ZSTD_DDict), customMem); if (ddict == NULL) return NULL; ddict->cMem = customMem; { size_t const initResult = ZSTD_initDDict_internal(ddict, @@ -198,7 +198,7 @@ const ZSTD_DDict* ZSTD_initStaticDDict( if ((size_t)sBuffer & 7) return NULL; /* 8-aligned */ if (sBufferSize < neededSpace) return NULL; if (dictLoadMethod == ZSTD_dlm_byCopy) { - ZSTD_memcpy(ddict+1, dict, dictSize); /* local copy */ + ZSTD_memcpy(ddict+1, dict, dictSize); /* local copy */ dict = ddict+1; } if (ZSTD_isError( ZSTD_initDDict_internal(ddict, @@ -213,8 +213,8 @@ size_t ZSTD_freeDDict(ZSTD_DDict* ddict) { if (ddict==NULL) return 0; /* support free on NULL */ { ZSTD_customMem const cMem = ddict->cMem; - ZSTD_customFree(ddict->dictBuffer, cMem); - ZSTD_customFree(ddict, cMem); + ZSTD_customFree(ddict->dictBuffer, cMem); + ZSTD_customFree(ddict, cMem); return 0; } } diff --git a/contrib/libs/zstd/lib/decompress/zstd_ddict.h b/contrib/libs/zstd/lib/decompress/zstd_ddict.h index bd03268b50..7bff4f3374 100644 --- a/contrib/libs/zstd/lib/decompress/zstd_ddict.h +++ b/contrib/libs/zstd/lib/decompress/zstd_ddict.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -15,7 +15,7 @@ /*-******************************************************* * Dependencies *********************************************************/ -#include "../common/zstd_deps.h" /* size_t */ +#include "../common/zstd_deps.h" /* size_t */ #include "../zstd.h" /* ZSTD_DDict, and several public functions */ diff --git a/contrib/libs/zstd/lib/decompress/zstd_decompress.c b/contrib/libs/zstd/lib/decompress/zstd_decompress.c index b8bbefd538..46ec6f950f 100644 --- a/contrib/libs/zstd/lib/decompress/zstd_decompress.c +++ b/contrib/libs/zstd/lib/decompress/zstd_decompress.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -55,13 +55,13 @@ /*-******************************************************* * Dependencies *********************************************************/ -#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memmove, ZSTD_memset */ +#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memmove, ZSTD_memset */ #include "../common/mem.h" /* low level memory routines */ #define FSE_STATIC_LINKING_ONLY #include "../common/fse.h" #define HUF_STATIC_LINKING_ONLY #include "../common/huf.h" -#include <contrib/libs/xxhash/xxhash.h> /* XXH64_reset, XXH64_update, XXH64_digest, XXH64 */ +#include <contrib/libs/xxhash/xxhash.h> /* XXH64_reset, XXH64_update, XXH64_digest, XXH64 */ #include "../common/zstd_internal.h" /* blockProperties_t */ #include "zstd_decompress_internal.h" /* ZSTD_DCtx */ #include "zstd_ddict.h" /* ZSTD_DDictDictContent */ @@ -72,147 +72,147 @@ #endif - -/************************************* - * Multiple DDicts Hashset internals * - *************************************/ - -#define DDICT_HASHSET_MAX_LOAD_FACTOR_COUNT_MULT 4 -#define DDICT_HASHSET_MAX_LOAD_FACTOR_SIZE_MULT 3 /* These two constants represent SIZE_MULT/COUNT_MULT load factor without using a float. - * Currently, that means a 0.75 load factor. - * So, if count * COUNT_MULT / size * SIZE_MULT != 0, then we've exceeded - * the load factor of the ddict hash set. - */ - -#define DDICT_HASHSET_TABLE_BASE_SIZE 64 -#define DDICT_HASHSET_RESIZE_FACTOR 2 - -/* Hash function to determine starting position of dict insertion within the table - * Returns an index between [0, hashSet->ddictPtrTableSize] - */ -static size_t ZSTD_DDictHashSet_getIndex(const ZSTD_DDictHashSet* hashSet, U32 dictID) { - const U64 hash = XXH64(&dictID, sizeof(U32), 0); - /* DDict ptr table size is a multiple of 2, use size - 1 as mask to get index within [0, hashSet->ddictPtrTableSize) */ - return hash & (hashSet->ddictPtrTableSize - 1); -} - -/* Adds DDict to a hashset without resizing it. - * If inserting a DDict with a dictID that already exists in the set, replaces the one in the set. - * Returns 0 if successful, or a zstd error code if something went wrong. - */ -static size_t ZSTD_DDictHashSet_emplaceDDict(ZSTD_DDictHashSet* hashSet, const ZSTD_DDict* ddict) { - const U32 dictID = ZSTD_getDictID_fromDDict(ddict); - size_t idx = ZSTD_DDictHashSet_getIndex(hashSet, dictID); - const size_t idxRangeMask = hashSet->ddictPtrTableSize - 1; - RETURN_ERROR_IF(hashSet->ddictPtrCount == hashSet->ddictPtrTableSize, GENERIC, "Hash set is full!"); - DEBUGLOG(4, "Hashed index: for dictID: %u is %zu", dictID, idx); - while (hashSet->ddictPtrTable[idx] != NULL) { - /* Replace existing ddict if inserting ddict with same dictID */ - if (ZSTD_getDictID_fromDDict(hashSet->ddictPtrTable[idx]) == dictID) { - DEBUGLOG(4, "DictID already exists, replacing rather than adding"); - hashSet->ddictPtrTable[idx] = ddict; - return 0; - } - idx &= idxRangeMask; - idx++; - } - DEBUGLOG(4, "Final idx after probing for dictID %u is: %zu", dictID, idx); - hashSet->ddictPtrTable[idx] = ddict; - hashSet->ddictPtrCount++; - return 0; -} - -/* Expands hash table by factor of DDICT_HASHSET_RESIZE_FACTOR and - * rehashes all values, allocates new table, frees old table. - * Returns 0 on success, otherwise a zstd error code. - */ -static size_t ZSTD_DDictHashSet_expand(ZSTD_DDictHashSet* hashSet, ZSTD_customMem customMem) { - size_t newTableSize = hashSet->ddictPtrTableSize * DDICT_HASHSET_RESIZE_FACTOR; - const ZSTD_DDict** newTable = (const ZSTD_DDict**)ZSTD_customCalloc(sizeof(ZSTD_DDict*) * newTableSize, customMem); - const ZSTD_DDict** oldTable = hashSet->ddictPtrTable; - size_t oldTableSize = hashSet->ddictPtrTableSize; - size_t i; - - DEBUGLOG(4, "Expanding DDict hash table! Old size: %zu new size: %zu", oldTableSize, newTableSize); - RETURN_ERROR_IF(!newTable, memory_allocation, "Expanded hashset allocation failed!"); - hashSet->ddictPtrTable = newTable; - hashSet->ddictPtrTableSize = newTableSize; - hashSet->ddictPtrCount = 0; - for (i = 0; i < oldTableSize; ++i) { - if (oldTable[i] != NULL) { - FORWARD_IF_ERROR(ZSTD_DDictHashSet_emplaceDDict(hashSet, oldTable[i]), ""); - } - } - ZSTD_customFree((void*)oldTable, customMem); - DEBUGLOG(4, "Finished re-hash"); - return 0; -} - -/* Fetches a DDict with the given dictID - * Returns the ZSTD_DDict* with the requested dictID. If it doesn't exist, then returns NULL. - */ -static const ZSTD_DDict* ZSTD_DDictHashSet_getDDict(ZSTD_DDictHashSet* hashSet, U32 dictID) { - size_t idx = ZSTD_DDictHashSet_getIndex(hashSet, dictID); - const size_t idxRangeMask = hashSet->ddictPtrTableSize - 1; - DEBUGLOG(4, "Hashed index: for dictID: %u is %zu", dictID, idx); - for (;;) { - size_t currDictID = ZSTD_getDictID_fromDDict(hashSet->ddictPtrTable[idx]); - if (currDictID == dictID || currDictID == 0) { - /* currDictID == 0 implies a NULL ddict entry */ - break; - } else { - idx &= idxRangeMask; /* Goes to start of table when we reach the end */ - idx++; - } - } - DEBUGLOG(4, "Final idx after probing for dictID %u is: %zu", dictID, idx); - return hashSet->ddictPtrTable[idx]; -} - -/* Allocates space for and returns a ddict hash set - * The hash set's ZSTD_DDict* table has all values automatically set to NULL to begin with. - * Returns NULL if allocation failed. - */ -static ZSTD_DDictHashSet* ZSTD_createDDictHashSet(ZSTD_customMem customMem) { - ZSTD_DDictHashSet* ret = (ZSTD_DDictHashSet*)ZSTD_customMalloc(sizeof(ZSTD_DDictHashSet), customMem); - DEBUGLOG(4, "Allocating new hash set"); + +/************************************* + * Multiple DDicts Hashset internals * + *************************************/ + +#define DDICT_HASHSET_MAX_LOAD_FACTOR_COUNT_MULT 4 +#define DDICT_HASHSET_MAX_LOAD_FACTOR_SIZE_MULT 3 /* These two constants represent SIZE_MULT/COUNT_MULT load factor without using a float. + * Currently, that means a 0.75 load factor. + * So, if count * COUNT_MULT / size * SIZE_MULT != 0, then we've exceeded + * the load factor of the ddict hash set. + */ + +#define DDICT_HASHSET_TABLE_BASE_SIZE 64 +#define DDICT_HASHSET_RESIZE_FACTOR 2 + +/* Hash function to determine starting position of dict insertion within the table + * Returns an index between [0, hashSet->ddictPtrTableSize] + */ +static size_t ZSTD_DDictHashSet_getIndex(const ZSTD_DDictHashSet* hashSet, U32 dictID) { + const U64 hash = XXH64(&dictID, sizeof(U32), 0); + /* DDict ptr table size is a multiple of 2, use size - 1 as mask to get index within [0, hashSet->ddictPtrTableSize) */ + return hash & (hashSet->ddictPtrTableSize - 1); +} + +/* Adds DDict to a hashset without resizing it. + * If inserting a DDict with a dictID that already exists in the set, replaces the one in the set. + * Returns 0 if successful, or a zstd error code if something went wrong. + */ +static size_t ZSTD_DDictHashSet_emplaceDDict(ZSTD_DDictHashSet* hashSet, const ZSTD_DDict* ddict) { + const U32 dictID = ZSTD_getDictID_fromDDict(ddict); + size_t idx = ZSTD_DDictHashSet_getIndex(hashSet, dictID); + const size_t idxRangeMask = hashSet->ddictPtrTableSize - 1; + RETURN_ERROR_IF(hashSet->ddictPtrCount == hashSet->ddictPtrTableSize, GENERIC, "Hash set is full!"); + DEBUGLOG(4, "Hashed index: for dictID: %u is %zu", dictID, idx); + while (hashSet->ddictPtrTable[idx] != NULL) { + /* Replace existing ddict if inserting ddict with same dictID */ + if (ZSTD_getDictID_fromDDict(hashSet->ddictPtrTable[idx]) == dictID) { + DEBUGLOG(4, "DictID already exists, replacing rather than adding"); + hashSet->ddictPtrTable[idx] = ddict; + return 0; + } + idx &= idxRangeMask; + idx++; + } + DEBUGLOG(4, "Final idx after probing for dictID %u is: %zu", dictID, idx); + hashSet->ddictPtrTable[idx] = ddict; + hashSet->ddictPtrCount++; + return 0; +} + +/* Expands hash table by factor of DDICT_HASHSET_RESIZE_FACTOR and + * rehashes all values, allocates new table, frees old table. + * Returns 0 on success, otherwise a zstd error code. + */ +static size_t ZSTD_DDictHashSet_expand(ZSTD_DDictHashSet* hashSet, ZSTD_customMem customMem) { + size_t newTableSize = hashSet->ddictPtrTableSize * DDICT_HASHSET_RESIZE_FACTOR; + const ZSTD_DDict** newTable = (const ZSTD_DDict**)ZSTD_customCalloc(sizeof(ZSTD_DDict*) * newTableSize, customMem); + const ZSTD_DDict** oldTable = hashSet->ddictPtrTable; + size_t oldTableSize = hashSet->ddictPtrTableSize; + size_t i; + + DEBUGLOG(4, "Expanding DDict hash table! Old size: %zu new size: %zu", oldTableSize, newTableSize); + RETURN_ERROR_IF(!newTable, memory_allocation, "Expanded hashset allocation failed!"); + hashSet->ddictPtrTable = newTable; + hashSet->ddictPtrTableSize = newTableSize; + hashSet->ddictPtrCount = 0; + for (i = 0; i < oldTableSize; ++i) { + if (oldTable[i] != NULL) { + FORWARD_IF_ERROR(ZSTD_DDictHashSet_emplaceDDict(hashSet, oldTable[i]), ""); + } + } + ZSTD_customFree((void*)oldTable, customMem); + DEBUGLOG(4, "Finished re-hash"); + return 0; +} + +/* Fetches a DDict with the given dictID + * Returns the ZSTD_DDict* with the requested dictID. If it doesn't exist, then returns NULL. + */ +static const ZSTD_DDict* ZSTD_DDictHashSet_getDDict(ZSTD_DDictHashSet* hashSet, U32 dictID) { + size_t idx = ZSTD_DDictHashSet_getIndex(hashSet, dictID); + const size_t idxRangeMask = hashSet->ddictPtrTableSize - 1; + DEBUGLOG(4, "Hashed index: for dictID: %u is %zu", dictID, idx); + for (;;) { + size_t currDictID = ZSTD_getDictID_fromDDict(hashSet->ddictPtrTable[idx]); + if (currDictID == dictID || currDictID == 0) { + /* currDictID == 0 implies a NULL ddict entry */ + break; + } else { + idx &= idxRangeMask; /* Goes to start of table when we reach the end */ + idx++; + } + } + DEBUGLOG(4, "Final idx after probing for dictID %u is: %zu", dictID, idx); + return hashSet->ddictPtrTable[idx]; +} + +/* Allocates space for and returns a ddict hash set + * The hash set's ZSTD_DDict* table has all values automatically set to NULL to begin with. + * Returns NULL if allocation failed. + */ +static ZSTD_DDictHashSet* ZSTD_createDDictHashSet(ZSTD_customMem customMem) { + ZSTD_DDictHashSet* ret = (ZSTD_DDictHashSet*)ZSTD_customMalloc(sizeof(ZSTD_DDictHashSet), customMem); + DEBUGLOG(4, "Allocating new hash set"); if (!ret) return NULL; - ret->ddictPtrTable = (const ZSTD_DDict**)ZSTD_customCalloc(DDICT_HASHSET_TABLE_BASE_SIZE * sizeof(ZSTD_DDict*), customMem); + ret->ddictPtrTable = (const ZSTD_DDict**)ZSTD_customCalloc(DDICT_HASHSET_TABLE_BASE_SIZE * sizeof(ZSTD_DDict*), customMem); if (!ret->ddictPtrTable) { ZSTD_customFree(ret, customMem); return NULL; } - ret->ddictPtrTableSize = DDICT_HASHSET_TABLE_BASE_SIZE; - ret->ddictPtrCount = 0; - return ret; -} - -/* Frees the table of ZSTD_DDict* within a hashset, then frees the hashset itself. - * Note: The ZSTD_DDict* within the table are NOT freed. - */ -static void ZSTD_freeDDictHashSet(ZSTD_DDictHashSet* hashSet, ZSTD_customMem customMem) { - DEBUGLOG(4, "Freeing ddict hash set"); - if (hashSet && hashSet->ddictPtrTable) { - ZSTD_customFree((void*)hashSet->ddictPtrTable, customMem); - } - if (hashSet) { - ZSTD_customFree(hashSet, customMem); - } -} - -/* Public function: Adds a DDict into the ZSTD_DDictHashSet, possibly triggering a resize of the hash set. - * Returns 0 on success, or a ZSTD error. - */ -static size_t ZSTD_DDictHashSet_addDDict(ZSTD_DDictHashSet* hashSet, const ZSTD_DDict* ddict, ZSTD_customMem customMem) { - DEBUGLOG(4, "Adding dict ID: %u to hashset with - Count: %zu Tablesize: %zu", ZSTD_getDictID_fromDDict(ddict), hashSet->ddictPtrCount, hashSet->ddictPtrTableSize); - if (hashSet->ddictPtrCount * DDICT_HASHSET_MAX_LOAD_FACTOR_COUNT_MULT / hashSet->ddictPtrTableSize * DDICT_HASHSET_MAX_LOAD_FACTOR_SIZE_MULT != 0) { - FORWARD_IF_ERROR(ZSTD_DDictHashSet_expand(hashSet, customMem), ""); - } - FORWARD_IF_ERROR(ZSTD_DDictHashSet_emplaceDDict(hashSet, ddict), ""); - return 0; -} - + ret->ddictPtrTableSize = DDICT_HASHSET_TABLE_BASE_SIZE; + ret->ddictPtrCount = 0; + return ret; +} + +/* Frees the table of ZSTD_DDict* within a hashset, then frees the hashset itself. + * Note: The ZSTD_DDict* within the table are NOT freed. + */ +static void ZSTD_freeDDictHashSet(ZSTD_DDictHashSet* hashSet, ZSTD_customMem customMem) { + DEBUGLOG(4, "Freeing ddict hash set"); + if (hashSet && hashSet->ddictPtrTable) { + ZSTD_customFree((void*)hashSet->ddictPtrTable, customMem); + } + if (hashSet) { + ZSTD_customFree(hashSet, customMem); + } +} + +/* Public function: Adds a DDict into the ZSTD_DDictHashSet, possibly triggering a resize of the hash set. + * Returns 0 on success, or a ZSTD error. + */ +static size_t ZSTD_DDictHashSet_addDDict(ZSTD_DDictHashSet* hashSet, const ZSTD_DDict* ddict, ZSTD_customMem customMem) { + DEBUGLOG(4, "Adding dict ID: %u to hashset with - Count: %zu Tablesize: %zu", ZSTD_getDictID_fromDDict(ddict), hashSet->ddictPtrCount, hashSet->ddictPtrTableSize); + if (hashSet->ddictPtrCount * DDICT_HASHSET_MAX_LOAD_FACTOR_COUNT_MULT / hashSet->ddictPtrTableSize * DDICT_HASHSET_MAX_LOAD_FACTOR_SIZE_MULT != 0) { + FORWARD_IF_ERROR(ZSTD_DDictHashSet_expand(hashSet, customMem), ""); + } + FORWARD_IF_ERROR(ZSTD_DDictHashSet_emplaceDDict(hashSet, ddict), ""); + return 0; +} + /*-************************************************************* * Context management ***************************************************************/ @@ -235,16 +235,16 @@ static size_t ZSTD_startingInputLength(ZSTD_format_e format) return startingInputLength; } -static void ZSTD_DCtx_resetParameters(ZSTD_DCtx* dctx) -{ - assert(dctx->streamStage == zdss_init); - dctx->format = ZSTD_f_zstd1; - dctx->maxWindowSize = ZSTD_MAXWINDOWSIZE_DEFAULT; - dctx->outBufferMode = ZSTD_bm_buffered; - dctx->forceIgnoreChecksum = ZSTD_d_validateChecksum; - dctx->refMultipleDDicts = ZSTD_rmd_refSingleDDict; -} - +static void ZSTD_DCtx_resetParameters(ZSTD_DCtx* dctx) +{ + assert(dctx->streamStage == zdss_init); + dctx->format = ZSTD_f_zstd1; + dctx->maxWindowSize = ZSTD_MAXWINDOWSIZE_DEFAULT; + dctx->outBufferMode = ZSTD_bm_buffered; + dctx->forceIgnoreChecksum = ZSTD_d_validateChecksum; + dctx->refMultipleDDicts = ZSTD_rmd_refSingleDDict; +} + static void ZSTD_initDCtx_internal(ZSTD_DCtx* dctx) { dctx->staticSize = 0; @@ -266,8 +266,8 @@ static void ZSTD_initDCtx_internal(ZSTD_DCtx* dctx) #if DYNAMIC_BMI2 dctx->bmi2 = ZSTD_cpuSupportsBmi2(); #endif - dctx->ddictSet = NULL; - ZSTD_DCtx_resetParameters(dctx); + dctx->ddictSet = NULL; + ZSTD_DCtx_resetParameters(dctx); #ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION dctx->dictContentEndForFuzzing = NULL; #endif @@ -287,9 +287,9 @@ ZSTD_DCtx* ZSTD_initStaticDCtx(void *workspace, size_t workspaceSize) } static ZSTD_DCtx* ZSTD_createDCtx_internal(ZSTD_customMem customMem) { - if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL; + if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL; - { ZSTD_DCtx* const dctx = (ZSTD_DCtx*)ZSTD_customMalloc(sizeof(*dctx), customMem); + { ZSTD_DCtx* const dctx = (ZSTD_DCtx*)ZSTD_customMalloc(sizeof(*dctx), customMem); if (!dctx) return NULL; dctx->customMem = customMem; ZSTD_initDCtx_internal(dctx); @@ -322,17 +322,17 @@ size_t ZSTD_freeDCtx(ZSTD_DCtx* dctx) RETURN_ERROR_IF(dctx->staticSize, memory_allocation, "not compatible with static DCtx"); { ZSTD_customMem const cMem = dctx->customMem; ZSTD_clearDict(dctx); - ZSTD_customFree(dctx->inBuff, cMem); + ZSTD_customFree(dctx->inBuff, cMem); dctx->inBuff = NULL; #if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT >= 1) if (dctx->legacyContext) ZSTD_freeLegacyStreamContext(dctx->legacyContext, dctx->previousLegacyVersion); #endif - if (dctx->ddictSet) { - ZSTD_freeDDictHashSet(dctx->ddictSet, cMem); - dctx->ddictSet = NULL; - } - ZSTD_customFree(dctx, cMem); + if (dctx->ddictSet) { + ZSTD_freeDDictHashSet(dctx->ddictSet, cMem); + dctx->ddictSet = NULL; + } + ZSTD_customFree(dctx, cMem); return 0; } } @@ -341,33 +341,33 @@ size_t ZSTD_freeDCtx(ZSTD_DCtx* dctx) void ZSTD_copyDCtx(ZSTD_DCtx* dstDCtx, const ZSTD_DCtx* srcDCtx) { size_t const toCopy = (size_t)((char*)(&dstDCtx->inBuff) - (char*)dstDCtx); - ZSTD_memcpy(dstDCtx, srcDCtx, toCopy); /* no need to copy workspace */ -} - -/* Given a dctx with a digested frame params, re-selects the correct ZSTD_DDict based on - * the requested dict ID from the frame. If there exists a reference to the correct ZSTD_DDict, then - * accordingly sets the ddict to be used to decompress the frame. - * - * If no DDict is found, then no action is taken, and the ZSTD_DCtx::ddict remains as-is. - * - * ZSTD_d_refMultipleDDicts must be enabled for this function to be called. - */ -static void ZSTD_DCtx_selectFrameDDict(ZSTD_DCtx* dctx) { - assert(dctx->refMultipleDDicts && dctx->ddictSet); - DEBUGLOG(4, "Adjusting DDict based on requested dict ID from frame"); - if (dctx->ddict) { - const ZSTD_DDict* frameDDict = ZSTD_DDictHashSet_getDDict(dctx->ddictSet, dctx->fParams.dictID); - if (frameDDict) { - DEBUGLOG(4, "DDict found!"); - ZSTD_clearDict(dctx); - dctx->dictID = dctx->fParams.dictID; - dctx->ddict = frameDDict; - dctx->dictUses = ZSTD_use_indefinitely; - } - } -} - - + ZSTD_memcpy(dstDCtx, srcDCtx, toCopy); /* no need to copy workspace */ +} + +/* Given a dctx with a digested frame params, re-selects the correct ZSTD_DDict based on + * the requested dict ID from the frame. If there exists a reference to the correct ZSTD_DDict, then + * accordingly sets the ddict to be used to decompress the frame. + * + * If no DDict is found, then no action is taken, and the ZSTD_DCtx::ddict remains as-is. + * + * ZSTD_d_refMultipleDDicts must be enabled for this function to be called. + */ +static void ZSTD_DCtx_selectFrameDDict(ZSTD_DCtx* dctx) { + assert(dctx->refMultipleDDicts && dctx->ddictSet); + DEBUGLOG(4, "Adjusting DDict based on requested dict ID from frame"); + if (dctx->ddict) { + const ZSTD_DDict* frameDDict = ZSTD_DDictHashSet_getDDict(dctx->ddictSet, dctx->fParams.dictID); + if (frameDDict) { + DEBUGLOG(4, "DDict found!"); + ZSTD_clearDict(dctx); + dctx->dictID = dctx->fParams.dictID; + dctx->ddict = frameDDict; + dctx->dictUses = ZSTD_use_indefinitely; + } + } +} + + /*-************************************************************* * Frame header decoding ***************************************************************/ @@ -444,7 +444,7 @@ size_t ZSTD_getFrameHeader_advanced(ZSTD_frameHeader* zfhPtr, const void* src, s const BYTE* ip = (const BYTE*)src; size_t const minInputSize = ZSTD_startingInputLength(format); - ZSTD_memset(zfhPtr, 0, sizeof(*zfhPtr)); /* not strictly necessary, but static analyzer do not understand that zfhPtr is only going to be read only if return value is zero, since they are 2 different signals */ + ZSTD_memset(zfhPtr, 0, sizeof(*zfhPtr)); /* not strictly necessary, but static analyzer do not understand that zfhPtr is only going to be read only if return value is zero, since they are 2 different signals */ if (srcSize < minInputSize) return minInputSize; RETURN_ERROR_IF(src==NULL, GENERIC, "invalid parameter"); @@ -454,7 +454,7 @@ size_t ZSTD_getFrameHeader_advanced(ZSTD_frameHeader* zfhPtr, const void* src, s /* skippable frame */ if (srcSize < ZSTD_SKIPPABLEHEADERSIZE) return ZSTD_SKIPPABLEHEADERSIZE; /* magic number + frame length */ - ZSTD_memset(zfhPtr, 0, sizeof(*zfhPtr)); + ZSTD_memset(zfhPtr, 0, sizeof(*zfhPtr)); zfhPtr->frameContentSize = MEM_readLE32((const char *)src + ZSTD_FRAMEIDSIZE); zfhPtr->frameType = ZSTD_skippableFrame; return 0; @@ -665,19 +665,19 @@ unsigned long long ZSTD_getDecompressedSize(const void* src, size_t srcSize) /** ZSTD_decodeFrameHeader() : * `headerSize` must be the size provided by ZSTD_frameHeaderSize(). - * If multiple DDict references are enabled, also will choose the correct DDict to use. + * If multiple DDict references are enabled, also will choose the correct DDict to use. * @return : 0 if success, or an error code, which can be tested using ZSTD_isError() */ static size_t ZSTD_decodeFrameHeader(ZSTD_DCtx* dctx, const void* src, size_t headerSize) { size_t const result = ZSTD_getFrameHeader_advanced(&(dctx->fParams), src, headerSize, dctx->format); if (ZSTD_isError(result)) return result; /* invalid header */ RETURN_ERROR_IF(result>0, srcSize_wrong, "headerSize too small"); - - /* Reference DDict requested by frame if dctx references multiple ddicts */ - if (dctx->refMultipleDDicts == ZSTD_rmd_refMultipleDDicts && dctx->ddictSet) { - ZSTD_DCtx_selectFrameDDict(dctx); - } - + + /* Reference DDict requested by frame if dctx references multiple ddicts */ + if (dctx->refMultipleDDicts == ZSTD_rmd_refMultipleDDicts && dctx->ddictSet) { + ZSTD_DCtx_selectFrameDDict(dctx); + } + #ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION /* Skip the dictID check in fuzzing mode, because it makes the search * harder. @@ -685,9 +685,9 @@ static size_t ZSTD_decodeFrameHeader(ZSTD_DCtx* dctx, const void* src, size_t he RETURN_ERROR_IF(dctx->fParams.dictID && (dctx->dictID != dctx->fParams.dictID), dictionary_wrong, ""); #endif - dctx->validateChecksum = (dctx->fParams.checksumFlag && !dctx->forceIgnoreChecksum) ? 1 : 0; - if (dctx->validateChecksum) XXH64_reset(&dctx->xxhState, 0); - dctx->processedCSize += headerSize; + dctx->validateChecksum = (dctx->fParams.checksumFlag && !dctx->forceIgnoreChecksum) ? 1 : 0; + if (dctx->validateChecksum) XXH64_reset(&dctx->xxhState, 0); + dctx->processedCSize += headerSize; return 0; } @@ -702,7 +702,7 @@ static ZSTD_frameSizeInfo ZSTD_errorFrameSizeInfo(size_t ret) static ZSTD_frameSizeInfo ZSTD_findFrameSizeInfo(const void* src, size_t srcSize) { ZSTD_frameSizeInfo frameSizeInfo; - ZSTD_memset(&frameSizeInfo, 0, sizeof(ZSTD_frameSizeInfo)); + ZSTD_memset(&frameSizeInfo, 0, sizeof(ZSTD_frameSizeInfo)); #if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT >= 1) if (ZSTD_isLegacy(src, srcSize)) @@ -757,7 +757,7 @@ static ZSTD_frameSizeInfo ZSTD_findFrameSizeInfo(const void* src, size_t srcSize ip += 4; } - frameSizeInfo.compressedSize = (size_t)(ip - ipstart); + frameSizeInfo.compressedSize = (size_t)(ip - ipstart); frameSizeInfo.decompressedBound = (zfh.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN) ? zfh.frameContentSize : nbBlocks * zfh.blockSizeMax; @@ -810,7 +810,7 @@ unsigned long long ZSTD_decompressBound(const void* src, size_t srcSize) size_t ZSTD_insertBlock(ZSTD_DCtx* dctx, const void* blockStart, size_t blockSize) { DEBUGLOG(5, "ZSTD_insertBlock: %u bytes", (unsigned)blockSize); - ZSTD_checkContinuity(dctx, blockStart, blockSize); + ZSTD_checkContinuity(dctx, blockStart, blockSize); dctx->previousDstEnd = (const char*)blockStart + blockSize; return blockSize; } @@ -820,12 +820,12 @@ static size_t ZSTD_copyRawBlock(void* dst, size_t dstCapacity, const void* src, size_t srcSize) { DEBUGLOG(5, "ZSTD_copyRawBlock"); - RETURN_ERROR_IF(srcSize > dstCapacity, dstSize_tooSmall, ""); + RETURN_ERROR_IF(srcSize > dstCapacity, dstSize_tooSmall, ""); if (dst == NULL) { if (srcSize == 0) return 0; RETURN_ERROR(dstBuffer_null, ""); } - ZSTD_memcpy(dst, src, srcSize); + ZSTD_memcpy(dst, src, srcSize); return srcSize; } @@ -833,42 +833,42 @@ static size_t ZSTD_setRleBlock(void* dst, size_t dstCapacity, BYTE b, size_t regenSize) { - RETURN_ERROR_IF(regenSize > dstCapacity, dstSize_tooSmall, ""); + RETURN_ERROR_IF(regenSize > dstCapacity, dstSize_tooSmall, ""); if (dst == NULL) { if (regenSize == 0) return 0; RETURN_ERROR(dstBuffer_null, ""); } - ZSTD_memset(dst, b, regenSize); + ZSTD_memset(dst, b, regenSize); return regenSize; } -static void ZSTD_DCtx_trace_end(ZSTD_DCtx const* dctx, U64 uncompressedSize, U64 compressedSize, unsigned streaming) -{ -#if ZSTD_TRACE - if (dctx->traceCtx && ZSTD_trace_decompress_end != NULL) { - ZSTD_Trace trace; - ZSTD_memset(&trace, 0, sizeof(trace)); - trace.version = ZSTD_VERSION_NUMBER; - trace.streaming = streaming; - if (dctx->ddict) { - trace.dictionaryID = ZSTD_getDictID_fromDDict(dctx->ddict); - trace.dictionarySize = ZSTD_DDict_dictSize(dctx->ddict); - trace.dictionaryIsCold = dctx->ddictIsCold; - } - trace.uncompressedSize = (size_t)uncompressedSize; - trace.compressedSize = (size_t)compressedSize; - trace.dctx = dctx; - ZSTD_trace_decompress_end(dctx->traceCtx, &trace); - } -#else - (void)dctx; - (void)uncompressedSize; - (void)compressedSize; - (void)streaming; -#endif -} - - +static void ZSTD_DCtx_trace_end(ZSTD_DCtx const* dctx, U64 uncompressedSize, U64 compressedSize, unsigned streaming) +{ +#if ZSTD_TRACE + if (dctx->traceCtx && ZSTD_trace_decompress_end != NULL) { + ZSTD_Trace trace; + ZSTD_memset(&trace, 0, sizeof(trace)); + trace.version = ZSTD_VERSION_NUMBER; + trace.streaming = streaming; + if (dctx->ddict) { + trace.dictionaryID = ZSTD_getDictID_fromDDict(dctx->ddict); + trace.dictionarySize = ZSTD_DDict_dictSize(dctx->ddict); + trace.dictionaryIsCold = dctx->ddictIsCold; + } + trace.uncompressedSize = (size_t)uncompressedSize; + trace.compressedSize = (size_t)compressedSize; + trace.dctx = dctx; + ZSTD_trace_decompress_end(dctx->traceCtx, &trace); + } +#else + (void)dctx; + (void)uncompressedSize; + (void)compressedSize; + (void)streaming; +#endif +} + + /*! ZSTD_decompressFrame() : * @dctx must be properly initialized * will update *srcPtr and *srcSizePtr, @@ -877,9 +877,9 @@ static size_t ZSTD_decompressFrame(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void** srcPtr, size_t *srcSizePtr) { - const BYTE* const istart = (const BYTE*)(*srcPtr); - const BYTE* ip = istart; - BYTE* const ostart = (BYTE*)dst; + const BYTE* const istart = (const BYTE*)(*srcPtr); + const BYTE* ip = istart; + BYTE* const ostart = (BYTE*)dst; BYTE* const oend = dstCapacity != 0 ? ostart + dstCapacity : ostart; BYTE* op = ostart; size_t remainingSrcSize = *srcSizePtr; @@ -918,10 +918,10 @@ static size_t ZSTD_decompressFrame(ZSTD_DCtx* dctx, decodedSize = ZSTD_decompressBlock_internal(dctx, op, (size_t)(oend-op), ip, cBlockSize, /* frame */ 1, not_streaming); break; case bt_raw : - decodedSize = ZSTD_copyRawBlock(op, (size_t)(oend-op), ip, cBlockSize); + decodedSize = ZSTD_copyRawBlock(op, (size_t)(oend-op), ip, cBlockSize); break; case bt_rle : - decodedSize = ZSTD_setRleBlock(op, (size_t)(oend-op), *ip, blockProperties.origSize); + decodedSize = ZSTD_setRleBlock(op, (size_t)(oend-op), *ip, blockProperties.origSize); break; case bt_reserved : default: @@ -929,7 +929,7 @@ static size_t ZSTD_decompressFrame(ZSTD_DCtx* dctx, } if (ZSTD_isError(decodedSize)) return decodedSize; - if (dctx->validateChecksum) + if (dctx->validateChecksum) XXH64_update(&dctx->xxhState, op, decodedSize); if (decodedSize != 0) op += decodedSize; @@ -945,20 +945,20 @@ static size_t ZSTD_decompressFrame(ZSTD_DCtx* dctx, } if (dctx->fParams.checksumFlag) { /* Frame content checksum verification */ RETURN_ERROR_IF(remainingSrcSize<4, checksum_wrong, ""); - if (!dctx->forceIgnoreChecksum) { - U32 const checkCalc = (U32)XXH64_digest(&dctx->xxhState); - U32 checkRead; - checkRead = MEM_readLE32(ip); - RETURN_ERROR_IF(checkRead != checkCalc, checksum_wrong, ""); - } + if (!dctx->forceIgnoreChecksum) { + U32 const checkCalc = (U32)XXH64_digest(&dctx->xxhState); + U32 checkRead; + checkRead = MEM_readLE32(ip); + RETURN_ERROR_IF(checkRead != checkCalc, checksum_wrong, ""); + } ip += 4; remainingSrcSize -= 4; } - ZSTD_DCtx_trace_end(dctx, (U64)(op-ostart), (U64)(ip-istart), /* streaming */ 0); + ZSTD_DCtx_trace_end(dctx, (U64)(op-ostart), (U64)(ip-istart), /* streaming */ 0); /* Allow caller to get size read */ *srcPtr = ip; *srcSizePtr = remainingSrcSize; - return (size_t)(op-ostart); + return (size_t)(op-ostart); } static size_t ZSTD_decompressMultiFrame(ZSTD_DCtx* dctx, @@ -991,7 +991,7 @@ static size_t ZSTD_decompressMultiFrame(ZSTD_DCtx* dctx, decodedSize = ZSTD_decompressLegacy(dst, dstCapacity, src, frameSize, dict, dictSize); if (ZSTD_isError(decodedSize)) return decodedSize; - assert(decodedSize <= dstCapacity); + assert(decodedSize <= dstCapacity); dst = (BYTE*)dst + decodedSize; dstCapacity -= decodedSize; @@ -1023,7 +1023,7 @@ static size_t ZSTD_decompressMultiFrame(ZSTD_DCtx* dctx, * use this in all cases but ddict */ FORWARD_IF_ERROR(ZSTD_decompressBegin_usingDict(dctx, dict, dictSize), ""); } - ZSTD_checkContinuity(dctx, dst, dstCapacity); + ZSTD_checkContinuity(dctx, dst, dstCapacity); { const size_t res = ZSTD_decompressFrame(dctx, dst, dstCapacity, &src, &srcSize); @@ -1031,13 +1031,13 @@ static size_t ZSTD_decompressMultiFrame(ZSTD_DCtx* dctx, (ZSTD_getErrorCode(res) == ZSTD_error_prefix_unknown) && (moreThan1Frame==1), srcSize_wrong, - "At least one frame successfully completed, " - "but following bytes are garbage: " - "it's more likely to be a srcSize error, " - "specifying more input bytes than size of frame(s). " - "Note: one could be unlucky, it might be a corruption error instead, " - "happening right at the place where we expect zstd magic bytes. " - "But this is _much_ less likely than a srcSize field error."); + "At least one frame successfully completed, " + "but following bytes are garbage: " + "it's more likely to be a srcSize error, " + "specifying more input bytes than size of frame(s). " + "Note: one could be unlucky, it might be a corruption error instead, " + "happening right at the place where we expect zstd magic bytes. " + "But this is _much_ less likely than a srcSize field error."); if (ZSTD_isError(res)) return res; assert(res <= dstCapacity); if (res != 0) @@ -1049,7 +1049,7 @@ static size_t ZSTD_decompressMultiFrame(ZSTD_DCtx* dctx, RETURN_ERROR_IF(srcSize, srcSize_wrong, "input not entirely consumed"); - return (size_t)((BYTE*)dst - (BYTE*)dststart); + return (size_t)((BYTE*)dst - (BYTE*)dststart); } size_t ZSTD_decompress_usingDict(ZSTD_DCtx* dctx, @@ -1161,10 +1161,10 @@ size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, c DEBUGLOG(5, "ZSTD_decompressContinue (srcSize:%u)", (unsigned)srcSize); /* Sanity check */ RETURN_ERROR_IF(srcSize != ZSTD_nextSrcSizeToDecompressWithInputSize(dctx, srcSize), srcSize_wrong, "not allowed"); - ZSTD_checkContinuity(dctx, dst, dstCapacity); - - dctx->processedCSize += srcSize; + ZSTD_checkContinuity(dctx, dst, dstCapacity); + dctx->processedCSize += srcSize; + switch (dctx->stage) { case ZSTDds_getFrameHeaderSize : @@ -1172,21 +1172,21 @@ size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, c if (dctx->format == ZSTD_f_zstd1) { /* allows header */ assert(srcSize >= ZSTD_FRAMEIDSIZE); /* to read skippable magic number */ if ((MEM_readLE32(src) & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { /* skippable frame */ - ZSTD_memcpy(dctx->headerBuffer, src, srcSize); + ZSTD_memcpy(dctx->headerBuffer, src, srcSize); dctx->expected = ZSTD_SKIPPABLEHEADERSIZE - srcSize; /* remaining to load to get full skippable frame header */ dctx->stage = ZSTDds_decodeSkippableHeader; return 0; } } dctx->headerSize = ZSTD_frameHeaderSize_internal(src, srcSize, dctx->format); if (ZSTD_isError(dctx->headerSize)) return dctx->headerSize; - ZSTD_memcpy(dctx->headerBuffer, src, srcSize); + ZSTD_memcpy(dctx->headerBuffer, src, srcSize); dctx->expected = dctx->headerSize - srcSize; dctx->stage = ZSTDds_decodeFrameHeader; return 0; case ZSTDds_decodeFrameHeader: assert(src != NULL); - ZSTD_memcpy(dctx->headerBuffer + (dctx->headerSize - srcSize), src, srcSize); + ZSTD_memcpy(dctx->headerBuffer + (dctx->headerSize - srcSize), src, srcSize); FORWARD_IF_ERROR(ZSTD_decodeFrameHeader(dctx, dctx->headerBuffer, dctx->headerSize), ""); dctx->expected = ZSTD_blockHeaderSize; dctx->stage = ZSTDds_decodeBlockHeader; @@ -1250,7 +1250,7 @@ size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, c RETURN_ERROR_IF(rSize > dctx->fParams.blockSizeMax, corruption_detected, "Decompressed Block Size Exceeds Maximum"); DEBUGLOG(5, "ZSTD_decompressContinue: decoded size from block : %u", (unsigned)rSize); dctx->decodedSize += rSize; - if (dctx->validateChecksum) XXH64_update(&dctx->xxhState, dst, rSize); + if (dctx->validateChecksum) XXH64_update(&dctx->xxhState, dst, rSize); dctx->previousDstEnd = (char*)dst + rSize; /* Stay on the same stage until we are finished streaming the block. */ @@ -1268,7 +1268,7 @@ size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, c dctx->expected = 4; dctx->stage = ZSTDds_checkChecksum; } else { - ZSTD_DCtx_trace_end(dctx, dctx->decodedSize, dctx->processedCSize, /* streaming */ 1); + ZSTD_DCtx_trace_end(dctx, dctx->decodedSize, dctx->processedCSize, /* streaming */ 1); dctx->expected = 0; /* ends here */ dctx->stage = ZSTDds_getFrameHeaderSize; } @@ -1281,14 +1281,14 @@ size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, c case ZSTDds_checkChecksum: assert(srcSize == 4); /* guaranteed by dctx->expected */ - { - if (dctx->validateChecksum) { - U32 const h32 = (U32)XXH64_digest(&dctx->xxhState); - U32 const check32 = MEM_readLE32(src); - DEBUGLOG(4, "ZSTD_decompressContinue: checksum : calculated %08X :: %08X read", (unsigned)h32, (unsigned)check32); - RETURN_ERROR_IF(check32 != h32, checksum_wrong, ""); - } - ZSTD_DCtx_trace_end(dctx, dctx->decodedSize, dctx->processedCSize, /* streaming */ 1); + { + if (dctx->validateChecksum) { + U32 const h32 = (U32)XXH64_digest(&dctx->xxhState); + U32 const check32 = MEM_readLE32(src); + DEBUGLOG(4, "ZSTD_decompressContinue: checksum : calculated %08X :: %08X read", (unsigned)h32, (unsigned)check32); + RETURN_ERROR_IF(check32 != h32, checksum_wrong, ""); + } + ZSTD_DCtx_trace_end(dctx, dctx->decodedSize, dctx->processedCSize, /* streaming */ 1); dctx->expected = 0; dctx->stage = ZSTDds_getFrameHeaderSize; return 0; @@ -1297,7 +1297,7 @@ size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, c case ZSTDds_decodeSkippableHeader: assert(src != NULL); assert(srcSize <= ZSTD_SKIPPABLEHEADERSIZE); - ZSTD_memcpy(dctx->headerBuffer + (ZSTD_SKIPPABLEHEADERSIZE - srcSize), src, srcSize); /* complete skippable header */ + ZSTD_memcpy(dctx->headerBuffer + (ZSTD_SKIPPABLEHEADERSIZE - srcSize), src, srcSize); /* complete skippable header */ dctx->expected = MEM_readLE32(dctx->headerBuffer + ZSTD_FRAMEIDSIZE); /* note : dctx->expected can grow seriously large, beyond local buffer size */ dctx->stage = ZSTDds_skipFrame; return 0; @@ -1353,7 +1353,7 @@ ZSTD_loadDEntropy(ZSTD_entropyDTables_t* entropy, workspace, workspaceSize); #else size_t const hSize = HUF_readDTableX2_wksp(entropy->hufTable, - dictPtr, (size_t)(dictEnd - dictPtr), + dictPtr, (size_t)(dictEnd - dictPtr), workspace, workspaceSize); #endif RETURN_ERROR_IF(HUF_isError(hSize), dictionary_corrupted, ""); @@ -1362,46 +1362,46 @@ ZSTD_loadDEntropy(ZSTD_entropyDTables_t* entropy, { short offcodeNCount[MaxOff+1]; unsigned offcodeMaxValue = MaxOff, offcodeLog; - size_t const offcodeHeaderSize = FSE_readNCount(offcodeNCount, &offcodeMaxValue, &offcodeLog, dictPtr, (size_t)(dictEnd-dictPtr)); + size_t const offcodeHeaderSize = FSE_readNCount(offcodeNCount, &offcodeMaxValue, &offcodeLog, dictPtr, (size_t)(dictEnd-dictPtr)); RETURN_ERROR_IF(FSE_isError(offcodeHeaderSize), dictionary_corrupted, ""); RETURN_ERROR_IF(offcodeMaxValue > MaxOff, dictionary_corrupted, ""); RETURN_ERROR_IF(offcodeLog > OffFSELog, dictionary_corrupted, ""); ZSTD_buildFSETable( entropy->OFTable, offcodeNCount, offcodeMaxValue, OF_base, OF_bits, - offcodeLog, - entropy->workspace, sizeof(entropy->workspace), - /* bmi2 */0); + offcodeLog, + entropy->workspace, sizeof(entropy->workspace), + /* bmi2 */0); dictPtr += offcodeHeaderSize; } { short matchlengthNCount[MaxML+1]; unsigned matchlengthMaxValue = MaxML, matchlengthLog; - size_t const matchlengthHeaderSize = FSE_readNCount(matchlengthNCount, &matchlengthMaxValue, &matchlengthLog, dictPtr, (size_t)(dictEnd-dictPtr)); + size_t const matchlengthHeaderSize = FSE_readNCount(matchlengthNCount, &matchlengthMaxValue, &matchlengthLog, dictPtr, (size_t)(dictEnd-dictPtr)); RETURN_ERROR_IF(FSE_isError(matchlengthHeaderSize), dictionary_corrupted, ""); RETURN_ERROR_IF(matchlengthMaxValue > MaxML, dictionary_corrupted, ""); RETURN_ERROR_IF(matchlengthLog > MLFSELog, dictionary_corrupted, ""); ZSTD_buildFSETable( entropy->MLTable, matchlengthNCount, matchlengthMaxValue, ML_base, ML_bits, - matchlengthLog, - entropy->workspace, sizeof(entropy->workspace), - /* bmi2 */ 0); + matchlengthLog, + entropy->workspace, sizeof(entropy->workspace), + /* bmi2 */ 0); dictPtr += matchlengthHeaderSize; } { short litlengthNCount[MaxLL+1]; unsigned litlengthMaxValue = MaxLL, litlengthLog; - size_t const litlengthHeaderSize = FSE_readNCount(litlengthNCount, &litlengthMaxValue, &litlengthLog, dictPtr, (size_t)(dictEnd-dictPtr)); + size_t const litlengthHeaderSize = FSE_readNCount(litlengthNCount, &litlengthMaxValue, &litlengthLog, dictPtr, (size_t)(dictEnd-dictPtr)); RETURN_ERROR_IF(FSE_isError(litlengthHeaderSize), dictionary_corrupted, ""); RETURN_ERROR_IF(litlengthMaxValue > MaxLL, dictionary_corrupted, ""); RETURN_ERROR_IF(litlengthLog > LLFSELog, dictionary_corrupted, ""); ZSTD_buildFSETable( entropy->LLTable, litlengthNCount, litlengthMaxValue, LL_base, LL_bits, - litlengthLog, - entropy->workspace, sizeof(entropy->workspace), - /* bmi2 */ 0); + litlengthLog, + entropy->workspace, sizeof(entropy->workspace), + /* bmi2 */ 0); dictPtr += litlengthHeaderSize; } @@ -1415,7 +1415,7 @@ ZSTD_loadDEntropy(ZSTD_entropyDTables_t* entropy, entropy->rep[i] = rep; } } - return (size_t)(dictPtr - (const BYTE*)dict); + return (size_t)(dictPtr - (const BYTE*)dict); } static size_t ZSTD_decompress_insertDictionary(ZSTD_DCtx* dctx, const void* dict, size_t dictSize) @@ -1442,12 +1442,12 @@ static size_t ZSTD_decompress_insertDictionary(ZSTD_DCtx* dctx, const void* dict size_t ZSTD_decompressBegin(ZSTD_DCtx* dctx) { assert(dctx != NULL); -#if ZSTD_TRACE - dctx->traceCtx = (ZSTD_trace_decompress_begin != NULL) ? ZSTD_trace_decompress_begin(dctx) : 0; -#endif +#if ZSTD_TRACE + dctx->traceCtx = (ZSTD_trace_decompress_begin != NULL) ? ZSTD_trace_decompress_begin(dctx) : 0; +#endif dctx->expected = ZSTD_startingInputLength(dctx->format); /* dctx->format must be properly set */ dctx->stage = ZSTDds_getFrameHeaderSize; - dctx->processedCSize = 0; + dctx->processedCSize = 0; dctx->decodedSize = 0; dctx->previousDstEnd = NULL; dctx->prefixStart = NULL; @@ -1458,7 +1458,7 @@ size_t ZSTD_decompressBegin(ZSTD_DCtx* dctx) dctx->dictID = 0; dctx->bType = bt_reserved; ZSTD_STATIC_ASSERT(sizeof(dctx->entropy.rep) == sizeof(repStartValue)); - ZSTD_memcpy(dctx->entropy.rep, repStartValue, sizeof(repStartValue)); /* initial repcodes */ + ZSTD_memcpy(dctx->entropy.rep, repStartValue, sizeof(repStartValue)); /* initial repcodes */ dctx->LLTptr = dctx->entropy.LLTable; dctx->MLTptr = dctx->entropy.MLTable; dctx->OFTptr = dctx->entropy.OFTable; @@ -1661,16 +1661,16 @@ size_t ZSTD_DCtx_refDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict) if (ddict) { dctx->ddict = ddict; dctx->dictUses = ZSTD_use_indefinitely; - if (dctx->refMultipleDDicts == ZSTD_rmd_refMultipleDDicts) { - if (dctx->ddictSet == NULL) { - dctx->ddictSet = ZSTD_createDDictHashSet(dctx->customMem); - if (!dctx->ddictSet) { - RETURN_ERROR(memory_allocation, "Failed to allocate memory for hash set!"); - } - } - assert(!dctx->staticSize); /* Impossible: ddictSet cannot have been allocated if static dctx */ - FORWARD_IF_ERROR(ZSTD_DDictHashSet_addDDict(dctx->ddictSet, ddict, dctx->customMem), ""); - } + if (dctx->refMultipleDDicts == ZSTD_rmd_refMultipleDDicts) { + if (dctx->ddictSet == NULL) { + dctx->ddictSet = ZSTD_createDDictHashSet(dctx->customMem); + if (!dctx->ddictSet) { + RETURN_ERROR(memory_allocation, "Failed to allocate memory for hash set!"); + } + } + assert(!dctx->staticSize); /* Impossible: ddictSet cannot have been allocated if static dctx */ + FORWARD_IF_ERROR(ZSTD_DDictHashSet_addDDict(dctx->ddictSet, ddict, dctx->customMem), ""); + } } return 0; } @@ -1692,7 +1692,7 @@ size_t ZSTD_DCtx_setMaxWindowSize(ZSTD_DCtx* dctx, size_t maxWindowSize) size_t ZSTD_DCtx_setFormat(ZSTD_DCtx* dctx, ZSTD_format_e format) { - return ZSTD_DCtx_setParameter(dctx, ZSTD_d_format, (int)format); + return ZSTD_DCtx_setParameter(dctx, ZSTD_d_format, (int)format); } ZSTD_bounds ZSTD_dParam_getBounds(ZSTD_dParameter dParam) @@ -1709,17 +1709,17 @@ ZSTD_bounds ZSTD_dParam_getBounds(ZSTD_dParameter dParam) ZSTD_STATIC_ASSERT(ZSTD_f_zstd1 < ZSTD_f_zstd1_magicless); return bounds; case ZSTD_d_stableOutBuffer: - bounds.lowerBound = (int)ZSTD_bm_buffered; - bounds.upperBound = (int)ZSTD_bm_stable; - return bounds; - case ZSTD_d_forceIgnoreChecksum: - bounds.lowerBound = (int)ZSTD_d_validateChecksum; - bounds.upperBound = (int)ZSTD_d_ignoreChecksum; - return bounds; - case ZSTD_d_refMultipleDDicts: - bounds.lowerBound = (int)ZSTD_rmd_refSingleDDict; - bounds.upperBound = (int)ZSTD_rmd_refMultipleDDicts; + bounds.lowerBound = (int)ZSTD_bm_buffered; + bounds.upperBound = (int)ZSTD_bm_stable; return bounds; + case ZSTD_d_forceIgnoreChecksum: + bounds.lowerBound = (int)ZSTD_d_validateChecksum; + bounds.upperBound = (int)ZSTD_d_ignoreChecksum; + return bounds; + case ZSTD_d_refMultipleDDicts: + bounds.lowerBound = (int)ZSTD_rmd_refSingleDDict; + bounds.upperBound = (int)ZSTD_rmd_refMultipleDDicts; + return bounds; default:; } bounds.error = ERROR(parameter_unsupported); @@ -1742,29 +1742,29 @@ static int ZSTD_dParam_withinBounds(ZSTD_dParameter dParam, int value) RETURN_ERROR_IF(!ZSTD_dParam_withinBounds(p, v), parameter_outOfBound, ""); \ } -size_t ZSTD_DCtx_getParameter(ZSTD_DCtx* dctx, ZSTD_dParameter param, int* value) -{ - switch (param) { - case ZSTD_d_windowLogMax: - *value = (int)ZSTD_highbit32((U32)dctx->maxWindowSize); - return 0; - case ZSTD_d_format: - *value = (int)dctx->format; - return 0; - case ZSTD_d_stableOutBuffer: - *value = (int)dctx->outBufferMode; - return 0; - case ZSTD_d_forceIgnoreChecksum: - *value = (int)dctx->forceIgnoreChecksum; - return 0; - case ZSTD_d_refMultipleDDicts: - *value = (int)dctx->refMultipleDDicts; - return 0; - default:; - } - RETURN_ERROR(parameter_unsupported, ""); -} - +size_t ZSTD_DCtx_getParameter(ZSTD_DCtx* dctx, ZSTD_dParameter param, int* value) +{ + switch (param) { + case ZSTD_d_windowLogMax: + *value = (int)ZSTD_highbit32((U32)dctx->maxWindowSize); + return 0; + case ZSTD_d_format: + *value = (int)dctx->format; + return 0; + case ZSTD_d_stableOutBuffer: + *value = (int)dctx->outBufferMode; + return 0; + case ZSTD_d_forceIgnoreChecksum: + *value = (int)dctx->forceIgnoreChecksum; + return 0; + case ZSTD_d_refMultipleDDicts: + *value = (int)dctx->refMultipleDDicts; + return 0; + default:; + } + RETURN_ERROR(parameter_unsupported, ""); +} + size_t ZSTD_DCtx_setParameter(ZSTD_DCtx* dctx, ZSTD_dParameter dParam, int value) { RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong, ""); @@ -1780,19 +1780,19 @@ size_t ZSTD_DCtx_setParameter(ZSTD_DCtx* dctx, ZSTD_dParameter dParam, int value return 0; case ZSTD_d_stableOutBuffer: CHECK_DBOUNDS(ZSTD_d_stableOutBuffer, value); - dctx->outBufferMode = (ZSTD_bufferMode_e)value; - return 0; - case ZSTD_d_forceIgnoreChecksum: - CHECK_DBOUNDS(ZSTD_d_forceIgnoreChecksum, value); - dctx->forceIgnoreChecksum = (ZSTD_forceIgnoreChecksum_e)value; - return 0; - case ZSTD_d_refMultipleDDicts: - CHECK_DBOUNDS(ZSTD_d_refMultipleDDicts, value); - if (dctx->staticSize != 0) { - RETURN_ERROR(parameter_unsupported, "Static dctx does not support multiple DDicts!"); - } - dctx->refMultipleDDicts = (ZSTD_refMultipleDDicts_e)value; + dctx->outBufferMode = (ZSTD_bufferMode_e)value; return 0; + case ZSTD_d_forceIgnoreChecksum: + CHECK_DBOUNDS(ZSTD_d_forceIgnoreChecksum, value); + dctx->forceIgnoreChecksum = (ZSTD_forceIgnoreChecksum_e)value; + return 0; + case ZSTD_d_refMultipleDDicts: + CHECK_DBOUNDS(ZSTD_d_refMultipleDDicts, value); + if (dctx->staticSize != 0) { + RETURN_ERROR(parameter_unsupported, "Static dctx does not support multiple DDicts!"); + } + dctx->refMultipleDDicts = (ZSTD_refMultipleDDicts_e)value; + return 0; default:; } RETURN_ERROR(parameter_unsupported, ""); @@ -1809,7 +1809,7 @@ size_t ZSTD_DCtx_reset(ZSTD_DCtx* dctx, ZSTD_ResetDirective reset) || (reset == ZSTD_reset_session_and_parameters) ) { RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong, ""); ZSTD_clearDict(dctx); - ZSTD_DCtx_resetParameters(dctx); + ZSTD_DCtx_resetParameters(dctx); } return 0; } @@ -1864,7 +1864,7 @@ static void ZSTD_DCtx_updateOversizedDuration(ZSTD_DStream* zds, size_t const ne { if (ZSTD_DCtx_isOverflow(zds, neededInBuffSize, neededOutBuffSize)) zds->oversizedDuration++; - else + else zds->oversizedDuration = 0; } @@ -1878,7 +1878,7 @@ static size_t ZSTD_checkOutBuffer(ZSTD_DStream const* zds, ZSTD_outBuffer const* { ZSTD_outBuffer const expect = zds->expectedOutBuffer; /* No requirement when ZSTD_obm_stable is not enabled. */ - if (zds->outBufferMode != ZSTD_bm_stable) + if (zds->outBufferMode != ZSTD_bm_stable) return 0; /* Any buffer is allowed in zdss_init, this must be the same for every other call until * the context is reset. @@ -1888,7 +1888,7 @@ static size_t ZSTD_checkOutBuffer(ZSTD_DStream const* zds, ZSTD_outBuffer const* /* The buffer must match our expectation exactly. */ if (expect.dst == output->dst && expect.pos == output->pos && expect.size == output->size) return 0; - RETURN_ERROR(dstBuffer_wrong, "ZSTD_d_stableOutBuffer enabled but output differs!"); + RETURN_ERROR(dstBuffer_wrong, "ZSTD_d_stableOutBuffer enabled but output differs!"); } /* Calls ZSTD_decompressContinue() with the right parameters for ZSTD_decompressStream() @@ -1900,7 +1900,7 @@ static size_t ZSTD_decompressContinueStream( ZSTD_DStream* zds, char** op, char* oend, void const* src, size_t srcSize) { int const isSkipFrame = ZSTD_isSkipFrame(zds); - if (zds->outBufferMode == ZSTD_bm_buffered) { + if (zds->outBufferMode == ZSTD_bm_buffered) { size_t const dstSize = isSkipFrame ? 0 : zds->outBuffSize - zds->outStart; size_t const decodedSize = ZSTD_decompressContinue(zds, zds->outBuff + zds->outStart, dstSize, src, srcSize); @@ -1913,14 +1913,14 @@ static size_t ZSTD_decompressContinueStream( } } else { /* Write directly into the output buffer */ - size_t const dstSize = isSkipFrame ? 0 : (size_t)(oend - *op); + size_t const dstSize = isSkipFrame ? 0 : (size_t)(oend - *op); size_t const decodedSize = ZSTD_decompressContinue(zds, *op, dstSize, src, srcSize); FORWARD_IF_ERROR(decodedSize, ""); *op += decodedSize; /* Flushing is not needed. */ zds->streamStage = zdss_read; assert(*op <= oend); - assert(zds->outBufferMode == ZSTD_bm_stable); + assert(zds->outBufferMode == ZSTD_bm_stable); } return 0; } @@ -1977,9 +1977,9 @@ size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inB } } #endif { size_t const hSize = ZSTD_getFrameHeader_advanced(&zds->fParams, zds->headerBuffer, zds->lhSize, zds->format); - if (zds->refMultipleDDicts && zds->ddictSet) { - ZSTD_DCtx_selectFrameDDict(zds); - } + if (zds->refMultipleDDicts && zds->ddictSet) { + ZSTD_DCtx_selectFrameDDict(zds); + } DEBUGLOG(5, "header size : %u", (U32)hSize); if (ZSTD_isError(hSize)) { #if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>=1) @@ -2008,14 +2008,14 @@ size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inB assert(iend >= ip); if (toLoad > remainingInput) { /* not enough input to load full header */ if (remainingInput > 0) { - ZSTD_memcpy(zds->headerBuffer + zds->lhSize, ip, remainingInput); + ZSTD_memcpy(zds->headerBuffer + zds->lhSize, ip, remainingInput); zds->lhSize += remainingInput; } input->pos = input->size; return (MAX((size_t)ZSTD_FRAMEHEADERSIZE_MIN(zds->format), hSize) - zds->lhSize) + ZSTD_blockHeaderSize; /* remaining header bytes + next block header */ } assert(ip != NULL); - ZSTD_memcpy(zds->headerBuffer + zds->lhSize, ip, toLoad); zds->lhSize = hSize; ip += toLoad; + ZSTD_memcpy(zds->headerBuffer + zds->lhSize, ip, toLoad); zds->lhSize = hSize; ip += toLoad; break; } } @@ -2023,10 +2023,10 @@ size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inB if (zds->fParams.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN && zds->fParams.frameType != ZSTD_skippableFrame && (U64)(size_t)(oend-op) >= zds->fParams.frameContentSize) { - size_t const cSize = ZSTD_findFrameCompressedSize(istart, (size_t)(iend-istart)); + size_t const cSize = ZSTD_findFrameCompressedSize(istart, (size_t)(iend-istart)); if (cSize <= (size_t)(iend-istart)) { /* shortcut : using single-pass mode */ - size_t const decompressedSize = ZSTD_decompress_usingDDict(zds, op, (size_t)(oend-op), istart, cSize, ZSTD_getDDict(zds)); + size_t const decompressedSize = ZSTD_decompress_usingDDict(zds, op, (size_t)(oend-op), istart, cSize, ZSTD_getDDict(zds)); if (ZSTD_isError(decompressedSize)) return decompressedSize; DEBUGLOG(4, "shortcut to single-pass ZSTD_decompress_usingDDict()") ip = istart + cSize; @@ -2038,7 +2038,7 @@ size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inB } } /* Check output buffer is large enough for ZSTD_odm_stable. */ - if (zds->outBufferMode == ZSTD_bm_stable + if (zds->outBufferMode == ZSTD_bm_stable && zds->fParams.frameType != ZSTD_skippableFrame && zds->fParams.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN && (U64)(size_t)(oend-op) < zds->fParams.frameContentSize) { @@ -2068,7 +2068,7 @@ size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inB /* Adapt buffer sizes to frame header instructions */ { size_t const neededInBuffSize = MAX(zds->fParams.blockSizeMax, 4 /* frame checksum */); - size_t const neededOutBuffSize = zds->outBufferMode == ZSTD_bm_buffered + size_t const neededOutBuffSize = zds->outBufferMode == ZSTD_bm_buffered ? ZSTD_decodingBufferSize_min(zds->fParams.windowSize, zds->fParams.frameContentSize) : 0; @@ -2076,7 +2076,7 @@ size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inB { int const tooSmall = (zds->inBuffSize < neededInBuffSize) || (zds->outBuffSize < neededOutBuffSize); int const tooLarge = ZSTD_DCtx_isOversizedTooLong(zds); - + if (tooSmall || tooLarge) { size_t const bufferSize = neededInBuffSize + neededOutBuffSize; DEBUGLOG(4, "inBuff : from %u to %u", @@ -2090,10 +2090,10 @@ size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inB bufferSize > zds->staticSize - sizeof(ZSTD_DCtx), memory_allocation, ""); } else { - ZSTD_customFree(zds->inBuff, zds->customMem); + ZSTD_customFree(zds->inBuff, zds->customMem); zds->inBuffSize = 0; zds->outBuffSize = 0; - zds->inBuff = (char*)ZSTD_customMalloc(bufferSize, zds->customMem); + zds->inBuff = (char*)ZSTD_customMalloc(bufferSize, zds->customMem); RETURN_ERROR_IF(zds->inBuff == NULL, memory_allocation, ""); } zds->inBuffSize = neededInBuffSize; @@ -2105,7 +2105,7 @@ size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inB case zdss_read: DEBUGLOG(5, "stage zdss_read"); - { size_t const neededInSize = ZSTD_nextSrcSizeToDecompressWithInputSize(zds, (size_t)(iend - ip)); + { size_t const neededInSize = ZSTD_nextSrcSizeToDecompressWithInputSize(zds, (size_t)(iend - ip)); DEBUGLOG(5, "neededInSize = %u", (U32)neededInSize); if (neededInSize==0) { /* end of frame */ zds->streamStage = zdss_init; @@ -2135,7 +2135,7 @@ size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inB RETURN_ERROR_IF(toLoad > zds->inBuffSize - zds->inPos, corruption_detected, "should never happen"); - loadedSize = ZSTD_limitCopy(zds->inBuff + zds->inPos, toLoad, ip, (size_t)(iend-ip)); + loadedSize = ZSTD_limitCopy(zds->inBuff + zds->inPos, toLoad, ip, (size_t)(iend-ip)); } ip += loadedSize; zds->inPos += loadedSize; @@ -2149,7 +2149,7 @@ size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inB } case zdss_flush: { size_t const toFlushSize = zds->outEnd - zds->outStart; - size_t const flushedSize = ZSTD_limitCopy(op, (size_t)(oend-op), zds->outBuff + zds->outStart, toFlushSize); + size_t const flushedSize = ZSTD_limitCopy(op, (size_t)(oend-op), zds->outBuff + zds->outStart, toFlushSize); op += flushedSize; zds->outStart += flushedSize; if (flushedSize == toFlushSize) { /* flush completed */ diff --git a/contrib/libs/zstd/lib/decompress/zstd_decompress_block.c b/contrib/libs/zstd/lib/decompress/zstd_decompress_block.c index 2e44d30d2f..96e24c142b 100644 --- a/contrib/libs/zstd/lib/decompress/zstd_decompress_block.c +++ b/contrib/libs/zstd/lib/decompress/zstd_decompress_block.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -14,7 +14,7 @@ /*-******************************************************* * Dependencies *********************************************************/ -#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memmove, ZSTD_memset */ +#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memmove, ZSTD_memset */ #include "../common/compiler.h" /* prefetch */ #include "../common/cpu.h" /* bmi2 */ #include "../common/mem.h" /* low level memory routines */ @@ -44,7 +44,7 @@ /*_******************************************************* * Memory operations **********************************************************/ -static void ZSTD_copy4(void* dst, const void* src) { ZSTD_memcpy(dst, src, 4); } +static void ZSTD_copy4(void* dst, const void* src) { ZSTD_memcpy(dst, src, 4); } /*-************************************************************* @@ -312,7 +312,7 @@ size_t ZSTD_decodeLiteralsBlock(ZSTD_DCtx* dctx, /* Default FSE distribution tables. * These are pre-calculated FSE decoding tables using default distributions as defined in specification : - * https://github.com/facebook/zstd/blob/release/doc/zstd_compression_format.md#default-distributions + * https://github.com/facebook/zstd/blob/release/doc/zstd_compression_format.md#default-distributions * They were generated programmatically with following method : * - start from default distributions, present in /lib/common/zstd_internal.h * - generate tables normally, using ZSTD_buildFSETable() @@ -440,26 +440,26 @@ static void ZSTD_buildSeqTable_rle(ZSTD_seqSymbol* dt, U32 baseValue, U8 nbAddBi * generate FSE decoding table for one symbol (ll, ml or off) * cannot fail if input is valid => * all inputs are presumed validated at this stage */ -FORCE_INLINE_TEMPLATE -void ZSTD_buildFSETable_body(ZSTD_seqSymbol* dt, +FORCE_INLINE_TEMPLATE +void ZSTD_buildFSETable_body(ZSTD_seqSymbol* dt, const short* normalizedCounter, unsigned maxSymbolValue, const U32* baseValue, const U8* nbAdditionalBits, - unsigned tableLog, void* wksp, size_t wkspSize) + unsigned tableLog, void* wksp, size_t wkspSize) { ZSTD_seqSymbol* const tableDecode = dt+1; U32 const maxSV1 = maxSymbolValue + 1; U32 const tableSize = 1 << tableLog; - U16* symbolNext = (U16*)wksp; - BYTE* spread = (BYTE*)(symbolNext + MaxSeq + 1); - U32 highThreshold = tableSize - 1; - - + U16* symbolNext = (U16*)wksp; + BYTE* spread = (BYTE*)(symbolNext + MaxSeq + 1); + U32 highThreshold = tableSize - 1; + + /* Sanity Checks */ assert(maxSymbolValue <= MaxSeq); assert(tableLog <= MaxFSELog); - assert(wkspSize >= ZSTD_BUILD_FSE_TABLE_WKSP_SIZE); - (void)wkspSize; + assert(wkspSize >= ZSTD_BUILD_FSE_TABLE_WKSP_SIZE); + (void)wkspSize; /* Init, lay down lowprob symbols */ { ZSTD_seqSymbol_header DTableH; DTableH.tableLog = tableLog; @@ -475,69 +475,69 @@ void ZSTD_buildFSETable_body(ZSTD_seqSymbol* dt, assert(normalizedCounter[s]>=0); symbolNext[s] = (U16)normalizedCounter[s]; } } } - ZSTD_memcpy(dt, &DTableH, sizeof(DTableH)); + ZSTD_memcpy(dt, &DTableH, sizeof(DTableH)); } /* Spread symbols */ - assert(tableSize <= 512); - /* Specialized symbol spreading for the case when there are - * no low probability (-1 count) symbols. When compressing - * small blocks we avoid low probability symbols to hit this - * case, since header decoding speed matters more. - */ - if (highThreshold == tableSize - 1) { - size_t const tableMask = tableSize-1; - size_t const step = FSE_TABLESTEP(tableSize); - /* First lay down the symbols in order. - * We use a uint64_t to lay down 8 bytes at a time. This reduces branch - * misses since small blocks generally have small table logs, so nearly - * all symbols have counts <= 8. We ensure we have 8 bytes at the end of - * our buffer to handle the over-write. - */ - { - U64 const add = 0x0101010101010101ull; - size_t pos = 0; - U64 sv = 0; - U32 s; - for (s=0; s<maxSV1; ++s, sv += add) { - int i; - int const n = normalizedCounter[s]; - MEM_write64(spread + pos, sv); - for (i = 8; i < n; i += 8) { - MEM_write64(spread + pos + i, sv); - } - pos += n; - } - } - /* Now we spread those positions across the table. - * The benefit of doing it in two stages is that we avoid the the - * variable size inner loop, which caused lots of branch misses. - * Now we can run through all the positions without any branch misses. - * We unroll the loop twice, since that is what emperically worked best. - */ - { - size_t position = 0; - size_t s; - size_t const unroll = 2; - assert(tableSize % unroll == 0); /* FSE_MIN_TABLELOG is 5 */ - for (s = 0; s < (size_t)tableSize; s += unroll) { - size_t u; - for (u = 0; u < unroll; ++u) { - size_t const uPosition = (position + (u * step)) & tableMask; - tableDecode[uPosition].baseValue = spread[s + u]; - } - position = (position + (unroll * step)) & tableMask; - } - assert(position == 0); - } - } else { - U32 const tableMask = tableSize-1; + assert(tableSize <= 512); + /* Specialized symbol spreading for the case when there are + * no low probability (-1 count) symbols. When compressing + * small blocks we avoid low probability symbols to hit this + * case, since header decoding speed matters more. + */ + if (highThreshold == tableSize - 1) { + size_t const tableMask = tableSize-1; + size_t const step = FSE_TABLESTEP(tableSize); + /* First lay down the symbols in order. + * We use a uint64_t to lay down 8 bytes at a time. This reduces branch + * misses since small blocks generally have small table logs, so nearly + * all symbols have counts <= 8. We ensure we have 8 bytes at the end of + * our buffer to handle the over-write. + */ + { + U64 const add = 0x0101010101010101ull; + size_t pos = 0; + U64 sv = 0; + U32 s; + for (s=0; s<maxSV1; ++s, sv += add) { + int i; + int const n = normalizedCounter[s]; + MEM_write64(spread + pos, sv); + for (i = 8; i < n; i += 8) { + MEM_write64(spread + pos + i, sv); + } + pos += n; + } + } + /* Now we spread those positions across the table. + * The benefit of doing it in two stages is that we avoid the the + * variable size inner loop, which caused lots of branch misses. + * Now we can run through all the positions without any branch misses. + * We unroll the loop twice, since that is what emperically worked best. + */ + { + size_t position = 0; + size_t s; + size_t const unroll = 2; + assert(tableSize % unroll == 0); /* FSE_MIN_TABLELOG is 5 */ + for (s = 0; s < (size_t)tableSize; s += unroll) { + size_t u; + for (u = 0; u < unroll; ++u) { + size_t const uPosition = (position + (u * step)) & tableMask; + tableDecode[uPosition].baseValue = spread[s + u]; + } + position = (position + (unroll * step)) & tableMask; + } + assert(position == 0); + } + } else { + U32 const tableMask = tableSize-1; U32 const step = FSE_TABLESTEP(tableSize); U32 s, position = 0; for (s=0; s<maxSV1; s++) { int i; - int const n = normalizedCounter[s]; - for (i=0; i<n; i++) { + int const n = normalizedCounter[s]; + for (i=0; i<n; i++) { tableDecode[position].baseValue = s; position = (position + step) & tableMask; while (position > highThreshold) position = (position + step) & tableMask; /* lowprob area */ @@ -546,8 +546,8 @@ void ZSTD_buildFSETable_body(ZSTD_seqSymbol* dt, } /* Build Decoding table */ - { - U32 u; + { + U32 u; for (u=0; u<tableSize; u++) { U32 const symbol = tableDecode[u].baseValue; U32 const nextState = symbolNext[symbol]++; @@ -556,49 +556,49 @@ void ZSTD_buildFSETable_body(ZSTD_seqSymbol* dt, assert(nbAdditionalBits[symbol] < 255); tableDecode[u].nbAdditionalBits = nbAdditionalBits[symbol]; tableDecode[u].baseValue = baseValue[symbol]; - } - } + } + } } -/* Avoids the FORCE_INLINE of the _body() function. */ -static void ZSTD_buildFSETable_body_default(ZSTD_seqSymbol* dt, - const short* normalizedCounter, unsigned maxSymbolValue, +/* Avoids the FORCE_INLINE of the _body() function. */ +static void ZSTD_buildFSETable_body_default(ZSTD_seqSymbol* dt, + const short* normalizedCounter, unsigned maxSymbolValue, const U32* baseValue, const U8* nbAdditionalBits, - unsigned tableLog, void* wksp, size_t wkspSize) -{ - ZSTD_buildFSETable_body(dt, normalizedCounter, maxSymbolValue, - baseValue, nbAdditionalBits, tableLog, wksp, wkspSize); -} + unsigned tableLog, void* wksp, size_t wkspSize) +{ + ZSTD_buildFSETable_body(dt, normalizedCounter, maxSymbolValue, + baseValue, nbAdditionalBits, tableLog, wksp, wkspSize); +} -#if DYNAMIC_BMI2 +#if DYNAMIC_BMI2 BMI2_TARGET_ATTRIBUTE static void ZSTD_buildFSETable_body_bmi2(ZSTD_seqSymbol* dt, - const short* normalizedCounter, unsigned maxSymbolValue, + const short* normalizedCounter, unsigned maxSymbolValue, const U32* baseValue, const U8* nbAdditionalBits, - unsigned tableLog, void* wksp, size_t wkspSize) -{ - ZSTD_buildFSETable_body(dt, normalizedCounter, maxSymbolValue, - baseValue, nbAdditionalBits, tableLog, wksp, wkspSize); -} -#endif - -void ZSTD_buildFSETable(ZSTD_seqSymbol* dt, - const short* normalizedCounter, unsigned maxSymbolValue, + unsigned tableLog, void* wksp, size_t wkspSize) +{ + ZSTD_buildFSETable_body(dt, normalizedCounter, maxSymbolValue, + baseValue, nbAdditionalBits, tableLog, wksp, wkspSize); +} +#endif + +void ZSTD_buildFSETable(ZSTD_seqSymbol* dt, + const short* normalizedCounter, unsigned maxSymbolValue, const U32* baseValue, const U8* nbAdditionalBits, - unsigned tableLog, void* wksp, size_t wkspSize, int bmi2) -{ -#if DYNAMIC_BMI2 - if (bmi2) { - ZSTD_buildFSETable_body_bmi2(dt, normalizedCounter, maxSymbolValue, - baseValue, nbAdditionalBits, tableLog, wksp, wkspSize); - return; - } -#endif - (void)bmi2; - ZSTD_buildFSETable_body_default(dt, normalizedCounter, maxSymbolValue, - baseValue, nbAdditionalBits, tableLog, wksp, wkspSize); -} - - + unsigned tableLog, void* wksp, size_t wkspSize, int bmi2) +{ +#if DYNAMIC_BMI2 + if (bmi2) { + ZSTD_buildFSETable_body_bmi2(dt, normalizedCounter, maxSymbolValue, + baseValue, nbAdditionalBits, tableLog, wksp, wkspSize); + return; + } +#endif + (void)bmi2; + ZSTD_buildFSETable_body_default(dt, normalizedCounter, maxSymbolValue, + baseValue, nbAdditionalBits, tableLog, wksp, wkspSize); +} + + /*! ZSTD_buildSeqTable() : * @return : nb bytes read from src, * or an error code if it fails */ @@ -607,8 +607,8 @@ static size_t ZSTD_buildSeqTable(ZSTD_seqSymbol* DTableSpace, const ZSTD_seqSymb const void* src, size_t srcSize, const U32* baseValue, const U8* nbAdditionalBits, const ZSTD_seqSymbol* defaultTable, U32 flagRepeatTable, - int ddictIsCold, int nbSeq, U32* wksp, size_t wkspSize, - int bmi2) + int ddictIsCold, int nbSeq, U32* wksp, size_t wkspSize, + int bmi2) { switch(type) { @@ -640,7 +640,7 @@ static size_t ZSTD_buildSeqTable(ZSTD_seqSymbol* DTableSpace, const ZSTD_seqSymb size_t const headerSize = FSE_readNCount(norm, &max, &tableLog, src, srcSize); RETURN_ERROR_IF(FSE_isError(headerSize), corruption_detected, ""); RETURN_ERROR_IF(tableLog > maxLog, corruption_detected, ""); - ZSTD_buildFSETable(DTableSpace, norm, max, baseValue, nbAdditionalBits, tableLog, wksp, wkspSize, bmi2); + ZSTD_buildFSETable(DTableSpace, norm, max, baseValue, nbAdditionalBits, tableLog, wksp, wkspSize, bmi2); *DTablePtr = DTableSpace; return headerSize; } @@ -653,7 +653,7 @@ static size_t ZSTD_buildSeqTable(ZSTD_seqSymbol* DTableSpace, const ZSTD_seqSymb size_t ZSTD_decodeSeqHeaders(ZSTD_DCtx* dctx, int* nbSeqPtr, const void* src, size_t srcSize) { - const BYTE* const istart = (const BYTE*)src; + const BYTE* const istart = (const BYTE*)src; const BYTE* const iend = istart + srcSize; const BYTE* ip = istart; int nbSeq; @@ -672,8 +672,8 @@ size_t ZSTD_decodeSeqHeaders(ZSTD_DCtx* dctx, int* nbSeqPtr, if (nbSeq > 0x7F) { if (nbSeq == 0xFF) { RETURN_ERROR_IF(ip+2 > iend, srcSize_wrong, ""); - nbSeq = MEM_readLE16(ip) + LONGNBSEQ; - ip+=2; + nbSeq = MEM_readLE16(ip) + LONGNBSEQ; + ip+=2; } else { RETURN_ERROR_IF(ip >= iend, srcSize_wrong, ""); nbSeq = ((nbSeq-0x80)<<8) + *ip++; @@ -694,8 +694,8 @@ size_t ZSTD_decodeSeqHeaders(ZSTD_DCtx* dctx, int* nbSeqPtr, ip, iend-ip, LL_base, LL_bits, LL_defaultDTable, dctx->fseEntropy, - dctx->ddictIsCold, nbSeq, - dctx->workspace, sizeof(dctx->workspace), + dctx->ddictIsCold, nbSeq, + dctx->workspace, sizeof(dctx->workspace), ZSTD_DCtx_get_bmi2(dctx)); RETURN_ERROR_IF(ZSTD_isError(llhSize), corruption_detected, "ZSTD_buildSeqTable failed"); ip += llhSize; @@ -706,8 +706,8 @@ size_t ZSTD_decodeSeqHeaders(ZSTD_DCtx* dctx, int* nbSeqPtr, ip, iend-ip, OF_base, OF_bits, OF_defaultDTable, dctx->fseEntropy, - dctx->ddictIsCold, nbSeq, - dctx->workspace, sizeof(dctx->workspace), + dctx->ddictIsCold, nbSeq, + dctx->workspace, sizeof(dctx->workspace), ZSTD_DCtx_get_bmi2(dctx)); RETURN_ERROR_IF(ZSTD_isError(ofhSize), corruption_detected, "ZSTD_buildSeqTable failed"); ip += ofhSize; @@ -718,8 +718,8 @@ size_t ZSTD_decodeSeqHeaders(ZSTD_DCtx* dctx, int* nbSeqPtr, ip, iend-ip, ML_base, ML_bits, ML_defaultDTable, dctx->fseEntropy, - dctx->ddictIsCold, nbSeq, - dctx->workspace, sizeof(dctx->workspace), + dctx->ddictIsCold, nbSeq, + dctx->workspace, sizeof(dctx->workspace), ZSTD_DCtx_get_bmi2(dctx)); RETURN_ERROR_IF(ZSTD_isError(mlhSize), corruption_detected, "ZSTD_buildSeqTable failed"); ip += mlhSize; @@ -886,7 +886,7 @@ size_t ZSTD_execSequenceEnd(BYTE* op, RETURN_ERROR_IF(sequence.offset > (size_t)(oLitEnd - virtualStart), corruption_detected, ""); match = dictEnd - (prefixStart - match); if (match + sequence.matchLength <= dictEnd) { - ZSTD_memmove(oLitEnd, match, sequence.matchLength); + ZSTD_memmove(oLitEnd, match, sequence.matchLength); return sequenceLength; } /* span extDict & currentPrefixSegment */ @@ -1093,12 +1093,12 @@ size_t ZSTD_execSequenceSplitLitBuffer(BYTE* op, RETURN_ERROR_IF(UNLIKELY(sequence.offset > (size_t)(oLitEnd - virtualStart)), corruption_detected, ""); match = dictEnd + (match - prefixStart); if (match + sequence.matchLength <= dictEnd) { - ZSTD_memmove(oLitEnd, match, sequence.matchLength); + ZSTD_memmove(oLitEnd, match, sequence.matchLength); return sequenceLength; } /* span extDict & currentPrefixSegment */ { size_t const length1 = dictEnd - match; - ZSTD_memmove(oLitEnd, match, length1); + ZSTD_memmove(oLitEnd, match, length1); op = oLitEnd + length1; sequence.matchLength -= length1; match = prefixStart; @@ -1166,7 +1166,7 @@ ZSTD_updateFseStateWithDInfo(ZSTD_fseState* DStatePtr, BIT_DStream_t* bitD, U16 typedef enum { ZSTD_lo_isRegularOffset, ZSTD_lo_isLongOffset=1 } ZSTD_longOffset_e; FORCE_INLINE_TEMPLATE seq_t -ZSTD_decodeSequence(seqState_t* seqState, const ZSTD_longOffset_e longOffsets) +ZSTD_decodeSequence(seqState_t* seqState, const ZSTD_longOffset_e longOffsets) { seq_t seq; const ZSTD_seqSymbol* const llDInfo = seqState->stateLL.table + seqState->stateLL.state; @@ -1269,7 +1269,7 @@ ZSTD_decodeSequence(seqState_t* seqState, const ZSTD_longOffset_e longOffsets) } #ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION -MEM_STATIC int ZSTD_dictionaryIsActive(ZSTD_DCtx const* dctx, BYTE const* prefixStart, BYTE const* oLitEnd) +MEM_STATIC int ZSTD_dictionaryIsActive(ZSTD_DCtx const* dctx, BYTE const* prefixStart, BYTE const* oLitEnd) { size_t const windowSize = dctx->fParams.windowSize; /* No dictionary used. */ @@ -1290,7 +1290,7 @@ MEM_STATIC void ZSTD_assertValidSequence( seq_t const seq, BYTE const* prefixStart, BYTE const* virtualStart) { -#if DEBUGLEVEL >= 1 +#if DEBUGLEVEL >= 1 size_t const windowSize = dctx->fParams.windowSize; size_t const sequenceSize = seq.litLength + seq.matchLength; BYTE const* const oLitEnd = op + seq.litLength; @@ -1308,9 +1308,9 @@ MEM_STATIC void ZSTD_assertValidSequence( /* Offset must be within our window. */ assert(seq.offset <= windowSize); } -#else - (void)dctx, (void)op, (void)oend, (void)seq, (void)prefixStart, (void)virtualStart; -#endif +#else + (void)dctx, (void)op, (void)oend, (void)seq, (void)prefixStart, (void)virtualStart; +#endif } #endif @@ -1327,7 +1327,7 @@ ZSTD_decompressSequences_bodySplitLitBuffer( ZSTD_DCtx* dctx, { const BYTE* ip = (const BYTE*)seqStart; const BYTE* const iend = ip + seqSize; - BYTE* const ostart = (BYTE*)dst; + BYTE* const ostart = (BYTE*)dst; BYTE* const oend = ostart + maxDstSize; BYTE* op = ostart; const BYTE* litPtr = dctx->litPtr; @@ -1478,11 +1478,11 @@ ZSTD_decompressSequences_bodySplitLitBuffer( ZSTD_DCtx* dctx, __asm__(".p2align 3"); # elif __GNUC__ >= 11 __asm__(".p2align 3"); -# else +# else __asm__(".p2align 5"); __asm__("nop"); __asm__(".p2align 3"); -# endif +# endif #endif for (; ; ) { @@ -1588,19 +1588,19 @@ ZSTD_decompressSequences_body(ZSTD_DCtx* dctx, #endif for ( ; ; ) { - seq_t const sequence = ZSTD_decodeSequence(&seqState, isLongOffset); + seq_t const sequence = ZSTD_decodeSequence(&seqState, isLongOffset); size_t const oneSeqSize = ZSTD_execSequence(op, oend, sequence, &litPtr, litEnd, prefixStart, vBase, dictEnd); #if defined(FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION) && defined(FUZZING_ASSERT_VALID_SEQUENCE) assert(!ZSTD_isError(oneSeqSize)); if (frame) ZSTD_assertValidSequence(dctx, op, oend, sequence, prefixStart, vBase); #endif - if (UNLIKELY(ZSTD_isError(oneSeqSize))) - return oneSeqSize; + if (UNLIKELY(ZSTD_isError(oneSeqSize))) + return oneSeqSize; DEBUGLOG(6, "regenerated sequence size : %u", (U32)oneSeqSize); - op += oneSeqSize; - if (UNLIKELY(!--nbSeq)) - break; - BIT_reloadDStream(&(seqState.DStream)); + op += oneSeqSize; + if (UNLIKELY(!--nbSeq)) + break; + BIT_reloadDStream(&(seqState.DStream)); } /* check if reached exact end */ @@ -1615,7 +1615,7 @@ ZSTD_decompressSequences_body(ZSTD_DCtx* dctx, { size_t const lastLLSize = litEnd - litPtr; RETURN_ERROR_IF(lastLLSize > (size_t)(oend-op), dstSize_tooSmall, ""); if (op != NULL) { - ZSTD_memcpy(op, litPtr, lastLLSize); + ZSTD_memcpy(op, litPtr, lastLLSize); op += lastLLSize; } } @@ -1645,25 +1645,25 @@ ZSTD_decompressSequencesSplitLitBuffer_default(ZSTD_DCtx* dctx, #endif /* ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG */ #ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT - -FORCE_INLINE_TEMPLATE size_t -ZSTD_prefetchMatch(size_t prefetchPos, seq_t const sequence, - const BYTE* const prefixStart, const BYTE* const dictEnd) -{ - prefetchPos += sequence.litLength; - { const BYTE* const matchBase = (sequence.offset > prefetchPos) ? dictEnd : prefixStart; - const BYTE* const match = matchBase + prefetchPos - sequence.offset; /* note : this operation can overflow when seq.offset is really too large, which can only happen when input is corrupted. - * No consequence though : memory address is only used for prefetching, not for dereferencing */ - PREFETCH_L1(match); PREFETCH_L1(match+CACHELINE_SIZE); /* note : it's safe to invoke PREFETCH() on any memory address, including invalid ones */ - } - return prefetchPos + sequence.matchLength; -} - -/* This decoding function employs prefetching - * to reduce latency impact of cache misses. - * It's generally employed when block contains a significant portion of long-distance matches - * or when coupled with a "cold" dictionary */ + FORCE_INLINE_TEMPLATE size_t +ZSTD_prefetchMatch(size_t prefetchPos, seq_t const sequence, + const BYTE* const prefixStart, const BYTE* const dictEnd) +{ + prefetchPos += sequence.litLength; + { const BYTE* const matchBase = (sequence.offset > prefetchPos) ? dictEnd : prefixStart; + const BYTE* const match = matchBase + prefetchPos - sequence.offset; /* note : this operation can overflow when seq.offset is really too large, which can only happen when input is corrupted. + * No consequence though : memory address is only used for prefetching, not for dereferencing */ + PREFETCH_L1(match); PREFETCH_L1(match+CACHELINE_SIZE); /* note : it's safe to invoke PREFETCH() on any memory address, including invalid ones */ + } + return prefetchPos + sequence.matchLength; +} + +/* This decoding function employs prefetching + * to reduce latency impact of cache misses. + * It's generally employed when block contains a significant portion of long-distance matches + * or when coupled with a "cold" dictionary */ +FORCE_INLINE_TEMPLATE size_t ZSTD_decompressSequencesLong_body( ZSTD_DCtx* dctx, void* dst, size_t maxDstSize, @@ -1673,7 +1673,7 @@ ZSTD_decompressSequencesLong_body( { const BYTE* ip = (const BYTE*)seqStart; const BYTE* const iend = ip + seqSize; - BYTE* const ostart = (BYTE*)dst; + BYTE* const ostart = (BYTE*)dst; BYTE* const oend = dctx->litBufferLocation == ZSTD_in_dst ? dctx->litBuffer : ostart + maxDstSize; BYTE* op = ostart; const BYTE* litPtr = dctx->litPtr; @@ -1685,15 +1685,15 @@ ZSTD_decompressSequencesLong_body( /* Regen sequences */ if (nbSeq) { -#define STORED_SEQS 8 +#define STORED_SEQS 8 #define STORED_SEQS_MASK (STORED_SEQS-1) -#define ADVANCED_SEQS STORED_SEQS +#define ADVANCED_SEQS STORED_SEQS seq_t sequences[STORED_SEQS]; int const seqAdvance = MIN(nbSeq, ADVANCED_SEQS); seqState_t seqState; int seqNb; - size_t prefetchPos = (size_t)(op-prefixStart); /* track position relative to prefixStart */ - + size_t prefetchPos = (size_t)(op-prefixStart); /* track position relative to prefixStart */ + dctx->fseEntropy = 1; { int i; for (i=0; i<ZSTD_REP_NUM; i++) seqState.prevOffset[i] = dctx->entropy.rep[i]; } assert(dst != NULL); @@ -1707,9 +1707,9 @@ ZSTD_decompressSequencesLong_body( /* prepare in advance */ for (seqNb=0; (BIT_reloadDStream(&seqState.DStream) <= BIT_DStream_completed) && (seqNb<seqAdvance); seqNb++) { - seq_t const sequence = ZSTD_decodeSequence(&seqState, isLongOffset); - prefetchPos = ZSTD_prefetchMatch(prefetchPos, sequence, prefixStart, dictEnd); - sequences[seqNb] = sequence; + seq_t const sequence = ZSTD_decodeSequence(&seqState, isLongOffset); + prefetchPos = ZSTD_prefetchMatch(prefetchPos, sequence, prefixStart, dictEnd); + sequences[seqNb] = sequence; } RETURN_ERROR_IF(seqNb<seqAdvance, corruption_detected, ""); @@ -1738,7 +1738,7 @@ ZSTD_decompressSequencesLong_body( if (frame) ZSTD_assertValidSequence(dctx, op, oend, sequences[(seqNb - ADVANCED_SEQS) & STORED_SEQS_MASK], prefixStart, dictStart); #endif if (ZSTD_isError(oneSeqSize)) return oneSeqSize; - + prefetchPos = ZSTD_prefetchMatch(prefetchPos, sequence, prefixStart, dictEnd); sequences[seqNb & STORED_SEQS_MASK] = sequence; op += oneSeqSize; @@ -2049,9 +2049,9 @@ ZSTD_decompressBlock_internal(ZSTD_DCtx* dctx, } -void ZSTD_checkContinuity(ZSTD_DCtx* dctx, const void* dst, size_t dstSize) +void ZSTD_checkContinuity(ZSTD_DCtx* dctx, const void* dst, size_t dstSize) { - if (dst != dctx->previousDstEnd && dstSize > 0) { /* not contiguous */ + if (dst != dctx->previousDstEnd && dstSize > 0) { /* not contiguous */ dctx->dictEnd = dctx->previousDstEnd; dctx->virtualStart = (const char*)dst - ((const char*)(dctx->previousDstEnd) - (const char*)(dctx->prefixStart)); dctx->prefixStart = dst; @@ -2065,7 +2065,7 @@ size_t ZSTD_decompressBlock(ZSTD_DCtx* dctx, const void* src, size_t srcSize) { size_t dSize; - ZSTD_checkContinuity(dctx, dst, dstCapacity); + ZSTD_checkContinuity(dctx, dst, dstCapacity); dSize = ZSTD_decompressBlock_internal(dctx, dst, dstCapacity, src, srcSize, /* frame */ 0, not_streaming); dctx->previousDstEnd = (char*)dst + dSize; return dSize; diff --git a/contrib/libs/zstd/lib/decompress/zstd_decompress_block.h b/contrib/libs/zstd/lib/decompress/zstd_decompress_block.h index c61a9d0c4b..5c48afb54b 100644 --- a/contrib/libs/zstd/lib/decompress/zstd_decompress_block.h +++ b/contrib/libs/zstd/lib/decompress/zstd_decompress_block.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -15,7 +15,7 @@ /*-******************************************************* * Dependencies *********************************************************/ -#include "../common/zstd_deps.h" /* size_t */ +#include "../common/zstd_deps.h" /* size_t */ #include "../zstd.h" /* DCtx, and some public functions */ #include "../common/zstd_internal.h" /* blockProperties_t, and some public functions */ #include "zstd_decompress_internal.h" /* ZSTD_seqSymbol */ @@ -54,15 +54,15 @@ size_t ZSTD_decompressBlock_internal(ZSTD_DCtx* dctx, * this function must be called with valid parameters only * (dt is large enough, normalizedCounter distribution total is a power of 2, max is within range, etc.) * in which case it cannot fail. - * The workspace must be 4-byte aligned and at least ZSTD_BUILD_FSE_TABLE_WKSP_SIZE bytes, which is - * defined in zstd_decompress_internal.h. + * The workspace must be 4-byte aligned and at least ZSTD_BUILD_FSE_TABLE_WKSP_SIZE bytes, which is + * defined in zstd_decompress_internal.h. * Internal use only. */ void ZSTD_buildFSETable(ZSTD_seqSymbol* dt, const short* normalizedCounter, unsigned maxSymbolValue, const U32* baseValue, const U8* nbAdditionalBits, - unsigned tableLog, void* wksp, size_t wkspSize, - int bmi2); + unsigned tableLog, void* wksp, size_t wkspSize, + int bmi2); #endif /* ZSTD_DEC_BLOCK_H */ diff --git a/contrib/libs/zstd/lib/decompress/zstd_decompress_internal.h b/contrib/libs/zstd/lib/decompress/zstd_decompress_internal.h index 2b5a53850a..23e81c2697 100644 --- a/contrib/libs/zstd/lib/decompress/zstd_decompress_internal.h +++ b/contrib/libs/zstd/lib/decompress/zstd_decompress_internal.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -27,14 +27,14 @@ /*-******************************************************* * Constants *********************************************************/ -static UNUSED_ATTR const U32 LL_base[MaxLL+1] = { +static UNUSED_ATTR const U32 LL_base[MaxLL+1] = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, 20, 22, 24, 28, 32, 40, 48, 64, 0x80, 0x100, 0x200, 0x400, 0x800, 0x1000, 0x2000, 0x4000, 0x8000, 0x10000 }; -static UNUSED_ATTR const U32 OF_base[MaxOff+1] = { +static UNUSED_ATTR const U32 OF_base[MaxOff+1] = { 0, 1, 1, 5, 0xD, 0x1D, 0x3D, 0x7D, 0xFD, 0x1FD, 0x3FD, 0x7FD, 0xFFD, 0x1FFD, 0x3FFD, 0x7FFD, 0xFFFD, 0x1FFFD, 0x3FFFD, 0x7FFFD, 0xFFFFD, 0x1FFFFD, 0x3FFFFD, 0x7FFFFD, @@ -46,7 +46,7 @@ static UNUSED_ATTR const U8 OF_bits[MaxOff+1] = { 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 }; -static UNUSED_ATTR const U32 ML_base[MaxML+1] = { +static UNUSED_ATTR const U32 ML_base[MaxML+1] = { 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, @@ -73,16 +73,16 @@ static UNUSED_ATTR const U32 ML_base[MaxML+1] = { #define SEQSYMBOL_TABLE_SIZE(log) (1 + (1 << (log))) -#define ZSTD_BUILD_FSE_TABLE_WKSP_SIZE (sizeof(S16) * (MaxSeq + 1) + (1u << MaxFSELog) + sizeof(U64)) -#define ZSTD_BUILD_FSE_TABLE_WKSP_SIZE_U32 ((ZSTD_BUILD_FSE_TABLE_WKSP_SIZE + sizeof(U32) - 1) / sizeof(U32)) - +#define ZSTD_BUILD_FSE_TABLE_WKSP_SIZE (sizeof(S16) * (MaxSeq + 1) + (1u << MaxFSELog) + sizeof(U64)) +#define ZSTD_BUILD_FSE_TABLE_WKSP_SIZE_U32 ((ZSTD_BUILD_FSE_TABLE_WKSP_SIZE + sizeof(U32) - 1) / sizeof(U32)) + typedef struct { ZSTD_seqSymbol LLTable[SEQSYMBOL_TABLE_SIZE(LLFSELog)]; /* Note : Space reserved for FSE Tables */ ZSTD_seqSymbol OFTable[SEQSYMBOL_TABLE_SIZE(OffFSELog)]; /* is also used as temporary workspace while building hufTable during DDict creation */ ZSTD_seqSymbol MLTable[SEQSYMBOL_TABLE_SIZE(MLFSELog)]; /* and therefore must be at least HUF_DECOMPRESS_WORKSPACE_SIZE large */ HUF_DTable hufTable[HUF_DTABLE_SIZE(HufLog)]; /* can accommodate HUF_decompress4X */ U32 rep[ZSTD_REP_NUM]; - U32 workspace[ZSTD_BUILD_FSE_TABLE_WKSP_SIZE_U32]; + U32 workspace[ZSTD_BUILD_FSE_TABLE_WKSP_SIZE_U32]; } ZSTD_entropyDTables_t; typedef enum { ZSTDds_getFrameHeaderSize, ZSTDds_decodeFrameHeader, @@ -99,13 +99,13 @@ typedef enum { ZSTD_use_once = 1 /* Use the dictionary once and set to ZSTD_dont_use */ } ZSTD_dictUses_e; -/* Hashset for storing references to multiple ZSTD_DDict within ZSTD_DCtx */ -typedef struct { - const ZSTD_DDict** ddictPtrTable; - size_t ddictPtrTableSize; - size_t ddictPtrCount; -} ZSTD_DDictHashSet; - +/* Hashset for storing references to multiple ZSTD_DDict within ZSTD_DCtx */ +typedef struct { + const ZSTD_DDict** ddictPtrTable; + size_t ddictPtrTableSize; + size_t ddictPtrCount; +} ZSTD_DDictHashSet; + #ifndef ZSTD_DECODER_INTERNAL_BUFFER # define ZSTD_DECODER_INTERNAL_BUFFER (1 << 16) #endif @@ -136,7 +136,7 @@ struct ZSTD_DCtx_s const void* dictEnd; /* end of previous segment */ size_t expected; ZSTD_frameHeader fParams; - U64 processedCSize; + U64 processedCSize; U64 decodedSize; blockType_e bType; /* used in ZSTD_decompressContinue(), store blockType between block header decoding and block decompression stages */ ZSTD_dStage stage; @@ -145,8 +145,8 @@ struct ZSTD_DCtx_s XXH64_state_t xxhState; size_t headerSize; ZSTD_format_e format; - ZSTD_forceIgnoreChecksum_e forceIgnoreChecksum; /* User specified: if == 1, will ignore checksums in compressed frame. Default == 0 */ - U32 validateChecksum; /* if == 1, will validate checksum. Is == 1 if (fParams.checksumFlag == 1) and (forceIgnoreChecksum == 0). */ + ZSTD_forceIgnoreChecksum_e forceIgnoreChecksum; /* User specified: if == 1, will ignore checksums in compressed frame. Default == 0 */ + U32 validateChecksum; /* if == 1, will validate checksum. Is == 1 if (fParams.checksumFlag == 1) and (forceIgnoreChecksum == 0). */ const BYTE* litPtr; ZSTD_customMem customMem; size_t litSize; @@ -162,8 +162,8 @@ struct ZSTD_DCtx_s U32 dictID; int ddictIsCold; /* if == 1 : dictionary is "new" for working context, and presumed "cold" (not in cpu cache) */ ZSTD_dictUses_e dictUses; - ZSTD_DDictHashSet* ddictSet; /* Hash set for multiple ddicts */ - ZSTD_refMultipleDDicts_e refMultipleDDicts; /* User specified: if == 1, will allow references to multiple DDicts. Default == 0 (disabled) */ + ZSTD_DDictHashSet* ddictSet; /* Hash set for multiple ddicts */ + ZSTD_refMultipleDDicts_e refMultipleDDicts; /* User specified: if == 1, will allow references to multiple DDicts. Default == 0 (disabled) */ /* streaming */ ZSTD_dStreamStage streamStage; @@ -183,7 +183,7 @@ struct ZSTD_DCtx_s #endif U32 hostageByte; int noForwardProgress; - ZSTD_bufferMode_e outBufferMode; + ZSTD_bufferMode_e outBufferMode; ZSTD_outBuffer expectedOutBuffer; /* workspace */ @@ -199,11 +199,11 @@ struct ZSTD_DCtx_s void const* dictContentBeginForFuzzing; void const* dictContentEndForFuzzing; #endif - - /* Tracing */ -#if ZSTD_TRACE - ZSTD_TraceCtx traceCtx; -#endif + + /* Tracing */ +#if ZSTD_TRACE + ZSTD_TraceCtx traceCtx; +#endif }; /* typedef'd to ZSTD_DCtx within "zstd.h" */ MEM_STATIC int ZSTD_DCtx_get_bmi2(const struct ZSTD_DCtx_s *dctx) { @@ -230,7 +230,7 @@ size_t ZSTD_loadDEntropy(ZSTD_entropyDTables_t* entropy, * If yes, do nothing (continue on current segment). * If not, classify previous segment as "external dictionary", and start a new segment. * This function cannot fail. */ -void ZSTD_checkContinuity(ZSTD_DCtx* dctx, const void* dst, size_t dstSize); +void ZSTD_checkContinuity(ZSTD_DCtx* dctx, const void* dst, size_t dstSize); #endif /* ZSTD_DECOMPRESS_INTERNAL_H */ diff --git a/contrib/libs/zstd/lib/dictBuilder/cover.c b/contrib/libs/zstd/lib/dictBuilder/cover.c index 028802a1b0..5b3f828889 100644 --- a/contrib/libs/zstd/lib/dictBuilder/cover.c +++ b/contrib/libs/zstd/lib/dictBuilder/cover.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -26,15 +26,15 @@ #include <string.h> /* memset */ #include <time.h> /* clock */ -#ifndef ZDICT_STATIC_LINKING_ONLY -# define ZDICT_STATIC_LINKING_ONLY -#endif - +#ifndef ZDICT_STATIC_LINKING_ONLY +# define ZDICT_STATIC_LINKING_ONLY +#endif + #include "../common/mem.h" /* read */ #include "../common/pool.h" #include "../common/threading.h" -#include "../common/zstd_internal.h" /* includes zstd.h */ -#include "../zdict.h" +#include "../common/zstd_internal.h" /* includes zstd.h */ +#include "../zdict.h" #include "cover.h" /*-************************************* @@ -48,41 +48,41 @@ * contiguous buffer, so 1GB is already a high limit. */ #define COVER_MAX_SAMPLES_SIZE (sizeof(size_t) == 8 ? ((unsigned)-1) : ((unsigned)1 GB)) -#define COVER_DEFAULT_SPLITPOINT 1.0 +#define COVER_DEFAULT_SPLITPOINT 1.0 /*-************************************* * Console display ***************************************/ -#ifndef LOCALDISPLAYLEVEL +#ifndef LOCALDISPLAYLEVEL static int g_displayLevel = 0; -#endif -#undef DISPLAY +#endif +#undef DISPLAY #define DISPLAY(...) \ { \ fprintf(stderr, __VA_ARGS__); \ fflush(stderr); \ } -#undef LOCALDISPLAYLEVEL +#undef LOCALDISPLAYLEVEL #define LOCALDISPLAYLEVEL(displayLevel, l, ...) \ if (displayLevel >= l) { \ DISPLAY(__VA_ARGS__); \ } /* 0 : no display; 1: errors; 2: default; 3: details; 4: debug */ -#undef DISPLAYLEVEL +#undef DISPLAYLEVEL #define DISPLAYLEVEL(l, ...) LOCALDISPLAYLEVEL(g_displayLevel, l, __VA_ARGS__) -#ifndef LOCALDISPLAYUPDATE -static const clock_t g_refreshRate = CLOCKS_PER_SEC * 15 / 100; -static clock_t g_time = 0; -#endif -#undef LOCALDISPLAYUPDATE +#ifndef LOCALDISPLAYUPDATE +static const clock_t g_refreshRate = CLOCKS_PER_SEC * 15 / 100; +static clock_t g_time = 0; +#endif +#undef LOCALDISPLAYUPDATE #define LOCALDISPLAYUPDATE(displayLevel, l, ...) \ if (displayLevel >= l) { \ - if ((clock() - g_time > g_refreshRate) || (displayLevel >= 4)) { \ + if ((clock() - g_time > g_refreshRate) || (displayLevel >= 4)) { \ g_time = clock(); \ DISPLAY(__VA_ARGS__); \ } \ } -#undef DISPLAYUPDATE +#undef DISPLAYUPDATE #define DISPLAYUPDATE(l, ...) LOCALDISPLAYUPDATE(g_displayLevel, l, __VA_ARGS__) /*-************************************* @@ -137,9 +137,9 @@ static int COVER_map_init(COVER_map_t *map, U32 size) { /** * Internal hash function */ -static const U32 COVER_prime4bytes = 2654435761U; +static const U32 COVER_prime4bytes = 2654435761U; static U32 COVER_map_hash(COVER_map_t *map, U32 key) { - return (key * COVER_prime4bytes) >> (32 - map->sizeLog); + return (key * COVER_prime4bytes) >> (32 - map->sizeLog); } /** @@ -232,7 +232,7 @@ typedef struct { } COVER_ctx_t; /* We need a global context for qsort... */ -static COVER_ctx_t *g_coverCtx = NULL; +static COVER_ctx_t *g_coverCtx = NULL; /*-************************************* * Helper functions @@ -275,11 +275,11 @@ static int COVER_cmp8(COVER_ctx_t *ctx, const void *lp, const void *rp) { /** * Same as COVER_cmp() except ties are broken by pointer value - * NOTE: g_coverCtx must be set to call this function. A global is required because + * NOTE: g_coverCtx must be set to call this function. A global is required because * qsort doesn't take an opaque pointer. */ -static int WIN_CDECL COVER_strict_cmp(const void *lp, const void *rp) { - int result = COVER_cmp(g_coverCtx, lp, rp); +static int WIN_CDECL COVER_strict_cmp(const void *lp, const void *rp) { + int result = COVER_cmp(g_coverCtx, lp, rp); if (result == 0) { result = lp < rp ? -1 : 1; } @@ -288,8 +288,8 @@ static int WIN_CDECL COVER_strict_cmp(const void *lp, const void *rp) { /** * Faster version for d <= 8. */ -static int WIN_CDECL COVER_strict_cmp8(const void *lp, const void *rp) { - int result = COVER_cmp8(g_coverCtx, lp, rp); +static int WIN_CDECL COVER_strict_cmp8(const void *lp, const void *rp) { + int result = COVER_cmp8(g_coverCtx, lp, rp); if (result == 0) { result = lp < rp ? -1 : 1; } @@ -620,7 +620,7 @@ static size_t COVER_ctx_init(COVER_ctx_t *ctx, const void *samplesBuffer, /* qsort doesn't take an opaque pointer, so pass as a global. * On OpenBSD qsort() is not guaranteed to be stable, their mergesort() is. */ - g_coverCtx = ctx; + g_coverCtx = ctx; #if defined(__OpenBSD__) mergesort(ctx->suffix, ctx->suffixSize, sizeof(U32), (ctx->d <= 8 ? &COVER_strict_cmp8 : &COVER_strict_cmp)); @@ -963,7 +963,7 @@ void COVER_dictSelectionFree(COVER_dictSelection_t selection){ free(selection.dictContent); } -COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent, size_t dictBufferCapacity, +COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent, size_t dictBufferCapacity, size_t dictContentSize, const BYTE* samplesBuffer, const size_t* samplesSizes, unsigned nbFinalizeSamples, size_t nbCheckSamples, size_t nbSamples, ZDICT_cover_params_t params, size_t* offsets, size_t totalCompressedSize) { @@ -971,8 +971,8 @@ COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent, size_t dictBuffe size_t largestCompressed = 0; BYTE* customDictContentEnd = customDictContent + dictContentSize; - BYTE * largestDictbuffer = (BYTE *)malloc(dictBufferCapacity); - BYTE * candidateDictBuffer = (BYTE *)malloc(dictBufferCapacity); + BYTE * largestDictbuffer = (BYTE *)malloc(dictBufferCapacity); + BYTE * candidateDictBuffer = (BYTE *)malloc(dictBufferCapacity); double regressionTolerance = ((double)params.shrinkDictMaxRegression / 100.0) + 1.00; if (!largestDictbuffer || !candidateDictBuffer) { @@ -984,7 +984,7 @@ COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent, size_t dictBuffe /* Initial dictionary size and compressed size */ memcpy(largestDictbuffer, customDictContent, dictContentSize); dictContentSize = ZDICT_finalizeDictionary( - largestDictbuffer, dictBufferCapacity, customDictContent, dictContentSize, + largestDictbuffer, dictBufferCapacity, customDictContent, dictContentSize, samplesBuffer, samplesSizes, nbFinalizeSamples, params.zParams); if (ZDICT_isError(dictContentSize)) { @@ -1018,7 +1018,7 @@ COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent, size_t dictBuffe while (dictContentSize < largestDict) { memcpy(candidateDictBuffer, largestDictbuffer, largestDict); dictContentSize = ZDICT_finalizeDictionary( - candidateDictBuffer, dictBufferCapacity, customDictContentEnd - dictContentSize, dictContentSize, + candidateDictBuffer, dictBufferCapacity, customDictContentEnd - dictContentSize, dictContentSize, samplesBuffer, samplesSizes, nbFinalizeSamples, params.zParams); if (ZDICT_isError(dictContentSize)) { @@ -1070,19 +1070,19 @@ typedef struct COVER_tryParameters_data_s { * This function is thread safe if zstd is compiled with multithreaded support. * It takes its parameters as an *OWNING* opaque pointer to support threading. */ -static void COVER_tryParameters(void *opaque) -{ +static void COVER_tryParameters(void *opaque) +{ /* Save parameters as local variables */ - COVER_tryParameters_data_t *const data = (COVER_tryParameters_data_t*)opaque; + COVER_tryParameters_data_t *const data = (COVER_tryParameters_data_t*)opaque; const COVER_ctx_t *const ctx = data->ctx; const ZDICT_cover_params_t parameters = data->parameters; size_t dictBufferCapacity = data->dictBufferCapacity; size_t totalCompressedSize = ERROR(GENERIC); /* Allocate space for hash table, dict, and freqs */ COVER_map_t activeDmers; - BYTE* const dict = (BYTE*)malloc(dictBufferCapacity); + BYTE* const dict = (BYTE*)malloc(dictBufferCapacity); COVER_dictSelection_t selection = COVER_dictSelectionError(ERROR(GENERIC)); - U32* const freqs = (U32*)malloc(ctx->suffixSize * sizeof(U32)); + U32* const freqs = (U32*)malloc(ctx->suffixSize * sizeof(U32)); if (!COVER_map_init(&activeDmers, parameters.k - parameters.d + 1)) { DISPLAYLEVEL(1, "Failed to allocate dmer map: out of memory\n"); goto _cleanup; @@ -1097,7 +1097,7 @@ static void COVER_tryParameters(void *opaque) { const size_t tail = COVER_buildDictionary(ctx, freqs, &activeDmers, dict, dictBufferCapacity, parameters); - selection = COVER_selectDict(dict + tail, dictBufferCapacity, dictBufferCapacity - tail, + selection = COVER_selectDict(dict + tail, dictBufferCapacity, dictBufferCapacity - tail, ctx->samples, ctx->samplesSizes, (unsigned)ctx->nbTrainSamples, ctx->nbTrainSamples, ctx->nbSamples, parameters, ctx->offsets, totalCompressedSize); @@ -1112,18 +1112,18 @@ _cleanup: free(data); COVER_map_destroy(&activeDmers); COVER_dictSelectionFree(selection); - free(freqs); + free(freqs); } ZDICTLIB_API size_t ZDICT_optimizeTrainFromBuffer_cover( - void* dictBuffer, size_t dictBufferCapacity, const void* samplesBuffer, - const size_t* samplesSizes, unsigned nbSamples, - ZDICT_cover_params_t* parameters) -{ + void* dictBuffer, size_t dictBufferCapacity, const void* samplesBuffer, + const size_t* samplesSizes, unsigned nbSamples, + ZDICT_cover_params_t* parameters) +{ /* constants */ const unsigned nbThreads = parameters->nbThreads; const double splitPoint = - parameters->splitPoint <= 0.0 ? COVER_DEFAULT_SPLITPOINT : parameters->splitPoint; + parameters->splitPoint <= 0.0 ? COVER_DEFAULT_SPLITPOINT : parameters->splitPoint; const unsigned kMinD = parameters->d == 0 ? 6 : parameters->d; const unsigned kMaxD = parameters->d == 0 ? 8 : parameters->d; const unsigned kMinK = parameters->k == 0 ? 50 : parameters->k; diff --git a/contrib/libs/zstd/lib/dictBuilder/cover.h b/contrib/libs/zstd/lib/dictBuilder/cover.h index 1aacdddd6f..0bce7d25c2 100644 --- a/contrib/libs/zstd/lib/dictBuilder/cover.h +++ b/contrib/libs/zstd/lib/dictBuilder/cover.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Facebook, Inc. + * Copyright (c) Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -8,10 +8,10 @@ * You may select, at your option, one of the above-listed licenses. */ -#ifndef ZDICT_STATIC_LINKING_ONLY -# define ZDICT_STATIC_LINKING_ONLY -#endif - +#ifndef ZDICT_STATIC_LINKING_ONLY +# define ZDICT_STATIC_LINKING_ONLY +#endif + #include <stdio.h> /* fprintf */ #include <stdlib.h> /* malloc, free, qsort */ #include <string.h> /* memset */ @@ -20,7 +20,7 @@ #include "../common/pool.h" #include "../common/threading.h" #include "../common/zstd_internal.h" /* includes zstd.h */ -#include "../zdict.h" +#include "../zdict.h" /** * COVER_best_t is used for two purposes: @@ -153,6 +153,6 @@ void COVER_dictSelectionFree(COVER_dictSelection_t selection); * smallest dictionary within a specified regression of the compressed size * from the largest dictionary. */ - COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent, size_t dictBufferCapacity, + COVER_dictSelection_t COVER_selectDict(BYTE* customDictContent, size_t dictBufferCapacity, size_t dictContentSize, const BYTE* samplesBuffer, const size_t* samplesSizes, unsigned nbFinalizeSamples, size_t nbCheckSamples, size_t nbSamples, ZDICT_cover_params_t params, size_t* offsets, size_t totalCompressedSize); diff --git a/contrib/libs/zstd/lib/dictBuilder/divsufsort.c b/contrib/libs/zstd/lib/dictBuilder/divsufsort.c index a2870fb3ba..f96e7aee3c 100644 --- a/contrib/libs/zstd/lib/dictBuilder/divsufsort.c +++ b/contrib/libs/zstd/lib/dictBuilder/divsufsort.c @@ -1576,7 +1576,7 @@ note: /* Construct the inverse suffix array of type B* suffixes using trsort. */ trsort(ISAb, SA, m, 1); - /* Set the sorted order of type B* suffixes. */ + /* Set the sorted order of type B* suffixes. */ for(i = n - 1, j = m, c0 = T[n - 1]; 0 <= i;) { for(--i, c1 = c0; (0 <= i) && ((c0 = T[i]) >= c1); --i, c1 = c0) { } if(0 <= i) { diff --git a/contrib/libs/zstd/lib/dictBuilder/fastcover.c b/contrib/libs/zstd/lib/dictBuilder/fastcover.c index 3352859ada..8fef0c9025 100644 --- a/contrib/libs/zstd/lib/dictBuilder/fastcover.c +++ b/contrib/libs/zstd/lib/dictBuilder/fastcover.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Facebook, Inc. + * Copyright (c) Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -16,17 +16,17 @@ #include <string.h> /* memset */ #include <time.h> /* clock */ -#ifndef ZDICT_STATIC_LINKING_ONLY -# define ZDICT_STATIC_LINKING_ONLY -#endif - +#ifndef ZDICT_STATIC_LINKING_ONLY +# define ZDICT_STATIC_LINKING_ONLY +#endif + #include "../common/mem.h" /* read */ #include "../common/pool.h" #include "../common/threading.h" #include "../common/zstd_internal.h" /* includes zstd.h */ -#include "../compress/zstd_compress_internal.h" /* ZSTD_hash*() */ -#include "../zdict.h" -#include "cover.h" +#include "../compress/zstd_compress_internal.h" /* ZSTD_hash*() */ +#include "../zdict.h" +#include "cover.h" /*-************************************* @@ -42,7 +42,7 @@ #define FASTCOVER_MAX_SAMPLES_SIZE (sizeof(size_t) == 8 ? ((unsigned)-1) : ((unsigned)1 GB)) #define FASTCOVER_MAX_F 31 #define FASTCOVER_MAX_ACCEL 10 -#define FASTCOVER_DEFAULT_SPLITPOINT 0.75 +#define FASTCOVER_DEFAULT_SPLITPOINT 0.75 #define DEFAULT_F 20 #define DEFAULT_ACCEL 1 @@ -50,36 +50,36 @@ /*-************************************* * Console display ***************************************/ -#ifndef LOCALDISPLAYLEVEL +#ifndef LOCALDISPLAYLEVEL static int g_displayLevel = 0; -#endif -#undef DISPLAY +#endif +#undef DISPLAY #define DISPLAY(...) \ { \ fprintf(stderr, __VA_ARGS__); \ fflush(stderr); \ } -#undef LOCALDISPLAYLEVEL +#undef LOCALDISPLAYLEVEL #define LOCALDISPLAYLEVEL(displayLevel, l, ...) \ if (displayLevel >= l) { \ DISPLAY(__VA_ARGS__); \ } /* 0 : no display; 1: errors; 2: default; 3: details; 4: debug */ -#undef DISPLAYLEVEL +#undef DISPLAYLEVEL #define DISPLAYLEVEL(l, ...) LOCALDISPLAYLEVEL(g_displayLevel, l, __VA_ARGS__) -#ifndef LOCALDISPLAYUPDATE -static const clock_t g_refreshRate = CLOCKS_PER_SEC * 15 / 100; -static clock_t g_time = 0; -#endif -#undef LOCALDISPLAYUPDATE +#ifndef LOCALDISPLAYUPDATE +static const clock_t g_refreshRate = CLOCKS_PER_SEC * 15 / 100; +static clock_t g_time = 0; +#endif +#undef LOCALDISPLAYUPDATE #define LOCALDISPLAYUPDATE(displayLevel, l, ...) \ if (displayLevel >= l) { \ - if ((clock() - g_time > g_refreshRate) || (displayLevel >= 4)) { \ + if ((clock() - g_time > g_refreshRate) || (displayLevel >= 4)) { \ g_time = clock(); \ DISPLAY(__VA_ARGS__); \ } \ } -#undef DISPLAYUPDATE +#undef DISPLAYUPDATE #define DISPLAYUPDATE(l, ...) LOCALDISPLAYUPDATE(g_displayLevel, l, __VA_ARGS__) @@ -87,13 +87,13 @@ static clock_t g_time = 0; * Hash Functions ***************************************/ /** - * Hash the d-byte value pointed to by p and mod 2^f into the frequency vector + * Hash the d-byte value pointed to by p and mod 2^f into the frequency vector */ -static size_t FASTCOVER_hashPtrToIndex(const void* p, U32 f, unsigned d) { +static size_t FASTCOVER_hashPtrToIndex(const void* p, U32 f, unsigned d) { if (d == 6) { - return ZSTD_hash6Ptr(p, f); + return ZSTD_hash6Ptr(p, f); } - return ZSTD_hash8Ptr(p, f); + return ZSTD_hash8Ptr(p, f); } @@ -470,20 +470,20 @@ typedef struct FASTCOVER_tryParameters_data_s { * This function is thread safe if zstd is compiled with multithreaded support. * It takes its parameters as an *OWNING* opaque pointer to support threading. */ -static void FASTCOVER_tryParameters(void* opaque) +static void FASTCOVER_tryParameters(void* opaque) { /* Save parameters as local variables */ - FASTCOVER_tryParameters_data_t *const data = (FASTCOVER_tryParameters_data_t*)opaque; + FASTCOVER_tryParameters_data_t *const data = (FASTCOVER_tryParameters_data_t*)opaque; const FASTCOVER_ctx_t *const ctx = data->ctx; const ZDICT_cover_params_t parameters = data->parameters; size_t dictBufferCapacity = data->dictBufferCapacity; size_t totalCompressedSize = ERROR(GENERIC); /* Initialize array to keep track of frequency of dmer within activeSegment */ - U16* segmentFreqs = (U16*)calloc(((U64)1 << ctx->f), sizeof(U16)); + U16* segmentFreqs = (U16*)calloc(((U64)1 << ctx->f), sizeof(U16)); /* Allocate space for hash table, dict, and freqs */ - BYTE *const dict = (BYTE*)malloc(dictBufferCapacity); + BYTE *const dict = (BYTE*)malloc(dictBufferCapacity); COVER_dictSelection_t selection = COVER_dictSelectionError(ERROR(GENERIC)); - U32* freqs = (U32*) malloc(((U64)1 << ctx->f) * sizeof(U32)); + U32* freqs = (U32*) malloc(((U64)1 << ctx->f) * sizeof(U32)); if (!segmentFreqs || !dict || !freqs) { DISPLAYLEVEL(1, "Failed to allocate buffers: out of memory\n"); goto _cleanup; @@ -495,7 +495,7 @@ static void FASTCOVER_tryParameters(void* opaque) parameters, segmentFreqs); const unsigned nbFinalizeSamples = (unsigned)(ctx->nbTrainSamples * ctx->accelParams.finalize / 100); - selection = COVER_selectDict(dict + tail, dictBufferCapacity, dictBufferCapacity - tail, + selection = COVER_selectDict(dict + tail, dictBufferCapacity, dictBufferCapacity - tail, ctx->samples, ctx->samplesSizes, nbFinalizeSamples, ctx->nbTrainSamples, ctx->nbSamples, parameters, ctx->offsets, totalCompressedSize); @@ -626,7 +626,7 @@ ZDICT_optimizeTrainFromBuffer_fastCover( /* constants */ const unsigned nbThreads = parameters->nbThreads; const double splitPoint = - parameters->splitPoint <= 0.0 ? FASTCOVER_DEFAULT_SPLITPOINT : parameters->splitPoint; + parameters->splitPoint <= 0.0 ? FASTCOVER_DEFAULT_SPLITPOINT : parameters->splitPoint; const unsigned kMinD = parameters->d == 0 ? 6 : parameters->d; const unsigned kMaxD = parameters->d == 0 ? 8 : parameters->d; const unsigned kMinK = parameters->k == 0 ? 50 : parameters->k; diff --git a/contrib/libs/zstd/lib/dictBuilder/zdict.c b/contrib/libs/zstd/lib/dictBuilder/zdict.c index 587df6b861..ac23934464 100644 --- a/contrib/libs/zstd/lib/dictBuilder/zdict.c +++ b/contrib/libs/zstd/lib/dictBuilder/zdict.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -23,13 +23,13 @@ /* Unix Large Files support (>4GB) */ #define _FILE_OFFSET_BITS 64 #if (defined(__sun__) && (!defined(__LP64__))) /* Sun Solaris 32-bits requires specific definitions */ -# ifndef _LARGEFILE_SOURCE +# ifndef _LARGEFILE_SOURCE # define _LARGEFILE_SOURCE -# endif +# endif #elif ! defined(__LP64__) /* No point defining Large file for 64 bit */ -# ifndef _LARGEFILE64_SOURCE +# ifndef _LARGEFILE64_SOURCE # define _LARGEFILE64_SOURCE -# endif +# endif #endif @@ -41,18 +41,18 @@ #include <stdio.h> /* fprintf, fopen, ftello64 */ #include <time.h> /* clock */ -#ifndef ZDICT_STATIC_LINKING_ONLY -# define ZDICT_STATIC_LINKING_ONLY -#endif -#define HUF_STATIC_LINKING_ONLY - +#ifndef ZDICT_STATIC_LINKING_ONLY +# define ZDICT_STATIC_LINKING_ONLY +#endif +#define HUF_STATIC_LINKING_ONLY + #include "../common/mem.h" /* read */ #include "../common/fse.h" /* FSE_normalizeCount, FSE_writeNCount */ #include "../common/huf.h" /* HUF_buildCTable, HUF_writeCTable */ #include "../common/zstd_internal.h" /* includes zstd.h */ #include <contrib/libs/xxhash/xxhash.h> /* XXH64 */ -#include "../compress/zstd_compress_internal.h" /* ZSTD_loadCEntropy() */ -#include "../zdict.h" +#include "../compress/zstd_compress_internal.h" /* ZSTD_loadCEntropy() */ +#include "../zdict.h" #include "divsufsort.h" @@ -73,9 +73,9 @@ static const U32 g_selectivity_default = 9; /*-************************************* * Console display ***************************************/ -#undef DISPLAY +#undef DISPLAY #define DISPLAY(...) { fprintf(stderr, __VA_ARGS__); fflush( stderr ); } -#undef DISPLAYLEVEL +#undef DISPLAYLEVEL #define DISPLAYLEVEL(l, ...) if (notificationLevel>=l) { DISPLAY(__VA_ARGS__); } /* 0 : no display; 1: errors; 2: default; 3: details; 4: debug */ static clock_t ZDICT_clockSpan(clock_t nPrevious) { return clock() - nPrevious; } @@ -111,13 +111,13 @@ size_t ZDICT_getDictHeaderSize(const void* dictBuffer, size_t dictSize) size_t headerSize; if (dictSize <= 8 || MEM_readLE32(dictBuffer) != ZSTD_MAGIC_DICTIONARY) return ERROR(dictionary_corrupted); - { ZSTD_compressedBlockState_t* bs = (ZSTD_compressedBlockState_t*)malloc(sizeof(ZSTD_compressedBlockState_t)); + { ZSTD_compressedBlockState_t* bs = (ZSTD_compressedBlockState_t*)malloc(sizeof(ZSTD_compressedBlockState_t)); U32* wksp = (U32*)malloc(HUF_WORKSPACE_SIZE); - if (!bs || !wksp) { + if (!bs || !wksp) { headerSize = ERROR(memory_allocation); } else { ZSTD_reset_compressedBlockState(bs); - headerSize = ZSTD_loadCEntropy(bs, wksp, dictBuffer, dictSize); + headerSize = ZSTD_loadCEntropy(bs, wksp, dictBuffer, dictSize); } free(bs); @@ -555,7 +555,7 @@ static size_t ZDICT_trainBuffer_legacy(dictItem* dictList, U32 dictListSize, clock_t displayClock = 0; clock_t const refreshRate = CLOCKS_PER_SEC * 3 / 10; -# undef DISPLAYUPDATE +# undef DISPLAYUPDATE # define DISPLAYUPDATE(l, ...) if (notificationLevel>=l) { \ if (ZDICT_clockSpan(displayClock) > refreshRate) \ { displayClock = clock(); DISPLAY(__VA_ARGS__); \ @@ -730,7 +730,7 @@ static void ZDICT_flatLit(unsigned* countLit) #define OFFCODE_MAX 30 /* only applicable to first block */ static size_t ZDICT_analyzeEntropy(void* dstBuffer, size_t maxDstSize, - int compressionLevel, + int compressionLevel, const void* srcBuffer, const size_t* fileSizes, unsigned nbFiles, const void* dictBuffer, size_t dictBufferSize, unsigned notificationLevel) @@ -765,7 +765,7 @@ static size_t ZDICT_analyzeEntropy(void* dstBuffer, size_t maxDstSize, memset(repOffset, 0, sizeof(repOffset)); repOffset[1] = repOffset[4] = repOffset[8] = 1; memset(bestRepOffset, 0, sizeof(bestRepOffset)); - if (compressionLevel==0) compressionLevel = ZSTD_CLEVEL_DEFAULT; + if (compressionLevel==0) compressionLevel = ZSTD_CLEVEL_DEFAULT; params = ZSTD_getParams(compressionLevel, averageSampleSize, dictBufferSize); esr.dict = ZSTD_createCDict_advanced(dictBuffer, dictBufferSize, ZSTD_dlm_byRef, ZSTD_dct_rawContent, params.cParams, ZSTD_defaultCMem); @@ -817,7 +817,7 @@ static size_t ZDICT_analyzeEntropy(void* dstBuffer, size_t maxDstSize, /* note : the result of this phase should be used to better appreciate the impact on statistics */ total=0; for (u=0; u<=offcodeMax; u++) total+=offcodeCount[u]; - errorCode = FSE_normalizeCount(offcodeNCount, Offlog, offcodeCount, total, offcodeMax, /* useLowProbCount */ 1); + errorCode = FSE_normalizeCount(offcodeNCount, Offlog, offcodeCount, total, offcodeMax, /* useLowProbCount */ 1); if (FSE_isError(errorCode)) { eSize = errorCode; DISPLAYLEVEL(1, "FSE_normalizeCount error with offcodeCount \n"); @@ -826,7 +826,7 @@ static size_t ZDICT_analyzeEntropy(void* dstBuffer, size_t maxDstSize, Offlog = (U32)errorCode; total=0; for (u=0; u<=MaxML; u++) total+=matchLengthCount[u]; - errorCode = FSE_normalizeCount(matchLengthNCount, mlLog, matchLengthCount, total, MaxML, /* useLowProbCount */ 1); + errorCode = FSE_normalizeCount(matchLengthNCount, mlLog, matchLengthCount, total, MaxML, /* useLowProbCount */ 1); if (FSE_isError(errorCode)) { eSize = errorCode; DISPLAYLEVEL(1, "FSE_normalizeCount error with matchLengthCount \n"); @@ -835,7 +835,7 @@ static size_t ZDICT_analyzeEntropy(void* dstBuffer, size_t maxDstSize, mlLog = (U32)errorCode; total=0; for (u=0; u<=MaxLL; u++) total+=litLengthCount[u]; - errorCode = FSE_normalizeCount(litLengthNCount, llLog, litLengthCount, total, MaxLL, /* useLowProbCount */ 1); + errorCode = FSE_normalizeCount(litLengthNCount, llLog, litLengthCount, total, MaxLL, /* useLowProbCount */ 1); if (FSE_isError(errorCode)) { eSize = errorCode; DISPLAYLEVEL(1, "FSE_normalizeCount error with litLengthCount \n"); @@ -935,7 +935,7 @@ size_t ZDICT_finalizeDictionary(void* dictBuffer, size_t dictBufferCapacity, size_t hSize; #define HBUFFSIZE 256 /* should prove large enough for all entropy headers */ BYTE header[HBUFFSIZE]; - int const compressionLevel = (params.compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : params.compressionLevel; + int const compressionLevel = (params.compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : params.compressionLevel; U32 const notificationLevel = params.notificationLevel; /* The final dictionary content must be at least as large as the largest repcode */ size_t const minContentSize = (size_t)ZDICT_maxRep(repStartValue); @@ -1014,7 +1014,7 @@ static size_t ZDICT_addEntropyTablesFromBuffer_advanced( const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples, ZDICT_params_t params) { - int const compressionLevel = (params.compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : params.compressionLevel; + int const compressionLevel = (params.compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : params.compressionLevel; U32 const notificationLevel = params.notificationLevel; size_t hSize = 8; @@ -1044,10 +1044,10 @@ static size_t ZDICT_addEntropyTablesFromBuffer_advanced( } /*! ZDICT_trainFromBuffer_unsafe_legacy() : -* Warning : `samplesBuffer` must be followed by noisy guard band !!! +* Warning : `samplesBuffer` must be followed by noisy guard band !!! * @return : size of dictionary, or an error code which can be tested with ZDICT_isError() */ -static size_t ZDICT_trainFromBuffer_unsafe_legacy( +static size_t ZDICT_trainFromBuffer_unsafe_legacy( void* dictBuffer, size_t maxDictSize, const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples, ZDICT_legacy_params_t params) @@ -1184,8 +1184,8 @@ size_t ZDICT_trainFromBuffer(void* dictBuffer, size_t dictBufferCapacity, memset(¶ms, 0, sizeof(params)); params.d = 8; params.steps = 4; - /* Use default level since no compression level information is available */ - params.zParams.compressionLevel = ZSTD_CLEVEL_DEFAULT; + /* Use default level since no compression level information is available */ + params.zParams.compressionLevel = ZSTD_CLEVEL_DEFAULT; #if defined(DEBUGLEVEL) && (DEBUGLEVEL>=1) params.zParams.notificationLevel = DEBUGLEVEL; #endif diff --git a/contrib/libs/zstd/lib/legacy/zstd_legacy.h b/contrib/libs/zstd/lib/legacy/zstd_legacy.h index a6f1174b82..e21ba5cbe0 100644 --- a/contrib/libs/zstd/lib/legacy/zstd_legacy.h +++ b/contrib/libs/zstd/lib/legacy/zstd_legacy.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the diff --git a/contrib/libs/zstd/lib/legacy/zstd_v01.c b/contrib/libs/zstd/lib/legacy/zstd_v01.c index 23caaef564..3e272675e4 100644 --- a/contrib/libs/zstd/lib/legacy/zstd_v01.c +++ b/contrib/libs/zstd/lib/legacy/zstd_v01.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -1276,11 +1276,11 @@ static size_t HUF_decompress (void* dst, size_t maxDstSize, const void* cSrc, si * Basic Types *********************************************************/ #if defined (__STDC_VERSION__) && __STDC_VERSION__ >= 199901L /* C99 */ -# if defined(_AIX) -# include <inttypes.h> -# else -# include <stdint.h> /* intptr_t */ -# endif +# if defined(_AIX) +# include <inttypes.h> +# else +# include <stdint.h> /* intptr_t */ +# endif typedef uint8_t BYTE; typedef uint16_t U16; typedef int16_t S16; diff --git a/contrib/libs/zstd/lib/legacy/zstd_v01.h b/contrib/libs/zstd/lib/legacy/zstd_v01.h index f777eb6e4c..bc2c5e641d 100644 --- a/contrib/libs/zstd/lib/legacy/zstd_v01.h +++ b/contrib/libs/zstd/lib/legacy/zstd_v01.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the diff --git a/contrib/libs/zstd/lib/legacy/zstd_v02.c b/contrib/libs/zstd/lib/legacy/zstd_v02.c index 2f473a7573..1d6e908657 100644 --- a/contrib/libs/zstd/lib/legacy/zstd_v02.c +++ b/contrib/libs/zstd/lib/legacy/zstd_v02.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -89,11 +89,11 @@ extern "C" { * Basic Types *****************************************************************/ #if defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) -# if defined(_AIX) -# include <inttypes.h> -# else -# include <stdint.h> /* intptr_t */ -# endif +# if defined(_AIX) +# include <inttypes.h> +# else +# include <stdint.h> /* intptr_t */ +# endif typedef uint8_t BYTE; typedef uint16_t U16; typedef int16_t S16; diff --git a/contrib/libs/zstd/lib/legacy/zstd_v02.h b/contrib/libs/zstd/lib/legacy/zstd_v02.h index 1b371953b7..8113f5c014 100644 --- a/contrib/libs/zstd/lib/legacy/zstd_v02.h +++ b/contrib/libs/zstd/lib/legacy/zstd_v02.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the diff --git a/contrib/libs/zstd/lib/legacy/zstd_v03.c b/contrib/libs/zstd/lib/legacy/zstd_v03.c index 6625f4df1c..5e7eaaa007 100644 --- a/contrib/libs/zstd/lib/legacy/zstd_v03.c +++ b/contrib/libs/zstd/lib/legacy/zstd_v03.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -90,11 +90,11 @@ extern "C" { * Basic Types *****************************************************************/ #if defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) -# if defined(_AIX) -# include <inttypes.h> -# else -# include <stdint.h> /* intptr_t */ -# endif +# if defined(_AIX) +# include <inttypes.h> +# else +# include <stdint.h> /* intptr_t */ +# endif typedef uint8_t BYTE; typedef uint16_t U16; typedef int16_t S16; diff --git a/contrib/libs/zstd/lib/legacy/zstd_v03.h b/contrib/libs/zstd/lib/legacy/zstd_v03.h index 7a00d4304b..509baa77ba 100644 --- a/contrib/libs/zstd/lib/legacy/zstd_v03.h +++ b/contrib/libs/zstd/lib/legacy/zstd_v03.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the diff --git a/contrib/libs/zstd/lib/legacy/zstd_v04.c b/contrib/libs/zstd/lib/legacy/zstd_v04.c index 8d305c7eae..795b0da53b 100644 --- a/contrib/libs/zstd/lib/legacy/zstd_v04.c +++ b/contrib/libs/zstd/lib/legacy/zstd_v04.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -52,11 +52,11 @@ extern "C" { * Basic Types *****************************************************************/ #if defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) -# if defined(_AIX) -# include <inttypes.h> -# else -# include <stdint.h> /* intptr_t */ -# endif +# if defined(_AIX) +# include <inttypes.h> +# else +# include <stdint.h> /* intptr_t */ +# endif typedef uint8_t BYTE; typedef uint16_t U16; typedef int16_t S16; diff --git a/contrib/libs/zstd/lib/legacy/zstd_v04.h b/contrib/libs/zstd/lib/legacy/zstd_v04.h index 66b97ab8e6..49cc5c7f4c 100644 --- a/contrib/libs/zstd/lib/legacy/zstd_v04.h +++ b/contrib/libs/zstd/lib/legacy/zstd_v04.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the diff --git a/contrib/libs/zstd/lib/legacy/zstd_v05.c b/contrib/libs/zstd/lib/legacy/zstd_v05.c index 795dfb410c..d3cbc6f6f3 100644 --- a/contrib/libs/zstd/lib/legacy/zstd_v05.c +++ b/contrib/libs/zstd/lib/legacy/zstd_v05.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -80,11 +80,11 @@ extern "C" { * Basic Types *****************************************************************/ #if defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) -# if defined(_AIX) -# include <inttypes.h> -# else -# include <stdint.h> /* intptr_t */ -# endif +# if defined(_AIX) +# include <inttypes.h> +# else +# include <stdint.h> /* intptr_t */ +# endif typedef uint8_t BYTE; typedef uint16_t U16; typedef int16_t S16; @@ -2829,7 +2829,7 @@ static size_t ZSTDv05_decodeFrameHeader_Part2(ZSTDv05_DCtx* zc, const void* src, static size_t ZSTDv05_getcBlockSize(const void* src, size_t srcSize, blockProperties_t* bpPtr) { - const BYTE* const in = (const BYTE*)src; + const BYTE* const in = (const BYTE*)src; BYTE headerFlags; U32 cSize; @@ -2998,7 +2998,7 @@ static size_t ZSTDv05_decodeSeqHeaders(int* nbSeq, const BYTE** dumpsPtr, size_t FSEv05_DTable* DTableLL, FSEv05_DTable* DTableML, FSEv05_DTable* DTableOffb, const void* src, size_t srcSize, U32 flagStaticTable) { - const BYTE* const istart = (const BYTE*)src; + const BYTE* const istart = (const BYTE*)src; const BYTE* ip = istart; const BYTE* const iend = istart + srcSize; U32 LLtype, Offtype, MLtype; @@ -3306,7 +3306,7 @@ static size_t ZSTDv05_decompressSequences( { const BYTE* ip = (const BYTE*)seqStart; const BYTE* const iend = ip + seqSize; - BYTE* const ostart = (BYTE*)dst; + BYTE* const ostart = (BYTE*)dst; BYTE* op = ostart; BYTE* const oend = ostart + maxDstSize; size_t errorCode, dumpsLength=0; @@ -3419,7 +3419,7 @@ static size_t ZSTDv05_decompress_continueDCtx(ZSTDv05_DCtx* dctx, { const BYTE* ip = (const BYTE*)src; const BYTE* iend = ip + srcSize; - BYTE* const ostart = (BYTE*)dst; + BYTE* const ostart = (BYTE*)dst; BYTE* op = ostart; BYTE* const oend = ostart + maxDstSize; size_t remainingSize = srcSize; diff --git a/contrib/libs/zstd/lib/legacy/zstd_v05.h b/contrib/libs/zstd/lib/legacy/zstd_v05.h index bd423bfc1b..dba2b1b28f 100644 --- a/contrib/libs/zstd/lib/legacy/zstd_v05.h +++ b/contrib/libs/zstd/lib/legacy/zstd_v05.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the diff --git a/contrib/libs/zstd/lib/legacy/zstd_v06.c b/contrib/libs/zstd/lib/legacy/zstd_v06.c index ead213c484..46983600e9 100644 --- a/contrib/libs/zstd/lib/legacy/zstd_v06.c +++ b/contrib/libs/zstd/lib/legacy/zstd_v06.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -82,11 +82,11 @@ extern "C" { * Basic Types *****************************************************************/ #if !defined (__VMS) && (defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) ) -# if defined(_AIX) -# include <inttypes.h> -# else -# include <stdint.h> /* intptr_t */ -# endif +# if defined(_AIX) +# include <inttypes.h> +# else +# include <stdint.h> /* intptr_t */ +# endif typedef uint8_t BYTE; typedef uint16_t U16; typedef int16_t S16; @@ -3025,7 +3025,7 @@ typedef struct * Provides the size of compressed block from block header `src` */ static size_t ZSTDv06_getcBlockSize(const void* src, size_t srcSize, blockProperties_t* bpPtr) { - const BYTE* const in = (const BYTE*)src; + const BYTE* const in = (const BYTE*)src; U32 cSize; if (srcSize < ZSTDv06_blockHeaderSize) return ERROR(srcSize_wrong); @@ -3219,7 +3219,7 @@ static size_t ZSTDv06_decodeSeqHeaders(int* nbSeqPtr, FSEv06_DTable* DTableLL, FSEv06_DTable* DTableML, FSEv06_DTable* DTableOffb, U32 flagRepeatTable, const void* src, size_t srcSize) { - const BYTE* const istart = (const BYTE*)src; + const BYTE* const istart = (const BYTE*)src; const BYTE* const iend = istart + srcSize; const BYTE* ip = istart; @@ -3441,7 +3441,7 @@ static size_t ZSTDv06_decompressSequences( { const BYTE* ip = (const BYTE*)seqStart; const BYTE* const iend = ip + seqSize; - BYTE* const ostart = (BYTE*)dst; + BYTE* const ostart = (BYTE*)dst; BYTE* const oend = ostart + maxDstSize; BYTE* op = ostart; const BYTE* litPtr = dctx->litPtr; @@ -3557,7 +3557,7 @@ static size_t ZSTDv06_decompressFrame(ZSTDv06_DCtx* dctx, { const BYTE* ip = (const BYTE*)src; const BYTE* const iend = ip + srcSize; - BYTE* const ostart = (BYTE*)dst; + BYTE* const ostart = (BYTE*)dst; BYTE* op = ostart; BYTE* const oend = ostart + dstCapacity; size_t remainingSize = srcSize; diff --git a/contrib/libs/zstd/lib/legacy/zstd_v06.h b/contrib/libs/zstd/lib/legacy/zstd_v06.h index 9e32b76e08..50c0267a85 100644 --- a/contrib/libs/zstd/lib/legacy/zstd_v06.h +++ b/contrib/libs/zstd/lib/legacy/zstd_v06.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the diff --git a/contrib/libs/zstd/lib/legacy/zstd_v07.c b/contrib/libs/zstd/lib/legacy/zstd_v07.c index 189f6ede69..591e0119d6 100644 --- a/contrib/libs/zstd/lib/legacy/zstd_v07.c +++ b/contrib/libs/zstd/lib/legacy/zstd_v07.c @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -242,11 +242,11 @@ extern "C" { * Basic Types *****************************************************************/ #if !defined (__VMS) && (defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) ) -# if defined(_AIX) -# include <inttypes.h> -# else -# include <stdint.h> /* intptr_t */ -# endif +# if defined(_AIX) +# include <inttypes.h> +# else +# include <stdint.h> /* intptr_t */ +# endif typedef uint8_t BYTE; typedef uint16_t U16; typedef int16_t S16; @@ -3254,7 +3254,7 @@ typedef struct * Provides the size of compressed block from block header `src` */ static size_t ZSTDv07_getcBlockSize(const void* src, size_t srcSize, blockProperties_t* bpPtr) { - const BYTE* const in = (const BYTE*)src; + const BYTE* const in = (const BYTE*)src; U32 cSize; if (srcSize < ZSTDv07_blockHeaderSize) return ERROR(srcSize_wrong); @@ -3449,7 +3449,7 @@ static size_t ZSTDv07_decodeSeqHeaders(int* nbSeqPtr, FSEv07_DTable* DTableLL, FSEv07_DTable* DTableML, FSEv07_DTable* DTableOffb, U32 flagRepeatTable, const void* src, size_t srcSize) { - const BYTE* const istart = (const BYTE*)src; + const BYTE* const istart = (const BYTE*)src; const BYTE* const iend = istart + srcSize; const BYTE* ip = istart; @@ -3668,7 +3668,7 @@ static size_t ZSTDv07_decompressSequences( { const BYTE* ip = (const BYTE*)seqStart; const BYTE* const iend = ip + seqSize; - BYTE* const ostart = (BYTE*)dst; + BYTE* const ostart = (BYTE*)dst; BYTE* const oend = ostart + maxDstSize; BYTE* op = ostart; const BYTE* litPtr = dctx->litPtr; @@ -3795,7 +3795,7 @@ static size_t ZSTDv07_decompressFrame(ZSTDv07_DCtx* dctx, { const BYTE* ip = (const BYTE*)src; const BYTE* const iend = ip + srcSize; - BYTE* const ostart = (BYTE*)dst; + BYTE* const ostart = (BYTE*)dst; BYTE* const oend = ostart + dstCapacity; BYTE* op = ostart; size_t remainingSize = srcSize; diff --git a/contrib/libs/zstd/lib/legacy/zstd_v07.h b/contrib/libs/zstd/lib/legacy/zstd_v07.h index bc35cfa6a3..baa3b3a5f7 100644 --- a/contrib/libs/zstd/lib/legacy/zstd_v07.h +++ b/contrib/libs/zstd/lib/legacy/zstd_v07.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the diff --git a/contrib/libs/zstd/lib/zdict.h b/contrib/libs/zstd/lib/zdict.h index f1e139a40d..cce8285b77 100644 --- a/contrib/libs/zstd/lib/zdict.h +++ b/contrib/libs/zstd/lib/zdict.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -36,146 +36,146 @@ extern "C" { # define ZDICTLIB_API ZDICTLIB_VISIBILITY #endif -/******************************************************************************* - * Zstd dictionary builder - * - * FAQ - * === - * Why should I use a dictionary? - * ------------------------------ - * - * Zstd can use dictionaries to improve compression ratio of small data. - * Traditionally small files don't compress well because there is very little +/******************************************************************************* + * Zstd dictionary builder + * + * FAQ + * === + * Why should I use a dictionary? + * ------------------------------ + * + * Zstd can use dictionaries to improve compression ratio of small data. + * Traditionally small files don't compress well because there is very little * repetition in a single sample, since it is small. But, if you are compressing - * many similar files, like a bunch of JSON records that share the same - * structure, you can train a dictionary on ahead of time on some samples of - * these files. Then, zstd can use the dictionary to find repetitions that are - * present across samples. This can vastly improve compression ratio. - * - * When is a dictionary useful? - * ---------------------------- - * - * Dictionaries are useful when compressing many small files that are similar. - * The larger a file is, the less benefit a dictionary will have. Generally, - * we don't expect dictionary compression to be effective past 100KB. And the - * smaller a file is, the more we would expect the dictionary to help. - * - * How do I use a dictionary? - * -------------------------- - * - * Simply pass the dictionary to the zstd compressor with - * `ZSTD_CCtx_loadDictionary()`. The same dictionary must then be passed to - * the decompressor, using `ZSTD_DCtx_loadDictionary()`. There are other - * more advanced functions that allow selecting some options, see zstd.h for - * complete documentation. - * - * What is a zstd dictionary? - * -------------------------- - * - * A zstd dictionary has two pieces: Its header, and its content. The header - * contains a magic number, the dictionary ID, and entropy tables. These - * entropy tables allow zstd to save on header costs in the compressed file, - * which really matters for small data. The content is just bytes, which are - * repeated content that is common across many samples. - * - * What is a raw content dictionary? - * --------------------------------- - * - * A raw content dictionary is just bytes. It doesn't have a zstd dictionary - * header, a dictionary ID, or entropy tables. Any buffer is a valid raw - * content dictionary. - * - * How do I train a dictionary? - * ---------------------------- - * - * Gather samples from your use case. These samples should be similar to each - * other. If you have several use cases, you could try to train one dictionary - * per use case. - * - * Pass those samples to `ZDICT_trainFromBuffer()` and that will train your - * dictionary. There are a few advanced versions of this function, but this - * is a great starting point. If you want to further tune your dictionary - * you could try `ZDICT_optimizeTrainFromBuffer_cover()`. If that is too slow - * you can try `ZDICT_optimizeTrainFromBuffer_fastCover()`. - * - * If the dictionary training function fails, that is likely because you - * either passed too few samples, or a dictionary would not be effective - * for your data. Look at the messages that the dictionary trainer printed, - * if it doesn't say too few samples, then a dictionary would not be effective. - * - * How large should my dictionary be? - * ---------------------------------- - * - * A reasonable dictionary size, the `dictBufferCapacity`, is about 100KB. - * The zstd CLI defaults to a 110KB dictionary. You likely don't need a - * dictionary larger than that. But, most use cases can get away with a - * smaller dictionary. The advanced dictionary builders can automatically - * shrink the dictionary for you, and select a the smallest size that - * doesn't hurt compression ratio too much. See the `shrinkDict` parameter. - * A smaller dictionary can save memory, and potentially speed up - * compression. - * - * How many samples should I provide to the dictionary builder? - * ------------------------------------------------------------ - * - * We generally recommend passing ~100x the size of the dictionary - * in samples. A few thousand should suffice. Having too few samples - * can hurt the dictionaries effectiveness. Having more samples will - * only improve the dictionaries effectiveness. But having too many - * samples can slow down the dictionary builder. - * - * How do I determine if a dictionary will be effective? - * ----------------------------------------------------- - * - * Simply train a dictionary and try it out. You can use zstd's built in - * benchmarking tool to test the dictionary effectiveness. - * - * # Benchmark levels 1-3 without a dictionary - * zstd -b1e3 -r /path/to/my/files + * many similar files, like a bunch of JSON records that share the same + * structure, you can train a dictionary on ahead of time on some samples of + * these files. Then, zstd can use the dictionary to find repetitions that are + * present across samples. This can vastly improve compression ratio. + * + * When is a dictionary useful? + * ---------------------------- + * + * Dictionaries are useful when compressing many small files that are similar. + * The larger a file is, the less benefit a dictionary will have. Generally, + * we don't expect dictionary compression to be effective past 100KB. And the + * smaller a file is, the more we would expect the dictionary to help. + * + * How do I use a dictionary? + * -------------------------- + * + * Simply pass the dictionary to the zstd compressor with + * `ZSTD_CCtx_loadDictionary()`. The same dictionary must then be passed to + * the decompressor, using `ZSTD_DCtx_loadDictionary()`. There are other + * more advanced functions that allow selecting some options, see zstd.h for + * complete documentation. + * + * What is a zstd dictionary? + * -------------------------- + * + * A zstd dictionary has two pieces: Its header, and its content. The header + * contains a magic number, the dictionary ID, and entropy tables. These + * entropy tables allow zstd to save on header costs in the compressed file, + * which really matters for small data. The content is just bytes, which are + * repeated content that is common across many samples. + * + * What is a raw content dictionary? + * --------------------------------- + * + * A raw content dictionary is just bytes. It doesn't have a zstd dictionary + * header, a dictionary ID, or entropy tables. Any buffer is a valid raw + * content dictionary. + * + * How do I train a dictionary? + * ---------------------------- + * + * Gather samples from your use case. These samples should be similar to each + * other. If you have several use cases, you could try to train one dictionary + * per use case. + * + * Pass those samples to `ZDICT_trainFromBuffer()` and that will train your + * dictionary. There are a few advanced versions of this function, but this + * is a great starting point. If you want to further tune your dictionary + * you could try `ZDICT_optimizeTrainFromBuffer_cover()`. If that is too slow + * you can try `ZDICT_optimizeTrainFromBuffer_fastCover()`. + * + * If the dictionary training function fails, that is likely because you + * either passed too few samples, or a dictionary would not be effective + * for your data. Look at the messages that the dictionary trainer printed, + * if it doesn't say too few samples, then a dictionary would not be effective. + * + * How large should my dictionary be? + * ---------------------------------- + * + * A reasonable dictionary size, the `dictBufferCapacity`, is about 100KB. + * The zstd CLI defaults to a 110KB dictionary. You likely don't need a + * dictionary larger than that. But, most use cases can get away with a + * smaller dictionary. The advanced dictionary builders can automatically + * shrink the dictionary for you, and select a the smallest size that + * doesn't hurt compression ratio too much. See the `shrinkDict` parameter. + * A smaller dictionary can save memory, and potentially speed up + * compression. + * + * How many samples should I provide to the dictionary builder? + * ------------------------------------------------------------ + * + * We generally recommend passing ~100x the size of the dictionary + * in samples. A few thousand should suffice. Having too few samples + * can hurt the dictionaries effectiveness. Having more samples will + * only improve the dictionaries effectiveness. But having too many + * samples can slow down the dictionary builder. + * + * How do I determine if a dictionary will be effective? + * ----------------------------------------------------- + * + * Simply train a dictionary and try it out. You can use zstd's built in + * benchmarking tool to test the dictionary effectiveness. + * + * # Benchmark levels 1-3 without a dictionary + * zstd -b1e3 -r /path/to/my/files * # Benchmark levels 1-3 with a dictionary - * zstd -b1e3 -r /path/to/my/files -D /path/to/my/dictionary - * - * When should I retrain a dictionary? - * ----------------------------------- - * - * You should retrain a dictionary when its effectiveness drops. Dictionary - * effectiveness drops as the data you are compressing changes. Generally, we do - * expect dictionaries to "decay" over time, as your data changes, but the rate - * at which they decay depends on your use case. Internally, we regularly - * retrain dictionaries, and if the new dictionary performs significantly - * better than the old dictionary, we will ship the new dictionary. - * - * I have a raw content dictionary, how do I turn it into a zstd dictionary? - * ------------------------------------------------------------------------- - * - * If you have a raw content dictionary, e.g. by manually constructing it, or - * using a third-party dictionary builder, you can turn it into a zstd - * dictionary by using `ZDICT_finalizeDictionary()`. You'll also have to - * provide some samples of the data. It will add the zstd header to the - * raw content, which contains a dictionary ID and entropy tables, which - * will improve compression ratio, and allow zstd to write the dictionary ID - * into the frame, if you so choose. - * - * Do I have to use zstd's dictionary builder? - * ------------------------------------------- - * - * No! You can construct dictionary content however you please, it is just - * bytes. It will always be valid as a raw content dictionary. If you want - * a zstd dictionary, which can improve compression ratio, use - * `ZDICT_finalizeDictionary()`. - * - * What is the attack surface of a zstd dictionary? - * ------------------------------------------------ - * - * Zstd is heavily fuzz tested, including loading fuzzed dictionaries, so - * zstd should never crash, or access out-of-bounds memory no matter what - * the dictionary is. However, if an attacker can control the dictionary - * during decompression, they can cause zstd to generate arbitrary bytes, - * just like if they controlled the compressed data. - * - ******************************************************************************/ - - + * zstd -b1e3 -r /path/to/my/files -D /path/to/my/dictionary + * + * When should I retrain a dictionary? + * ----------------------------------- + * + * You should retrain a dictionary when its effectiveness drops. Dictionary + * effectiveness drops as the data you are compressing changes. Generally, we do + * expect dictionaries to "decay" over time, as your data changes, but the rate + * at which they decay depends on your use case. Internally, we regularly + * retrain dictionaries, and if the new dictionary performs significantly + * better than the old dictionary, we will ship the new dictionary. + * + * I have a raw content dictionary, how do I turn it into a zstd dictionary? + * ------------------------------------------------------------------------- + * + * If you have a raw content dictionary, e.g. by manually constructing it, or + * using a third-party dictionary builder, you can turn it into a zstd + * dictionary by using `ZDICT_finalizeDictionary()`. You'll also have to + * provide some samples of the data. It will add the zstd header to the + * raw content, which contains a dictionary ID and entropy tables, which + * will improve compression ratio, and allow zstd to write the dictionary ID + * into the frame, if you so choose. + * + * Do I have to use zstd's dictionary builder? + * ------------------------------------------- + * + * No! You can construct dictionary content however you please, it is just + * bytes. It will always be valid as a raw content dictionary. If you want + * a zstd dictionary, which can improve compression ratio, use + * `ZDICT_finalizeDictionary()`. + * + * What is the attack surface of a zstd dictionary? + * ------------------------------------------------ + * + * Zstd is heavily fuzz tested, including loading fuzzed dictionaries, so + * zstd should never crash, or access out-of-bounds memory no matter what + * the dictionary is. However, if an attacker can control the dictionary + * during decompression, they can cause zstd to generate arbitrary bytes, + * just like if they controlled the compressed data. + * + ******************************************************************************/ + + /*! ZDICT_trainFromBuffer(): * Train a dictionary from an array of samples. * Redirect towards ZDICT_optimizeTrainFromBuffer_fastCover() single-threaded, with d=8, steps=4, @@ -203,14 +203,14 @@ ZDICTLIB_API size_t ZDICT_trainFromBuffer(void* dictBuffer, size_t dictBufferCap typedef struct { int compressionLevel; /*< optimize for a specific zstd compression level; 0 means default */ unsigned notificationLevel; /*< Write log to stderr; 0 = none (default); 1 = errors; 2 = progression; 3 = details; 4 = debug; */ - unsigned dictID; /*< force dictID value; 0 means auto mode (32-bits random value) - * NOTE: The zstd format reserves some dictionary IDs for future use. - * You may use them in private settings, but be warned that they - * may be used by zstd in a public dictionary registry in the future. - * These dictionary IDs are: - * - low range : <= 32767 - * - high range : >= (2^31) - */ + unsigned dictID; /*< force dictID value; 0 means auto mode (32-bits random value) + * NOTE: The zstd format reserves some dictionary IDs for future use. + * You may use them in private settings, but be warned that they + * may be used by zstd in a public dictionary registry in the future. + * These dictionary IDs are: + * - low range : <= 32767 + * - high range : >= (2^31) + */ } ZDICT_params_t; /*! ZDICT_finalizeDictionary(): @@ -410,11 +410,11 @@ typedef struct { * Note: ZDICT_trainFromBuffer_legacy() will send notifications into stderr if instructed to, using notificationLevel>0. */ ZDICTLIB_API size_t ZDICT_trainFromBuffer_legacy( - void* dictBuffer, size_t dictBufferCapacity, - const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples, + void* dictBuffer, size_t dictBufferCapacity, + const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples, ZDICT_legacy_params_t parameters); - + /* Deprecation warnings */ /* It is generally possible to disable deprecation warnings from compiler, for example with -Wno-deprecated-declarations for gcc @@ -426,7 +426,7 @@ ZDICTLIB_API size_t ZDICT_trainFromBuffer_legacy( # define ZDICT_GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__) # if defined (__cplusplus) && (__cplusplus >= 201402) /* C++14 or greater */ # define ZDICT_DEPRECATED(message) [[deprecated(message)]] ZDICTLIB_API -# elif defined(__clang__) || (ZDICT_GCC_VERSION >= 405) +# elif defined(__clang__) || (ZDICT_GCC_VERSION >= 405) # define ZDICT_DEPRECATED(message) ZDICTLIB_API __attribute__((deprecated(message))) # elif (ZDICT_GCC_VERSION >= 301) # define ZDICT_DEPRECATED(message) ZDICTLIB_API __attribute__((deprecated)) diff --git a/contrib/libs/zstd/lib/zstd.h b/contrib/libs/zstd/lib/zstd.h index a88ae7bf8e..f6550f5908 100644 --- a/contrib/libs/zstd/lib/zstd.h +++ b/contrib/libs/zstd/lib/zstd.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -73,23 +73,23 @@ extern "C" { /*------ Version ------*/ #define ZSTD_VERSION_MAJOR 1 -#define ZSTD_VERSION_MINOR 5 +#define ZSTD_VERSION_MINOR 5 #define ZSTD_VERSION_RELEASE 2 #define ZSTD_VERSION_NUMBER (ZSTD_VERSION_MAJOR *100*100 + ZSTD_VERSION_MINOR *100 + ZSTD_VERSION_RELEASE) -/*! ZSTD_versionNumber() : - * Return runtime library version, the value is (MAJOR*100*100 + MINOR*100 + RELEASE). */ -ZSTDLIB_API unsigned ZSTD_versionNumber(void); - +/*! ZSTD_versionNumber() : + * Return runtime library version, the value is (MAJOR*100*100 + MINOR*100 + RELEASE). */ +ZSTDLIB_API unsigned ZSTD_versionNumber(void); + #define ZSTD_LIB_VERSION ZSTD_VERSION_MAJOR.ZSTD_VERSION_MINOR.ZSTD_VERSION_RELEASE #define ZSTD_QUOTE(str) #str #define ZSTD_EXPAND_AND_QUOTE(str) ZSTD_QUOTE(str) #define ZSTD_VERSION_STRING ZSTD_EXPAND_AND_QUOTE(ZSTD_LIB_VERSION) -/*! ZSTD_versionString() : - * Return runtime library version, like "1.4.5". Requires v1.3.0+. */ -ZSTDLIB_API const char* ZSTD_versionString(void); - +/*! ZSTD_versionString() : + * Return runtime library version, like "1.4.5". Requires v1.3.0+. */ +ZSTDLIB_API const char* ZSTD_versionString(void); + /* ************************************* * Default constant ***************************************/ @@ -167,7 +167,7 @@ ZSTDLIB_API unsigned long long ZSTD_getFrameContentSize(const void *src, size_t * @return : decompressed size of `src` frame content _if known and not empty_, 0 otherwise. */ ZSTDLIB_API unsigned long long ZSTD_getDecompressedSize(const void* src, size_t srcSize); -/*! ZSTD_findFrameCompressedSize() : Requires v1.4.0+ +/*! ZSTD_findFrameCompressedSize() : Requires v1.4.0+ * `src` should point to the start of a ZSTD frame or skippable frame. * `srcSize` must be >= first frame size * @return : the compressed size of the first frame starting at `src`, @@ -181,9 +181,9 @@ ZSTDLIB_API size_t ZSTD_findFrameCompressedSize(const void* src, size_t srcSize) ZSTDLIB_API size_t ZSTD_compressBound(size_t srcSize); /*!< maximum compressed size in worst case single-pass scenario */ ZSTDLIB_API unsigned ZSTD_isError(size_t code); /*!< tells if a `size_t` function result is an error code */ ZSTDLIB_API const char* ZSTD_getErrorName(size_t code); /*!< provides readable string from an error code */ -ZSTDLIB_API int ZSTD_minCLevel(void); /*!< minimum negative compression level allowed, requires v1.4.0+ */ +ZSTDLIB_API int ZSTD_minCLevel(void); /*!< minimum negative compression level allowed, requires v1.4.0+ */ ZSTDLIB_API int ZSTD_maxCLevel(void); /*!< maximum compression level available */ -ZSTDLIB_API int ZSTD_defaultCLevel(void); /*!< default compression level, specified by ZSTD_CLEVEL_DEFAULT, requires v1.5.0+ */ +ZSTDLIB_API int ZSTD_defaultCLevel(void); /*!< default compression level, specified by ZSTD_CLEVEL_DEFAULT, requires v1.5.0+ */ /*************************************** @@ -201,7 +201,7 @@ ZSTDLIB_API int ZSTD_defaultCLevel(void); /*!< default compres */ typedef struct ZSTD_CCtx_s ZSTD_CCtx; ZSTDLIB_API ZSTD_CCtx* ZSTD_createCCtx(void); -ZSTDLIB_API size_t ZSTD_freeCCtx(ZSTD_CCtx* cctx); /* accept NULL pointer */ +ZSTDLIB_API size_t ZSTD_freeCCtx(ZSTD_CCtx* cctx); /* accept NULL pointer */ /*! ZSTD_compressCCtx() : * Same as ZSTD_compress(), using an explicit ZSTD_CCtx. @@ -224,7 +224,7 @@ ZSTDLIB_API size_t ZSTD_compressCCtx(ZSTD_CCtx* cctx, * Use one context per thread for parallel execution. */ typedef struct ZSTD_DCtx_s ZSTD_DCtx; ZSTDLIB_API ZSTD_DCtx* ZSTD_createDCtx(void); -ZSTDLIB_API size_t ZSTD_freeDCtx(ZSTD_DCtx* dctx); /* accept NULL pointer */ +ZSTDLIB_API size_t ZSTD_freeDCtx(ZSTD_DCtx* dctx); /* accept NULL pointer */ /*! ZSTD_decompressDCtx() : * Same as ZSTD_decompress(), @@ -236,9 +236,9 @@ ZSTDLIB_API size_t ZSTD_decompressDCtx(ZSTD_DCtx* dctx, const void* src, size_t srcSize); -/********************************************* -* Advanced compression API (Requires v1.4.0+) -**********************************************/ +/********************************************* +* Advanced compression API (Requires v1.4.0+) +**********************************************/ /* API design : * Parameters are pushed one by one into an existing context, @@ -339,9 +339,9 @@ typedef enum { * for large inputs, by finding large matches at long distance. * It increases memory usage and window size. * Note: enabling this parameter increases default ZSTD_c_windowLog to 128 MB - * except when expressly set to a different value. - * Note: will be enabled by default if ZSTD_c_windowLog >= 128 MB and - * compression strategy >= ZSTD_btopt (== compression level 16+) */ + * except when expressly set to a different value. + * Note: will be enabled by default if ZSTD_c_windowLog >= 128 MB and + * compression strategy >= ZSTD_btopt (== compression level 16+) */ ZSTD_c_ldmHashLog=161, /* Size of the table for long distance matching, as a power of 2. * Larger values increase memory usage and compression ratio, * but decrease compression speed. @@ -372,24 +372,24 @@ typedef enum { ZSTD_c_dictIDFlag=202, /* When applicable, dictionary's ID is written into frame header (default:1) */ /* multi-threading parameters */ - /* These parameters are only active if multi-threading is enabled (compiled with build macro ZSTD_MULTITHREAD). - * Otherwise, trying to set any other value than default (0) will be a no-op and return an error. - * In a situation where it's unknown if the linked library supports multi-threading or not, - * setting ZSTD_c_nbWorkers to any value >= 1 and consulting the return value provides a quick way to check this property. - */ + /* These parameters are only active if multi-threading is enabled (compiled with build macro ZSTD_MULTITHREAD). + * Otherwise, trying to set any other value than default (0) will be a no-op and return an error. + * In a situation where it's unknown if the linked library supports multi-threading or not, + * setting ZSTD_c_nbWorkers to any value >= 1 and consulting the return value provides a quick way to check this property. + */ ZSTD_c_nbWorkers=400, /* Select how many threads will be spawned to compress in parallel. - * When nbWorkers >= 1, triggers asynchronous mode when invoking ZSTD_compressStream*() : + * When nbWorkers >= 1, triggers asynchronous mode when invoking ZSTD_compressStream*() : * ZSTD_compressStream*() consumes input and flush output if possible, but immediately gives back control to caller, - * while compression is performed in parallel, within worker thread(s). + * while compression is performed in parallel, within worker thread(s). * (note : a strong exception to this rule is when first invocation of ZSTD_compressStream2() sets ZSTD_e_end : * in which case, ZSTD_compressStream2() delegates to ZSTD_compress2(), which is always a blocking call). * More workers improve speed, but also increase memory usage. - * Default value is `0`, aka "single-threaded mode" : no worker is spawned, - * compression is performed inside Caller's thread, and all invocations are blocking */ + * Default value is `0`, aka "single-threaded mode" : no worker is spawned, + * compression is performed inside Caller's thread, and all invocations are blocking */ ZSTD_c_jobSize=401, /* Size of a compression job. This value is enforced only when nbWorkers >= 1. * Each compression job is completed in parallel, so this value can indirectly impact the nb of active threads. * 0 means default, which is dynamically determined based on compression parameters. - * Job size must be a minimum of overlap size, or ZSTDMT_JOBSIZE_MIN (= 512 KB), whichever is largest. + * Job size must be a minimum of overlap size, or ZSTDMT_JOBSIZE_MIN (= 512 KB), whichever is largest. * The minimum size is automatically and transparently enforced. */ ZSTD_c_overlapLog=402, /* Control the overlap size, as a fraction of window size. * The overlap size is an amount of data reloaded from previous job at the beginning of a new job. @@ -414,13 +414,13 @@ typedef enum { * ZSTD_c_literalCompressionMode * ZSTD_c_targetCBlockSize * ZSTD_c_srcSizeHint - * ZSTD_c_enableDedicatedDictSearch - * ZSTD_c_stableInBuffer - * ZSTD_c_stableOutBuffer - * ZSTD_c_blockDelimiters - * ZSTD_c_validateSequences + * ZSTD_c_enableDedicatedDictSearch + * ZSTD_c_stableInBuffer + * ZSTD_c_stableOutBuffer + * ZSTD_c_blockDelimiters + * ZSTD_c_validateSequences * ZSTD_c_useBlockSplitter - * ZSTD_c_useRowMatchFinder + * ZSTD_c_useRowMatchFinder * Because they are not stable, it's necessary to define ZSTD_STATIC_LINKING_ONLY to access them. * note : never ever use experimentalParam? names directly; * also, the enums values themselves are unstable and can still change. @@ -431,15 +431,15 @@ typedef enum { ZSTD_c_experimentalParam4=1001, ZSTD_c_experimentalParam5=1002, ZSTD_c_experimentalParam6=1003, - ZSTD_c_experimentalParam7=1004, - ZSTD_c_experimentalParam8=1005, - ZSTD_c_experimentalParam9=1006, - ZSTD_c_experimentalParam10=1007, - ZSTD_c_experimentalParam11=1008, - ZSTD_c_experimentalParam12=1009, - ZSTD_c_experimentalParam13=1010, - ZSTD_c_experimentalParam14=1011, - ZSTD_c_experimentalParam15=1012 + ZSTD_c_experimentalParam7=1004, + ZSTD_c_experimentalParam8=1005, + ZSTD_c_experimentalParam9=1006, + ZSTD_c_experimentalParam10=1007, + ZSTD_c_experimentalParam11=1008, + ZSTD_c_experimentalParam12=1009, + ZSTD_c_experimentalParam13=1010, + ZSTD_c_experimentalParam14=1011, + ZSTD_c_experimentalParam15=1012 } ZSTD_cParameter; typedef struct { @@ -524,9 +524,9 @@ ZSTDLIB_API size_t ZSTD_compress2( ZSTD_CCtx* cctx, const void* src, size_t srcSize); -/*********************************************** -* Advanced decompression API (Requires v1.4.0+) -************************************************/ +/*********************************************** +* Advanced decompression API (Requires v1.4.0+) +************************************************/ /* The advanced API pushes parameters one by one into an existing DCtx context. * Parameters are sticky, and remain valid for all following frames @@ -550,15 +550,15 @@ typedef enum { * At the time of this writing, they include : * ZSTD_d_format * ZSTD_d_stableOutBuffer - * ZSTD_d_forceIgnoreChecksum - * ZSTD_d_refMultipleDDicts + * ZSTD_d_forceIgnoreChecksum + * ZSTD_d_refMultipleDDicts * Because they are not stable, it's necessary to define ZSTD_STATIC_LINKING_ONLY to access them. * note : never ever use experimentalParam? names directly */ ZSTD_d_experimentalParam1=1000, - ZSTD_d_experimentalParam2=1001, - ZSTD_d_experimentalParam3=1002, - ZSTD_d_experimentalParam4=1003 + ZSTD_d_experimentalParam2=1001, + ZSTD_d_experimentalParam3=1002, + ZSTD_d_experimentalParam4=1003 } ZSTD_dParameter; @@ -672,7 +672,7 @@ typedef ZSTD_CCtx ZSTD_CStream; /**< CCtx and CStream are now effectively same /* Continue to distinguish them for compatibility with older versions <= v1.2.0 */ /*===== ZSTD_CStream management functions =====*/ ZSTDLIB_API ZSTD_CStream* ZSTD_createCStream(void); -ZSTDLIB_API size_t ZSTD_freeCStream(ZSTD_CStream* zcs); /* accept NULL pointer */ +ZSTDLIB_API size_t ZSTD_freeCStream(ZSTD_CStream* zcs); /* accept NULL pointer */ /*===== Streaming compression functions =====*/ typedef enum { @@ -688,15 +688,15 @@ typedef enum { : note : multithreaded compression will block to flush as much output as possible. */ } ZSTD_EndDirective; -/*! ZSTD_compressStream2() : Requires v1.4.0+ +/*! ZSTD_compressStream2() : Requires v1.4.0+ * Behaves about the same as ZSTD_compressStream, with additional control on end directive. * - Compression parameters are pushed into CCtx before starting compression, using ZSTD_CCtx_set*() * - Compression parameters cannot be changed once compression is started (save a list of exceptions in multi-threading mode) * - output->pos must be <= dstCapacity, input->pos must be <= srcSize * - output->pos and input->pos will be updated. They are guaranteed to remain below their respective limit. - * - endOp must be a valid directive + * - endOp must be a valid directive * - When nbWorkers==0 (default), function is blocking : it completes its job before returning to caller. - * - When nbWorkers>=1, function is non-blocking : it copies a portion of input, distributes jobs to internal worker threads, flush to output whatever is available, + * - When nbWorkers>=1, function is non-blocking : it copies a portion of input, distributes jobs to internal worker threads, flush to output whatever is available, * and then immediately returns, just indicating that there is some data remaining to be flushed. * The function nonetheless guarantees forward progress : it will return only after it reads or write at least 1+ byte. * - Exception : if the first call requests a ZSTD_e_end directive and provides enough dstCapacity, the function delegates to ZSTD_compress2() which is always blocking. @@ -734,11 +734,11 @@ ZSTDLIB_API size_t ZSTD_CStreamOutSize(void); /**< recommended size for output /* ***************************************************************************** - * This following is a legacy streaming API, available since v1.0+ . + * This following is a legacy streaming API, available since v1.0+ . * It can be replaced by ZSTD_CCtx_reset() and ZSTD_compressStream2(). * It is redundant, but remains fully supported. - * Streaming in combination with advanced parameters and dictionary compression - * can only be used through the new API. + * Streaming in combination with advanced parameters and dictionary compression + * can only be used through the new API. ******************************************************************************/ /*! @@ -793,7 +793,7 @@ typedef ZSTD_DCtx ZSTD_DStream; /**< DCtx and DStream are now effectively same /* For compatibility with versions <= v1.2.0, prefer differentiating them. */ /*===== ZSTD_DStream management functions =====*/ ZSTDLIB_API ZSTD_DStream* ZSTD_createDStream(void); -ZSTDLIB_API size_t ZSTD_freeDStream(ZSTD_DStream* zds); /* accept NULL pointer */ +ZSTDLIB_API size_t ZSTD_freeDStream(ZSTD_DStream* zds); /* accept NULL pointer */ /*===== Streaming decompression functions =====*/ @@ -816,7 +816,7 @@ ZSTDLIB_API size_t ZSTD_DStreamOutSize(void); /*!< recommended size for output /*! ZSTD_compress_usingDict() : * Compression at an explicit compression level using a Dictionary. * A dictionary can be any arbitrary data segment (also called a prefix), - * or a buffer with specified information (see zdict.h). + * or a buffer with specified information (see zdict.h). * Note : This function loads the dictionary, resulting in significant startup delay. * It's intended for a dictionary used only once. * Note 2 : When `dict == NULL || dictSize < 8` no dictionary is used. */ @@ -859,8 +859,8 @@ ZSTDLIB_API ZSTD_CDict* ZSTD_createCDict(const void* dictBuffer, size_t dictSize int compressionLevel); /*! ZSTD_freeCDict() : - * Function frees memory allocated by ZSTD_createCDict(). - * If a NULL pointer is passed, no operation is performed. */ + * Function frees memory allocated by ZSTD_createCDict(). + * If a NULL pointer is passed, no operation is performed. */ ZSTDLIB_API size_t ZSTD_freeCDict(ZSTD_CDict* CDict); /*! ZSTD_compress_usingCDict() : @@ -882,8 +882,8 @@ typedef struct ZSTD_DDict_s ZSTD_DDict; ZSTDLIB_API ZSTD_DDict* ZSTD_createDDict(const void* dictBuffer, size_t dictSize); /*! ZSTD_freeDDict() : - * Function frees memory allocated with ZSTD_createDDict() - * If a NULL pointer is passed, no operation is performed. */ + * Function frees memory allocated with ZSTD_createDDict() + * If a NULL pointer is passed, no operation is performed. */ ZSTDLIB_API size_t ZSTD_freeDDict(ZSTD_DDict* ddict); /*! ZSTD_decompress_usingDDict() : @@ -899,25 +899,25 @@ ZSTDLIB_API size_t ZSTD_decompress_usingDDict(ZSTD_DCtx* dctx, * Dictionary helper functions *******************************/ -/*! ZSTD_getDictID_fromDict() : Requires v1.4.0+ +/*! ZSTD_getDictID_fromDict() : Requires v1.4.0+ * Provides the dictID stored within dictionary. * if @return == 0, the dictionary is not conformant with Zstandard specification. * It can still be loaded, but as a content-only dictionary. */ ZSTDLIB_API unsigned ZSTD_getDictID_fromDict(const void* dict, size_t dictSize); -/*! ZSTD_getDictID_fromCDict() : Requires v1.5.0+ - * Provides the dictID of the dictionary loaded into `cdict`. - * If @return == 0, the dictionary is not conformant to Zstandard specification, or empty. - * Non-conformant dictionaries can still be loaded, but as content-only dictionaries. */ -ZSTDLIB_API unsigned ZSTD_getDictID_fromCDict(const ZSTD_CDict* cdict); - -/*! ZSTD_getDictID_fromDDict() : Requires v1.4.0+ +/*! ZSTD_getDictID_fromCDict() : Requires v1.5.0+ + * Provides the dictID of the dictionary loaded into `cdict`. + * If @return == 0, the dictionary is not conformant to Zstandard specification, or empty. + * Non-conformant dictionaries can still be loaded, but as content-only dictionaries. */ +ZSTDLIB_API unsigned ZSTD_getDictID_fromCDict(const ZSTD_CDict* cdict); + +/*! ZSTD_getDictID_fromDDict() : Requires v1.4.0+ * Provides the dictID of the dictionary loaded into `ddict`. * If @return == 0, the dictionary is not conformant to Zstandard specification, or empty. * Non-conformant dictionaries can still be loaded, but as content-only dictionaries. */ ZSTDLIB_API unsigned ZSTD_getDictID_fromDDict(const ZSTD_DDict* ddict); -/*! ZSTD_getDictID_fromFrame() : Requires v1.4.0+ +/*! ZSTD_getDictID_fromFrame() : Requires v1.4.0+ * Provides the dictID required to decompressed the frame stored within `src`. * If @return == 0, the dictID could not be decoded. * This could for one of the following reasons : @@ -931,7 +931,7 @@ ZSTDLIB_API unsigned ZSTD_getDictID_fromFrame(const void* src, size_t srcSize); /******************************************************************************* - * Advanced dictionary and prefix API (Requires v1.4.0+) + * Advanced dictionary and prefix API (Requires v1.4.0+) * * This API allows dictionaries to be used with ZSTD_compress2(), * ZSTD_compressStream2(), and ZSTD_decompressDCtx(). Dictionaries are sticky, and @@ -940,7 +940,7 @@ ZSTDLIB_API unsigned ZSTD_getDictID_fromFrame(const void* src, size_t srcSize); ******************************************************************************/ -/*! ZSTD_CCtx_loadDictionary() : Requires v1.4.0+ +/*! ZSTD_CCtx_loadDictionary() : Requires v1.4.0+ * Create an internal CDict from `dict` buffer. * Decompression will have to use same dictionary. * @result : 0, or an error code (which can be tested with ZSTD_isError()). @@ -959,11 +959,11 @@ ZSTDLIB_API unsigned ZSTD_getDictID_fromFrame(const void* src, size_t srcSize); * to precisely select how dictionary content must be interpreted. */ ZSTDLIB_API size_t ZSTD_CCtx_loadDictionary(ZSTD_CCtx* cctx, const void* dict, size_t dictSize); -/*! ZSTD_CCtx_refCDict() : Requires v1.4.0+ +/*! ZSTD_CCtx_refCDict() : Requires v1.4.0+ * Reference a prepared dictionary, to be used for all next compressed frames. * Note that compression parameters are enforced from within CDict, * and supersede any compression parameter previously set within CCtx. - * The parameters ignored are labelled as "superseded-by-cdict" in the ZSTD_cParameter enum docs. + * The parameters ignored are labelled as "superseded-by-cdict" in the ZSTD_cParameter enum docs. * The ignored parameters will be used again if the CCtx is returned to no-dictionary mode. * The dictionary will remain valid for future compressed frames using same CCtx. * @result : 0, or an error code (which can be tested with ZSTD_isError()). @@ -973,7 +973,7 @@ ZSTDLIB_API size_t ZSTD_CCtx_loadDictionary(ZSTD_CCtx* cctx, const void* dict, s * Note 2 : CDict is just referenced, its lifetime must outlive its usage within CCtx. */ ZSTDLIB_API size_t ZSTD_CCtx_refCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict); -/*! ZSTD_CCtx_refPrefix() : Requires v1.4.0+ +/*! ZSTD_CCtx_refPrefix() : Requires v1.4.0+ * Reference a prefix (single-usage dictionary) for next compressed frame. * A prefix is **only used once**. Tables are discarded at end of frame (ZSTD_e_end). * Decompression will need same prefix to properly regenerate data. @@ -994,7 +994,7 @@ ZSTDLIB_API size_t ZSTD_CCtx_refCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict); ZSTDLIB_API size_t ZSTD_CCtx_refPrefix(ZSTD_CCtx* cctx, const void* prefix, size_t prefixSize); -/*! ZSTD_DCtx_loadDictionary() : Requires v1.4.0+ +/*! ZSTD_DCtx_loadDictionary() : Requires v1.4.0+ * Create an internal DDict from dict buffer, * to be used to decompress next frames. * The dictionary remains valid for all future frames, until explicitly invalidated. @@ -1011,16 +1011,16 @@ ZSTDLIB_API size_t ZSTD_CCtx_refPrefix(ZSTD_CCtx* cctx, */ ZSTDLIB_API size_t ZSTD_DCtx_loadDictionary(ZSTD_DCtx* dctx, const void* dict, size_t dictSize); -/*! ZSTD_DCtx_refDDict() : Requires v1.4.0+ +/*! ZSTD_DCtx_refDDict() : Requires v1.4.0+ * Reference a prepared dictionary, to be used to decompress next frames. * The dictionary remains active for decompression of future frames using same DCtx. - * - * If called with ZSTD_d_refMultipleDDicts enabled, repeated calls of this function - * will store the DDict references in a table, and the DDict used for decompression - * will be determined at decompression time, as per the dict ID in the frame. - * The memory for the table is allocated on the first call to refDDict, and can be - * freed with ZSTD_freeDCtx(). - * + * + * If called with ZSTD_d_refMultipleDDicts enabled, repeated calls of this function + * will store the DDict references in a table, and the DDict used for decompression + * will be determined at decompression time, as per the dict ID in the frame. + * The memory for the table is allocated on the first call to refDDict, and can be + * freed with ZSTD_freeDCtx(). + * * @result : 0, or an error code (which can be tested with ZSTD_isError()). * Note 1 : Currently, only one dictionary can be managed. * Referencing a new dictionary effectively "discards" any previous one. @@ -1029,7 +1029,7 @@ ZSTDLIB_API size_t ZSTD_DCtx_loadDictionary(ZSTD_DCtx* dctx, const void* dict, s */ ZSTDLIB_API size_t ZSTD_DCtx_refDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict); -/*! ZSTD_DCtx_refPrefix() : Requires v1.4.0+ +/*! ZSTD_DCtx_refPrefix() : Requires v1.4.0+ * Reference a prefix (single-usage dictionary) to decompress next frame. * This is the reverse operation of ZSTD_CCtx_refPrefix(), * and must use the same prefix as the one used during compression. @@ -1050,7 +1050,7 @@ ZSTDLIB_API size_t ZSTD_DCtx_refPrefix(ZSTD_DCtx* dctx, /* === Memory management === */ -/*! ZSTD_sizeof_*() : Requires v1.4.0+ +/*! ZSTD_sizeof_*() : Requires v1.4.0+ * These functions give the _current_ memory usage of selected object. * Note that object memory usage can evolve (increase or decrease) over time. */ ZSTDLIB_API size_t ZSTD_sizeof_CCtx(const ZSTD_CCtx* cctx); @@ -1086,28 +1086,28 @@ ZSTDLIB_API size_t ZSTD_sizeof_DDict(const ZSTD_DDict* ddict); # endif #endif -/* Deprecation warnings : - * Should these warnings be a problem, it is generally possible to disable them, - * typically with -Wno-deprecated-declarations for gcc or _CRT_SECURE_NO_WARNINGS in Visual. - * Otherwise, it's also possible to define ZSTD_DISABLE_DEPRECATE_WARNINGS. - */ -#ifdef ZSTD_DISABLE_DEPRECATE_WARNINGS +/* Deprecation warnings : + * Should these warnings be a problem, it is generally possible to disable them, + * typically with -Wno-deprecated-declarations for gcc or _CRT_SECURE_NO_WARNINGS in Visual. + * Otherwise, it's also possible to define ZSTD_DISABLE_DEPRECATE_WARNINGS. + */ +#ifdef ZSTD_DISABLE_DEPRECATE_WARNINGS # define ZSTD_DEPRECATED(message) ZSTDLIB_STATIC_API /* disable deprecation warnings */ -#else -# if defined (__cplusplus) && (__cplusplus >= 201402) /* C++14 or greater */ +#else +# if defined (__cplusplus) && (__cplusplus >= 201402) /* C++14 or greater */ # define ZSTD_DEPRECATED(message) [[deprecated(message)]] ZSTDLIB_STATIC_API -# elif (defined(GNUC) && (GNUC > 4 || (GNUC == 4 && GNUC_MINOR >= 5))) || defined(__clang__) +# elif (defined(GNUC) && (GNUC > 4 || (GNUC == 4 && GNUC_MINOR >= 5))) || defined(__clang__) # define ZSTD_DEPRECATED(message) ZSTDLIB_STATIC_API __attribute__((deprecated(message))) -# elif defined(__GNUC__) && (__GNUC__ >= 3) +# elif defined(__GNUC__) && (__GNUC__ >= 3) # define ZSTD_DEPRECATED(message) ZSTDLIB_STATIC_API __attribute__((deprecated)) -# elif defined(_MSC_VER) +# elif defined(_MSC_VER) # define ZSTD_DEPRECATED(message) ZSTDLIB_STATIC_API __declspec(deprecated(message)) -# else -# pragma message("WARNING: You need to implement ZSTD_DEPRECATED for this compiler") +# else +# pragma message("WARNING: You need to implement ZSTD_DEPRECATED for this compiler") # define ZSTD_DEPRECATED(message) ZSTDLIB_STATIC_API -# endif -#endif /* ZSTD_DISABLE_DEPRECATE_WARNINGS */ - +# endif +#endif /* ZSTD_DISABLE_DEPRECATE_WARNINGS */ + /**************************************************************************************** * experimental API (static linking only) **************************************************************************************** @@ -1176,40 +1176,40 @@ ZSTDLIB_API size_t ZSTD_sizeof_DDict(const ZSTD_DDict* ddict); typedef struct ZSTD_CCtx_params_s ZSTD_CCtx_params; typedef struct { - unsigned int offset; /* The offset of the match. (NOT the same as the offset code) - * If offset == 0 and matchLength == 0, this sequence represents the last - * literals in the block of litLength size. - */ - - unsigned int litLength; /* Literal length of the sequence. */ - unsigned int matchLength; /* Match length of the sequence. */ - - /* Note: Users of this API may provide a sequence with matchLength == litLength == offset == 0. - * In this case, we will treat the sequence as a marker for a block boundary. - */ - - unsigned int rep; /* Represents which repeat offset is represented by the field 'offset'. - * Ranges from [0, 3]. - * - * Repeat offsets are essentially previous offsets from previous sequences sorted in - * recency order. For more detail, see doc/zstd_compression_format.md - * - * If rep == 0, then 'offset' does not contain a repeat offset. - * If rep > 0: - * If litLength != 0: - * rep == 1 --> offset == repeat_offset_1 - * rep == 2 --> offset == repeat_offset_2 - * rep == 3 --> offset == repeat_offset_3 - * If litLength == 0: - * rep == 1 --> offset == repeat_offset_2 - * rep == 2 --> offset == repeat_offset_3 - * rep == 3 --> offset == repeat_offset_1 - 1 - * - * Note: This field is optional. ZSTD_generateSequences() will calculate the value of - * 'rep', but repeat offsets do not necessarily need to be calculated from an external - * sequence provider's perspective. For example, ZSTD_compressSequences() does not - * use this 'rep' field at all (as of now). - */ + unsigned int offset; /* The offset of the match. (NOT the same as the offset code) + * If offset == 0 and matchLength == 0, this sequence represents the last + * literals in the block of litLength size. + */ + + unsigned int litLength; /* Literal length of the sequence. */ + unsigned int matchLength; /* Match length of the sequence. */ + + /* Note: Users of this API may provide a sequence with matchLength == litLength == offset == 0. + * In this case, we will treat the sequence as a marker for a block boundary. + */ + + unsigned int rep; /* Represents which repeat offset is represented by the field 'offset'. + * Ranges from [0, 3]. + * + * Repeat offsets are essentially previous offsets from previous sequences sorted in + * recency order. For more detail, see doc/zstd_compression_format.md + * + * If rep == 0, then 'offset' does not contain a repeat offset. + * If rep > 0: + * If litLength != 0: + * rep == 1 --> offset == repeat_offset_1 + * rep == 2 --> offset == repeat_offset_2 + * rep == 3 --> offset == repeat_offset_3 + * If litLength == 0: + * rep == 1 --> offset == repeat_offset_2 + * rep == 2 --> offset == repeat_offset_3 + * rep == 3 --> offset == repeat_offset_1 - 1 + * + * Note: This field is optional. ZSTD_generateSequences() will calculate the value of + * 'rep', but repeat offsets do not necessarily need to be calculated from an external + * sequence provider's perspective. For example, ZSTD_compressSequences() does not + * use this 'rep' field at all (as of now). + */ } ZSTD_Sequence; typedef struct { @@ -1252,18 +1252,18 @@ typedef enum { } ZSTD_format_e; typedef enum { - /* Note: this enum controls ZSTD_d_forceIgnoreChecksum */ - ZSTD_d_validateChecksum = 0, - ZSTD_d_ignoreChecksum = 1 -} ZSTD_forceIgnoreChecksum_e; - -typedef enum { - /* Note: this enum controls ZSTD_d_refMultipleDDicts */ - ZSTD_rmd_refSingleDDict = 0, - ZSTD_rmd_refMultipleDDicts = 1 -} ZSTD_refMultipleDDicts_e; - -typedef enum { + /* Note: this enum controls ZSTD_d_forceIgnoreChecksum */ + ZSTD_d_validateChecksum = 0, + ZSTD_d_ignoreChecksum = 1 +} ZSTD_forceIgnoreChecksum_e; + +typedef enum { + /* Note: this enum controls ZSTD_d_refMultipleDDicts */ + ZSTD_rmd_refSingleDDict = 0, + ZSTD_rmd_refMultipleDDicts = 1 +} ZSTD_refMultipleDDicts_e; + +typedef enum { /* Note: this enum and the behavior it controls are effectively internal * implementation details of the compressor. They are expected to continue * to evolve and should be considered only in the context of extremely @@ -1311,7 +1311,7 @@ typedef enum { ZSTD_lcm_uncompressed = 2 /**< Always emit uncompressed literals. */ } ZSTD_literalCompressionMode_e; -typedef enum { +typedef enum { /* Note: This enum controls features which are conditionally beneficial. Zstd typically will make a final * decision on whether or not to enable the feature (ZSTD_ps_auto), but setting the switch to ZSTD_ps_enable * or ZSTD_ps_disable allow for a force enable/disable the feature. @@ -1353,7 +1353,7 @@ ZSTDLIB_STATIC_API unsigned long long ZSTD_findDecompressedSize(const void* src, * `srcSize` must be the _exact_ size of this series * (i.e. there should be a frame boundary at `src + srcSize`) * @return : - upper-bound for the decompressed size of all data in all successive frames - * - if an error occurred: ZSTD_CONTENTSIZE_ERROR + * - if an error occurred: ZSTD_CONTENTSIZE_ERROR * * note 1 : an error can occur if `src` contains an invalid or incorrectly formatted frame. * note 2 : the upper-bound is exact when the decompressed size field is available in every ZSTD encoded frame of `src`. @@ -1369,92 +1369,92 @@ ZSTDLIB_STATIC_API unsigned long long ZSTD_decompressBound(const void* src, size * or an error code (if srcSize is too small) */ ZSTDLIB_STATIC_API size_t ZSTD_frameHeaderSize(const void* src, size_t srcSize); -typedef enum { - ZSTD_sf_noBlockDelimiters = 0, /* Representation of ZSTD_Sequence has no block delimiters, sequences only */ - ZSTD_sf_explicitBlockDelimiters = 1 /* Representation of ZSTD_Sequence contains explicit block delimiters */ -} ZSTD_sequenceFormat_e; - -/*! ZSTD_generateSequences() : - * Generate sequences using ZSTD_compress2, given a source buffer. - * - * Each block will end with a dummy sequence - * with offset == 0, matchLength == 0, and litLength == length of last literals. - * litLength may be == 0, and if so, then the sequence of (of: 0 ml: 0 ll: 0) - * simply acts as a block delimiter. - * +typedef enum { + ZSTD_sf_noBlockDelimiters = 0, /* Representation of ZSTD_Sequence has no block delimiters, sequences only */ + ZSTD_sf_explicitBlockDelimiters = 1 /* Representation of ZSTD_Sequence contains explicit block delimiters */ +} ZSTD_sequenceFormat_e; + +/*! ZSTD_generateSequences() : + * Generate sequences using ZSTD_compress2, given a source buffer. + * + * Each block will end with a dummy sequence + * with offset == 0, matchLength == 0, and litLength == length of last literals. + * litLength may be == 0, and if so, then the sequence of (of: 0 ml: 0 ll: 0) + * simply acts as a block delimiter. + * * zc can be used to insert custom compression params. * This function invokes ZSTD_compress2 - * - * The output of this function can be fed into ZSTD_compressSequences() with CCtx - * setting of ZSTD_c_blockDelimiters as ZSTD_sf_explicitBlockDelimiters - * @return : number of sequences generated + * + * The output of this function can be fed into ZSTD_compressSequences() with CCtx + * setting of ZSTD_c_blockDelimiters as ZSTD_sf_explicitBlockDelimiters + * @return : number of sequences generated */ ZSTDLIB_STATIC_API size_t ZSTD_generateSequences(ZSTD_CCtx* zc, ZSTD_Sequence* outSeqs, - size_t outSeqsSize, const void* src, size_t srcSize); - -/*! ZSTD_mergeBlockDelimiters() : - * Given an array of ZSTD_Sequence, remove all sequences that represent block delimiters/last literals - * by merging them into into the literals of the next sequence. - * - * As such, the final generated result has no explicit representation of block boundaries, - * and the final last literals segment is not represented in the sequences. - * - * The output of this function can be fed into ZSTD_compressSequences() with CCtx - * setting of ZSTD_c_blockDelimiters as ZSTD_sf_noBlockDelimiters - * @return : number of sequences left after merging - */ + size_t outSeqsSize, const void* src, size_t srcSize); + +/*! ZSTD_mergeBlockDelimiters() : + * Given an array of ZSTD_Sequence, remove all sequences that represent block delimiters/last literals + * by merging them into into the literals of the next sequence. + * + * As such, the final generated result has no explicit representation of block boundaries, + * and the final last literals segment is not represented in the sequences. + * + * The output of this function can be fed into ZSTD_compressSequences() with CCtx + * setting of ZSTD_c_blockDelimiters as ZSTD_sf_noBlockDelimiters + * @return : number of sequences left after merging + */ ZSTDLIB_STATIC_API size_t ZSTD_mergeBlockDelimiters(ZSTD_Sequence* sequences, size_t seqsSize); - -/*! ZSTD_compressSequences() : - * Compress an array of ZSTD_Sequence, generated from the original source buffer, into dst. - * If a dictionary is included, then the cctx should reference the dict. (see: ZSTD_CCtx_refCDict(), ZSTD_CCtx_loadDictionary(), etc.) - * The entire source is compressed into a single frame. - * - * The compression behavior changes based on cctx params. In particular: - * If ZSTD_c_blockDelimiters == ZSTD_sf_noBlockDelimiters, the array of ZSTD_Sequence is expected to contain - * no block delimiters (defined in ZSTD_Sequence). Block boundaries are roughly determined based on - * the block size derived from the cctx, and sequences may be split. This is the default setting. - * - * If ZSTD_c_blockDelimiters == ZSTD_sf_explicitBlockDelimiters, the array of ZSTD_Sequence is expected to contain - * block delimiters (defined in ZSTD_Sequence). Behavior is undefined if no block delimiters are provided. - * - * If ZSTD_c_validateSequences == 0, this function will blindly accept the sequences provided. Invalid sequences cause undefined - * behavior. If ZSTD_c_validateSequences == 1, then if sequence is invalid (see doc/zstd_compression_format.md for - * specifics regarding offset/matchlength requirements) then the function will bail out and return an error. - * - * In addition to the two adjustable experimental params, there are other important cctx params. - * - ZSTD_c_minMatch MUST be set as less than or equal to the smallest match generated by the match finder. It has a minimum value of ZSTD_MINMATCH_MIN. - * - ZSTD_c_compressionLevel accordingly adjusts the strength of the entropy coder, as it would in typical compression. - * - ZSTD_c_windowLog affects offset validation: this function will return an error at higher debug levels if a provided offset - * is larger than what the spec allows for a given window log and dictionary (if present). See: doc/zstd_compression_format.md - * - * Note: Repcodes are, as of now, always re-calculated within this function, so ZSTD_Sequence::rep is unused. - * Note 2: Once we integrate ability to ingest repcodes, the explicit block delims mode must respect those repcodes exactly, - * and cannot emit an RLE block that disagrees with the repcode history - * @return : final compressed size or a ZSTD error. - */ + +/*! ZSTD_compressSequences() : + * Compress an array of ZSTD_Sequence, generated from the original source buffer, into dst. + * If a dictionary is included, then the cctx should reference the dict. (see: ZSTD_CCtx_refCDict(), ZSTD_CCtx_loadDictionary(), etc.) + * The entire source is compressed into a single frame. + * + * The compression behavior changes based on cctx params. In particular: + * If ZSTD_c_blockDelimiters == ZSTD_sf_noBlockDelimiters, the array of ZSTD_Sequence is expected to contain + * no block delimiters (defined in ZSTD_Sequence). Block boundaries are roughly determined based on + * the block size derived from the cctx, and sequences may be split. This is the default setting. + * + * If ZSTD_c_blockDelimiters == ZSTD_sf_explicitBlockDelimiters, the array of ZSTD_Sequence is expected to contain + * block delimiters (defined in ZSTD_Sequence). Behavior is undefined if no block delimiters are provided. + * + * If ZSTD_c_validateSequences == 0, this function will blindly accept the sequences provided. Invalid sequences cause undefined + * behavior. If ZSTD_c_validateSequences == 1, then if sequence is invalid (see doc/zstd_compression_format.md for + * specifics regarding offset/matchlength requirements) then the function will bail out and return an error. + * + * In addition to the two adjustable experimental params, there are other important cctx params. + * - ZSTD_c_minMatch MUST be set as less than or equal to the smallest match generated by the match finder. It has a minimum value of ZSTD_MINMATCH_MIN. + * - ZSTD_c_compressionLevel accordingly adjusts the strength of the entropy coder, as it would in typical compression. + * - ZSTD_c_windowLog affects offset validation: this function will return an error at higher debug levels if a provided offset + * is larger than what the spec allows for a given window log and dictionary (if present). See: doc/zstd_compression_format.md + * + * Note: Repcodes are, as of now, always re-calculated within this function, so ZSTD_Sequence::rep is unused. + * Note 2: Once we integrate ability to ingest repcodes, the explicit block delims mode must respect those repcodes exactly, + * and cannot emit an RLE block that disagrees with the repcode history + * @return : final compressed size or a ZSTD error. + */ ZSTDLIB_STATIC_API size_t ZSTD_compressSequences(ZSTD_CCtx* const cctx, void* dst, size_t dstSize, - const ZSTD_Sequence* inSeqs, size_t inSeqsSize, - const void* src, size_t srcSize); - - -/*! ZSTD_writeSkippableFrame() : - * Generates a zstd skippable frame containing data given by src, and writes it to dst buffer. - * - * Skippable frames begin with a a 4-byte magic number. There are 16 possible choices of magic number, - * ranging from ZSTD_MAGIC_SKIPPABLE_START to ZSTD_MAGIC_SKIPPABLE_START+15. - * As such, the parameter magicVariant controls the exact skippable frame magic number variant used, so - * the magic number used will be ZSTD_MAGIC_SKIPPABLE_START + magicVariant. - * - * Returns an error if destination buffer is not large enough, if the source size is not representable - * with a 4-byte unsigned int, or if the parameter magicVariant is greater than 15 (and therefore invalid). - * - * @return : number of bytes written or a ZSTD error. - */ + const ZSTD_Sequence* inSeqs, size_t inSeqsSize, + const void* src, size_t srcSize); + + +/*! ZSTD_writeSkippableFrame() : + * Generates a zstd skippable frame containing data given by src, and writes it to dst buffer. + * + * Skippable frames begin with a a 4-byte magic number. There are 16 possible choices of magic number, + * ranging from ZSTD_MAGIC_SKIPPABLE_START to ZSTD_MAGIC_SKIPPABLE_START+15. + * As such, the parameter magicVariant controls the exact skippable frame magic number variant used, so + * the magic number used will be ZSTD_MAGIC_SKIPPABLE_START + magicVariant. + * + * Returns an error if destination buffer is not large enough, if the source size is not representable + * with a 4-byte unsigned int, or if the parameter magicVariant is greater than 15 (and therefore invalid). + * + * @return : number of bytes written or a ZSTD error. + */ ZSTDLIB_STATIC_API size_t ZSTD_writeSkippableFrame(void* dst, size_t dstCapacity, - const void* src, size_t srcSize, unsigned magicVariant); - + const void* src, size_t srcSize, unsigned magicVariant); + /*! ZSTD_readSkippableFrame() : * Retrieves a zstd skippable frame containing data given by src, and writes it to dst buffer. * @@ -1468,7 +1468,7 @@ ZSTDLIB_STATIC_API size_t ZSTD_writeSkippableFrame(void* dst, size_t dstCapacity */ ZSTDLIB_API size_t ZSTD_readSkippableFrame(void* dst, size_t dstCapacity, unsigned* magicVariant, const void* src, size_t srcSize); - + /*! ZSTD_isSkippableFrame() : * Tells if the content of `buffer` starts with a valid Frame Identifier for a skippable frame. */ @@ -1585,11 +1585,11 @@ ZSTDLIB_STATIC_API const ZSTD_DDict* ZSTD_initStaticDDict( typedef void* (*ZSTD_allocFunction) (void* opaque, size_t size); typedef void (*ZSTD_freeFunction) (void* opaque, void* address); typedef struct { ZSTD_allocFunction customAlloc; ZSTD_freeFunction customFree; void* opaque; } ZSTD_customMem; -static -#ifdef __GNUC__ -__attribute__((__unused__)) -#endif -ZSTD_customMem const ZSTD_defaultCMem = { NULL, NULL, NULL }; /**< this constant defers to stdlib's functions */ +static +#ifdef __GNUC__ +__attribute__((__unused__)) +#endif +ZSTD_customMem const ZSTD_defaultCMem = { NULL, NULL, NULL }; /**< this constant defers to stdlib's functions */ ZSTDLIB_STATIC_API ZSTD_CCtx* ZSTD_createCCtx_advanced(ZSTD_customMem customMem); ZSTDLIB_STATIC_API ZSTD_CStream* ZSTD_createCStream_advanced(ZSTD_customMem customMem); @@ -1611,30 +1611,30 @@ ZSTDLIB_STATIC_API ZSTD_CDict* ZSTD_createCDict_advanced(const void* dict, size_ * ZSTD_CCtx_refThreadPool assigns a thread pool to a context (use NULL argument value * to use an internal thread pool). * ZSTD_freeThreadPool frees a thread pool, accepts NULL pointer. - */ -typedef struct POOL_ctx_s ZSTD_threadPool; + */ +typedef struct POOL_ctx_s ZSTD_threadPool; ZSTDLIB_STATIC_API ZSTD_threadPool* ZSTD_createThreadPool(size_t numThreads); ZSTDLIB_STATIC_API void ZSTD_freeThreadPool (ZSTD_threadPool* pool); /* accept NULL pointer */ ZSTDLIB_STATIC_API size_t ZSTD_CCtx_refThreadPool(ZSTD_CCtx* cctx, ZSTD_threadPool* pool); - - -/* - * This API is temporary and is expected to change or disappear in the future! - */ + + +/* + * This API is temporary and is expected to change or disappear in the future! + */ ZSTDLIB_STATIC_API ZSTD_CDict* ZSTD_createCDict_advanced2( - const void* dict, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType, - const ZSTD_CCtx_params* cctxParams, - ZSTD_customMem customMem); - + const void* dict, size_t dictSize, + ZSTD_dictLoadMethod_e dictLoadMethod, + ZSTD_dictContentType_e dictContentType, + const ZSTD_CCtx_params* cctxParams, + ZSTD_customMem customMem); + ZSTDLIB_STATIC_API ZSTD_DDict* ZSTD_createDDict_advanced( - const void* dict, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType, - ZSTD_customMem customMem); - + const void* dict, size_t dictSize, + ZSTD_dictLoadMethod_e dictLoadMethod, + ZSTD_dictContentType_e dictContentType, + ZSTD_customMem customMem); + /*************************************** * Advanced compression functions ***************************************/ @@ -1673,20 +1673,20 @@ ZSTDLIB_STATIC_API ZSTD_compressionParameters ZSTD_adjustCParams(ZSTD_compressio /*! ZSTD_compress_advanced() : * Note : this function is now DEPRECATED. * It can be replaced by ZSTD_compress2(), in combination with ZSTD_CCtx_setParameter() and other parameter setters. - * This prototype will generate compilation warnings. */ -ZSTD_DEPRECATED("use ZSTD_compress2") -size_t ZSTD_compress_advanced(ZSTD_CCtx* cctx, + * This prototype will generate compilation warnings. */ +ZSTD_DEPRECATED("use ZSTD_compress2") +size_t ZSTD_compress_advanced(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize, const void* dict,size_t dictSize, ZSTD_parameters params); /*! ZSTD_compress_usingCDict_advanced() : - * Note : this function is now DEPRECATED. + * Note : this function is now DEPRECATED. * It can be replaced by ZSTD_compress2(), in combination with ZSTD_CCtx_loadDictionary() and other parameter setters. - * This prototype will generate compilation warnings. */ -ZSTD_DEPRECATED("use ZSTD_compress2 with ZSTD_CCtx_loadDictionary") -size_t ZSTD_compress_usingCDict_advanced(ZSTD_CCtx* cctx, + * This prototype will generate compilation warnings. */ +ZSTD_DEPRECATED("use ZSTD_compress2 with ZSTD_CCtx_loadDictionary") +size_t ZSTD_compress_usingCDict_advanced(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize, const ZSTD_CDict* cdict, @@ -1769,186 +1769,186 @@ ZSTDLIB_STATIC_API size_t ZSTD_CCtx_refPrefix_advanced(ZSTD_CCtx* cctx, const vo * but compression ratio may regress significantly if guess considerably underestimates */ #define ZSTD_c_srcSizeHint ZSTD_c_experimentalParam7 -/* Controls whether the new and experimental "dedicated dictionary search - * structure" can be used. This feature is still rough around the edges, be - * prepared for surprising behavior! - * - * How to use it: - * - * When using a CDict, whether to use this feature or not is controlled at - * CDict creation, and it must be set in a CCtxParams set passed into that - * construction (via ZSTD_createCDict_advanced2()). A compression will then - * use the feature or not based on how the CDict was constructed; the value of - * this param, set in the CCtx, will have no effect. - * - * However, when a dictionary buffer is passed into a CCtx, such as via - * ZSTD_CCtx_loadDictionary(), this param can be set on the CCtx to control - * whether the CDict that is created internally can use the feature or not. - * - * What it does: - * - * Normally, the internal data structures of the CDict are analogous to what - * would be stored in a CCtx after compressing the contents of a dictionary. - * To an approximation, a compression using a dictionary can then use those - * data structures to simply continue what is effectively a streaming - * compression where the simulated compression of the dictionary left off. - * Which is to say, the search structures in the CDict are normally the same - * format as in the CCtx. - * - * It is possible to do better, since the CDict is not like a CCtx: the search - * structures are written once during CDict creation, and then are only read - * after that, while the search structures in the CCtx are both read and - * written as the compression goes along. This means we can choose a search - * structure for the dictionary that is read-optimized. - * - * This feature enables the use of that different structure. - * - * Note that some of the members of the ZSTD_compressionParameters struct have - * different semantics and constraints in the dedicated search structure. It is - * highly recommended that you simply set a compression level in the CCtxParams - * you pass into the CDict creation call, and avoid messing with the cParams - * directly. - * - * Effects: - * - * This will only have any effect when the selected ZSTD_strategy - * implementation supports this feature. Currently, that's limited to - * ZSTD_greedy, ZSTD_lazy, and ZSTD_lazy2. - * - * Note that this means that the CDict tables can no longer be copied into the - * CCtx, so the dict attachment mode ZSTD_dictForceCopy will no longer be +/* Controls whether the new and experimental "dedicated dictionary search + * structure" can be used. This feature is still rough around the edges, be + * prepared for surprising behavior! + * + * How to use it: + * + * When using a CDict, whether to use this feature or not is controlled at + * CDict creation, and it must be set in a CCtxParams set passed into that + * construction (via ZSTD_createCDict_advanced2()). A compression will then + * use the feature or not based on how the CDict was constructed; the value of + * this param, set in the CCtx, will have no effect. + * + * However, when a dictionary buffer is passed into a CCtx, such as via + * ZSTD_CCtx_loadDictionary(), this param can be set on the CCtx to control + * whether the CDict that is created internally can use the feature or not. + * + * What it does: + * + * Normally, the internal data structures of the CDict are analogous to what + * would be stored in a CCtx after compressing the contents of a dictionary. + * To an approximation, a compression using a dictionary can then use those + * data structures to simply continue what is effectively a streaming + * compression where the simulated compression of the dictionary left off. + * Which is to say, the search structures in the CDict are normally the same + * format as in the CCtx. + * + * It is possible to do better, since the CDict is not like a CCtx: the search + * structures are written once during CDict creation, and then are only read + * after that, while the search structures in the CCtx are both read and + * written as the compression goes along. This means we can choose a search + * structure for the dictionary that is read-optimized. + * + * This feature enables the use of that different structure. + * + * Note that some of the members of the ZSTD_compressionParameters struct have + * different semantics and constraints in the dedicated search structure. It is + * highly recommended that you simply set a compression level in the CCtxParams + * you pass into the CDict creation call, and avoid messing with the cParams + * directly. + * + * Effects: + * + * This will only have any effect when the selected ZSTD_strategy + * implementation supports this feature. Currently, that's limited to + * ZSTD_greedy, ZSTD_lazy, and ZSTD_lazy2. + * + * Note that this means that the CDict tables can no longer be copied into the + * CCtx, so the dict attachment mode ZSTD_dictForceCopy will no longer be * usable. The dictionary can only be attached or reloaded. - * - * In general, you should expect compression to be faster--sometimes very much - * so--and CDict creation to be slightly slower. Eventually, we will probably - * make this mode the default. - */ -#define ZSTD_c_enableDedicatedDictSearch ZSTD_c_experimentalParam8 - -/* ZSTD_c_stableInBuffer - * Experimental parameter. - * Default is 0 == disabled. Set to 1 to enable. - * - * Tells the compressor that the ZSTD_inBuffer will ALWAYS be the same - * between calls, except for the modifications that zstd makes to pos (the - * caller must not modify pos). This is checked by the compressor, and - * compression will fail if it ever changes. This means the only flush - * mode that makes sense is ZSTD_e_end, so zstd will error if ZSTD_e_end - * is not used. The data in the ZSTD_inBuffer in the range [src, src + pos) - * MUST not be modified during compression or you will get data corruption. - * - * When this flag is enabled zstd won't allocate an input window buffer, - * because the user guarantees it can reference the ZSTD_inBuffer until - * the frame is complete. But, it will still allocate an output buffer - * large enough to fit a block (see ZSTD_c_stableOutBuffer). This will also - * avoid the memcpy() from the input buffer to the input window buffer. - * - * NOTE: ZSTD_compressStream2() will error if ZSTD_e_end is not used. - * That means this flag cannot be used with ZSTD_compressStream(). - * - * NOTE: So long as the ZSTD_inBuffer always points to valid memory, using - * this flag is ALWAYS memory safe, and will never access out-of-bounds - * memory. However, compression WILL fail if you violate the preconditions. - * - * WARNING: The data in the ZSTD_inBuffer in the range [dst, dst + pos) MUST - * not be modified during compression or you will get data corruption. This - * is because zstd needs to reference data in the ZSTD_inBuffer to find - * matches. Normally zstd maintains its own window buffer for this purpose, - * but passing this flag tells zstd to use the user provided buffer. - */ -#define ZSTD_c_stableInBuffer ZSTD_c_experimentalParam9 - -/* ZSTD_c_stableOutBuffer - * Experimental parameter. - * Default is 0 == disabled. Set to 1 to enable. - * - * Tells he compressor that the ZSTD_outBuffer will not be resized between - * calls. Specifically: (out.size - out.pos) will never grow. This gives the - * compressor the freedom to say: If the compressed data doesn't fit in the - * output buffer then return ZSTD_error_dstSizeTooSmall. This allows us to - * always decompress directly into the output buffer, instead of decompressing - * into an internal buffer and copying to the output buffer. - * - * When this flag is enabled zstd won't allocate an output buffer, because - * it can write directly to the ZSTD_outBuffer. It will still allocate the - * input window buffer (see ZSTD_c_stableInBuffer). - * - * Zstd will check that (out.size - out.pos) never grows and return an error - * if it does. While not strictly necessary, this should prevent surprises. - */ -#define ZSTD_c_stableOutBuffer ZSTD_c_experimentalParam10 - -/* ZSTD_c_blockDelimiters - * Default is 0 == ZSTD_sf_noBlockDelimiters. - * - * For use with sequence compression API: ZSTD_compressSequences(). - * - * Designates whether or not the given array of ZSTD_Sequence contains block delimiters - * and last literals, which are defined as sequences with offset == 0 and matchLength == 0. - * See the definition of ZSTD_Sequence for more specifics. - */ -#define ZSTD_c_blockDelimiters ZSTD_c_experimentalParam11 - -/* ZSTD_c_validateSequences - * Default is 0 == disabled. Set to 1 to enable sequence validation. - * - * For use with sequence compression API: ZSTD_compressSequences(). - * Designates whether or not we validate sequences provided to ZSTD_compressSequences() - * during function execution. - * - * Without validation, providing a sequence that does not conform to the zstd spec will cause - * undefined behavior, and may produce a corrupted block. - * - * With validation enabled, a if sequence is invalid (see doc/zstd_compression_format.md for - * specifics regarding offset/matchlength requirements) then the function will bail out and - * return an error. - * - */ -#define ZSTD_c_validateSequences ZSTD_c_experimentalParam12 - + * + * In general, you should expect compression to be faster--sometimes very much + * so--and CDict creation to be slightly slower. Eventually, we will probably + * make this mode the default. + */ +#define ZSTD_c_enableDedicatedDictSearch ZSTD_c_experimentalParam8 + +/* ZSTD_c_stableInBuffer + * Experimental parameter. + * Default is 0 == disabled. Set to 1 to enable. + * + * Tells the compressor that the ZSTD_inBuffer will ALWAYS be the same + * between calls, except for the modifications that zstd makes to pos (the + * caller must not modify pos). This is checked by the compressor, and + * compression will fail if it ever changes. This means the only flush + * mode that makes sense is ZSTD_e_end, so zstd will error if ZSTD_e_end + * is not used. The data in the ZSTD_inBuffer in the range [src, src + pos) + * MUST not be modified during compression or you will get data corruption. + * + * When this flag is enabled zstd won't allocate an input window buffer, + * because the user guarantees it can reference the ZSTD_inBuffer until + * the frame is complete. But, it will still allocate an output buffer + * large enough to fit a block (see ZSTD_c_stableOutBuffer). This will also + * avoid the memcpy() from the input buffer to the input window buffer. + * + * NOTE: ZSTD_compressStream2() will error if ZSTD_e_end is not used. + * That means this flag cannot be used with ZSTD_compressStream(). + * + * NOTE: So long as the ZSTD_inBuffer always points to valid memory, using + * this flag is ALWAYS memory safe, and will never access out-of-bounds + * memory. However, compression WILL fail if you violate the preconditions. + * + * WARNING: The data in the ZSTD_inBuffer in the range [dst, dst + pos) MUST + * not be modified during compression or you will get data corruption. This + * is because zstd needs to reference data in the ZSTD_inBuffer to find + * matches. Normally zstd maintains its own window buffer for this purpose, + * but passing this flag tells zstd to use the user provided buffer. + */ +#define ZSTD_c_stableInBuffer ZSTD_c_experimentalParam9 + +/* ZSTD_c_stableOutBuffer + * Experimental parameter. + * Default is 0 == disabled. Set to 1 to enable. + * + * Tells he compressor that the ZSTD_outBuffer will not be resized between + * calls. Specifically: (out.size - out.pos) will never grow. This gives the + * compressor the freedom to say: If the compressed data doesn't fit in the + * output buffer then return ZSTD_error_dstSizeTooSmall. This allows us to + * always decompress directly into the output buffer, instead of decompressing + * into an internal buffer and copying to the output buffer. + * + * When this flag is enabled zstd won't allocate an output buffer, because + * it can write directly to the ZSTD_outBuffer. It will still allocate the + * input window buffer (see ZSTD_c_stableInBuffer). + * + * Zstd will check that (out.size - out.pos) never grows and return an error + * if it does. While not strictly necessary, this should prevent surprises. + */ +#define ZSTD_c_stableOutBuffer ZSTD_c_experimentalParam10 + +/* ZSTD_c_blockDelimiters + * Default is 0 == ZSTD_sf_noBlockDelimiters. + * + * For use with sequence compression API: ZSTD_compressSequences(). + * + * Designates whether or not the given array of ZSTD_Sequence contains block delimiters + * and last literals, which are defined as sequences with offset == 0 and matchLength == 0. + * See the definition of ZSTD_Sequence for more specifics. + */ +#define ZSTD_c_blockDelimiters ZSTD_c_experimentalParam11 + +/* ZSTD_c_validateSequences + * Default is 0 == disabled. Set to 1 to enable sequence validation. + * + * For use with sequence compression API: ZSTD_compressSequences(). + * Designates whether or not we validate sequences provided to ZSTD_compressSequences() + * during function execution. + * + * Without validation, providing a sequence that does not conform to the zstd spec will cause + * undefined behavior, and may produce a corrupted block. + * + * With validation enabled, a if sequence is invalid (see doc/zstd_compression_format.md for + * specifics regarding offset/matchlength requirements) then the function will bail out and + * return an error. + * + */ +#define ZSTD_c_validateSequences ZSTD_c_experimentalParam12 + /* ZSTD_c_useBlockSplitter * Controlled with ZSTD_paramSwitch_e enum. * Default is ZSTD_ps_auto. * Set to ZSTD_ps_disable to never use block splitter. * Set to ZSTD_ps_enable to always use block splitter. - * + * * By default, in ZSTD_ps_auto, the library will decide at runtime whether to use * block splitting based on the compression parameters. - */ + */ #define ZSTD_c_useBlockSplitter ZSTD_c_experimentalParam13 - -/* ZSTD_c_useRowMatchFinder + +/* ZSTD_c_useRowMatchFinder * Controlled with ZSTD_paramSwitch_e enum. * Default is ZSTD_ps_auto. * Set to ZSTD_ps_disable to never use row-based matchfinder. * Set to ZSTD_ps_enable to force usage of row-based matchfinder. - * + * * By default, in ZSTD_ps_auto, the library will decide at runtime whether to use * the row-based matchfinder based on support for SIMD instructions and the window log. * Note that this only pertains to compression strategies: greedy, lazy, and lazy2 - */ -#define ZSTD_c_useRowMatchFinder ZSTD_c_experimentalParam14 - -/* ZSTD_c_deterministicRefPrefix - * Default is 0 == disabled. Set to 1 to enable. - * - * Zstd produces different results for prefix compression when the prefix is - * directly adjacent to the data about to be compressed vs. when it isn't. - * This is because zstd detects that the two buffers are contiguous and it can - * use a more efficient match finding algorithm. However, this produces different - * results than when the two buffers are non-contiguous. This flag forces zstd - * to always load the prefix in non-contiguous mode, even if it happens to be - * adjacent to the data, to guarantee determinism. - * - * If you really care about determinism when using a dictionary or prefix, - * like when doing delta compression, you should select this option. It comes - * at a speed penalty of about ~2.5% if the dictionary and data happened to be - * contiguous, and is free if they weren't contiguous. We don't expect that - * intentionally making the dictionary and data contiguous will be worth the - * cost to memcpy() the data. - */ -#define ZSTD_c_deterministicRefPrefix ZSTD_c_experimentalParam15 - + */ +#define ZSTD_c_useRowMatchFinder ZSTD_c_experimentalParam14 + +/* ZSTD_c_deterministicRefPrefix + * Default is 0 == disabled. Set to 1 to enable. + * + * Zstd produces different results for prefix compression when the prefix is + * directly adjacent to the data about to be compressed vs. when it isn't. + * This is because zstd detects that the two buffers are contiguous and it can + * use a more efficient match finding algorithm. However, this produces different + * results than when the two buffers are non-contiguous. This flag forces zstd + * to always load the prefix in non-contiguous mode, even if it happens to be + * adjacent to the data, to guarantee determinism. + * + * If you really care about determinism when using a dictionary or prefix, + * like when doing delta compression, you should select this option. It comes + * at a speed penalty of about ~2.5% if the dictionary and data happened to be + * contiguous, and is free if they weren't contiguous. We don't expect that + * intentionally making the dictionary and data contiguous will be worth the + * cost to memcpy() the data. + */ +#define ZSTD_c_deterministicRefPrefix ZSTD_c_experimentalParam15 + /*! ZSTD_CCtx_getParameter() : * Get the requested compression parameter value, selected by enum ZSTD_cParameter, * and store it into int* value. @@ -1969,7 +1969,7 @@ ZSTDLIB_STATIC_API size_t ZSTD_CCtx_getParameter(const ZSTD_CCtx* cctx, ZSTD_cPa * These parameters will be applied to * all subsequent frames. * - ZSTD_compressStream2() : Do compression using the CCtx. - * - ZSTD_freeCCtxParams() : Free the memory, accept NULL pointer. + * - ZSTD_freeCCtxParams() : Free the memory, accept NULL pointer. * * This can be used with ZSTD_estimateCCtxSize_advanced_usingCCtxParams() * for static allocation of CCtx for single-threaded compression. @@ -1994,13 +1994,13 @@ ZSTDLIB_STATIC_API size_t ZSTD_CCtxParams_init(ZSTD_CCtx_params* cctxParams, int */ ZSTDLIB_STATIC_API size_t ZSTD_CCtxParams_init_advanced(ZSTD_CCtx_params* cctxParams, ZSTD_parameters params); -/*! ZSTD_CCtxParams_setParameter() : Requires v1.4.0+ +/*! ZSTD_CCtxParams_setParameter() : Requires v1.4.0+ * Similar to ZSTD_CCtx_setParameter. * Set one compression parameter, selected by enum ZSTD_cParameter. - * Parameters must be applied to a ZSTD_CCtx using - * ZSTD_CCtx_setParametersUsingCCtxParams(). - * @result : a code representing success or failure (which can be tested with - * ZSTD_isError()). + * Parameters must be applied to a ZSTD_CCtx using + * ZSTD_CCtx_setParametersUsingCCtxParams(). + * @result : a code representing success or failure (which can be tested with + * ZSTD_isError()). */ ZSTDLIB_STATIC_API size_t ZSTD_CCtxParams_setParameter(ZSTD_CCtx_params* params, ZSTD_cParameter param, int value); @@ -2080,13 +2080,13 @@ ZSTDLIB_STATIC_API size_t ZSTD_DCtx_refPrefix_advanced(ZSTD_DCtx* dctx, const vo */ ZSTDLIB_STATIC_API size_t ZSTD_DCtx_setMaxWindowSize(ZSTD_DCtx* dctx, size_t maxWindowSize); -/*! ZSTD_DCtx_getParameter() : - * Get the requested decompression parameter value, selected by enum ZSTD_dParameter, - * and store it into int* value. - * @return : 0, or an error code (which can be tested with ZSTD_isError()). - */ +/*! ZSTD_DCtx_getParameter() : + * Get the requested decompression parameter value, selected by enum ZSTD_dParameter, + * and store it into int* value. + * @return : 0, or an error code (which can be tested with ZSTD_isError()). + */ ZSTDLIB_STATIC_API size_t ZSTD_DCtx_getParameter(ZSTD_DCtx* dctx, ZSTD_dParameter param, int* value); - + /* ZSTD_d_format * experimental parameter, * allowing selection between ZSTD_format_e input compression formats @@ -2124,49 +2124,49 @@ ZSTDLIB_STATIC_API size_t ZSTD_DCtx_getParameter(ZSTD_DCtx* dctx, ZSTD_dParamete */ #define ZSTD_d_stableOutBuffer ZSTD_d_experimentalParam2 -/* ZSTD_d_forceIgnoreChecksum - * Experimental parameter. - * Default is 0 == disabled. Set to 1 to enable - * - * Tells the decompressor to skip checksum validation during decompression, regardless - * of whether checksumming was specified during compression. This offers some - * slight performance benefits, and may be useful for debugging. - * Param has values of type ZSTD_forceIgnoreChecksum_e - */ -#define ZSTD_d_forceIgnoreChecksum ZSTD_d_experimentalParam3 - -/* ZSTD_d_refMultipleDDicts - * Experimental parameter. - * Default is 0 == disabled. Set to 1 to enable - * - * If enabled and dctx is allocated on the heap, then additional memory will be allocated - * to store references to multiple ZSTD_DDict. That is, multiple calls of ZSTD_refDDict() - * using a given ZSTD_DCtx, rather than overwriting the previous DDict reference, will instead - * store all references. At decompression time, the appropriate dictID is selected - * from the set of DDicts based on the dictID in the frame. - * - * Usage is simply calling ZSTD_refDDict() on multiple dict buffers. - * - * Param has values of byte ZSTD_refMultipleDDicts_e - * - * WARNING: Enabling this parameter and calling ZSTD_DCtx_refDDict(), will trigger memory - * allocation for the hash table. ZSTD_freeDCtx() also frees this memory. - * Memory is allocated as per ZSTD_DCtx::customMem. - * - * Although this function allocates memory for the table, the user is still responsible for - * memory management of the underlying ZSTD_DDict* themselves. - */ -#define ZSTD_d_refMultipleDDicts ZSTD_d_experimentalParam4 - - +/* ZSTD_d_forceIgnoreChecksum + * Experimental parameter. + * Default is 0 == disabled. Set to 1 to enable + * + * Tells the decompressor to skip checksum validation during decompression, regardless + * of whether checksumming was specified during compression. This offers some + * slight performance benefits, and may be useful for debugging. + * Param has values of type ZSTD_forceIgnoreChecksum_e + */ +#define ZSTD_d_forceIgnoreChecksum ZSTD_d_experimentalParam3 + +/* ZSTD_d_refMultipleDDicts + * Experimental parameter. + * Default is 0 == disabled. Set to 1 to enable + * + * If enabled and dctx is allocated on the heap, then additional memory will be allocated + * to store references to multiple ZSTD_DDict. That is, multiple calls of ZSTD_refDDict() + * using a given ZSTD_DCtx, rather than overwriting the previous DDict reference, will instead + * store all references. At decompression time, the appropriate dictID is selected + * from the set of DDicts based on the dictID in the frame. + * + * Usage is simply calling ZSTD_refDDict() on multiple dict buffers. + * + * Param has values of byte ZSTD_refMultipleDDicts_e + * + * WARNING: Enabling this parameter and calling ZSTD_DCtx_refDDict(), will trigger memory + * allocation for the hash table. ZSTD_freeDCtx() also frees this memory. + * Memory is allocated as per ZSTD_DCtx::customMem. + * + * Although this function allocates memory for the table, the user is still responsible for + * memory management of the underlying ZSTD_DDict* themselves. + */ +#define ZSTD_d_refMultipleDDicts ZSTD_d_experimentalParam4 + + /*! ZSTD_DCtx_setFormat() : - * This function is REDUNDANT. Prefer ZSTD_DCtx_setParameter(). + * This function is REDUNDANT. Prefer ZSTD_DCtx_setParameter(). * Instruct the decoder context about what kind of data to decode next. * This instruction is mandatory to decode data without a fully-formed header, * such ZSTD_f_zstd1_magicless for example. * @return : 0, or an error code (which can be tested using ZSTD_isError()). */ -ZSTD_DEPRECATED("use ZSTD_DCtx_setParameter() instead") -size_t ZSTD_DCtx_setFormat(ZSTD_DCtx* dctx, ZSTD_format_e format); +ZSTD_DEPRECATED("use ZSTD_DCtx_setParameter() instead") +size_t ZSTD_DCtx_setFormat(ZSTD_DCtx* dctx, ZSTD_format_e format); /*! ZSTD_decompressStream_simpleArgs() : * Same as ZSTD_decompressStream(), @@ -2188,9 +2188,9 @@ ZSTDLIB_STATIC_API size_t ZSTD_decompressStream_simpleArgs ( ********************************************************************/ /*===== Advanced Streaming compression functions =====*/ - -/*! ZSTD_initCStream_srcSize() : - * This function is DEPRECATED, and equivalent to: + +/*! ZSTD_initCStream_srcSize() : + * This function is DEPRECATED, and equivalent to: * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); * ZSTD_CCtx_refCDict(zcs, NULL); // clear the dictionary (if any) * ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel); @@ -2199,15 +2199,15 @@ ZSTDLIB_STATIC_API size_t ZSTD_decompressStream_simpleArgs ( * pledgedSrcSize must be correct. If it is not known at init time, use * ZSTD_CONTENTSIZE_UNKNOWN. Note that, for compatibility with older programs, * "0" also disables frame content size field. It may be enabled in the future. - * This prototype will generate compilation warnings. + * This prototype will generate compilation warnings. */ -ZSTD_DEPRECATED("use ZSTD_CCtx_reset, see zstd.h for detailed instructions") -size_t ZSTD_initCStream_srcSize(ZSTD_CStream* zcs, +ZSTD_DEPRECATED("use ZSTD_CCtx_reset, see zstd.h for detailed instructions") +size_t ZSTD_initCStream_srcSize(ZSTD_CStream* zcs, int compressionLevel, unsigned long long pledgedSrcSize); -/*! ZSTD_initCStream_usingDict() : - * This function is DEPRECATED, and is equivalent to: +/*! ZSTD_initCStream_usingDict() : + * This function is DEPRECATED, and is equivalent to: * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); * ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel); * ZSTD_CCtx_loadDictionary(zcs, dict, dictSize); @@ -2216,15 +2216,15 @@ size_t ZSTD_initCStream_srcSize(ZSTD_CStream* zcs, * dict == NULL or dictSize < 8, in which case no dict is used. * Note: dict is loaded with ZSTD_dct_auto (treated as a full zstd dictionary if * it begins with ZSTD_MAGIC_DICTIONARY, else as raw content) and ZSTD_dlm_byCopy. - * This prototype will generate compilation warnings. + * This prototype will generate compilation warnings. */ -ZSTD_DEPRECATED("use ZSTD_CCtx_reset, see zstd.h for detailed instructions") -size_t ZSTD_initCStream_usingDict(ZSTD_CStream* zcs, +ZSTD_DEPRECATED("use ZSTD_CCtx_reset, see zstd.h for detailed instructions") +size_t ZSTD_initCStream_usingDict(ZSTD_CStream* zcs, const void* dict, size_t dictSize, int compressionLevel); -/*! ZSTD_initCStream_advanced() : - * This function is DEPRECATED, and is approximately equivalent to: +/*! ZSTD_initCStream_advanced() : + * This function is DEPRECATED, and is approximately equivalent to: * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); * // Pseudocode: Set each zstd parameter and leave the rest as-is. * for ((param, value) : params) { @@ -2236,26 +2236,26 @@ size_t ZSTD_initCStream_usingDict(ZSTD_CStream* zcs, * dict is loaded with ZSTD_dct_auto and ZSTD_dlm_byCopy. * pledgedSrcSize must be correct. * If srcSize is not known at init time, use value ZSTD_CONTENTSIZE_UNKNOWN. - * This prototype will generate compilation warnings. + * This prototype will generate compilation warnings. */ -ZSTD_DEPRECATED("use ZSTD_CCtx_reset, see zstd.h for detailed instructions") -size_t ZSTD_initCStream_advanced(ZSTD_CStream* zcs, +ZSTD_DEPRECATED("use ZSTD_CCtx_reset, see zstd.h for detailed instructions") +size_t ZSTD_initCStream_advanced(ZSTD_CStream* zcs, const void* dict, size_t dictSize, ZSTD_parameters params, unsigned long long pledgedSrcSize); -/*! ZSTD_initCStream_usingCDict() : - * This function is DEPRECATED, and equivalent to: +/*! ZSTD_initCStream_usingCDict() : + * This function is DEPRECATED, and equivalent to: * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); * ZSTD_CCtx_refCDict(zcs, cdict); * * note : cdict will just be referenced, and must outlive compression session - * This prototype will generate compilation warnings. + * This prototype will generate compilation warnings. */ -ZSTD_DEPRECATED("use ZSTD_CCtx_reset and ZSTD_CCtx_refCDict, see zstd.h for detailed instructions") -size_t ZSTD_initCStream_usingCDict(ZSTD_CStream* zcs, const ZSTD_CDict* cdict); +ZSTD_DEPRECATED("use ZSTD_CCtx_reset and ZSTD_CCtx_refCDict, see zstd.h for detailed instructions") +size_t ZSTD_initCStream_usingCDict(ZSTD_CStream* zcs, const ZSTD_CDict* cdict); -/*! ZSTD_initCStream_usingCDict_advanced() : +/*! ZSTD_initCStream_usingCDict_advanced() : * This function is DEPRECATED, and is approximately equivalent to: * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); * // Pseudocode: Set each zstd frame parameter and leave the rest as-is. @@ -2268,21 +2268,21 @@ size_t ZSTD_initCStream_usingCDict(ZSTD_CStream* zcs, const ZSTD_CDict* cdict); * same as ZSTD_initCStream_usingCDict(), with control over frame parameters. * pledgedSrcSize must be correct. If srcSize is not known at init time, use * value ZSTD_CONTENTSIZE_UNKNOWN. - * This prototype will generate compilation warnings. + * This prototype will generate compilation warnings. */ -ZSTD_DEPRECATED("use ZSTD_CCtx_reset and ZSTD_CCtx_refCDict, see zstd.h for detailed instructions") -size_t ZSTD_initCStream_usingCDict_advanced(ZSTD_CStream* zcs, +ZSTD_DEPRECATED("use ZSTD_CCtx_reset and ZSTD_CCtx_refCDict, see zstd.h for detailed instructions") +size_t ZSTD_initCStream_usingCDict_advanced(ZSTD_CStream* zcs, const ZSTD_CDict* cdict, ZSTD_frameParameters fParams, unsigned long long pledgedSrcSize); /*! ZSTD_resetCStream() : - * This function is DEPRECATED, and is equivalent to: + * This function is DEPRECATED, and is equivalent to: * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); * ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize); - * Note: ZSTD_resetCStream() interprets pledgedSrcSize == 0 as ZSTD_CONTENTSIZE_UNKNOWN, but - * ZSTD_CCtx_setPledgedSrcSize() does not do the same, so ZSTD_CONTENTSIZE_UNKNOWN must be - * explicitly specified. + * Note: ZSTD_resetCStream() interprets pledgedSrcSize == 0 as ZSTD_CONTENTSIZE_UNKNOWN, but + * ZSTD_CCtx_setPledgedSrcSize() does not do the same, so ZSTD_CONTENTSIZE_UNKNOWN must be + * explicitly specified. * * start a new frame, using same parameters from previous frame. * This is typically useful to skip dictionary loading stage, since it will re-use it in-place. @@ -2292,10 +2292,10 @@ size_t ZSTD_initCStream_usingCDict_advanced(ZSTD_CStream* zcs, * For the time being, pledgedSrcSize==0 is interpreted as "srcSize unknown" for compatibility with older programs, * but it will change to mean "empty" in future version, so use macro ZSTD_CONTENTSIZE_UNKNOWN instead. * @return : 0, or an error code (which can be tested using ZSTD_isError()) - * This prototype will generate compilation warnings. + * This prototype will generate compilation warnings. */ -ZSTD_DEPRECATED("use ZSTD_CCtx_reset, see zstd.h for detailed instructions") -size_t ZSTD_resetCStream(ZSTD_CStream* zcs, unsigned long long pledgedSrcSize); +ZSTD_DEPRECATED("use ZSTD_CCtx_reset, see zstd.h for detailed instructions") +size_t ZSTD_resetCStream(ZSTD_CStream* zcs, unsigned long long pledgedSrcSize); typedef struct { @@ -2332,8 +2332,8 @@ ZSTDLIB_STATIC_API size_t ZSTD_toFlushNow(ZSTD_CCtx* cctx); /*===== Advanced Streaming decompression functions =====*/ - -/*! + +/*! * This function is deprecated, and is equivalent to: * * ZSTD_DCtx_reset(zds, ZSTD_reset_session_only); @@ -2344,7 +2344,7 @@ ZSTDLIB_STATIC_API size_t ZSTD_toFlushNow(ZSTD_CCtx* cctx); */ ZSTDLIB_STATIC_API size_t ZSTD_initDStream_usingDict(ZSTD_DStream* zds, const void* dict, size_t dictSize); -/*! +/*! * This function is deprecated, and is equivalent to: * * ZSTD_DCtx_reset(zds, ZSTD_reset_session_only); @@ -2355,7 +2355,7 @@ ZSTDLIB_STATIC_API size_t ZSTD_initDStream_usingDict(ZSTD_DStream* zds, const vo */ ZSTDLIB_STATIC_API size_t ZSTD_initDStream_usingDDict(ZSTD_DStream* zds, const ZSTD_DDict* ddict); -/*! +/*! * This function is deprecated, and is equivalent to: * * ZSTD_DCtx_reset(zds, ZSTD_reset_session_only); @@ -2382,7 +2382,7 @@ ZSTDLIB_STATIC_API size_t ZSTD_resetDStream(ZSTD_DStream* zds); ZSTD_CCtx object can be re-used multiple times within successive compression operations. Start by initializing a context. - Use ZSTD_compressBegin(), or ZSTD_compressBegin_usingDict() for dictionary compression. + Use ZSTD_compressBegin(), or ZSTD_compressBegin_usingDict() for dictionary compression. It's also possible to duplicate a reference context which has already been initialized, using ZSTD_copyCCtx() Then, consume your input using ZSTD_compressContinue(). @@ -2413,12 +2413,12 @@ ZSTDLIB_STATIC_API size_t ZSTD_copyCCtx(ZSTD_CCtx* cctx, const ZSTD_CCtx* prepar ZSTDLIB_STATIC_API size_t ZSTD_compressContinue(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); ZSTDLIB_STATIC_API size_t ZSTD_compressEnd(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); -/* The ZSTD_compressBegin_advanced() and ZSTD_compressBegin_usingCDict_advanced() are now DEPRECATED and will generate a compiler warning */ -ZSTD_DEPRECATED("use advanced API to access custom parameters") -size_t ZSTD_compressBegin_advanced(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, ZSTD_parameters params, unsigned long long pledgedSrcSize); /**< pledgedSrcSize : If srcSize is not known at init time, use ZSTD_CONTENTSIZE_UNKNOWN */ -ZSTD_DEPRECATED("use advanced API to access custom parameters") -size_t ZSTD_compressBegin_usingCDict_advanced(ZSTD_CCtx* const cctx, const ZSTD_CDict* const cdict, ZSTD_frameParameters const fParams, unsigned long long const pledgedSrcSize); /* compression parameters are already set within cdict. pledgedSrcSize must be correct. If srcSize is not known, use macro ZSTD_CONTENTSIZE_UNKNOWN */ -/** +/* The ZSTD_compressBegin_advanced() and ZSTD_compressBegin_usingCDict_advanced() are now DEPRECATED and will generate a compiler warning */ +ZSTD_DEPRECATED("use advanced API to access custom parameters") +size_t ZSTD_compressBegin_advanced(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, ZSTD_parameters params, unsigned long long pledgedSrcSize); /**< pledgedSrcSize : If srcSize is not known at init time, use ZSTD_CONTENTSIZE_UNKNOWN */ +ZSTD_DEPRECATED("use advanced API to access custom parameters") +size_t ZSTD_compressBegin_usingCDict_advanced(ZSTD_CCtx* const cctx, const ZSTD_CDict* const cdict, ZSTD_frameParameters const fParams, unsigned long long const pledgedSrcSize); /* compression parameters are already set within cdict. pledgedSrcSize must be correct. If srcSize is not known, use macro ZSTD_CONTENTSIZE_UNKNOWN */ +/** Buffer-less streaming decompression (synchronous mode) A ZSTD_DCtx object is required to track streaming operations. diff --git a/contrib/libs/zstd/lib/zstd_errors.h b/contrib/libs/zstd/lib/zstd_errors.h index fa3686b772..fb227eeb0f 100644 --- a/contrib/libs/zstd/lib/zstd_errors.h +++ b/contrib/libs/zstd/lib/zstd_errors.h @@ -1,5 +1,5 @@ /* - * Copyright (c) Yann Collet, Facebook, Inc. + * Copyright (c) Yann Collet, Facebook, Inc. * All rights reserved. * * This source code is licensed under both the BSD-style license (found in the @@ -77,7 +77,7 @@ typedef enum { ZSTD_error_frameIndex_tooLarge = 100, ZSTD_error_seekableIO = 102, ZSTD_error_dstBuffer_wrong = 104, - ZSTD_error_srcBuffer_wrong = 105, + ZSTD_error_srcBuffer_wrong = 105, ZSTD_error_maxCode = 120 /* never EVER use this value directly, it can change in future versions! Use ZSTD_isError() instead */ } ZSTD_ErrorCode; diff --git a/contrib/libs/zstd/programs/README.md b/contrib/libs/zstd/programs/README.md index 5570f90c3b..aaff49db5f 100644 --- a/contrib/libs/zstd/programs/README.md +++ b/contrib/libs/zstd/programs/README.md @@ -1,301 +1,301 @@ -Command Line Interface for Zstandard library -============================================ - -Command Line Interface (CLI) can be created using the `make` command without any additional parameters. -There are however other Makefile targets that create different variations of CLI: -- `zstd` : default CLI supporting gzip-like arguments; includes dictionary builder, benchmark, and supports decompression of legacy zstd formats -- `zstd_nolegacy` : Same as `zstd` but without support for legacy zstd formats -- `zstd-small` : CLI optimized for minimal size; no dictionary builder, no benchmark, and no support for legacy zstd formats -- `zstd-compress` : version of CLI which can only compress into zstd format -- `zstd-decompress` : version of CLI which can only decompress zstd format - - -### Compilation variables -`zstd` scope can be altered by modifying the following `make` variables : - -- __HAVE_THREAD__ : multithreading is automatically enabled when `pthread` is detected. - It's possible to disable multithread support, by setting `HAVE_THREAD=0`. - Example : `make zstd HAVE_THREAD=0` - It's also possible to force multithread support, using `HAVE_THREAD=1`. - In which case, linking stage will fail if neither `pthread` nor `windows.h` library can be found. - This is useful to ensure this feature is not silently disabled. - -- __ZSTD_LEGACY_SUPPORT__ : `zstd` can decompress files compressed by older versions of `zstd`. - Starting v0.8.0, all versions of `zstd` produce frames compliant with the [specification](../doc/zstd_compression_format.md), and are therefore compatible. - But older versions (< v0.8.0) produced different, incompatible, frames. - By default, `zstd` supports decoding legacy formats >= v0.4.0 (`ZSTD_LEGACY_SUPPORT=4`). - This can be altered by modifying this compilation variable. - `ZSTD_LEGACY_SUPPORT=1` means "support all formats >= v0.1.0". - `ZSTD_LEGACY_SUPPORT=2` means "support all formats >= v0.2.0", and so on. - `ZSTD_LEGACY_SUPPORT=0` means _DO NOT_ support any legacy format. - if `ZSTD_LEGACY_SUPPORT >= 8`, it's the same as `0`, since there is no legacy format after `7`. - Note : `zstd` only supports decoding older formats, and cannot generate any legacy format. - -- __HAVE_ZLIB__ : `zstd` can compress and decompress files in `.gz` format. - This is ordered through command `--format=gzip`. - Alternatively, symlinks named `gzip` or `gunzip` will mimic intended behavior. - `.gz` support is automatically enabled when `zlib` library is detected at build time. - It's possible to disable `.gz` support, by setting `HAVE_ZLIB=0`. - Example : `make zstd HAVE_ZLIB=0` - It's also possible to force compilation with zlib support, using `HAVE_ZLIB=1`. - In which case, linking stage will fail if `zlib` library cannot be found. - This is useful to prevent silent feature disabling. - -- __HAVE_LZMA__ : `zstd` can compress and decompress files in `.xz` and `.lzma` formats. - This is ordered through commands `--format=xz` and `--format=lzma` respectively. - Alternatively, symlinks named `xz`, `unxz`, `lzma`, or `unlzma` will mimic intended behavior. - `.xz` and `.lzma` support is automatically enabled when `lzma` library is detected at build time. - It's possible to disable `.xz` and `.lzma` support, by setting `HAVE_LZMA=0`. - Example : `make zstd HAVE_LZMA=0` - It's also possible to force compilation with lzma support, using `HAVE_LZMA=1`. - In which case, linking stage will fail if `lzma` library cannot be found. - This is useful to prevent silent feature disabling. - -- __HAVE_LZ4__ : `zstd` can compress and decompress files in `.lz4` formats. - This is ordered through commands `--format=lz4`. - Alternatively, symlinks named `lz4`, or `unlz4` will mimic intended behavior. - `.lz4` support is automatically enabled when `lz4` library is detected at build time. - It's possible to disable `.lz4` support, by setting `HAVE_LZ4=0` . - Example : `make zstd HAVE_LZ4=0` - It's also possible to force compilation with lz4 support, using `HAVE_LZ4=1`. - In which case, linking stage will fail if `lz4` library cannot be found. - This is useful to prevent silent feature disabling. - -- __ZSTD_NOBENCH__ : `zstd` cli will be compiled without its integrated benchmark module. - This can be useful to produce smaller binaries. - In this case, the corresponding unit can also be excluded from compilation target. - -- __ZSTD_NODICT__ : `zstd` cli will be compiled without support for the integrated dictionary builder. - This can be useful to produce smaller binaries. - In this case, the corresponding unit can also be excluded from compilation target. - -- __ZSTD_NOCOMPRESS__ : `zstd` cli will be compiled without support for compression. - The resulting binary will only be able to decompress files. - This can be useful to produce smaller binaries. - A corresponding `Makefile` target using this ability is `zstd-decompress`. - -- __ZSTD_NODECOMPRESS__ : `zstd` cli will be compiled without support for decompression. - The resulting binary will only be able to compress files. - This can be useful to produce smaller binaries. - A corresponding `Makefile` target using this ability is `zstd-compress`. - -- __BACKTRACE__ : `zstd` can display a stack backtrace when execution - generates a runtime exception. By default, this feature may be - degraded/disabled on some platforms unless additional compiler directives are - applied. When triaging a runtime issue, enabling this feature can provide - more context to determine the location of the fault. - Example : `make zstd BACKTRACE=1` - - -### Aggregation of parameters -CLI supports aggregation of parameters i.e. `-b1`, `-e18`, and `-i1` can be joined into `-b1e18i1`. - - -### Symlink shortcuts -It's possible to invoke `zstd` through a symlink. -When the name of the symlink has a specific value, it triggers an associated behavior. -- `zstdmt` : compress using all cores available on local system. -- `zcat` : will decompress and output target file using any of the supported formats. `gzcat` and `zstdcat` are also equivalent. -- `gzip` : if zlib support is enabled, will mimic `gzip` by compressing file using `.gz` format, removing source file by default (use `--keep` to preserve). If zlib is not supported, triggers an error. -- `xz` : if lzma support is enabled, will mimic `xz` by compressing file using `.xz` format, removing source file by default (use `--keep` to preserve). If xz is not supported, triggers an error. -- `lzma` : if lzma support is enabled, will mimic `lzma` by compressing file using `.lzma` format, removing source file by default (use `--keep` to preserve). If lzma is not supported, triggers an error. -- `lz4` : if lz4 support is enabled, will mimic `lz4` by compressing file using `.lz4` format. If lz4 is not supported, triggers an error. -- `unzstd` and `unlz4` will decompress any of the supported format. -- `ungz`, `unxz` and `unlzma` will do the same, and will also remove source file by default (use `--keep` to preserve). - - -### Dictionary builder in Command Line Interface -Zstd offers a training mode, which can be used to tune the algorithm for a selected -type of data, by providing it with a few samples. The result of the training is stored -in a file selected with the `-o` option (default name is `dictionary`), -which can be loaded before compression and decompression. - -Using a dictionary, the compression ratio achievable on small data improves dramatically. -These compression gains are achieved while simultaneously providing faster compression and decompression speeds. -Dictionary work if there is some correlation in a family of small data (there is no universal dictionary). -Hence, deploying one dictionary per type of data will provide the greater benefits. -Dictionary gains are mostly effective in the first few KB. Then, the compression algorithm -will rely more and more on previously decoded content to compress the rest of the file. - -Usage of the dictionary builder and created dictionaries with CLI: - -1. Create the dictionary : `zstd --train PathToTrainingSet/* -o dictionaryName` -2. Compress with the dictionary: `zstd FILE -D dictionaryName` -3. Decompress with the dictionary: `zstd --decompress FILE.zst -D dictionaryName` - - -### Benchmark in Command Line Interface -CLI includes in-memory compression benchmark module for zstd. -The benchmark is conducted using given filenames. The files are read into memory and joined together. -It makes benchmark more precise as it eliminates I/O overhead. -Multiple filenames can be supplied, as multiple parameters, with wildcards, -or names of directories can be used as parameters with `-r` option. - -The benchmark measures ratio, compressed size, compression and decompression speed. -One can select compression levels starting from `-b` and ending with `-e`. -The `-i` parameter selects minimal time used for each of tested levels. - - -### Usage of Command Line Interface -The full list of options can be obtained with `-h` or `-H` parameter: -``` -Usage : - zstd [args] [FILE(s)] [-o file] - -FILE : a filename - with no FILE, or when FILE is - , read standard input -Arguments : - -# : # compression level (1-19, default: 3) - -d : decompression - -D DICT: use DICT as Dictionary for compression or decompression - -o file: result stored into `file` (only 1 output file) - -f : overwrite output without prompting, also (de)compress links ---rm : remove source file(s) after successful de/compression - -k : preserve source file(s) (default) - -h/-H : display help/long help and exit - -Advanced arguments : - -V : display Version number and exit +Command Line Interface for Zstandard library +============================================ + +Command Line Interface (CLI) can be created using the `make` command without any additional parameters. +There are however other Makefile targets that create different variations of CLI: +- `zstd` : default CLI supporting gzip-like arguments; includes dictionary builder, benchmark, and supports decompression of legacy zstd formats +- `zstd_nolegacy` : Same as `zstd` but without support for legacy zstd formats +- `zstd-small` : CLI optimized for minimal size; no dictionary builder, no benchmark, and no support for legacy zstd formats +- `zstd-compress` : version of CLI which can only compress into zstd format +- `zstd-decompress` : version of CLI which can only decompress zstd format + + +### Compilation variables +`zstd` scope can be altered by modifying the following `make` variables : + +- __HAVE_THREAD__ : multithreading is automatically enabled when `pthread` is detected. + It's possible to disable multithread support, by setting `HAVE_THREAD=0`. + Example : `make zstd HAVE_THREAD=0` + It's also possible to force multithread support, using `HAVE_THREAD=1`. + In which case, linking stage will fail if neither `pthread` nor `windows.h` library can be found. + This is useful to ensure this feature is not silently disabled. + +- __ZSTD_LEGACY_SUPPORT__ : `zstd` can decompress files compressed by older versions of `zstd`. + Starting v0.8.0, all versions of `zstd` produce frames compliant with the [specification](../doc/zstd_compression_format.md), and are therefore compatible. + But older versions (< v0.8.0) produced different, incompatible, frames. + By default, `zstd` supports decoding legacy formats >= v0.4.0 (`ZSTD_LEGACY_SUPPORT=4`). + This can be altered by modifying this compilation variable. + `ZSTD_LEGACY_SUPPORT=1` means "support all formats >= v0.1.0". + `ZSTD_LEGACY_SUPPORT=2` means "support all formats >= v0.2.0", and so on. + `ZSTD_LEGACY_SUPPORT=0` means _DO NOT_ support any legacy format. + if `ZSTD_LEGACY_SUPPORT >= 8`, it's the same as `0`, since there is no legacy format after `7`. + Note : `zstd` only supports decoding older formats, and cannot generate any legacy format. + +- __HAVE_ZLIB__ : `zstd` can compress and decompress files in `.gz` format. + This is ordered through command `--format=gzip`. + Alternatively, symlinks named `gzip` or `gunzip` will mimic intended behavior. + `.gz` support is automatically enabled when `zlib` library is detected at build time. + It's possible to disable `.gz` support, by setting `HAVE_ZLIB=0`. + Example : `make zstd HAVE_ZLIB=0` + It's also possible to force compilation with zlib support, using `HAVE_ZLIB=1`. + In which case, linking stage will fail if `zlib` library cannot be found. + This is useful to prevent silent feature disabling. + +- __HAVE_LZMA__ : `zstd` can compress and decompress files in `.xz` and `.lzma` formats. + This is ordered through commands `--format=xz` and `--format=lzma` respectively. + Alternatively, symlinks named `xz`, `unxz`, `lzma`, or `unlzma` will mimic intended behavior. + `.xz` and `.lzma` support is automatically enabled when `lzma` library is detected at build time. + It's possible to disable `.xz` and `.lzma` support, by setting `HAVE_LZMA=0`. + Example : `make zstd HAVE_LZMA=0` + It's also possible to force compilation with lzma support, using `HAVE_LZMA=1`. + In which case, linking stage will fail if `lzma` library cannot be found. + This is useful to prevent silent feature disabling. + +- __HAVE_LZ4__ : `zstd` can compress and decompress files in `.lz4` formats. + This is ordered through commands `--format=lz4`. + Alternatively, symlinks named `lz4`, or `unlz4` will mimic intended behavior. + `.lz4` support is automatically enabled when `lz4` library is detected at build time. + It's possible to disable `.lz4` support, by setting `HAVE_LZ4=0` . + Example : `make zstd HAVE_LZ4=0` + It's also possible to force compilation with lz4 support, using `HAVE_LZ4=1`. + In which case, linking stage will fail if `lz4` library cannot be found. + This is useful to prevent silent feature disabling. + +- __ZSTD_NOBENCH__ : `zstd` cli will be compiled without its integrated benchmark module. + This can be useful to produce smaller binaries. + In this case, the corresponding unit can also be excluded from compilation target. + +- __ZSTD_NODICT__ : `zstd` cli will be compiled without support for the integrated dictionary builder. + This can be useful to produce smaller binaries. + In this case, the corresponding unit can also be excluded from compilation target. + +- __ZSTD_NOCOMPRESS__ : `zstd` cli will be compiled without support for compression. + The resulting binary will only be able to decompress files. + This can be useful to produce smaller binaries. + A corresponding `Makefile` target using this ability is `zstd-decompress`. + +- __ZSTD_NODECOMPRESS__ : `zstd` cli will be compiled without support for decompression. + The resulting binary will only be able to compress files. + This can be useful to produce smaller binaries. + A corresponding `Makefile` target using this ability is `zstd-compress`. + +- __BACKTRACE__ : `zstd` can display a stack backtrace when execution + generates a runtime exception. By default, this feature may be + degraded/disabled on some platforms unless additional compiler directives are + applied. When triaging a runtime issue, enabling this feature can provide + more context to determine the location of the fault. + Example : `make zstd BACKTRACE=1` + + +### Aggregation of parameters +CLI supports aggregation of parameters i.e. `-b1`, `-e18`, and `-i1` can be joined into `-b1e18i1`. + + +### Symlink shortcuts +It's possible to invoke `zstd` through a symlink. +When the name of the symlink has a specific value, it triggers an associated behavior. +- `zstdmt` : compress using all cores available on local system. +- `zcat` : will decompress and output target file using any of the supported formats. `gzcat` and `zstdcat` are also equivalent. +- `gzip` : if zlib support is enabled, will mimic `gzip` by compressing file using `.gz` format, removing source file by default (use `--keep` to preserve). If zlib is not supported, triggers an error. +- `xz` : if lzma support is enabled, will mimic `xz` by compressing file using `.xz` format, removing source file by default (use `--keep` to preserve). If xz is not supported, triggers an error. +- `lzma` : if lzma support is enabled, will mimic `lzma` by compressing file using `.lzma` format, removing source file by default (use `--keep` to preserve). If lzma is not supported, triggers an error. +- `lz4` : if lz4 support is enabled, will mimic `lz4` by compressing file using `.lz4` format. If lz4 is not supported, triggers an error. +- `unzstd` and `unlz4` will decompress any of the supported format. +- `ungz`, `unxz` and `unlzma` will do the same, and will also remove source file by default (use `--keep` to preserve). + + +### Dictionary builder in Command Line Interface +Zstd offers a training mode, which can be used to tune the algorithm for a selected +type of data, by providing it with a few samples. The result of the training is stored +in a file selected with the `-o` option (default name is `dictionary`), +which can be loaded before compression and decompression. + +Using a dictionary, the compression ratio achievable on small data improves dramatically. +These compression gains are achieved while simultaneously providing faster compression and decompression speeds. +Dictionary work if there is some correlation in a family of small data (there is no universal dictionary). +Hence, deploying one dictionary per type of data will provide the greater benefits. +Dictionary gains are mostly effective in the first few KB. Then, the compression algorithm +will rely more and more on previously decoded content to compress the rest of the file. + +Usage of the dictionary builder and created dictionaries with CLI: + +1. Create the dictionary : `zstd --train PathToTrainingSet/* -o dictionaryName` +2. Compress with the dictionary: `zstd FILE -D dictionaryName` +3. Decompress with the dictionary: `zstd --decompress FILE.zst -D dictionaryName` + + +### Benchmark in Command Line Interface +CLI includes in-memory compression benchmark module for zstd. +The benchmark is conducted using given filenames. The files are read into memory and joined together. +It makes benchmark more precise as it eliminates I/O overhead. +Multiple filenames can be supplied, as multiple parameters, with wildcards, +or names of directories can be used as parameters with `-r` option. + +The benchmark measures ratio, compressed size, compression and decompression speed. +One can select compression levels starting from `-b` and ending with `-e`. +The `-i` parameter selects minimal time used for each of tested levels. + + +### Usage of Command Line Interface +The full list of options can be obtained with `-h` or `-H` parameter: +``` +Usage : + zstd [args] [FILE(s)] [-o file] + +FILE : a filename + with no FILE, or when FILE is - , read standard input +Arguments : + -# : # compression level (1-19, default: 3) + -d : decompression + -D DICT: use DICT as Dictionary for compression or decompression + -o file: result stored into `file` (only 1 output file) + -f : overwrite output without prompting, also (de)compress links +--rm : remove source file(s) after successful de/compression + -k : preserve source file(s) (default) + -h/-H : display help/long help and exit + +Advanced arguments : + -V : display Version number and exit -c : write to standard output (even if it is the console) - -v : verbose mode; specify multiple times to increase verbosity - -q : suppress warnings; specify twice to suppress errors too ---no-progress : do not display the progress counter - -r : operate recursively on directories ---filelist FILE : read list of files to operate upon from FILE ---output-dir-flat DIR : processed files are stored into DIR ---output-dir-mirror DIR : processed files are stored into DIR respecting original directory structure ---[no-]check : during compression, add XXH64 integrity checksum to frame (default: enabled). If specified with -d, decompressor will ignore/validate checksums in compressed frame (default: validate). --- : All arguments after "--" are treated as files - -Advanced compression arguments : ---ultra : enable levels beyond 19, up to 22 (requires more memory) ---long[=#]: enable long distance matching with given window log (default: 27) ---fast[=#]: switch to very fast compression levels (default: 1) ---adapt : dynamically adapt compression level to I/O conditions + -v : verbose mode; specify multiple times to increase verbosity + -q : suppress warnings; specify twice to suppress errors too +--no-progress : do not display the progress counter + -r : operate recursively on directories +--filelist FILE : read list of files to operate upon from FILE +--output-dir-flat DIR : processed files are stored into DIR +--output-dir-mirror DIR : processed files are stored into DIR respecting original directory structure +--[no-]check : during compression, add XXH64 integrity checksum to frame (default: enabled). If specified with -d, decompressor will ignore/validate checksums in compressed frame (default: validate). +-- : All arguments after "--" are treated as files + +Advanced compression arguments : +--ultra : enable levels beyond 19, up to 22 (requires more memory) +--long[=#]: enable long distance matching with given window log (default: 27) +--fast[=#]: switch to very fast compression levels (default: 1) +--adapt : dynamically adapt compression level to I/O conditions --patch-from=FILE : specify the file to be used as a reference point for zstd's diff engine - -T# : spawns # compression threads (default: 1, 0==# cores) - -B# : select size of each job (default: 0==automatic) ---single-thread : use a single thread for both I/O and compression (result slightly different than -T1) ---rsyncable : compress using a rsync-friendly method (-B sets block size) ---exclude-compressed: only compress files that are not already compressed ---stream-size=# : specify size of streaming input from `stdin` ---size-hint=# optimize compression parameters for streaming input of approximately this size ---target-compressed-block-size=# : generate compressed block of approximately targeted size ---no-dictID : don't write dictID into header (dictionary compression only) ---[no-]compress-literals : force (un)compressed literals ---format=zstd : compress files to the .zst format (default) ---format=gzip : compress files to the .gz format ---format=xz : compress files to the .xz format ---format=lzma : compress files to the .lzma format ---format=lz4 : compress files to the .lz4 format - -Advanced decompression arguments : - -l : print information about zstd compressed files ---test : test compressed file integrity - -M# : Set a memory usage limit for decompression ---[no-]sparse : sparse mode (default: disabled) - -Dictionary builder : ---train ## : create a dictionary from a training set of files ---train-cover[=k=#,d=#,steps=#,split=#,shrink[=#]] : use the cover algorithm with optional args ---train-fastcover[=k=#,d=#,f=#,steps=#,split=#,accel=#,shrink[=#]] : use the fast cover algorithm with optional args ---train-legacy[=s=#] : use the legacy algorithm with selectivity (default: 9) - -o DICT : DICT is dictionary name (default: dictionary) ---maxdict=# : limit dictionary to specified size (default: 112640) ---dictID=# : force dictionary ID to specified value (default: random) - -Benchmark arguments : - -b# : benchmark file(s), using # compression level (default: 3) - -e# : test all compression levels successively from -b# to -e# (default: 1) - -i# : minimum evaluation time in seconds (default: 3s) - -B# : cut file into independent blocks of size # (default: no block) - -S : output one benchmark result per input file (default: consolidated result) ---priority=rt : set process priority to real-time -``` - -### Passing parameters through Environment Variables -There is no "generic" way to pass "any kind of parameter" to `zstd` in a pass-through manner. -Using environment variables for this purpose has security implications. -Therefore, this avenue is intentionally restricted and only supports `ZSTD_CLEVEL` and `ZSTD_NBTHREADS`. - -`ZSTD_CLEVEL` can be used to modify the default compression level of `zstd` -(usually set to `3`) to another value between 1 and 19 (the "normal" range). - -`ZSTD_NBTHREADS` can be used to specify a number of threads -that `zstd` will use for compression, which by default is `1`. -This functionality only exists when `zstd` is compiled with multithread support. -`0` means "use as many threads as detected cpu cores on local system". -The max # of threads is capped at `ZSTDMT_NBWORKERS_MAX`, -which is either 64 in 32-bit mode, or 256 for 64-bit environments. - -This functionality can be useful when `zstd` CLI is invoked in a way that doesn't allow passing arguments. -One such scenario is `tar --zstd`. -As `ZSTD_CLEVEL` and `ZSTD_NBTHREADS` only replace the default compression level -and number of threads respectively, they can both be overridden by corresponding command line arguments: -`-#` for compression level and `-T#` for number of threads. - - -### Long distance matching mode -The long distance matching mode, enabled with `--long`, is designed to improve -the compression ratio for files with long matches at a large distance (up to the -maximum window size, `128 MiB`) while still maintaining compression speed. - -Enabling this mode sets the window size to `128 MiB` and thus increases the memory -usage for both the compressor and decompressor. Performance in terms of speed is -dependent on long matches being found. Compression speed may degrade if few long -matches are found. Decompression speed usually improves when there are many long -distance matches. - -Below are graphs comparing the compression speed, compression ratio, and -decompression speed with and without long distance matching on an ideal use -case: a tar of four versions of clang (versions `3.4.1`, `3.4.2`, `3.5.0`, -`3.5.1`) with a total size of `244889600 B`. This is an ideal use case as there -are many long distance matches within the maximum window size of `128 MiB` (each -version is less than `128 MiB`). - -Compression Speed vs Ratio | Decompression Speed ----------------------------|--------------------- - |  - -| Method | Compression ratio | Compression speed | Decompression speed | -|:-------|------------------:|-------------------------:|---------------------------:| -| `zstd -1` | `5.065` | `284.8 MB/s` | `759.3 MB/s` | -| `zstd -5` | `5.826` | `124.9 MB/s` | `674.0 MB/s` | -| `zstd -10` | `6.504` | `29.5 MB/s` | `771.3 MB/s` | -| `zstd -1 --long` | `17.426` | `220.6 MB/s` | `1638.4 MB/s` | -| `zstd -5 --long` | `19.661` | `165.5 MB/s` | `1530.6 MB/s` | -| `zstd -10 --long`| `21.949` | `75.6 MB/s` | `1632.6 MB/s` | - -On this file, the compression ratio improves significantly with minimal impact -on compression speed, and the decompression speed doubles. - -On the other extreme, compressing a file with few long distance matches (such as -the [Silesia compression corpus]) will likely lead to a deterioration in -compression speed (for lower levels) with minimal change in compression ratio. - -The below table illustrates this on the [Silesia compression corpus]. - -[Silesia compression corpus]: http://sun.aei.polsl.pl/~sdeor/index.php?page=silesia - -| Method | Compression ratio | Compression speed | Decompression speed | -|:-------|------------------:|------------------:|---------------------:| -| `zstd -1` | `2.878` | `231.7 MB/s` | `594.4 MB/s` | -| `zstd -1 --long` | `2.929` | `106.5 MB/s` | `517.9 MB/s` | -| `zstd -5` | `3.274` | `77.1 MB/s` | `464.2 MB/s` | -| `zstd -5 --long` | `3.319` | `51.7 MB/s` | `371.9 MB/s` | -| `zstd -10` | `3.523` | `16.4 MB/s` | `489.2 MB/s` | -| `zstd -10 --long`| `3.566` | `16.2 MB/s` | `415.7 MB/s` | - - -### zstdgrep - -`zstdgrep` is a utility which makes it possible to `grep` directly a `.zst` compressed file. -It's used the same way as normal `grep`, for example : -`zstdgrep pattern file.zst` - -`zstdgrep` is _not_ compatible with dictionary compression. - -To search into a file compressed with a dictionary, -it's necessary to decompress it using `zstd` or `zstdcat`, -and then pipe the result to `grep`. For example : -`zstdcat -D dictionary -qc -- file.zst | grep pattern` + -T# : spawns # compression threads (default: 1, 0==# cores) + -B# : select size of each job (default: 0==automatic) +--single-thread : use a single thread for both I/O and compression (result slightly different than -T1) +--rsyncable : compress using a rsync-friendly method (-B sets block size) +--exclude-compressed: only compress files that are not already compressed +--stream-size=# : specify size of streaming input from `stdin` +--size-hint=# optimize compression parameters for streaming input of approximately this size +--target-compressed-block-size=# : generate compressed block of approximately targeted size +--no-dictID : don't write dictID into header (dictionary compression only) +--[no-]compress-literals : force (un)compressed literals +--format=zstd : compress files to the .zst format (default) +--format=gzip : compress files to the .gz format +--format=xz : compress files to the .xz format +--format=lzma : compress files to the .lzma format +--format=lz4 : compress files to the .lz4 format + +Advanced decompression arguments : + -l : print information about zstd compressed files +--test : test compressed file integrity + -M# : Set a memory usage limit for decompression +--[no-]sparse : sparse mode (default: disabled) + +Dictionary builder : +--train ## : create a dictionary from a training set of files +--train-cover[=k=#,d=#,steps=#,split=#,shrink[=#]] : use the cover algorithm with optional args +--train-fastcover[=k=#,d=#,f=#,steps=#,split=#,accel=#,shrink[=#]] : use the fast cover algorithm with optional args +--train-legacy[=s=#] : use the legacy algorithm with selectivity (default: 9) + -o DICT : DICT is dictionary name (default: dictionary) +--maxdict=# : limit dictionary to specified size (default: 112640) +--dictID=# : force dictionary ID to specified value (default: random) + +Benchmark arguments : + -b# : benchmark file(s), using # compression level (default: 3) + -e# : test all compression levels successively from -b# to -e# (default: 1) + -i# : minimum evaluation time in seconds (default: 3s) + -B# : cut file into independent blocks of size # (default: no block) + -S : output one benchmark result per input file (default: consolidated result) +--priority=rt : set process priority to real-time +``` + +### Passing parameters through Environment Variables +There is no "generic" way to pass "any kind of parameter" to `zstd` in a pass-through manner. +Using environment variables for this purpose has security implications. +Therefore, this avenue is intentionally restricted and only supports `ZSTD_CLEVEL` and `ZSTD_NBTHREADS`. + +`ZSTD_CLEVEL` can be used to modify the default compression level of `zstd` +(usually set to `3`) to another value between 1 and 19 (the "normal" range). + +`ZSTD_NBTHREADS` can be used to specify a number of threads +that `zstd` will use for compression, which by default is `1`. +This functionality only exists when `zstd` is compiled with multithread support. +`0` means "use as many threads as detected cpu cores on local system". +The max # of threads is capped at `ZSTDMT_NBWORKERS_MAX`, +which is either 64 in 32-bit mode, or 256 for 64-bit environments. + +This functionality can be useful when `zstd` CLI is invoked in a way that doesn't allow passing arguments. +One such scenario is `tar --zstd`. +As `ZSTD_CLEVEL` and `ZSTD_NBTHREADS` only replace the default compression level +and number of threads respectively, they can both be overridden by corresponding command line arguments: +`-#` for compression level and `-T#` for number of threads. + + +### Long distance matching mode +The long distance matching mode, enabled with `--long`, is designed to improve +the compression ratio for files with long matches at a large distance (up to the +maximum window size, `128 MiB`) while still maintaining compression speed. + +Enabling this mode sets the window size to `128 MiB` and thus increases the memory +usage for both the compressor and decompressor. Performance in terms of speed is +dependent on long matches being found. Compression speed may degrade if few long +matches are found. Decompression speed usually improves when there are many long +distance matches. + +Below are graphs comparing the compression speed, compression ratio, and +decompression speed with and without long distance matching on an ideal use +case: a tar of four versions of clang (versions `3.4.1`, `3.4.2`, `3.5.0`, +`3.5.1`) with a total size of `244889600 B`. This is an ideal use case as there +are many long distance matches within the maximum window size of `128 MiB` (each +version is less than `128 MiB`). + +Compression Speed vs Ratio | Decompression Speed +---------------------------|--------------------- + |  + +| Method | Compression ratio | Compression speed | Decompression speed | +|:-------|------------------:|-------------------------:|---------------------------:| +| `zstd -1` | `5.065` | `284.8 MB/s` | `759.3 MB/s` | +| `zstd -5` | `5.826` | `124.9 MB/s` | `674.0 MB/s` | +| `zstd -10` | `6.504` | `29.5 MB/s` | `771.3 MB/s` | +| `zstd -1 --long` | `17.426` | `220.6 MB/s` | `1638.4 MB/s` | +| `zstd -5 --long` | `19.661` | `165.5 MB/s` | `1530.6 MB/s` | +| `zstd -10 --long`| `21.949` | `75.6 MB/s` | `1632.6 MB/s` | + +On this file, the compression ratio improves significantly with minimal impact +on compression speed, and the decompression speed doubles. + +On the other extreme, compressing a file with few long distance matches (such as +the [Silesia compression corpus]) will likely lead to a deterioration in +compression speed (for lower levels) with minimal change in compression ratio. + +The below table illustrates this on the [Silesia compression corpus]. + +[Silesia compression corpus]: http://sun.aei.polsl.pl/~sdeor/index.php?page=silesia + +| Method | Compression ratio | Compression speed | Decompression speed | +|:-------|------------------:|------------------:|---------------------:| +| `zstd -1` | `2.878` | `231.7 MB/s` | `594.4 MB/s` | +| `zstd -1 --long` | `2.929` | `106.5 MB/s` | `517.9 MB/s` | +| `zstd -5` | `3.274` | `77.1 MB/s` | `464.2 MB/s` | +| `zstd -5 --long` | `3.319` | `51.7 MB/s` | `371.9 MB/s` | +| `zstd -10` | `3.523` | `16.4 MB/s` | `489.2 MB/s` | +| `zstd -10 --long`| `3.566` | `16.2 MB/s` | `415.7 MB/s` | + + +### zstdgrep + +`zstdgrep` is a utility which makes it possible to `grep` directly a `.zst` compressed file. +It's used the same way as normal `grep`, for example : +`zstdgrep pattern file.zst` + +`zstdgrep` is _not_ compatible with dictionary compression. + +To search into a file compressed with a dictionary, +it's necessary to decompress it using `zstd` or `zstdcat`, +and then pipe the result to `grep`. For example : +`zstdcat -D dictionary -qc -- file.zst | grep pattern` diff --git a/contrib/libs/zstd/programs/benchfn.c b/contrib/libs/zstd/programs/benchfn.c index 1aadbdd913..70855f1197 100644 --- a/contrib/libs/zstd/programs/benchfn.c +++ b/contrib/libs/zstd/programs/benchfn.c @@ -1,256 +1,256 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - - -/* ************************************* -* Includes -***************************************/ -#include <stdlib.h> /* malloc, free */ -#include <string.h> /* memset */ -#include <assert.h> /* assert */ - -#include "timefn.h" /* UTIL_time_t, UTIL_getTime */ -#include "benchfn.h" - - -/* ************************************* -* Constants -***************************************/ -#define TIMELOOP_MICROSEC SEC_TO_MICRO /* 1 second */ -#define TIMELOOP_NANOSEC (1*1000000000ULL) /* 1 second */ - -#define KB *(1 <<10) -#define MB *(1 <<20) -#define GB *(1U<<30) - - -/* ************************************* -* Debug errors -***************************************/ -#if defined(DEBUG) && (DEBUG >= 1) -# include <stdio.h> /* fprintf */ -# define DISPLAY(...) fprintf(stderr, __VA_ARGS__) -# define DEBUGOUTPUT(...) { if (DEBUG) DISPLAY(__VA_ARGS__); } -#else -# define DEBUGOUTPUT(...) -#endif - - -/* error without displaying */ -#define RETURN_QUIET_ERROR(retValue, ...) { \ - DEBUGOUTPUT("%s: %i: \n", __FILE__, __LINE__); \ - DEBUGOUTPUT("Error : "); \ - DEBUGOUTPUT(__VA_ARGS__); \ - DEBUGOUTPUT(" \n"); \ - return retValue; \ -} - -/* Abort execution if a condition is not met */ -#define CONTROL(c) { if (!(c)) { DEBUGOUTPUT("error: %s \n", #c); abort(); } } - - -/* ************************************* -* Benchmarking an arbitrary function -***************************************/ - -int BMK_isSuccessful_runOutcome(BMK_runOutcome_t outcome) -{ - return outcome.error_tag_never_ever_use_directly == 0; -} - -/* warning : this function will stop program execution if outcome is invalid ! - * check outcome validity first, using BMK_isValid_runResult() */ -BMK_runTime_t BMK_extract_runTime(BMK_runOutcome_t outcome) -{ - CONTROL(outcome.error_tag_never_ever_use_directly == 0); - return outcome.internal_never_ever_use_directly; -} - -size_t BMK_extract_errorResult(BMK_runOutcome_t outcome) -{ - CONTROL(outcome.error_tag_never_ever_use_directly != 0); - return outcome.error_result_never_ever_use_directly; -} - -static BMK_runOutcome_t BMK_runOutcome_error(size_t errorResult) -{ - BMK_runOutcome_t b; - memset(&b, 0, sizeof(b)); - b.error_tag_never_ever_use_directly = 1; - b.error_result_never_ever_use_directly = errorResult; - return b; -} - -static BMK_runOutcome_t BMK_setValid_runTime(BMK_runTime_t runTime) -{ - BMK_runOutcome_t outcome; - outcome.error_tag_never_ever_use_directly = 0; - outcome.internal_never_ever_use_directly = runTime; - return outcome; -} - - -/* initFn will be measured once, benchFn will be measured `nbLoops` times */ -/* initFn is optional, provide NULL if none */ -/* benchFn must return a size_t value that errorFn can interpret */ -/* takes # of blocks and list of size & stuff for each. */ -/* can report result of benchFn for each block into blockResult. */ -/* blockResult is optional, provide NULL if this information is not required */ -/* note : time per loop can be reported as zero if run time < timer resolution */ -BMK_runOutcome_t BMK_benchFunction(BMK_benchParams_t p, - unsigned nbLoops) -{ - size_t dstSize = 0; - nbLoops += !nbLoops; /* minimum nbLoops is 1 */ - - /* init */ - { size_t i; - for(i = 0; i < p.blockCount; i++) { - memset(p.dstBuffers[i], 0xE5, p.dstCapacities[i]); /* warm up and erase result buffer */ - } } - - /* benchmark */ - { UTIL_time_t const clockStart = UTIL_getTime(); - unsigned loopNb, blockNb; - if (p.initFn != NULL) p.initFn(p.initPayload); - for (loopNb = 0; loopNb < nbLoops; loopNb++) { - for (blockNb = 0; blockNb < p.blockCount; blockNb++) { - size_t const res = p.benchFn(p.srcBuffers[blockNb], p.srcSizes[blockNb], - p.dstBuffers[blockNb], p.dstCapacities[blockNb], - p.benchPayload); - if (loopNb == 0) { - if (p.blockResults != NULL) p.blockResults[blockNb] = res; - if ((p.errorFn != NULL) && (p.errorFn(res))) { - RETURN_QUIET_ERROR(BMK_runOutcome_error(res), - "Function benchmark failed on block %u (of size %u) with error %i", - blockNb, (unsigned)p.srcSizes[blockNb], (int)res); - } - dstSize += res; - } } - } /* for (loopNb = 0; loopNb < nbLoops; loopNb++) */ - - { PTime const totalTime = UTIL_clockSpanNano(clockStart); - BMK_runTime_t rt; - rt.nanoSecPerRun = (double)totalTime / nbLoops; - rt.sumOfReturn = dstSize; - return BMK_setValid_runTime(rt); - } } -} - - -/* ==== Benchmarking any function, providing intermediate results ==== */ - -struct BMK_timedFnState_s { - PTime timeSpent_ns; - PTime timeBudget_ns; - PTime runBudget_ns; - BMK_runTime_t fastestRun; - unsigned nbLoops; - UTIL_time_t coolTime; -}; /* typedef'd to BMK_timedFnState_t within bench.h */ - -BMK_timedFnState_t* BMK_createTimedFnState(unsigned total_ms, unsigned run_ms) -{ - BMK_timedFnState_t* const r = (BMK_timedFnState_t*)malloc(sizeof(*r)); - if (r == NULL) return NULL; /* malloc() error */ - BMK_resetTimedFnState(r, total_ms, run_ms); - return r; -} - -void BMK_freeTimedFnState(BMK_timedFnState_t* state) { free(state); } - -BMK_timedFnState_t* -BMK_initStatic_timedFnState(void* buffer, size_t size, unsigned total_ms, unsigned run_ms) -{ - typedef char check_size[ 2 * (sizeof(BMK_timedFnState_shell) >= sizeof(struct BMK_timedFnState_s)) - 1]; /* static assert : a compilation failure indicates that BMK_timedFnState_shell is not large enough */ - typedef struct { check_size c; BMK_timedFnState_t tfs; } tfs_align; /* force tfs to be aligned at its next best position */ - size_t const tfs_alignment = offsetof(tfs_align, tfs); /* provides the minimal alignment restriction for BMK_timedFnState_t */ - BMK_timedFnState_t* const r = (BMK_timedFnState_t*)buffer; - if (buffer == NULL) return NULL; - if (size < sizeof(struct BMK_timedFnState_s)) return NULL; - if ((size_t)buffer % tfs_alignment) return NULL; /* buffer must be properly aligned */ - BMK_resetTimedFnState(r, total_ms, run_ms); - return r; -} - -void BMK_resetTimedFnState(BMK_timedFnState_t* timedFnState, unsigned total_ms, unsigned run_ms) -{ - if (!total_ms) total_ms = 1 ; - if (!run_ms) run_ms = 1; - if (run_ms > total_ms) run_ms = total_ms; - timedFnState->timeSpent_ns = 0; - timedFnState->timeBudget_ns = (PTime)total_ms * TIMELOOP_NANOSEC / 1000; - timedFnState->runBudget_ns = (PTime)run_ms * TIMELOOP_NANOSEC / 1000; - timedFnState->fastestRun.nanoSecPerRun = (double)TIMELOOP_NANOSEC * 2000000000; /* hopefully large enough : must be larger than any potential measurement */ - timedFnState->fastestRun.sumOfReturn = (size_t)(-1LL); - timedFnState->nbLoops = 1; - timedFnState->coolTime = UTIL_getTime(); -} - -/* Tells if nb of seconds set in timedFnState for all runs is spent. - * note : this function will return 1 if BMK_benchFunctionTimed() has actually errored. */ -int BMK_isCompleted_TimedFn(const BMK_timedFnState_t* timedFnState) -{ - return (timedFnState->timeSpent_ns >= timedFnState->timeBudget_ns); -} - - -#undef MIN -#define MIN(a,b) ( (a) < (b) ? (a) : (b) ) - -#define MINUSABLETIME (TIMELOOP_NANOSEC / 2) /* 0.5 seconds */ - -BMK_runOutcome_t BMK_benchTimedFn(BMK_timedFnState_t* cont, - BMK_benchParams_t p) -{ - PTime const runBudget_ns = cont->runBudget_ns; - PTime const runTimeMin_ns = runBudget_ns / 2; - int completed = 0; - BMK_runTime_t bestRunTime = cont->fastestRun; - - while (!completed) { - BMK_runOutcome_t const runResult = BMK_benchFunction(p, cont->nbLoops); - - if(!BMK_isSuccessful_runOutcome(runResult)) { /* error : move out */ - return runResult; - } - - { BMK_runTime_t const newRunTime = BMK_extract_runTime(runResult); - double const loopDuration_ns = newRunTime.nanoSecPerRun * cont->nbLoops; - - cont->timeSpent_ns += (unsigned long long)loopDuration_ns; - - /* estimate nbLoops for next run to last approximately 1 second */ - if (loopDuration_ns > (runBudget_ns / 50)) { - double const fastestRun_ns = MIN(bestRunTime.nanoSecPerRun, newRunTime.nanoSecPerRun); - cont->nbLoops = (unsigned)(runBudget_ns / fastestRun_ns) + 1; - } else { - /* previous run was too short : blindly increase workload by x multiplier */ - const unsigned multiplier = 10; - assert(cont->nbLoops < ((unsigned)-1) / multiplier); /* avoid overflow */ - cont->nbLoops *= multiplier; - } - - if(loopDuration_ns < runTimeMin_ns) { - /* don't report results for which benchmark run time was too small : increased risks of rounding errors */ - assert(completed == 0); - continue; - } else { - if(newRunTime.nanoSecPerRun < bestRunTime.nanoSecPerRun) { - bestRunTime = newRunTime; - } - completed = 1; - } - } - } /* while (!completed) */ - - return BMK_setValid_runTime(bestRunTime); -} +/* + * Copyright (c) Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + + + +/* ************************************* +* Includes +***************************************/ +#include <stdlib.h> /* malloc, free */ +#include <string.h> /* memset */ +#include <assert.h> /* assert */ + +#include "timefn.h" /* UTIL_time_t, UTIL_getTime */ +#include "benchfn.h" + + +/* ************************************* +* Constants +***************************************/ +#define TIMELOOP_MICROSEC SEC_TO_MICRO /* 1 second */ +#define TIMELOOP_NANOSEC (1*1000000000ULL) /* 1 second */ + +#define KB *(1 <<10) +#define MB *(1 <<20) +#define GB *(1U<<30) + + +/* ************************************* +* Debug errors +***************************************/ +#if defined(DEBUG) && (DEBUG >= 1) +# include <stdio.h> /* fprintf */ +# define DISPLAY(...) fprintf(stderr, __VA_ARGS__) +# define DEBUGOUTPUT(...) { if (DEBUG) DISPLAY(__VA_ARGS__); } +#else +# define DEBUGOUTPUT(...) +#endif + + +/* error without displaying */ +#define RETURN_QUIET_ERROR(retValue, ...) { \ + DEBUGOUTPUT("%s: %i: \n", __FILE__, __LINE__); \ + DEBUGOUTPUT("Error : "); \ + DEBUGOUTPUT(__VA_ARGS__); \ + DEBUGOUTPUT(" \n"); \ + return retValue; \ +} + +/* Abort execution if a condition is not met */ +#define CONTROL(c) { if (!(c)) { DEBUGOUTPUT("error: %s \n", #c); abort(); } } + + +/* ************************************* +* Benchmarking an arbitrary function +***************************************/ + +int BMK_isSuccessful_runOutcome(BMK_runOutcome_t outcome) +{ + return outcome.error_tag_never_ever_use_directly == 0; +} + +/* warning : this function will stop program execution if outcome is invalid ! + * check outcome validity first, using BMK_isValid_runResult() */ +BMK_runTime_t BMK_extract_runTime(BMK_runOutcome_t outcome) +{ + CONTROL(outcome.error_tag_never_ever_use_directly == 0); + return outcome.internal_never_ever_use_directly; +} + +size_t BMK_extract_errorResult(BMK_runOutcome_t outcome) +{ + CONTROL(outcome.error_tag_never_ever_use_directly != 0); + return outcome.error_result_never_ever_use_directly; +} + +static BMK_runOutcome_t BMK_runOutcome_error(size_t errorResult) +{ + BMK_runOutcome_t b; + memset(&b, 0, sizeof(b)); + b.error_tag_never_ever_use_directly = 1; + b.error_result_never_ever_use_directly = errorResult; + return b; +} + +static BMK_runOutcome_t BMK_setValid_runTime(BMK_runTime_t runTime) +{ + BMK_runOutcome_t outcome; + outcome.error_tag_never_ever_use_directly = 0; + outcome.internal_never_ever_use_directly = runTime; + return outcome; +} + + +/* initFn will be measured once, benchFn will be measured `nbLoops` times */ +/* initFn is optional, provide NULL if none */ +/* benchFn must return a size_t value that errorFn can interpret */ +/* takes # of blocks and list of size & stuff for each. */ +/* can report result of benchFn for each block into blockResult. */ +/* blockResult is optional, provide NULL if this information is not required */ +/* note : time per loop can be reported as zero if run time < timer resolution */ +BMK_runOutcome_t BMK_benchFunction(BMK_benchParams_t p, + unsigned nbLoops) +{ + size_t dstSize = 0; + nbLoops += !nbLoops; /* minimum nbLoops is 1 */ + + /* init */ + { size_t i; + for(i = 0; i < p.blockCount; i++) { + memset(p.dstBuffers[i], 0xE5, p.dstCapacities[i]); /* warm up and erase result buffer */ + } } + + /* benchmark */ + { UTIL_time_t const clockStart = UTIL_getTime(); + unsigned loopNb, blockNb; + if (p.initFn != NULL) p.initFn(p.initPayload); + for (loopNb = 0; loopNb < nbLoops; loopNb++) { + for (blockNb = 0; blockNb < p.blockCount; blockNb++) { + size_t const res = p.benchFn(p.srcBuffers[blockNb], p.srcSizes[blockNb], + p.dstBuffers[blockNb], p.dstCapacities[blockNb], + p.benchPayload); + if (loopNb == 0) { + if (p.blockResults != NULL) p.blockResults[blockNb] = res; + if ((p.errorFn != NULL) && (p.errorFn(res))) { + RETURN_QUIET_ERROR(BMK_runOutcome_error(res), + "Function benchmark failed on block %u (of size %u) with error %i", + blockNb, (unsigned)p.srcSizes[blockNb], (int)res); + } + dstSize += res; + } } + } /* for (loopNb = 0; loopNb < nbLoops; loopNb++) */ + + { PTime const totalTime = UTIL_clockSpanNano(clockStart); + BMK_runTime_t rt; + rt.nanoSecPerRun = (double)totalTime / nbLoops; + rt.sumOfReturn = dstSize; + return BMK_setValid_runTime(rt); + } } +} + + +/* ==== Benchmarking any function, providing intermediate results ==== */ + +struct BMK_timedFnState_s { + PTime timeSpent_ns; + PTime timeBudget_ns; + PTime runBudget_ns; + BMK_runTime_t fastestRun; + unsigned nbLoops; + UTIL_time_t coolTime; +}; /* typedef'd to BMK_timedFnState_t within bench.h */ + +BMK_timedFnState_t* BMK_createTimedFnState(unsigned total_ms, unsigned run_ms) +{ + BMK_timedFnState_t* const r = (BMK_timedFnState_t*)malloc(sizeof(*r)); + if (r == NULL) return NULL; /* malloc() error */ + BMK_resetTimedFnState(r, total_ms, run_ms); + return r; +} + +void BMK_freeTimedFnState(BMK_timedFnState_t* state) { free(state); } + +BMK_timedFnState_t* +BMK_initStatic_timedFnState(void* buffer, size_t size, unsigned total_ms, unsigned run_ms) +{ + typedef char check_size[ 2 * (sizeof(BMK_timedFnState_shell) >= sizeof(struct BMK_timedFnState_s)) - 1]; /* static assert : a compilation failure indicates that BMK_timedFnState_shell is not large enough */ + typedef struct { check_size c; BMK_timedFnState_t tfs; } tfs_align; /* force tfs to be aligned at its next best position */ + size_t const tfs_alignment = offsetof(tfs_align, tfs); /* provides the minimal alignment restriction for BMK_timedFnState_t */ + BMK_timedFnState_t* const r = (BMK_timedFnState_t*)buffer; + if (buffer == NULL) return NULL; + if (size < sizeof(struct BMK_timedFnState_s)) return NULL; + if ((size_t)buffer % tfs_alignment) return NULL; /* buffer must be properly aligned */ + BMK_resetTimedFnState(r, total_ms, run_ms); + return r; +} + +void BMK_resetTimedFnState(BMK_timedFnState_t* timedFnState, unsigned total_ms, unsigned run_ms) +{ + if (!total_ms) total_ms = 1 ; + if (!run_ms) run_ms = 1; + if (run_ms > total_ms) run_ms = total_ms; + timedFnState->timeSpent_ns = 0; + timedFnState->timeBudget_ns = (PTime)total_ms * TIMELOOP_NANOSEC / 1000; + timedFnState->runBudget_ns = (PTime)run_ms * TIMELOOP_NANOSEC / 1000; + timedFnState->fastestRun.nanoSecPerRun = (double)TIMELOOP_NANOSEC * 2000000000; /* hopefully large enough : must be larger than any potential measurement */ + timedFnState->fastestRun.sumOfReturn = (size_t)(-1LL); + timedFnState->nbLoops = 1; + timedFnState->coolTime = UTIL_getTime(); +} + +/* Tells if nb of seconds set in timedFnState for all runs is spent. + * note : this function will return 1 if BMK_benchFunctionTimed() has actually errored. */ +int BMK_isCompleted_TimedFn(const BMK_timedFnState_t* timedFnState) +{ + return (timedFnState->timeSpent_ns >= timedFnState->timeBudget_ns); +} + + +#undef MIN +#define MIN(a,b) ( (a) < (b) ? (a) : (b) ) + +#define MINUSABLETIME (TIMELOOP_NANOSEC / 2) /* 0.5 seconds */ + +BMK_runOutcome_t BMK_benchTimedFn(BMK_timedFnState_t* cont, + BMK_benchParams_t p) +{ + PTime const runBudget_ns = cont->runBudget_ns; + PTime const runTimeMin_ns = runBudget_ns / 2; + int completed = 0; + BMK_runTime_t bestRunTime = cont->fastestRun; + + while (!completed) { + BMK_runOutcome_t const runResult = BMK_benchFunction(p, cont->nbLoops); + + if(!BMK_isSuccessful_runOutcome(runResult)) { /* error : move out */ + return runResult; + } + + { BMK_runTime_t const newRunTime = BMK_extract_runTime(runResult); + double const loopDuration_ns = newRunTime.nanoSecPerRun * cont->nbLoops; + + cont->timeSpent_ns += (unsigned long long)loopDuration_ns; + + /* estimate nbLoops for next run to last approximately 1 second */ + if (loopDuration_ns > (runBudget_ns / 50)) { + double const fastestRun_ns = MIN(bestRunTime.nanoSecPerRun, newRunTime.nanoSecPerRun); + cont->nbLoops = (unsigned)(runBudget_ns / fastestRun_ns) + 1; + } else { + /* previous run was too short : blindly increase workload by x multiplier */ + const unsigned multiplier = 10; + assert(cont->nbLoops < ((unsigned)-1) / multiplier); /* avoid overflow */ + cont->nbLoops *= multiplier; + } + + if(loopDuration_ns < runTimeMin_ns) { + /* don't report results for which benchmark run time was too small : increased risks of rounding errors */ + assert(completed == 0); + continue; + } else { + if(newRunTime.nanoSecPerRun < bestRunTime.nanoSecPerRun) { + bestRunTime = newRunTime; + } + completed = 1; + } + } + } /* while (!completed) */ + + return BMK_setValid_runTime(bestRunTime); +} diff --git a/contrib/libs/zstd/programs/benchfn.h b/contrib/libs/zstd/programs/benchfn.h index 590f292eaa..7df8d608ba 100644 --- a/contrib/libs/zstd/programs/benchfn.h +++ b/contrib/libs/zstd/programs/benchfn.h @@ -1,183 +1,183 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - -/* benchfn : - * benchmark any function on a set of input - * providing result in nanoSecPerRun - * or detecting and returning an error - */ - -#if defined (__cplusplus) -extern "C" { -#endif - -#ifndef BENCH_FN_H_23876 -#define BENCH_FN_H_23876 - -/* === Dependencies === */ -#include <stddef.h> /* size_t */ - - -/* ==== Benchmark any function, iterated on a set of blocks ==== */ - -/* BMK_runTime_t: valid result return type */ - -typedef struct { - double nanoSecPerRun; /* time per iteration (over all blocks) */ - size_t sumOfReturn; /* sum of return values */ -} BMK_runTime_t; - - -/* BMK_runOutcome_t: - * type expressing the outcome of a benchmark run by BMK_benchFunction(), - * which can be either valid or invalid. - * benchmark outcome can be invalid if errorFn is provided. - * BMK_runOutcome_t must be considered "opaque" : never access its members directly. - * Instead, use its assigned methods : - * BMK_isSuccessful_runOutcome, BMK_extract_runTime, BMK_extract_errorResult. - * The structure is only described here to allow its allocation on stack. */ - -typedef struct { - BMK_runTime_t internal_never_ever_use_directly; - size_t error_result_never_ever_use_directly; - int error_tag_never_ever_use_directly; -} BMK_runOutcome_t; - - -/* prototypes for benchmarked functions */ -typedef size_t (*BMK_benchFn_t)(const void* src, size_t srcSize, void* dst, size_t dstCapacity, void* customPayload); -typedef size_t (*BMK_initFn_t)(void* initPayload); -typedef unsigned (*BMK_errorFn_t)(size_t); - - -/* BMK_benchFunction() parameters are provided via the following structure. - * A structure is preferable for readability, - * as the number of parameters required is fairly large. - * No initializer is provided, because it doesn't make sense to provide some "default" : - * all parameters must be specified by the caller. - * optional parameters are labelled explicitly, and accept value NULL when not used */ -typedef struct { - BMK_benchFn_t benchFn; /* the function to benchmark, over the set of blocks */ - void* benchPayload; /* pass custom parameters to benchFn : - * (*benchFn)(srcBuffers[i], srcSizes[i], dstBuffers[i], dstCapacities[i], benchPayload) */ - BMK_initFn_t initFn; /* (*initFn)(initPayload) is run once per run, at the beginning. */ - void* initPayload; /* Both arguments can be NULL, in which case nothing is run. */ - BMK_errorFn_t errorFn; /* errorFn will check each return value of benchFn over each block, to determine if it failed or not. - * errorFn can be NULL, in which case no check is performed. - * errorFn must return 0 when benchFn was successful, and >= 1 if it detects an error. - * Execution is stopped as soon as an error is detected. - * the triggering return value can be retrieved using BMK_extract_errorResult(). */ - size_t blockCount; /* number of blocks to operate benchFn on. - * It's also the size of all array parameters : - * srcBuffers, srcSizes, dstBuffers, dstCapacities, blockResults */ - const void *const * srcBuffers; /* read-only array of buffers to be operated on by benchFn */ - const size_t* srcSizes; /* read-only array containing sizes of srcBuffers */ - void *const * dstBuffers; /* array of buffers to be written into by benchFn. This array is not optional, it must be provided even if unused by benchfn. */ - const size_t* dstCapacities; /* read-only array containing capacities of dstBuffers. This array must be present. */ - size_t* blockResults; /* Optional: store the return value of benchFn for each block. Use NULL if this result is not requested. */ -} BMK_benchParams_t; - - -/* BMK_benchFunction() : - * This function benchmarks benchFn and initFn, providing a result. - * - * params : see description of BMK_benchParams_t above. - * nbLoops: defines number of times benchFn is run over the full set of blocks. - * Minimum value is 1. A 0 is interpreted as a 1. - * - * @return: can express either an error or a successful result. - * Use BMK_isSuccessful_runOutcome() to check if benchmark was successful. - * If yes, extract the result with BMK_extract_runTime(), - * it will contain : - * .sumOfReturn : the sum of all return values of benchFn through all of blocks - * .nanoSecPerRun : time per run of benchFn + (time for initFn / nbLoops) - * .sumOfReturn is generally intended for functions which return a # of bytes written into dstBuffer, - * in which case, this value will be the total amount of bytes written into dstBuffer. - * - * blockResults : when provided (!= NULL), and when benchmark is successful, - * params.blockResults contains all return values of `benchFn` over all blocks. - * when provided (!= NULL), and when benchmark failed, - * params.blockResults contains return values of `benchFn` over all blocks preceding and including the failed block. - */ -BMK_runOutcome_t BMK_benchFunction(BMK_benchParams_t params, unsigned nbLoops); - - - -/* check first if the benchmark was successful or not */ -int BMK_isSuccessful_runOutcome(BMK_runOutcome_t outcome); - -/* If the benchmark was successful, extract the result. - * note : this function will abort() program execution if benchmark failed ! - * always check if benchmark was successful first ! - */ -BMK_runTime_t BMK_extract_runTime(BMK_runOutcome_t outcome); - -/* when benchmark failed, it means one invocation of `benchFn` failed. - * The failure was detected by `errorFn`, operating on return values of `benchFn`. - * Returns the faulty return value. - * note : this function will abort() program execution if benchmark did not failed. - * always check if benchmark failed first ! - */ -size_t BMK_extract_errorResult(BMK_runOutcome_t outcome); - - - -/* ==== Benchmark any function, returning intermediate results ==== */ - -/* state information tracking benchmark session */ -typedef struct BMK_timedFnState_s BMK_timedFnState_t; - -/* BMK_benchTimedFn() : - * Similar to BMK_benchFunction(), most arguments being identical. - * Automatically determines `nbLoops` so that each result is regularly produced at interval of about run_ms. - * Note : minimum `nbLoops` is 1, therefore a run may last more than run_ms, and possibly even more than total_ms. - * Usage - initialize timedFnState, select benchmark duration (total_ms) and each measurement duration (run_ms) - * call BMK_benchTimedFn() repetitively, each measurement is supposed to last about run_ms - * Check if total time budget is spent or exceeded, using BMK_isCompleted_TimedFn() - */ -BMK_runOutcome_t BMK_benchTimedFn(BMK_timedFnState_t* timedFnState, - BMK_benchParams_t params); - -/* Tells if duration of all benchmark runs has exceeded total_ms - */ -int BMK_isCompleted_TimedFn(const BMK_timedFnState_t* timedFnState); - -/* BMK_createTimedFnState() and BMK_resetTimedFnState() : - * Create/Set BMK_timedFnState_t for next benchmark session, - * which shall last a minimum of total_ms milliseconds, - * producing intermediate results, paced at interval of (approximately) run_ms. - */ -BMK_timedFnState_t* BMK_createTimedFnState(unsigned total_ms, unsigned run_ms); -void BMK_resetTimedFnState(BMK_timedFnState_t* timedFnState, unsigned total_ms, unsigned run_ms); -void BMK_freeTimedFnState(BMK_timedFnState_t* state); - - -/* BMK_timedFnState_shell and BMK_initStatic_timedFnState() : - * Makes it possible to statically allocate a BMK_timedFnState_t on stack. - * BMK_timedFnState_shell is only there to allocate space, - * never ever access its members. - * BMK_timedFnState_t() actually accepts any buffer. - * It will check if provided buffer is large enough and is correctly aligned, - * and will return NULL if conditions are not respected. - */ -#define BMK_TIMEDFNSTATE_SIZE 64 -typedef union { - char never_access_space[BMK_TIMEDFNSTATE_SIZE]; - long long alignment_enforcer; /* must be aligned on 8-bytes boundaries */ -} BMK_timedFnState_shell; -BMK_timedFnState_t* BMK_initStatic_timedFnState(void* buffer, size_t size, unsigned total_ms, unsigned run_ms); - - -#endif /* BENCH_FN_H_23876 */ - -#if defined (__cplusplus) -} -#endif +/* + * Copyright (c) Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + + +/* benchfn : + * benchmark any function on a set of input + * providing result in nanoSecPerRun + * or detecting and returning an error + */ + +#if defined (__cplusplus) +extern "C" { +#endif + +#ifndef BENCH_FN_H_23876 +#define BENCH_FN_H_23876 + +/* === Dependencies === */ +#include <stddef.h> /* size_t */ + + +/* ==== Benchmark any function, iterated on a set of blocks ==== */ + +/* BMK_runTime_t: valid result return type */ + +typedef struct { + double nanoSecPerRun; /* time per iteration (over all blocks) */ + size_t sumOfReturn; /* sum of return values */ +} BMK_runTime_t; + + +/* BMK_runOutcome_t: + * type expressing the outcome of a benchmark run by BMK_benchFunction(), + * which can be either valid or invalid. + * benchmark outcome can be invalid if errorFn is provided. + * BMK_runOutcome_t must be considered "opaque" : never access its members directly. + * Instead, use its assigned methods : + * BMK_isSuccessful_runOutcome, BMK_extract_runTime, BMK_extract_errorResult. + * The structure is only described here to allow its allocation on stack. */ + +typedef struct { + BMK_runTime_t internal_never_ever_use_directly; + size_t error_result_never_ever_use_directly; + int error_tag_never_ever_use_directly; +} BMK_runOutcome_t; + + +/* prototypes for benchmarked functions */ +typedef size_t (*BMK_benchFn_t)(const void* src, size_t srcSize, void* dst, size_t dstCapacity, void* customPayload); +typedef size_t (*BMK_initFn_t)(void* initPayload); +typedef unsigned (*BMK_errorFn_t)(size_t); + + +/* BMK_benchFunction() parameters are provided via the following structure. + * A structure is preferable for readability, + * as the number of parameters required is fairly large. + * No initializer is provided, because it doesn't make sense to provide some "default" : + * all parameters must be specified by the caller. + * optional parameters are labelled explicitly, and accept value NULL when not used */ +typedef struct { + BMK_benchFn_t benchFn; /* the function to benchmark, over the set of blocks */ + void* benchPayload; /* pass custom parameters to benchFn : + * (*benchFn)(srcBuffers[i], srcSizes[i], dstBuffers[i], dstCapacities[i], benchPayload) */ + BMK_initFn_t initFn; /* (*initFn)(initPayload) is run once per run, at the beginning. */ + void* initPayload; /* Both arguments can be NULL, in which case nothing is run. */ + BMK_errorFn_t errorFn; /* errorFn will check each return value of benchFn over each block, to determine if it failed or not. + * errorFn can be NULL, in which case no check is performed. + * errorFn must return 0 when benchFn was successful, and >= 1 if it detects an error. + * Execution is stopped as soon as an error is detected. + * the triggering return value can be retrieved using BMK_extract_errorResult(). */ + size_t blockCount; /* number of blocks to operate benchFn on. + * It's also the size of all array parameters : + * srcBuffers, srcSizes, dstBuffers, dstCapacities, blockResults */ + const void *const * srcBuffers; /* read-only array of buffers to be operated on by benchFn */ + const size_t* srcSizes; /* read-only array containing sizes of srcBuffers */ + void *const * dstBuffers; /* array of buffers to be written into by benchFn. This array is not optional, it must be provided even if unused by benchfn. */ + const size_t* dstCapacities; /* read-only array containing capacities of dstBuffers. This array must be present. */ + size_t* blockResults; /* Optional: store the return value of benchFn for each block. Use NULL if this result is not requested. */ +} BMK_benchParams_t; + + +/* BMK_benchFunction() : + * This function benchmarks benchFn and initFn, providing a result. + * + * params : see description of BMK_benchParams_t above. + * nbLoops: defines number of times benchFn is run over the full set of blocks. + * Minimum value is 1. A 0 is interpreted as a 1. + * + * @return: can express either an error or a successful result. + * Use BMK_isSuccessful_runOutcome() to check if benchmark was successful. + * If yes, extract the result with BMK_extract_runTime(), + * it will contain : + * .sumOfReturn : the sum of all return values of benchFn through all of blocks + * .nanoSecPerRun : time per run of benchFn + (time for initFn / nbLoops) + * .sumOfReturn is generally intended for functions which return a # of bytes written into dstBuffer, + * in which case, this value will be the total amount of bytes written into dstBuffer. + * + * blockResults : when provided (!= NULL), and when benchmark is successful, + * params.blockResults contains all return values of `benchFn` over all blocks. + * when provided (!= NULL), and when benchmark failed, + * params.blockResults contains return values of `benchFn` over all blocks preceding and including the failed block. + */ +BMK_runOutcome_t BMK_benchFunction(BMK_benchParams_t params, unsigned nbLoops); + + + +/* check first if the benchmark was successful or not */ +int BMK_isSuccessful_runOutcome(BMK_runOutcome_t outcome); + +/* If the benchmark was successful, extract the result. + * note : this function will abort() program execution if benchmark failed ! + * always check if benchmark was successful first ! + */ +BMK_runTime_t BMK_extract_runTime(BMK_runOutcome_t outcome); + +/* when benchmark failed, it means one invocation of `benchFn` failed. + * The failure was detected by `errorFn`, operating on return values of `benchFn`. + * Returns the faulty return value. + * note : this function will abort() program execution if benchmark did not failed. + * always check if benchmark failed first ! + */ +size_t BMK_extract_errorResult(BMK_runOutcome_t outcome); + + + +/* ==== Benchmark any function, returning intermediate results ==== */ + +/* state information tracking benchmark session */ +typedef struct BMK_timedFnState_s BMK_timedFnState_t; + +/* BMK_benchTimedFn() : + * Similar to BMK_benchFunction(), most arguments being identical. + * Automatically determines `nbLoops` so that each result is regularly produced at interval of about run_ms. + * Note : minimum `nbLoops` is 1, therefore a run may last more than run_ms, and possibly even more than total_ms. + * Usage - initialize timedFnState, select benchmark duration (total_ms) and each measurement duration (run_ms) + * call BMK_benchTimedFn() repetitively, each measurement is supposed to last about run_ms + * Check if total time budget is spent or exceeded, using BMK_isCompleted_TimedFn() + */ +BMK_runOutcome_t BMK_benchTimedFn(BMK_timedFnState_t* timedFnState, + BMK_benchParams_t params); + +/* Tells if duration of all benchmark runs has exceeded total_ms + */ +int BMK_isCompleted_TimedFn(const BMK_timedFnState_t* timedFnState); + +/* BMK_createTimedFnState() and BMK_resetTimedFnState() : + * Create/Set BMK_timedFnState_t for next benchmark session, + * which shall last a minimum of total_ms milliseconds, + * producing intermediate results, paced at interval of (approximately) run_ms. + */ +BMK_timedFnState_t* BMK_createTimedFnState(unsigned total_ms, unsigned run_ms); +void BMK_resetTimedFnState(BMK_timedFnState_t* timedFnState, unsigned total_ms, unsigned run_ms); +void BMK_freeTimedFnState(BMK_timedFnState_t* state); + + +/* BMK_timedFnState_shell and BMK_initStatic_timedFnState() : + * Makes it possible to statically allocate a BMK_timedFnState_t on stack. + * BMK_timedFnState_shell is only there to allocate space, + * never ever access its members. + * BMK_timedFnState_t() actually accepts any buffer. + * It will check if provided buffer is large enough and is correctly aligned, + * and will return NULL if conditions are not respected. + */ +#define BMK_TIMEDFNSTATE_SIZE 64 +typedef union { + char never_access_space[BMK_TIMEDFNSTATE_SIZE]; + long long alignment_enforcer; /* must be aligned on 8-bytes boundaries */ +} BMK_timedFnState_shell; +BMK_timedFnState_t* BMK_initStatic_timedFnState(void* buffer, size_t size, unsigned total_ms, unsigned run_ms); + + +#endif /* BENCH_FN_H_23876 */ + +#if defined (__cplusplus) +} +#endif diff --git a/contrib/libs/zstd/programs/benchzstd.c b/contrib/libs/zstd/programs/benchzstd.c index 9dc76a6f3e..cced9d09ce 100644 --- a/contrib/libs/zstd/programs/benchzstd.c +++ b/contrib/libs/zstd/programs/benchzstd.c @@ -1,892 +1,892 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - -/* ************************************** -* Tuning parameters -****************************************/ -#ifndef BMK_TIMETEST_DEFAULT_S /* default minimum time per test */ -#define BMK_TIMETEST_DEFAULT_S 3 -#endif - - -/* ************************************* -* Includes -***************************************/ -#include "platform.h" /* Large Files support */ -#include "util.h" /* UTIL_getFileSize, UTIL_sleep */ -#include <stdlib.h> /* malloc, free */ -#include <string.h> /* memset, strerror */ -#include <stdio.h> /* fprintf, fopen */ -#include <errno.h> -#include <assert.h> /* assert */ - -#include "timefn.h" /* UTIL_time_t */ -#include "benchfn.h" -#include "../lib/common/mem.h" +/* + * Copyright (c) Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + + +/* ************************************** +* Tuning parameters +****************************************/ +#ifndef BMK_TIMETEST_DEFAULT_S /* default minimum time per test */ +#define BMK_TIMETEST_DEFAULT_S 3 +#endif + + +/* ************************************* +* Includes +***************************************/ +#include "platform.h" /* Large Files support */ +#include "util.h" /* UTIL_getFileSize, UTIL_sleep */ +#include <stdlib.h> /* malloc, free */ +#include <string.h> /* memset, strerror */ +#include <stdio.h> /* fprintf, fopen */ +#include <errno.h> +#include <assert.h> /* assert */ + +#include "timefn.h" /* UTIL_time_t */ +#include "benchfn.h" +#include "../lib/common/mem.h" #ifndef ZSTD_STATIC_LINKING_ONLY -#define ZSTD_STATIC_LINKING_ONLY +#define ZSTD_STATIC_LINKING_ONLY #endif -#include "../lib/zstd.h" -#include "datagen.h" /* RDG_genBuffer */ +#include "../lib/zstd.h" +#include "datagen.h" /* RDG_genBuffer */ #ifndef XXH_INLINE_ALL #define XXH_INLINE_ALL #endif -#include <contrib/libs/xxhash/xxhash.h> -#include "benchzstd.h" -#include "../lib/zstd_errors.h" - - -/* ************************************* -* Constants -***************************************/ -#ifndef ZSTD_GIT_COMMIT -# define ZSTD_GIT_COMMIT_STRING "" -#else -# define ZSTD_GIT_COMMIT_STRING ZSTD_EXPAND_AND_QUOTE(ZSTD_GIT_COMMIT) -#endif - -#define TIMELOOP_MICROSEC (1*1000000ULL) /* 1 second */ -#define TIMELOOP_NANOSEC (1*1000000000ULL) /* 1 second */ -#define ACTIVEPERIOD_MICROSEC (70*TIMELOOP_MICROSEC) /* 70 seconds */ -#define COOLPERIOD_SEC 10 - -#define KB *(1 <<10) -#define MB *(1 <<20) -#define GB *(1U<<30) - -#define BMK_RUNTEST_DEFAULT_MS 1000 - -static const size_t maxMemory = (sizeof(size_t)==4) ? - /* 32-bit */ (2 GB - 64 MB) : - /* 64-bit */ (size_t)(1ULL << ((sizeof(size_t)*8)-31)); - - -/* ************************************* -* console display -***************************************/ -#define DISPLAY(...) { fprintf(stderr, __VA_ARGS__); fflush(NULL); } -#define DISPLAYLEVEL(l, ...) if (displayLevel>=l) { DISPLAY(__VA_ARGS__); } -/* 0 : no display; 1: errors; 2 : + result + interaction + warnings; 3 : + progression; 4 : + information */ +#include <contrib/libs/xxhash/xxhash.h> +#include "benchzstd.h" +#include "../lib/zstd_errors.h" + + +/* ************************************* +* Constants +***************************************/ +#ifndef ZSTD_GIT_COMMIT +# define ZSTD_GIT_COMMIT_STRING "" +#else +# define ZSTD_GIT_COMMIT_STRING ZSTD_EXPAND_AND_QUOTE(ZSTD_GIT_COMMIT) +#endif + +#define TIMELOOP_MICROSEC (1*1000000ULL) /* 1 second */ +#define TIMELOOP_NANOSEC (1*1000000000ULL) /* 1 second */ +#define ACTIVEPERIOD_MICROSEC (70*TIMELOOP_MICROSEC) /* 70 seconds */ +#define COOLPERIOD_SEC 10 + +#define KB *(1 <<10) +#define MB *(1 <<20) +#define GB *(1U<<30) + +#define BMK_RUNTEST_DEFAULT_MS 1000 + +static const size_t maxMemory = (sizeof(size_t)==4) ? + /* 32-bit */ (2 GB - 64 MB) : + /* 64-bit */ (size_t)(1ULL << ((sizeof(size_t)*8)-31)); + + +/* ************************************* +* console display +***************************************/ +#define DISPLAY(...) { fprintf(stderr, __VA_ARGS__); fflush(NULL); } +#define DISPLAYLEVEL(l, ...) if (displayLevel>=l) { DISPLAY(__VA_ARGS__); } +/* 0 : no display; 1: errors; 2 : + result + interaction + warnings; 3 : + progression; 4 : + information */ #define OUTPUT(...) { fprintf(stdout, __VA_ARGS__); fflush(NULL); } #define OUTPUTLEVEL(l, ...) if (displayLevel>=l) { OUTPUT(__VA_ARGS__); } - - -/* ************************************* -* Exceptions -***************************************/ -#ifndef DEBUG -# define DEBUG 0 -#endif -#define DEBUGOUTPUT(...) { if (DEBUG) DISPLAY(__VA_ARGS__); } - -#define RETURN_ERROR_INT(errorNum, ...) { \ - DEBUGOUTPUT("%s: %i: \n", __FILE__, __LINE__); \ - DISPLAYLEVEL(1, "Error %i : ", errorNum); \ - DISPLAYLEVEL(1, __VA_ARGS__); \ - DISPLAYLEVEL(1, " \n"); \ - return errorNum; \ -} - -#define CHECK_Z(zf) { \ - size_t const zerr = zf; \ - if (ZSTD_isError(zerr)) { \ - DEBUGOUTPUT("%s: %i: \n", __FILE__, __LINE__); \ - DISPLAY("Error : "); \ - DISPLAY("%s failed : %s", \ - #zf, ZSTD_getErrorName(zerr)); \ - DISPLAY(" \n"); \ - exit(1); \ - } \ -} - -#define RETURN_ERROR(errorNum, retType, ...) { \ - retType r; \ - memset(&r, 0, sizeof(retType)); \ - DEBUGOUTPUT("%s: %i: \n", __FILE__, __LINE__); \ - DISPLAYLEVEL(1, "Error %i : ", errorNum); \ - DISPLAYLEVEL(1, __VA_ARGS__); \ - DISPLAYLEVEL(1, " \n"); \ - r.tag = errorNum; \ - return r; \ -} - - -/* ************************************* -* Benchmark Parameters -***************************************/ - -BMK_advancedParams_t BMK_initAdvancedParams(void) { - BMK_advancedParams_t const res = { - BMK_both, /* mode */ - BMK_TIMETEST_DEFAULT_S, /* nbSeconds */ - 0, /* blockSize */ - 0, /* nbWorkers */ - 0, /* realTime */ - 0, /* additionalParam */ - 0, /* ldmFlag */ - 0, /* ldmMinMatch */ - 0, /* ldmHashLog */ - 0, /* ldmBuckSizeLog */ - 0, /* ldmHashRateLog */ + + +/* ************************************* +* Exceptions +***************************************/ +#ifndef DEBUG +# define DEBUG 0 +#endif +#define DEBUGOUTPUT(...) { if (DEBUG) DISPLAY(__VA_ARGS__); } + +#define RETURN_ERROR_INT(errorNum, ...) { \ + DEBUGOUTPUT("%s: %i: \n", __FILE__, __LINE__); \ + DISPLAYLEVEL(1, "Error %i : ", errorNum); \ + DISPLAYLEVEL(1, __VA_ARGS__); \ + DISPLAYLEVEL(1, " \n"); \ + return errorNum; \ +} + +#define CHECK_Z(zf) { \ + size_t const zerr = zf; \ + if (ZSTD_isError(zerr)) { \ + DEBUGOUTPUT("%s: %i: \n", __FILE__, __LINE__); \ + DISPLAY("Error : "); \ + DISPLAY("%s failed : %s", \ + #zf, ZSTD_getErrorName(zerr)); \ + DISPLAY(" \n"); \ + exit(1); \ + } \ +} + +#define RETURN_ERROR(errorNum, retType, ...) { \ + retType r; \ + memset(&r, 0, sizeof(retType)); \ + DEBUGOUTPUT("%s: %i: \n", __FILE__, __LINE__); \ + DISPLAYLEVEL(1, "Error %i : ", errorNum); \ + DISPLAYLEVEL(1, __VA_ARGS__); \ + DISPLAYLEVEL(1, " \n"); \ + r.tag = errorNum; \ + return r; \ +} + + +/* ************************************* +* Benchmark Parameters +***************************************/ + +BMK_advancedParams_t BMK_initAdvancedParams(void) { + BMK_advancedParams_t const res = { + BMK_both, /* mode */ + BMK_TIMETEST_DEFAULT_S, /* nbSeconds */ + 0, /* blockSize */ + 0, /* nbWorkers */ + 0, /* realTime */ + 0, /* additionalParam */ + 0, /* ldmFlag */ + 0, /* ldmMinMatch */ + 0, /* ldmHashLog */ + 0, /* ldmBuckSizeLog */ + 0, /* ldmHashRateLog */ ZSTD_ps_auto, /* literalCompressionMode */ - 0 /* useRowMatchFinder */ - }; - return res; -} - - -/* ******************************************************** -* Bench functions -**********************************************************/ -typedef struct { - const void* srcPtr; - size_t srcSize; - void* cPtr; - size_t cRoom; - size_t cSize; - void* resPtr; - size_t resSize; -} blockParam_t; - -#undef MIN -#undef MAX -#define MIN(a,b) ((a) < (b) ? (a) : (b)) -#define MAX(a,b) ((a) > (b) ? (a) : (b)) - -static void -BMK_initCCtx(ZSTD_CCtx* ctx, - const void* dictBuffer, size_t dictBufferSize, - int cLevel, - const ZSTD_compressionParameters* comprParams, - const BMK_advancedParams_t* adv) -{ - ZSTD_CCtx_reset(ctx, ZSTD_reset_session_and_parameters); - if (adv->nbWorkers==1) { - CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_nbWorkers, 0)); - } else { - CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_nbWorkers, adv->nbWorkers)); - } - CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_compressionLevel, cLevel)); - CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_useRowMatchFinder, adv->useRowMatchFinder)); - CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_enableLongDistanceMatching, adv->ldmFlag)); - CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_ldmMinMatch, adv->ldmMinMatch)); - CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_ldmHashLog, adv->ldmHashLog)); - CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_ldmBucketSizeLog, adv->ldmBucketSizeLog)); - CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_ldmHashRateLog, adv->ldmHashRateLog)); - CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_windowLog, (int)comprParams->windowLog)); - CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_hashLog, (int)comprParams->hashLog)); - CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_chainLog, (int)comprParams->chainLog)); - CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_searchLog, (int)comprParams->searchLog)); - CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_minMatch, (int)comprParams->minMatch)); - CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_targetLength, (int)comprParams->targetLength)); - CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_literalCompressionMode, (int)adv->literalCompressionMode)); + 0 /* useRowMatchFinder */ + }; + return res; +} + + +/* ******************************************************** +* Bench functions +**********************************************************/ +typedef struct { + const void* srcPtr; + size_t srcSize; + void* cPtr; + size_t cRoom; + size_t cSize; + void* resPtr; + size_t resSize; +} blockParam_t; + +#undef MIN +#undef MAX +#define MIN(a,b) ((a) < (b) ? (a) : (b)) +#define MAX(a,b) ((a) > (b) ? (a) : (b)) + +static void +BMK_initCCtx(ZSTD_CCtx* ctx, + const void* dictBuffer, size_t dictBufferSize, + int cLevel, + const ZSTD_compressionParameters* comprParams, + const BMK_advancedParams_t* adv) +{ + ZSTD_CCtx_reset(ctx, ZSTD_reset_session_and_parameters); + if (adv->nbWorkers==1) { + CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_nbWorkers, 0)); + } else { + CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_nbWorkers, adv->nbWorkers)); + } + CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_compressionLevel, cLevel)); + CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_useRowMatchFinder, adv->useRowMatchFinder)); + CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_enableLongDistanceMatching, adv->ldmFlag)); + CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_ldmMinMatch, adv->ldmMinMatch)); + CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_ldmHashLog, adv->ldmHashLog)); + CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_ldmBucketSizeLog, adv->ldmBucketSizeLog)); + CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_ldmHashRateLog, adv->ldmHashRateLog)); + CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_windowLog, (int)comprParams->windowLog)); + CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_hashLog, (int)comprParams->hashLog)); + CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_chainLog, (int)comprParams->chainLog)); + CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_searchLog, (int)comprParams->searchLog)); + CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_minMatch, (int)comprParams->minMatch)); + CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_targetLength, (int)comprParams->targetLength)); + CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_literalCompressionMode, (int)adv->literalCompressionMode)); CHECK_Z(ZSTD_CCtx_setParameter(ctx, ZSTD_c_strategy, (int)comprParams->strategy)); - CHECK_Z(ZSTD_CCtx_loadDictionary(ctx, dictBuffer, dictBufferSize)); -} - -static void BMK_initDCtx(ZSTD_DCtx* dctx, - const void* dictBuffer, size_t dictBufferSize) { - CHECK_Z(ZSTD_DCtx_reset(dctx, ZSTD_reset_session_and_parameters)); - CHECK_Z(ZSTD_DCtx_loadDictionary(dctx, dictBuffer, dictBufferSize)); -} - - -typedef struct { - ZSTD_CCtx* cctx; - const void* dictBuffer; - size_t dictBufferSize; - int cLevel; - const ZSTD_compressionParameters* comprParams; - const BMK_advancedParams_t* adv; -} BMK_initCCtxArgs; - -static size_t local_initCCtx(void* payload) { - BMK_initCCtxArgs* ag = (BMK_initCCtxArgs*)payload; - BMK_initCCtx(ag->cctx, ag->dictBuffer, ag->dictBufferSize, ag->cLevel, ag->comprParams, ag->adv); - return 0; -} - -typedef struct { - ZSTD_DCtx* dctx; - const void* dictBuffer; - size_t dictBufferSize; -} BMK_initDCtxArgs; - -static size_t local_initDCtx(void* payload) { - BMK_initDCtxArgs* ag = (BMK_initDCtxArgs*)payload; - BMK_initDCtx(ag->dctx, ag->dictBuffer, ag->dictBufferSize); - return 0; -} - - -/* `addArgs` is the context */ -static size_t local_defaultCompress( - const void* srcBuffer, size_t srcSize, - void* dstBuffer, size_t dstSize, - void* addArgs) -{ - ZSTD_CCtx* const cctx = (ZSTD_CCtx*)addArgs; - return ZSTD_compress2(cctx, dstBuffer, dstSize, srcBuffer, srcSize); -} - -/* `addArgs` is the context */ -static size_t local_defaultDecompress( - const void* srcBuffer, size_t srcSize, - void* dstBuffer, size_t dstCapacity, - void* addArgs) -{ - size_t moreToFlush = 1; - ZSTD_DCtx* const dctx = (ZSTD_DCtx*)addArgs; - ZSTD_inBuffer in; - ZSTD_outBuffer out; - in.src = srcBuffer; in.size = srcSize; in.pos = 0; - out.dst = dstBuffer; out.size = dstCapacity; out.pos = 0; - while (moreToFlush) { - if(out.pos == out.size) { - return (size_t)-ZSTD_error_dstSize_tooSmall; - } - moreToFlush = ZSTD_decompressStream(dctx, &out, &in); - if (ZSTD_isError(moreToFlush)) { - return moreToFlush; - } - } - return out.pos; - -} - - -/* ================================================================= */ -/* Benchmark Zstandard, mem-to-mem scenarios */ -/* ================================================================= */ - -int BMK_isSuccessful_benchOutcome(BMK_benchOutcome_t outcome) -{ - return outcome.tag == 0; -} - -BMK_benchResult_t BMK_extract_benchResult(BMK_benchOutcome_t outcome) -{ - assert(outcome.tag == 0); - return outcome.internal_never_use_directly; -} - -static BMK_benchOutcome_t BMK_benchOutcome_error(void) -{ - BMK_benchOutcome_t b; - memset(&b, 0, sizeof(b)); - b.tag = 1; - return b; -} - -static BMK_benchOutcome_t BMK_benchOutcome_setValidResult(BMK_benchResult_t result) -{ - BMK_benchOutcome_t b; - b.tag = 0; - b.internal_never_use_directly = result; - return b; -} - - -/* benchMem with no allocation */ -static BMK_benchOutcome_t -BMK_benchMemAdvancedNoAlloc( - const void** srcPtrs, size_t* srcSizes, - void** cPtrs, size_t* cCapacities, size_t* cSizes, - void** resPtrs, size_t* resSizes, - void** resultBufferPtr, void* compressedBuffer, - size_t maxCompressedSize, - BMK_timedFnState_t* timeStateCompress, - BMK_timedFnState_t* timeStateDecompress, - - const void* srcBuffer, size_t srcSize, - const size_t* fileSizes, unsigned nbFiles, - const int cLevel, - const ZSTD_compressionParameters* comprParams, - const void* dictBuffer, size_t dictBufferSize, - ZSTD_CCtx* cctx, ZSTD_DCtx* dctx, - int displayLevel, const char* displayName, - const BMK_advancedParams_t* adv) -{ - size_t const blockSize = ((adv->blockSize>=32 && (adv->mode != BMK_decodeOnly)) ? adv->blockSize : srcSize) + (!srcSize); /* avoid div by 0 */ - BMK_benchResult_t benchResult; - size_t const loadedCompressedSize = srcSize; - size_t cSize = 0; - double ratio = 0.; - U32 nbBlocks; - - assert(cctx != NULL); assert(dctx != NULL); - - /* init */ - memset(&benchResult, 0, sizeof(benchResult)); - if (strlen(displayName)>17) displayName += strlen(displayName) - 17; /* display last 17 characters */ - if (adv->mode == BMK_decodeOnly) { /* benchmark only decompression : source must be already compressed */ - const char* srcPtr = (const char*)srcBuffer; - U64 totalDSize64 = 0; - U32 fileNb; - for (fileNb=0; fileNb<nbFiles; fileNb++) { - U64 const fSize64 = ZSTD_findDecompressedSize(srcPtr, fileSizes[fileNb]); - if (fSize64==0) RETURN_ERROR(32, BMK_benchOutcome_t, "Impossible to determine original size "); - totalDSize64 += fSize64; - srcPtr += fileSizes[fileNb]; - } - { size_t const decodedSize = (size_t)totalDSize64; - assert((U64)decodedSize == totalDSize64); /* check overflow */ - free(*resultBufferPtr); - *resultBufferPtr = malloc(decodedSize); - if (!(*resultBufferPtr)) { - RETURN_ERROR(33, BMK_benchOutcome_t, "not enough memory"); - } - if (totalDSize64 > decodedSize) { /* size_t overflow */ - free(*resultBufferPtr); - RETURN_ERROR(32, BMK_benchOutcome_t, "original size is too large"); - } - cSize = srcSize; - srcSize = decodedSize; - ratio = (double)srcSize / (double)cSize; - } - } - - /* Init data blocks */ - { const char* srcPtr = (const char*)srcBuffer; - char* cPtr = (char*)compressedBuffer; - char* resPtr = (char*)(*resultBufferPtr); - U32 fileNb; - for (nbBlocks=0, fileNb=0; fileNb<nbFiles; fileNb++) { - size_t remaining = fileSizes[fileNb]; - U32 const nbBlocksforThisFile = (adv->mode == BMK_decodeOnly) ? 1 : (U32)((remaining + (blockSize-1)) / blockSize); - U32 const blockEnd = nbBlocks + nbBlocksforThisFile; - for ( ; nbBlocks<blockEnd; nbBlocks++) { - size_t const thisBlockSize = MIN(remaining, blockSize); - srcPtrs[nbBlocks] = srcPtr; - srcSizes[nbBlocks] = thisBlockSize; - cPtrs[nbBlocks] = cPtr; - cCapacities[nbBlocks] = (adv->mode == BMK_decodeOnly) ? thisBlockSize : ZSTD_compressBound(thisBlockSize); - resPtrs[nbBlocks] = resPtr; - resSizes[nbBlocks] = (adv->mode == BMK_decodeOnly) ? (size_t) ZSTD_findDecompressedSize(srcPtr, thisBlockSize) : thisBlockSize; - srcPtr += thisBlockSize; - cPtr += cCapacities[nbBlocks]; - resPtr += thisBlockSize; - remaining -= thisBlockSize; - if (adv->mode == BMK_decodeOnly) { - cSizes[nbBlocks] = thisBlockSize; - benchResult.cSize = thisBlockSize; + CHECK_Z(ZSTD_CCtx_loadDictionary(ctx, dictBuffer, dictBufferSize)); +} + +static void BMK_initDCtx(ZSTD_DCtx* dctx, + const void* dictBuffer, size_t dictBufferSize) { + CHECK_Z(ZSTD_DCtx_reset(dctx, ZSTD_reset_session_and_parameters)); + CHECK_Z(ZSTD_DCtx_loadDictionary(dctx, dictBuffer, dictBufferSize)); +} + + +typedef struct { + ZSTD_CCtx* cctx; + const void* dictBuffer; + size_t dictBufferSize; + int cLevel; + const ZSTD_compressionParameters* comprParams; + const BMK_advancedParams_t* adv; +} BMK_initCCtxArgs; + +static size_t local_initCCtx(void* payload) { + BMK_initCCtxArgs* ag = (BMK_initCCtxArgs*)payload; + BMK_initCCtx(ag->cctx, ag->dictBuffer, ag->dictBufferSize, ag->cLevel, ag->comprParams, ag->adv); + return 0; +} + +typedef struct { + ZSTD_DCtx* dctx; + const void* dictBuffer; + size_t dictBufferSize; +} BMK_initDCtxArgs; + +static size_t local_initDCtx(void* payload) { + BMK_initDCtxArgs* ag = (BMK_initDCtxArgs*)payload; + BMK_initDCtx(ag->dctx, ag->dictBuffer, ag->dictBufferSize); + return 0; +} + + +/* `addArgs` is the context */ +static size_t local_defaultCompress( + const void* srcBuffer, size_t srcSize, + void* dstBuffer, size_t dstSize, + void* addArgs) +{ + ZSTD_CCtx* const cctx = (ZSTD_CCtx*)addArgs; + return ZSTD_compress2(cctx, dstBuffer, dstSize, srcBuffer, srcSize); +} + +/* `addArgs` is the context */ +static size_t local_defaultDecompress( + const void* srcBuffer, size_t srcSize, + void* dstBuffer, size_t dstCapacity, + void* addArgs) +{ + size_t moreToFlush = 1; + ZSTD_DCtx* const dctx = (ZSTD_DCtx*)addArgs; + ZSTD_inBuffer in; + ZSTD_outBuffer out; + in.src = srcBuffer; in.size = srcSize; in.pos = 0; + out.dst = dstBuffer; out.size = dstCapacity; out.pos = 0; + while (moreToFlush) { + if(out.pos == out.size) { + return (size_t)-ZSTD_error_dstSize_tooSmall; + } + moreToFlush = ZSTD_decompressStream(dctx, &out, &in); + if (ZSTD_isError(moreToFlush)) { + return moreToFlush; + } + } + return out.pos; + +} + + +/* ================================================================= */ +/* Benchmark Zstandard, mem-to-mem scenarios */ +/* ================================================================= */ + +int BMK_isSuccessful_benchOutcome(BMK_benchOutcome_t outcome) +{ + return outcome.tag == 0; +} + +BMK_benchResult_t BMK_extract_benchResult(BMK_benchOutcome_t outcome) +{ + assert(outcome.tag == 0); + return outcome.internal_never_use_directly; +} + +static BMK_benchOutcome_t BMK_benchOutcome_error(void) +{ + BMK_benchOutcome_t b; + memset(&b, 0, sizeof(b)); + b.tag = 1; + return b; +} + +static BMK_benchOutcome_t BMK_benchOutcome_setValidResult(BMK_benchResult_t result) +{ + BMK_benchOutcome_t b; + b.tag = 0; + b.internal_never_use_directly = result; + return b; +} + + +/* benchMem with no allocation */ +static BMK_benchOutcome_t +BMK_benchMemAdvancedNoAlloc( + const void** srcPtrs, size_t* srcSizes, + void** cPtrs, size_t* cCapacities, size_t* cSizes, + void** resPtrs, size_t* resSizes, + void** resultBufferPtr, void* compressedBuffer, + size_t maxCompressedSize, + BMK_timedFnState_t* timeStateCompress, + BMK_timedFnState_t* timeStateDecompress, + + const void* srcBuffer, size_t srcSize, + const size_t* fileSizes, unsigned nbFiles, + const int cLevel, + const ZSTD_compressionParameters* comprParams, + const void* dictBuffer, size_t dictBufferSize, + ZSTD_CCtx* cctx, ZSTD_DCtx* dctx, + int displayLevel, const char* displayName, + const BMK_advancedParams_t* adv) +{ + size_t const blockSize = ((adv->blockSize>=32 && (adv->mode != BMK_decodeOnly)) ? adv->blockSize : srcSize) + (!srcSize); /* avoid div by 0 */ + BMK_benchResult_t benchResult; + size_t const loadedCompressedSize = srcSize; + size_t cSize = 0; + double ratio = 0.; + U32 nbBlocks; + + assert(cctx != NULL); assert(dctx != NULL); + + /* init */ + memset(&benchResult, 0, sizeof(benchResult)); + if (strlen(displayName)>17) displayName += strlen(displayName) - 17; /* display last 17 characters */ + if (adv->mode == BMK_decodeOnly) { /* benchmark only decompression : source must be already compressed */ + const char* srcPtr = (const char*)srcBuffer; + U64 totalDSize64 = 0; + U32 fileNb; + for (fileNb=0; fileNb<nbFiles; fileNb++) { + U64 const fSize64 = ZSTD_findDecompressedSize(srcPtr, fileSizes[fileNb]); + if (fSize64==0) RETURN_ERROR(32, BMK_benchOutcome_t, "Impossible to determine original size "); + totalDSize64 += fSize64; + srcPtr += fileSizes[fileNb]; + } + { size_t const decodedSize = (size_t)totalDSize64; + assert((U64)decodedSize == totalDSize64); /* check overflow */ + free(*resultBufferPtr); + *resultBufferPtr = malloc(decodedSize); + if (!(*resultBufferPtr)) { + RETURN_ERROR(33, BMK_benchOutcome_t, "not enough memory"); + } + if (totalDSize64 > decodedSize) { /* size_t overflow */ + free(*resultBufferPtr); + RETURN_ERROR(32, BMK_benchOutcome_t, "original size is too large"); + } + cSize = srcSize; + srcSize = decodedSize; + ratio = (double)srcSize / (double)cSize; + } + } + + /* Init data blocks */ + { const char* srcPtr = (const char*)srcBuffer; + char* cPtr = (char*)compressedBuffer; + char* resPtr = (char*)(*resultBufferPtr); + U32 fileNb; + for (nbBlocks=0, fileNb=0; fileNb<nbFiles; fileNb++) { + size_t remaining = fileSizes[fileNb]; + U32 const nbBlocksforThisFile = (adv->mode == BMK_decodeOnly) ? 1 : (U32)((remaining + (blockSize-1)) / blockSize); + U32 const blockEnd = nbBlocks + nbBlocksforThisFile; + for ( ; nbBlocks<blockEnd; nbBlocks++) { + size_t const thisBlockSize = MIN(remaining, blockSize); + srcPtrs[nbBlocks] = srcPtr; + srcSizes[nbBlocks] = thisBlockSize; + cPtrs[nbBlocks] = cPtr; + cCapacities[nbBlocks] = (adv->mode == BMK_decodeOnly) ? thisBlockSize : ZSTD_compressBound(thisBlockSize); + resPtrs[nbBlocks] = resPtr; + resSizes[nbBlocks] = (adv->mode == BMK_decodeOnly) ? (size_t) ZSTD_findDecompressedSize(srcPtr, thisBlockSize) : thisBlockSize; + srcPtr += thisBlockSize; + cPtr += cCapacities[nbBlocks]; + resPtr += thisBlockSize; + remaining -= thisBlockSize; + if (adv->mode == BMK_decodeOnly) { + cSizes[nbBlocks] = thisBlockSize; + benchResult.cSize = thisBlockSize; } } } } - - /* warming up `compressedBuffer` */ - if (adv->mode == BMK_decodeOnly) { - memcpy(compressedBuffer, srcBuffer, loadedCompressedSize); - } else { - RDG_genBuffer(compressedBuffer, maxCompressedSize, 0.10, 0.50, 1); - } - - /* Bench */ - { U64 const crcOrig = (adv->mode == BMK_decodeOnly) ? 0 : XXH64(srcBuffer, srcSize, 0); -# define NB_MARKS 4 - const char* marks[NB_MARKS] = { " |", " /", " =", " \\" }; - U32 markNb = 0; - int compressionCompleted = (adv->mode == BMK_decodeOnly); - int decompressionCompleted = (adv->mode == BMK_compressOnly); - BMK_benchParams_t cbp, dbp; - BMK_initCCtxArgs cctxprep; - BMK_initDCtxArgs dctxprep; - - cbp.benchFn = local_defaultCompress; /* ZSTD_compress2 */ - cbp.benchPayload = cctx; - cbp.initFn = local_initCCtx; /* BMK_initCCtx */ - cbp.initPayload = &cctxprep; - cbp.errorFn = ZSTD_isError; - cbp.blockCount = nbBlocks; - cbp.srcBuffers = srcPtrs; - cbp.srcSizes = srcSizes; - cbp.dstBuffers = cPtrs; - cbp.dstCapacities = cCapacities; - cbp.blockResults = cSizes; - - cctxprep.cctx = cctx; - cctxprep.dictBuffer = dictBuffer; - cctxprep.dictBufferSize = dictBufferSize; - cctxprep.cLevel = cLevel; - cctxprep.comprParams = comprParams; - cctxprep.adv = adv; - - dbp.benchFn = local_defaultDecompress; - dbp.benchPayload = dctx; - dbp.initFn = local_initDCtx; - dbp.initPayload = &dctxprep; - dbp.errorFn = ZSTD_isError; - dbp.blockCount = nbBlocks; - dbp.srcBuffers = (const void* const *) cPtrs; - dbp.srcSizes = cSizes; - dbp.dstBuffers = resPtrs; - dbp.dstCapacities = resSizes; - dbp.blockResults = NULL; - - dctxprep.dctx = dctx; - dctxprep.dictBuffer = dictBuffer; - dctxprep.dictBufferSize = dictBufferSize; - + + /* warming up `compressedBuffer` */ + if (adv->mode == BMK_decodeOnly) { + memcpy(compressedBuffer, srcBuffer, loadedCompressedSize); + } else { + RDG_genBuffer(compressedBuffer, maxCompressedSize, 0.10, 0.50, 1); + } + + /* Bench */ + { U64 const crcOrig = (adv->mode == BMK_decodeOnly) ? 0 : XXH64(srcBuffer, srcSize, 0); +# define NB_MARKS 4 + const char* marks[NB_MARKS] = { " |", " /", " =", " \\" }; + U32 markNb = 0; + int compressionCompleted = (adv->mode == BMK_decodeOnly); + int decompressionCompleted = (adv->mode == BMK_compressOnly); + BMK_benchParams_t cbp, dbp; + BMK_initCCtxArgs cctxprep; + BMK_initDCtxArgs dctxprep; + + cbp.benchFn = local_defaultCompress; /* ZSTD_compress2 */ + cbp.benchPayload = cctx; + cbp.initFn = local_initCCtx; /* BMK_initCCtx */ + cbp.initPayload = &cctxprep; + cbp.errorFn = ZSTD_isError; + cbp.blockCount = nbBlocks; + cbp.srcBuffers = srcPtrs; + cbp.srcSizes = srcSizes; + cbp.dstBuffers = cPtrs; + cbp.dstCapacities = cCapacities; + cbp.blockResults = cSizes; + + cctxprep.cctx = cctx; + cctxprep.dictBuffer = dictBuffer; + cctxprep.dictBufferSize = dictBufferSize; + cctxprep.cLevel = cLevel; + cctxprep.comprParams = comprParams; + cctxprep.adv = adv; + + dbp.benchFn = local_defaultDecompress; + dbp.benchPayload = dctx; + dbp.initFn = local_initDCtx; + dbp.initPayload = &dctxprep; + dbp.errorFn = ZSTD_isError; + dbp.blockCount = nbBlocks; + dbp.srcBuffers = (const void* const *) cPtrs; + dbp.srcSizes = cSizes; + dbp.dstBuffers = resPtrs; + dbp.dstCapacities = resSizes; + dbp.blockResults = NULL; + + dctxprep.dctx = dctx; + dctxprep.dictBuffer = dictBuffer; + dctxprep.dictBufferSize = dictBufferSize; + OUTPUTLEVEL(2, "\r%70s\r", ""); /* blank line */ assert(srcSize < UINT_MAX); OUTPUTLEVEL(2, "%2s-%-17.17s :%10u -> \r", marks[markNb], displayName, (unsigned)srcSize); - - while (!(compressionCompleted && decompressionCompleted)) { - if (!compressionCompleted) { - BMK_runOutcome_t const cOutcome = BMK_benchTimedFn( timeStateCompress, cbp); - - if (!BMK_isSuccessful_runOutcome(cOutcome)) { - return BMK_benchOutcome_error(); - } - - { BMK_runTime_t const cResult = BMK_extract_runTime(cOutcome); - cSize = cResult.sumOfReturn; + + while (!(compressionCompleted && decompressionCompleted)) { + if (!compressionCompleted) { + BMK_runOutcome_t const cOutcome = BMK_benchTimedFn( timeStateCompress, cbp); + + if (!BMK_isSuccessful_runOutcome(cOutcome)) { + return BMK_benchOutcome_error(); + } + + { BMK_runTime_t const cResult = BMK_extract_runTime(cOutcome); + cSize = cResult.sumOfReturn; ratio = (double)srcSize / (double)cSize; - { BMK_benchResult_t newResult; - newResult.cSpeed = (U64)((double)srcSize * TIMELOOP_NANOSEC / cResult.nanoSecPerRun); - benchResult.cSize = cSize; - if (newResult.cSpeed > benchResult.cSpeed) - benchResult.cSpeed = newResult.cSpeed; - } } - - { int const ratioAccuracy = (ratio < 10.) ? 3 : 2; + { BMK_benchResult_t newResult; + newResult.cSpeed = (U64)((double)srcSize * TIMELOOP_NANOSEC / cResult.nanoSecPerRun); + benchResult.cSize = cSize; + if (newResult.cSpeed > benchResult.cSpeed) + benchResult.cSpeed = newResult.cSpeed; + } } + + { int const ratioAccuracy = (ratio < 10.) ? 3 : 2; assert(cSize < UINT_MAX); OUTPUTLEVEL(2, "%2s-%-17.17s :%10u ->%10u (x%5.*f), %6.*f MB/s \r", - marks[markNb], displayName, - (unsigned)srcSize, (unsigned)cSize, - ratioAccuracy, ratio, + marks[markNb], displayName, + (unsigned)srcSize, (unsigned)cSize, + ratioAccuracy, ratio, benchResult.cSpeed < (10 * MB_UNIT) ? 2 : 1, (double)benchResult.cSpeed / MB_UNIT); - } - compressionCompleted = BMK_isCompleted_TimedFn(timeStateCompress); - } - - if(!decompressionCompleted) { - BMK_runOutcome_t const dOutcome = BMK_benchTimedFn(timeStateDecompress, dbp); - - if(!BMK_isSuccessful_runOutcome(dOutcome)) { - return BMK_benchOutcome_error(); - } - - { BMK_runTime_t const dResult = BMK_extract_runTime(dOutcome); - U64 const newDSpeed = (U64)((double)srcSize * TIMELOOP_NANOSEC / dResult.nanoSecPerRun); - if (newDSpeed > benchResult.dSpeed) - benchResult.dSpeed = newDSpeed; - } - - { int const ratioAccuracy = (ratio < 10.) ? 3 : 2; + } + compressionCompleted = BMK_isCompleted_TimedFn(timeStateCompress); + } + + if(!decompressionCompleted) { + BMK_runOutcome_t const dOutcome = BMK_benchTimedFn(timeStateDecompress, dbp); + + if(!BMK_isSuccessful_runOutcome(dOutcome)) { + return BMK_benchOutcome_error(); + } + + { BMK_runTime_t const dResult = BMK_extract_runTime(dOutcome); + U64 const newDSpeed = (U64)((double)srcSize * TIMELOOP_NANOSEC / dResult.nanoSecPerRun); + if (newDSpeed > benchResult.dSpeed) + benchResult.dSpeed = newDSpeed; + } + + { int const ratioAccuracy = (ratio < 10.) ? 3 : 2; OUTPUTLEVEL(2, "%2s-%-17.17s :%10u ->%10u (x%5.*f), %6.*f MB/s, %6.1f MB/s\r", - marks[markNb], displayName, - (unsigned)srcSize, (unsigned)cSize, - ratioAccuracy, ratio, + marks[markNb], displayName, + (unsigned)srcSize, (unsigned)cSize, + ratioAccuracy, ratio, benchResult.cSpeed < (10 * MB_UNIT) ? 2 : 1, (double)benchResult.cSpeed / MB_UNIT, - (double)benchResult.dSpeed / MB_UNIT); - } - decompressionCompleted = BMK_isCompleted_TimedFn(timeStateDecompress); - } - markNb = (markNb+1) % NB_MARKS; - } /* while (!(compressionCompleted && decompressionCompleted)) */ - - /* CRC Checking */ - { const BYTE* resultBuffer = (const BYTE*)(*resultBufferPtr); - U64 const crcCheck = XXH64(resultBuffer, srcSize, 0); - if ((adv->mode == BMK_both) && (crcOrig!=crcCheck)) { - size_t u; - DISPLAY("!!! WARNING !!! %14s : Invalid Checksum : %x != %x \n", - displayName, (unsigned)crcOrig, (unsigned)crcCheck); - for (u=0; u<srcSize; u++) { - if (((const BYTE*)srcBuffer)[u] != resultBuffer[u]) { - unsigned segNb, bNb, pos; - size_t bacc = 0; - DISPLAY("Decoding error at pos %u ", (unsigned)u); - for (segNb = 0; segNb < nbBlocks; segNb++) { - if (bacc + srcSizes[segNb] > u) break; - bacc += srcSizes[segNb]; - } - pos = (U32)(u - bacc); - bNb = pos / (128 KB); - DISPLAY("(sample %u, block %u, pos %u) \n", segNb, bNb, pos); - { size_t const lowest = (u>5) ? 5 : u; - size_t n; - DISPLAY("origin: "); + (double)benchResult.dSpeed / MB_UNIT); + } + decompressionCompleted = BMK_isCompleted_TimedFn(timeStateDecompress); + } + markNb = (markNb+1) % NB_MARKS; + } /* while (!(compressionCompleted && decompressionCompleted)) */ + + /* CRC Checking */ + { const BYTE* resultBuffer = (const BYTE*)(*resultBufferPtr); + U64 const crcCheck = XXH64(resultBuffer, srcSize, 0); + if ((adv->mode == BMK_both) && (crcOrig!=crcCheck)) { + size_t u; + DISPLAY("!!! WARNING !!! %14s : Invalid Checksum : %x != %x \n", + displayName, (unsigned)crcOrig, (unsigned)crcCheck); + for (u=0; u<srcSize; u++) { + if (((const BYTE*)srcBuffer)[u] != resultBuffer[u]) { + unsigned segNb, bNb, pos; + size_t bacc = 0; + DISPLAY("Decoding error at pos %u ", (unsigned)u); + for (segNb = 0; segNb < nbBlocks; segNb++) { + if (bacc + srcSizes[segNb] > u) break; + bacc += srcSizes[segNb]; + } + pos = (U32)(u - bacc); + bNb = pos / (128 KB); + DISPLAY("(sample %u, block %u, pos %u) \n", segNb, bNb, pos); + { size_t const lowest = (u>5) ? 5 : u; + size_t n; + DISPLAY("origin: "); + for (n=lowest; n>0; n--) + DISPLAY("%02X ", ((const BYTE*)srcBuffer)[u-n]); + DISPLAY(" :%02X: ", ((const BYTE*)srcBuffer)[u]); + for (n=1; n<3; n++) + DISPLAY("%02X ", ((const BYTE*)srcBuffer)[u+n]); + DISPLAY(" \n"); + DISPLAY("decode: "); for (n=lowest; n>0; n--) - DISPLAY("%02X ", ((const BYTE*)srcBuffer)[u-n]); - DISPLAY(" :%02X: ", ((const BYTE*)srcBuffer)[u]); - for (n=1; n<3; n++) - DISPLAY("%02X ", ((const BYTE*)srcBuffer)[u+n]); - DISPLAY(" \n"); - DISPLAY("decode: "); - for (n=lowest; n>0; n--) - DISPLAY("%02X ", resultBuffer[u-n]); - DISPLAY(" :%02X: ", resultBuffer[u]); - for (n=1; n<3; n++) - DISPLAY("%02X ", resultBuffer[u+n]); - DISPLAY(" \n"); - } - break; - } - if (u==srcSize-1) { /* should never happen */ - DISPLAY("no difference detected\n"); - } - } /* for (u=0; u<srcSize; u++) */ - } /* if ((adv->mode == BMK_both) && (crcOrig!=crcCheck)) */ - } /* CRC Checking */ - - if (displayLevel == 1) { /* hidden display mode -q, used by python speed benchmark */ - double const cSpeed = (double)benchResult.cSpeed / MB_UNIT; - double const dSpeed = (double)benchResult.dSpeed / MB_UNIT; - if (adv->additionalParam) { + DISPLAY("%02X ", resultBuffer[u-n]); + DISPLAY(" :%02X: ", resultBuffer[u]); + for (n=1; n<3; n++) + DISPLAY("%02X ", resultBuffer[u+n]); + DISPLAY(" \n"); + } + break; + } + if (u==srcSize-1) { /* should never happen */ + DISPLAY("no difference detected\n"); + } + } /* for (u=0; u<srcSize; u++) */ + } /* if ((adv->mode == BMK_both) && (crcOrig!=crcCheck)) */ + } /* CRC Checking */ + + if (displayLevel == 1) { /* hidden display mode -q, used by python speed benchmark */ + double const cSpeed = (double)benchResult.cSpeed / MB_UNIT; + double const dSpeed = (double)benchResult.dSpeed / MB_UNIT; + if (adv->additionalParam) { OUTPUT("-%-3i%11i (%5.3f) %6.2f MB/s %6.1f MB/s %s (param=%d)\n", cLevel, (int)cSize, ratio, cSpeed, dSpeed, displayName, adv->additionalParam); - } else { + } else { OUTPUT("-%-3i%11i (%5.3f) %6.2f MB/s %6.1f MB/s %s\n", cLevel, (int)cSize, ratio, cSpeed, dSpeed, displayName); - } - } - + } + } + OUTPUTLEVEL(2, "%2i#\n", cLevel); - } /* Bench */ - - benchResult.cMem = (1ULL << (comprParams->windowLog)) + ZSTD_sizeof_CCtx(cctx); - return BMK_benchOutcome_setValidResult(benchResult); -} - -BMK_benchOutcome_t BMK_benchMemAdvanced(const void* srcBuffer, size_t srcSize, - void* dstBuffer, size_t dstCapacity, - const size_t* fileSizes, unsigned nbFiles, - int cLevel, const ZSTD_compressionParameters* comprParams, - const void* dictBuffer, size_t dictBufferSize, - int displayLevel, const char* displayName, const BMK_advancedParams_t* adv) - -{ - int const dstParamsError = !dstBuffer ^ !dstCapacity; /* must be both NULL or none */ - - size_t const blockSize = ((adv->blockSize>=32 && (adv->mode != BMK_decodeOnly)) ? adv->blockSize : srcSize) + (!srcSize) /* avoid div by 0 */ ; - U32 const maxNbBlocks = (U32) ((srcSize + (blockSize-1)) / blockSize) + nbFiles; - - /* these are the blockTable parameters, just split up */ - const void ** const srcPtrs = (const void**)malloc(maxNbBlocks * sizeof(void*)); - size_t* const srcSizes = (size_t*)malloc(maxNbBlocks * sizeof(size_t)); - - - void ** const cPtrs = (void**)malloc(maxNbBlocks * sizeof(void*)); - size_t* const cSizes = (size_t*)malloc(maxNbBlocks * sizeof(size_t)); - size_t* const cCapacities = (size_t*)malloc(maxNbBlocks * sizeof(size_t)); - - void ** const resPtrs = (void**)malloc(maxNbBlocks * sizeof(void*)); - size_t* const resSizes = (size_t*)malloc(maxNbBlocks * sizeof(size_t)); - - BMK_timedFnState_t* timeStateCompress = BMK_createTimedFnState(adv->nbSeconds * 1000, BMK_RUNTEST_DEFAULT_MS); - BMK_timedFnState_t* timeStateDecompress = BMK_createTimedFnState(adv->nbSeconds * 1000, BMK_RUNTEST_DEFAULT_MS); - - ZSTD_CCtx* const cctx = ZSTD_createCCtx(); - ZSTD_DCtx* const dctx = ZSTD_createDCtx(); - - const size_t maxCompressedSize = dstCapacity ? dstCapacity : ZSTD_compressBound(srcSize) + (maxNbBlocks * 1024); - - void* const internalDstBuffer = dstBuffer ? NULL : malloc(maxCompressedSize); - void* const compressedBuffer = dstBuffer ? dstBuffer : internalDstBuffer; - - BMK_benchOutcome_t outcome = BMK_benchOutcome_error(); /* error by default */ - - void* resultBuffer = srcSize ? malloc(srcSize) : NULL; - - int allocationincomplete = !srcPtrs || !srcSizes || !cPtrs || - !cSizes || !cCapacities || !resPtrs || !resSizes || - !timeStateCompress || !timeStateDecompress || - !cctx || !dctx || - !compressedBuffer || !resultBuffer; - - - if (!allocationincomplete && !dstParamsError) { - outcome = BMK_benchMemAdvancedNoAlloc(srcPtrs, srcSizes, - cPtrs, cCapacities, cSizes, - resPtrs, resSizes, - &resultBuffer, - compressedBuffer, maxCompressedSize, - timeStateCompress, timeStateDecompress, - srcBuffer, srcSize, - fileSizes, nbFiles, - cLevel, comprParams, - dictBuffer, dictBufferSize, - cctx, dctx, - displayLevel, displayName, adv); - } - - /* clean up */ - BMK_freeTimedFnState(timeStateCompress); - BMK_freeTimedFnState(timeStateDecompress); - - ZSTD_freeCCtx(cctx); - ZSTD_freeDCtx(dctx); - - free(internalDstBuffer); - free(resultBuffer); - - free((void*)srcPtrs); - free(srcSizes); - free(cPtrs); - free(cSizes); - free(cCapacities); - free(resPtrs); - free(resSizes); - - if(allocationincomplete) { - RETURN_ERROR(31, BMK_benchOutcome_t, "allocation error : not enough memory"); - } - - if(dstParamsError) { - RETURN_ERROR(32, BMK_benchOutcome_t, "Dst parameters not coherent"); - } - return outcome; -} - -BMK_benchOutcome_t BMK_benchMem(const void* srcBuffer, size_t srcSize, - const size_t* fileSizes, unsigned nbFiles, - int cLevel, const ZSTD_compressionParameters* comprParams, - const void* dictBuffer, size_t dictBufferSize, - int displayLevel, const char* displayName) { - - BMK_advancedParams_t const adv = BMK_initAdvancedParams(); - return BMK_benchMemAdvanced(srcBuffer, srcSize, - NULL, 0, - fileSizes, nbFiles, - cLevel, comprParams, - dictBuffer, dictBufferSize, - displayLevel, displayName, &adv); -} - -static BMK_benchOutcome_t BMK_benchCLevel(const void* srcBuffer, size_t benchedSize, - const size_t* fileSizes, unsigned nbFiles, - int cLevel, const ZSTD_compressionParameters* comprParams, - const void* dictBuffer, size_t dictBufferSize, - int displayLevel, const char* displayName, - BMK_advancedParams_t const * const adv) -{ - const char* pch = strrchr(displayName, '\\'); /* Windows */ - if (!pch) pch = strrchr(displayName, '/'); /* Linux */ - if (pch) displayName = pch+1; - - if (adv->realTime) { - DISPLAYLEVEL(2, "Note : switching to real-time priority \n"); - SET_REALTIME_PRIORITY; - } - - if (displayLevel == 1 && !adv->additionalParam) /* --quiet mode */ + } /* Bench */ + + benchResult.cMem = (1ULL << (comprParams->windowLog)) + ZSTD_sizeof_CCtx(cctx); + return BMK_benchOutcome_setValidResult(benchResult); +} + +BMK_benchOutcome_t BMK_benchMemAdvanced(const void* srcBuffer, size_t srcSize, + void* dstBuffer, size_t dstCapacity, + const size_t* fileSizes, unsigned nbFiles, + int cLevel, const ZSTD_compressionParameters* comprParams, + const void* dictBuffer, size_t dictBufferSize, + int displayLevel, const char* displayName, const BMK_advancedParams_t* adv) + +{ + int const dstParamsError = !dstBuffer ^ !dstCapacity; /* must be both NULL or none */ + + size_t const blockSize = ((adv->blockSize>=32 && (adv->mode != BMK_decodeOnly)) ? adv->blockSize : srcSize) + (!srcSize) /* avoid div by 0 */ ; + U32 const maxNbBlocks = (U32) ((srcSize + (blockSize-1)) / blockSize) + nbFiles; + + /* these are the blockTable parameters, just split up */ + const void ** const srcPtrs = (const void**)malloc(maxNbBlocks * sizeof(void*)); + size_t* const srcSizes = (size_t*)malloc(maxNbBlocks * sizeof(size_t)); + + + void ** const cPtrs = (void**)malloc(maxNbBlocks * sizeof(void*)); + size_t* const cSizes = (size_t*)malloc(maxNbBlocks * sizeof(size_t)); + size_t* const cCapacities = (size_t*)malloc(maxNbBlocks * sizeof(size_t)); + + void ** const resPtrs = (void**)malloc(maxNbBlocks * sizeof(void*)); + size_t* const resSizes = (size_t*)malloc(maxNbBlocks * sizeof(size_t)); + + BMK_timedFnState_t* timeStateCompress = BMK_createTimedFnState(adv->nbSeconds * 1000, BMK_RUNTEST_DEFAULT_MS); + BMK_timedFnState_t* timeStateDecompress = BMK_createTimedFnState(adv->nbSeconds * 1000, BMK_RUNTEST_DEFAULT_MS); + + ZSTD_CCtx* const cctx = ZSTD_createCCtx(); + ZSTD_DCtx* const dctx = ZSTD_createDCtx(); + + const size_t maxCompressedSize = dstCapacity ? dstCapacity : ZSTD_compressBound(srcSize) + (maxNbBlocks * 1024); + + void* const internalDstBuffer = dstBuffer ? NULL : malloc(maxCompressedSize); + void* const compressedBuffer = dstBuffer ? dstBuffer : internalDstBuffer; + + BMK_benchOutcome_t outcome = BMK_benchOutcome_error(); /* error by default */ + + void* resultBuffer = srcSize ? malloc(srcSize) : NULL; + + int allocationincomplete = !srcPtrs || !srcSizes || !cPtrs || + !cSizes || !cCapacities || !resPtrs || !resSizes || + !timeStateCompress || !timeStateDecompress || + !cctx || !dctx || + !compressedBuffer || !resultBuffer; + + + if (!allocationincomplete && !dstParamsError) { + outcome = BMK_benchMemAdvancedNoAlloc(srcPtrs, srcSizes, + cPtrs, cCapacities, cSizes, + resPtrs, resSizes, + &resultBuffer, + compressedBuffer, maxCompressedSize, + timeStateCompress, timeStateDecompress, + srcBuffer, srcSize, + fileSizes, nbFiles, + cLevel, comprParams, + dictBuffer, dictBufferSize, + cctx, dctx, + displayLevel, displayName, adv); + } + + /* clean up */ + BMK_freeTimedFnState(timeStateCompress); + BMK_freeTimedFnState(timeStateDecompress); + + ZSTD_freeCCtx(cctx); + ZSTD_freeDCtx(dctx); + + free(internalDstBuffer); + free(resultBuffer); + + free((void*)srcPtrs); + free(srcSizes); + free(cPtrs); + free(cSizes); + free(cCapacities); + free(resPtrs); + free(resSizes); + + if(allocationincomplete) { + RETURN_ERROR(31, BMK_benchOutcome_t, "allocation error : not enough memory"); + } + + if(dstParamsError) { + RETURN_ERROR(32, BMK_benchOutcome_t, "Dst parameters not coherent"); + } + return outcome; +} + +BMK_benchOutcome_t BMK_benchMem(const void* srcBuffer, size_t srcSize, + const size_t* fileSizes, unsigned nbFiles, + int cLevel, const ZSTD_compressionParameters* comprParams, + const void* dictBuffer, size_t dictBufferSize, + int displayLevel, const char* displayName) { + + BMK_advancedParams_t const adv = BMK_initAdvancedParams(); + return BMK_benchMemAdvanced(srcBuffer, srcSize, + NULL, 0, + fileSizes, nbFiles, + cLevel, comprParams, + dictBuffer, dictBufferSize, + displayLevel, displayName, &adv); +} + +static BMK_benchOutcome_t BMK_benchCLevel(const void* srcBuffer, size_t benchedSize, + const size_t* fileSizes, unsigned nbFiles, + int cLevel, const ZSTD_compressionParameters* comprParams, + const void* dictBuffer, size_t dictBufferSize, + int displayLevel, const char* displayName, + BMK_advancedParams_t const * const adv) +{ + const char* pch = strrchr(displayName, '\\'); /* Windows */ + if (!pch) pch = strrchr(displayName, '/'); /* Linux */ + if (pch) displayName = pch+1; + + if (adv->realTime) { + DISPLAYLEVEL(2, "Note : switching to real-time priority \n"); + SET_REALTIME_PRIORITY; + } + + if (displayLevel == 1 && !adv->additionalParam) /* --quiet mode */ OUTPUT("bench %s %s: input %u bytes, %u seconds, %u KB blocks\n", - ZSTD_VERSION_STRING, ZSTD_GIT_COMMIT_STRING, - (unsigned)benchedSize, adv->nbSeconds, (unsigned)(adv->blockSize>>10)); - - return BMK_benchMemAdvanced(srcBuffer, benchedSize, - NULL, 0, - fileSizes, nbFiles, - cLevel, comprParams, - dictBuffer, dictBufferSize, - displayLevel, displayName, adv); -} - -BMK_benchOutcome_t BMK_syntheticTest(int cLevel, double compressibility, - const ZSTD_compressionParameters* compressionParams, - int displayLevel, const BMK_advancedParams_t* adv) -{ - char name[20] = {0}; - size_t const benchedSize = 10000000; - void* srcBuffer; - BMK_benchOutcome_t res; - - if (cLevel > ZSTD_maxCLevel()) { - RETURN_ERROR(15, BMK_benchOutcome_t, "Invalid Compression Level"); - } - - /* Memory allocation */ - srcBuffer = malloc(benchedSize); - if (!srcBuffer) RETURN_ERROR(21, BMK_benchOutcome_t, "not enough memory"); - - /* Fill input buffer */ - RDG_genBuffer(srcBuffer, benchedSize, compressibility, 0.0, 0); - - /* Bench */ - snprintf (name, sizeof(name), "Synthetic %2u%%", (unsigned)(compressibility*100)); - res = BMK_benchCLevel(srcBuffer, benchedSize, - &benchedSize /* ? */, 1 /* ? */, - cLevel, compressionParams, - NULL, 0, /* dictionary */ - displayLevel, name, adv); - - /* clean up */ - free(srcBuffer); - - return res; -} - - - -static size_t BMK_findMaxMem(U64 requiredMem) -{ - size_t const step = 64 MB; - BYTE* testmem = NULL; - - requiredMem = (((requiredMem >> 26) + 1) << 26); - requiredMem += step; - if (requiredMem > maxMemory) requiredMem = maxMemory; - - do { - testmem = (BYTE*)malloc((size_t)requiredMem); - requiredMem -= step; - } while (!testmem && requiredMem > 0); - - free(testmem); - return (size_t)(requiredMem); -} - -/*! BMK_loadFiles() : - * Loads `buffer` with content of files listed within `fileNamesTable`. - * At most, fills `buffer` entirely. */ -static int BMK_loadFiles(void* buffer, size_t bufferSize, - size_t* fileSizes, - const char* const * fileNamesTable, unsigned nbFiles, - int displayLevel) -{ - size_t pos = 0, totalSize = 0; - unsigned n; - for (n=0; n<nbFiles; n++) { - U64 fileSize = UTIL_getFileSize(fileNamesTable[n]); /* last file may be shortened */ - if (UTIL_isDirectory(fileNamesTable[n])) { - DISPLAYLEVEL(2, "Ignoring %s directory... \n", fileNamesTable[n]); - fileSizes[n] = 0; - continue; - } - if (fileSize == UTIL_FILESIZE_UNKNOWN) { - DISPLAYLEVEL(2, "Cannot evaluate size of %s, ignoring ... \n", fileNamesTable[n]); - fileSizes[n] = 0; - continue; - } - { FILE* const f = fopen(fileNamesTable[n], "rb"); - if (f==NULL) RETURN_ERROR_INT(10, "impossible to open file %s", fileNamesTable[n]); + ZSTD_VERSION_STRING, ZSTD_GIT_COMMIT_STRING, + (unsigned)benchedSize, adv->nbSeconds, (unsigned)(adv->blockSize>>10)); + + return BMK_benchMemAdvanced(srcBuffer, benchedSize, + NULL, 0, + fileSizes, nbFiles, + cLevel, comprParams, + dictBuffer, dictBufferSize, + displayLevel, displayName, adv); +} + +BMK_benchOutcome_t BMK_syntheticTest(int cLevel, double compressibility, + const ZSTD_compressionParameters* compressionParams, + int displayLevel, const BMK_advancedParams_t* adv) +{ + char name[20] = {0}; + size_t const benchedSize = 10000000; + void* srcBuffer; + BMK_benchOutcome_t res; + + if (cLevel > ZSTD_maxCLevel()) { + RETURN_ERROR(15, BMK_benchOutcome_t, "Invalid Compression Level"); + } + + /* Memory allocation */ + srcBuffer = malloc(benchedSize); + if (!srcBuffer) RETURN_ERROR(21, BMK_benchOutcome_t, "not enough memory"); + + /* Fill input buffer */ + RDG_genBuffer(srcBuffer, benchedSize, compressibility, 0.0, 0); + + /* Bench */ + snprintf (name, sizeof(name), "Synthetic %2u%%", (unsigned)(compressibility*100)); + res = BMK_benchCLevel(srcBuffer, benchedSize, + &benchedSize /* ? */, 1 /* ? */, + cLevel, compressionParams, + NULL, 0, /* dictionary */ + displayLevel, name, adv); + + /* clean up */ + free(srcBuffer); + + return res; +} + + + +static size_t BMK_findMaxMem(U64 requiredMem) +{ + size_t const step = 64 MB; + BYTE* testmem = NULL; + + requiredMem = (((requiredMem >> 26) + 1) << 26); + requiredMem += step; + if (requiredMem > maxMemory) requiredMem = maxMemory; + + do { + testmem = (BYTE*)malloc((size_t)requiredMem); + requiredMem -= step; + } while (!testmem && requiredMem > 0); + + free(testmem); + return (size_t)(requiredMem); +} + +/*! BMK_loadFiles() : + * Loads `buffer` with content of files listed within `fileNamesTable`. + * At most, fills `buffer` entirely. */ +static int BMK_loadFiles(void* buffer, size_t bufferSize, + size_t* fileSizes, + const char* const * fileNamesTable, unsigned nbFiles, + int displayLevel) +{ + size_t pos = 0, totalSize = 0; + unsigned n; + for (n=0; n<nbFiles; n++) { + U64 fileSize = UTIL_getFileSize(fileNamesTable[n]); /* last file may be shortened */ + if (UTIL_isDirectory(fileNamesTable[n])) { + DISPLAYLEVEL(2, "Ignoring %s directory... \n", fileNamesTable[n]); + fileSizes[n] = 0; + continue; + } + if (fileSize == UTIL_FILESIZE_UNKNOWN) { + DISPLAYLEVEL(2, "Cannot evaluate size of %s, ignoring ... \n", fileNamesTable[n]); + fileSizes[n] = 0; + continue; + } + { FILE* const f = fopen(fileNamesTable[n], "rb"); + if (f==NULL) RETURN_ERROR_INT(10, "impossible to open file %s", fileNamesTable[n]); OUTPUTLEVEL(2, "Loading %s... \r", fileNamesTable[n]); - if (fileSize > bufferSize-pos) fileSize = bufferSize-pos, nbFiles=n; /* buffer too small - stop after this file */ - { size_t const readSize = fread(((char*)buffer)+pos, 1, (size_t)fileSize, f); - if (readSize != (size_t)fileSize) RETURN_ERROR_INT(11, "could not read %s", fileNamesTable[n]); - pos += readSize; - } - fileSizes[n] = (size_t)fileSize; - totalSize += (size_t)fileSize; - fclose(f); - } } - - if (totalSize == 0) RETURN_ERROR_INT(12, "no data to bench"); - return 0; -} - -BMK_benchOutcome_t BMK_benchFilesAdvanced( - const char* const * fileNamesTable, unsigned nbFiles, - const char* dictFileName, int cLevel, - const ZSTD_compressionParameters* compressionParams, - int displayLevel, const BMK_advancedParams_t* adv) -{ - void* srcBuffer = NULL; - size_t benchedSize; - void* dictBuffer = NULL; - size_t dictBufferSize = 0; - size_t* fileSizes = NULL; - BMK_benchOutcome_t res; - U64 const totalSizeToLoad = UTIL_getTotalFileSize(fileNamesTable, nbFiles); - - if (!nbFiles) { - RETURN_ERROR(14, BMK_benchOutcome_t, "No Files to Benchmark"); - } - - if (cLevel > ZSTD_maxCLevel()) { - RETURN_ERROR(15, BMK_benchOutcome_t, "Invalid Compression Level"); - } - + if (fileSize > bufferSize-pos) fileSize = bufferSize-pos, nbFiles=n; /* buffer too small - stop after this file */ + { size_t const readSize = fread(((char*)buffer)+pos, 1, (size_t)fileSize, f); + if (readSize != (size_t)fileSize) RETURN_ERROR_INT(11, "could not read %s", fileNamesTable[n]); + pos += readSize; + } + fileSizes[n] = (size_t)fileSize; + totalSize += (size_t)fileSize; + fclose(f); + } } + + if (totalSize == 0) RETURN_ERROR_INT(12, "no data to bench"); + return 0; +} + +BMK_benchOutcome_t BMK_benchFilesAdvanced( + const char* const * fileNamesTable, unsigned nbFiles, + const char* dictFileName, int cLevel, + const ZSTD_compressionParameters* compressionParams, + int displayLevel, const BMK_advancedParams_t* adv) +{ + void* srcBuffer = NULL; + size_t benchedSize; + void* dictBuffer = NULL; + size_t dictBufferSize = 0; + size_t* fileSizes = NULL; + BMK_benchOutcome_t res; + U64 const totalSizeToLoad = UTIL_getTotalFileSize(fileNamesTable, nbFiles); + + if (!nbFiles) { + RETURN_ERROR(14, BMK_benchOutcome_t, "No Files to Benchmark"); + } + + if (cLevel > ZSTD_maxCLevel()) { + RETURN_ERROR(15, BMK_benchOutcome_t, "Invalid Compression Level"); + } + if (totalSizeToLoad == UTIL_FILESIZE_UNKNOWN) { RETURN_ERROR(9, BMK_benchOutcome_t, "Error loading files"); } - fileSizes = (size_t*)calloc(nbFiles, sizeof(size_t)); - if (!fileSizes) RETURN_ERROR(12, BMK_benchOutcome_t, "not enough memory for fileSizes"); - - /* Load dictionary */ - if (dictFileName != NULL) { - U64 const dictFileSize = UTIL_getFileSize(dictFileName); - if (dictFileSize == UTIL_FILESIZE_UNKNOWN) { - DISPLAYLEVEL(1, "error loading %s : %s \n", dictFileName, strerror(errno)); - free(fileSizes); - RETURN_ERROR(9, BMK_benchOutcome_t, "benchmark aborted"); - } - if (dictFileSize > 64 MB) { - free(fileSizes); - RETURN_ERROR(10, BMK_benchOutcome_t, "dictionary file %s too large", dictFileName); - } - dictBufferSize = (size_t)dictFileSize; - dictBuffer = malloc(dictBufferSize); - if (dictBuffer==NULL) { - free(fileSizes); - RETURN_ERROR(11, BMK_benchOutcome_t, "not enough memory for dictionary (%u bytes)", - (unsigned)dictBufferSize); - } - - { int const errorCode = BMK_loadFiles(dictBuffer, dictBufferSize, - fileSizes, &dictFileName /*?*/, - 1 /*?*/, displayLevel); - if (errorCode) { - res = BMK_benchOutcome_error(); - goto _cleanUp; - } } - } - - /* Memory allocation & restrictions */ - benchedSize = BMK_findMaxMem(totalSizeToLoad * 3) / 3; - if ((U64)benchedSize > totalSizeToLoad) benchedSize = (size_t)totalSizeToLoad; - if (benchedSize < totalSizeToLoad) - DISPLAY("Not enough memory; testing %u MB only...\n", (unsigned)(benchedSize >> 20)); - - srcBuffer = benchedSize ? malloc(benchedSize) : NULL; - if (!srcBuffer) { - free(dictBuffer); - free(fileSizes); - RETURN_ERROR(12, BMK_benchOutcome_t, "not enough memory"); - } - - /* Load input buffer */ - { int const errorCode = BMK_loadFiles(srcBuffer, benchedSize, - fileSizes, fileNamesTable, nbFiles, - displayLevel); - if (errorCode) { - res = BMK_benchOutcome_error(); - goto _cleanUp; - } } - - /* Bench */ - { char mfName[20] = {0}; - snprintf (mfName, sizeof(mfName), " %u files", nbFiles); - { const char* const displayName = (nbFiles > 1) ? mfName : fileNamesTable[0]; - res = BMK_benchCLevel(srcBuffer, benchedSize, - fileSizes, nbFiles, - cLevel, compressionParams, - dictBuffer, dictBufferSize, - displayLevel, displayName, - adv); - } } - -_cleanUp: - free(srcBuffer); - free(dictBuffer); - free(fileSizes); - return res; -} - - -BMK_benchOutcome_t BMK_benchFiles( - const char* const * fileNamesTable, unsigned nbFiles, - const char* dictFileName, - int cLevel, const ZSTD_compressionParameters* compressionParams, - int displayLevel) -{ - BMK_advancedParams_t const adv = BMK_initAdvancedParams(); - return BMK_benchFilesAdvanced(fileNamesTable, nbFiles, dictFileName, cLevel, compressionParams, displayLevel, &adv); -} + fileSizes = (size_t*)calloc(nbFiles, sizeof(size_t)); + if (!fileSizes) RETURN_ERROR(12, BMK_benchOutcome_t, "not enough memory for fileSizes"); + + /* Load dictionary */ + if (dictFileName != NULL) { + U64 const dictFileSize = UTIL_getFileSize(dictFileName); + if (dictFileSize == UTIL_FILESIZE_UNKNOWN) { + DISPLAYLEVEL(1, "error loading %s : %s \n", dictFileName, strerror(errno)); + free(fileSizes); + RETURN_ERROR(9, BMK_benchOutcome_t, "benchmark aborted"); + } + if (dictFileSize > 64 MB) { + free(fileSizes); + RETURN_ERROR(10, BMK_benchOutcome_t, "dictionary file %s too large", dictFileName); + } + dictBufferSize = (size_t)dictFileSize; + dictBuffer = malloc(dictBufferSize); + if (dictBuffer==NULL) { + free(fileSizes); + RETURN_ERROR(11, BMK_benchOutcome_t, "not enough memory for dictionary (%u bytes)", + (unsigned)dictBufferSize); + } + + { int const errorCode = BMK_loadFiles(dictBuffer, dictBufferSize, + fileSizes, &dictFileName /*?*/, + 1 /*?*/, displayLevel); + if (errorCode) { + res = BMK_benchOutcome_error(); + goto _cleanUp; + } } + } + + /* Memory allocation & restrictions */ + benchedSize = BMK_findMaxMem(totalSizeToLoad * 3) / 3; + if ((U64)benchedSize > totalSizeToLoad) benchedSize = (size_t)totalSizeToLoad; + if (benchedSize < totalSizeToLoad) + DISPLAY("Not enough memory; testing %u MB only...\n", (unsigned)(benchedSize >> 20)); + + srcBuffer = benchedSize ? malloc(benchedSize) : NULL; + if (!srcBuffer) { + free(dictBuffer); + free(fileSizes); + RETURN_ERROR(12, BMK_benchOutcome_t, "not enough memory"); + } + + /* Load input buffer */ + { int const errorCode = BMK_loadFiles(srcBuffer, benchedSize, + fileSizes, fileNamesTable, nbFiles, + displayLevel); + if (errorCode) { + res = BMK_benchOutcome_error(); + goto _cleanUp; + } } + + /* Bench */ + { char mfName[20] = {0}; + snprintf (mfName, sizeof(mfName), " %u files", nbFiles); + { const char* const displayName = (nbFiles > 1) ? mfName : fileNamesTable[0]; + res = BMK_benchCLevel(srcBuffer, benchedSize, + fileSizes, nbFiles, + cLevel, compressionParams, + dictBuffer, dictBufferSize, + displayLevel, displayName, + adv); + } } + +_cleanUp: + free(srcBuffer); + free(dictBuffer); + free(fileSizes); + return res; +} + + +BMK_benchOutcome_t BMK_benchFiles( + const char* const * fileNamesTable, unsigned nbFiles, + const char* dictFileName, + int cLevel, const ZSTD_compressionParameters* compressionParams, + int displayLevel) +{ + BMK_advancedParams_t const adv = BMK_initAdvancedParams(); + return BMK_benchFilesAdvanced(fileNamesTable, nbFiles, dictFileName, cLevel, compressionParams, displayLevel, &adv); +} diff --git a/contrib/libs/zstd/programs/benchzstd.h b/contrib/libs/zstd/programs/benchzstd.h index 11ac85da7f..8dcecd08ec 100644 --- a/contrib/libs/zstd/programs/benchzstd.h +++ b/contrib/libs/zstd/programs/benchzstd.h @@ -1,213 +1,213 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - /* benchzstd : - * benchmark Zstandard compression / decompression - * over a set of files or buffers - * and display progress result and final summary - */ - -#if defined (__cplusplus) -extern "C" { -#endif - -#ifndef BENCH_ZSTD_H_3242387 -#define BENCH_ZSTD_H_3242387 - -/* === Dependencies === */ -#include <stddef.h> /* size_t */ -#define ZSTD_STATIC_LINKING_ONLY /* ZSTD_compressionParameters */ -#include "../lib/zstd.h" /* ZSTD_compressionParameters */ - - -/* === Constants === */ - -#define MB_UNIT 1000000 - - -/* === Benchmark functions === */ - -/* Creates a variant `typeName`, able to express "error or valid result". - * Functions with return type `typeName` - * must first check if result is valid, using BMK_isSuccessful_*(), - * and only then can extract `baseType`. - */ -#define VARIANT_ERROR_RESULT(baseType, variantName) \ - \ -typedef struct { \ - baseType internal_never_use_directly; \ - int tag; \ -} variantName - - -typedef struct { - size_t cSize; - unsigned long long cSpeed; /* bytes / sec */ - unsigned long long dSpeed; - size_t cMem; /* memory usage during compression */ -} BMK_benchResult_t; - -VARIANT_ERROR_RESULT(BMK_benchResult_t, BMK_benchOutcome_t); - -/* check first if the return structure represents an error or a valid result */ -int BMK_isSuccessful_benchOutcome(BMK_benchOutcome_t outcome); - -/* extract result from variant type. - * note : this function will abort() program execution if result is not valid - * check result validity first, by using BMK_isSuccessful_benchOutcome() - */ -BMK_benchResult_t BMK_extract_benchResult(BMK_benchOutcome_t outcome); - - -/*! BMK_benchFiles() -- called by zstdcli */ -/* Loads files from fileNamesTable into memory, - * and an optional dictionary from dictFileName (can be NULL), - * then uses benchMem(). - * fileNamesTable - name of files to benchmark. - * nbFiles - number of files (size of fileNamesTable), must be > 0. - * dictFileName - name of dictionary file to load. - * cLevel - compression level to benchmark, errors if invalid. - * compressionParams - advanced compression Parameters. - * displayLevel - what gets printed: - * 0 : no display; - * 1 : errors; - * 2 : + result + interaction + warnings; - * 3 : + information; - * 4 : + debug - * @return: - * a variant, which expresses either an error, or a valid result. - * Use BMK_isSuccessful_benchOutcome() to check if function was successful. - * If yes, extract the valid result with BMK_extract_benchResult(), - * it will contain : - * .cSpeed: compression speed in bytes per second, - * .dSpeed: decompression speed in bytes per second, - * .cSize : compressed size, in bytes - * .cMem : memory budget required for the compression context - */ -BMK_benchOutcome_t BMK_benchFiles( - const char* const * fileNamesTable, unsigned nbFiles, - const char* dictFileName, - int cLevel, const ZSTD_compressionParameters* compressionParams, - int displayLevel); - - -typedef enum { - BMK_both = 0, - BMK_decodeOnly = 1, - BMK_compressOnly = 2 -} BMK_mode_t; - -typedef struct { - BMK_mode_t mode; /* 0: all, 1: compress only 2: decode only */ - unsigned nbSeconds; /* default timing is in nbSeconds */ - size_t blockSize; /* Maximum size of each block*/ - int nbWorkers; /* multithreading */ - unsigned realTime; /* real time priority */ - int additionalParam; /* used by python speed benchmark */ - int ldmFlag; /* enables long distance matching */ - int ldmMinMatch; /* below: parameters for long distance matching, see zstd.1.md */ - int ldmHashLog; - int ldmBucketSizeLog; - int ldmHashRateLog; +/* + * Copyright (c) Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + + /* benchzstd : + * benchmark Zstandard compression / decompression + * over a set of files or buffers + * and display progress result and final summary + */ + +#if defined (__cplusplus) +extern "C" { +#endif + +#ifndef BENCH_ZSTD_H_3242387 +#define BENCH_ZSTD_H_3242387 + +/* === Dependencies === */ +#include <stddef.h> /* size_t */ +#define ZSTD_STATIC_LINKING_ONLY /* ZSTD_compressionParameters */ +#include "../lib/zstd.h" /* ZSTD_compressionParameters */ + + +/* === Constants === */ + +#define MB_UNIT 1000000 + + +/* === Benchmark functions === */ + +/* Creates a variant `typeName`, able to express "error or valid result". + * Functions with return type `typeName` + * must first check if result is valid, using BMK_isSuccessful_*(), + * and only then can extract `baseType`. + */ +#define VARIANT_ERROR_RESULT(baseType, variantName) \ + \ +typedef struct { \ + baseType internal_never_use_directly; \ + int tag; \ +} variantName + + +typedef struct { + size_t cSize; + unsigned long long cSpeed; /* bytes / sec */ + unsigned long long dSpeed; + size_t cMem; /* memory usage during compression */ +} BMK_benchResult_t; + +VARIANT_ERROR_RESULT(BMK_benchResult_t, BMK_benchOutcome_t); + +/* check first if the return structure represents an error or a valid result */ +int BMK_isSuccessful_benchOutcome(BMK_benchOutcome_t outcome); + +/* extract result from variant type. + * note : this function will abort() program execution if result is not valid + * check result validity first, by using BMK_isSuccessful_benchOutcome() + */ +BMK_benchResult_t BMK_extract_benchResult(BMK_benchOutcome_t outcome); + + +/*! BMK_benchFiles() -- called by zstdcli */ +/* Loads files from fileNamesTable into memory, + * and an optional dictionary from dictFileName (can be NULL), + * then uses benchMem(). + * fileNamesTable - name of files to benchmark. + * nbFiles - number of files (size of fileNamesTable), must be > 0. + * dictFileName - name of dictionary file to load. + * cLevel - compression level to benchmark, errors if invalid. + * compressionParams - advanced compression Parameters. + * displayLevel - what gets printed: + * 0 : no display; + * 1 : errors; + * 2 : + result + interaction + warnings; + * 3 : + information; + * 4 : + debug + * @return: + * a variant, which expresses either an error, or a valid result. + * Use BMK_isSuccessful_benchOutcome() to check if function was successful. + * If yes, extract the valid result with BMK_extract_benchResult(), + * it will contain : + * .cSpeed: compression speed in bytes per second, + * .dSpeed: decompression speed in bytes per second, + * .cSize : compressed size, in bytes + * .cMem : memory budget required for the compression context + */ +BMK_benchOutcome_t BMK_benchFiles( + const char* const * fileNamesTable, unsigned nbFiles, + const char* dictFileName, + int cLevel, const ZSTD_compressionParameters* compressionParams, + int displayLevel); + + +typedef enum { + BMK_both = 0, + BMK_decodeOnly = 1, + BMK_compressOnly = 2 +} BMK_mode_t; + +typedef struct { + BMK_mode_t mode; /* 0: all, 1: compress only 2: decode only */ + unsigned nbSeconds; /* default timing is in nbSeconds */ + size_t blockSize; /* Maximum size of each block*/ + int nbWorkers; /* multithreading */ + unsigned realTime; /* real time priority */ + int additionalParam; /* used by python speed benchmark */ + int ldmFlag; /* enables long distance matching */ + int ldmMinMatch; /* below: parameters for long distance matching, see zstd.1.md */ + int ldmHashLog; + int ldmBucketSizeLog; + int ldmHashRateLog; ZSTD_paramSwitch_e literalCompressionMode; - int useRowMatchFinder; /* use row-based matchfinder if possible */ -} BMK_advancedParams_t; - -/* returns default parameters used by nonAdvanced functions */ -BMK_advancedParams_t BMK_initAdvancedParams(void); - -/*! BMK_benchFilesAdvanced(): - * Same as BMK_benchFiles(), - * with more controls, provided through advancedParams_t structure */ -BMK_benchOutcome_t BMK_benchFilesAdvanced( - const char* const * fileNamesTable, unsigned nbFiles, - const char* dictFileName, - int cLevel, const ZSTD_compressionParameters* compressionParams, - int displayLevel, const BMK_advancedParams_t* adv); - -/*! BMK_syntheticTest() -- called from zstdcli */ -/* Generates a sample with datagen, using compressibility argument */ -/* cLevel - compression level to benchmark, errors if invalid - * compressibility - determines compressibility of sample - * compressionParams - basic compression Parameters - * displayLevel - see benchFiles - * adv - see advanced_Params_t - * @return: - * a variant, which expresses either an error, or a valid result. - * Use BMK_isSuccessful_benchOutcome() to check if function was successful. - * If yes, extract the valid result with BMK_extract_benchResult(), - * it will contain : - * .cSpeed: compression speed in bytes per second, - * .dSpeed: decompression speed in bytes per second, - * .cSize : compressed size, in bytes - * .cMem : memory budget required for the compression context - */ -BMK_benchOutcome_t BMK_syntheticTest( - int cLevel, double compressibility, - const ZSTD_compressionParameters* compressionParams, - int displayLevel, const BMK_advancedParams_t* adv); - - - -/* === Benchmark Zstandard in a memory-to-memory scenario === */ - -/** BMK_benchMem() -- core benchmarking function, called in paramgrill - * applies ZSTD_compress_generic() and ZSTD_decompress_generic() on data in srcBuffer - * with specific compression parameters provided by other arguments using benchFunction - * (cLevel, comprParams + adv in advanced Mode) */ -/* srcBuffer - data source, expected to be valid compressed data if in Decode Only Mode - * srcSize - size of data in srcBuffer - * fileSizes - srcBuffer is considered cut into 1+ segments, to compress separately. - * note : sum(fileSizes) must be == srcSize. (<== ensure it's properly checked) - * nbFiles - nb of segments - * cLevel - compression level - * comprParams - basic compression parameters - * dictBuffer - a dictionary if used, null otherwise - * dictBufferSize - size of dictBuffer, 0 otherwise - * displayLevel - see BMK_benchFiles - * displayName - name used by display - * @return: - * a variant, which expresses either an error, or a valid result. - * Use BMK_isSuccessful_benchOutcome() to check if function was successful. - * If yes, extract the valid result with BMK_extract_benchResult(), - * it will contain : - * .cSpeed: compression speed in bytes per second, - * .dSpeed: decompression speed in bytes per second, - * .cSize : compressed size, in bytes - * .cMem : memory budget required for the compression context - */ -BMK_benchOutcome_t BMK_benchMem(const void* srcBuffer, size_t srcSize, - const size_t* fileSizes, unsigned nbFiles, - int cLevel, const ZSTD_compressionParameters* comprParams, - const void* dictBuffer, size_t dictBufferSize, - int displayLevel, const char* displayName); - - -/* BMK_benchMemAdvanced() : same as BMK_benchMem() - * with following additional options : - * dstBuffer - destination buffer to write compressed output in, NULL if none provided. - * dstCapacity - capacity of destination buffer, give 0 if dstBuffer = NULL - * adv = see advancedParams_t - */ -BMK_benchOutcome_t BMK_benchMemAdvanced(const void* srcBuffer, size_t srcSize, - void* dstBuffer, size_t dstCapacity, - const size_t* fileSizes, unsigned nbFiles, - int cLevel, const ZSTD_compressionParameters* comprParams, - const void* dictBuffer, size_t dictBufferSize, - int displayLevel, const char* displayName, - const BMK_advancedParams_t* adv); - - - -#endif /* BENCH_ZSTD_H_3242387 */ - -#if defined (__cplusplus) -} -#endif + int useRowMatchFinder; /* use row-based matchfinder if possible */ +} BMK_advancedParams_t; + +/* returns default parameters used by nonAdvanced functions */ +BMK_advancedParams_t BMK_initAdvancedParams(void); + +/*! BMK_benchFilesAdvanced(): + * Same as BMK_benchFiles(), + * with more controls, provided through advancedParams_t structure */ +BMK_benchOutcome_t BMK_benchFilesAdvanced( + const char* const * fileNamesTable, unsigned nbFiles, + const char* dictFileName, + int cLevel, const ZSTD_compressionParameters* compressionParams, + int displayLevel, const BMK_advancedParams_t* adv); + +/*! BMK_syntheticTest() -- called from zstdcli */ +/* Generates a sample with datagen, using compressibility argument */ +/* cLevel - compression level to benchmark, errors if invalid + * compressibility - determines compressibility of sample + * compressionParams - basic compression Parameters + * displayLevel - see benchFiles + * adv - see advanced_Params_t + * @return: + * a variant, which expresses either an error, or a valid result. + * Use BMK_isSuccessful_benchOutcome() to check if function was successful. + * If yes, extract the valid result with BMK_extract_benchResult(), + * it will contain : + * .cSpeed: compression speed in bytes per second, + * .dSpeed: decompression speed in bytes per second, + * .cSize : compressed size, in bytes + * .cMem : memory budget required for the compression context + */ +BMK_benchOutcome_t BMK_syntheticTest( + int cLevel, double compressibility, + const ZSTD_compressionParameters* compressionParams, + int displayLevel, const BMK_advancedParams_t* adv); + + + +/* === Benchmark Zstandard in a memory-to-memory scenario === */ + +/** BMK_benchMem() -- core benchmarking function, called in paramgrill + * applies ZSTD_compress_generic() and ZSTD_decompress_generic() on data in srcBuffer + * with specific compression parameters provided by other arguments using benchFunction + * (cLevel, comprParams + adv in advanced Mode) */ +/* srcBuffer - data source, expected to be valid compressed data if in Decode Only Mode + * srcSize - size of data in srcBuffer + * fileSizes - srcBuffer is considered cut into 1+ segments, to compress separately. + * note : sum(fileSizes) must be == srcSize. (<== ensure it's properly checked) + * nbFiles - nb of segments + * cLevel - compression level + * comprParams - basic compression parameters + * dictBuffer - a dictionary if used, null otherwise + * dictBufferSize - size of dictBuffer, 0 otherwise + * displayLevel - see BMK_benchFiles + * displayName - name used by display + * @return: + * a variant, which expresses either an error, or a valid result. + * Use BMK_isSuccessful_benchOutcome() to check if function was successful. + * If yes, extract the valid result with BMK_extract_benchResult(), + * it will contain : + * .cSpeed: compression speed in bytes per second, + * .dSpeed: decompression speed in bytes per second, + * .cSize : compressed size, in bytes + * .cMem : memory budget required for the compression context + */ +BMK_benchOutcome_t BMK_benchMem(const void* srcBuffer, size_t srcSize, + const size_t* fileSizes, unsigned nbFiles, + int cLevel, const ZSTD_compressionParameters* comprParams, + const void* dictBuffer, size_t dictBufferSize, + int displayLevel, const char* displayName); + + +/* BMK_benchMemAdvanced() : same as BMK_benchMem() + * with following additional options : + * dstBuffer - destination buffer to write compressed output in, NULL if none provided. + * dstCapacity - capacity of destination buffer, give 0 if dstBuffer = NULL + * adv = see advancedParams_t + */ +BMK_benchOutcome_t BMK_benchMemAdvanced(const void* srcBuffer, size_t srcSize, + void* dstBuffer, size_t dstCapacity, + const size_t* fileSizes, unsigned nbFiles, + int cLevel, const ZSTD_compressionParameters* comprParams, + const void* dictBuffer, size_t dictBufferSize, + int displayLevel, const char* displayName, + const BMK_advancedParams_t* adv); + + + +#endif /* BENCH_ZSTD_H_3242387 */ + +#if defined (__cplusplus) +} +#endif diff --git a/contrib/libs/zstd/programs/datagen.c b/contrib/libs/zstd/programs/datagen.c index 3b4f9e5c7b..21b4d036a8 100644 --- a/contrib/libs/zstd/programs/datagen.c +++ b/contrib/libs/zstd/programs/datagen.c @@ -1,186 +1,186 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - - -/*-************************************ -* Dependencies -**************************************/ -#include "datagen.h" -#include "platform.h" /* SET_BINARY_MODE */ -#include <stdlib.h> /* malloc, free */ -#include <stdio.h> /* FILE, fwrite, fprintf */ -#include <string.h> /* memcpy */ -#include "../lib/common/mem.h" /* U32 */ - - -/*-************************************ -* Macros -**************************************/ -#define KB *(1 <<10) -#define MIN(a,b) ( (a) < (b) ? (a) : (b) ) - -#define RDG_DEBUG 0 -#define TRACE(...) if (RDG_DEBUG) fprintf(stderr, __VA_ARGS__ ) - - -/*-************************************ -* Local constants -**************************************/ -#define LTLOG 13 -#define LTSIZE (1<<LTLOG) -#define LTMASK (LTSIZE-1) - - -/*-******************************************************* -* Local Functions -*********************************************************/ -#define RDG_rotl32(x,r) ((x << r) | (x >> (32 - r))) -static U32 RDG_rand(U32* src) -{ - static const U32 prime1 = 2654435761U; - static const U32 prime2 = 2246822519U; - U32 rand32 = *src; - rand32 *= prime1; - rand32 ^= prime2; - rand32 = RDG_rotl32(rand32, 13); - *src = rand32; - return rand32 >> 5; -} - -typedef U32 fixedPoint_24_8; - -static void RDG_fillLiteralDistrib(BYTE* ldt, fixedPoint_24_8 ld) -{ - BYTE const firstChar = (ld<=0.0) ? 0 : '('; - BYTE const lastChar = (ld<=0.0) ? 255 : '}'; - BYTE character = (ld<=0.0) ? 0 : '0'; - U32 u; - - if (ld<=0) ld = 0; - for (u=0; u<LTSIZE; ) { - U32 const weight = (((LTSIZE - u) * ld) >> 8) + 1; - U32 const end = MIN ( u + weight , LTSIZE); - while (u < end) ldt[u++] = character; - character++; - if (character > lastChar) character = firstChar; - } -} - - -static BYTE RDG_genChar(U32* seed, const BYTE* ldt) -{ - U32 const id = RDG_rand(seed) & LTMASK; - return ldt[id]; /* memory-sanitizer fails here, stating "uninitialized value" when table initialized with P==0.0. Checked : table is fully initialized */ -} - - -static U32 RDG_rand15Bits (U32* seedPtr) -{ - return RDG_rand(seedPtr) & 0x7FFF; -} - -static U32 RDG_randLength(U32* seedPtr) -{ - if (RDG_rand(seedPtr) & 7) return (RDG_rand(seedPtr) & 0xF); /* small length */ - return (RDG_rand(seedPtr) & 0x1FF) + 0xF; -} - -static void RDG_genBlock(void* buffer, size_t buffSize, size_t prefixSize, - double matchProba, const BYTE* ldt, U32* seedPtr) -{ - BYTE* const buffPtr = (BYTE*)buffer; - U32 const matchProba32 = (U32)(32768 * matchProba); - size_t pos = prefixSize; - U32 prevOffset = 1; - - /* special case : sparse content */ - while (matchProba >= 1.0) { - size_t size0 = RDG_rand(seedPtr) & 3; - size0 = (size_t)1 << (16 + size0 * 2); - size0 += RDG_rand(seedPtr) & (size0-1); /* because size0 is power of 2*/ - if (buffSize < pos + size0) { - memset(buffPtr+pos, 0, buffSize-pos); - return; - } - memset(buffPtr+pos, 0, size0); - pos += size0; - buffPtr[pos-1] = RDG_genChar(seedPtr, ldt); - continue; - } - - /* init */ - if (pos==0) buffPtr[0] = RDG_genChar(seedPtr, ldt), pos=1; - - /* Generate compressible data */ - while (pos < buffSize) { - /* Select : Literal (char) or Match (within 32K) */ - if (RDG_rand15Bits(seedPtr) < matchProba32) { - /* Copy (within 32K) */ - U32 const length = RDG_randLength(seedPtr) + 4; - U32 const d = (U32) MIN(pos + length , buffSize); - U32 const repeatOffset = (RDG_rand(seedPtr) & 15) == 2; - U32 const randOffset = RDG_rand15Bits(seedPtr) + 1; - U32 const offset = repeatOffset ? prevOffset : (U32) MIN(randOffset , pos); - size_t match = pos - offset; - while (pos < d) { buffPtr[pos++] = buffPtr[match++]; /* correctly manages overlaps */ } - prevOffset = offset; - } else { - /* Literal (noise) */ - U32 const length = RDG_randLength(seedPtr); - U32 const d = (U32) MIN(pos + length, buffSize); - while (pos < d) { buffPtr[pos++] = RDG_genChar(seedPtr, ldt); } - } } -} - - -void RDG_genBuffer(void* buffer, size_t size, double matchProba, double litProba, unsigned seed) -{ - U32 seed32 = seed; - BYTE ldt[LTSIZE]; - memset(ldt, '0', sizeof(ldt)); /* yes, character '0', this is intentional */ - if (litProba<=0.0) litProba = matchProba / 4.5; - RDG_fillLiteralDistrib(ldt, (fixedPoint_24_8)(litProba * 256 + 0.001)); - RDG_genBlock(buffer, size, 0, matchProba, ldt, &seed32); -} - - -void RDG_genStdout(unsigned long long size, double matchProba, double litProba, unsigned seed) -{ - U32 seed32 = seed; - size_t const stdBlockSize = 128 KB; - size_t const stdDictSize = 32 KB; - BYTE* const buff = (BYTE*)malloc(stdDictSize + stdBlockSize); - U64 total = 0; - BYTE ldt[LTSIZE]; /* literals distribution table */ - - /* init */ - if (buff==NULL) { perror("datagen"); exit(1); } - if (litProba<=0.0) litProba = matchProba / 4.5; - memset(ldt, '0', sizeof(ldt)); /* yes, character '0', this is intentional */ - RDG_fillLiteralDistrib(ldt, (fixedPoint_24_8)(litProba * 256 + 0.001)); - SET_BINARY_MODE(stdout); - - /* Generate initial dict */ - RDG_genBlock(buff, stdDictSize, 0, matchProba, ldt, &seed32); - - /* Generate compressible data */ - while (total < size) { - size_t const genBlockSize = (size_t) (MIN (stdBlockSize, size-total)); - RDG_genBlock(buff, stdDictSize+stdBlockSize, stdDictSize, matchProba, ldt, &seed32); - total += genBlockSize; - { size_t const unused = fwrite(buff, 1, genBlockSize, stdout); (void)unused; } - /* update dict */ - memcpy(buff, buff + stdBlockSize, stdDictSize); - } - - /* cleanup */ - free(buff); -} +/* + * Copyright (c) Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + + + +/*-************************************ +* Dependencies +**************************************/ +#include "datagen.h" +#include "platform.h" /* SET_BINARY_MODE */ +#include <stdlib.h> /* malloc, free */ +#include <stdio.h> /* FILE, fwrite, fprintf */ +#include <string.h> /* memcpy */ +#include "../lib/common/mem.h" /* U32 */ + + +/*-************************************ +* Macros +**************************************/ +#define KB *(1 <<10) +#define MIN(a,b) ( (a) < (b) ? (a) : (b) ) + +#define RDG_DEBUG 0 +#define TRACE(...) if (RDG_DEBUG) fprintf(stderr, __VA_ARGS__ ) + + +/*-************************************ +* Local constants +**************************************/ +#define LTLOG 13 +#define LTSIZE (1<<LTLOG) +#define LTMASK (LTSIZE-1) + + +/*-******************************************************* +* Local Functions +*********************************************************/ +#define RDG_rotl32(x,r) ((x << r) | (x >> (32 - r))) +static U32 RDG_rand(U32* src) +{ + static const U32 prime1 = 2654435761U; + static const U32 prime2 = 2246822519U; + U32 rand32 = *src; + rand32 *= prime1; + rand32 ^= prime2; + rand32 = RDG_rotl32(rand32, 13); + *src = rand32; + return rand32 >> 5; +} + +typedef U32 fixedPoint_24_8; + +static void RDG_fillLiteralDistrib(BYTE* ldt, fixedPoint_24_8 ld) +{ + BYTE const firstChar = (ld<=0.0) ? 0 : '('; + BYTE const lastChar = (ld<=0.0) ? 255 : '}'; + BYTE character = (ld<=0.0) ? 0 : '0'; + U32 u; + + if (ld<=0) ld = 0; + for (u=0; u<LTSIZE; ) { + U32 const weight = (((LTSIZE - u) * ld) >> 8) + 1; + U32 const end = MIN ( u + weight , LTSIZE); + while (u < end) ldt[u++] = character; + character++; + if (character > lastChar) character = firstChar; + } +} + + +static BYTE RDG_genChar(U32* seed, const BYTE* ldt) +{ + U32 const id = RDG_rand(seed) & LTMASK; + return ldt[id]; /* memory-sanitizer fails here, stating "uninitialized value" when table initialized with P==0.0. Checked : table is fully initialized */ +} + + +static U32 RDG_rand15Bits (U32* seedPtr) +{ + return RDG_rand(seedPtr) & 0x7FFF; +} + +static U32 RDG_randLength(U32* seedPtr) +{ + if (RDG_rand(seedPtr) & 7) return (RDG_rand(seedPtr) & 0xF); /* small length */ + return (RDG_rand(seedPtr) & 0x1FF) + 0xF; +} + +static void RDG_genBlock(void* buffer, size_t buffSize, size_t prefixSize, + double matchProba, const BYTE* ldt, U32* seedPtr) +{ + BYTE* const buffPtr = (BYTE*)buffer; + U32 const matchProba32 = (U32)(32768 * matchProba); + size_t pos = prefixSize; + U32 prevOffset = 1; + + /* special case : sparse content */ + while (matchProba >= 1.0) { + size_t size0 = RDG_rand(seedPtr) & 3; + size0 = (size_t)1 << (16 + size0 * 2); + size0 += RDG_rand(seedPtr) & (size0-1); /* because size0 is power of 2*/ + if (buffSize < pos + size0) { + memset(buffPtr+pos, 0, buffSize-pos); + return; + } + memset(buffPtr+pos, 0, size0); + pos += size0; + buffPtr[pos-1] = RDG_genChar(seedPtr, ldt); + continue; + } + + /* init */ + if (pos==0) buffPtr[0] = RDG_genChar(seedPtr, ldt), pos=1; + + /* Generate compressible data */ + while (pos < buffSize) { + /* Select : Literal (char) or Match (within 32K) */ + if (RDG_rand15Bits(seedPtr) < matchProba32) { + /* Copy (within 32K) */ + U32 const length = RDG_randLength(seedPtr) + 4; + U32 const d = (U32) MIN(pos + length , buffSize); + U32 const repeatOffset = (RDG_rand(seedPtr) & 15) == 2; + U32 const randOffset = RDG_rand15Bits(seedPtr) + 1; + U32 const offset = repeatOffset ? prevOffset : (U32) MIN(randOffset , pos); + size_t match = pos - offset; + while (pos < d) { buffPtr[pos++] = buffPtr[match++]; /* correctly manages overlaps */ } + prevOffset = offset; + } else { + /* Literal (noise) */ + U32 const length = RDG_randLength(seedPtr); + U32 const d = (U32) MIN(pos + length, buffSize); + while (pos < d) { buffPtr[pos++] = RDG_genChar(seedPtr, ldt); } + } } +} + + +void RDG_genBuffer(void* buffer, size_t size, double matchProba, double litProba, unsigned seed) +{ + U32 seed32 = seed; + BYTE ldt[LTSIZE]; + memset(ldt, '0', sizeof(ldt)); /* yes, character '0', this is intentional */ + if (litProba<=0.0) litProba = matchProba / 4.5; + RDG_fillLiteralDistrib(ldt, (fixedPoint_24_8)(litProba * 256 + 0.001)); + RDG_genBlock(buffer, size, 0, matchProba, ldt, &seed32); +} + + +void RDG_genStdout(unsigned long long size, double matchProba, double litProba, unsigned seed) +{ + U32 seed32 = seed; + size_t const stdBlockSize = 128 KB; + size_t const stdDictSize = 32 KB; + BYTE* const buff = (BYTE*)malloc(stdDictSize + stdBlockSize); + U64 total = 0; + BYTE ldt[LTSIZE]; /* literals distribution table */ + + /* init */ + if (buff==NULL) { perror("datagen"); exit(1); } + if (litProba<=0.0) litProba = matchProba / 4.5; + memset(ldt, '0', sizeof(ldt)); /* yes, character '0', this is intentional */ + RDG_fillLiteralDistrib(ldt, (fixedPoint_24_8)(litProba * 256 + 0.001)); + SET_BINARY_MODE(stdout); + + /* Generate initial dict */ + RDG_genBlock(buff, stdDictSize, 0, matchProba, ldt, &seed32); + + /* Generate compressible data */ + while (total < size) { + size_t const genBlockSize = (size_t) (MIN (stdBlockSize, size-total)); + RDG_genBlock(buff, stdDictSize+stdBlockSize, stdDictSize, matchProba, ldt, &seed32); + total += genBlockSize; + { size_t const unused = fwrite(buff, 1, genBlockSize, stdout); (void)unused; } + /* update dict */ + memcpy(buff, buff + stdBlockSize, stdDictSize); + } + + /* cleanup */ + free(buff); +} diff --git a/contrib/libs/zstd/programs/datagen.h b/contrib/libs/zstd/programs/datagen.h index b76ae2a222..ebb2b10c3a 100644 --- a/contrib/libs/zstd/programs/datagen.h +++ b/contrib/libs/zstd/programs/datagen.h @@ -1,30 +1,30 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - -#ifndef DATAGEN_H -#define DATAGEN_H - -#include <stddef.h> /* size_t */ - -void RDG_genStdout(unsigned long long size, double matchProba, double litProba, unsigned seed); -void RDG_genBuffer(void* buffer, size_t size, double matchProba, double litProba, unsigned seed); -/*!RDG_genBuffer - Generate 'size' bytes of compressible data into 'buffer'. - Compressibility can be controlled using 'matchProba', which is floating point value between 0 and 1. - 'LitProba' is optional, it affect variability of individual bytes. If litProba==0.0, default value will be used. - Generated data pattern can be modified using different 'seed'. - For a triplet (matchProba, litProba, seed), the function always generate the same content. - - RDG_genStdout - Same as RDG_genBuffer, but generates data into stdout -*/ - -#endif +/* + * Copyright (c) Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + + +#ifndef DATAGEN_H +#define DATAGEN_H + +#include <stddef.h> /* size_t */ + +void RDG_genStdout(unsigned long long size, double matchProba, double litProba, unsigned seed); +void RDG_genBuffer(void* buffer, size_t size, double matchProba, double litProba, unsigned seed); +/*!RDG_genBuffer + Generate 'size' bytes of compressible data into 'buffer'. + Compressibility can be controlled using 'matchProba', which is floating point value between 0 and 1. + 'LitProba' is optional, it affect variability of individual bytes. If litProba==0.0, default value will be used. + Generated data pattern can be modified using different 'seed'. + For a triplet (matchProba, litProba, seed), the function always generate the same content. + + RDG_genStdout + Same as RDG_genBuffer, but generates data into stdout +*/ + +#endif diff --git a/contrib/libs/zstd/programs/dibio.c b/contrib/libs/zstd/programs/dibio.c index d19f954486..32a51e1704 100644 --- a/contrib/libs/zstd/programs/dibio.c +++ b/contrib/libs/zstd/programs/dibio.c @@ -1,93 +1,93 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - - -/* ************************************** -* Compiler Warnings -****************************************/ -#ifdef _MSC_VER -# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */ -#endif - - -/*-************************************* -* Includes -***************************************/ -#include "platform.h" /* Large Files support */ -#include "util.h" /* UTIL_getFileSize, UTIL_getTotalFileSize */ -#include <stdlib.h> /* malloc, free */ -#include <string.h> /* memset */ -#include <stdio.h> /* fprintf, fopen, ftello64 */ -#include <errno.h> /* errno */ -#include <assert.h> - -#include "timefn.h" /* UTIL_time_t, UTIL_clockSpanMicro, UTIL_getTime */ -#include "../lib/common/mem.h" /* read */ -#include "dibio.h" - - -/*-************************************* -* Constants -***************************************/ -#define KB *(1 <<10) -#define MB *(1 <<20) -#define GB *(1U<<30) - -#define SAMPLESIZE_MAX (128 KB) -#define MEMMULT 11 /* rough estimation : memory cost to analyze 1 byte of sample */ -#define COVER_MEMMULT 9 /* rough estimation : memory cost to analyze 1 byte of sample */ -#define FASTCOVER_MEMMULT 1 /* rough estimation : memory cost to analyze 1 byte of sample */ -static const size_t g_maxMemory = (sizeof(size_t) == 4) ? (2 GB - 64 MB) : ((size_t)(512 MB) << sizeof(size_t)); - -#define NOISELENGTH 32 +/* + * Copyright (c) Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + + + +/* ************************************** +* Compiler Warnings +****************************************/ +#ifdef _MSC_VER +# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */ +#endif + + +/*-************************************* +* Includes +***************************************/ +#include "platform.h" /* Large Files support */ +#include "util.h" /* UTIL_getFileSize, UTIL_getTotalFileSize */ +#include <stdlib.h> /* malloc, free */ +#include <string.h> /* memset */ +#include <stdio.h> /* fprintf, fopen, ftello64 */ +#include <errno.h> /* errno */ +#include <assert.h> + +#include "timefn.h" /* UTIL_time_t, UTIL_clockSpanMicro, UTIL_getTime */ +#include "../lib/common/mem.h" /* read */ +#include "dibio.h" + + +/*-************************************* +* Constants +***************************************/ +#define KB *(1 <<10) +#define MB *(1 <<20) +#define GB *(1U<<30) + +#define SAMPLESIZE_MAX (128 KB) +#define MEMMULT 11 /* rough estimation : memory cost to analyze 1 byte of sample */ +#define COVER_MEMMULT 9 /* rough estimation : memory cost to analyze 1 byte of sample */ +#define FASTCOVER_MEMMULT 1 /* rough estimation : memory cost to analyze 1 byte of sample */ +static const size_t g_maxMemory = (sizeof(size_t) == 4) ? (2 GB - 64 MB) : ((size_t)(512 MB) << sizeof(size_t)); + +#define NOISELENGTH 32 #define MAX_SAMPLES_SIZE (2 GB) /* training dataset limited to 2GB */ - - -/*-************************************* -* Console display -***************************************/ -#define DISPLAY(...) fprintf(stderr, __VA_ARGS__) -#define DISPLAYLEVEL(l, ...) if (displayLevel>=l) { DISPLAY(__VA_ARGS__); } - -static const U64 g_refreshRate = SEC_TO_MICRO / 6; -static UTIL_time_t g_displayClock = UTIL_TIME_INITIALIZER; - -#define DISPLAYUPDATE(l, ...) { if (displayLevel>=l) { \ - if ((UTIL_clockSpanMicro(g_displayClock) > g_refreshRate) || (displayLevel>=4)) \ - { g_displayClock = UTIL_getTime(); DISPLAY(__VA_ARGS__); \ - if (displayLevel>=4) fflush(stderr); } } } - -/*-************************************* -* Exceptions -***************************************/ -#ifndef DEBUG -# define DEBUG 0 -#endif -#define DEBUGOUTPUT(...) if (DEBUG) DISPLAY(__VA_ARGS__); -#define EXM_THROW(error, ...) \ -{ \ - DEBUGOUTPUT("Error defined at %s, line %i : \n", __FILE__, __LINE__); \ - DISPLAY("Error %i : ", error); \ - DISPLAY(__VA_ARGS__); \ - DISPLAY("\n"); \ - exit(error); \ -} - - -/* ******************************************************** -* Helper functions -**********************************************************/ -#undef MIN -#define MIN(a,b) ((a) < (b) ? (a) : (b)) - + + +/*-************************************* +* Console display +***************************************/ +#define DISPLAY(...) fprintf(stderr, __VA_ARGS__) +#define DISPLAYLEVEL(l, ...) if (displayLevel>=l) { DISPLAY(__VA_ARGS__); } + +static const U64 g_refreshRate = SEC_TO_MICRO / 6; +static UTIL_time_t g_displayClock = UTIL_TIME_INITIALIZER; + +#define DISPLAYUPDATE(l, ...) { if (displayLevel>=l) { \ + if ((UTIL_clockSpanMicro(g_displayClock) > g_refreshRate) || (displayLevel>=4)) \ + { g_displayClock = UTIL_getTime(); DISPLAY(__VA_ARGS__); \ + if (displayLevel>=4) fflush(stderr); } } } + +/*-************************************* +* Exceptions +***************************************/ +#ifndef DEBUG +# define DEBUG 0 +#endif +#define DEBUGOUTPUT(...) if (DEBUG) DISPLAY(__VA_ARGS__); +#define EXM_THROW(error, ...) \ +{ \ + DEBUGOUTPUT("Error defined at %s, line %i : \n", __FILE__, __LINE__); \ + DISPLAY("Error %i : ", error); \ + DISPLAY(__VA_ARGS__); \ + DISPLAY("\n"); \ + exit(error); \ +} + + +/* ******************************************************** +* Helper functions +**********************************************************/ +#undef MIN +#define MIN(a,b) ((a) < (b) ? (a) : (b)) + /** Returns the size of a file. If error returns -1. @@ -97,31 +97,31 @@ static S64 DiB_getFileSize (const char * fileName) U64 const fileSize = UTIL_getFileSize(fileName); return (fileSize == UTIL_FILESIZE_UNKNOWN) ? -1 : (S64)fileSize; } - -/* ******************************************************** -* File related operations -**********************************************************/ -/** DiB_loadFiles() : - * load samples from files listed in fileNamesTable into buffer. - * works even if buffer is too small to load all samples. - * Also provides the size of each sample into sampleSizes table - * which must be sized correctly, using DiB_fileStats(). - * @return : nb of samples effectively loaded into `buffer` - * *bufferSizePtr is modified, it provides the amount data loaded within buffer. - * sampleSizes is filled with the size of each sample. - */ + +/* ******************************************************** +* File related operations +**********************************************************/ +/** DiB_loadFiles() : + * load samples from files listed in fileNamesTable into buffer. + * works even if buffer is too small to load all samples. + * Also provides the size of each sample into sampleSizes table + * which must be sized correctly, using DiB_fileStats(). + * @return : nb of samples effectively loaded into `buffer` + * *bufferSizePtr is modified, it provides the amount data loaded within buffer. + * sampleSizes is filled with the size of each sample. + */ static int DiB_loadFiles( void* buffer, size_t* bufferSizePtr, size_t* sampleSizes, int sstSize, const char** fileNamesTable, int nbFiles, size_t targetChunkSize, int displayLevel ) -{ - char* const buff = (char*)buffer; +{ + char* const buff = (char*)buffer; size_t totalDataLoaded = 0; int nbSamplesLoaded = 0; int fileIndex = 0; FILE * f = NULL; - + assert(targetChunkSize <= SAMPLESIZE_MAX); while ( nbSamplesLoaded < sstSize && fileIndex < nbFiles ) { @@ -151,7 +151,7 @@ static int DiB_loadFiles( while( (S64)fileDataLoaded < fileSize && nbSamplesLoaded < sstSize ) { size_t const chunkSize = MIN((size_t)(fileSize-fileDataLoaded), targetChunkSize); if (totalDataLoaded + chunkSize > *bufferSizePtr) /* buffer is full */ - break; + break; if (fread( buff+totalDataLoaded, 1, chunkSize, f ) != chunkSize) EXM_THROW(11, "Pb reading %s", fileNamesTable[fileIndex]); @@ -164,115 +164,115 @@ static int DiB_loadFiles( fclose(f); f = NULL; } if (f != NULL) - fclose(f); + fclose(f); - DISPLAYLEVEL(2, "\r%79s\r", ""); + DISPLAYLEVEL(2, "\r%79s\r", ""); DISPLAYLEVEL(4, "Loaded %d KB total training data, %d nb samples \n", (int)(totalDataLoaded / (1 KB)), nbSamplesLoaded ); *bufferSizePtr = totalDataLoaded; return nbSamplesLoaded; -} - -#define DiB_rotl32(x,r) ((x << r) | (x >> (32 - r))) -static U32 DiB_rand(U32* src) -{ - static const U32 prime1 = 2654435761U; - static const U32 prime2 = 2246822519U; - U32 rand32 = *src; - rand32 *= prime1; - rand32 ^= prime2; - rand32 = DiB_rotl32(rand32, 13); - *src = rand32; - return rand32 >> 5; -} - -/* DiB_shuffle() : - * shuffle a table of file names in a semi-random way - * It improves dictionary quality by reducing "locality" impact, so if sample set is very large, - * it will load random elements from it, instead of just the first ones. */ -static void DiB_shuffle(const char** fileNamesTable, unsigned nbFiles) { - U32 seed = 0xFD2FB528; - unsigned i; - assert(nbFiles >= 1); - for (i = nbFiles - 1; i > 0; --i) { - unsigned const j = DiB_rand(&seed) % (i + 1); - const char* const tmp = fileNamesTable[j]; - fileNamesTable[j] = fileNamesTable[i]; - fileNamesTable[i] = tmp; - } -} - - -/*-******************************************************** -* Dictionary training functions -**********************************************************/ -static size_t DiB_findMaxMem(unsigned long long requiredMem) -{ - size_t const step = 8 MB; - void* testmem = NULL; - - requiredMem = (((requiredMem >> 23) + 1) << 23); - requiredMem += step; - if (requiredMem > g_maxMemory) requiredMem = g_maxMemory; - - while (!testmem) { - testmem = malloc((size_t)requiredMem); - requiredMem -= step; - } - - free(testmem); - return (size_t)requiredMem; -} - - -static void DiB_fillNoise(void* buffer, size_t length) -{ - unsigned const prime1 = 2654435761U; - unsigned const prime2 = 2246822519U; - unsigned acc = prime1; - size_t p=0; - - for (p=0; p<length; p++) { - acc *= prime2; - ((unsigned char*)buffer)[p] = (unsigned char)(acc >> 21); - } -} - - -static void DiB_saveDict(const char* dictFileName, - const void* buff, size_t buffSize) -{ - FILE* const f = fopen(dictFileName, "wb"); - if (f==NULL) EXM_THROW(3, "cannot open %s ", dictFileName); - - { size_t const n = fwrite(buff, 1, buffSize, f); - if (n!=buffSize) EXM_THROW(4, "%s : write error", dictFileName) } - - { size_t const n = (size_t)fclose(f); - if (n!=0) EXM_THROW(5, "%s : flush error", dictFileName) } -} - -typedef struct { +} + +#define DiB_rotl32(x,r) ((x << r) | (x >> (32 - r))) +static U32 DiB_rand(U32* src) +{ + static const U32 prime1 = 2654435761U; + static const U32 prime2 = 2246822519U; + U32 rand32 = *src; + rand32 *= prime1; + rand32 ^= prime2; + rand32 = DiB_rotl32(rand32, 13); + *src = rand32; + return rand32 >> 5; +} + +/* DiB_shuffle() : + * shuffle a table of file names in a semi-random way + * It improves dictionary quality by reducing "locality" impact, so if sample set is very large, + * it will load random elements from it, instead of just the first ones. */ +static void DiB_shuffle(const char** fileNamesTable, unsigned nbFiles) { + U32 seed = 0xFD2FB528; + unsigned i; + assert(nbFiles >= 1); + for (i = nbFiles - 1; i > 0; --i) { + unsigned const j = DiB_rand(&seed) % (i + 1); + const char* const tmp = fileNamesTable[j]; + fileNamesTable[j] = fileNamesTable[i]; + fileNamesTable[i] = tmp; + } +} + + +/*-******************************************************** +* Dictionary training functions +**********************************************************/ +static size_t DiB_findMaxMem(unsigned long long requiredMem) +{ + size_t const step = 8 MB; + void* testmem = NULL; + + requiredMem = (((requiredMem >> 23) + 1) << 23); + requiredMem += step; + if (requiredMem > g_maxMemory) requiredMem = g_maxMemory; + + while (!testmem) { + testmem = malloc((size_t)requiredMem); + requiredMem -= step; + } + + free(testmem); + return (size_t)requiredMem; +} + + +static void DiB_fillNoise(void* buffer, size_t length) +{ + unsigned const prime1 = 2654435761U; + unsigned const prime2 = 2246822519U; + unsigned acc = prime1; + size_t p=0; + + for (p=0; p<length; p++) { + acc *= prime2; + ((unsigned char*)buffer)[p] = (unsigned char)(acc >> 21); + } +} + + +static void DiB_saveDict(const char* dictFileName, + const void* buff, size_t buffSize) +{ + FILE* const f = fopen(dictFileName, "wb"); + if (f==NULL) EXM_THROW(3, "cannot open %s ", dictFileName); + + { size_t const n = fwrite(buff, 1, buffSize, f); + if (n!=buffSize) EXM_THROW(4, "%s : write error", dictFileName) } + + { size_t const n = (size_t)fclose(f); + if (n!=0) EXM_THROW(5, "%s : flush error", dictFileName) } +} + +typedef struct { S64 totalSizeToLoad; int nbSamples; int oneSampleTooLarge; -} fileStats; - -/*! DiB_fileStats() : - * Given a list of files, and a chunkSize (0 == no chunk, whole files) - * provides the amount of data to be loaded and the resulting nb of samples. - * This is useful primarily for allocation purpose => sample buffer, and sample sizes table. - */ +} fileStats; + +/*! DiB_fileStats() : + * Given a list of files, and a chunkSize (0 == no chunk, whole files) + * provides the amount of data to be loaded and the resulting nb of samples. + * This is useful primarily for allocation purpose => sample buffer, and sample sizes table. + */ static fileStats DiB_fileStats(const char** fileNamesTable, int nbFiles, size_t chunkSize, int displayLevel) -{ - fileStats fs; +{ + fileStats fs; int n; - memset(&fs, 0, sizeof(fs)); + memset(&fs, 0, sizeof(fs)); // We assume that if chunking is requested, the chunk size is < SAMPLESIZE_MAX assert( chunkSize <= SAMPLESIZE_MAX ); - for (n=0; n<nbFiles; n++) { + for (n=0; n<nbFiles; n++) { S64 const fileSize = DiB_getFileSize(fileNamesTable[n]); // TODO: is there a minimum sample size? What if the file is 1-byte? if (fileSize == 0) { @@ -300,24 +300,24 @@ static fileStats DiB_fileStats(const char** fileNamesTable, int nbFiles, size_t fs.nbSamples += 1; fs.totalSizeToLoad += MIN(fileSize, SAMPLESIZE_MAX); } - } + } DISPLAYLEVEL(4, "Found training data %d files, %d KB, %d samples\n", nbFiles, (int)(fs.totalSizeToLoad / (1 KB)), fs.nbSamples); - return fs; -} - + return fs; +} + int DiB_trainFromFiles(const char* dictFileName, size_t maxDictSize, const char** fileNamesTable, int nbFiles, size_t chunkSize, - ZDICT_legacy_params_t* params, ZDICT_cover_params_t* coverParams, + ZDICT_legacy_params_t* params, ZDICT_cover_params_t* coverParams, ZDICT_fastCover_params_t* fastCoverParams, int optimize, unsigned memLimit) -{ +{ fileStats fs; size_t* sampleSizes; /* vector of sample sizes. Each sample can be up to SAMPLESIZE_MAX */ int nbSamplesLoaded; /* nb of samples effectively loaded in srcBuffer */ size_t loadedSize; /* total data loaded in srcBuffer for all samples */ void* srcBuffer /* contiguous buffer with training data/samples */; - void* const dictBuffer = malloc(maxDictSize); - int result = 0; - + void* const dictBuffer = malloc(maxDictSize); + int result = 0; + int const displayLevel = params ? params->zParams.notificationLevel : coverParams ? coverParams->zParams.notificationLevel : fastCoverParams ? fastCoverParams->zParams.notificationLevel : 0; @@ -349,87 +349,87 @@ int DiB_trainFromFiles(const char* dictFileName, size_t maxDictSize, sampleSizes = (size_t*)malloc(fs.nbSamples * sizeof(size_t)); } - /* Checks */ - if ((!sampleSizes) || (!srcBuffer) || (!dictBuffer)) - EXM_THROW(12, "not enough memory for DiB_trainFiles"); /* should not happen */ - if (fs.oneSampleTooLarge) { - DISPLAYLEVEL(2, "! Warning : some sample(s) are very large \n"); - DISPLAYLEVEL(2, "! Note that dictionary is only useful for small samples. \n"); - DISPLAYLEVEL(2, "! As a consequence, only the first %u bytes of each sample are loaded \n", SAMPLESIZE_MAX); - } - if (fs.nbSamples < 5) { - DISPLAYLEVEL(2, "! Warning : nb of samples too low for proper processing ! \n"); - DISPLAYLEVEL(2, "! Please provide _one file per sample_. \n"); - DISPLAYLEVEL(2, "! Alternatively, split files into fixed-size blocks representative of samples, with -B# \n"); - EXM_THROW(14, "nb of samples too low"); /* we now clearly forbid this case */ - } + /* Checks */ + if ((!sampleSizes) || (!srcBuffer) || (!dictBuffer)) + EXM_THROW(12, "not enough memory for DiB_trainFiles"); /* should not happen */ + if (fs.oneSampleTooLarge) { + DISPLAYLEVEL(2, "! Warning : some sample(s) are very large \n"); + DISPLAYLEVEL(2, "! Note that dictionary is only useful for small samples. \n"); + DISPLAYLEVEL(2, "! As a consequence, only the first %u bytes of each sample are loaded \n", SAMPLESIZE_MAX); + } + if (fs.nbSamples < 5) { + DISPLAYLEVEL(2, "! Warning : nb of samples too low for proper processing ! \n"); + DISPLAYLEVEL(2, "! Please provide _one file per sample_. \n"); + DISPLAYLEVEL(2, "! Alternatively, split files into fixed-size blocks representative of samples, with -B# \n"); + EXM_THROW(14, "nb of samples too low"); /* we now clearly forbid this case */ + } if (fs.totalSizeToLoad < (S64)maxDictSize * 8) { - DISPLAYLEVEL(2, "! Warning : data size of samples too small for target dictionary size \n"); - DISPLAYLEVEL(2, "! Samples should be about 100x larger than target dictionary size \n"); - } - - /* init */ + DISPLAYLEVEL(2, "! Warning : data size of samples too small for target dictionary size \n"); + DISPLAYLEVEL(2, "! Samples should be about 100x larger than target dictionary size \n"); + } + + /* init */ if ((S64)loadedSize < fs.totalSizeToLoad) DISPLAYLEVEL(1, "Training samples set too large (%u MB); training on %u MB only...\n", (unsigned)(fs.totalSizeToLoad / (1 MB)), (unsigned)(loadedSize / (1 MB))); - - /* Load input buffer */ + + /* Load input buffer */ nbSamplesLoaded = DiB_loadFiles( srcBuffer, &loadedSize, sampleSizes, fs.nbSamples, fileNamesTable, nbFiles, chunkSize, displayLevel); - - { size_t dictSize; - if (params) { - DiB_fillNoise((char*)srcBuffer + loadedSize, NOISELENGTH); /* guard band, for end of buffer condition */ - dictSize = ZDICT_trainFromBuffer_legacy(dictBuffer, maxDictSize, + + { size_t dictSize; + if (params) { + DiB_fillNoise((char*)srcBuffer + loadedSize, NOISELENGTH); /* guard band, for end of buffer condition */ + dictSize = ZDICT_trainFromBuffer_legacy(dictBuffer, maxDictSize, srcBuffer, sampleSizes, nbSamplesLoaded, - *params); - } else if (coverParams) { - if (optimize) { - dictSize = ZDICT_optimizeTrainFromBuffer_cover(dictBuffer, maxDictSize, + *params); + } else if (coverParams) { + if (optimize) { + dictSize = ZDICT_optimizeTrainFromBuffer_cover(dictBuffer, maxDictSize, srcBuffer, sampleSizes, nbSamplesLoaded, - coverParams); - if (!ZDICT_isError(dictSize)) { - unsigned splitPercentage = (unsigned)(coverParams->splitPoint * 100); - DISPLAYLEVEL(2, "k=%u\nd=%u\nsteps=%u\nsplit=%u\n", coverParams->k, coverParams->d, - coverParams->steps, splitPercentage); - } - } else { - dictSize = ZDICT_trainFromBuffer_cover(dictBuffer, maxDictSize, srcBuffer, + coverParams); + if (!ZDICT_isError(dictSize)) { + unsigned splitPercentage = (unsigned)(coverParams->splitPoint * 100); + DISPLAYLEVEL(2, "k=%u\nd=%u\nsteps=%u\nsplit=%u\n", coverParams->k, coverParams->d, + coverParams->steps, splitPercentage); + } + } else { + dictSize = ZDICT_trainFromBuffer_cover(dictBuffer, maxDictSize, srcBuffer, sampleSizes, nbSamplesLoaded, *coverParams); - } - } else { - assert(fastCoverParams != NULL); - if (optimize) { - dictSize = ZDICT_optimizeTrainFromBuffer_fastCover(dictBuffer, maxDictSize, + } + } else { + assert(fastCoverParams != NULL); + if (optimize) { + dictSize = ZDICT_optimizeTrainFromBuffer_fastCover(dictBuffer, maxDictSize, srcBuffer, sampleSizes, nbSamplesLoaded, - fastCoverParams); - if (!ZDICT_isError(dictSize)) { - unsigned splitPercentage = (unsigned)(fastCoverParams->splitPoint * 100); - DISPLAYLEVEL(2, "k=%u\nd=%u\nf=%u\nsteps=%u\nsplit=%u\naccel=%u\n", fastCoverParams->k, - fastCoverParams->d, fastCoverParams->f, fastCoverParams->steps, splitPercentage, - fastCoverParams->accel); - } - } else { - dictSize = ZDICT_trainFromBuffer_fastCover(dictBuffer, maxDictSize, srcBuffer, + fastCoverParams); + if (!ZDICT_isError(dictSize)) { + unsigned splitPercentage = (unsigned)(fastCoverParams->splitPoint * 100); + DISPLAYLEVEL(2, "k=%u\nd=%u\nf=%u\nsteps=%u\nsplit=%u\naccel=%u\n", fastCoverParams->k, + fastCoverParams->d, fastCoverParams->f, fastCoverParams->steps, splitPercentage, + fastCoverParams->accel); + } + } else { + dictSize = ZDICT_trainFromBuffer_fastCover(dictBuffer, maxDictSize, srcBuffer, sampleSizes, nbSamplesLoaded, *fastCoverParams); - } - } - if (ZDICT_isError(dictSize)) { - DISPLAYLEVEL(1, "dictionary training failed : %s \n", ZDICT_getErrorName(dictSize)); /* should not happen */ - result = 1; - goto _cleanup; - } - /* save dict */ - DISPLAYLEVEL(2, "Save dictionary of size %u into file %s \n", (unsigned)dictSize, dictFileName); - DiB_saveDict(dictFileName, dictBuffer, dictSize); - } - - /* clean up */ -_cleanup: - free(srcBuffer); - free(sampleSizes); - free(dictBuffer); - return result; -} + } + } + if (ZDICT_isError(dictSize)) { + DISPLAYLEVEL(1, "dictionary training failed : %s \n", ZDICT_getErrorName(dictSize)); /* should not happen */ + result = 1; + goto _cleanup; + } + /* save dict */ + DISPLAYLEVEL(2, "Save dictionary of size %u into file %s \n", (unsigned)dictSize, dictFileName); + DiB_saveDict(dictFileName, dictBuffer, dictSize); + } + + /* clean up */ +_cleanup: + free(srcBuffer); + free(sampleSizes); + free(dictBuffer); + return result; +} diff --git a/contrib/libs/zstd/programs/dibio.h b/contrib/libs/zstd/programs/dibio.h index 666c1e6618..9be3d57055 100644 --- a/contrib/libs/zstd/programs/dibio.h +++ b/contrib/libs/zstd/programs/dibio.h @@ -1,39 +1,39 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -/* This library is designed for a single-threaded console application. -* It exit() and printf() into stderr when it encounters an error condition. */ - -#ifndef DIBIO_H_003 -#define DIBIO_H_003 - - -/*-************************************* -* Dependencies -***************************************/ -#define ZDICT_STATIC_LINKING_ONLY -#include "../lib/zdict.h" /* ZDICT_params_t */ - - -/*-************************************* -* Public functions -***************************************/ -/*! DiB_trainFromFiles() : - Train a dictionary from a set of files provided by `fileNamesTable`. - Resulting dictionary is written into file `dictFileName`. - `parameters` is optional and can be provided with values set to 0, meaning "default". - @return : 0 == ok. Any other : error. -*/ +/* + * Copyright (c) Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + +/* This library is designed for a single-threaded console application. +* It exit() and printf() into stderr when it encounters an error condition. */ + +#ifndef DIBIO_H_003 +#define DIBIO_H_003 + + +/*-************************************* +* Dependencies +***************************************/ +#define ZDICT_STATIC_LINKING_ONLY +#include "../lib/zdict.h" /* ZDICT_params_t */ + + +/*-************************************* +* Public functions +***************************************/ +/*! DiB_trainFromFiles() : + Train a dictionary from a set of files provided by `fileNamesTable`. + Resulting dictionary is written into file `dictFileName`. + `parameters` is optional and can be provided with values set to 0, meaning "default". + @return : 0 == ok. Any other : error. +*/ int DiB_trainFromFiles(const char* dictFileName, size_t maxDictSize, const char** fileNamesTable, int nbFiles, size_t chunkSize, - ZDICT_legacy_params_t* params, ZDICT_cover_params_t* coverParams, + ZDICT_legacy_params_t* params, ZDICT_cover_params_t* coverParams, ZDICT_fastCover_params_t* fastCoverParams, int optimize, unsigned memLimit); - -#endif + +#endif diff --git a/contrib/libs/zstd/programs/fileio.c b/contrib/libs/zstd/programs/fileio.c index 0a0dfc42cb..eec90d13e1 100644 --- a/contrib/libs/zstd/programs/fileio.c +++ b/contrib/libs/zstd/programs/fileio.c @@ -1,744 +1,744 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - -/* ************************************* -* Compiler Options -***************************************/ -#ifdef _MSC_VER /* Visual */ -# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */ -# pragma warning(disable : 4204) /* non-constant aggregate initializer */ -#endif -#if defined(__MINGW32__) && !defined(_POSIX_SOURCE) -# define _POSIX_SOURCE 1 /* disable %llu warnings with MinGW on Windows */ -#endif - -/*-************************************* -* Includes -***************************************/ -#include "platform.h" /* Large Files support, SET_BINARY_MODE */ -#include "util.h" /* UTIL_getFileSize, UTIL_isRegularFile, UTIL_isSameFile */ -#include <stdio.h> /* fprintf, open, fdopen, fread, _fileno, stdin, stdout */ -#include <stdlib.h> /* malloc, free */ -#include <string.h> /* strcmp, strlen */ -#include <fcntl.h> /* O_WRONLY */ -#include <assert.h> -#include <errno.h> /* errno */ -#include <limits.h> /* INT_MAX */ -#include <signal.h> -#include "timefn.h" /* UTIL_getTime, UTIL_clockSpanMicro */ - -#if defined (_MSC_VER) -# include <sys/stat.h> -# include <io.h> -#endif - -#include "../lib/common/mem.h" /* U32, U64 */ -#include "fileio.h" - -#define ZSTD_STATIC_LINKING_ONLY /* ZSTD_magicNumber, ZSTD_frameHeaderSize_max */ -#include "../lib/zstd.h" -#include "../lib/zstd_errors.h" /* ZSTD_error_frameParameter_windowTooLarge */ - -#if defined(ZSTD_GZCOMPRESS) || defined(ZSTD_GZDECOMPRESS) -# error #include <zlib.h> -# if !defined(z_const) -# define z_const -# endif -#endif - -#if defined(ZSTD_LZMACOMPRESS) || defined(ZSTD_LZMADECOMPRESS) -# error #include <lzma.h> -#endif - -#define LZ4_MAGICNUMBER 0x184D2204 -#if defined(ZSTD_LZ4COMPRESS) || defined(ZSTD_LZ4DECOMPRESS) -# define LZ4F_ENABLE_OBSOLETE_ENUMS -# error #include <lz4frame.h> -# error #include <lz4.h> -#endif - - -/*-************************************* -* Constants -***************************************/ -#define ADAPT_WINDOWLOG_DEFAULT 23 /* 8 MB */ -#define DICTSIZE_MAX (32 MB) /* protection against large input (attack scenario) */ - -#define FNSPACE 30 - -/* Default file permissions 0666 (modulated by umask) */ -#if !defined(_WIN32) -/* These macros aren't defined on windows. */ -#define DEFAULT_FILE_PERMISSIONS (S_IRUSR|S_IWUSR|S_IRGRP|S_IWGRP|S_IROTH|S_IWOTH) -#else -#define DEFAULT_FILE_PERMISSIONS (0666) -#endif - -/*-************************************* -* Macros -***************************************/ -#define KB *(1 <<10) -#define MB *(1 <<20) -#define GB *(1U<<30) -#undef MAX -#define MAX(a,b) ((a)>(b) ? (a) : (b)) - -struct FIO_display_prefs_s { - int displayLevel; /* 0 : no display; 1: errors; 2: + result + interaction + warnings; 3: + progression; 4: + information */ - FIO_progressSetting_e progressSetting; -}; - -static FIO_display_prefs_t g_display_prefs = {2, FIO_ps_auto}; - -#define DISPLAY(...) fprintf(stderr, __VA_ARGS__) -#define DISPLAYOUT(...) fprintf(stdout, __VA_ARGS__) -#define DISPLAYLEVEL(l, ...) { if (g_display_prefs.displayLevel>=l) { DISPLAY(__VA_ARGS__); } } - -static const U64 g_refreshRate = SEC_TO_MICRO / 6; -static UTIL_time_t g_displayClock = UTIL_TIME_INITIALIZER; - -#define READY_FOR_UPDATE() ((g_display_prefs.progressSetting != FIO_ps_never) && UTIL_clockSpanMicro(g_displayClock) > g_refreshRate) -#define DELAY_NEXT_UPDATE() { g_displayClock = UTIL_getTime(); } -#define DISPLAYUPDATE(l, ...) { \ - if (g_display_prefs.displayLevel>=l && (g_display_prefs.progressSetting != FIO_ps_never)) { \ - if (READY_FOR_UPDATE() || (g_display_prefs.displayLevel>=4)) { \ - DELAY_NEXT_UPDATE(); \ - DISPLAY(__VA_ARGS__); \ - if (g_display_prefs.displayLevel>=4) fflush(stderr); \ - } } } - -#undef MIN /* in case it would be already defined */ -#define MIN(a,b) ((a) < (b) ? (a) : (b)) - - -#define EXM_THROW(error, ...) \ -{ \ - DISPLAYLEVEL(1, "zstd: "); \ - DISPLAYLEVEL(5, "Error defined at %s, line %i : \n", __FILE__, __LINE__); \ - DISPLAYLEVEL(1, "error %i : ", error); \ - DISPLAYLEVEL(1, __VA_ARGS__); \ - DISPLAYLEVEL(1, " \n"); \ - exit(error); \ -} - -#define CHECK_V(v, f) \ - v = f; \ - if (ZSTD_isError(v)) { \ - DISPLAYLEVEL(5, "%s \n", #f); \ - EXM_THROW(11, "%s", ZSTD_getErrorName(v)); \ - } -#define CHECK(f) { size_t err; CHECK_V(err, f); } - - -/*-************************************ -* Signal (Ctrl-C trapping) -**************************************/ -static const char* g_artefact = NULL; -static void INThandler(int sig) -{ - assert(sig==SIGINT); (void)sig; -#if !defined(_MSC_VER) - signal(sig, SIG_IGN); /* this invocation generates a buggy warning in Visual Studio */ -#endif - if (g_artefact) { - assert(UTIL_isRegularFile(g_artefact)); - remove(g_artefact); - } - DISPLAY("\n"); - exit(2); -} -static void addHandler(char const* dstFileName) -{ - if (UTIL_isRegularFile(dstFileName)) { - g_artefact = dstFileName; - signal(SIGINT, INThandler); - } else { - g_artefact = NULL; - } -} -/* Idempotent */ -static void clearHandler(void) -{ - if (g_artefact) signal(SIGINT, SIG_DFL); - g_artefact = NULL; -} - - -/*-********************************************************* -* Termination signal trapping (Print debug stack trace) -***********************************************************/ -#if defined(__has_feature) && !defined(BACKTRACE_ENABLE) /* Clang compiler */ -# if (__has_feature(address_sanitizer)) -# define BACKTRACE_ENABLE 0 -# endif /* __has_feature(address_sanitizer) */ -#elif defined(__SANITIZE_ADDRESS__) && !defined(BACKTRACE_ENABLE) /* GCC compiler */ -# define BACKTRACE_ENABLE 0 -#endif - -#if !defined(BACKTRACE_ENABLE) -/* automatic detector : backtrace enabled by default on linux+glibc and osx */ -# if (defined(__linux__) && (defined(__GLIBC__) && !defined(__UCLIBC__))) \ - || (defined(__APPLE__) && defined(__MACH__)) -# define BACKTRACE_ENABLE 1 -# else -# define BACKTRACE_ENABLE 0 -# endif -#endif - -/* note : after this point, BACKTRACE_ENABLE is necessarily defined */ - - -#if BACKTRACE_ENABLE - -#include <execinfo.h> /* backtrace, backtrace_symbols */ - -#define MAX_STACK_FRAMES 50 - -static void ABRThandler(int sig) { - const char* name; - void* addrlist[MAX_STACK_FRAMES]; - char** symbollist; - int addrlen, i; - - switch (sig) { - case SIGABRT: name = "SIGABRT"; break; - case SIGFPE: name = "SIGFPE"; break; - case SIGILL: name = "SIGILL"; break; - case SIGINT: name = "SIGINT"; break; - case SIGSEGV: name = "SIGSEGV"; break; - default: name = "UNKNOWN"; - } - - DISPLAY("Caught %s signal, printing stack:\n", name); - /* Retrieve current stack addresses. */ - addrlen = backtrace(addrlist, MAX_STACK_FRAMES); - if (addrlen == 0) { - DISPLAY("\n"); - return; - } - /* Create readable strings to each frame. */ - symbollist = backtrace_symbols(addrlist, addrlen); - /* Print the stack trace, excluding calls handling the signal. */ - for (i = ZSTD_START_SYMBOLLIST_FRAME; i < addrlen; i++) { - DISPLAY("%s\n", symbollist[i]); - } - free(symbollist); - /* Reset and raise the signal so default handler runs. */ - signal(sig, SIG_DFL); - raise(sig); -} -#endif - -void FIO_addAbortHandler() -{ -#if BACKTRACE_ENABLE - signal(SIGABRT, ABRThandler); - signal(SIGFPE, ABRThandler); - signal(SIGILL, ABRThandler); - signal(SIGSEGV, ABRThandler); - signal(SIGBUS, ABRThandler); -#endif -} - - -/*-************************************************************ -* Avoid fseek()'s 2GiB barrier with MSVC, macOS, *BSD, MinGW -***************************************************************/ -#if defined(_MSC_VER) && _MSC_VER >= 1400 -# define LONG_SEEK _fseeki64 -# define LONG_TELL _ftelli64 -#elif !defined(__64BIT__) && (PLATFORM_POSIX_VERSION >= 200112L) /* No point defining Large file for 64 bit */ -# define LONG_SEEK fseeko -# define LONG_TELL ftello -#elif defined(__MINGW32__) && !defined(__STRICT_ANSI__) && !defined(__NO_MINGW_LFS) && defined(__MSVCRT__) -# define LONG_SEEK fseeko64 -# define LONG_TELL ftello64 -#elif defined(_WIN32) && !defined(__DJGPP__) -# include <windows.h> - static int LONG_SEEK(FILE* file, __int64 offset, int origin) { - LARGE_INTEGER off; - DWORD method; - off.QuadPart = offset; - if (origin == SEEK_END) - method = FILE_END; - else if (origin == SEEK_CUR) - method = FILE_CURRENT; - else - method = FILE_BEGIN; - - if (SetFilePointerEx((HANDLE) _get_osfhandle(_fileno(file)), off, NULL, method)) - return 0; - else - return -1; - } - static __int64 LONG_TELL(FILE* file) { - LARGE_INTEGER off, newOff; - off.QuadPart = 0; - newOff.QuadPart = 0; - SetFilePointerEx((HANDLE) _get_osfhandle(_fileno(file)), off, &newOff, FILE_CURRENT); - return newOff.QuadPart; - } -#else -# define LONG_SEEK fseek -# define LONG_TELL ftell -#endif - - -/*-************************************* -* Parameters: FIO_prefs_t -***************************************/ - -/* typedef'd to FIO_prefs_t within fileio.h */ -struct FIO_prefs_s { - - /* Algorithm preferences */ - FIO_compressionType_t compressionType; - U32 sparseFileSupport; /* 0: no sparse allowed; 1: auto (file yes, stdout no); 2: force sparse */ - int dictIDFlag; - int checksumFlag; - int blockSize; - int overlapLog; - U32 adaptiveMode; - U32 useRowMatchFinder; - int rsyncable; - int minAdaptLevel; - int maxAdaptLevel; - int ldmFlag; - int ldmHashLog; - int ldmMinMatch; - int ldmBucketSizeLog; - int ldmHashRateLog; - size_t streamSrcSize; - size_t targetCBlockSize; - int srcSizeHint; - int testMode; +/* + * Copyright (c) Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + + +/* ************************************* +* Compiler Options +***************************************/ +#ifdef _MSC_VER /* Visual */ +# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */ +# pragma warning(disable : 4204) /* non-constant aggregate initializer */ +#endif +#if defined(__MINGW32__) && !defined(_POSIX_SOURCE) +# define _POSIX_SOURCE 1 /* disable %llu warnings with MinGW on Windows */ +#endif + +/*-************************************* +* Includes +***************************************/ +#include "platform.h" /* Large Files support, SET_BINARY_MODE */ +#include "util.h" /* UTIL_getFileSize, UTIL_isRegularFile, UTIL_isSameFile */ +#include <stdio.h> /* fprintf, open, fdopen, fread, _fileno, stdin, stdout */ +#include <stdlib.h> /* malloc, free */ +#include <string.h> /* strcmp, strlen */ +#include <fcntl.h> /* O_WRONLY */ +#include <assert.h> +#include <errno.h> /* errno */ +#include <limits.h> /* INT_MAX */ +#include <signal.h> +#include "timefn.h" /* UTIL_getTime, UTIL_clockSpanMicro */ + +#if defined (_MSC_VER) +# include <sys/stat.h> +# include <io.h> +#endif + +#include "../lib/common/mem.h" /* U32, U64 */ +#include "fileio.h" + +#define ZSTD_STATIC_LINKING_ONLY /* ZSTD_magicNumber, ZSTD_frameHeaderSize_max */ +#include "../lib/zstd.h" +#include "../lib/zstd_errors.h" /* ZSTD_error_frameParameter_windowTooLarge */ + +#if defined(ZSTD_GZCOMPRESS) || defined(ZSTD_GZDECOMPRESS) +# error #include <zlib.h> +# if !defined(z_const) +# define z_const +# endif +#endif + +#if defined(ZSTD_LZMACOMPRESS) || defined(ZSTD_LZMADECOMPRESS) +# error #include <lzma.h> +#endif + +#define LZ4_MAGICNUMBER 0x184D2204 +#if defined(ZSTD_LZ4COMPRESS) || defined(ZSTD_LZ4DECOMPRESS) +# define LZ4F_ENABLE_OBSOLETE_ENUMS +# error #include <lz4frame.h> +# error #include <lz4.h> +#endif + + +/*-************************************* +* Constants +***************************************/ +#define ADAPT_WINDOWLOG_DEFAULT 23 /* 8 MB */ +#define DICTSIZE_MAX (32 MB) /* protection against large input (attack scenario) */ + +#define FNSPACE 30 + +/* Default file permissions 0666 (modulated by umask) */ +#if !defined(_WIN32) +/* These macros aren't defined on windows. */ +#define DEFAULT_FILE_PERMISSIONS (S_IRUSR|S_IWUSR|S_IRGRP|S_IWGRP|S_IROTH|S_IWOTH) +#else +#define DEFAULT_FILE_PERMISSIONS (0666) +#endif + +/*-************************************* +* Macros +***************************************/ +#define KB *(1 <<10) +#define MB *(1 <<20) +#define GB *(1U<<30) +#undef MAX +#define MAX(a,b) ((a)>(b) ? (a) : (b)) + +struct FIO_display_prefs_s { + int displayLevel; /* 0 : no display; 1: errors; 2: + result + interaction + warnings; 3: + progression; 4: + information */ + FIO_progressSetting_e progressSetting; +}; + +static FIO_display_prefs_t g_display_prefs = {2, FIO_ps_auto}; + +#define DISPLAY(...) fprintf(stderr, __VA_ARGS__) +#define DISPLAYOUT(...) fprintf(stdout, __VA_ARGS__) +#define DISPLAYLEVEL(l, ...) { if (g_display_prefs.displayLevel>=l) { DISPLAY(__VA_ARGS__); } } + +static const U64 g_refreshRate = SEC_TO_MICRO / 6; +static UTIL_time_t g_displayClock = UTIL_TIME_INITIALIZER; + +#define READY_FOR_UPDATE() ((g_display_prefs.progressSetting != FIO_ps_never) && UTIL_clockSpanMicro(g_displayClock) > g_refreshRate) +#define DELAY_NEXT_UPDATE() { g_displayClock = UTIL_getTime(); } +#define DISPLAYUPDATE(l, ...) { \ + if (g_display_prefs.displayLevel>=l && (g_display_prefs.progressSetting != FIO_ps_never)) { \ + if (READY_FOR_UPDATE() || (g_display_prefs.displayLevel>=4)) { \ + DELAY_NEXT_UPDATE(); \ + DISPLAY(__VA_ARGS__); \ + if (g_display_prefs.displayLevel>=4) fflush(stderr); \ + } } } + +#undef MIN /* in case it would be already defined */ +#define MIN(a,b) ((a) < (b) ? (a) : (b)) + + +#define EXM_THROW(error, ...) \ +{ \ + DISPLAYLEVEL(1, "zstd: "); \ + DISPLAYLEVEL(5, "Error defined at %s, line %i : \n", __FILE__, __LINE__); \ + DISPLAYLEVEL(1, "error %i : ", error); \ + DISPLAYLEVEL(1, __VA_ARGS__); \ + DISPLAYLEVEL(1, " \n"); \ + exit(error); \ +} + +#define CHECK_V(v, f) \ + v = f; \ + if (ZSTD_isError(v)) { \ + DISPLAYLEVEL(5, "%s \n", #f); \ + EXM_THROW(11, "%s", ZSTD_getErrorName(v)); \ + } +#define CHECK(f) { size_t err; CHECK_V(err, f); } + + +/*-************************************ +* Signal (Ctrl-C trapping) +**************************************/ +static const char* g_artefact = NULL; +static void INThandler(int sig) +{ + assert(sig==SIGINT); (void)sig; +#if !defined(_MSC_VER) + signal(sig, SIG_IGN); /* this invocation generates a buggy warning in Visual Studio */ +#endif + if (g_artefact) { + assert(UTIL_isRegularFile(g_artefact)); + remove(g_artefact); + } + DISPLAY("\n"); + exit(2); +} +static void addHandler(char const* dstFileName) +{ + if (UTIL_isRegularFile(dstFileName)) { + g_artefact = dstFileName; + signal(SIGINT, INThandler); + } else { + g_artefact = NULL; + } +} +/* Idempotent */ +static void clearHandler(void) +{ + if (g_artefact) signal(SIGINT, SIG_DFL); + g_artefact = NULL; +} + + +/*-********************************************************* +* Termination signal trapping (Print debug stack trace) +***********************************************************/ +#if defined(__has_feature) && !defined(BACKTRACE_ENABLE) /* Clang compiler */ +# if (__has_feature(address_sanitizer)) +# define BACKTRACE_ENABLE 0 +# endif /* __has_feature(address_sanitizer) */ +#elif defined(__SANITIZE_ADDRESS__) && !defined(BACKTRACE_ENABLE) /* GCC compiler */ +# define BACKTRACE_ENABLE 0 +#endif + +#if !defined(BACKTRACE_ENABLE) +/* automatic detector : backtrace enabled by default on linux+glibc and osx */ +# if (defined(__linux__) && (defined(__GLIBC__) && !defined(__UCLIBC__))) \ + || (defined(__APPLE__) && defined(__MACH__)) +# define BACKTRACE_ENABLE 1 +# else +# define BACKTRACE_ENABLE 0 +# endif +#endif + +/* note : after this point, BACKTRACE_ENABLE is necessarily defined */ + + +#if BACKTRACE_ENABLE + +#include <execinfo.h> /* backtrace, backtrace_symbols */ + +#define MAX_STACK_FRAMES 50 + +static void ABRThandler(int sig) { + const char* name; + void* addrlist[MAX_STACK_FRAMES]; + char** symbollist; + int addrlen, i; + + switch (sig) { + case SIGABRT: name = "SIGABRT"; break; + case SIGFPE: name = "SIGFPE"; break; + case SIGILL: name = "SIGILL"; break; + case SIGINT: name = "SIGINT"; break; + case SIGSEGV: name = "SIGSEGV"; break; + default: name = "UNKNOWN"; + } + + DISPLAY("Caught %s signal, printing stack:\n", name); + /* Retrieve current stack addresses. */ + addrlen = backtrace(addrlist, MAX_STACK_FRAMES); + if (addrlen == 0) { + DISPLAY("\n"); + return; + } + /* Create readable strings to each frame. */ + symbollist = backtrace_symbols(addrlist, addrlen); + /* Print the stack trace, excluding calls handling the signal. */ + for (i = ZSTD_START_SYMBOLLIST_FRAME; i < addrlen; i++) { + DISPLAY("%s\n", symbollist[i]); + } + free(symbollist); + /* Reset and raise the signal so default handler runs. */ + signal(sig, SIG_DFL); + raise(sig); +} +#endif + +void FIO_addAbortHandler() +{ +#if BACKTRACE_ENABLE + signal(SIGABRT, ABRThandler); + signal(SIGFPE, ABRThandler); + signal(SIGILL, ABRThandler); + signal(SIGSEGV, ABRThandler); + signal(SIGBUS, ABRThandler); +#endif +} + + +/*-************************************************************ +* Avoid fseek()'s 2GiB barrier with MSVC, macOS, *BSD, MinGW +***************************************************************/ +#if defined(_MSC_VER) && _MSC_VER >= 1400 +# define LONG_SEEK _fseeki64 +# define LONG_TELL _ftelli64 +#elif !defined(__64BIT__) && (PLATFORM_POSIX_VERSION >= 200112L) /* No point defining Large file for 64 bit */ +# define LONG_SEEK fseeko +# define LONG_TELL ftello +#elif defined(__MINGW32__) && !defined(__STRICT_ANSI__) && !defined(__NO_MINGW_LFS) && defined(__MSVCRT__) +# define LONG_SEEK fseeko64 +# define LONG_TELL ftello64 +#elif defined(_WIN32) && !defined(__DJGPP__) +# include <windows.h> + static int LONG_SEEK(FILE* file, __int64 offset, int origin) { + LARGE_INTEGER off; + DWORD method; + off.QuadPart = offset; + if (origin == SEEK_END) + method = FILE_END; + else if (origin == SEEK_CUR) + method = FILE_CURRENT; + else + method = FILE_BEGIN; + + if (SetFilePointerEx((HANDLE) _get_osfhandle(_fileno(file)), off, NULL, method)) + return 0; + else + return -1; + } + static __int64 LONG_TELL(FILE* file) { + LARGE_INTEGER off, newOff; + off.QuadPart = 0; + newOff.QuadPart = 0; + SetFilePointerEx((HANDLE) _get_osfhandle(_fileno(file)), off, &newOff, FILE_CURRENT); + return newOff.QuadPart; + } +#else +# define LONG_SEEK fseek +# define LONG_TELL ftell +#endif + + +/*-************************************* +* Parameters: FIO_prefs_t +***************************************/ + +/* typedef'd to FIO_prefs_t within fileio.h */ +struct FIO_prefs_s { + + /* Algorithm preferences */ + FIO_compressionType_t compressionType; + U32 sparseFileSupport; /* 0: no sparse allowed; 1: auto (file yes, stdout no); 2: force sparse */ + int dictIDFlag; + int checksumFlag; + int blockSize; + int overlapLog; + U32 adaptiveMode; + U32 useRowMatchFinder; + int rsyncable; + int minAdaptLevel; + int maxAdaptLevel; + int ldmFlag; + int ldmHashLog; + int ldmMinMatch; + int ldmBucketSizeLog; + int ldmHashRateLog; + size_t streamSrcSize; + size_t targetCBlockSize; + int srcSizeHint; + int testMode; ZSTD_paramSwitch_e literalCompressionMode; - - /* IO preferences */ - U32 removeSrcFile; - U32 overwrite; - - /* Computation resources preferences */ - unsigned memLimit; - int nbWorkers; - - int excludeCompressedFiles; - int patchFromMode; - int contentSize; - int allowBlockDevices; -}; - -/*-************************************* -* Parameters: FIO_ctx_t -***************************************/ - -/* typedef'd to FIO_ctx_t within fileio.h */ -struct FIO_ctx_s { - - /* file i/o info */ - int nbFilesTotal; - int hasStdinInput; - int hasStdoutOutput; - - /* file i/o state */ - int currFileIdx; - int nbFilesProcessed; - size_t totalBytesInput; - size_t totalBytesOutput; -}; - - -/*-************************************* -* Parameters: Initialization -***************************************/ - -#define FIO_OVERLAP_LOG_NOTSET 9999 -#define FIO_LDM_PARAM_NOTSET 9999 - - -FIO_prefs_t* FIO_createPreferences(void) -{ - FIO_prefs_t* const ret = (FIO_prefs_t*)malloc(sizeof(FIO_prefs_t)); - if (!ret) EXM_THROW(21, "Allocation error : not enough memory"); - - ret->compressionType = FIO_zstdCompression; - ret->overwrite = 0; - ret->sparseFileSupport = ZSTD_SPARSE_DEFAULT; - ret->dictIDFlag = 1; - ret->checksumFlag = 1; - ret->removeSrcFile = 0; - ret->memLimit = 0; - ret->nbWorkers = 1; - ret->blockSize = 0; - ret->overlapLog = FIO_OVERLAP_LOG_NOTSET; - ret->adaptiveMode = 0; - ret->rsyncable = 0; - ret->minAdaptLevel = -50; /* initializing this value requires a constant, so ZSTD_minCLevel() doesn't work */ - ret->maxAdaptLevel = 22; /* initializing this value requires a constant, so ZSTD_maxCLevel() doesn't work */ - ret->ldmFlag = 0; - ret->ldmHashLog = 0; - ret->ldmMinMatch = 0; - ret->ldmBucketSizeLog = FIO_LDM_PARAM_NOTSET; - ret->ldmHashRateLog = FIO_LDM_PARAM_NOTSET; - ret->streamSrcSize = 0; - ret->targetCBlockSize = 0; - ret->srcSizeHint = 0; - ret->testMode = 0; + + /* IO preferences */ + U32 removeSrcFile; + U32 overwrite; + + /* Computation resources preferences */ + unsigned memLimit; + int nbWorkers; + + int excludeCompressedFiles; + int patchFromMode; + int contentSize; + int allowBlockDevices; +}; + +/*-************************************* +* Parameters: FIO_ctx_t +***************************************/ + +/* typedef'd to FIO_ctx_t within fileio.h */ +struct FIO_ctx_s { + + /* file i/o info */ + int nbFilesTotal; + int hasStdinInput; + int hasStdoutOutput; + + /* file i/o state */ + int currFileIdx; + int nbFilesProcessed; + size_t totalBytesInput; + size_t totalBytesOutput; +}; + + +/*-************************************* +* Parameters: Initialization +***************************************/ + +#define FIO_OVERLAP_LOG_NOTSET 9999 +#define FIO_LDM_PARAM_NOTSET 9999 + + +FIO_prefs_t* FIO_createPreferences(void) +{ + FIO_prefs_t* const ret = (FIO_prefs_t*)malloc(sizeof(FIO_prefs_t)); + if (!ret) EXM_THROW(21, "Allocation error : not enough memory"); + + ret->compressionType = FIO_zstdCompression; + ret->overwrite = 0; + ret->sparseFileSupport = ZSTD_SPARSE_DEFAULT; + ret->dictIDFlag = 1; + ret->checksumFlag = 1; + ret->removeSrcFile = 0; + ret->memLimit = 0; + ret->nbWorkers = 1; + ret->blockSize = 0; + ret->overlapLog = FIO_OVERLAP_LOG_NOTSET; + ret->adaptiveMode = 0; + ret->rsyncable = 0; + ret->minAdaptLevel = -50; /* initializing this value requires a constant, so ZSTD_minCLevel() doesn't work */ + ret->maxAdaptLevel = 22; /* initializing this value requires a constant, so ZSTD_maxCLevel() doesn't work */ + ret->ldmFlag = 0; + ret->ldmHashLog = 0; + ret->ldmMinMatch = 0; + ret->ldmBucketSizeLog = FIO_LDM_PARAM_NOTSET; + ret->ldmHashRateLog = FIO_LDM_PARAM_NOTSET; + ret->streamSrcSize = 0; + ret->targetCBlockSize = 0; + ret->srcSizeHint = 0; + ret->testMode = 0; ret->literalCompressionMode = ZSTD_ps_auto; - ret->excludeCompressedFiles = 0; - ret->allowBlockDevices = 0; - return ret; -} - -FIO_ctx_t* FIO_createContext(void) -{ - FIO_ctx_t* const ret = (FIO_ctx_t*)malloc(sizeof(FIO_ctx_t)); - if (!ret) EXM_THROW(21, "Allocation error : not enough memory"); - - ret->currFileIdx = 0; - ret->hasStdinInput = 0; - ret->hasStdoutOutput = 0; - ret->nbFilesTotal = 1; - ret->nbFilesProcessed = 0; - ret->totalBytesInput = 0; - ret->totalBytesOutput = 0; - return ret; -} - -void FIO_freePreferences(FIO_prefs_t* const prefs) -{ - free(prefs); -} - -void FIO_freeContext(FIO_ctx_t* const fCtx) -{ - free(fCtx); -} - - -/*-************************************* -* Parameters: Display Options -***************************************/ - -void FIO_setNotificationLevel(int level) { g_display_prefs.displayLevel=level; } - -void FIO_setProgressSetting(FIO_progressSetting_e setting) { g_display_prefs.progressSetting = setting; } - - -/*-************************************* -* Parameters: Setters -***************************************/ - -/* FIO_prefs_t functions */ - -void FIO_setCompressionType(FIO_prefs_t* const prefs, FIO_compressionType_t compressionType) { prefs->compressionType = compressionType; } - -void FIO_overwriteMode(FIO_prefs_t* const prefs) { prefs->overwrite = 1; } - -void FIO_setSparseWrite(FIO_prefs_t* const prefs, unsigned sparse) { prefs->sparseFileSupport = sparse; } - -void FIO_setDictIDFlag(FIO_prefs_t* const prefs, int dictIDFlag) { prefs->dictIDFlag = dictIDFlag; } - -void FIO_setChecksumFlag(FIO_prefs_t* const prefs, int checksumFlag) { prefs->checksumFlag = checksumFlag; } - -void FIO_setRemoveSrcFile(FIO_prefs_t* const prefs, unsigned flag) { prefs->removeSrcFile = (flag>0); } - -void FIO_setMemLimit(FIO_prefs_t* const prefs, unsigned memLimit) { prefs->memLimit = memLimit; } - -void FIO_setNbWorkers(FIO_prefs_t* const prefs, int nbWorkers) { -#ifndef ZSTD_MULTITHREAD - if (nbWorkers > 0) DISPLAYLEVEL(2, "Note : multi-threading is disabled \n"); -#endif - prefs->nbWorkers = nbWorkers; -} - -void FIO_setExcludeCompressedFile(FIO_prefs_t* const prefs, int excludeCompressedFiles) { prefs->excludeCompressedFiles = excludeCompressedFiles; } - -void FIO_setAllowBlockDevices(FIO_prefs_t* const prefs, int allowBlockDevices) { prefs->allowBlockDevices = allowBlockDevices; } - -void FIO_setBlockSize(FIO_prefs_t* const prefs, int blockSize) { - if (blockSize && prefs->nbWorkers==0) - DISPLAYLEVEL(2, "Setting block size is useless in single-thread mode \n"); - prefs->blockSize = blockSize; -} - -void FIO_setOverlapLog(FIO_prefs_t* const prefs, int overlapLog){ - if (overlapLog && prefs->nbWorkers==0) - DISPLAYLEVEL(2, "Setting overlapLog is useless in single-thread mode \n"); - prefs->overlapLog = overlapLog; -} - -void FIO_setAdaptiveMode(FIO_prefs_t* const prefs, unsigned adapt) { - if ((adapt>0) && (prefs->nbWorkers==0)) - EXM_THROW(1, "Adaptive mode is not compatible with single thread mode \n"); - prefs->adaptiveMode = adapt; -} - -void FIO_setUseRowMatchFinder(FIO_prefs_t* const prefs, int useRowMatchFinder) { - prefs->useRowMatchFinder = useRowMatchFinder; -} - -void FIO_setRsyncable(FIO_prefs_t* const prefs, int rsyncable) { - if ((rsyncable>0) && (prefs->nbWorkers==0)) - EXM_THROW(1, "Rsyncable mode is not compatible with single thread mode \n"); - prefs->rsyncable = rsyncable; -} - -void FIO_setStreamSrcSize(FIO_prefs_t* const prefs, size_t streamSrcSize) { - prefs->streamSrcSize = streamSrcSize; -} - -void FIO_setTargetCBlockSize(FIO_prefs_t* const prefs, size_t targetCBlockSize) { - prefs->targetCBlockSize = targetCBlockSize; -} - -void FIO_setSrcSizeHint(FIO_prefs_t* const prefs, size_t srcSizeHint) { - prefs->srcSizeHint = (int)MIN((size_t)INT_MAX, srcSizeHint); -} - -void FIO_setTestMode(FIO_prefs_t* const prefs, int testMode) { - prefs->testMode = (testMode!=0); -} - -void FIO_setLiteralCompressionMode( - FIO_prefs_t* const prefs, + ret->excludeCompressedFiles = 0; + ret->allowBlockDevices = 0; + return ret; +} + +FIO_ctx_t* FIO_createContext(void) +{ + FIO_ctx_t* const ret = (FIO_ctx_t*)malloc(sizeof(FIO_ctx_t)); + if (!ret) EXM_THROW(21, "Allocation error : not enough memory"); + + ret->currFileIdx = 0; + ret->hasStdinInput = 0; + ret->hasStdoutOutput = 0; + ret->nbFilesTotal = 1; + ret->nbFilesProcessed = 0; + ret->totalBytesInput = 0; + ret->totalBytesOutput = 0; + return ret; +} + +void FIO_freePreferences(FIO_prefs_t* const prefs) +{ + free(prefs); +} + +void FIO_freeContext(FIO_ctx_t* const fCtx) +{ + free(fCtx); +} + + +/*-************************************* +* Parameters: Display Options +***************************************/ + +void FIO_setNotificationLevel(int level) { g_display_prefs.displayLevel=level; } + +void FIO_setProgressSetting(FIO_progressSetting_e setting) { g_display_prefs.progressSetting = setting; } + + +/*-************************************* +* Parameters: Setters +***************************************/ + +/* FIO_prefs_t functions */ + +void FIO_setCompressionType(FIO_prefs_t* const prefs, FIO_compressionType_t compressionType) { prefs->compressionType = compressionType; } + +void FIO_overwriteMode(FIO_prefs_t* const prefs) { prefs->overwrite = 1; } + +void FIO_setSparseWrite(FIO_prefs_t* const prefs, unsigned sparse) { prefs->sparseFileSupport = sparse; } + +void FIO_setDictIDFlag(FIO_prefs_t* const prefs, int dictIDFlag) { prefs->dictIDFlag = dictIDFlag; } + +void FIO_setChecksumFlag(FIO_prefs_t* const prefs, int checksumFlag) { prefs->checksumFlag = checksumFlag; } + +void FIO_setRemoveSrcFile(FIO_prefs_t* const prefs, unsigned flag) { prefs->removeSrcFile = (flag>0); } + +void FIO_setMemLimit(FIO_prefs_t* const prefs, unsigned memLimit) { prefs->memLimit = memLimit; } + +void FIO_setNbWorkers(FIO_prefs_t* const prefs, int nbWorkers) { +#ifndef ZSTD_MULTITHREAD + if (nbWorkers > 0) DISPLAYLEVEL(2, "Note : multi-threading is disabled \n"); +#endif + prefs->nbWorkers = nbWorkers; +} + +void FIO_setExcludeCompressedFile(FIO_prefs_t* const prefs, int excludeCompressedFiles) { prefs->excludeCompressedFiles = excludeCompressedFiles; } + +void FIO_setAllowBlockDevices(FIO_prefs_t* const prefs, int allowBlockDevices) { prefs->allowBlockDevices = allowBlockDevices; } + +void FIO_setBlockSize(FIO_prefs_t* const prefs, int blockSize) { + if (blockSize && prefs->nbWorkers==0) + DISPLAYLEVEL(2, "Setting block size is useless in single-thread mode \n"); + prefs->blockSize = blockSize; +} + +void FIO_setOverlapLog(FIO_prefs_t* const prefs, int overlapLog){ + if (overlapLog && prefs->nbWorkers==0) + DISPLAYLEVEL(2, "Setting overlapLog is useless in single-thread mode \n"); + prefs->overlapLog = overlapLog; +} + +void FIO_setAdaptiveMode(FIO_prefs_t* const prefs, unsigned adapt) { + if ((adapt>0) && (prefs->nbWorkers==0)) + EXM_THROW(1, "Adaptive mode is not compatible with single thread mode \n"); + prefs->adaptiveMode = adapt; +} + +void FIO_setUseRowMatchFinder(FIO_prefs_t* const prefs, int useRowMatchFinder) { + prefs->useRowMatchFinder = useRowMatchFinder; +} + +void FIO_setRsyncable(FIO_prefs_t* const prefs, int rsyncable) { + if ((rsyncable>0) && (prefs->nbWorkers==0)) + EXM_THROW(1, "Rsyncable mode is not compatible with single thread mode \n"); + prefs->rsyncable = rsyncable; +} + +void FIO_setStreamSrcSize(FIO_prefs_t* const prefs, size_t streamSrcSize) { + prefs->streamSrcSize = streamSrcSize; +} + +void FIO_setTargetCBlockSize(FIO_prefs_t* const prefs, size_t targetCBlockSize) { + prefs->targetCBlockSize = targetCBlockSize; +} + +void FIO_setSrcSizeHint(FIO_prefs_t* const prefs, size_t srcSizeHint) { + prefs->srcSizeHint = (int)MIN((size_t)INT_MAX, srcSizeHint); +} + +void FIO_setTestMode(FIO_prefs_t* const prefs, int testMode) { + prefs->testMode = (testMode!=0); +} + +void FIO_setLiteralCompressionMode( + FIO_prefs_t* const prefs, ZSTD_paramSwitch_e mode) { - prefs->literalCompressionMode = mode; -} - -void FIO_setAdaptMin(FIO_prefs_t* const prefs, int minCLevel) -{ -#ifndef ZSTD_NOCOMPRESS - assert(minCLevel >= ZSTD_minCLevel()); -#endif - prefs->minAdaptLevel = minCLevel; -} - -void FIO_setAdaptMax(FIO_prefs_t* const prefs, int maxCLevel) -{ - prefs->maxAdaptLevel = maxCLevel; -} - -void FIO_setLdmFlag(FIO_prefs_t* const prefs, unsigned ldmFlag) { - prefs->ldmFlag = (ldmFlag>0); -} - -void FIO_setLdmHashLog(FIO_prefs_t* const prefs, int ldmHashLog) { - prefs->ldmHashLog = ldmHashLog; -} - -void FIO_setLdmMinMatch(FIO_prefs_t* const prefs, int ldmMinMatch) { - prefs->ldmMinMatch = ldmMinMatch; -} - -void FIO_setLdmBucketSizeLog(FIO_prefs_t* const prefs, int ldmBucketSizeLog) { - prefs->ldmBucketSizeLog = ldmBucketSizeLog; -} - - -void FIO_setLdmHashRateLog(FIO_prefs_t* const prefs, int ldmHashRateLog) { - prefs->ldmHashRateLog = ldmHashRateLog; -} - -void FIO_setPatchFromMode(FIO_prefs_t* const prefs, int value) -{ - prefs->patchFromMode = value != 0; -} - -void FIO_setContentSize(FIO_prefs_t* const prefs, int value) -{ - prefs->contentSize = value != 0; -} - -/* FIO_ctx_t functions */ - -void FIO_setHasStdoutOutput(FIO_ctx_t* const fCtx, int value) { - fCtx->hasStdoutOutput = value; -} - -void FIO_setNbFilesTotal(FIO_ctx_t* const fCtx, int value) -{ - fCtx->nbFilesTotal = value; -} - -void FIO_determineHasStdinInput(FIO_ctx_t* const fCtx, const FileNamesTable* const filenames) { - size_t i = 0; - for ( ; i < filenames->tableSize; ++i) { - if (!strcmp(stdinmark, filenames->fileNames[i])) { - fCtx->hasStdinInput = 1; - return; - } - } -} - -/*-************************************* -* Functions -***************************************/ -/** FIO_removeFile() : - * @result : Unlink `fileName`, even if it's read-only */ -static int FIO_removeFile(const char* path) -{ - stat_t statbuf; - if (!UTIL_stat(path, &statbuf)) { - DISPLAYLEVEL(2, "zstd: Failed to stat %s while trying to remove it\n", path); - return 0; - } - if (!UTIL_isRegularFileStat(&statbuf)) { - DISPLAYLEVEL(2, "zstd: Refusing to remove non-regular file %s\n", path); - return 0; - } -#if defined(_WIN32) || defined(WIN32) - /* windows doesn't allow remove read-only files, - * so try to make it writable first */ - if (!(statbuf.st_mode & _S_IWRITE)) { - UTIL_chmod(path, &statbuf, _S_IWRITE); - } -#endif - return remove(path); -} - -/** FIO_openSrcFile() : - * condition : `srcFileName` must be non-NULL. `prefs` may be NULL. - * @result : FILE* to `srcFileName`, or NULL if it fails */ -static FILE* FIO_openSrcFile(const FIO_prefs_t* const prefs, const char* srcFileName) -{ - stat_t statbuf; - int allowBlockDevices = prefs != NULL ? prefs->allowBlockDevices : 0; - assert(srcFileName != NULL); - if (!strcmp (srcFileName, stdinmark)) { - DISPLAYLEVEL(4,"Using stdin for input \n"); - SET_BINARY_MODE(stdin); - return stdin; - } - - if (!UTIL_stat(srcFileName, &statbuf)) { - DISPLAYLEVEL(1, "zstd: can't stat %s : %s -- ignored \n", - srcFileName, strerror(errno)); - return NULL; - } - - if (!UTIL_isRegularFileStat(&statbuf) - && !UTIL_isFIFOStat(&statbuf) - && !(allowBlockDevices && UTIL_isBlockDevStat(&statbuf)) - ) { - DISPLAYLEVEL(1, "zstd: %s is not a regular file -- ignored \n", - srcFileName); - return NULL; - } - - { FILE* const f = fopen(srcFileName, "rb"); - if (f == NULL) - DISPLAYLEVEL(1, "zstd: %s: %s \n", srcFileName, strerror(errno)); - return f; - } -} - -/** FIO_openDstFile() : - * condition : `dstFileName` must be non-NULL. - * @result : FILE* to `dstFileName`, or NULL if it fails */ -static FILE* -FIO_openDstFile(FIO_ctx_t* fCtx, FIO_prefs_t* const prefs, - const char* srcFileName, const char* dstFileName, - const int mode) -{ - if (prefs->testMode) return NULL; /* do not open file in test mode */ - - assert(dstFileName != NULL); - if (!strcmp (dstFileName, stdoutmark)) { - DISPLAYLEVEL(4,"Using stdout for output \n"); - SET_BINARY_MODE(stdout); - if (prefs->sparseFileSupport == 1) { - prefs->sparseFileSupport = 0; - DISPLAYLEVEL(4, "Sparse File Support is automatically disabled on stdout ; try --sparse \n"); - } - return stdout; - } - - /* ensure dst is not the same as src */ - if (srcFileName != NULL && UTIL_isSameFile(srcFileName, dstFileName)) { - DISPLAYLEVEL(1, "zstd: Refusing to open an output file which will overwrite the input file \n"); - return NULL; - } - - if (prefs->sparseFileSupport == 1) { - prefs->sparseFileSupport = ZSTD_SPARSE_DEFAULT; - } - - if (UTIL_isRegularFile(dstFileName)) { - /* Check if destination file already exists */ -#if !defined(_WIN32) - /* this test does not work on Windows : - * `NUL` and `nul` are detected as regular files */ - if (!strcmp(dstFileName, nulmark)) { - EXM_THROW(40, "%s is unexpectedly categorized as a regular file", - dstFileName); - } -#endif - if (!prefs->overwrite) { - if (g_display_prefs.displayLevel <= 1) { - /* No interaction possible */ - DISPLAY("zstd: %s already exists; not overwritten \n", - dstFileName); - return NULL; - } - DISPLAY("zstd: %s already exists; ", dstFileName); - if (UTIL_requireUserConfirmation("overwrite (y/n) ? ", "Not overwritten \n", "yY", fCtx->hasStdinInput)) - return NULL; - } - /* need to unlink */ - FIO_removeFile(dstFileName); - } - - { -#if defined(_WIN32) - /* Windows requires opening the file as a "binary" file to avoid - * mangling. This macro doesn't exist on unix. */ - const int openflags = O_WRONLY|O_CREAT|O_TRUNC|O_BINARY; - const int fd = _open(dstFileName, openflags, mode); - FILE* f = NULL; - if (fd != -1) { - f = _fdopen(fd, "wb"); - } -#else - const int openflags = O_WRONLY|O_CREAT|O_TRUNC; - const int fd = open(dstFileName, openflags, mode); - FILE* f = NULL; - if (fd != -1) { - f = fdopen(fd, "wb"); - } -#endif - if (f == NULL) { - DISPLAYLEVEL(1, "zstd: %s: %s\n", dstFileName, strerror(errno)); - } - return f; - } -} - -/*! FIO_createDictBuffer() : - * creates a buffer, pointed by `*bufferPtr`, - * loads `filename` content into it, up to DICTSIZE_MAX bytes. - * @return : loaded size - * if fileName==NULL, returns 0 and a NULL pointer - */ -static size_t FIO_createDictBuffer(void** bufferPtr, const char* fileName, FIO_prefs_t* const prefs) -{ - FILE* fileHandle; - U64 fileSize; + prefs->literalCompressionMode = mode; +} + +void FIO_setAdaptMin(FIO_prefs_t* const prefs, int minCLevel) +{ +#ifndef ZSTD_NOCOMPRESS + assert(minCLevel >= ZSTD_minCLevel()); +#endif + prefs->minAdaptLevel = minCLevel; +} + +void FIO_setAdaptMax(FIO_prefs_t* const prefs, int maxCLevel) +{ + prefs->maxAdaptLevel = maxCLevel; +} + +void FIO_setLdmFlag(FIO_prefs_t* const prefs, unsigned ldmFlag) { + prefs->ldmFlag = (ldmFlag>0); +} + +void FIO_setLdmHashLog(FIO_prefs_t* const prefs, int ldmHashLog) { + prefs->ldmHashLog = ldmHashLog; +} + +void FIO_setLdmMinMatch(FIO_prefs_t* const prefs, int ldmMinMatch) { + prefs->ldmMinMatch = ldmMinMatch; +} + +void FIO_setLdmBucketSizeLog(FIO_prefs_t* const prefs, int ldmBucketSizeLog) { + prefs->ldmBucketSizeLog = ldmBucketSizeLog; +} + + +void FIO_setLdmHashRateLog(FIO_prefs_t* const prefs, int ldmHashRateLog) { + prefs->ldmHashRateLog = ldmHashRateLog; +} + +void FIO_setPatchFromMode(FIO_prefs_t* const prefs, int value) +{ + prefs->patchFromMode = value != 0; +} + +void FIO_setContentSize(FIO_prefs_t* const prefs, int value) +{ + prefs->contentSize = value != 0; +} + +/* FIO_ctx_t functions */ + +void FIO_setHasStdoutOutput(FIO_ctx_t* const fCtx, int value) { + fCtx->hasStdoutOutput = value; +} + +void FIO_setNbFilesTotal(FIO_ctx_t* const fCtx, int value) +{ + fCtx->nbFilesTotal = value; +} + +void FIO_determineHasStdinInput(FIO_ctx_t* const fCtx, const FileNamesTable* const filenames) { + size_t i = 0; + for ( ; i < filenames->tableSize; ++i) { + if (!strcmp(stdinmark, filenames->fileNames[i])) { + fCtx->hasStdinInput = 1; + return; + } + } +} + +/*-************************************* +* Functions +***************************************/ +/** FIO_removeFile() : + * @result : Unlink `fileName`, even if it's read-only */ +static int FIO_removeFile(const char* path) +{ + stat_t statbuf; + if (!UTIL_stat(path, &statbuf)) { + DISPLAYLEVEL(2, "zstd: Failed to stat %s while trying to remove it\n", path); + return 0; + } + if (!UTIL_isRegularFileStat(&statbuf)) { + DISPLAYLEVEL(2, "zstd: Refusing to remove non-regular file %s\n", path); + return 0; + } +#if defined(_WIN32) || defined(WIN32) + /* windows doesn't allow remove read-only files, + * so try to make it writable first */ + if (!(statbuf.st_mode & _S_IWRITE)) { + UTIL_chmod(path, &statbuf, _S_IWRITE); + } +#endif + return remove(path); +} + +/** FIO_openSrcFile() : + * condition : `srcFileName` must be non-NULL. `prefs` may be NULL. + * @result : FILE* to `srcFileName`, or NULL if it fails */ +static FILE* FIO_openSrcFile(const FIO_prefs_t* const prefs, const char* srcFileName) +{ + stat_t statbuf; + int allowBlockDevices = prefs != NULL ? prefs->allowBlockDevices : 0; + assert(srcFileName != NULL); + if (!strcmp (srcFileName, stdinmark)) { + DISPLAYLEVEL(4,"Using stdin for input \n"); + SET_BINARY_MODE(stdin); + return stdin; + } + + if (!UTIL_stat(srcFileName, &statbuf)) { + DISPLAYLEVEL(1, "zstd: can't stat %s : %s -- ignored \n", + srcFileName, strerror(errno)); + return NULL; + } + + if (!UTIL_isRegularFileStat(&statbuf) + && !UTIL_isFIFOStat(&statbuf) + && !(allowBlockDevices && UTIL_isBlockDevStat(&statbuf)) + ) { + DISPLAYLEVEL(1, "zstd: %s is not a regular file -- ignored \n", + srcFileName); + return NULL; + } + + { FILE* const f = fopen(srcFileName, "rb"); + if (f == NULL) + DISPLAYLEVEL(1, "zstd: %s: %s \n", srcFileName, strerror(errno)); + return f; + } +} + +/** FIO_openDstFile() : + * condition : `dstFileName` must be non-NULL. + * @result : FILE* to `dstFileName`, or NULL if it fails */ +static FILE* +FIO_openDstFile(FIO_ctx_t* fCtx, FIO_prefs_t* const prefs, + const char* srcFileName, const char* dstFileName, + const int mode) +{ + if (prefs->testMode) return NULL; /* do not open file in test mode */ + + assert(dstFileName != NULL); + if (!strcmp (dstFileName, stdoutmark)) { + DISPLAYLEVEL(4,"Using stdout for output \n"); + SET_BINARY_MODE(stdout); + if (prefs->sparseFileSupport == 1) { + prefs->sparseFileSupport = 0; + DISPLAYLEVEL(4, "Sparse File Support is automatically disabled on stdout ; try --sparse \n"); + } + return stdout; + } + + /* ensure dst is not the same as src */ + if (srcFileName != NULL && UTIL_isSameFile(srcFileName, dstFileName)) { + DISPLAYLEVEL(1, "zstd: Refusing to open an output file which will overwrite the input file \n"); + return NULL; + } + + if (prefs->sparseFileSupport == 1) { + prefs->sparseFileSupport = ZSTD_SPARSE_DEFAULT; + } + + if (UTIL_isRegularFile(dstFileName)) { + /* Check if destination file already exists */ +#if !defined(_WIN32) + /* this test does not work on Windows : + * `NUL` and `nul` are detected as regular files */ + if (!strcmp(dstFileName, nulmark)) { + EXM_THROW(40, "%s is unexpectedly categorized as a regular file", + dstFileName); + } +#endif + if (!prefs->overwrite) { + if (g_display_prefs.displayLevel <= 1) { + /* No interaction possible */ + DISPLAY("zstd: %s already exists; not overwritten \n", + dstFileName); + return NULL; + } + DISPLAY("zstd: %s already exists; ", dstFileName); + if (UTIL_requireUserConfirmation("overwrite (y/n) ? ", "Not overwritten \n", "yY", fCtx->hasStdinInput)) + return NULL; + } + /* need to unlink */ + FIO_removeFile(dstFileName); + } + + { +#if defined(_WIN32) + /* Windows requires opening the file as a "binary" file to avoid + * mangling. This macro doesn't exist on unix. */ + const int openflags = O_WRONLY|O_CREAT|O_TRUNC|O_BINARY; + const int fd = _open(dstFileName, openflags, mode); + FILE* f = NULL; + if (fd != -1) { + f = _fdopen(fd, "wb"); + } +#else + const int openflags = O_WRONLY|O_CREAT|O_TRUNC; + const int fd = open(dstFileName, openflags, mode); + FILE* f = NULL; + if (fd != -1) { + f = fdopen(fd, "wb"); + } +#endif + if (f == NULL) { + DISPLAYLEVEL(1, "zstd: %s: %s\n", dstFileName, strerror(errno)); + } + return f; + } +} + +/*! FIO_createDictBuffer() : + * creates a buffer, pointed by `*bufferPtr`, + * loads `filename` content into it, up to DICTSIZE_MAX bytes. + * @return : loaded size + * if fileName==NULL, returns 0 and a NULL pointer + */ +static size_t FIO_createDictBuffer(void** bufferPtr, const char* fileName, FIO_prefs_t* const prefs) +{ + FILE* fileHandle; + U64 fileSize; stat_t statbuf; - - assert(bufferPtr != NULL); - *bufferPtr = NULL; - if (fileName == NULL) return 0; - - DISPLAYLEVEL(4,"Loading %s as dictionary \n", fileName); + + assert(bufferPtr != NULL); + *bufferPtr = NULL; + if (fileName == NULL) return 0; + + DISPLAYLEVEL(4,"Loading %s as dictionary \n", fileName); if (!UTIL_stat(fileName, &statbuf)) { EXM_THROW(31, "Stat failed on dictionary file %s: %s", fileName, strerror(errno)); @@ -748,627 +748,627 @@ static size_t FIO_createDictBuffer(void** bufferPtr, const char* fileName, FIO_p EXM_THROW(32, "Dictionary %s must be a regular file.", fileName); } - fileHandle = fopen(fileName, "rb"); - + fileHandle = fopen(fileName, "rb"); + if (fileHandle == NULL) { EXM_THROW(33, "Couldn't open dictionary %s: %s", fileName, strerror(errno)); } fileSize = UTIL_getFileSizeStat(&statbuf); - { - size_t const dictSizeMax = prefs->patchFromMode ? prefs->memLimit : DICTSIZE_MAX; - if (fileSize > dictSizeMax) { + { + size_t const dictSizeMax = prefs->patchFromMode ? prefs->memLimit : DICTSIZE_MAX; + if (fileSize > dictSizeMax) { EXM_THROW(34, "Dictionary file %s is too large (> %u bytes)", - fileName, (unsigned)dictSizeMax); /* avoid extreme cases */ - } - } - *bufferPtr = malloc((size_t)fileSize); - if (*bufferPtr==NULL) EXM_THROW(34, "%s", strerror(errno)); - { size_t const readSize = fread(*bufferPtr, 1, (size_t)fileSize, fileHandle); + fileName, (unsigned)dictSizeMax); /* avoid extreme cases */ + } + } + *bufferPtr = malloc((size_t)fileSize); + if (*bufferPtr==NULL) EXM_THROW(34, "%s", strerror(errno)); + { size_t const readSize = fread(*bufferPtr, 1, (size_t)fileSize, fileHandle); if (readSize != fileSize) { - EXM_THROW(35, "Error reading dictionary file %s : %s", - fileName, strerror(errno)); - } - } - fclose(fileHandle); - return (size_t)fileSize; -} - - - -/* FIO_checkFilenameCollisions() : - * Checks for and warns if there are any files that would have the same output path - */ -int FIO_checkFilenameCollisions(const char** filenameTable, unsigned nbFiles) { - const char **filenameTableSorted, *prevElem, *filename; - unsigned u; - - filenameTableSorted = (const char**) malloc(sizeof(char*) * nbFiles); - if (!filenameTableSorted) { - DISPLAY("Unable to malloc new str array, not checking for name collisions\n"); - return 1; - } - - for (u = 0; u < nbFiles; ++u) { - filename = strrchr(filenameTable[u], PATH_SEP); - if (filename == NULL) { - filenameTableSorted[u] = filenameTable[u]; - } else { - filenameTableSorted[u] = filename+1; + EXM_THROW(35, "Error reading dictionary file %s : %s", + fileName, strerror(errno)); } - } - - qsort((void*)filenameTableSorted, nbFiles, sizeof(char*), UTIL_compareStr); - prevElem = filenameTableSorted[0]; - for (u = 1; u < nbFiles; ++u) { - if (strcmp(prevElem, filenameTableSorted[u]) == 0) { - DISPLAY("WARNING: Two files have same filename: %s\n", prevElem); - } - prevElem = filenameTableSorted[u]; - } - - free((void*)filenameTableSorted); - return 0; -} - -static const char* -extractFilename(const char* path, char separator) -{ - const char* search = strrchr(path, separator); - if (search == NULL) return path; - return search+1; -} - -/* FIO_createFilename_fromOutDir() : - * Takes a source file name and specified output directory, and - * allocates memory for and returns a pointer to final path. - * This function never returns an error (it may abort() in case of pb) - */ -static char* -FIO_createFilename_fromOutDir(const char* path, const char* outDirName, const size_t suffixLen) -{ - const char* filenameStart; - char separator; - char* result; - -#if defined(_MSC_VER) || defined(__MINGW32__) || defined (__MSVCRT__) /* windows support */ - separator = '\\'; -#else - separator = '/'; -#endif - - filenameStart = extractFilename(path, separator); -#if defined(_MSC_VER) || defined(__MINGW32__) || defined (__MSVCRT__) /* windows support */ - filenameStart = extractFilename(filenameStart, '/'); /* sometimes, '/' separator is also used on Windows (mingw+msys2) */ -#endif - - result = (char*) calloc(1, strlen(outDirName) + 1 + strlen(filenameStart) + suffixLen + 1); - if (!result) { - EXM_THROW(30, "zstd: FIO_createFilename_fromOutDir: %s", strerror(errno)); - } - - memcpy(result, outDirName, strlen(outDirName)); - if (outDirName[strlen(outDirName)-1] == separator) { - memcpy(result + strlen(outDirName), filenameStart, strlen(filenameStart)); - } else { - memcpy(result + strlen(outDirName), &separator, 1); - memcpy(result + strlen(outDirName) + 1, filenameStart, strlen(filenameStart)); - } - - return result; -} - -/* FIO_highbit64() : - * gives position of highest bit. - * note : only works for v > 0 ! - */ -static unsigned FIO_highbit64(unsigned long long v) -{ - unsigned count = 0; - assert(v != 0); - v >>= 1; - while (v) { v >>= 1; count++; } - return count; -} - -static void FIO_adjustMemLimitForPatchFromMode(FIO_prefs_t* const prefs, - unsigned long long const dictSize, - unsigned long long const maxSrcFileSize) -{ - unsigned long long maxSize = MAX(prefs->memLimit, MAX(dictSize, maxSrcFileSize)); - unsigned const maxWindowSize = (1U << ZSTD_WINDOWLOG_MAX); - if (maxSize == UTIL_FILESIZE_UNKNOWN) - EXM_THROW(42, "Using --patch-from with stdin requires --stream-size"); - assert(maxSize != UTIL_FILESIZE_UNKNOWN); - if (maxSize > maxWindowSize) - EXM_THROW(42, "Can't handle files larger than %u GB\n", maxWindowSize/(1 GB)); - FIO_setMemLimit(prefs, (unsigned)maxSize); -} - -/* FIO_removeMultiFilesWarning() : - * Returns 1 if the console should abort, 0 if console should proceed. - * This function handles logic when processing multiple files with -o, displaying the appropriate warnings/prompts. - * - * If -f is specified, or there is just 1 file, zstd will always proceed as usual. - * If --rm is specified, there will be a prompt asking for user confirmation. - * If -f is specified with --rm, zstd will proceed as usual - * If -q is specified with --rm, zstd will abort pre-emptively - * If neither flag is specified, zstd will prompt the user for confirmation to proceed. - * If --rm is not specified, then zstd will print a warning to the user (which can be silenced with -q). - * However, if the output is stdout, we will always abort rather than displaying the warning prompt. - */ -static int FIO_removeMultiFilesWarning(FIO_ctx_t* const fCtx, const FIO_prefs_t* const prefs, const char* outFileName, int displayLevelCutoff) -{ - int error = 0; - if (fCtx->nbFilesTotal > 1 && !prefs->overwrite) { - if (g_display_prefs.displayLevel <= displayLevelCutoff) { - if (prefs->removeSrcFile) { + } + fclose(fileHandle); + return (size_t)fileSize; +} + + + +/* FIO_checkFilenameCollisions() : + * Checks for and warns if there are any files that would have the same output path + */ +int FIO_checkFilenameCollisions(const char** filenameTable, unsigned nbFiles) { + const char **filenameTableSorted, *prevElem, *filename; + unsigned u; + + filenameTableSorted = (const char**) malloc(sizeof(char*) * nbFiles); + if (!filenameTableSorted) { + DISPLAY("Unable to malloc new str array, not checking for name collisions\n"); + return 1; + } + + for (u = 0; u < nbFiles; ++u) { + filename = strrchr(filenameTable[u], PATH_SEP); + if (filename == NULL) { + filenameTableSorted[u] = filenameTable[u]; + } else { + filenameTableSorted[u] = filename+1; + } + } + + qsort((void*)filenameTableSorted, nbFiles, sizeof(char*), UTIL_compareStr); + prevElem = filenameTableSorted[0]; + for (u = 1; u < nbFiles; ++u) { + if (strcmp(prevElem, filenameTableSorted[u]) == 0) { + DISPLAY("WARNING: Two files have same filename: %s\n", prevElem); + } + prevElem = filenameTableSorted[u]; + } + + free((void*)filenameTableSorted); + return 0; +} + +static const char* +extractFilename(const char* path, char separator) +{ + const char* search = strrchr(path, separator); + if (search == NULL) return path; + return search+1; +} + +/* FIO_createFilename_fromOutDir() : + * Takes a source file name and specified output directory, and + * allocates memory for and returns a pointer to final path. + * This function never returns an error (it may abort() in case of pb) + */ +static char* +FIO_createFilename_fromOutDir(const char* path, const char* outDirName, const size_t suffixLen) +{ + const char* filenameStart; + char separator; + char* result; + +#if defined(_MSC_VER) || defined(__MINGW32__) || defined (__MSVCRT__) /* windows support */ + separator = '\\'; +#else + separator = '/'; +#endif + + filenameStart = extractFilename(path, separator); +#if defined(_MSC_VER) || defined(__MINGW32__) || defined (__MSVCRT__) /* windows support */ + filenameStart = extractFilename(filenameStart, '/'); /* sometimes, '/' separator is also used on Windows (mingw+msys2) */ +#endif + + result = (char*) calloc(1, strlen(outDirName) + 1 + strlen(filenameStart) + suffixLen + 1); + if (!result) { + EXM_THROW(30, "zstd: FIO_createFilename_fromOutDir: %s", strerror(errno)); + } + + memcpy(result, outDirName, strlen(outDirName)); + if (outDirName[strlen(outDirName)-1] == separator) { + memcpy(result + strlen(outDirName), filenameStart, strlen(filenameStart)); + } else { + memcpy(result + strlen(outDirName), &separator, 1); + memcpy(result + strlen(outDirName) + 1, filenameStart, strlen(filenameStart)); + } + + return result; +} + +/* FIO_highbit64() : + * gives position of highest bit. + * note : only works for v > 0 ! + */ +static unsigned FIO_highbit64(unsigned long long v) +{ + unsigned count = 0; + assert(v != 0); + v >>= 1; + while (v) { v >>= 1; count++; } + return count; +} + +static void FIO_adjustMemLimitForPatchFromMode(FIO_prefs_t* const prefs, + unsigned long long const dictSize, + unsigned long long const maxSrcFileSize) +{ + unsigned long long maxSize = MAX(prefs->memLimit, MAX(dictSize, maxSrcFileSize)); + unsigned const maxWindowSize = (1U << ZSTD_WINDOWLOG_MAX); + if (maxSize == UTIL_FILESIZE_UNKNOWN) + EXM_THROW(42, "Using --patch-from with stdin requires --stream-size"); + assert(maxSize != UTIL_FILESIZE_UNKNOWN); + if (maxSize > maxWindowSize) + EXM_THROW(42, "Can't handle files larger than %u GB\n", maxWindowSize/(1 GB)); + FIO_setMemLimit(prefs, (unsigned)maxSize); +} + +/* FIO_removeMultiFilesWarning() : + * Returns 1 if the console should abort, 0 if console should proceed. + * This function handles logic when processing multiple files with -o, displaying the appropriate warnings/prompts. + * + * If -f is specified, or there is just 1 file, zstd will always proceed as usual. + * If --rm is specified, there will be a prompt asking for user confirmation. + * If -f is specified with --rm, zstd will proceed as usual + * If -q is specified with --rm, zstd will abort pre-emptively + * If neither flag is specified, zstd will prompt the user for confirmation to proceed. + * If --rm is not specified, then zstd will print a warning to the user (which can be silenced with -q). + * However, if the output is stdout, we will always abort rather than displaying the warning prompt. + */ +static int FIO_removeMultiFilesWarning(FIO_ctx_t* const fCtx, const FIO_prefs_t* const prefs, const char* outFileName, int displayLevelCutoff) +{ + int error = 0; + if (fCtx->nbFilesTotal > 1 && !prefs->overwrite) { + if (g_display_prefs.displayLevel <= displayLevelCutoff) { + if (prefs->removeSrcFile) { DISPLAYLEVEL(1, "zstd: Aborting... not deleting files and processing into dst: %s\n", outFileName); - error = 1; - } - } else { - if (!strcmp(outFileName, stdoutmark)) { + error = 1; + } + } else { + if (!strcmp(outFileName, stdoutmark)) { DISPLAYLEVEL(2, "zstd: WARNING: all input files will be processed and concatenated into stdout. \n"); - } else { + } else { DISPLAYLEVEL(2, "zstd: WARNING: all input files will be processed and concatenated into a single output file: %s \n", outFileName); - } + } DISPLAYLEVEL(2, "The concatenated output CANNOT regenerate the original directory tree. \n") - if (prefs->removeSrcFile) { - if (fCtx->hasStdoutOutput) { + if (prefs->removeSrcFile) { + if (fCtx->hasStdoutOutput) { DISPLAYLEVEL(1, "Aborting. Use -f if you really want to delete the files and output to stdout\n"); - error = 1; - } else { - error = g_display_prefs.displayLevel > displayLevelCutoff && UTIL_requireUserConfirmation("This is a destructive operation. Proceed? (y/n): ", "Aborting...", "yY", fCtx->hasStdinInput); - } - } - } - } - return error; -} - -#ifndef ZSTD_NOCOMPRESS - -/* ********************************************************************** - * Compression - ************************************************************************/ -typedef struct { - FILE* srcFile; - FILE* dstFile; - void* srcBuffer; - size_t srcBufferSize; - void* dstBuffer; - size_t dstBufferSize; - void* dictBuffer; - size_t dictBufferSize; - const char* dictFileName; - ZSTD_CStream* cctx; -} cRess_t; - -/** ZSTD_cycleLog() : - * condition for correct operation : hashLog > 1 */ -static U32 ZSTD_cycleLog(U32 hashLog, ZSTD_strategy strat) -{ - U32 const btScale = ((U32)strat >= (U32)ZSTD_btlazy2); - assert(hashLog > 1); - return hashLog - btScale; -} - -static void FIO_adjustParamsForPatchFromMode(FIO_prefs_t* const prefs, - ZSTD_compressionParameters* comprParams, - unsigned long long const dictSize, - unsigned long long const maxSrcFileSize, - int cLevel) -{ - unsigned const fileWindowLog = FIO_highbit64(maxSrcFileSize) + 1; - ZSTD_compressionParameters const cParams = ZSTD_getCParams(cLevel, (size_t)maxSrcFileSize, (size_t)dictSize); - FIO_adjustMemLimitForPatchFromMode(prefs, dictSize, maxSrcFileSize); - if (fileWindowLog > ZSTD_WINDOWLOG_MAX) - DISPLAYLEVEL(1, "Max window log exceeded by file (compression ratio will suffer)\n"); - comprParams->windowLog = MAX(ZSTD_WINDOWLOG_MIN, MIN(ZSTD_WINDOWLOG_MAX, fileWindowLog)); - if (fileWindowLog > ZSTD_cycleLog(cParams.chainLog, cParams.strategy)) { - if (!prefs->ldmFlag) - DISPLAYLEVEL(1, "long mode automatically triggered\n"); - FIO_setLdmFlag(prefs, 1); - } - if (cParams.strategy >= ZSTD_btopt) { - DISPLAYLEVEL(1, "[Optimal parser notes] Consider the following to improve patch size at the cost of speed:\n"); - DISPLAYLEVEL(1, "- Use --single-thread mode in the zstd cli\n"); - DISPLAYLEVEL(1, "- Set a larger targetLength (eg. --zstd=targetLength=4096)\n"); - DISPLAYLEVEL(1, "- Set a larger chainLog (eg. --zstd=chainLog=%u)\n", ZSTD_CHAINLOG_MAX); + error = 1; + } else { + error = g_display_prefs.displayLevel > displayLevelCutoff && UTIL_requireUserConfirmation("This is a destructive operation. Proceed? (y/n): ", "Aborting...", "yY", fCtx->hasStdinInput); + } + } + } + } + return error; +} + +#ifndef ZSTD_NOCOMPRESS + +/* ********************************************************************** + * Compression + ************************************************************************/ +typedef struct { + FILE* srcFile; + FILE* dstFile; + void* srcBuffer; + size_t srcBufferSize; + void* dstBuffer; + size_t dstBufferSize; + void* dictBuffer; + size_t dictBufferSize; + const char* dictFileName; + ZSTD_CStream* cctx; +} cRess_t; + +/** ZSTD_cycleLog() : + * condition for correct operation : hashLog > 1 */ +static U32 ZSTD_cycleLog(U32 hashLog, ZSTD_strategy strat) +{ + U32 const btScale = ((U32)strat >= (U32)ZSTD_btlazy2); + assert(hashLog > 1); + return hashLog - btScale; +} + +static void FIO_adjustParamsForPatchFromMode(FIO_prefs_t* const prefs, + ZSTD_compressionParameters* comprParams, + unsigned long long const dictSize, + unsigned long long const maxSrcFileSize, + int cLevel) +{ + unsigned const fileWindowLog = FIO_highbit64(maxSrcFileSize) + 1; + ZSTD_compressionParameters const cParams = ZSTD_getCParams(cLevel, (size_t)maxSrcFileSize, (size_t)dictSize); + FIO_adjustMemLimitForPatchFromMode(prefs, dictSize, maxSrcFileSize); + if (fileWindowLog > ZSTD_WINDOWLOG_MAX) + DISPLAYLEVEL(1, "Max window log exceeded by file (compression ratio will suffer)\n"); + comprParams->windowLog = MAX(ZSTD_WINDOWLOG_MIN, MIN(ZSTD_WINDOWLOG_MAX, fileWindowLog)); + if (fileWindowLog > ZSTD_cycleLog(cParams.chainLog, cParams.strategy)) { + if (!prefs->ldmFlag) + DISPLAYLEVEL(1, "long mode automatically triggered\n"); + FIO_setLdmFlag(prefs, 1); + } + if (cParams.strategy >= ZSTD_btopt) { + DISPLAYLEVEL(1, "[Optimal parser notes] Consider the following to improve patch size at the cost of speed:\n"); + DISPLAYLEVEL(1, "- Use --single-thread mode in the zstd cli\n"); + DISPLAYLEVEL(1, "- Set a larger targetLength (eg. --zstd=targetLength=4096)\n"); + DISPLAYLEVEL(1, "- Set a larger chainLog (eg. --zstd=chainLog=%u)\n", ZSTD_CHAINLOG_MAX); DISPLAYLEVEL(1, "Also consider playing around with searchLog and hashLog\n"); - } -} - -static cRess_t FIO_createCResources(FIO_prefs_t* const prefs, - const char* dictFileName, unsigned long long const maxSrcFileSize, - int cLevel, ZSTD_compressionParameters comprParams) { - cRess_t ress; - memset(&ress, 0, sizeof(ress)); - - DISPLAYLEVEL(6, "FIO_createCResources \n"); - ress.cctx = ZSTD_createCCtx(); - if (ress.cctx == NULL) - EXM_THROW(30, "allocation error (%s): can't create ZSTD_CCtx", - strerror(errno)); - ress.srcBufferSize = ZSTD_CStreamInSize(); - ress.srcBuffer = malloc(ress.srcBufferSize); - ress.dstBufferSize = ZSTD_CStreamOutSize(); - - /* need to update memLimit before calling createDictBuffer - * because of memLimit check inside it */ - if (prefs->patchFromMode) { - unsigned long long const ssSize = (unsigned long long)prefs->streamSrcSize; - FIO_adjustParamsForPatchFromMode(prefs, &comprParams, UTIL_getFileSize(dictFileName), ssSize > 0 ? ssSize : maxSrcFileSize, cLevel); - } - ress.dstBuffer = malloc(ress.dstBufferSize); - ress.dictBufferSize = FIO_createDictBuffer(&ress.dictBuffer, dictFileName, prefs); /* works with dictFileName==NULL */ - if (!ress.srcBuffer || !ress.dstBuffer) - EXM_THROW(31, "allocation error : not enough memory"); - - /* Advanced parameters, including dictionary */ - if (dictFileName && (ress.dictBuffer==NULL)) - EXM_THROW(32, "allocation error : can't create dictBuffer"); - ress.dictFileName = dictFileName; - - if (prefs->adaptiveMode && !prefs->ldmFlag && !comprParams.windowLog) - comprParams.windowLog = ADAPT_WINDOWLOG_DEFAULT; - - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_contentSizeFlag, prefs->contentSize) ); /* always enable content size when available (note: supposed to be default) */ - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_dictIDFlag, prefs->dictIDFlag) ); - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_checksumFlag, prefs->checksumFlag) ); - /* compression level */ - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_compressionLevel, cLevel) ); - /* max compressed block size */ - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_targetCBlockSize, (int)prefs->targetCBlockSize) ); - /* source size hint */ - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_srcSizeHint, (int)prefs->srcSizeHint) ); - /* long distance matching */ - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_enableLongDistanceMatching, prefs->ldmFlag) ); - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_ldmHashLog, prefs->ldmHashLog) ); - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_ldmMinMatch, prefs->ldmMinMatch) ); - if (prefs->ldmBucketSizeLog != FIO_LDM_PARAM_NOTSET) { - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_ldmBucketSizeLog, prefs->ldmBucketSizeLog) ); - } - if (prefs->ldmHashRateLog != FIO_LDM_PARAM_NOTSET) { - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_ldmHashRateLog, prefs->ldmHashRateLog) ); - } - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_useRowMatchFinder, prefs->useRowMatchFinder)); - /* compression parameters */ - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_windowLog, (int)comprParams.windowLog) ); - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_chainLog, (int)comprParams.chainLog) ); - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_hashLog, (int)comprParams.hashLog) ); - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_searchLog, (int)comprParams.searchLog) ); - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_minMatch, (int)comprParams.minMatch) ); - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_targetLength, (int)comprParams.targetLength) ); - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_strategy, (int)comprParams.strategy) ); - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_literalCompressionMode, (int)prefs->literalCompressionMode) ); - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_enableDedicatedDictSearch, 1) ); - /* multi-threading */ -#ifdef ZSTD_MULTITHREAD - DISPLAYLEVEL(5,"set nb workers = %u \n", prefs->nbWorkers); - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_nbWorkers, prefs->nbWorkers) ); - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_jobSize, prefs->blockSize) ); - if (prefs->overlapLog != FIO_OVERLAP_LOG_NOTSET) { - DISPLAYLEVEL(3,"set overlapLog = %u \n", prefs->overlapLog); - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_overlapLog, prefs->overlapLog) ); - } - CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_rsyncable, prefs->rsyncable) ); -#endif - /* dictionary */ - if (prefs->patchFromMode) { - CHECK( ZSTD_CCtx_refPrefix(ress.cctx, ress.dictBuffer, ress.dictBufferSize) ); - } else { - CHECK( ZSTD_CCtx_loadDictionary(ress.cctx, ress.dictBuffer, ress.dictBufferSize) ); - } - - return ress; -} - -static void FIO_freeCResources(const cRess_t* const ress) -{ - free(ress->srcBuffer); - free(ress->dstBuffer); - free(ress->dictBuffer); - ZSTD_freeCStream(ress->cctx); /* never fails */ -} - - -#ifdef ZSTD_GZCOMPRESS -static unsigned long long -FIO_compressGzFrame(const cRess_t* ress, /* buffers & handlers are used, but not changed */ - const char* srcFileName, U64 const srcFileSize, - int compressionLevel, U64* readsize) -{ - unsigned long long inFileSize = 0, outFileSize = 0; - z_stream strm; - - if (compressionLevel > Z_BEST_COMPRESSION) - compressionLevel = Z_BEST_COMPRESSION; - - strm.zalloc = Z_NULL; - strm.zfree = Z_NULL; - strm.opaque = Z_NULL; - - { int const ret = deflateInit2(&strm, compressionLevel, Z_DEFLATED, - 15 /* maxWindowLogSize */ + 16 /* gzip only */, - 8, Z_DEFAULT_STRATEGY); /* see http://www.zlib.net/manual.html */ - if (ret != Z_OK) { - EXM_THROW(71, "zstd: %s: deflateInit2 error %d \n", srcFileName, ret); - } } - - strm.next_in = 0; - strm.avail_in = 0; - strm.next_out = (Bytef*)ress->dstBuffer; - strm.avail_out = (uInt)ress->dstBufferSize; - - while (1) { - int ret; - if (strm.avail_in == 0) { - size_t const inSize = fread(ress->srcBuffer, 1, ress->srcBufferSize, ress->srcFile); - if (inSize == 0) break; - inFileSize += inSize; - strm.next_in = (z_const unsigned char*)ress->srcBuffer; - strm.avail_in = (uInt)inSize; - } - ret = deflate(&strm, Z_NO_FLUSH); - if (ret != Z_OK) - EXM_THROW(72, "zstd: %s: deflate error %d \n", srcFileName, ret); - { size_t const cSize = ress->dstBufferSize - strm.avail_out; - if (cSize) { - if (fwrite(ress->dstBuffer, 1, cSize, ress->dstFile) != cSize) - EXM_THROW(73, "Write error : cannot write to output file : %s ", strerror(errno)); - outFileSize += cSize; - strm.next_out = (Bytef*)ress->dstBuffer; - strm.avail_out = (uInt)ress->dstBufferSize; - } } - if (srcFileSize == UTIL_FILESIZE_UNKNOWN) { - DISPLAYUPDATE(2, "\rRead : %u MB ==> %.2f%% ", - (unsigned)(inFileSize>>20), - (double)outFileSize/inFileSize*100) - } else { - DISPLAYUPDATE(2, "\rRead : %u / %u MB ==> %.2f%% ", - (unsigned)(inFileSize>>20), (unsigned)(srcFileSize>>20), - (double)outFileSize/inFileSize*100); - } } - - while (1) { - int const ret = deflate(&strm, Z_FINISH); - { size_t const cSize = ress->dstBufferSize - strm.avail_out; - if (cSize) { - if (fwrite(ress->dstBuffer, 1, cSize, ress->dstFile) != cSize) - EXM_THROW(75, "Write error : %s ", strerror(errno)); - outFileSize += cSize; - strm.next_out = (Bytef*)ress->dstBuffer; - strm.avail_out = (uInt)ress->dstBufferSize; - } } - if (ret == Z_STREAM_END) break; - if (ret != Z_BUF_ERROR) - EXM_THROW(77, "zstd: %s: deflate error %d \n", srcFileName, ret); - } - - { int const ret = deflateEnd(&strm); - if (ret != Z_OK) { - EXM_THROW(79, "zstd: %s: deflateEnd error %d \n", srcFileName, ret); - } } - *readsize = inFileSize; - return outFileSize; -} -#endif - - -#ifdef ZSTD_LZMACOMPRESS -static unsigned long long -FIO_compressLzmaFrame(cRess_t* ress, - const char* srcFileName, U64 const srcFileSize, - int compressionLevel, U64* readsize, int plain_lzma) -{ - unsigned long long inFileSize = 0, outFileSize = 0; - lzma_stream strm = LZMA_STREAM_INIT; - lzma_action action = LZMA_RUN; - lzma_ret ret; - - if (compressionLevel < 0) compressionLevel = 0; - if (compressionLevel > 9) compressionLevel = 9; - - if (plain_lzma) { - lzma_options_lzma opt_lzma; - if (lzma_lzma_preset(&opt_lzma, compressionLevel)) - EXM_THROW(81, "zstd: %s: lzma_lzma_preset error", srcFileName); - ret = lzma_alone_encoder(&strm, &opt_lzma); /* LZMA */ - if (ret != LZMA_OK) - EXM_THROW(82, "zstd: %s: lzma_alone_encoder error %d", srcFileName, ret); - } else { - ret = lzma_easy_encoder(&strm, compressionLevel, LZMA_CHECK_CRC64); /* XZ */ - if (ret != LZMA_OK) - EXM_THROW(83, "zstd: %s: lzma_easy_encoder error %d", srcFileName, ret); - } - - strm.next_in = 0; - strm.avail_in = 0; - strm.next_out = (BYTE*)ress->dstBuffer; - strm.avail_out = ress->dstBufferSize; - - while (1) { - if (strm.avail_in == 0) { - size_t const inSize = fread(ress->srcBuffer, 1, ress->srcBufferSize, ress->srcFile); - if (inSize == 0) action = LZMA_FINISH; - inFileSize += inSize; - strm.next_in = (BYTE const*)ress->srcBuffer; - strm.avail_in = inSize; - } - - ret = lzma_code(&strm, action); - - if (ret != LZMA_OK && ret != LZMA_STREAM_END) - EXM_THROW(84, "zstd: %s: lzma_code encoding error %d", srcFileName, ret); - { size_t const compBytes = ress->dstBufferSize - strm.avail_out; - if (compBytes) { - if (fwrite(ress->dstBuffer, 1, compBytes, ress->dstFile) != compBytes) - EXM_THROW(85, "Write error : %s", strerror(errno)); - outFileSize += compBytes; - strm.next_out = (BYTE*)ress->dstBuffer; - strm.avail_out = ress->dstBufferSize; - } } - if (srcFileSize == UTIL_FILESIZE_UNKNOWN) - DISPLAYUPDATE(2, "\rRead : %u MB ==> %.2f%%", - (unsigned)(inFileSize>>20), - (double)outFileSize/inFileSize*100) - else - DISPLAYUPDATE(2, "\rRead : %u / %u MB ==> %.2f%%", - (unsigned)(inFileSize>>20), (unsigned)(srcFileSize>>20), - (double)outFileSize/inFileSize*100); - if (ret == LZMA_STREAM_END) break; - } - - lzma_end(&strm); - *readsize = inFileSize; - - return outFileSize; -} -#endif - -#ifdef ZSTD_LZ4COMPRESS - -#if LZ4_VERSION_NUMBER <= 10600 -#define LZ4F_blockLinked blockLinked -#define LZ4F_max64KB max64KB -#endif - -static int FIO_LZ4_GetBlockSize_FromBlockId (int id) { return (1 << (8 + (2 * id))); } - -static unsigned long long -FIO_compressLz4Frame(cRess_t* ress, - const char* srcFileName, U64 const srcFileSize, - int compressionLevel, int checksumFlag, - U64* readsize) -{ - const size_t blockSize = FIO_LZ4_GetBlockSize_FromBlockId(LZ4F_max64KB); - unsigned long long inFileSize = 0, outFileSize = 0; - - LZ4F_preferences_t prefs; - LZ4F_compressionContext_t ctx; - - LZ4F_errorCode_t const errorCode = LZ4F_createCompressionContext(&ctx, LZ4F_VERSION); - if (LZ4F_isError(errorCode)) - EXM_THROW(31, "zstd: failed to create lz4 compression context"); - - memset(&prefs, 0, sizeof(prefs)); - - assert(blockSize <= ress->srcBufferSize); - - prefs.autoFlush = 1; - prefs.compressionLevel = compressionLevel; - prefs.frameInfo.blockMode = LZ4F_blockLinked; - prefs.frameInfo.blockSizeID = LZ4F_max64KB; - prefs.frameInfo.contentChecksumFlag = (contentChecksum_t)checksumFlag; -#if LZ4_VERSION_NUMBER >= 10600 - prefs.frameInfo.contentSize = (srcFileSize==UTIL_FILESIZE_UNKNOWN) ? 0 : srcFileSize; -#endif - assert(LZ4F_compressBound(blockSize, &prefs) <= ress->dstBufferSize); - - { - size_t readSize; - size_t headerSize = LZ4F_compressBegin(ctx, ress->dstBuffer, ress->dstBufferSize, &prefs); - if (LZ4F_isError(headerSize)) - EXM_THROW(33, "File header generation failed : %s", - LZ4F_getErrorName(headerSize)); - if (fwrite(ress->dstBuffer, 1, headerSize, ress->dstFile) != headerSize) - EXM_THROW(34, "Write error : %s (cannot write header)", strerror(errno)); - outFileSize += headerSize; - - /* Read first block */ - readSize = fread(ress->srcBuffer, (size_t)1, (size_t)blockSize, ress->srcFile); - inFileSize += readSize; - - /* Main Loop */ - while (readSize>0) { - size_t const outSize = LZ4F_compressUpdate(ctx, - ress->dstBuffer, ress->dstBufferSize, - ress->srcBuffer, readSize, NULL); - if (LZ4F_isError(outSize)) - EXM_THROW(35, "zstd: %s: lz4 compression failed : %s", - srcFileName, LZ4F_getErrorName(outSize)); - outFileSize += outSize; - if (srcFileSize == UTIL_FILESIZE_UNKNOWN) { - DISPLAYUPDATE(2, "\rRead : %u MB ==> %.2f%%", - (unsigned)(inFileSize>>20), - (double)outFileSize/inFileSize*100) - } else { - DISPLAYUPDATE(2, "\rRead : %u / %u MB ==> %.2f%%", - (unsigned)(inFileSize>>20), (unsigned)(srcFileSize>>20), - (double)outFileSize/inFileSize*100); - } - - /* Write Block */ - { size_t const sizeCheck = fwrite(ress->dstBuffer, 1, outSize, ress->dstFile); - if (sizeCheck != outSize) - EXM_THROW(36, "Write error : %s", strerror(errno)); - } - - /* Read next block */ - readSize = fread(ress->srcBuffer, (size_t)1, (size_t)blockSize, ress->srcFile); - inFileSize += readSize; - } - if (ferror(ress->srcFile)) EXM_THROW(37, "Error reading %s ", srcFileName); - - /* End of Stream mark */ - headerSize = LZ4F_compressEnd(ctx, ress->dstBuffer, ress->dstBufferSize, NULL); - if (LZ4F_isError(headerSize)) - EXM_THROW(38, "zstd: %s: lz4 end of file generation failed : %s", - srcFileName, LZ4F_getErrorName(headerSize)); - - { size_t const sizeCheck = fwrite(ress->dstBuffer, 1, headerSize, ress->dstFile); - if (sizeCheck != headerSize) - EXM_THROW(39, "Write error : %s (cannot write end of stream)", - strerror(errno)); - } - outFileSize += headerSize; - } - - *readsize = inFileSize; - LZ4F_freeCompressionContext(ctx); - - return outFileSize; -} -#endif - - -static unsigned long long -FIO_compressZstdFrame(FIO_ctx_t* const fCtx, - FIO_prefs_t* const prefs, - const cRess_t* ressPtr, - const char* srcFileName, U64 fileSize, - int compressionLevel, U64* readsize) -{ - cRess_t const ress = *ressPtr; - FILE* const srcFile = ress.srcFile; - FILE* const dstFile = ress.dstFile; - U64 compressedfilesize = 0; - ZSTD_EndDirective directive = ZSTD_e_continue; + } +} + +static cRess_t FIO_createCResources(FIO_prefs_t* const prefs, + const char* dictFileName, unsigned long long const maxSrcFileSize, + int cLevel, ZSTD_compressionParameters comprParams) { + cRess_t ress; + memset(&ress, 0, sizeof(ress)); + + DISPLAYLEVEL(6, "FIO_createCResources \n"); + ress.cctx = ZSTD_createCCtx(); + if (ress.cctx == NULL) + EXM_THROW(30, "allocation error (%s): can't create ZSTD_CCtx", + strerror(errno)); + ress.srcBufferSize = ZSTD_CStreamInSize(); + ress.srcBuffer = malloc(ress.srcBufferSize); + ress.dstBufferSize = ZSTD_CStreamOutSize(); + + /* need to update memLimit before calling createDictBuffer + * because of memLimit check inside it */ + if (prefs->patchFromMode) { + unsigned long long const ssSize = (unsigned long long)prefs->streamSrcSize; + FIO_adjustParamsForPatchFromMode(prefs, &comprParams, UTIL_getFileSize(dictFileName), ssSize > 0 ? ssSize : maxSrcFileSize, cLevel); + } + ress.dstBuffer = malloc(ress.dstBufferSize); + ress.dictBufferSize = FIO_createDictBuffer(&ress.dictBuffer, dictFileName, prefs); /* works with dictFileName==NULL */ + if (!ress.srcBuffer || !ress.dstBuffer) + EXM_THROW(31, "allocation error : not enough memory"); + + /* Advanced parameters, including dictionary */ + if (dictFileName && (ress.dictBuffer==NULL)) + EXM_THROW(32, "allocation error : can't create dictBuffer"); + ress.dictFileName = dictFileName; + + if (prefs->adaptiveMode && !prefs->ldmFlag && !comprParams.windowLog) + comprParams.windowLog = ADAPT_WINDOWLOG_DEFAULT; + + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_contentSizeFlag, prefs->contentSize) ); /* always enable content size when available (note: supposed to be default) */ + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_dictIDFlag, prefs->dictIDFlag) ); + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_checksumFlag, prefs->checksumFlag) ); + /* compression level */ + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_compressionLevel, cLevel) ); + /* max compressed block size */ + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_targetCBlockSize, (int)prefs->targetCBlockSize) ); + /* source size hint */ + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_srcSizeHint, (int)prefs->srcSizeHint) ); + /* long distance matching */ + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_enableLongDistanceMatching, prefs->ldmFlag) ); + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_ldmHashLog, prefs->ldmHashLog) ); + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_ldmMinMatch, prefs->ldmMinMatch) ); + if (prefs->ldmBucketSizeLog != FIO_LDM_PARAM_NOTSET) { + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_ldmBucketSizeLog, prefs->ldmBucketSizeLog) ); + } + if (prefs->ldmHashRateLog != FIO_LDM_PARAM_NOTSET) { + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_ldmHashRateLog, prefs->ldmHashRateLog) ); + } + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_useRowMatchFinder, prefs->useRowMatchFinder)); + /* compression parameters */ + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_windowLog, (int)comprParams.windowLog) ); + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_chainLog, (int)comprParams.chainLog) ); + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_hashLog, (int)comprParams.hashLog) ); + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_searchLog, (int)comprParams.searchLog) ); + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_minMatch, (int)comprParams.minMatch) ); + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_targetLength, (int)comprParams.targetLength) ); + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_strategy, (int)comprParams.strategy) ); + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_literalCompressionMode, (int)prefs->literalCompressionMode) ); + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_enableDedicatedDictSearch, 1) ); + /* multi-threading */ +#ifdef ZSTD_MULTITHREAD + DISPLAYLEVEL(5,"set nb workers = %u \n", prefs->nbWorkers); + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_nbWorkers, prefs->nbWorkers) ); + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_jobSize, prefs->blockSize) ); + if (prefs->overlapLog != FIO_OVERLAP_LOG_NOTSET) { + DISPLAYLEVEL(3,"set overlapLog = %u \n", prefs->overlapLog); + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_overlapLog, prefs->overlapLog) ); + } + CHECK( ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_rsyncable, prefs->rsyncable) ); +#endif + /* dictionary */ + if (prefs->patchFromMode) { + CHECK( ZSTD_CCtx_refPrefix(ress.cctx, ress.dictBuffer, ress.dictBufferSize) ); + } else { + CHECK( ZSTD_CCtx_loadDictionary(ress.cctx, ress.dictBuffer, ress.dictBufferSize) ); + } + + return ress; +} + +static void FIO_freeCResources(const cRess_t* const ress) +{ + free(ress->srcBuffer); + free(ress->dstBuffer); + free(ress->dictBuffer); + ZSTD_freeCStream(ress->cctx); /* never fails */ +} + + +#ifdef ZSTD_GZCOMPRESS +static unsigned long long +FIO_compressGzFrame(const cRess_t* ress, /* buffers & handlers are used, but not changed */ + const char* srcFileName, U64 const srcFileSize, + int compressionLevel, U64* readsize) +{ + unsigned long long inFileSize = 0, outFileSize = 0; + z_stream strm; + + if (compressionLevel > Z_BEST_COMPRESSION) + compressionLevel = Z_BEST_COMPRESSION; + + strm.zalloc = Z_NULL; + strm.zfree = Z_NULL; + strm.opaque = Z_NULL; + + { int const ret = deflateInit2(&strm, compressionLevel, Z_DEFLATED, + 15 /* maxWindowLogSize */ + 16 /* gzip only */, + 8, Z_DEFAULT_STRATEGY); /* see http://www.zlib.net/manual.html */ + if (ret != Z_OK) { + EXM_THROW(71, "zstd: %s: deflateInit2 error %d \n", srcFileName, ret); + } } + + strm.next_in = 0; + strm.avail_in = 0; + strm.next_out = (Bytef*)ress->dstBuffer; + strm.avail_out = (uInt)ress->dstBufferSize; + + while (1) { + int ret; + if (strm.avail_in == 0) { + size_t const inSize = fread(ress->srcBuffer, 1, ress->srcBufferSize, ress->srcFile); + if (inSize == 0) break; + inFileSize += inSize; + strm.next_in = (z_const unsigned char*)ress->srcBuffer; + strm.avail_in = (uInt)inSize; + } + ret = deflate(&strm, Z_NO_FLUSH); + if (ret != Z_OK) + EXM_THROW(72, "zstd: %s: deflate error %d \n", srcFileName, ret); + { size_t const cSize = ress->dstBufferSize - strm.avail_out; + if (cSize) { + if (fwrite(ress->dstBuffer, 1, cSize, ress->dstFile) != cSize) + EXM_THROW(73, "Write error : cannot write to output file : %s ", strerror(errno)); + outFileSize += cSize; + strm.next_out = (Bytef*)ress->dstBuffer; + strm.avail_out = (uInt)ress->dstBufferSize; + } } + if (srcFileSize == UTIL_FILESIZE_UNKNOWN) { + DISPLAYUPDATE(2, "\rRead : %u MB ==> %.2f%% ", + (unsigned)(inFileSize>>20), + (double)outFileSize/inFileSize*100) + } else { + DISPLAYUPDATE(2, "\rRead : %u / %u MB ==> %.2f%% ", + (unsigned)(inFileSize>>20), (unsigned)(srcFileSize>>20), + (double)outFileSize/inFileSize*100); + } } + + while (1) { + int const ret = deflate(&strm, Z_FINISH); + { size_t const cSize = ress->dstBufferSize - strm.avail_out; + if (cSize) { + if (fwrite(ress->dstBuffer, 1, cSize, ress->dstFile) != cSize) + EXM_THROW(75, "Write error : %s ", strerror(errno)); + outFileSize += cSize; + strm.next_out = (Bytef*)ress->dstBuffer; + strm.avail_out = (uInt)ress->dstBufferSize; + } } + if (ret == Z_STREAM_END) break; + if (ret != Z_BUF_ERROR) + EXM_THROW(77, "zstd: %s: deflate error %d \n", srcFileName, ret); + } + + { int const ret = deflateEnd(&strm); + if (ret != Z_OK) { + EXM_THROW(79, "zstd: %s: deflateEnd error %d \n", srcFileName, ret); + } } + *readsize = inFileSize; + return outFileSize; +} +#endif + + +#ifdef ZSTD_LZMACOMPRESS +static unsigned long long +FIO_compressLzmaFrame(cRess_t* ress, + const char* srcFileName, U64 const srcFileSize, + int compressionLevel, U64* readsize, int plain_lzma) +{ + unsigned long long inFileSize = 0, outFileSize = 0; + lzma_stream strm = LZMA_STREAM_INIT; + lzma_action action = LZMA_RUN; + lzma_ret ret; + + if (compressionLevel < 0) compressionLevel = 0; + if (compressionLevel > 9) compressionLevel = 9; + + if (plain_lzma) { + lzma_options_lzma opt_lzma; + if (lzma_lzma_preset(&opt_lzma, compressionLevel)) + EXM_THROW(81, "zstd: %s: lzma_lzma_preset error", srcFileName); + ret = lzma_alone_encoder(&strm, &opt_lzma); /* LZMA */ + if (ret != LZMA_OK) + EXM_THROW(82, "zstd: %s: lzma_alone_encoder error %d", srcFileName, ret); + } else { + ret = lzma_easy_encoder(&strm, compressionLevel, LZMA_CHECK_CRC64); /* XZ */ + if (ret != LZMA_OK) + EXM_THROW(83, "zstd: %s: lzma_easy_encoder error %d", srcFileName, ret); + } + + strm.next_in = 0; + strm.avail_in = 0; + strm.next_out = (BYTE*)ress->dstBuffer; + strm.avail_out = ress->dstBufferSize; + + while (1) { + if (strm.avail_in == 0) { + size_t const inSize = fread(ress->srcBuffer, 1, ress->srcBufferSize, ress->srcFile); + if (inSize == 0) action = LZMA_FINISH; + inFileSize += inSize; + strm.next_in = (BYTE const*)ress->srcBuffer; + strm.avail_in = inSize; + } + + ret = lzma_code(&strm, action); + + if (ret != LZMA_OK && ret != LZMA_STREAM_END) + EXM_THROW(84, "zstd: %s: lzma_code encoding error %d", srcFileName, ret); + { size_t const compBytes = ress->dstBufferSize - strm.avail_out; + if (compBytes) { + if (fwrite(ress->dstBuffer, 1, compBytes, ress->dstFile) != compBytes) + EXM_THROW(85, "Write error : %s", strerror(errno)); + outFileSize += compBytes; + strm.next_out = (BYTE*)ress->dstBuffer; + strm.avail_out = ress->dstBufferSize; + } } + if (srcFileSize == UTIL_FILESIZE_UNKNOWN) + DISPLAYUPDATE(2, "\rRead : %u MB ==> %.2f%%", + (unsigned)(inFileSize>>20), + (double)outFileSize/inFileSize*100) + else + DISPLAYUPDATE(2, "\rRead : %u / %u MB ==> %.2f%%", + (unsigned)(inFileSize>>20), (unsigned)(srcFileSize>>20), + (double)outFileSize/inFileSize*100); + if (ret == LZMA_STREAM_END) break; + } + + lzma_end(&strm); + *readsize = inFileSize; + + return outFileSize; +} +#endif + +#ifdef ZSTD_LZ4COMPRESS + +#if LZ4_VERSION_NUMBER <= 10600 +#define LZ4F_blockLinked blockLinked +#define LZ4F_max64KB max64KB +#endif + +static int FIO_LZ4_GetBlockSize_FromBlockId (int id) { return (1 << (8 + (2 * id))); } + +static unsigned long long +FIO_compressLz4Frame(cRess_t* ress, + const char* srcFileName, U64 const srcFileSize, + int compressionLevel, int checksumFlag, + U64* readsize) +{ + const size_t blockSize = FIO_LZ4_GetBlockSize_FromBlockId(LZ4F_max64KB); + unsigned long long inFileSize = 0, outFileSize = 0; + + LZ4F_preferences_t prefs; + LZ4F_compressionContext_t ctx; + + LZ4F_errorCode_t const errorCode = LZ4F_createCompressionContext(&ctx, LZ4F_VERSION); + if (LZ4F_isError(errorCode)) + EXM_THROW(31, "zstd: failed to create lz4 compression context"); + + memset(&prefs, 0, sizeof(prefs)); + + assert(blockSize <= ress->srcBufferSize); + + prefs.autoFlush = 1; + prefs.compressionLevel = compressionLevel; + prefs.frameInfo.blockMode = LZ4F_blockLinked; + prefs.frameInfo.blockSizeID = LZ4F_max64KB; + prefs.frameInfo.contentChecksumFlag = (contentChecksum_t)checksumFlag; +#if LZ4_VERSION_NUMBER >= 10600 + prefs.frameInfo.contentSize = (srcFileSize==UTIL_FILESIZE_UNKNOWN) ? 0 : srcFileSize; +#endif + assert(LZ4F_compressBound(blockSize, &prefs) <= ress->dstBufferSize); + + { + size_t readSize; + size_t headerSize = LZ4F_compressBegin(ctx, ress->dstBuffer, ress->dstBufferSize, &prefs); + if (LZ4F_isError(headerSize)) + EXM_THROW(33, "File header generation failed : %s", + LZ4F_getErrorName(headerSize)); + if (fwrite(ress->dstBuffer, 1, headerSize, ress->dstFile) != headerSize) + EXM_THROW(34, "Write error : %s (cannot write header)", strerror(errno)); + outFileSize += headerSize; + + /* Read first block */ + readSize = fread(ress->srcBuffer, (size_t)1, (size_t)blockSize, ress->srcFile); + inFileSize += readSize; + + /* Main Loop */ + while (readSize>0) { + size_t const outSize = LZ4F_compressUpdate(ctx, + ress->dstBuffer, ress->dstBufferSize, + ress->srcBuffer, readSize, NULL); + if (LZ4F_isError(outSize)) + EXM_THROW(35, "zstd: %s: lz4 compression failed : %s", + srcFileName, LZ4F_getErrorName(outSize)); + outFileSize += outSize; + if (srcFileSize == UTIL_FILESIZE_UNKNOWN) { + DISPLAYUPDATE(2, "\rRead : %u MB ==> %.2f%%", + (unsigned)(inFileSize>>20), + (double)outFileSize/inFileSize*100) + } else { + DISPLAYUPDATE(2, "\rRead : %u / %u MB ==> %.2f%%", + (unsigned)(inFileSize>>20), (unsigned)(srcFileSize>>20), + (double)outFileSize/inFileSize*100); + } + + /* Write Block */ + { size_t const sizeCheck = fwrite(ress->dstBuffer, 1, outSize, ress->dstFile); + if (sizeCheck != outSize) + EXM_THROW(36, "Write error : %s", strerror(errno)); + } + + /* Read next block */ + readSize = fread(ress->srcBuffer, (size_t)1, (size_t)blockSize, ress->srcFile); + inFileSize += readSize; + } + if (ferror(ress->srcFile)) EXM_THROW(37, "Error reading %s ", srcFileName); + + /* End of Stream mark */ + headerSize = LZ4F_compressEnd(ctx, ress->dstBuffer, ress->dstBufferSize, NULL); + if (LZ4F_isError(headerSize)) + EXM_THROW(38, "zstd: %s: lz4 end of file generation failed : %s", + srcFileName, LZ4F_getErrorName(headerSize)); + + { size_t const sizeCheck = fwrite(ress->dstBuffer, 1, headerSize, ress->dstFile); + if (sizeCheck != headerSize) + EXM_THROW(39, "Write error : %s (cannot write end of stream)", + strerror(errno)); + } + outFileSize += headerSize; + } + + *readsize = inFileSize; + LZ4F_freeCompressionContext(ctx); + + return outFileSize; +} +#endif + + +static unsigned long long +FIO_compressZstdFrame(FIO_ctx_t* const fCtx, + FIO_prefs_t* const prefs, + const cRess_t* ressPtr, + const char* srcFileName, U64 fileSize, + int compressionLevel, U64* readsize) +{ + cRess_t const ress = *ressPtr; + FILE* const srcFile = ress.srcFile; + FILE* const dstFile = ress.dstFile; + U64 compressedfilesize = 0; + ZSTD_EndDirective directive = ZSTD_e_continue; U64 pledgedSrcSize = ZSTD_CONTENTSIZE_UNKNOWN; - - /* stats */ - ZSTD_frameProgression previous_zfp_update = { 0, 0, 0, 0, 0, 0 }; - ZSTD_frameProgression previous_zfp_correction = { 0, 0, 0, 0, 0, 0 }; - typedef enum { noChange, slower, faster } speedChange_e; - speedChange_e speedChange = noChange; - unsigned flushWaiting = 0; - unsigned inputPresented = 0; - unsigned inputBlocked = 0; - unsigned lastJobID = 0; + + /* stats */ + ZSTD_frameProgression previous_zfp_update = { 0, 0, 0, 0, 0, 0 }; + ZSTD_frameProgression previous_zfp_correction = { 0, 0, 0, 0, 0, 0 }; + typedef enum { noChange, slower, faster } speedChange_e; + speedChange_e speedChange = noChange; + unsigned flushWaiting = 0; + unsigned inputPresented = 0; + unsigned inputBlocked = 0; + unsigned lastJobID = 0; UTIL_HumanReadableSize_t const file_hrs = UTIL_makeHumanReadableSize(fileSize); - - DISPLAYLEVEL(6, "compression using zstd format \n"); - - /* init */ - if (fileSize != UTIL_FILESIZE_UNKNOWN) { + + DISPLAYLEVEL(6, "compression using zstd format \n"); + + /* init */ + if (fileSize != UTIL_FILESIZE_UNKNOWN) { pledgedSrcSize = fileSize; - CHECK(ZSTD_CCtx_setPledgedSrcSize(ress.cctx, fileSize)); - } else if (prefs->streamSrcSize > 0) { - /* unknown source size; use the declared stream size */ + CHECK(ZSTD_CCtx_setPledgedSrcSize(ress.cctx, fileSize)); + } else if (prefs->streamSrcSize > 0) { + /* unknown source size; use the declared stream size */ pledgedSrcSize = prefs->streamSrcSize; - CHECK( ZSTD_CCtx_setPledgedSrcSize(ress.cctx, prefs->streamSrcSize) ); - } + CHECK( ZSTD_CCtx_setPledgedSrcSize(ress.cctx, prefs->streamSrcSize) ); + } { int windowLog; @@ -1381,421 +1381,421 @@ FIO_compressZstdFrame(FIO_ctx_t* const fCtx, windowSize = UTIL_makeHumanReadableSize(MAX(1ULL, MIN(1ULL << windowLog, pledgedSrcSize))); DISPLAYLEVEL(4, "Decompression will require %.*f%s of memory\n", windowSize.precision, windowSize.value, windowSize.suffix); } - (void)srcFileName; - - /* Main compression loop */ - do { - size_t stillToFlush; - /* Fill input Buffer */ - size_t const inSize = fread(ress.srcBuffer, (size_t)1, ress.srcBufferSize, srcFile); - ZSTD_inBuffer inBuff = { ress.srcBuffer, inSize, 0 }; - DISPLAYLEVEL(6, "fread %u bytes from source \n", (unsigned)inSize); - *readsize += inSize; - - if ((inSize == 0) || (*readsize == fileSize)) - directive = ZSTD_e_end; - - stillToFlush = 1; - while ((inBuff.pos != inBuff.size) /* input buffer must be entirely ingested */ - || (directive == ZSTD_e_end && stillToFlush != 0) ) { - - size_t const oldIPos = inBuff.pos; - ZSTD_outBuffer outBuff = { ress.dstBuffer, ress.dstBufferSize, 0 }; - size_t const toFlushNow = ZSTD_toFlushNow(ress.cctx); - CHECK_V(stillToFlush, ZSTD_compressStream2(ress.cctx, &outBuff, &inBuff, directive)); - - /* count stats */ - inputPresented++; - if (oldIPos == inBuff.pos) inputBlocked++; /* input buffer is full and can't take any more : input speed is faster than consumption rate */ - if (!toFlushNow) flushWaiting = 1; - - /* Write compressed stream */ - DISPLAYLEVEL(6, "ZSTD_compress_generic(end:%u) => input pos(%u)<=(%u)size ; output generated %u bytes \n", - (unsigned)directive, (unsigned)inBuff.pos, (unsigned)inBuff.size, (unsigned)outBuff.pos); - if (outBuff.pos) { - size_t const sizeCheck = fwrite(ress.dstBuffer, 1, outBuff.pos, dstFile); - if (sizeCheck != outBuff.pos) - EXM_THROW(25, "Write error : %s (cannot write compressed block)", - strerror(errno)); - compressedfilesize += outBuff.pos; - } - - /* display notification; and adapt compression level */ - if (READY_FOR_UPDATE()) { - ZSTD_frameProgression const zfp = ZSTD_getFrameProgression(ress.cctx); - double const cShare = (double)zfp.produced / (double)(zfp.consumed + !zfp.consumed/*avoid div0*/) * 100; + (void)srcFileName; + + /* Main compression loop */ + do { + size_t stillToFlush; + /* Fill input Buffer */ + size_t const inSize = fread(ress.srcBuffer, (size_t)1, ress.srcBufferSize, srcFile); + ZSTD_inBuffer inBuff = { ress.srcBuffer, inSize, 0 }; + DISPLAYLEVEL(6, "fread %u bytes from source \n", (unsigned)inSize); + *readsize += inSize; + + if ((inSize == 0) || (*readsize == fileSize)) + directive = ZSTD_e_end; + + stillToFlush = 1; + while ((inBuff.pos != inBuff.size) /* input buffer must be entirely ingested */ + || (directive == ZSTD_e_end && stillToFlush != 0) ) { + + size_t const oldIPos = inBuff.pos; + ZSTD_outBuffer outBuff = { ress.dstBuffer, ress.dstBufferSize, 0 }; + size_t const toFlushNow = ZSTD_toFlushNow(ress.cctx); + CHECK_V(stillToFlush, ZSTD_compressStream2(ress.cctx, &outBuff, &inBuff, directive)); + + /* count stats */ + inputPresented++; + if (oldIPos == inBuff.pos) inputBlocked++; /* input buffer is full and can't take any more : input speed is faster than consumption rate */ + if (!toFlushNow) flushWaiting = 1; + + /* Write compressed stream */ + DISPLAYLEVEL(6, "ZSTD_compress_generic(end:%u) => input pos(%u)<=(%u)size ; output generated %u bytes \n", + (unsigned)directive, (unsigned)inBuff.pos, (unsigned)inBuff.size, (unsigned)outBuff.pos); + if (outBuff.pos) { + size_t const sizeCheck = fwrite(ress.dstBuffer, 1, outBuff.pos, dstFile); + if (sizeCheck != outBuff.pos) + EXM_THROW(25, "Write error : %s (cannot write compressed block)", + strerror(errno)); + compressedfilesize += outBuff.pos; + } + + /* display notification; and adapt compression level */ + if (READY_FOR_UPDATE()) { + ZSTD_frameProgression const zfp = ZSTD_getFrameProgression(ress.cctx); + double const cShare = (double)zfp.produced / (double)(zfp.consumed + !zfp.consumed/*avoid div0*/) * 100; UTIL_HumanReadableSize_t const buffered_hrs = UTIL_makeHumanReadableSize(zfp.ingested - zfp.consumed); UTIL_HumanReadableSize_t const consumed_hrs = UTIL_makeHumanReadableSize(zfp.consumed); UTIL_HumanReadableSize_t const produced_hrs = UTIL_makeHumanReadableSize(zfp.produced); - - /* display progress notifications */ - if (g_display_prefs.displayLevel >= 3) { + + /* display progress notifications */ + if (g_display_prefs.displayLevel >= 3) { DISPLAYUPDATE(3, "\r(L%i) Buffered :%6.*f%4s - Consumed :%6.*f%4s - Compressed :%6.*f%4s => %.2f%% ", - compressionLevel, + compressionLevel, buffered_hrs.precision, buffered_hrs.value, buffered_hrs.suffix, consumed_hrs.precision, consumed_hrs.value, consumed_hrs.suffix, produced_hrs.precision, produced_hrs.value, produced_hrs.suffix, - cShare ); - } else if (g_display_prefs.displayLevel >= 2 || g_display_prefs.progressSetting == FIO_ps_always) { - /* Require level 2 or forcibly displayed progress counter for summarized updates */ - DISPLAYLEVEL(1, "\r%79s\r", ""); /* Clear out the current displayed line */ - if (fCtx->nbFilesTotal > 1) { - size_t srcFileNameSize = strlen(srcFileName); - /* Ensure that the string we print is roughly the same size each time */ - if (srcFileNameSize > 18) { - const char* truncatedSrcFileName = srcFileName + srcFileNameSize - 15; - DISPLAYLEVEL(1, "Compress: %u/%u files. Current: ...%s ", - fCtx->currFileIdx+1, fCtx->nbFilesTotal, truncatedSrcFileName); - } else { - DISPLAYLEVEL(1, "Compress: %u/%u files. Current: %*s ", - fCtx->currFileIdx+1, fCtx->nbFilesTotal, (int)(18-srcFileNameSize), srcFileName); - } - } + cShare ); + } else if (g_display_prefs.displayLevel >= 2 || g_display_prefs.progressSetting == FIO_ps_always) { + /* Require level 2 or forcibly displayed progress counter for summarized updates */ + DISPLAYLEVEL(1, "\r%79s\r", ""); /* Clear out the current displayed line */ + if (fCtx->nbFilesTotal > 1) { + size_t srcFileNameSize = strlen(srcFileName); + /* Ensure that the string we print is roughly the same size each time */ + if (srcFileNameSize > 18) { + const char* truncatedSrcFileName = srcFileName + srcFileNameSize - 15; + DISPLAYLEVEL(1, "Compress: %u/%u files. Current: ...%s ", + fCtx->currFileIdx+1, fCtx->nbFilesTotal, truncatedSrcFileName); + } else { + DISPLAYLEVEL(1, "Compress: %u/%u files. Current: %*s ", + fCtx->currFileIdx+1, fCtx->nbFilesTotal, (int)(18-srcFileNameSize), srcFileName); + } + } DISPLAYLEVEL(1, "Read:%6.*f%4s ", consumed_hrs.precision, consumed_hrs.value, consumed_hrs.suffix); - if (fileSize != UTIL_FILESIZE_UNKNOWN) + if (fileSize != UTIL_FILESIZE_UNKNOWN) DISPLAYLEVEL(2, "/%6.*f%4s", file_hrs.precision, file_hrs.value, file_hrs.suffix); DISPLAYLEVEL(1, " ==> %2.f%%", cShare); - DELAY_NEXT_UPDATE(); - } - - /* adaptive mode : statistics measurement and speed correction */ - if (prefs->adaptiveMode) { - - /* check output speed */ - if (zfp.currentJobID > 1) { /* only possible if nbWorkers >= 1 */ - - unsigned long long newlyProduced = zfp.produced - previous_zfp_update.produced; - unsigned long long newlyFlushed = zfp.flushed - previous_zfp_update.flushed; - assert(zfp.produced >= previous_zfp_update.produced); - assert(prefs->nbWorkers >= 1); - - /* test if compression is blocked - * either because output is slow and all buffers are full - * or because input is slow and no job can start while waiting for at least one buffer to be filled. - * note : exclude starting part, since currentJobID > 1 */ - if ( (zfp.consumed == previous_zfp_update.consumed) /* no data compressed : no data available, or no more buffer to compress to, OR compression is really slow (compression of a single block is slower than update rate)*/ - && (zfp.nbActiveWorkers == 0) /* confirmed : no compression ongoing */ - ) { - DISPLAYLEVEL(6, "all buffers full : compression stopped => slow down \n") - speedChange = slower; - } - - previous_zfp_update = zfp; - - if ( (newlyProduced > (newlyFlushed * 9 / 8)) /* compression produces more data than output can flush (though production can be spiky, due to work unit : (N==4)*block sizes) */ - && (flushWaiting == 0) /* flush speed was never slowed by lack of production, so it's operating at max capacity */ - ) { - DISPLAYLEVEL(6, "compression faster than flush (%llu > %llu), and flushed was never slowed down by lack of production => slow down \n", newlyProduced, newlyFlushed); - speedChange = slower; - } - flushWaiting = 0; - } - - /* course correct only if there is at least one new job completed */ - if (zfp.currentJobID > lastJobID) { - DISPLAYLEVEL(6, "compression level adaptation check \n") - - /* check input speed */ - if (zfp.currentJobID > (unsigned)(prefs->nbWorkers+1)) { /* warm up period, to fill all workers */ - if (inputBlocked <= 0) { - DISPLAYLEVEL(6, "input is never blocked => input is slower than ingestion \n"); - speedChange = slower; - } else if (speedChange == noChange) { - unsigned long long newlyIngested = zfp.ingested - previous_zfp_correction.ingested; - unsigned long long newlyConsumed = zfp.consumed - previous_zfp_correction.consumed; - unsigned long long newlyProduced = zfp.produced - previous_zfp_correction.produced; - unsigned long long newlyFlushed = zfp.flushed - previous_zfp_correction.flushed; - previous_zfp_correction = zfp; - assert(inputPresented > 0); - DISPLAYLEVEL(6, "input blocked %u/%u(%.2f) - ingested:%u vs %u:consumed - flushed:%u vs %u:produced \n", - inputBlocked, inputPresented, (double)inputBlocked/inputPresented*100, - (unsigned)newlyIngested, (unsigned)newlyConsumed, - (unsigned)newlyFlushed, (unsigned)newlyProduced); - if ( (inputBlocked > inputPresented / 8) /* input is waiting often, because input buffers is full : compression or output too slow */ - && (newlyFlushed * 33 / 32 > newlyProduced) /* flush everything that is produced */ - && (newlyIngested * 33 / 32 > newlyConsumed) /* input speed as fast or faster than compression speed */ - ) { - DISPLAYLEVEL(6, "recommend faster as in(%llu) >= (%llu)comp(%llu) <= out(%llu) \n", - newlyIngested, newlyConsumed, newlyProduced, newlyFlushed); - speedChange = faster; - } - } - inputBlocked = 0; - inputPresented = 0; - } - - if (speedChange == slower) { - DISPLAYLEVEL(6, "slower speed , higher compression \n") - compressionLevel ++; - if (compressionLevel > ZSTD_maxCLevel()) compressionLevel = ZSTD_maxCLevel(); - if (compressionLevel > prefs->maxAdaptLevel) compressionLevel = prefs->maxAdaptLevel; - compressionLevel += (compressionLevel == 0); /* skip 0 */ - ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_compressionLevel, compressionLevel); - } - if (speedChange == faster) { - DISPLAYLEVEL(6, "faster speed , lighter compression \n") - compressionLevel --; - if (compressionLevel < prefs->minAdaptLevel) compressionLevel = prefs->minAdaptLevel; - compressionLevel -= (compressionLevel == 0); /* skip 0 */ - ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_compressionLevel, compressionLevel); - } - speedChange = noChange; - - lastJobID = zfp.currentJobID; - } /* if (zfp.currentJobID > lastJobID) */ - } /* if (g_adaptiveMode) */ - } /* if (READY_FOR_UPDATE()) */ - } /* while ((inBuff.pos != inBuff.size) */ - } while (directive != ZSTD_e_end); - - if (ferror(srcFile)) { - EXM_THROW(26, "Read error : I/O error"); - } - if (fileSize != UTIL_FILESIZE_UNKNOWN && *readsize != fileSize) { - EXM_THROW(27, "Read error : Incomplete read : %llu / %llu B", - (unsigned long long)*readsize, (unsigned long long)fileSize); - } - - return compressedfilesize; -} - -/*! FIO_compressFilename_internal() : - * same as FIO_compressFilename_extRess(), with `ress.desFile` already opened. - * @return : 0 : compression completed correctly, - * 1 : missing or pb opening srcFileName - */ -static int -FIO_compressFilename_internal(FIO_ctx_t* const fCtx, - FIO_prefs_t* const prefs, - cRess_t ress, - const char* dstFileName, const char* srcFileName, - int compressionLevel) -{ - UTIL_time_t const timeStart = UTIL_getTime(); - clock_t const cpuStart = clock(); - U64 readsize = 0; - U64 compressedfilesize = 0; - U64 const fileSize = UTIL_getFileSize(srcFileName); - DISPLAYLEVEL(5, "%s: %llu bytes \n", srcFileName, (unsigned long long)fileSize); - - /* compression format selection */ - switch (prefs->compressionType) { - default: - case FIO_zstdCompression: - compressedfilesize = FIO_compressZstdFrame(fCtx, prefs, &ress, srcFileName, fileSize, compressionLevel, &readsize); - break; - - case FIO_gzipCompression: -#ifdef ZSTD_GZCOMPRESS - compressedfilesize = FIO_compressGzFrame(&ress, srcFileName, fileSize, compressionLevel, &readsize); -#else - (void)compressionLevel; - EXM_THROW(20, "zstd: %s: file cannot be compressed as gzip (zstd compiled without ZSTD_GZCOMPRESS) -- ignored \n", - srcFileName); -#endif - break; - - case FIO_xzCompression: - case FIO_lzmaCompression: -#ifdef ZSTD_LZMACOMPRESS - compressedfilesize = FIO_compressLzmaFrame(&ress, srcFileName, fileSize, compressionLevel, &readsize, prefs->compressionType==FIO_lzmaCompression); -#else - (void)compressionLevel; - EXM_THROW(20, "zstd: %s: file cannot be compressed as xz/lzma (zstd compiled without ZSTD_LZMACOMPRESS) -- ignored \n", - srcFileName); -#endif - break; - - case FIO_lz4Compression: -#ifdef ZSTD_LZ4COMPRESS - compressedfilesize = FIO_compressLz4Frame(&ress, srcFileName, fileSize, compressionLevel, prefs->checksumFlag, &readsize); -#else - (void)compressionLevel; - EXM_THROW(20, "zstd: %s: file cannot be compressed as lz4 (zstd compiled without ZSTD_LZ4COMPRESS) -- ignored \n", - srcFileName); -#endif - break; - } - - /* Status */ - fCtx->totalBytesInput += (size_t)readsize; - fCtx->totalBytesOutput += (size_t)compressedfilesize; - DISPLAYLEVEL(2, "\r%79s\r", ""); - if (g_display_prefs.displayLevel >= 2 && - !fCtx->hasStdoutOutput && - (g_display_prefs.displayLevel >= 3 || fCtx->nbFilesTotal <= 1)) { + DELAY_NEXT_UPDATE(); + } + + /* adaptive mode : statistics measurement and speed correction */ + if (prefs->adaptiveMode) { + + /* check output speed */ + if (zfp.currentJobID > 1) { /* only possible if nbWorkers >= 1 */ + + unsigned long long newlyProduced = zfp.produced - previous_zfp_update.produced; + unsigned long long newlyFlushed = zfp.flushed - previous_zfp_update.flushed; + assert(zfp.produced >= previous_zfp_update.produced); + assert(prefs->nbWorkers >= 1); + + /* test if compression is blocked + * either because output is slow and all buffers are full + * or because input is slow and no job can start while waiting for at least one buffer to be filled. + * note : exclude starting part, since currentJobID > 1 */ + if ( (zfp.consumed == previous_zfp_update.consumed) /* no data compressed : no data available, or no more buffer to compress to, OR compression is really slow (compression of a single block is slower than update rate)*/ + && (zfp.nbActiveWorkers == 0) /* confirmed : no compression ongoing */ + ) { + DISPLAYLEVEL(6, "all buffers full : compression stopped => slow down \n") + speedChange = slower; + } + + previous_zfp_update = zfp; + + if ( (newlyProduced > (newlyFlushed * 9 / 8)) /* compression produces more data than output can flush (though production can be spiky, due to work unit : (N==4)*block sizes) */ + && (flushWaiting == 0) /* flush speed was never slowed by lack of production, so it's operating at max capacity */ + ) { + DISPLAYLEVEL(6, "compression faster than flush (%llu > %llu), and flushed was never slowed down by lack of production => slow down \n", newlyProduced, newlyFlushed); + speedChange = slower; + } + flushWaiting = 0; + } + + /* course correct only if there is at least one new job completed */ + if (zfp.currentJobID > lastJobID) { + DISPLAYLEVEL(6, "compression level adaptation check \n") + + /* check input speed */ + if (zfp.currentJobID > (unsigned)(prefs->nbWorkers+1)) { /* warm up period, to fill all workers */ + if (inputBlocked <= 0) { + DISPLAYLEVEL(6, "input is never blocked => input is slower than ingestion \n"); + speedChange = slower; + } else if (speedChange == noChange) { + unsigned long long newlyIngested = zfp.ingested - previous_zfp_correction.ingested; + unsigned long long newlyConsumed = zfp.consumed - previous_zfp_correction.consumed; + unsigned long long newlyProduced = zfp.produced - previous_zfp_correction.produced; + unsigned long long newlyFlushed = zfp.flushed - previous_zfp_correction.flushed; + previous_zfp_correction = zfp; + assert(inputPresented > 0); + DISPLAYLEVEL(6, "input blocked %u/%u(%.2f) - ingested:%u vs %u:consumed - flushed:%u vs %u:produced \n", + inputBlocked, inputPresented, (double)inputBlocked/inputPresented*100, + (unsigned)newlyIngested, (unsigned)newlyConsumed, + (unsigned)newlyFlushed, (unsigned)newlyProduced); + if ( (inputBlocked > inputPresented / 8) /* input is waiting often, because input buffers is full : compression or output too slow */ + && (newlyFlushed * 33 / 32 > newlyProduced) /* flush everything that is produced */ + && (newlyIngested * 33 / 32 > newlyConsumed) /* input speed as fast or faster than compression speed */ + ) { + DISPLAYLEVEL(6, "recommend faster as in(%llu) >= (%llu)comp(%llu) <= out(%llu) \n", + newlyIngested, newlyConsumed, newlyProduced, newlyFlushed); + speedChange = faster; + } + } + inputBlocked = 0; + inputPresented = 0; + } + + if (speedChange == slower) { + DISPLAYLEVEL(6, "slower speed , higher compression \n") + compressionLevel ++; + if (compressionLevel > ZSTD_maxCLevel()) compressionLevel = ZSTD_maxCLevel(); + if (compressionLevel > prefs->maxAdaptLevel) compressionLevel = prefs->maxAdaptLevel; + compressionLevel += (compressionLevel == 0); /* skip 0 */ + ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_compressionLevel, compressionLevel); + } + if (speedChange == faster) { + DISPLAYLEVEL(6, "faster speed , lighter compression \n") + compressionLevel --; + if (compressionLevel < prefs->minAdaptLevel) compressionLevel = prefs->minAdaptLevel; + compressionLevel -= (compressionLevel == 0); /* skip 0 */ + ZSTD_CCtx_setParameter(ress.cctx, ZSTD_c_compressionLevel, compressionLevel); + } + speedChange = noChange; + + lastJobID = zfp.currentJobID; + } /* if (zfp.currentJobID > lastJobID) */ + } /* if (g_adaptiveMode) */ + } /* if (READY_FOR_UPDATE()) */ + } /* while ((inBuff.pos != inBuff.size) */ + } while (directive != ZSTD_e_end); + + if (ferror(srcFile)) { + EXM_THROW(26, "Read error : I/O error"); + } + if (fileSize != UTIL_FILESIZE_UNKNOWN && *readsize != fileSize) { + EXM_THROW(27, "Read error : Incomplete read : %llu / %llu B", + (unsigned long long)*readsize, (unsigned long long)fileSize); + } + + return compressedfilesize; +} + +/*! FIO_compressFilename_internal() : + * same as FIO_compressFilename_extRess(), with `ress.desFile` already opened. + * @return : 0 : compression completed correctly, + * 1 : missing or pb opening srcFileName + */ +static int +FIO_compressFilename_internal(FIO_ctx_t* const fCtx, + FIO_prefs_t* const prefs, + cRess_t ress, + const char* dstFileName, const char* srcFileName, + int compressionLevel) +{ + UTIL_time_t const timeStart = UTIL_getTime(); + clock_t const cpuStart = clock(); + U64 readsize = 0; + U64 compressedfilesize = 0; + U64 const fileSize = UTIL_getFileSize(srcFileName); + DISPLAYLEVEL(5, "%s: %llu bytes \n", srcFileName, (unsigned long long)fileSize); + + /* compression format selection */ + switch (prefs->compressionType) { + default: + case FIO_zstdCompression: + compressedfilesize = FIO_compressZstdFrame(fCtx, prefs, &ress, srcFileName, fileSize, compressionLevel, &readsize); + break; + + case FIO_gzipCompression: +#ifdef ZSTD_GZCOMPRESS + compressedfilesize = FIO_compressGzFrame(&ress, srcFileName, fileSize, compressionLevel, &readsize); +#else + (void)compressionLevel; + EXM_THROW(20, "zstd: %s: file cannot be compressed as gzip (zstd compiled without ZSTD_GZCOMPRESS) -- ignored \n", + srcFileName); +#endif + break; + + case FIO_xzCompression: + case FIO_lzmaCompression: +#ifdef ZSTD_LZMACOMPRESS + compressedfilesize = FIO_compressLzmaFrame(&ress, srcFileName, fileSize, compressionLevel, &readsize, prefs->compressionType==FIO_lzmaCompression); +#else + (void)compressionLevel; + EXM_THROW(20, "zstd: %s: file cannot be compressed as xz/lzma (zstd compiled without ZSTD_LZMACOMPRESS) -- ignored \n", + srcFileName); +#endif + break; + + case FIO_lz4Compression: +#ifdef ZSTD_LZ4COMPRESS + compressedfilesize = FIO_compressLz4Frame(&ress, srcFileName, fileSize, compressionLevel, prefs->checksumFlag, &readsize); +#else + (void)compressionLevel; + EXM_THROW(20, "zstd: %s: file cannot be compressed as lz4 (zstd compiled without ZSTD_LZ4COMPRESS) -- ignored \n", + srcFileName); +#endif + break; + } + + /* Status */ + fCtx->totalBytesInput += (size_t)readsize; + fCtx->totalBytesOutput += (size_t)compressedfilesize; + DISPLAYLEVEL(2, "\r%79s\r", ""); + if (g_display_prefs.displayLevel >= 2 && + !fCtx->hasStdoutOutput && + (g_display_prefs.displayLevel >= 3 || fCtx->nbFilesTotal <= 1)) { UTIL_HumanReadableSize_t hr_isize = UTIL_makeHumanReadableSize((U64) readsize); UTIL_HumanReadableSize_t hr_osize = UTIL_makeHumanReadableSize((U64) compressedfilesize); - if (readsize == 0) { + if (readsize == 0) { DISPLAYLEVEL(2,"%-20s : (%6.*f%4s => %6.*f%4s, %s) \n", - srcFileName, + srcFileName, hr_isize.precision, hr_isize.value, hr_isize.suffix, hr_osize.precision, hr_osize.value, hr_osize.suffix, - dstFileName); - } else { + dstFileName); + } else { DISPLAYLEVEL(2,"%-20s :%6.2f%% (%6.*f%4s => %6.*f%4s, %s) \n", - srcFileName, - (double)compressedfilesize / (double)readsize * 100, + srcFileName, + (double)compressedfilesize / (double)readsize * 100, hr_isize.precision, hr_isize.value, hr_isize.suffix, hr_osize.precision, hr_osize.value, hr_osize.suffix, - dstFileName); - } - } - - /* Elapsed Time and CPU Load */ - { clock_t const cpuEnd = clock(); - double const cpuLoad_s = (double)(cpuEnd - cpuStart) / CLOCKS_PER_SEC; - U64 const timeLength_ns = UTIL_clockSpanNano(timeStart); - double const timeLength_s = (double)timeLength_ns / 1000000000; - double const cpuLoad_pct = (cpuLoad_s / timeLength_s) * 100; - DISPLAYLEVEL(4, "%-20s : Completed in %.2f sec (cpu load : %.0f%%)\n", - srcFileName, timeLength_s, cpuLoad_pct); - } - return 0; -} - - -/*! FIO_compressFilename_dstFile() : - * open dstFileName, or pass-through if ress.dstFile != NULL, - * then start compression with FIO_compressFilename_internal(). - * Manages source removal (--rm) and file permissions transfer. - * note : ress.srcFile must be != NULL, - * so reach this function through FIO_compressFilename_srcFile(). - * @return : 0 : compression completed correctly, - * 1 : pb - */ -static int FIO_compressFilename_dstFile(FIO_ctx_t* const fCtx, - FIO_prefs_t* const prefs, - cRess_t ress, - const char* dstFileName, - const char* srcFileName, - int compressionLevel) -{ - int closeDstFile = 0; - int result; - stat_t statbuf; + dstFileName); + } + } + + /* Elapsed Time and CPU Load */ + { clock_t const cpuEnd = clock(); + double const cpuLoad_s = (double)(cpuEnd - cpuStart) / CLOCKS_PER_SEC; + U64 const timeLength_ns = UTIL_clockSpanNano(timeStart); + double const timeLength_s = (double)timeLength_ns / 1000000000; + double const cpuLoad_pct = (cpuLoad_s / timeLength_s) * 100; + DISPLAYLEVEL(4, "%-20s : Completed in %.2f sec (cpu load : %.0f%%)\n", + srcFileName, timeLength_s, cpuLoad_pct); + } + return 0; +} + + +/*! FIO_compressFilename_dstFile() : + * open dstFileName, or pass-through if ress.dstFile != NULL, + * then start compression with FIO_compressFilename_internal(). + * Manages source removal (--rm) and file permissions transfer. + * note : ress.srcFile must be != NULL, + * so reach this function through FIO_compressFilename_srcFile(). + * @return : 0 : compression completed correctly, + * 1 : pb + */ +static int FIO_compressFilename_dstFile(FIO_ctx_t* const fCtx, + FIO_prefs_t* const prefs, + cRess_t ress, + const char* dstFileName, + const char* srcFileName, + int compressionLevel) +{ + int closeDstFile = 0; + int result; + stat_t statbuf; int transferMTime = 0; - assert(ress.srcFile != NULL); - if (ress.dstFile == NULL) { - int dstFilePermissions = DEFAULT_FILE_PERMISSIONS; - if ( strcmp (srcFileName, stdinmark) + assert(ress.srcFile != NULL); + if (ress.dstFile == NULL) { + int dstFilePermissions = DEFAULT_FILE_PERMISSIONS; + if ( strcmp (srcFileName, stdinmark) && strcmp (dstFileName, stdoutmark) - && UTIL_stat(srcFileName, &statbuf) - && UTIL_isRegularFileStat(&statbuf) ) { - dstFilePermissions = statbuf.st_mode; + && UTIL_stat(srcFileName, &statbuf) + && UTIL_isRegularFileStat(&statbuf) ) { + dstFilePermissions = statbuf.st_mode; transferMTime = 1; - } - - closeDstFile = 1; - DISPLAYLEVEL(6, "FIO_compressFilename_dstFile: opening dst: %s \n", dstFileName); - ress.dstFile = FIO_openDstFile(fCtx, prefs, srcFileName, dstFileName, dstFilePermissions); - if (ress.dstFile==NULL) return 1; /* could not open dstFileName */ - /* Must only be added after FIO_openDstFile() succeeds. - * Otherwise we may delete the destination file if it already exists, - * and the user presses Ctrl-C when asked if they wish to overwrite. - */ - addHandler(dstFileName); - } - - result = FIO_compressFilename_internal(fCtx, prefs, ress, dstFileName, srcFileName, compressionLevel); - - if (closeDstFile) { - FILE* const dstFile = ress.dstFile; - ress.dstFile = NULL; - - clearHandler(); - - DISPLAYLEVEL(6, "FIO_compressFilename_dstFile: closing dst: %s \n", dstFileName); - if (fclose(dstFile)) { /* error closing dstFile */ - DISPLAYLEVEL(1, "zstd: %s: %s \n", dstFileName, strerror(errno)); - result=1; - } + } + + closeDstFile = 1; + DISPLAYLEVEL(6, "FIO_compressFilename_dstFile: opening dst: %s \n", dstFileName); + ress.dstFile = FIO_openDstFile(fCtx, prefs, srcFileName, dstFileName, dstFilePermissions); + if (ress.dstFile==NULL) return 1; /* could not open dstFileName */ + /* Must only be added after FIO_openDstFile() succeeds. + * Otherwise we may delete the destination file if it already exists, + * and the user presses Ctrl-C when asked if they wish to overwrite. + */ + addHandler(dstFileName); + } + + result = FIO_compressFilename_internal(fCtx, prefs, ress, dstFileName, srcFileName, compressionLevel); + + if (closeDstFile) { + FILE* const dstFile = ress.dstFile; + ress.dstFile = NULL; + + clearHandler(); + + DISPLAYLEVEL(6, "FIO_compressFilename_dstFile: closing dst: %s \n", dstFileName); + if (fclose(dstFile)) { /* error closing dstFile */ + DISPLAYLEVEL(1, "zstd: %s: %s \n", dstFileName, strerror(errno)); + result=1; + } if (transferMTime) { UTIL_utime(dstFileName, &statbuf); } - if ( (result != 0) /* operation failure */ - && strcmp(dstFileName, stdoutmark) /* special case : don't remove() stdout */ - ) { - FIO_removeFile(dstFileName); /* remove compression artefact; note don't do anything special if remove() fails */ - } - } - - return result; -} - -/* List used to compare file extensions (used with --exclude-compressed flag) -* Different from the suffixList and should only apply to ZSTD compress operationResult -*/ -static const char *compressedFileExtensions[] = { - ZSTD_EXTENSION, - TZSTD_EXTENSION, - GZ_EXTENSION, - TGZ_EXTENSION, - LZMA_EXTENSION, - XZ_EXTENSION, - TXZ_EXTENSION, - LZ4_EXTENSION, - TLZ4_EXTENSION, - NULL -}; - -/*! FIO_compressFilename_srcFile() : - * @return : 0 : compression completed correctly, - * 1 : missing or pb opening srcFileName - */ -static int -FIO_compressFilename_srcFile(FIO_ctx_t* const fCtx, - FIO_prefs_t* const prefs, - cRess_t ress, - const char* dstFileName, - const char* srcFileName, - int compressionLevel) -{ - int result; - DISPLAYLEVEL(6, "FIO_compressFilename_srcFile: %s \n", srcFileName); - - /* ensure src is not a directory */ - if (UTIL_isDirectory(srcFileName)) { - DISPLAYLEVEL(1, "zstd: %s is a directory -- ignored \n", srcFileName); - return 1; - } - - /* ensure src is not the same as dict (if present) */ - if (ress.dictFileName != NULL && UTIL_isSameFile(srcFileName, ress.dictFileName)) { - DISPLAYLEVEL(1, "zstd: cannot use %s as an input file and dictionary \n", srcFileName); - return 1; - } - - /* Check if "srcFile" is compressed. Only done if --exclude-compressed flag is used - * YES => ZSTD will skip compression of the file and will return 0. - * NO => ZSTD will resume with compress operation. - */ - if (prefs->excludeCompressedFiles == 1 && UTIL_isCompressedFile(srcFileName, compressedFileExtensions)) { - DISPLAYLEVEL(4, "File is already compressed : %s \n", srcFileName); - return 0; - } - - ress.srcFile = FIO_openSrcFile(prefs, srcFileName); - if (ress.srcFile == NULL) return 1; /* srcFile could not be opened */ - - result = FIO_compressFilename_dstFile(fCtx, prefs, ress, dstFileName, srcFileName, compressionLevel); - - fclose(ress.srcFile); - ress.srcFile = NULL; - if ( prefs->removeSrcFile /* --rm */ - && result == 0 /* success */ - && strcmp(srcFileName, stdinmark) /* exception : don't erase stdin */ - ) { - /* We must clear the handler, since after this point calling it would - * delete both the source and destination files. - */ - clearHandler(); - if (FIO_removeFile(srcFileName)) - EXM_THROW(1, "zstd: %s: %s", srcFileName, strerror(errno)); - } - return result; -} - + if ( (result != 0) /* operation failure */ + && strcmp(dstFileName, stdoutmark) /* special case : don't remove() stdout */ + ) { + FIO_removeFile(dstFileName); /* remove compression artefact; note don't do anything special if remove() fails */ + } + } + + return result; +} + +/* List used to compare file extensions (used with --exclude-compressed flag) +* Different from the suffixList and should only apply to ZSTD compress operationResult +*/ +static const char *compressedFileExtensions[] = { + ZSTD_EXTENSION, + TZSTD_EXTENSION, + GZ_EXTENSION, + TGZ_EXTENSION, + LZMA_EXTENSION, + XZ_EXTENSION, + TXZ_EXTENSION, + LZ4_EXTENSION, + TLZ4_EXTENSION, + NULL +}; + +/*! FIO_compressFilename_srcFile() : + * @return : 0 : compression completed correctly, + * 1 : missing or pb opening srcFileName + */ +static int +FIO_compressFilename_srcFile(FIO_ctx_t* const fCtx, + FIO_prefs_t* const prefs, + cRess_t ress, + const char* dstFileName, + const char* srcFileName, + int compressionLevel) +{ + int result; + DISPLAYLEVEL(6, "FIO_compressFilename_srcFile: %s \n", srcFileName); + + /* ensure src is not a directory */ + if (UTIL_isDirectory(srcFileName)) { + DISPLAYLEVEL(1, "zstd: %s is a directory -- ignored \n", srcFileName); + return 1; + } + + /* ensure src is not the same as dict (if present) */ + if (ress.dictFileName != NULL && UTIL_isSameFile(srcFileName, ress.dictFileName)) { + DISPLAYLEVEL(1, "zstd: cannot use %s as an input file and dictionary \n", srcFileName); + return 1; + } + + /* Check if "srcFile" is compressed. Only done if --exclude-compressed flag is used + * YES => ZSTD will skip compression of the file and will return 0. + * NO => ZSTD will resume with compress operation. + */ + if (prefs->excludeCompressedFiles == 1 && UTIL_isCompressedFile(srcFileName, compressedFileExtensions)) { + DISPLAYLEVEL(4, "File is already compressed : %s \n", srcFileName); + return 0; + } + + ress.srcFile = FIO_openSrcFile(prefs, srcFileName); + if (ress.srcFile == NULL) return 1; /* srcFile could not be opened */ + + result = FIO_compressFilename_dstFile(fCtx, prefs, ress, dstFileName, srcFileName, compressionLevel); + + fclose(ress.srcFile); + ress.srcFile = NULL; + if ( prefs->removeSrcFile /* --rm */ + && result == 0 /* success */ + && strcmp(srcFileName, stdinmark) /* exception : don't erase stdin */ + ) { + /* We must clear the handler, since after this point calling it would + * delete both the source and destination files. + */ + clearHandler(); + if (FIO_removeFile(srcFileName)) + EXM_THROW(1, "zstd: %s: %s", srcFileName, strerror(errno)); + } + return result; +} + static const char* checked_index(const char* options[], size_t length, size_t index) { assert(index < length); // Necessary to avoid warnings since -O3 will omit the above `assert` @@ -1840,1382 +1840,1382 @@ void FIO_displayCompressionParameters(const FIO_prefs_t* prefs) { #undef INDEX -int FIO_compressFilename(FIO_ctx_t* const fCtx, FIO_prefs_t* const prefs, const char* dstFileName, - const char* srcFileName, const char* dictFileName, - int compressionLevel, ZSTD_compressionParameters comprParams) -{ - cRess_t const ress = FIO_createCResources(prefs, dictFileName, UTIL_getFileSize(srcFileName), compressionLevel, comprParams); - int const result = FIO_compressFilename_srcFile(fCtx, prefs, ress, dstFileName, srcFileName, compressionLevel); - -#define DISPLAY_LEVEL_DEFAULT 2 - - FIO_freeCResources(&ress); - return result; -} - -/* FIO_determineCompressedName() : - * create a destination filename for compressed srcFileName. - * @return a pointer to it. - * This function never returns an error (it may abort() in case of pb) - */ -static const char* -FIO_determineCompressedName(const char* srcFileName, const char* outDirName, const char* suffix) -{ - static size_t dfnbCapacity = 0; - static char* dstFileNameBuffer = NULL; /* using static allocation : this function cannot be multi-threaded */ - char* outDirFilename = NULL; - size_t sfnSize = strlen(srcFileName); - size_t const srcSuffixLen = strlen(suffix); - if (outDirName) { - outDirFilename = FIO_createFilename_fromOutDir(srcFileName, outDirName, srcSuffixLen); - sfnSize = strlen(outDirFilename); - assert(outDirFilename != NULL); - } - - if (dfnbCapacity <= sfnSize+srcSuffixLen+1) { - /* resize buffer for dstName */ - free(dstFileNameBuffer); - dfnbCapacity = sfnSize + srcSuffixLen + 30; - dstFileNameBuffer = (char*)malloc(dfnbCapacity); - if (!dstFileNameBuffer) { - EXM_THROW(30, "zstd: %s", strerror(errno)); - } - } - assert(dstFileNameBuffer != NULL); - - if (outDirFilename) { - memcpy(dstFileNameBuffer, outDirFilename, sfnSize); - free(outDirFilename); - } else { - memcpy(dstFileNameBuffer, srcFileName, sfnSize); - } - memcpy(dstFileNameBuffer+sfnSize, suffix, srcSuffixLen+1 /* Include terminating null */); - return dstFileNameBuffer; -} - -static unsigned long long FIO_getLargestFileSize(const char** inFileNames, unsigned nbFiles) -{ - size_t i; - unsigned long long fileSize, maxFileSize = 0; - for (i = 0; i < nbFiles; i++) { - fileSize = UTIL_getFileSize(inFileNames[i]); - maxFileSize = fileSize > maxFileSize ? fileSize : maxFileSize; - } - return maxFileSize; -} - -/* FIO_compressMultipleFilenames() : - * compress nbFiles files - * into either one destination (outFileName), - * or into one file each (outFileName == NULL, but suffix != NULL), - * or into a destination folder (specified with -O) - */ -int FIO_compressMultipleFilenames(FIO_ctx_t* const fCtx, - FIO_prefs_t* const prefs, - const char** inFileNamesTable, - const char* outMirroredRootDirName, - const char* outDirName, - const char* outFileName, const char* suffix, - const char* dictFileName, int compressionLevel, - ZSTD_compressionParameters comprParams) -{ - int status; - int error = 0; - cRess_t ress = FIO_createCResources(prefs, dictFileName, - FIO_getLargestFileSize(inFileNamesTable, (unsigned)fCtx->nbFilesTotal), - compressionLevel, comprParams); - - /* init */ - assert(outFileName != NULL || suffix != NULL); - if (outFileName != NULL) { /* output into a single destination (stdout typically) */ - if (FIO_removeMultiFilesWarning(fCtx, prefs, outFileName, 1 /* displayLevelCutoff */)) { - FIO_freeCResources(&ress); - return 1; - } - ress.dstFile = FIO_openDstFile(fCtx, prefs, NULL, outFileName, DEFAULT_FILE_PERMISSIONS); - if (ress.dstFile == NULL) { /* could not open outFileName */ - error = 1; - } else { - for (; fCtx->currFileIdx < fCtx->nbFilesTotal; ++fCtx->currFileIdx) { - status = FIO_compressFilename_srcFile(fCtx, prefs, ress, outFileName, inFileNamesTable[fCtx->currFileIdx], compressionLevel); - if (!status) fCtx->nbFilesProcessed++; - error |= status; - } - if (fclose(ress.dstFile)) - EXM_THROW(29, "Write error (%s) : cannot properly close %s", - strerror(errno), outFileName); - ress.dstFile = NULL; - } - } else { - if (outMirroredRootDirName) - UTIL_mirrorSourceFilesDirectories(inFileNamesTable, (unsigned)fCtx->nbFilesTotal, outMirroredRootDirName); - - for (; fCtx->currFileIdx < fCtx->nbFilesTotal; ++fCtx->currFileIdx) { - const char* const srcFileName = inFileNamesTable[fCtx->currFileIdx]; - const char* dstFileName = NULL; - if (outMirroredRootDirName) { - char* validMirroredDirName = UTIL_createMirroredDestDirName(srcFileName, outMirroredRootDirName); - if (validMirroredDirName) { - dstFileName = FIO_determineCompressedName(srcFileName, validMirroredDirName, suffix); - free(validMirroredDirName); - } else { - DISPLAYLEVEL(2, "zstd: --output-dir-mirror cannot compress '%s' into '%s' \n", srcFileName, outMirroredRootDirName); - error=1; - continue; - } - } else { - dstFileName = FIO_determineCompressedName(srcFileName, outDirName, suffix); /* cannot fail */ - } - status = FIO_compressFilename_srcFile(fCtx, prefs, ress, dstFileName, srcFileName, compressionLevel); - if (!status) fCtx->nbFilesProcessed++; - error |= status; - } - - if (outDirName) - FIO_checkFilenameCollisions(inFileNamesTable , (unsigned)fCtx->nbFilesTotal); - } - - if (fCtx->nbFilesProcessed >= 1 && fCtx->nbFilesTotal > 1 && fCtx->totalBytesInput != 0) { +int FIO_compressFilename(FIO_ctx_t* const fCtx, FIO_prefs_t* const prefs, const char* dstFileName, + const char* srcFileName, const char* dictFileName, + int compressionLevel, ZSTD_compressionParameters comprParams) +{ + cRess_t const ress = FIO_createCResources(prefs, dictFileName, UTIL_getFileSize(srcFileName), compressionLevel, comprParams); + int const result = FIO_compressFilename_srcFile(fCtx, prefs, ress, dstFileName, srcFileName, compressionLevel); + +#define DISPLAY_LEVEL_DEFAULT 2 + + FIO_freeCResources(&ress); + return result; +} + +/* FIO_determineCompressedName() : + * create a destination filename for compressed srcFileName. + * @return a pointer to it. + * This function never returns an error (it may abort() in case of pb) + */ +static const char* +FIO_determineCompressedName(const char* srcFileName, const char* outDirName, const char* suffix) +{ + static size_t dfnbCapacity = 0; + static char* dstFileNameBuffer = NULL; /* using static allocation : this function cannot be multi-threaded */ + char* outDirFilename = NULL; + size_t sfnSize = strlen(srcFileName); + size_t const srcSuffixLen = strlen(suffix); + if (outDirName) { + outDirFilename = FIO_createFilename_fromOutDir(srcFileName, outDirName, srcSuffixLen); + sfnSize = strlen(outDirFilename); + assert(outDirFilename != NULL); + } + + if (dfnbCapacity <= sfnSize+srcSuffixLen+1) { + /* resize buffer for dstName */ + free(dstFileNameBuffer); + dfnbCapacity = sfnSize + srcSuffixLen + 30; + dstFileNameBuffer = (char*)malloc(dfnbCapacity); + if (!dstFileNameBuffer) { + EXM_THROW(30, "zstd: %s", strerror(errno)); + } + } + assert(dstFileNameBuffer != NULL); + + if (outDirFilename) { + memcpy(dstFileNameBuffer, outDirFilename, sfnSize); + free(outDirFilename); + } else { + memcpy(dstFileNameBuffer, srcFileName, sfnSize); + } + memcpy(dstFileNameBuffer+sfnSize, suffix, srcSuffixLen+1 /* Include terminating null */); + return dstFileNameBuffer; +} + +static unsigned long long FIO_getLargestFileSize(const char** inFileNames, unsigned nbFiles) +{ + size_t i; + unsigned long long fileSize, maxFileSize = 0; + for (i = 0; i < nbFiles; i++) { + fileSize = UTIL_getFileSize(inFileNames[i]); + maxFileSize = fileSize > maxFileSize ? fileSize : maxFileSize; + } + return maxFileSize; +} + +/* FIO_compressMultipleFilenames() : + * compress nbFiles files + * into either one destination (outFileName), + * or into one file each (outFileName == NULL, but suffix != NULL), + * or into a destination folder (specified with -O) + */ +int FIO_compressMultipleFilenames(FIO_ctx_t* const fCtx, + FIO_prefs_t* const prefs, + const char** inFileNamesTable, + const char* outMirroredRootDirName, + const char* outDirName, + const char* outFileName, const char* suffix, + const char* dictFileName, int compressionLevel, + ZSTD_compressionParameters comprParams) +{ + int status; + int error = 0; + cRess_t ress = FIO_createCResources(prefs, dictFileName, + FIO_getLargestFileSize(inFileNamesTable, (unsigned)fCtx->nbFilesTotal), + compressionLevel, comprParams); + + /* init */ + assert(outFileName != NULL || suffix != NULL); + if (outFileName != NULL) { /* output into a single destination (stdout typically) */ + if (FIO_removeMultiFilesWarning(fCtx, prefs, outFileName, 1 /* displayLevelCutoff */)) { + FIO_freeCResources(&ress); + return 1; + } + ress.dstFile = FIO_openDstFile(fCtx, prefs, NULL, outFileName, DEFAULT_FILE_PERMISSIONS); + if (ress.dstFile == NULL) { /* could not open outFileName */ + error = 1; + } else { + for (; fCtx->currFileIdx < fCtx->nbFilesTotal; ++fCtx->currFileIdx) { + status = FIO_compressFilename_srcFile(fCtx, prefs, ress, outFileName, inFileNamesTable[fCtx->currFileIdx], compressionLevel); + if (!status) fCtx->nbFilesProcessed++; + error |= status; + } + if (fclose(ress.dstFile)) + EXM_THROW(29, "Write error (%s) : cannot properly close %s", + strerror(errno), outFileName); + ress.dstFile = NULL; + } + } else { + if (outMirroredRootDirName) + UTIL_mirrorSourceFilesDirectories(inFileNamesTable, (unsigned)fCtx->nbFilesTotal, outMirroredRootDirName); + + for (; fCtx->currFileIdx < fCtx->nbFilesTotal; ++fCtx->currFileIdx) { + const char* const srcFileName = inFileNamesTable[fCtx->currFileIdx]; + const char* dstFileName = NULL; + if (outMirroredRootDirName) { + char* validMirroredDirName = UTIL_createMirroredDestDirName(srcFileName, outMirroredRootDirName); + if (validMirroredDirName) { + dstFileName = FIO_determineCompressedName(srcFileName, validMirroredDirName, suffix); + free(validMirroredDirName); + } else { + DISPLAYLEVEL(2, "zstd: --output-dir-mirror cannot compress '%s' into '%s' \n", srcFileName, outMirroredRootDirName); + error=1; + continue; + } + } else { + dstFileName = FIO_determineCompressedName(srcFileName, outDirName, suffix); /* cannot fail */ + } + status = FIO_compressFilename_srcFile(fCtx, prefs, ress, dstFileName, srcFileName, compressionLevel); + if (!status) fCtx->nbFilesProcessed++; + error |= status; + } + + if (outDirName) + FIO_checkFilenameCollisions(inFileNamesTable , (unsigned)fCtx->nbFilesTotal); + } + + if (fCtx->nbFilesProcessed >= 1 && fCtx->nbFilesTotal > 1 && fCtx->totalBytesInput != 0) { UTIL_HumanReadableSize_t hr_isize = UTIL_makeHumanReadableSize((U64) fCtx->totalBytesInput); UTIL_HumanReadableSize_t hr_osize = UTIL_makeHumanReadableSize((U64) fCtx->totalBytesOutput); - DISPLAYLEVEL(2, "\r%79s\r", ""); + DISPLAYLEVEL(2, "\r%79s\r", ""); DISPLAYLEVEL(2, "%3d files compressed :%.2f%% (%6.*f%4s => %6.*f%4s)\n", fCtx->nbFilesProcessed, - (double)fCtx->totalBytesOutput/((double)fCtx->totalBytesInput)*100, + (double)fCtx->totalBytesOutput/((double)fCtx->totalBytesInput)*100, hr_isize.precision, hr_isize.value, hr_isize.suffix, hr_osize.precision, hr_osize.value, hr_osize.suffix); - } - - FIO_freeCResources(&ress); - return error; -} - -#endif /* #ifndef ZSTD_NOCOMPRESS */ - - - -#ifndef ZSTD_NODECOMPRESS - -/* ************************************************************************** - * Decompression - ***************************************************************************/ -typedef struct { - void* srcBuffer; - size_t srcBufferSize; - size_t srcBufferLoaded; - void* dstBuffer; - size_t dstBufferSize; - ZSTD_DStream* dctx; - FILE* dstFile; -} dRess_t; - -static dRess_t FIO_createDResources(FIO_prefs_t* const prefs, const char* dictFileName) -{ - dRess_t ress; - memset(&ress, 0, sizeof(ress)); - - if (prefs->patchFromMode) - FIO_adjustMemLimitForPatchFromMode(prefs, UTIL_getFileSize(dictFileName), 0 /* just use the dict size */); - - /* Allocation */ - ress.dctx = ZSTD_createDStream(); - if (ress.dctx==NULL) - EXM_THROW(60, "Error: %s : can't create ZSTD_DStream", strerror(errno)); - CHECK( ZSTD_DCtx_setMaxWindowSize(ress.dctx, prefs->memLimit) ); - CHECK( ZSTD_DCtx_setParameter(ress.dctx, ZSTD_d_forceIgnoreChecksum, !prefs->checksumFlag)); - - ress.srcBufferSize = ZSTD_DStreamInSize(); - ress.srcBuffer = malloc(ress.srcBufferSize); - ress.dstBufferSize = ZSTD_DStreamOutSize(); - ress.dstBuffer = malloc(ress.dstBufferSize); - if (!ress.srcBuffer || !ress.dstBuffer) - EXM_THROW(61, "Allocation error : not enough memory"); - - /* dictionary */ - { void* dictBuffer; - size_t const dictBufferSize = FIO_createDictBuffer(&dictBuffer, dictFileName, prefs); - CHECK( ZSTD_initDStream_usingDict(ress.dctx, dictBuffer, dictBufferSize) ); - free(dictBuffer); - } - - return ress; -} - -static void FIO_freeDResources(dRess_t ress) -{ - CHECK( ZSTD_freeDStream(ress.dctx) ); - free(ress.srcBuffer); - free(ress.dstBuffer); -} - - -/** FIO_fwriteSparse() : -* @return : storedSkips, -* argument for next call to FIO_fwriteSparse() or FIO_fwriteSparseEnd() */ -static unsigned -FIO_fwriteSparse(FILE* file, - const void* buffer, size_t bufferSize, - const FIO_prefs_t* const prefs, - unsigned storedSkips) -{ - const size_t* const bufferT = (const size_t*)buffer; /* Buffer is supposed malloc'ed, hence aligned on size_t */ - size_t bufferSizeT = bufferSize / sizeof(size_t); - const size_t* const bufferTEnd = bufferT + bufferSizeT; - const size_t* ptrT = bufferT; - static const size_t segmentSizeT = (32 KB) / sizeof(size_t); /* check every 32 KB */ - - if (prefs->testMode) return 0; /* do not output anything in test mode */ - - if (!prefs->sparseFileSupport) { /* normal write */ - size_t const sizeCheck = fwrite(buffer, 1, bufferSize, file); - if (sizeCheck != bufferSize) - EXM_THROW(70, "Write error : cannot write decoded block : %s", - strerror(errno)); - return 0; - } - - /* avoid int overflow */ - if (storedSkips > 1 GB) { - if (LONG_SEEK(file, 1 GB, SEEK_CUR) != 0) - EXM_THROW(91, "1 GB skip error (sparse file support)"); - storedSkips -= 1 GB; - } - - while (ptrT < bufferTEnd) { - size_t nb0T; - - /* adjust last segment if < 32 KB */ - size_t seg0SizeT = segmentSizeT; - if (seg0SizeT > bufferSizeT) seg0SizeT = bufferSizeT; - bufferSizeT -= seg0SizeT; - - /* count leading zeroes */ - for (nb0T=0; (nb0T < seg0SizeT) && (ptrT[nb0T] == 0); nb0T++) ; - storedSkips += (unsigned)(nb0T * sizeof(size_t)); - - if (nb0T != seg0SizeT) { /* not all 0s */ - size_t const nbNon0ST = seg0SizeT - nb0T; - /* skip leading zeros */ - if (LONG_SEEK(file, storedSkips, SEEK_CUR) != 0) - EXM_THROW(92, "Sparse skip error ; try --no-sparse"); - storedSkips = 0; - /* write the rest */ - if (fwrite(ptrT + nb0T, sizeof(size_t), nbNon0ST, file) != nbNon0ST) - EXM_THROW(93, "Write error : cannot write decoded block : %s", - strerror(errno)); - } - ptrT += seg0SizeT; - } - - { static size_t const maskT = sizeof(size_t)-1; - if (bufferSize & maskT) { - /* size not multiple of sizeof(size_t) : implies end of block */ - const char* const restStart = (const char*)bufferTEnd; - const char* restPtr = restStart; - const char* const restEnd = (const char*)buffer + bufferSize; - assert(restEnd > restStart && restEnd < restStart + sizeof(size_t)); - for ( ; (restPtr < restEnd) && (*restPtr == 0); restPtr++) ; - storedSkips += (unsigned) (restPtr - restStart); - if (restPtr != restEnd) { - /* not all remaining bytes are 0 */ - size_t const restSize = (size_t)(restEnd - restPtr); - if (LONG_SEEK(file, storedSkips, SEEK_CUR) != 0) - EXM_THROW(92, "Sparse skip error ; try --no-sparse"); - if (fwrite(restPtr, 1, restSize, file) != restSize) - EXM_THROW(95, "Write error : cannot write end of decoded block : %s", - strerror(errno)); - storedSkips = 0; - } } } - - return storedSkips; -} - -static void -FIO_fwriteSparseEnd(const FIO_prefs_t* const prefs, FILE* file, unsigned storedSkips) -{ - if (prefs->testMode) assert(storedSkips == 0); - if (storedSkips>0) { - assert(prefs->sparseFileSupport > 0); /* storedSkips>0 implies sparse support is enabled */ - (void)prefs; /* assert can be disabled, in which case prefs becomes unused */ - if (LONG_SEEK(file, storedSkips-1, SEEK_CUR) != 0) - EXM_THROW(69, "Final skip error (sparse file support)"); - /* last zero must be explicitly written, - * so that skipped ones get implicitly translated as zero by FS */ - { const char lastZeroByte[1] = { 0 }; - if (fwrite(lastZeroByte, 1, 1, file) != 1) - EXM_THROW(69, "Write error : cannot write last zero : %s", strerror(errno)); - } } -} - - -/** FIO_passThrough() : just copy input into output, for compatibility with gzip -df mode - @return : 0 (no error) */ -static int FIO_passThrough(const FIO_prefs_t* const prefs, - FILE* foutput, FILE* finput, - void* buffer, size_t bufferSize, - size_t alreadyLoaded) -{ - size_t const blockSize = MIN(64 KB, bufferSize); - size_t readFromInput; - unsigned storedSkips = 0; - - /* assumption : ress->srcBufferLoaded bytes already loaded and stored within buffer */ - { size_t const sizeCheck = fwrite(buffer, 1, alreadyLoaded, foutput); - if (sizeCheck != alreadyLoaded) { - DISPLAYLEVEL(1, "Pass-through write error : %s\n", strerror(errno)); - return 1; - } } - - do { - readFromInput = fread(buffer, 1, blockSize, finput); - storedSkips = FIO_fwriteSparse(foutput, buffer, readFromInput, prefs, storedSkips); - } while (readFromInput == blockSize); - if (ferror(finput)) { - DISPLAYLEVEL(1, "Pass-through read error : %s\n", strerror(errno)); - return 1; - } - assert(feof(finput)); - - FIO_fwriteSparseEnd(prefs, foutput, storedSkips); - return 0; -} - -/* FIO_zstdErrorHelp() : - * detailed error message when requested window size is too large */ -static void -FIO_zstdErrorHelp(const FIO_prefs_t* const prefs, - const dRess_t* ress, - size_t err, const char* srcFileName) -{ - ZSTD_frameHeader header; - - /* Help message only for one specific error */ - if (ZSTD_getErrorCode(err) != ZSTD_error_frameParameter_windowTooLarge) - return; - - /* Try to decode the frame header */ - err = ZSTD_getFrameHeader(&header, ress->srcBuffer, ress->srcBufferLoaded); - if (err == 0) { - unsigned long long const windowSize = header.windowSize; - unsigned const windowLog = FIO_highbit64(windowSize) + ((windowSize & (windowSize - 1)) != 0); - assert(prefs->memLimit > 0); - DISPLAYLEVEL(1, "%s : Window size larger than maximum : %llu > %u \n", - srcFileName, windowSize, prefs->memLimit); - if (windowLog <= ZSTD_WINDOWLOG_MAX) { - unsigned const windowMB = (unsigned)((windowSize >> 20) + ((windowSize & ((1 MB) - 1)) != 0)); - assert(windowSize < (U64)(1ULL << 52)); /* ensure now overflow for windowMB */ - DISPLAYLEVEL(1, "%s : Use --long=%u or --memory=%uMB \n", - srcFileName, windowLog, windowMB); - return; - } } - DISPLAYLEVEL(1, "%s : Window log larger than ZSTD_WINDOWLOG_MAX=%u; not supported \n", - srcFileName, ZSTD_WINDOWLOG_MAX); -} - -/** FIO_decompressFrame() : - * @return : size of decoded zstd frame, or an error code - */ -#define FIO_ERROR_FRAME_DECODING ((unsigned long long)(-2)) -static unsigned long long -FIO_decompressZstdFrame(FIO_ctx_t* const fCtx, dRess_t* ress, FILE* finput, - const FIO_prefs_t* const prefs, - const char* srcFileName, - U64 alreadyDecoded) /* for multi-frames streams */ -{ - U64 frameSize = 0; - U32 storedSkips = 0; - - /* display last 20 characters only */ - { size_t const srcFileLength = strlen(srcFileName); - if (srcFileLength>20) srcFileName += srcFileLength-20; - } - - ZSTD_DCtx_reset(ress->dctx, ZSTD_reset_session_only); - - /* Header loading : ensures ZSTD_getFrameHeader() will succeed */ - { size_t const toDecode = ZSTD_FRAMEHEADERSIZE_MAX; - if (ress->srcBufferLoaded < toDecode) { - size_t const toRead = toDecode - ress->srcBufferLoaded; - void* const startPosition = (char*)ress->srcBuffer + ress->srcBufferLoaded; - ress->srcBufferLoaded += fread(startPosition, 1, toRead, finput); - } } - - /* Main decompression Loop */ - while (1) { - ZSTD_inBuffer inBuff = { ress->srcBuffer, ress->srcBufferLoaded, 0 }; - ZSTD_outBuffer outBuff= { ress->dstBuffer, ress->dstBufferSize, 0 }; - size_t const readSizeHint = ZSTD_decompressStream(ress->dctx, &outBuff, &inBuff); + } + + FIO_freeCResources(&ress); + return error; +} + +#endif /* #ifndef ZSTD_NOCOMPRESS */ + + + +#ifndef ZSTD_NODECOMPRESS + +/* ************************************************************************** + * Decompression + ***************************************************************************/ +typedef struct { + void* srcBuffer; + size_t srcBufferSize; + size_t srcBufferLoaded; + void* dstBuffer; + size_t dstBufferSize; + ZSTD_DStream* dctx; + FILE* dstFile; +} dRess_t; + +static dRess_t FIO_createDResources(FIO_prefs_t* const prefs, const char* dictFileName) +{ + dRess_t ress; + memset(&ress, 0, sizeof(ress)); + + if (prefs->patchFromMode) + FIO_adjustMemLimitForPatchFromMode(prefs, UTIL_getFileSize(dictFileName), 0 /* just use the dict size */); + + /* Allocation */ + ress.dctx = ZSTD_createDStream(); + if (ress.dctx==NULL) + EXM_THROW(60, "Error: %s : can't create ZSTD_DStream", strerror(errno)); + CHECK( ZSTD_DCtx_setMaxWindowSize(ress.dctx, prefs->memLimit) ); + CHECK( ZSTD_DCtx_setParameter(ress.dctx, ZSTD_d_forceIgnoreChecksum, !prefs->checksumFlag)); + + ress.srcBufferSize = ZSTD_DStreamInSize(); + ress.srcBuffer = malloc(ress.srcBufferSize); + ress.dstBufferSize = ZSTD_DStreamOutSize(); + ress.dstBuffer = malloc(ress.dstBufferSize); + if (!ress.srcBuffer || !ress.dstBuffer) + EXM_THROW(61, "Allocation error : not enough memory"); + + /* dictionary */ + { void* dictBuffer; + size_t const dictBufferSize = FIO_createDictBuffer(&dictBuffer, dictFileName, prefs); + CHECK( ZSTD_initDStream_usingDict(ress.dctx, dictBuffer, dictBufferSize) ); + free(dictBuffer); + } + + return ress; +} + +static void FIO_freeDResources(dRess_t ress) +{ + CHECK( ZSTD_freeDStream(ress.dctx) ); + free(ress.srcBuffer); + free(ress.dstBuffer); +} + + +/** FIO_fwriteSparse() : +* @return : storedSkips, +* argument for next call to FIO_fwriteSparse() or FIO_fwriteSparseEnd() */ +static unsigned +FIO_fwriteSparse(FILE* file, + const void* buffer, size_t bufferSize, + const FIO_prefs_t* const prefs, + unsigned storedSkips) +{ + const size_t* const bufferT = (const size_t*)buffer; /* Buffer is supposed malloc'ed, hence aligned on size_t */ + size_t bufferSizeT = bufferSize / sizeof(size_t); + const size_t* const bufferTEnd = bufferT + bufferSizeT; + const size_t* ptrT = bufferT; + static const size_t segmentSizeT = (32 KB) / sizeof(size_t); /* check every 32 KB */ + + if (prefs->testMode) return 0; /* do not output anything in test mode */ + + if (!prefs->sparseFileSupport) { /* normal write */ + size_t const sizeCheck = fwrite(buffer, 1, bufferSize, file); + if (sizeCheck != bufferSize) + EXM_THROW(70, "Write error : cannot write decoded block : %s", + strerror(errno)); + return 0; + } + + /* avoid int overflow */ + if (storedSkips > 1 GB) { + if (LONG_SEEK(file, 1 GB, SEEK_CUR) != 0) + EXM_THROW(91, "1 GB skip error (sparse file support)"); + storedSkips -= 1 GB; + } + + while (ptrT < bufferTEnd) { + size_t nb0T; + + /* adjust last segment if < 32 KB */ + size_t seg0SizeT = segmentSizeT; + if (seg0SizeT > bufferSizeT) seg0SizeT = bufferSizeT; + bufferSizeT -= seg0SizeT; + + /* count leading zeroes */ + for (nb0T=0; (nb0T < seg0SizeT) && (ptrT[nb0T] == 0); nb0T++) ; + storedSkips += (unsigned)(nb0T * sizeof(size_t)); + + if (nb0T != seg0SizeT) { /* not all 0s */ + size_t const nbNon0ST = seg0SizeT - nb0T; + /* skip leading zeros */ + if (LONG_SEEK(file, storedSkips, SEEK_CUR) != 0) + EXM_THROW(92, "Sparse skip error ; try --no-sparse"); + storedSkips = 0; + /* write the rest */ + if (fwrite(ptrT + nb0T, sizeof(size_t), nbNon0ST, file) != nbNon0ST) + EXM_THROW(93, "Write error : cannot write decoded block : %s", + strerror(errno)); + } + ptrT += seg0SizeT; + } + + { static size_t const maskT = sizeof(size_t)-1; + if (bufferSize & maskT) { + /* size not multiple of sizeof(size_t) : implies end of block */ + const char* const restStart = (const char*)bufferTEnd; + const char* restPtr = restStart; + const char* const restEnd = (const char*)buffer + bufferSize; + assert(restEnd > restStart && restEnd < restStart + sizeof(size_t)); + for ( ; (restPtr < restEnd) && (*restPtr == 0); restPtr++) ; + storedSkips += (unsigned) (restPtr - restStart); + if (restPtr != restEnd) { + /* not all remaining bytes are 0 */ + size_t const restSize = (size_t)(restEnd - restPtr); + if (LONG_SEEK(file, storedSkips, SEEK_CUR) != 0) + EXM_THROW(92, "Sparse skip error ; try --no-sparse"); + if (fwrite(restPtr, 1, restSize, file) != restSize) + EXM_THROW(95, "Write error : cannot write end of decoded block : %s", + strerror(errno)); + storedSkips = 0; + } } } + + return storedSkips; +} + +static void +FIO_fwriteSparseEnd(const FIO_prefs_t* const prefs, FILE* file, unsigned storedSkips) +{ + if (prefs->testMode) assert(storedSkips == 0); + if (storedSkips>0) { + assert(prefs->sparseFileSupport > 0); /* storedSkips>0 implies sparse support is enabled */ + (void)prefs; /* assert can be disabled, in which case prefs becomes unused */ + if (LONG_SEEK(file, storedSkips-1, SEEK_CUR) != 0) + EXM_THROW(69, "Final skip error (sparse file support)"); + /* last zero must be explicitly written, + * so that skipped ones get implicitly translated as zero by FS */ + { const char lastZeroByte[1] = { 0 }; + if (fwrite(lastZeroByte, 1, 1, file) != 1) + EXM_THROW(69, "Write error : cannot write last zero : %s", strerror(errno)); + } } +} + + +/** FIO_passThrough() : just copy input into output, for compatibility with gzip -df mode + @return : 0 (no error) */ +static int FIO_passThrough(const FIO_prefs_t* const prefs, + FILE* foutput, FILE* finput, + void* buffer, size_t bufferSize, + size_t alreadyLoaded) +{ + size_t const blockSize = MIN(64 KB, bufferSize); + size_t readFromInput; + unsigned storedSkips = 0; + + /* assumption : ress->srcBufferLoaded bytes already loaded and stored within buffer */ + { size_t const sizeCheck = fwrite(buffer, 1, alreadyLoaded, foutput); + if (sizeCheck != alreadyLoaded) { + DISPLAYLEVEL(1, "Pass-through write error : %s\n", strerror(errno)); + return 1; + } } + + do { + readFromInput = fread(buffer, 1, blockSize, finput); + storedSkips = FIO_fwriteSparse(foutput, buffer, readFromInput, prefs, storedSkips); + } while (readFromInput == blockSize); + if (ferror(finput)) { + DISPLAYLEVEL(1, "Pass-through read error : %s\n", strerror(errno)); + return 1; + } + assert(feof(finput)); + + FIO_fwriteSparseEnd(prefs, foutput, storedSkips); + return 0; +} + +/* FIO_zstdErrorHelp() : + * detailed error message when requested window size is too large */ +static void +FIO_zstdErrorHelp(const FIO_prefs_t* const prefs, + const dRess_t* ress, + size_t err, const char* srcFileName) +{ + ZSTD_frameHeader header; + + /* Help message only for one specific error */ + if (ZSTD_getErrorCode(err) != ZSTD_error_frameParameter_windowTooLarge) + return; + + /* Try to decode the frame header */ + err = ZSTD_getFrameHeader(&header, ress->srcBuffer, ress->srcBufferLoaded); + if (err == 0) { + unsigned long long const windowSize = header.windowSize; + unsigned const windowLog = FIO_highbit64(windowSize) + ((windowSize & (windowSize - 1)) != 0); + assert(prefs->memLimit > 0); + DISPLAYLEVEL(1, "%s : Window size larger than maximum : %llu > %u \n", + srcFileName, windowSize, prefs->memLimit); + if (windowLog <= ZSTD_WINDOWLOG_MAX) { + unsigned const windowMB = (unsigned)((windowSize >> 20) + ((windowSize & ((1 MB) - 1)) != 0)); + assert(windowSize < (U64)(1ULL << 52)); /* ensure now overflow for windowMB */ + DISPLAYLEVEL(1, "%s : Use --long=%u or --memory=%uMB \n", + srcFileName, windowLog, windowMB); + return; + } } + DISPLAYLEVEL(1, "%s : Window log larger than ZSTD_WINDOWLOG_MAX=%u; not supported \n", + srcFileName, ZSTD_WINDOWLOG_MAX); +} + +/** FIO_decompressFrame() : + * @return : size of decoded zstd frame, or an error code + */ +#define FIO_ERROR_FRAME_DECODING ((unsigned long long)(-2)) +static unsigned long long +FIO_decompressZstdFrame(FIO_ctx_t* const fCtx, dRess_t* ress, FILE* finput, + const FIO_prefs_t* const prefs, + const char* srcFileName, + U64 alreadyDecoded) /* for multi-frames streams */ +{ + U64 frameSize = 0; + U32 storedSkips = 0; + + /* display last 20 characters only */ + { size_t const srcFileLength = strlen(srcFileName); + if (srcFileLength>20) srcFileName += srcFileLength-20; + } + + ZSTD_DCtx_reset(ress->dctx, ZSTD_reset_session_only); + + /* Header loading : ensures ZSTD_getFrameHeader() will succeed */ + { size_t const toDecode = ZSTD_FRAMEHEADERSIZE_MAX; + if (ress->srcBufferLoaded < toDecode) { + size_t const toRead = toDecode - ress->srcBufferLoaded; + void* const startPosition = (char*)ress->srcBuffer + ress->srcBufferLoaded; + ress->srcBufferLoaded += fread(startPosition, 1, toRead, finput); + } } + + /* Main decompression Loop */ + while (1) { + ZSTD_inBuffer inBuff = { ress->srcBuffer, ress->srcBufferLoaded, 0 }; + ZSTD_outBuffer outBuff= { ress->dstBuffer, ress->dstBufferSize, 0 }; + size_t const readSizeHint = ZSTD_decompressStream(ress->dctx, &outBuff, &inBuff); const int displayLevel = (g_display_prefs.progressSetting == FIO_ps_always) ? 1 : 2; UTIL_HumanReadableSize_t const hrs = UTIL_makeHumanReadableSize(alreadyDecoded+frameSize); - if (ZSTD_isError(readSizeHint)) { - DISPLAYLEVEL(1, "%s : Decoding error (36) : %s \n", - srcFileName, ZSTD_getErrorName(readSizeHint)); - FIO_zstdErrorHelp(prefs, ress, readSizeHint, srcFileName); - return FIO_ERROR_FRAME_DECODING; - } - - /* Write block */ - storedSkips = FIO_fwriteSparse(ress->dstFile, ress->dstBuffer, outBuff.pos, prefs, storedSkips); - frameSize += outBuff.pos; + if (ZSTD_isError(readSizeHint)) { + DISPLAYLEVEL(1, "%s : Decoding error (36) : %s \n", + srcFileName, ZSTD_getErrorName(readSizeHint)); + FIO_zstdErrorHelp(prefs, ress, readSizeHint, srcFileName); + return FIO_ERROR_FRAME_DECODING; + } + + /* Write block */ + storedSkips = FIO_fwriteSparse(ress->dstFile, ress->dstBuffer, outBuff.pos, prefs, storedSkips); + frameSize += outBuff.pos; if (fCtx->nbFilesTotal > 1) { size_t srcFileNameSize = strlen(srcFileName); if (srcFileNameSize > 18) { const char* truncatedSrcFileName = srcFileName + srcFileNameSize - 15; DISPLAYUPDATE(displayLevel, "\rDecompress: %2u/%2u files. Current: ...%s : %.*f%s... ", fCtx->currFileIdx+1, fCtx->nbFilesTotal, truncatedSrcFileName, hrs.precision, hrs.value, hrs.suffix); - } else { + } else { DISPLAYUPDATE(displayLevel, "\rDecompress: %2u/%2u files. Current: %s : %.*f%s... ", fCtx->currFileIdx+1, fCtx->nbFilesTotal, srcFileName, hrs.precision, hrs.value, hrs.suffix); - } + } } else { DISPLAYUPDATE(displayLevel, "\r%-20.20s : %.*f%s... ", srcFileName, hrs.precision, hrs.value, hrs.suffix); - } - - if (inBuff.pos > 0) { - memmove(ress->srcBuffer, (char*)ress->srcBuffer + inBuff.pos, inBuff.size - inBuff.pos); - ress->srcBufferLoaded -= inBuff.pos; - } - - if (readSizeHint == 0) break; /* end of frame */ - - /* Fill input buffer */ - { size_t const toDecode = MIN(readSizeHint, ress->srcBufferSize); /* support large skippable frames */ - if (ress->srcBufferLoaded < toDecode) { - size_t const toRead = toDecode - ress->srcBufferLoaded; /* > 0 */ - void* const startPosition = (char*)ress->srcBuffer + ress->srcBufferLoaded; - size_t const readSize = fread(startPosition, 1, toRead, finput); - if (readSize==0) { - DISPLAYLEVEL(1, "%s : Read error (39) : premature end \n", - srcFileName); - return FIO_ERROR_FRAME_DECODING; - } - ress->srcBufferLoaded += readSize; - } } } - - FIO_fwriteSparseEnd(prefs, ress->dstFile, storedSkips); - - return frameSize; -} - - -#ifdef ZSTD_GZDECOMPRESS -static unsigned long long -FIO_decompressGzFrame(dRess_t* ress, FILE* srcFile, - const FIO_prefs_t* const prefs, - const char* srcFileName) -{ - unsigned long long outFileSize = 0; - z_stream strm; - int flush = Z_NO_FLUSH; - int decodingError = 0; - unsigned storedSkips = 0; - - strm.zalloc = Z_NULL; - strm.zfree = Z_NULL; - strm.opaque = Z_NULL; - strm.next_in = 0; - strm.avail_in = 0; - /* see http://www.zlib.net/manual.html */ - if (inflateInit2(&strm, 15 /* maxWindowLogSize */ + 16 /* gzip only */) != Z_OK) - return FIO_ERROR_FRAME_DECODING; - - strm.next_out = (Bytef*)ress->dstBuffer; - strm.avail_out = (uInt)ress->dstBufferSize; - strm.avail_in = (uInt)ress->srcBufferLoaded; - strm.next_in = (z_const unsigned char*)ress->srcBuffer; - - for ( ; ; ) { - int ret; - if (strm.avail_in == 0) { - ress->srcBufferLoaded = fread(ress->srcBuffer, 1, ress->srcBufferSize, srcFile); - if (ress->srcBufferLoaded == 0) flush = Z_FINISH; - strm.next_in = (z_const unsigned char*)ress->srcBuffer; - strm.avail_in = (uInt)ress->srcBufferLoaded; - } - ret = inflate(&strm, flush); - if (ret == Z_BUF_ERROR) { - DISPLAYLEVEL(1, "zstd: %s: premature gz end \n", srcFileName); - decodingError = 1; break; - } - if (ret != Z_OK && ret != Z_STREAM_END) { - DISPLAYLEVEL(1, "zstd: %s: inflate error %d \n", srcFileName, ret); - decodingError = 1; break; - } - { size_t const decompBytes = ress->dstBufferSize - strm.avail_out; - if (decompBytes) { - storedSkips = FIO_fwriteSparse(ress->dstFile, ress->dstBuffer, decompBytes, prefs, storedSkips); - outFileSize += decompBytes; - strm.next_out = (Bytef*)ress->dstBuffer; - strm.avail_out = (uInt)ress->dstBufferSize; - } - } - if (ret == Z_STREAM_END) break; - } - - if (strm.avail_in > 0) - memmove(ress->srcBuffer, strm.next_in, strm.avail_in); - ress->srcBufferLoaded = strm.avail_in; - if ( (inflateEnd(&strm) != Z_OK) /* release resources ; error detected */ - && (decodingError==0) ) { - DISPLAYLEVEL(1, "zstd: %s: inflateEnd error \n", srcFileName); - decodingError = 1; - } - FIO_fwriteSparseEnd(prefs, ress->dstFile, storedSkips); - return decodingError ? FIO_ERROR_FRAME_DECODING : outFileSize; -} -#endif - - -#ifdef ZSTD_LZMADECOMPRESS -static unsigned long long -FIO_decompressLzmaFrame(dRess_t* ress, FILE* srcFile, - const FIO_prefs_t* const prefs, - const char* srcFileName, int plain_lzma) -{ - unsigned long long outFileSize = 0; - lzma_stream strm = LZMA_STREAM_INIT; - lzma_action action = LZMA_RUN; - lzma_ret initRet; - int decodingError = 0; - unsigned storedSkips = 0; - - strm.next_in = 0; - strm.avail_in = 0; - if (plain_lzma) { - initRet = lzma_alone_decoder(&strm, UINT64_MAX); /* LZMA */ - } else { - initRet = lzma_stream_decoder(&strm, UINT64_MAX, 0); /* XZ */ - } - - if (initRet != LZMA_OK) { - DISPLAYLEVEL(1, "zstd: %s: %s error %d \n", - plain_lzma ? "lzma_alone_decoder" : "lzma_stream_decoder", - srcFileName, initRet); - return FIO_ERROR_FRAME_DECODING; - } - - strm.next_out = (BYTE*)ress->dstBuffer; - strm.avail_out = ress->dstBufferSize; - strm.next_in = (BYTE const*)ress->srcBuffer; - strm.avail_in = ress->srcBufferLoaded; - - for ( ; ; ) { - lzma_ret ret; - if (strm.avail_in == 0) { - ress->srcBufferLoaded = fread(ress->srcBuffer, 1, ress->srcBufferSize, srcFile); - if (ress->srcBufferLoaded == 0) action = LZMA_FINISH; - strm.next_in = (BYTE const*)ress->srcBuffer; - strm.avail_in = ress->srcBufferLoaded; - } - ret = lzma_code(&strm, action); - - if (ret == LZMA_BUF_ERROR) { - DISPLAYLEVEL(1, "zstd: %s: premature lzma end \n", srcFileName); - decodingError = 1; break; - } - if (ret != LZMA_OK && ret != LZMA_STREAM_END) { - DISPLAYLEVEL(1, "zstd: %s: lzma_code decoding error %d \n", - srcFileName, ret); - decodingError = 1; break; - } - { size_t const decompBytes = ress->dstBufferSize - strm.avail_out; - if (decompBytes) { - storedSkips = FIO_fwriteSparse(ress->dstFile, ress->dstBuffer, decompBytes, prefs, storedSkips); - outFileSize += decompBytes; - strm.next_out = (BYTE*)ress->dstBuffer; - strm.avail_out = ress->dstBufferSize; - } } - if (ret == LZMA_STREAM_END) break; - } - - if (strm.avail_in > 0) - memmove(ress->srcBuffer, strm.next_in, strm.avail_in); - ress->srcBufferLoaded = strm.avail_in; - lzma_end(&strm); - FIO_fwriteSparseEnd(prefs, ress->dstFile, storedSkips); - return decodingError ? FIO_ERROR_FRAME_DECODING : outFileSize; -} -#endif - -#ifdef ZSTD_LZ4DECOMPRESS -static unsigned long long -FIO_decompressLz4Frame(dRess_t* ress, FILE* srcFile, - const FIO_prefs_t* const prefs, - const char* srcFileName) -{ - unsigned long long filesize = 0; - LZ4F_errorCode_t nextToLoad; - LZ4F_decompressionContext_t dCtx; - LZ4F_errorCode_t const errorCode = LZ4F_createDecompressionContext(&dCtx, LZ4F_VERSION); - int decodingError = 0; - unsigned storedSkips = 0; - - if (LZ4F_isError(errorCode)) { - DISPLAYLEVEL(1, "zstd: failed to create lz4 decompression context \n"); - return FIO_ERROR_FRAME_DECODING; - } - - /* Init feed with magic number (already consumed from FILE* sFile) */ - { size_t inSize = 4; - size_t outSize= 0; - MEM_writeLE32(ress->srcBuffer, LZ4_MAGICNUMBER); - nextToLoad = LZ4F_decompress(dCtx, ress->dstBuffer, &outSize, ress->srcBuffer, &inSize, NULL); - if (LZ4F_isError(nextToLoad)) { - DISPLAYLEVEL(1, "zstd: %s: lz4 header error : %s \n", - srcFileName, LZ4F_getErrorName(nextToLoad)); - LZ4F_freeDecompressionContext(dCtx); - return FIO_ERROR_FRAME_DECODING; - } } - - /* Main Loop */ - for (;nextToLoad;) { - size_t readSize; - size_t pos = 0; - size_t decodedBytes = ress->dstBufferSize; - - /* Read input */ - if (nextToLoad > ress->srcBufferSize) nextToLoad = ress->srcBufferSize; - readSize = fread(ress->srcBuffer, 1, nextToLoad, srcFile); - if (!readSize) break; /* reached end of file or stream */ - - while ((pos < readSize) || (decodedBytes == ress->dstBufferSize)) { /* still to read, or still to flush */ - /* Decode Input (at least partially) */ - size_t remaining = readSize - pos; - decodedBytes = ress->dstBufferSize; - nextToLoad = LZ4F_decompress(dCtx, ress->dstBuffer, &decodedBytes, (char*)(ress->srcBuffer)+pos, &remaining, NULL); - if (LZ4F_isError(nextToLoad)) { - DISPLAYLEVEL(1, "zstd: %s: lz4 decompression error : %s \n", - srcFileName, LZ4F_getErrorName(nextToLoad)); - decodingError = 1; nextToLoad = 0; break; - } - pos += remaining; - - /* Write Block */ - if (decodedBytes) { + } + + if (inBuff.pos > 0) { + memmove(ress->srcBuffer, (char*)ress->srcBuffer + inBuff.pos, inBuff.size - inBuff.pos); + ress->srcBufferLoaded -= inBuff.pos; + } + + if (readSizeHint == 0) break; /* end of frame */ + + /* Fill input buffer */ + { size_t const toDecode = MIN(readSizeHint, ress->srcBufferSize); /* support large skippable frames */ + if (ress->srcBufferLoaded < toDecode) { + size_t const toRead = toDecode - ress->srcBufferLoaded; /* > 0 */ + void* const startPosition = (char*)ress->srcBuffer + ress->srcBufferLoaded; + size_t const readSize = fread(startPosition, 1, toRead, finput); + if (readSize==0) { + DISPLAYLEVEL(1, "%s : Read error (39) : premature end \n", + srcFileName); + return FIO_ERROR_FRAME_DECODING; + } + ress->srcBufferLoaded += readSize; + } } } + + FIO_fwriteSparseEnd(prefs, ress->dstFile, storedSkips); + + return frameSize; +} + + +#ifdef ZSTD_GZDECOMPRESS +static unsigned long long +FIO_decompressGzFrame(dRess_t* ress, FILE* srcFile, + const FIO_prefs_t* const prefs, + const char* srcFileName) +{ + unsigned long long outFileSize = 0; + z_stream strm; + int flush = Z_NO_FLUSH; + int decodingError = 0; + unsigned storedSkips = 0; + + strm.zalloc = Z_NULL; + strm.zfree = Z_NULL; + strm.opaque = Z_NULL; + strm.next_in = 0; + strm.avail_in = 0; + /* see http://www.zlib.net/manual.html */ + if (inflateInit2(&strm, 15 /* maxWindowLogSize */ + 16 /* gzip only */) != Z_OK) + return FIO_ERROR_FRAME_DECODING; + + strm.next_out = (Bytef*)ress->dstBuffer; + strm.avail_out = (uInt)ress->dstBufferSize; + strm.avail_in = (uInt)ress->srcBufferLoaded; + strm.next_in = (z_const unsigned char*)ress->srcBuffer; + + for ( ; ; ) { + int ret; + if (strm.avail_in == 0) { + ress->srcBufferLoaded = fread(ress->srcBuffer, 1, ress->srcBufferSize, srcFile); + if (ress->srcBufferLoaded == 0) flush = Z_FINISH; + strm.next_in = (z_const unsigned char*)ress->srcBuffer; + strm.avail_in = (uInt)ress->srcBufferLoaded; + } + ret = inflate(&strm, flush); + if (ret == Z_BUF_ERROR) { + DISPLAYLEVEL(1, "zstd: %s: premature gz end \n", srcFileName); + decodingError = 1; break; + } + if (ret != Z_OK && ret != Z_STREAM_END) { + DISPLAYLEVEL(1, "zstd: %s: inflate error %d \n", srcFileName, ret); + decodingError = 1; break; + } + { size_t const decompBytes = ress->dstBufferSize - strm.avail_out; + if (decompBytes) { + storedSkips = FIO_fwriteSparse(ress->dstFile, ress->dstBuffer, decompBytes, prefs, storedSkips); + outFileSize += decompBytes; + strm.next_out = (Bytef*)ress->dstBuffer; + strm.avail_out = (uInt)ress->dstBufferSize; + } + } + if (ret == Z_STREAM_END) break; + } + + if (strm.avail_in > 0) + memmove(ress->srcBuffer, strm.next_in, strm.avail_in); + ress->srcBufferLoaded = strm.avail_in; + if ( (inflateEnd(&strm) != Z_OK) /* release resources ; error detected */ + && (decodingError==0) ) { + DISPLAYLEVEL(1, "zstd: %s: inflateEnd error \n", srcFileName); + decodingError = 1; + } + FIO_fwriteSparseEnd(prefs, ress->dstFile, storedSkips); + return decodingError ? FIO_ERROR_FRAME_DECODING : outFileSize; +} +#endif + + +#ifdef ZSTD_LZMADECOMPRESS +static unsigned long long +FIO_decompressLzmaFrame(dRess_t* ress, FILE* srcFile, + const FIO_prefs_t* const prefs, + const char* srcFileName, int plain_lzma) +{ + unsigned long long outFileSize = 0; + lzma_stream strm = LZMA_STREAM_INIT; + lzma_action action = LZMA_RUN; + lzma_ret initRet; + int decodingError = 0; + unsigned storedSkips = 0; + + strm.next_in = 0; + strm.avail_in = 0; + if (plain_lzma) { + initRet = lzma_alone_decoder(&strm, UINT64_MAX); /* LZMA */ + } else { + initRet = lzma_stream_decoder(&strm, UINT64_MAX, 0); /* XZ */ + } + + if (initRet != LZMA_OK) { + DISPLAYLEVEL(1, "zstd: %s: %s error %d \n", + plain_lzma ? "lzma_alone_decoder" : "lzma_stream_decoder", + srcFileName, initRet); + return FIO_ERROR_FRAME_DECODING; + } + + strm.next_out = (BYTE*)ress->dstBuffer; + strm.avail_out = ress->dstBufferSize; + strm.next_in = (BYTE const*)ress->srcBuffer; + strm.avail_in = ress->srcBufferLoaded; + + for ( ; ; ) { + lzma_ret ret; + if (strm.avail_in == 0) { + ress->srcBufferLoaded = fread(ress->srcBuffer, 1, ress->srcBufferSize, srcFile); + if (ress->srcBufferLoaded == 0) action = LZMA_FINISH; + strm.next_in = (BYTE const*)ress->srcBuffer; + strm.avail_in = ress->srcBufferLoaded; + } + ret = lzma_code(&strm, action); + + if (ret == LZMA_BUF_ERROR) { + DISPLAYLEVEL(1, "zstd: %s: premature lzma end \n", srcFileName); + decodingError = 1; break; + } + if (ret != LZMA_OK && ret != LZMA_STREAM_END) { + DISPLAYLEVEL(1, "zstd: %s: lzma_code decoding error %d \n", + srcFileName, ret); + decodingError = 1; break; + } + { size_t const decompBytes = ress->dstBufferSize - strm.avail_out; + if (decompBytes) { + storedSkips = FIO_fwriteSparse(ress->dstFile, ress->dstBuffer, decompBytes, prefs, storedSkips); + outFileSize += decompBytes; + strm.next_out = (BYTE*)ress->dstBuffer; + strm.avail_out = ress->dstBufferSize; + } } + if (ret == LZMA_STREAM_END) break; + } + + if (strm.avail_in > 0) + memmove(ress->srcBuffer, strm.next_in, strm.avail_in); + ress->srcBufferLoaded = strm.avail_in; + lzma_end(&strm); + FIO_fwriteSparseEnd(prefs, ress->dstFile, storedSkips); + return decodingError ? FIO_ERROR_FRAME_DECODING : outFileSize; +} +#endif + +#ifdef ZSTD_LZ4DECOMPRESS +static unsigned long long +FIO_decompressLz4Frame(dRess_t* ress, FILE* srcFile, + const FIO_prefs_t* const prefs, + const char* srcFileName) +{ + unsigned long long filesize = 0; + LZ4F_errorCode_t nextToLoad; + LZ4F_decompressionContext_t dCtx; + LZ4F_errorCode_t const errorCode = LZ4F_createDecompressionContext(&dCtx, LZ4F_VERSION); + int decodingError = 0; + unsigned storedSkips = 0; + + if (LZ4F_isError(errorCode)) { + DISPLAYLEVEL(1, "zstd: failed to create lz4 decompression context \n"); + return FIO_ERROR_FRAME_DECODING; + } + + /* Init feed with magic number (already consumed from FILE* sFile) */ + { size_t inSize = 4; + size_t outSize= 0; + MEM_writeLE32(ress->srcBuffer, LZ4_MAGICNUMBER); + nextToLoad = LZ4F_decompress(dCtx, ress->dstBuffer, &outSize, ress->srcBuffer, &inSize, NULL); + if (LZ4F_isError(nextToLoad)) { + DISPLAYLEVEL(1, "zstd: %s: lz4 header error : %s \n", + srcFileName, LZ4F_getErrorName(nextToLoad)); + LZ4F_freeDecompressionContext(dCtx); + return FIO_ERROR_FRAME_DECODING; + } } + + /* Main Loop */ + for (;nextToLoad;) { + size_t readSize; + size_t pos = 0; + size_t decodedBytes = ress->dstBufferSize; + + /* Read input */ + if (nextToLoad > ress->srcBufferSize) nextToLoad = ress->srcBufferSize; + readSize = fread(ress->srcBuffer, 1, nextToLoad, srcFile); + if (!readSize) break; /* reached end of file or stream */ + + while ((pos < readSize) || (decodedBytes == ress->dstBufferSize)) { /* still to read, or still to flush */ + /* Decode Input (at least partially) */ + size_t remaining = readSize - pos; + decodedBytes = ress->dstBufferSize; + nextToLoad = LZ4F_decompress(dCtx, ress->dstBuffer, &decodedBytes, (char*)(ress->srcBuffer)+pos, &remaining, NULL); + if (LZ4F_isError(nextToLoad)) { + DISPLAYLEVEL(1, "zstd: %s: lz4 decompression error : %s \n", + srcFileName, LZ4F_getErrorName(nextToLoad)); + decodingError = 1; nextToLoad = 0; break; + } + pos += remaining; + + /* Write Block */ + if (decodedBytes) { UTIL_HumanReadableSize_t hrs; - storedSkips = FIO_fwriteSparse(ress->dstFile, ress->dstBuffer, decodedBytes, prefs, storedSkips); - filesize += decodedBytes; + storedSkips = FIO_fwriteSparse(ress->dstFile, ress->dstBuffer, decodedBytes, prefs, storedSkips); + filesize += decodedBytes; hrs = UTIL_makeHumanReadableSize(filesize); DISPLAYUPDATE(2, "\rDecompressed : %.*f%s ", hrs.precision, hrs.value, hrs.suffix); - } - - if (!nextToLoad) break; - } - } - /* can be out because readSize == 0, which could be an fread() error */ - if (ferror(srcFile)) { - DISPLAYLEVEL(1, "zstd: %s: read error \n", srcFileName); - decodingError=1; - } - - if (nextToLoad!=0) { - DISPLAYLEVEL(1, "zstd: %s: unfinished lz4 stream \n", srcFileName); - decodingError=1; - } - - LZ4F_freeDecompressionContext(dCtx); - ress->srcBufferLoaded = 0; /* LZ4F will reach exact frame boundary */ - FIO_fwriteSparseEnd(prefs, ress->dstFile, storedSkips); - - return decodingError ? FIO_ERROR_FRAME_DECODING : filesize; -} -#endif - - - -/** FIO_decompressFrames() : - * Find and decode frames inside srcFile - * srcFile presumed opened and valid - * @return : 0 : OK - * 1 : error - */ -static int FIO_decompressFrames(FIO_ctx_t* const fCtx, - dRess_t ress, FILE* srcFile, - const FIO_prefs_t* const prefs, - const char* dstFileName, const char* srcFileName) -{ - unsigned readSomething = 0; - unsigned long long filesize = 0; - assert(srcFile != NULL); - - /* for each frame */ - for ( ; ; ) { - /* check magic number -> version */ - size_t const toRead = 4; - const BYTE* const buf = (const BYTE*)ress.srcBuffer; - if (ress.srcBufferLoaded < toRead) /* load up to 4 bytes for header */ - ress.srcBufferLoaded += fread((char*)ress.srcBuffer + ress.srcBufferLoaded, - (size_t)1, toRead - ress.srcBufferLoaded, srcFile); - if (ress.srcBufferLoaded==0) { - if (readSomething==0) { /* srcFile is empty (which is invalid) */ - DISPLAYLEVEL(1, "zstd: %s: unexpected end of file \n", srcFileName); - return 1; - } /* else, just reached frame boundary */ - break; /* no more input */ - } - readSomething = 1; /* there is at least 1 byte in srcFile */ - if (ress.srcBufferLoaded < toRead) { - DISPLAYLEVEL(1, "zstd: %s: unknown header \n", srcFileName); - return 1; - } - if (ZSTD_isFrame(buf, ress.srcBufferLoaded)) { - unsigned long long const frameSize = FIO_decompressZstdFrame(fCtx, &ress, srcFile, prefs, srcFileName, filesize); - if (frameSize == FIO_ERROR_FRAME_DECODING) return 1; - filesize += frameSize; - } else if (buf[0] == 31 && buf[1] == 139) { /* gz magic number */ -#ifdef ZSTD_GZDECOMPRESS - unsigned long long const frameSize = FIO_decompressGzFrame(&ress, srcFile, prefs, srcFileName); - if (frameSize == FIO_ERROR_FRAME_DECODING) return 1; - filesize += frameSize; -#else - DISPLAYLEVEL(1, "zstd: %s: gzip file cannot be uncompressed (zstd compiled without HAVE_ZLIB) -- ignored \n", srcFileName); - return 1; -#endif - } else if ((buf[0] == 0xFD && buf[1] == 0x37) /* xz magic number */ - || (buf[0] == 0x5D && buf[1] == 0x00)) { /* lzma header (no magic number) */ -#ifdef ZSTD_LZMADECOMPRESS - unsigned long long const frameSize = FIO_decompressLzmaFrame(&ress, srcFile, prefs, srcFileName, buf[0] != 0xFD); - if (frameSize == FIO_ERROR_FRAME_DECODING) return 1; - filesize += frameSize; -#else - DISPLAYLEVEL(1, "zstd: %s: xz/lzma file cannot be uncompressed (zstd compiled without HAVE_LZMA) -- ignored \n", srcFileName); - return 1; -#endif - } else if (MEM_readLE32(buf) == LZ4_MAGICNUMBER) { -#ifdef ZSTD_LZ4DECOMPRESS - unsigned long long const frameSize = FIO_decompressLz4Frame(&ress, srcFile, prefs, srcFileName); - if (frameSize == FIO_ERROR_FRAME_DECODING) return 1; - filesize += frameSize; -#else - DISPLAYLEVEL(1, "zstd: %s: lz4 file cannot be uncompressed (zstd compiled without HAVE_LZ4) -- ignored \n", srcFileName); - return 1; -#endif - } else if ((prefs->overwrite) && !strcmp (dstFileName, stdoutmark)) { /* pass-through mode */ - return FIO_passThrough(prefs, - ress.dstFile, srcFile, - ress.srcBuffer, ress.srcBufferSize, - ress.srcBufferLoaded); - } else { - DISPLAYLEVEL(1, "zstd: %s: unsupported format \n", srcFileName); - return 1; - } } /* for each frame */ - - /* Final Status */ - fCtx->totalBytesOutput += (size_t)filesize; - DISPLAYLEVEL(2, "\r%79s\r", ""); - /* No status message in pipe mode (stdin - stdout) or multi-files mode */ + } + + if (!nextToLoad) break; + } + } + /* can be out because readSize == 0, which could be an fread() error */ + if (ferror(srcFile)) { + DISPLAYLEVEL(1, "zstd: %s: read error \n", srcFileName); + decodingError=1; + } + + if (nextToLoad!=0) { + DISPLAYLEVEL(1, "zstd: %s: unfinished lz4 stream \n", srcFileName); + decodingError=1; + } + + LZ4F_freeDecompressionContext(dCtx); + ress->srcBufferLoaded = 0; /* LZ4F will reach exact frame boundary */ + FIO_fwriteSparseEnd(prefs, ress->dstFile, storedSkips); + + return decodingError ? FIO_ERROR_FRAME_DECODING : filesize; +} +#endif + + + +/** FIO_decompressFrames() : + * Find and decode frames inside srcFile + * srcFile presumed opened and valid + * @return : 0 : OK + * 1 : error + */ +static int FIO_decompressFrames(FIO_ctx_t* const fCtx, + dRess_t ress, FILE* srcFile, + const FIO_prefs_t* const prefs, + const char* dstFileName, const char* srcFileName) +{ + unsigned readSomething = 0; + unsigned long long filesize = 0; + assert(srcFile != NULL); + + /* for each frame */ + for ( ; ; ) { + /* check magic number -> version */ + size_t const toRead = 4; + const BYTE* const buf = (const BYTE*)ress.srcBuffer; + if (ress.srcBufferLoaded < toRead) /* load up to 4 bytes for header */ + ress.srcBufferLoaded += fread((char*)ress.srcBuffer + ress.srcBufferLoaded, + (size_t)1, toRead - ress.srcBufferLoaded, srcFile); + if (ress.srcBufferLoaded==0) { + if (readSomething==0) { /* srcFile is empty (which is invalid) */ + DISPLAYLEVEL(1, "zstd: %s: unexpected end of file \n", srcFileName); + return 1; + } /* else, just reached frame boundary */ + break; /* no more input */ + } + readSomething = 1; /* there is at least 1 byte in srcFile */ + if (ress.srcBufferLoaded < toRead) { + DISPLAYLEVEL(1, "zstd: %s: unknown header \n", srcFileName); + return 1; + } + if (ZSTD_isFrame(buf, ress.srcBufferLoaded)) { + unsigned long long const frameSize = FIO_decompressZstdFrame(fCtx, &ress, srcFile, prefs, srcFileName, filesize); + if (frameSize == FIO_ERROR_FRAME_DECODING) return 1; + filesize += frameSize; + } else if (buf[0] == 31 && buf[1] == 139) { /* gz magic number */ +#ifdef ZSTD_GZDECOMPRESS + unsigned long long const frameSize = FIO_decompressGzFrame(&ress, srcFile, prefs, srcFileName); + if (frameSize == FIO_ERROR_FRAME_DECODING) return 1; + filesize += frameSize; +#else + DISPLAYLEVEL(1, "zstd: %s: gzip file cannot be uncompressed (zstd compiled without HAVE_ZLIB) -- ignored \n", srcFileName); + return 1; +#endif + } else if ((buf[0] == 0xFD && buf[1] == 0x37) /* xz magic number */ + || (buf[0] == 0x5D && buf[1] == 0x00)) { /* lzma header (no magic number) */ +#ifdef ZSTD_LZMADECOMPRESS + unsigned long long const frameSize = FIO_decompressLzmaFrame(&ress, srcFile, prefs, srcFileName, buf[0] != 0xFD); + if (frameSize == FIO_ERROR_FRAME_DECODING) return 1; + filesize += frameSize; +#else + DISPLAYLEVEL(1, "zstd: %s: xz/lzma file cannot be uncompressed (zstd compiled without HAVE_LZMA) -- ignored \n", srcFileName); + return 1; +#endif + } else if (MEM_readLE32(buf) == LZ4_MAGICNUMBER) { +#ifdef ZSTD_LZ4DECOMPRESS + unsigned long long const frameSize = FIO_decompressLz4Frame(&ress, srcFile, prefs, srcFileName); + if (frameSize == FIO_ERROR_FRAME_DECODING) return 1; + filesize += frameSize; +#else + DISPLAYLEVEL(1, "zstd: %s: lz4 file cannot be uncompressed (zstd compiled without HAVE_LZ4) -- ignored \n", srcFileName); + return 1; +#endif + } else if ((prefs->overwrite) && !strcmp (dstFileName, stdoutmark)) { /* pass-through mode */ + return FIO_passThrough(prefs, + ress.dstFile, srcFile, + ress.srcBuffer, ress.srcBufferSize, + ress.srcBufferLoaded); + } else { + DISPLAYLEVEL(1, "zstd: %s: unsupported format \n", srcFileName); + return 1; + } } /* for each frame */ + + /* Final Status */ + fCtx->totalBytesOutput += (size_t)filesize; + DISPLAYLEVEL(2, "\r%79s\r", ""); + /* No status message in pipe mode (stdin - stdout) or multi-files mode */ if ((g_display_prefs.displayLevel >= 2 && fCtx->nbFilesTotal <= 1) || g_display_prefs.displayLevel >= 3 || g_display_prefs.progressSetting == FIO_ps_always) { DISPLAYLEVEL(1, "\r%-20s: %llu bytes \n", srcFileName, filesize); - } - - return 0; -} - -/** FIO_decompressDstFile() : - open `dstFileName`, - or path-through if ress.dstFile is already != 0, - then start decompression process (FIO_decompressFrames()). - @return : 0 : OK - 1 : operation aborted -*/ -static int FIO_decompressDstFile(FIO_ctx_t* const fCtx, - FIO_prefs_t* const prefs, - dRess_t ress, FILE* srcFile, - const char* dstFileName, const char* srcFileName) -{ - int result; - stat_t statbuf; - int releaseDstFile = 0; + } + + return 0; +} + +/** FIO_decompressDstFile() : + open `dstFileName`, + or path-through if ress.dstFile is already != 0, + then start decompression process (FIO_decompressFrames()). + @return : 0 : OK + 1 : operation aborted +*/ +static int FIO_decompressDstFile(FIO_ctx_t* const fCtx, + FIO_prefs_t* const prefs, + dRess_t ress, FILE* srcFile, + const char* dstFileName, const char* srcFileName) +{ + int result; + stat_t statbuf; + int releaseDstFile = 0; int transferMTime = 0; - - if ((ress.dstFile == NULL) && (prefs->testMode==0)) { - int dstFilePermissions = DEFAULT_FILE_PERMISSIONS; - if ( strcmp(srcFileName, stdinmark) /* special case : don't transfer permissions from stdin */ + + if ((ress.dstFile == NULL) && (prefs->testMode==0)) { + int dstFilePermissions = DEFAULT_FILE_PERMISSIONS; + if ( strcmp(srcFileName, stdinmark) /* special case : don't transfer permissions from stdin */ && strcmp(dstFileName, stdoutmark) - && UTIL_stat(srcFileName, &statbuf) - && UTIL_isRegularFileStat(&statbuf) ) { - dstFilePermissions = statbuf.st_mode; + && UTIL_stat(srcFileName, &statbuf) + && UTIL_isRegularFileStat(&statbuf) ) { + dstFilePermissions = statbuf.st_mode; transferMTime = 1; - } - - releaseDstFile = 1; - - ress.dstFile = FIO_openDstFile(fCtx, prefs, srcFileName, dstFileName, dstFilePermissions); - if (ress.dstFile==NULL) return 1; - - /* Must only be added after FIO_openDstFile() succeeds. - * Otherwise we may delete the destination file if it already exists, - * and the user presses Ctrl-C when asked if they wish to overwrite. - */ - addHandler(dstFileName); - } - - result = FIO_decompressFrames(fCtx, ress, srcFile, prefs, dstFileName, srcFileName); - - if (releaseDstFile) { - FILE* const dstFile = ress.dstFile; - clearHandler(); - ress.dstFile = NULL; - if (fclose(dstFile)) { - DISPLAYLEVEL(1, "zstd: %s: %s \n", dstFileName, strerror(errno)); - result = 1; - } - + } + + releaseDstFile = 1; + + ress.dstFile = FIO_openDstFile(fCtx, prefs, srcFileName, dstFileName, dstFilePermissions); + if (ress.dstFile==NULL) return 1; + + /* Must only be added after FIO_openDstFile() succeeds. + * Otherwise we may delete the destination file if it already exists, + * and the user presses Ctrl-C when asked if they wish to overwrite. + */ + addHandler(dstFileName); + } + + result = FIO_decompressFrames(fCtx, ress, srcFile, prefs, dstFileName, srcFileName); + + if (releaseDstFile) { + FILE* const dstFile = ress.dstFile; + clearHandler(); + ress.dstFile = NULL; + if (fclose(dstFile)) { + DISPLAYLEVEL(1, "zstd: %s: %s \n", dstFileName, strerror(errno)); + result = 1; + } + if (transferMTime) { UTIL_utime(dstFileName, &statbuf); } - if ( (result != 0) /* operation failure */ - && strcmp(dstFileName, stdoutmark) /* special case : don't remove() stdout */ - ) { - FIO_removeFile(dstFileName); /* remove decompression artefact; note: don't do anything special if remove() fails */ - } - } - - return result; -} - - -/** FIO_decompressSrcFile() : - Open `srcFileName`, transfer control to decompressDstFile() - @return : 0 : OK - 1 : error -*/ -static int FIO_decompressSrcFile(FIO_ctx_t* const fCtx, FIO_prefs_t* const prefs, dRess_t ress, const char* dstFileName, const char* srcFileName) -{ - FILE* srcFile; - int result; - - if (UTIL_isDirectory(srcFileName)) { - DISPLAYLEVEL(1, "zstd: %s is a directory -- ignored \n", srcFileName); - return 1; - } - - srcFile = FIO_openSrcFile(prefs, srcFileName); - if (srcFile==NULL) return 1; - ress.srcBufferLoaded = 0; - - result = FIO_decompressDstFile(fCtx, prefs, ress, srcFile, dstFileName, srcFileName); - - /* Close file */ - if (fclose(srcFile)) { - DISPLAYLEVEL(1, "zstd: %s: %s \n", srcFileName, strerror(errno)); /* error should not happen */ - return 1; - } - if ( prefs->removeSrcFile /* --rm */ - && (result==0) /* decompression successful */ - && strcmp(srcFileName, stdinmark) ) /* not stdin */ { - /* We must clear the handler, since after this point calling it would - * delete both the source and destination files. - */ - clearHandler(); - if (FIO_removeFile(srcFileName)) { - /* failed to remove src file */ - DISPLAYLEVEL(1, "zstd: %s: %s \n", srcFileName, strerror(errno)); - return 1; - } } - return result; -} - - - -int FIO_decompressFilename(FIO_ctx_t* const fCtx, FIO_prefs_t* const prefs, - const char* dstFileName, const char* srcFileName, - const char* dictFileName) -{ - dRess_t const ress = FIO_createDResources(prefs, dictFileName); - - int const decodingError = FIO_decompressSrcFile(fCtx, prefs, ress, dstFileName, srcFileName); - - FIO_freeDResources(ress); - return decodingError; -} - -static const char *suffixList[] = { - ZSTD_EXTENSION, - TZSTD_EXTENSION, -#ifndef ZSTD_NODECOMPRESS - ZSTD_ALT_EXTENSION, -#endif -#ifdef ZSTD_GZDECOMPRESS - GZ_EXTENSION, - TGZ_EXTENSION, -#endif -#ifdef ZSTD_LZMADECOMPRESS - LZMA_EXTENSION, - XZ_EXTENSION, - TXZ_EXTENSION, -#endif -#ifdef ZSTD_LZ4DECOMPRESS - LZ4_EXTENSION, - TLZ4_EXTENSION, -#endif - NULL -}; - -static const char *suffixListStr = - ZSTD_EXTENSION "/" TZSTD_EXTENSION -#ifdef ZSTD_GZDECOMPRESS - "/" GZ_EXTENSION "/" TGZ_EXTENSION -#endif -#ifdef ZSTD_LZMADECOMPRESS - "/" LZMA_EXTENSION "/" XZ_EXTENSION "/" TXZ_EXTENSION -#endif -#ifdef ZSTD_LZ4DECOMPRESS - "/" LZ4_EXTENSION "/" TLZ4_EXTENSION -#endif -; - -/* FIO_determineDstName() : - * create a destination filename from a srcFileName. - * @return a pointer to it. - * @return == NULL if there is an error */ -static const char* -FIO_determineDstName(const char* srcFileName, const char* outDirName) -{ - static size_t dfnbCapacity = 0; - static char* dstFileNameBuffer = NULL; /* using static allocation : this function cannot be multi-threaded */ - size_t dstFileNameEndPos; - char* outDirFilename = NULL; - const char* dstSuffix = ""; - size_t dstSuffixLen = 0; - - size_t sfnSize = strlen(srcFileName); - - size_t srcSuffixLen; - const char* const srcSuffix = strrchr(srcFileName, '.'); - if (srcSuffix == NULL) { - DISPLAYLEVEL(1, - "zstd: %s: unknown suffix (%s expected). " - "Can't derive the output file name. " - "Specify it with -o dstFileName. Ignoring.\n", - srcFileName, suffixListStr); - return NULL; - } - srcSuffixLen = strlen(srcSuffix); - - { - const char** matchedSuffixPtr; - for (matchedSuffixPtr = suffixList; *matchedSuffixPtr != NULL; matchedSuffixPtr++) { - if (!strcmp(*matchedSuffixPtr, srcSuffix)) { - break; - } - } - - /* check suffix is authorized */ - if (sfnSize <= srcSuffixLen || *matchedSuffixPtr == NULL) { - DISPLAYLEVEL(1, - "zstd: %s: unknown suffix (%s expected). " - "Can't derive the output file name. " - "Specify it with -o dstFileName. Ignoring.\n", - srcFileName, suffixListStr); - return NULL; - } - - if ((*matchedSuffixPtr)[1] == 't') { - dstSuffix = ".tar"; - dstSuffixLen = strlen(dstSuffix); - } - } - - if (outDirName) { - outDirFilename = FIO_createFilename_fromOutDir(srcFileName, outDirName, 0); - sfnSize = strlen(outDirFilename); - assert(outDirFilename != NULL); - } - - if (dfnbCapacity+srcSuffixLen <= sfnSize+1+dstSuffixLen) { - /* allocate enough space to write dstFilename into it */ - free(dstFileNameBuffer); - dfnbCapacity = sfnSize + 20; - dstFileNameBuffer = (char*)malloc(dfnbCapacity); - if (dstFileNameBuffer==NULL) - EXM_THROW(74, "%s : not enough memory for dstFileName", - strerror(errno)); - } - - /* return dst name == src name truncated from suffix */ - assert(dstFileNameBuffer != NULL); - dstFileNameEndPos = sfnSize - srcSuffixLen; - if (outDirFilename) { - memcpy(dstFileNameBuffer, outDirFilename, dstFileNameEndPos); - free(outDirFilename); - } else { - memcpy(dstFileNameBuffer, srcFileName, dstFileNameEndPos); - } - - /* The short tar extensions tzst, tgz, txz and tlz4 files should have "tar" - * extension on decompression. Also writes terminating null. */ - strcpy(dstFileNameBuffer + dstFileNameEndPos, dstSuffix); - return dstFileNameBuffer; - - /* note : dstFileNameBuffer memory is not going to be free */ -} - -int -FIO_decompressMultipleFilenames(FIO_ctx_t* const fCtx, - FIO_prefs_t* const prefs, - const char** srcNamesTable, - const char* outMirroredRootDirName, - const char* outDirName, const char* outFileName, - const char* dictFileName) -{ - int status; - int error = 0; - dRess_t ress = FIO_createDResources(prefs, dictFileName); - - if (outFileName) { - if (FIO_removeMultiFilesWarning(fCtx, prefs, outFileName, 1 /* displayLevelCutoff */)) { - FIO_freeDResources(ress); - return 1; - } - if (!prefs->testMode) { - ress.dstFile = FIO_openDstFile(fCtx, prefs, NULL, outFileName, DEFAULT_FILE_PERMISSIONS); - if (ress.dstFile == 0) EXM_THROW(19, "cannot open %s", outFileName); - } - for (; fCtx->currFileIdx < fCtx->nbFilesTotal; fCtx->currFileIdx++) { - status = FIO_decompressSrcFile(fCtx, prefs, ress, outFileName, srcNamesTable[fCtx->currFileIdx]); - if (!status) fCtx->nbFilesProcessed++; - error |= status; - } - if ((!prefs->testMode) && (fclose(ress.dstFile))) - EXM_THROW(72, "Write error : %s : cannot properly close output file", - strerror(errno)); - } else { - if (outMirroredRootDirName) - UTIL_mirrorSourceFilesDirectories(srcNamesTable, (unsigned)fCtx->nbFilesTotal, outMirroredRootDirName); - - for (; fCtx->currFileIdx < fCtx->nbFilesTotal; fCtx->currFileIdx++) { /* create dstFileName */ - const char* const srcFileName = srcNamesTable[fCtx->currFileIdx]; - const char* dstFileName = NULL; - if (outMirroredRootDirName) { - char* validMirroredDirName = UTIL_createMirroredDestDirName(srcFileName, outMirroredRootDirName); - if (validMirroredDirName) { - dstFileName = FIO_determineDstName(srcFileName, validMirroredDirName); - free(validMirroredDirName); - } else { - DISPLAYLEVEL(2, "zstd: --output-dir-mirror cannot decompress '%s' into '%s'\n", srcFileName, outMirroredRootDirName); - } - } else { - dstFileName = FIO_determineDstName(srcFileName, outDirName); - } - if (dstFileName == NULL) { error=1; continue; } - status = FIO_decompressSrcFile(fCtx, prefs, ress, dstFileName, srcFileName); - if (!status) fCtx->nbFilesProcessed++; - error |= status; - } - if (outDirName) - FIO_checkFilenameCollisions(srcNamesTable , (unsigned)fCtx->nbFilesTotal); - } - - if (fCtx->nbFilesProcessed >= 1 && fCtx->nbFilesTotal > 1 && fCtx->totalBytesOutput != 0) - DISPLAYLEVEL(2, "%d files decompressed : %6zu bytes total \n", fCtx->nbFilesProcessed, fCtx->totalBytesOutput); - - FIO_freeDResources(ress); - return error; -} - -/* ************************************************************************** - * .zst file info (--list command) - ***************************************************************************/ - -typedef struct { - U64 decompressedSize; - U64 compressedSize; - U64 windowSize; - int numActualFrames; - int numSkippableFrames; - int decompUnavailable; - int usesCheck; - U32 nbFiles; -} fileInfo_t; - -typedef enum { - info_success=0, - info_frame_error=1, - info_not_zstd=2, - info_file_error=3, - info_truncated_input=4, -} InfoError; - -#define ERROR_IF(c,n,...) { \ - if (c) { \ - DISPLAYLEVEL(1, __VA_ARGS__); \ - DISPLAYLEVEL(1, " \n"); \ - return n; \ - } \ -} - -static InfoError -FIO_analyzeFrames(fileInfo_t* info, FILE* const srcFile) -{ - /* begin analyzing frame */ - for ( ; ; ) { - BYTE headerBuffer[ZSTD_FRAMEHEADERSIZE_MAX]; - size_t const numBytesRead = fread(headerBuffer, 1, sizeof(headerBuffer), srcFile); - if (numBytesRead < ZSTD_FRAMEHEADERSIZE_MIN(ZSTD_f_zstd1)) { - if ( feof(srcFile) - && (numBytesRead == 0) - && (info->compressedSize > 0) - && (info->compressedSize != UTIL_FILESIZE_UNKNOWN) ) { - unsigned long long file_position = (unsigned long long) LONG_TELL(srcFile); - unsigned long long file_size = (unsigned long long) info->compressedSize; - ERROR_IF(file_position != file_size, info_truncated_input, - "Error: seeked to position %llu, which is beyond file size of %llu\n", - file_position, - file_size); - break; /* correct end of file => success */ - } - ERROR_IF(feof(srcFile), info_not_zstd, "Error: reached end of file with incomplete frame"); - ERROR_IF(1, info_frame_error, "Error: did not reach end of file but ran out of frames"); - } - { U32 const magicNumber = MEM_readLE32(headerBuffer); - /* Zstandard frame */ - if (magicNumber == ZSTD_MAGICNUMBER) { - ZSTD_frameHeader header; - U64 const frameContentSize = ZSTD_getFrameContentSize(headerBuffer, numBytesRead); - if ( frameContentSize == ZSTD_CONTENTSIZE_ERROR - || frameContentSize == ZSTD_CONTENTSIZE_UNKNOWN ) { - info->decompUnavailable = 1; - } else { - info->decompressedSize += frameContentSize; - } - ERROR_IF(ZSTD_getFrameHeader(&header, headerBuffer, numBytesRead) != 0, - info_frame_error, "Error: could not decode frame header"); - info->windowSize = header.windowSize; - /* move to the end of the frame header */ - { size_t const headerSize = ZSTD_frameHeaderSize(headerBuffer, numBytesRead); - ERROR_IF(ZSTD_isError(headerSize), info_frame_error, "Error: could not determine frame header size"); - ERROR_IF(fseek(srcFile, ((long)headerSize)-((long)numBytesRead), SEEK_CUR) != 0, - info_frame_error, "Error: could not move to end of frame header"); - } - - /* skip all blocks in the frame */ - { int lastBlock = 0; - do { - BYTE blockHeaderBuffer[3]; - ERROR_IF(fread(blockHeaderBuffer, 1, 3, srcFile) != 3, - info_frame_error, "Error while reading block header"); - { U32 const blockHeader = MEM_readLE24(blockHeaderBuffer); - U32 const blockTypeID = (blockHeader >> 1) & 3; - U32 const isRLE = (blockTypeID == 1); - U32 const isWrongBlock = (blockTypeID == 3); - long const blockSize = isRLE ? 1 : (long)(blockHeader >> 3); - ERROR_IF(isWrongBlock, info_frame_error, "Error: unsupported block type"); - lastBlock = blockHeader & 1; - ERROR_IF(fseek(srcFile, blockSize, SEEK_CUR) != 0, - info_frame_error, "Error: could not skip to end of block"); - } - } while (lastBlock != 1); - } - - /* check if checksum is used */ - { BYTE const frameHeaderDescriptor = headerBuffer[4]; - int const contentChecksumFlag = (frameHeaderDescriptor & (1 << 2)) >> 2; - if (contentChecksumFlag) { - info->usesCheck = 1; - ERROR_IF(fseek(srcFile, 4, SEEK_CUR) != 0, - info_frame_error, "Error: could not skip past checksum"); - } } - info->numActualFrames++; - } - /* Skippable frame */ - else if ((magicNumber & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { - U32 const frameSize = MEM_readLE32(headerBuffer + 4); - long const seek = (long)(8 + frameSize - numBytesRead); - ERROR_IF(LONG_SEEK(srcFile, seek, SEEK_CUR) != 0, - info_frame_error, "Error: could not find end of skippable frame"); - info->numSkippableFrames++; - } - /* unknown content */ - else { - return info_not_zstd; - } - } /* magic number analysis */ - } /* end analyzing frames */ - return info_success; -} - - -static InfoError -getFileInfo_fileConfirmed(fileInfo_t* info, const char* inFileName) -{ - InfoError status; - FILE* const srcFile = FIO_openSrcFile(NULL, inFileName); - ERROR_IF(srcFile == NULL, info_file_error, "Error: could not open source file %s", inFileName); - - info->compressedSize = UTIL_getFileSize(inFileName); - status = FIO_analyzeFrames(info, srcFile); - - fclose(srcFile); - info->nbFiles = 1; - return status; -} - - -/** getFileInfo() : - * Reads information from file, stores in *info - * @return : InfoError status - */ -static InfoError -getFileInfo(fileInfo_t* info, const char* srcFileName) -{ - ERROR_IF(!UTIL_isRegularFile(srcFileName), - info_file_error, "Error : %s is not a file", srcFileName); - return getFileInfo_fileConfirmed(info, srcFileName); -} - - -static void -displayInfo(const char* inFileName, const fileInfo_t* info, int displayLevel) -{ + if ( (result != 0) /* operation failure */ + && strcmp(dstFileName, stdoutmark) /* special case : don't remove() stdout */ + ) { + FIO_removeFile(dstFileName); /* remove decompression artefact; note: don't do anything special if remove() fails */ + } + } + + return result; +} + + +/** FIO_decompressSrcFile() : + Open `srcFileName`, transfer control to decompressDstFile() + @return : 0 : OK + 1 : error +*/ +static int FIO_decompressSrcFile(FIO_ctx_t* const fCtx, FIO_prefs_t* const prefs, dRess_t ress, const char* dstFileName, const char* srcFileName) +{ + FILE* srcFile; + int result; + + if (UTIL_isDirectory(srcFileName)) { + DISPLAYLEVEL(1, "zstd: %s is a directory -- ignored \n", srcFileName); + return 1; + } + + srcFile = FIO_openSrcFile(prefs, srcFileName); + if (srcFile==NULL) return 1; + ress.srcBufferLoaded = 0; + + result = FIO_decompressDstFile(fCtx, prefs, ress, srcFile, dstFileName, srcFileName); + + /* Close file */ + if (fclose(srcFile)) { + DISPLAYLEVEL(1, "zstd: %s: %s \n", srcFileName, strerror(errno)); /* error should not happen */ + return 1; + } + if ( prefs->removeSrcFile /* --rm */ + && (result==0) /* decompression successful */ + && strcmp(srcFileName, stdinmark) ) /* not stdin */ { + /* We must clear the handler, since after this point calling it would + * delete both the source and destination files. + */ + clearHandler(); + if (FIO_removeFile(srcFileName)) { + /* failed to remove src file */ + DISPLAYLEVEL(1, "zstd: %s: %s \n", srcFileName, strerror(errno)); + return 1; + } } + return result; +} + + + +int FIO_decompressFilename(FIO_ctx_t* const fCtx, FIO_prefs_t* const prefs, + const char* dstFileName, const char* srcFileName, + const char* dictFileName) +{ + dRess_t const ress = FIO_createDResources(prefs, dictFileName); + + int const decodingError = FIO_decompressSrcFile(fCtx, prefs, ress, dstFileName, srcFileName); + + FIO_freeDResources(ress); + return decodingError; +} + +static const char *suffixList[] = { + ZSTD_EXTENSION, + TZSTD_EXTENSION, +#ifndef ZSTD_NODECOMPRESS + ZSTD_ALT_EXTENSION, +#endif +#ifdef ZSTD_GZDECOMPRESS + GZ_EXTENSION, + TGZ_EXTENSION, +#endif +#ifdef ZSTD_LZMADECOMPRESS + LZMA_EXTENSION, + XZ_EXTENSION, + TXZ_EXTENSION, +#endif +#ifdef ZSTD_LZ4DECOMPRESS + LZ4_EXTENSION, + TLZ4_EXTENSION, +#endif + NULL +}; + +static const char *suffixListStr = + ZSTD_EXTENSION "/" TZSTD_EXTENSION +#ifdef ZSTD_GZDECOMPRESS + "/" GZ_EXTENSION "/" TGZ_EXTENSION +#endif +#ifdef ZSTD_LZMADECOMPRESS + "/" LZMA_EXTENSION "/" XZ_EXTENSION "/" TXZ_EXTENSION +#endif +#ifdef ZSTD_LZ4DECOMPRESS + "/" LZ4_EXTENSION "/" TLZ4_EXTENSION +#endif +; + +/* FIO_determineDstName() : + * create a destination filename from a srcFileName. + * @return a pointer to it. + * @return == NULL if there is an error */ +static const char* +FIO_determineDstName(const char* srcFileName, const char* outDirName) +{ + static size_t dfnbCapacity = 0; + static char* dstFileNameBuffer = NULL; /* using static allocation : this function cannot be multi-threaded */ + size_t dstFileNameEndPos; + char* outDirFilename = NULL; + const char* dstSuffix = ""; + size_t dstSuffixLen = 0; + + size_t sfnSize = strlen(srcFileName); + + size_t srcSuffixLen; + const char* const srcSuffix = strrchr(srcFileName, '.'); + if (srcSuffix == NULL) { + DISPLAYLEVEL(1, + "zstd: %s: unknown suffix (%s expected). " + "Can't derive the output file name. " + "Specify it with -o dstFileName. Ignoring.\n", + srcFileName, suffixListStr); + return NULL; + } + srcSuffixLen = strlen(srcSuffix); + + { + const char** matchedSuffixPtr; + for (matchedSuffixPtr = suffixList; *matchedSuffixPtr != NULL; matchedSuffixPtr++) { + if (!strcmp(*matchedSuffixPtr, srcSuffix)) { + break; + } + } + + /* check suffix is authorized */ + if (sfnSize <= srcSuffixLen || *matchedSuffixPtr == NULL) { + DISPLAYLEVEL(1, + "zstd: %s: unknown suffix (%s expected). " + "Can't derive the output file name. " + "Specify it with -o dstFileName. Ignoring.\n", + srcFileName, suffixListStr); + return NULL; + } + + if ((*matchedSuffixPtr)[1] == 't') { + dstSuffix = ".tar"; + dstSuffixLen = strlen(dstSuffix); + } + } + + if (outDirName) { + outDirFilename = FIO_createFilename_fromOutDir(srcFileName, outDirName, 0); + sfnSize = strlen(outDirFilename); + assert(outDirFilename != NULL); + } + + if (dfnbCapacity+srcSuffixLen <= sfnSize+1+dstSuffixLen) { + /* allocate enough space to write dstFilename into it */ + free(dstFileNameBuffer); + dfnbCapacity = sfnSize + 20; + dstFileNameBuffer = (char*)malloc(dfnbCapacity); + if (dstFileNameBuffer==NULL) + EXM_THROW(74, "%s : not enough memory for dstFileName", + strerror(errno)); + } + + /* return dst name == src name truncated from suffix */ + assert(dstFileNameBuffer != NULL); + dstFileNameEndPos = sfnSize - srcSuffixLen; + if (outDirFilename) { + memcpy(dstFileNameBuffer, outDirFilename, dstFileNameEndPos); + free(outDirFilename); + } else { + memcpy(dstFileNameBuffer, srcFileName, dstFileNameEndPos); + } + + /* The short tar extensions tzst, tgz, txz and tlz4 files should have "tar" + * extension on decompression. Also writes terminating null. */ + strcpy(dstFileNameBuffer + dstFileNameEndPos, dstSuffix); + return dstFileNameBuffer; + + /* note : dstFileNameBuffer memory is not going to be free */ +} + +int +FIO_decompressMultipleFilenames(FIO_ctx_t* const fCtx, + FIO_prefs_t* const prefs, + const char** srcNamesTable, + const char* outMirroredRootDirName, + const char* outDirName, const char* outFileName, + const char* dictFileName) +{ + int status; + int error = 0; + dRess_t ress = FIO_createDResources(prefs, dictFileName); + + if (outFileName) { + if (FIO_removeMultiFilesWarning(fCtx, prefs, outFileName, 1 /* displayLevelCutoff */)) { + FIO_freeDResources(ress); + return 1; + } + if (!prefs->testMode) { + ress.dstFile = FIO_openDstFile(fCtx, prefs, NULL, outFileName, DEFAULT_FILE_PERMISSIONS); + if (ress.dstFile == 0) EXM_THROW(19, "cannot open %s", outFileName); + } + for (; fCtx->currFileIdx < fCtx->nbFilesTotal; fCtx->currFileIdx++) { + status = FIO_decompressSrcFile(fCtx, prefs, ress, outFileName, srcNamesTable[fCtx->currFileIdx]); + if (!status) fCtx->nbFilesProcessed++; + error |= status; + } + if ((!prefs->testMode) && (fclose(ress.dstFile))) + EXM_THROW(72, "Write error : %s : cannot properly close output file", + strerror(errno)); + } else { + if (outMirroredRootDirName) + UTIL_mirrorSourceFilesDirectories(srcNamesTable, (unsigned)fCtx->nbFilesTotal, outMirroredRootDirName); + + for (; fCtx->currFileIdx < fCtx->nbFilesTotal; fCtx->currFileIdx++) { /* create dstFileName */ + const char* const srcFileName = srcNamesTable[fCtx->currFileIdx]; + const char* dstFileName = NULL; + if (outMirroredRootDirName) { + char* validMirroredDirName = UTIL_createMirroredDestDirName(srcFileName, outMirroredRootDirName); + if (validMirroredDirName) { + dstFileName = FIO_determineDstName(srcFileName, validMirroredDirName); + free(validMirroredDirName); + } else { + DISPLAYLEVEL(2, "zstd: --output-dir-mirror cannot decompress '%s' into '%s'\n", srcFileName, outMirroredRootDirName); + } + } else { + dstFileName = FIO_determineDstName(srcFileName, outDirName); + } + if (dstFileName == NULL) { error=1; continue; } + status = FIO_decompressSrcFile(fCtx, prefs, ress, dstFileName, srcFileName); + if (!status) fCtx->nbFilesProcessed++; + error |= status; + } + if (outDirName) + FIO_checkFilenameCollisions(srcNamesTable , (unsigned)fCtx->nbFilesTotal); + } + + if (fCtx->nbFilesProcessed >= 1 && fCtx->nbFilesTotal > 1 && fCtx->totalBytesOutput != 0) + DISPLAYLEVEL(2, "%d files decompressed : %6zu bytes total \n", fCtx->nbFilesProcessed, fCtx->totalBytesOutput); + + FIO_freeDResources(ress); + return error; +} + +/* ************************************************************************** + * .zst file info (--list command) + ***************************************************************************/ + +typedef struct { + U64 decompressedSize; + U64 compressedSize; + U64 windowSize; + int numActualFrames; + int numSkippableFrames; + int decompUnavailable; + int usesCheck; + U32 nbFiles; +} fileInfo_t; + +typedef enum { + info_success=0, + info_frame_error=1, + info_not_zstd=2, + info_file_error=3, + info_truncated_input=4, +} InfoError; + +#define ERROR_IF(c,n,...) { \ + if (c) { \ + DISPLAYLEVEL(1, __VA_ARGS__); \ + DISPLAYLEVEL(1, " \n"); \ + return n; \ + } \ +} + +static InfoError +FIO_analyzeFrames(fileInfo_t* info, FILE* const srcFile) +{ + /* begin analyzing frame */ + for ( ; ; ) { + BYTE headerBuffer[ZSTD_FRAMEHEADERSIZE_MAX]; + size_t const numBytesRead = fread(headerBuffer, 1, sizeof(headerBuffer), srcFile); + if (numBytesRead < ZSTD_FRAMEHEADERSIZE_MIN(ZSTD_f_zstd1)) { + if ( feof(srcFile) + && (numBytesRead == 0) + && (info->compressedSize > 0) + && (info->compressedSize != UTIL_FILESIZE_UNKNOWN) ) { + unsigned long long file_position = (unsigned long long) LONG_TELL(srcFile); + unsigned long long file_size = (unsigned long long) info->compressedSize; + ERROR_IF(file_position != file_size, info_truncated_input, + "Error: seeked to position %llu, which is beyond file size of %llu\n", + file_position, + file_size); + break; /* correct end of file => success */ + } + ERROR_IF(feof(srcFile), info_not_zstd, "Error: reached end of file with incomplete frame"); + ERROR_IF(1, info_frame_error, "Error: did not reach end of file but ran out of frames"); + } + { U32 const magicNumber = MEM_readLE32(headerBuffer); + /* Zstandard frame */ + if (magicNumber == ZSTD_MAGICNUMBER) { + ZSTD_frameHeader header; + U64 const frameContentSize = ZSTD_getFrameContentSize(headerBuffer, numBytesRead); + if ( frameContentSize == ZSTD_CONTENTSIZE_ERROR + || frameContentSize == ZSTD_CONTENTSIZE_UNKNOWN ) { + info->decompUnavailable = 1; + } else { + info->decompressedSize += frameContentSize; + } + ERROR_IF(ZSTD_getFrameHeader(&header, headerBuffer, numBytesRead) != 0, + info_frame_error, "Error: could not decode frame header"); + info->windowSize = header.windowSize; + /* move to the end of the frame header */ + { size_t const headerSize = ZSTD_frameHeaderSize(headerBuffer, numBytesRead); + ERROR_IF(ZSTD_isError(headerSize), info_frame_error, "Error: could not determine frame header size"); + ERROR_IF(fseek(srcFile, ((long)headerSize)-((long)numBytesRead), SEEK_CUR) != 0, + info_frame_error, "Error: could not move to end of frame header"); + } + + /* skip all blocks in the frame */ + { int lastBlock = 0; + do { + BYTE blockHeaderBuffer[3]; + ERROR_IF(fread(blockHeaderBuffer, 1, 3, srcFile) != 3, + info_frame_error, "Error while reading block header"); + { U32 const blockHeader = MEM_readLE24(blockHeaderBuffer); + U32 const blockTypeID = (blockHeader >> 1) & 3; + U32 const isRLE = (blockTypeID == 1); + U32 const isWrongBlock = (blockTypeID == 3); + long const blockSize = isRLE ? 1 : (long)(blockHeader >> 3); + ERROR_IF(isWrongBlock, info_frame_error, "Error: unsupported block type"); + lastBlock = blockHeader & 1; + ERROR_IF(fseek(srcFile, blockSize, SEEK_CUR) != 0, + info_frame_error, "Error: could not skip to end of block"); + } + } while (lastBlock != 1); + } + + /* check if checksum is used */ + { BYTE const frameHeaderDescriptor = headerBuffer[4]; + int const contentChecksumFlag = (frameHeaderDescriptor & (1 << 2)) >> 2; + if (contentChecksumFlag) { + info->usesCheck = 1; + ERROR_IF(fseek(srcFile, 4, SEEK_CUR) != 0, + info_frame_error, "Error: could not skip past checksum"); + } } + info->numActualFrames++; + } + /* Skippable frame */ + else if ((magicNumber & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { + U32 const frameSize = MEM_readLE32(headerBuffer + 4); + long const seek = (long)(8 + frameSize - numBytesRead); + ERROR_IF(LONG_SEEK(srcFile, seek, SEEK_CUR) != 0, + info_frame_error, "Error: could not find end of skippable frame"); + info->numSkippableFrames++; + } + /* unknown content */ + else { + return info_not_zstd; + } + } /* magic number analysis */ + } /* end analyzing frames */ + return info_success; +} + + +static InfoError +getFileInfo_fileConfirmed(fileInfo_t* info, const char* inFileName) +{ + InfoError status; + FILE* const srcFile = FIO_openSrcFile(NULL, inFileName); + ERROR_IF(srcFile == NULL, info_file_error, "Error: could not open source file %s", inFileName); + + info->compressedSize = UTIL_getFileSize(inFileName); + status = FIO_analyzeFrames(info, srcFile); + + fclose(srcFile); + info->nbFiles = 1; + return status; +} + + +/** getFileInfo() : + * Reads information from file, stores in *info + * @return : InfoError status + */ +static InfoError +getFileInfo(fileInfo_t* info, const char* srcFileName) +{ + ERROR_IF(!UTIL_isRegularFile(srcFileName), + info_file_error, "Error : %s is not a file", srcFileName); + return getFileInfo_fileConfirmed(info, srcFileName); +} + + +static void +displayInfo(const char* inFileName, const fileInfo_t* info, int displayLevel) +{ UTIL_HumanReadableSize_t const window_hrs = UTIL_makeHumanReadableSize(info->windowSize); UTIL_HumanReadableSize_t const compressed_hrs = UTIL_makeHumanReadableSize(info->compressedSize); UTIL_HumanReadableSize_t const decompressed_hrs = UTIL_makeHumanReadableSize(info->decompressedSize); - double const ratio = (info->compressedSize == 0) ? 0 : ((double)info->decompressedSize)/(double)info->compressedSize; - const char* const checkString = (info->usesCheck ? "XXH64" : "None"); - if (displayLevel <= 2) { - if (!info->decompUnavailable) { + double const ratio = (info->compressedSize == 0) ? 0 : ((double)info->decompressedSize)/(double)info->compressedSize; + const char* const checkString = (info->usesCheck ? "XXH64" : "None"); + if (displayLevel <= 2) { + if (!info->decompUnavailable) { DISPLAYOUT("%6d %5d %6.*f%4s %8.*f%4s %5.3f %5s %s\n", - info->numSkippableFrames + info->numActualFrames, - info->numSkippableFrames, + info->numSkippableFrames + info->numActualFrames, + info->numSkippableFrames, compressed_hrs.precision, compressed_hrs.value, compressed_hrs.suffix, decompressed_hrs.precision, decompressed_hrs.value, decompressed_hrs.suffix, - ratio, checkString, inFileName); - } else { + ratio, checkString, inFileName); + } else { DISPLAYOUT("%6d %5d %6.*f%4s %5s %s\n", - info->numSkippableFrames + info->numActualFrames, - info->numSkippableFrames, + info->numSkippableFrames + info->numActualFrames, + info->numSkippableFrames, compressed_hrs.precision, compressed_hrs.value, compressed_hrs.suffix, - checkString, inFileName); - } - } else { - DISPLAYOUT("%s \n", inFileName); - DISPLAYOUT("# Zstandard Frames: %d\n", info->numActualFrames); - if (info->numSkippableFrames) - DISPLAYOUT("# Skippable Frames: %d\n", info->numSkippableFrames); + checkString, inFileName); + } + } else { + DISPLAYOUT("%s \n", inFileName); + DISPLAYOUT("# Zstandard Frames: %d\n", info->numActualFrames); + if (info->numSkippableFrames) + DISPLAYOUT("# Skippable Frames: %d\n", info->numSkippableFrames); DISPLAYOUT("Window Size: %.*f%s (%llu B)\n", window_hrs.precision, window_hrs.value, window_hrs.suffix, - (unsigned long long)info->windowSize); + (unsigned long long)info->windowSize); DISPLAYOUT("Compressed Size: %.*f%s (%llu B)\n", compressed_hrs.precision, compressed_hrs.value, compressed_hrs.suffix, - (unsigned long long)info->compressedSize); - if (!info->decompUnavailable) { + (unsigned long long)info->compressedSize); + if (!info->decompUnavailable) { DISPLAYOUT("Decompressed Size: %.*f%s (%llu B)\n", decompressed_hrs.precision, decompressed_hrs.value, decompressed_hrs.suffix, - (unsigned long long)info->decompressedSize); - DISPLAYOUT("Ratio: %.4f\n", ratio); - } - DISPLAYOUT("Check: %s\n", checkString); - DISPLAYOUT("\n"); - } -} - -static fileInfo_t FIO_addFInfo(fileInfo_t fi1, fileInfo_t fi2) -{ - fileInfo_t total; - memset(&total, 0, sizeof(total)); - total.numActualFrames = fi1.numActualFrames + fi2.numActualFrames; - total.numSkippableFrames = fi1.numSkippableFrames + fi2.numSkippableFrames; - total.compressedSize = fi1.compressedSize + fi2.compressedSize; - total.decompressedSize = fi1.decompressedSize + fi2.decompressedSize; - total.decompUnavailable = fi1.decompUnavailable | fi2.decompUnavailable; - total.usesCheck = fi1.usesCheck & fi2.usesCheck; - total.nbFiles = fi1.nbFiles + fi2.nbFiles; - return total; -} - -static int -FIO_listFile(fileInfo_t* total, const char* inFileName, int displayLevel) -{ - fileInfo_t info; - memset(&info, 0, sizeof(info)); - { InfoError const error = getFileInfo(&info, inFileName); - switch (error) { - case info_frame_error: - /* display error, but provide output */ - DISPLAYLEVEL(1, "Error while parsing \"%s\" \n", inFileName); - break; - case info_not_zstd: - DISPLAYOUT("File \"%s\" not compressed by zstd \n", inFileName); - if (displayLevel > 2) DISPLAYOUT("\n"); - return 1; - case info_file_error: - /* error occurred while opening the file */ - if (displayLevel > 2) DISPLAYOUT("\n"); - return 1; - case info_truncated_input: - DISPLAYOUT("File \"%s\" is truncated \n", inFileName); - if (displayLevel > 2) DISPLAYOUT("\n"); - return 1; - case info_success: - default: - break; - } - - displayInfo(inFileName, &info, displayLevel); - *total = FIO_addFInfo(*total, info); - assert(error == info_success || error == info_frame_error); - return (int)error; - } -} - -int FIO_listMultipleFiles(unsigned numFiles, const char** filenameTable, int displayLevel) -{ - /* ensure no specified input is stdin (needs fseek() capability) */ - { unsigned u; - for (u=0; u<numFiles;u++) { - ERROR_IF(!strcmp (filenameTable[u], stdinmark), - 1, "zstd: --list does not support reading from standard input"); - } } - - if (numFiles == 0) { - if (!IS_CONSOLE(stdin)) { - DISPLAYLEVEL(1, "zstd: --list does not support reading from standard input \n"); - } - DISPLAYLEVEL(1, "No files given \n"); - return 1; - } - - if (displayLevel <= 2) { - DISPLAYOUT("Frames Skips Compressed Uncompressed Ratio Check Filename\n"); - } - { int error = 0; - fileInfo_t total; - memset(&total, 0, sizeof(total)); - total.usesCheck = 1; - /* --list each file, and check for any error */ - { unsigned u; - for (u=0; u<numFiles;u++) { - error |= FIO_listFile(&total, filenameTable[u], displayLevel); - } } - if (numFiles > 1 && displayLevel <= 2) { /* display total */ + (unsigned long long)info->decompressedSize); + DISPLAYOUT("Ratio: %.4f\n", ratio); + } + DISPLAYOUT("Check: %s\n", checkString); + DISPLAYOUT("\n"); + } +} + +static fileInfo_t FIO_addFInfo(fileInfo_t fi1, fileInfo_t fi2) +{ + fileInfo_t total; + memset(&total, 0, sizeof(total)); + total.numActualFrames = fi1.numActualFrames + fi2.numActualFrames; + total.numSkippableFrames = fi1.numSkippableFrames + fi2.numSkippableFrames; + total.compressedSize = fi1.compressedSize + fi2.compressedSize; + total.decompressedSize = fi1.decompressedSize + fi2.decompressedSize; + total.decompUnavailable = fi1.decompUnavailable | fi2.decompUnavailable; + total.usesCheck = fi1.usesCheck & fi2.usesCheck; + total.nbFiles = fi1.nbFiles + fi2.nbFiles; + return total; +} + +static int +FIO_listFile(fileInfo_t* total, const char* inFileName, int displayLevel) +{ + fileInfo_t info; + memset(&info, 0, sizeof(info)); + { InfoError const error = getFileInfo(&info, inFileName); + switch (error) { + case info_frame_error: + /* display error, but provide output */ + DISPLAYLEVEL(1, "Error while parsing \"%s\" \n", inFileName); + break; + case info_not_zstd: + DISPLAYOUT("File \"%s\" not compressed by zstd \n", inFileName); + if (displayLevel > 2) DISPLAYOUT("\n"); + return 1; + case info_file_error: + /* error occurred while opening the file */ + if (displayLevel > 2) DISPLAYOUT("\n"); + return 1; + case info_truncated_input: + DISPLAYOUT("File \"%s\" is truncated \n", inFileName); + if (displayLevel > 2) DISPLAYOUT("\n"); + return 1; + case info_success: + default: + break; + } + + displayInfo(inFileName, &info, displayLevel); + *total = FIO_addFInfo(*total, info); + assert(error == info_success || error == info_frame_error); + return (int)error; + } +} + +int FIO_listMultipleFiles(unsigned numFiles, const char** filenameTable, int displayLevel) +{ + /* ensure no specified input is stdin (needs fseek() capability) */ + { unsigned u; + for (u=0; u<numFiles;u++) { + ERROR_IF(!strcmp (filenameTable[u], stdinmark), + 1, "zstd: --list does not support reading from standard input"); + } } + + if (numFiles == 0) { + if (!IS_CONSOLE(stdin)) { + DISPLAYLEVEL(1, "zstd: --list does not support reading from standard input \n"); + } + DISPLAYLEVEL(1, "No files given \n"); + return 1; + } + + if (displayLevel <= 2) { + DISPLAYOUT("Frames Skips Compressed Uncompressed Ratio Check Filename\n"); + } + { int error = 0; + fileInfo_t total; + memset(&total, 0, sizeof(total)); + total.usesCheck = 1; + /* --list each file, and check for any error */ + { unsigned u; + for (u=0; u<numFiles;u++) { + error |= FIO_listFile(&total, filenameTable[u], displayLevel); + } } + if (numFiles > 1 && displayLevel <= 2) { /* display total */ UTIL_HumanReadableSize_t const compressed_hrs = UTIL_makeHumanReadableSize(total.compressedSize); UTIL_HumanReadableSize_t const decompressed_hrs = UTIL_makeHumanReadableSize(total.decompressedSize); - double const ratio = (total.compressedSize == 0) ? 0 : ((double)total.decompressedSize)/(double)total.compressedSize; - const char* const checkString = (total.usesCheck ? "XXH64" : ""); - DISPLAYOUT("----------------------------------------------------------------- \n"); - if (total.decompUnavailable) { + double const ratio = (total.compressedSize == 0) ? 0 : ((double)total.decompressedSize)/(double)total.compressedSize; + const char* const checkString = (total.usesCheck ? "XXH64" : ""); + DISPLAYOUT("----------------------------------------------------------------- \n"); + if (total.decompUnavailable) { DISPLAYOUT("%6d %5d %6.*f%4s %5s %u files\n", - total.numSkippableFrames + total.numActualFrames, - total.numSkippableFrames, + total.numSkippableFrames + total.numActualFrames, + total.numSkippableFrames, compressed_hrs.precision, compressed_hrs.value, compressed_hrs.suffix, - checkString, (unsigned)total.nbFiles); - } else { + checkString, (unsigned)total.nbFiles); + } else { DISPLAYOUT("%6d %5d %6.*f%4s %8.*f%4s %5.3f %5s %u files\n", - total.numSkippableFrames + total.numActualFrames, - total.numSkippableFrames, + total.numSkippableFrames + total.numActualFrames, + total.numSkippableFrames, compressed_hrs.precision, compressed_hrs.value, compressed_hrs.suffix, decompressed_hrs.precision, decompressed_hrs.value, decompressed_hrs.suffix, - ratio, checkString, (unsigned)total.nbFiles); - } } - return error; - } -} - - -#endif /* #ifndef ZSTD_NODECOMPRESS */ + ratio, checkString, (unsigned)total.nbFiles); + } } + return error; + } +} + + +#endif /* #ifndef ZSTD_NODECOMPRESS */ diff --git a/contrib/libs/zstd/programs/fileio.h b/contrib/libs/zstd/programs/fileio.h index 61094db83c..36ec19a395 100644 --- a/contrib/libs/zstd/programs/fileio.h +++ b/contrib/libs/zstd/programs/fileio.h @@ -1,180 +1,180 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - -#ifndef FILEIO_H_23981798732 -#define FILEIO_H_23981798732 - -#define ZSTD_STATIC_LINKING_ONLY /* ZSTD_compressionParameters */ -#include "../lib/zstd.h" /* ZSTD_* */ - -#if defined (__cplusplus) -extern "C" { -#endif - - -/* ************************************* -* Special i/o constants -**************************************/ -#define stdinmark "/*stdin*\\" -#define stdoutmark "/*stdout*\\" -#ifdef _WIN32 -# define nulmark "NUL" -#else -# define nulmark "/dev/null" -#endif - -/** - * We test whether the extension we found starts with 't', and if so, we append - * ".tar" to the end of the output name. - */ -#define LZMA_EXTENSION ".lzma" -#define XZ_EXTENSION ".xz" -#define TXZ_EXTENSION ".txz" - -#define GZ_EXTENSION ".gz" -#define TGZ_EXTENSION ".tgz" - -#define ZSTD_EXTENSION ".zst" -#define TZSTD_EXTENSION ".tzst" -#define ZSTD_ALT_EXTENSION ".zstd" /* allow decompression of .zstd files */ - -#define LZ4_EXTENSION ".lz4" -#define TLZ4_EXTENSION ".tlz4" - - -/*-************************************* -* Types -***************************************/ -typedef enum { FIO_zstdCompression, FIO_gzipCompression, FIO_xzCompression, FIO_lzmaCompression, FIO_lz4Compression } FIO_compressionType_t; - -typedef struct FIO_prefs_s FIO_prefs_t; - -FIO_prefs_t* FIO_createPreferences(void); -void FIO_freePreferences(FIO_prefs_t* const prefs); - -/* Mutable struct containing relevant context and state regarding (de)compression with respect to file I/O */ -typedef struct FIO_ctx_s FIO_ctx_t; - -FIO_ctx_t* FIO_createContext(void); -void FIO_freeContext(FIO_ctx_t* const fCtx); - -typedef struct FIO_display_prefs_s FIO_display_prefs_t; - -typedef enum { FIO_ps_auto, FIO_ps_never, FIO_ps_always } FIO_progressSetting_e; - -/*-************************************* -* Parameters -***************************************/ -/* FIO_prefs_t functions */ -void FIO_setCompressionType(FIO_prefs_t* const prefs, FIO_compressionType_t compressionType); -void FIO_overwriteMode(FIO_prefs_t* const prefs); -void FIO_setAdaptiveMode(FIO_prefs_t* const prefs, unsigned adapt); -void FIO_setAdaptMin(FIO_prefs_t* const prefs, int minCLevel); -void FIO_setAdaptMax(FIO_prefs_t* const prefs, int maxCLevel); -void FIO_setUseRowMatchFinder(FIO_prefs_t* const prefs, int useRowMatchFinder); -void FIO_setBlockSize(FIO_prefs_t* const prefs, int blockSize); -void FIO_setChecksumFlag(FIO_prefs_t* const prefs, int checksumFlag); -void FIO_setDictIDFlag(FIO_prefs_t* const prefs, int dictIDFlag); -void FIO_setLdmBucketSizeLog(FIO_prefs_t* const prefs, int ldmBucketSizeLog); -void FIO_setLdmFlag(FIO_prefs_t* const prefs, unsigned ldmFlag); -void FIO_setLdmHashRateLog(FIO_prefs_t* const prefs, int ldmHashRateLog); -void FIO_setLdmHashLog(FIO_prefs_t* const prefs, int ldmHashLog); -void FIO_setLdmMinMatch(FIO_prefs_t* const prefs, int ldmMinMatch); -void FIO_setMemLimit(FIO_prefs_t* const prefs, unsigned memLimit); -void FIO_setNbWorkers(FIO_prefs_t* const prefs, int nbWorkers); -void FIO_setOverlapLog(FIO_prefs_t* const prefs, int overlapLog); -void FIO_setRemoveSrcFile(FIO_prefs_t* const prefs, unsigned flag); -void FIO_setSparseWrite(FIO_prefs_t* const prefs, unsigned sparse); /**< 0: no sparse; 1: disable on stdout; 2: always enabled */ -void FIO_setRsyncable(FIO_prefs_t* const prefs, int rsyncable); -void FIO_setStreamSrcSize(FIO_prefs_t* const prefs, size_t streamSrcSize); -void FIO_setTargetCBlockSize(FIO_prefs_t* const prefs, size_t targetCBlockSize); -void FIO_setSrcSizeHint(FIO_prefs_t* const prefs, size_t srcSizeHint); -void FIO_setTestMode(FIO_prefs_t* const prefs, int testMode); -void FIO_setLiteralCompressionMode( - FIO_prefs_t* const prefs, +/* + * Copyright (c) Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + + +#ifndef FILEIO_H_23981798732 +#define FILEIO_H_23981798732 + +#define ZSTD_STATIC_LINKING_ONLY /* ZSTD_compressionParameters */ +#include "../lib/zstd.h" /* ZSTD_* */ + +#if defined (__cplusplus) +extern "C" { +#endif + + +/* ************************************* +* Special i/o constants +**************************************/ +#define stdinmark "/*stdin*\\" +#define stdoutmark "/*stdout*\\" +#ifdef _WIN32 +# define nulmark "NUL" +#else +# define nulmark "/dev/null" +#endif + +/** + * We test whether the extension we found starts with 't', and if so, we append + * ".tar" to the end of the output name. + */ +#define LZMA_EXTENSION ".lzma" +#define XZ_EXTENSION ".xz" +#define TXZ_EXTENSION ".txz" + +#define GZ_EXTENSION ".gz" +#define TGZ_EXTENSION ".tgz" + +#define ZSTD_EXTENSION ".zst" +#define TZSTD_EXTENSION ".tzst" +#define ZSTD_ALT_EXTENSION ".zstd" /* allow decompression of .zstd files */ + +#define LZ4_EXTENSION ".lz4" +#define TLZ4_EXTENSION ".tlz4" + + +/*-************************************* +* Types +***************************************/ +typedef enum { FIO_zstdCompression, FIO_gzipCompression, FIO_xzCompression, FIO_lzmaCompression, FIO_lz4Compression } FIO_compressionType_t; + +typedef struct FIO_prefs_s FIO_prefs_t; + +FIO_prefs_t* FIO_createPreferences(void); +void FIO_freePreferences(FIO_prefs_t* const prefs); + +/* Mutable struct containing relevant context and state regarding (de)compression with respect to file I/O */ +typedef struct FIO_ctx_s FIO_ctx_t; + +FIO_ctx_t* FIO_createContext(void); +void FIO_freeContext(FIO_ctx_t* const fCtx); + +typedef struct FIO_display_prefs_s FIO_display_prefs_t; + +typedef enum { FIO_ps_auto, FIO_ps_never, FIO_ps_always } FIO_progressSetting_e; + +/*-************************************* +* Parameters +***************************************/ +/* FIO_prefs_t functions */ +void FIO_setCompressionType(FIO_prefs_t* const prefs, FIO_compressionType_t compressionType); +void FIO_overwriteMode(FIO_prefs_t* const prefs); +void FIO_setAdaptiveMode(FIO_prefs_t* const prefs, unsigned adapt); +void FIO_setAdaptMin(FIO_prefs_t* const prefs, int minCLevel); +void FIO_setAdaptMax(FIO_prefs_t* const prefs, int maxCLevel); +void FIO_setUseRowMatchFinder(FIO_prefs_t* const prefs, int useRowMatchFinder); +void FIO_setBlockSize(FIO_prefs_t* const prefs, int blockSize); +void FIO_setChecksumFlag(FIO_prefs_t* const prefs, int checksumFlag); +void FIO_setDictIDFlag(FIO_prefs_t* const prefs, int dictIDFlag); +void FIO_setLdmBucketSizeLog(FIO_prefs_t* const prefs, int ldmBucketSizeLog); +void FIO_setLdmFlag(FIO_prefs_t* const prefs, unsigned ldmFlag); +void FIO_setLdmHashRateLog(FIO_prefs_t* const prefs, int ldmHashRateLog); +void FIO_setLdmHashLog(FIO_prefs_t* const prefs, int ldmHashLog); +void FIO_setLdmMinMatch(FIO_prefs_t* const prefs, int ldmMinMatch); +void FIO_setMemLimit(FIO_prefs_t* const prefs, unsigned memLimit); +void FIO_setNbWorkers(FIO_prefs_t* const prefs, int nbWorkers); +void FIO_setOverlapLog(FIO_prefs_t* const prefs, int overlapLog); +void FIO_setRemoveSrcFile(FIO_prefs_t* const prefs, unsigned flag); +void FIO_setSparseWrite(FIO_prefs_t* const prefs, unsigned sparse); /**< 0: no sparse; 1: disable on stdout; 2: always enabled */ +void FIO_setRsyncable(FIO_prefs_t* const prefs, int rsyncable); +void FIO_setStreamSrcSize(FIO_prefs_t* const prefs, size_t streamSrcSize); +void FIO_setTargetCBlockSize(FIO_prefs_t* const prefs, size_t targetCBlockSize); +void FIO_setSrcSizeHint(FIO_prefs_t* const prefs, size_t srcSizeHint); +void FIO_setTestMode(FIO_prefs_t* const prefs, int testMode); +void FIO_setLiteralCompressionMode( + FIO_prefs_t* const prefs, ZSTD_paramSwitch_e mode); - -void FIO_setProgressSetting(FIO_progressSetting_e progressSetting); -void FIO_setNotificationLevel(int level); -void FIO_setExcludeCompressedFile(FIO_prefs_t* const prefs, int excludeCompressedFiles); -void FIO_setAllowBlockDevices(FIO_prefs_t* const prefs, int allowBlockDevices); -void FIO_setPatchFromMode(FIO_prefs_t* const prefs, int value); -void FIO_setContentSize(FIO_prefs_t* const prefs, int value); + +void FIO_setProgressSetting(FIO_progressSetting_e progressSetting); +void FIO_setNotificationLevel(int level); +void FIO_setExcludeCompressedFile(FIO_prefs_t* const prefs, int excludeCompressedFiles); +void FIO_setAllowBlockDevices(FIO_prefs_t* const prefs, int allowBlockDevices); +void FIO_setPatchFromMode(FIO_prefs_t* const prefs, int value); +void FIO_setContentSize(FIO_prefs_t* const prefs, int value); void FIO_displayCompressionParameters(const FIO_prefs_t* prefs); - -/* FIO_ctx_t functions */ -void FIO_setNbFilesTotal(FIO_ctx_t* const fCtx, int value); -void FIO_setHasStdoutOutput(FIO_ctx_t* const fCtx, int value); -void FIO_determineHasStdinInput(FIO_ctx_t* const fCtx, const FileNamesTable* const filenames); - -/*-************************************* -* Single File functions -***************************************/ -/** FIO_compressFilename() : - * @return : 0 == ok; 1 == pb with src file. */ -int FIO_compressFilename (FIO_ctx_t* const fCtx, FIO_prefs_t* const prefs, - const char* outfilename, const char* infilename, - const char* dictFileName, int compressionLevel, - ZSTD_compressionParameters comprParams); - -/** FIO_decompressFilename() : - * @return : 0 == ok; 1 == pb with src file. */ -int FIO_decompressFilename (FIO_ctx_t* const fCtx, FIO_prefs_t* const prefs, - const char* outfilename, const char* infilename, const char* dictFileName); - -int FIO_listMultipleFiles(unsigned numFiles, const char** filenameTable, int displayLevel); - - -/*-************************************* -* Multiple File functions -***************************************/ -/** FIO_compressMultipleFilenames() : - * @return : nb of missing files */ -int FIO_compressMultipleFilenames(FIO_ctx_t* const fCtx, - FIO_prefs_t* const prefs, - const char** inFileNamesTable, - const char* outMirroredDirName, - const char* outDirName, - const char* outFileName, const char* suffix, - const char* dictFileName, int compressionLevel, - ZSTD_compressionParameters comprParams); - -/** FIO_decompressMultipleFilenames() : - * @return : nb of missing or skipped files */ -int FIO_decompressMultipleFilenames(FIO_ctx_t* const fCtx, - FIO_prefs_t* const prefs, - const char** srcNamesTable, - const char* outMirroredDirName, - const char* outDirName, - const char* outFileName, - const char* dictFileName); - -/* FIO_checkFilenameCollisions() : - * Checks for and warns if there are any files that would have the same output path - */ -int FIO_checkFilenameCollisions(const char** filenameTable, unsigned nbFiles); - - - -/*-************************************* -* Advanced stuff (should actually be hosted elsewhere) -***************************************/ - -/* custom crash signal handler */ -void FIO_addAbortHandler(void); - - - -#if defined (__cplusplus) -} -#endif - -#endif /* FILEIO_H_23981798732 */ + +/* FIO_ctx_t functions */ +void FIO_setNbFilesTotal(FIO_ctx_t* const fCtx, int value); +void FIO_setHasStdoutOutput(FIO_ctx_t* const fCtx, int value); +void FIO_determineHasStdinInput(FIO_ctx_t* const fCtx, const FileNamesTable* const filenames); + +/*-************************************* +* Single File functions +***************************************/ +/** FIO_compressFilename() : + * @return : 0 == ok; 1 == pb with src file. */ +int FIO_compressFilename (FIO_ctx_t* const fCtx, FIO_prefs_t* const prefs, + const char* outfilename, const char* infilename, + const char* dictFileName, int compressionLevel, + ZSTD_compressionParameters comprParams); + +/** FIO_decompressFilename() : + * @return : 0 == ok; 1 == pb with src file. */ +int FIO_decompressFilename (FIO_ctx_t* const fCtx, FIO_prefs_t* const prefs, + const char* outfilename, const char* infilename, const char* dictFileName); + +int FIO_listMultipleFiles(unsigned numFiles, const char** filenameTable, int displayLevel); + + +/*-************************************* +* Multiple File functions +***************************************/ +/** FIO_compressMultipleFilenames() : + * @return : nb of missing files */ +int FIO_compressMultipleFilenames(FIO_ctx_t* const fCtx, + FIO_prefs_t* const prefs, + const char** inFileNamesTable, + const char* outMirroredDirName, + const char* outDirName, + const char* outFileName, const char* suffix, + const char* dictFileName, int compressionLevel, + ZSTD_compressionParameters comprParams); + +/** FIO_decompressMultipleFilenames() : + * @return : nb of missing or skipped files */ +int FIO_decompressMultipleFilenames(FIO_ctx_t* const fCtx, + FIO_prefs_t* const prefs, + const char** srcNamesTable, + const char* outMirroredDirName, + const char* outDirName, + const char* outFileName, + const char* dictFileName); + +/* FIO_checkFilenameCollisions() : + * Checks for and warns if there are any files that would have the same output path + */ +int FIO_checkFilenameCollisions(const char** filenameTable, unsigned nbFiles); + + + +/*-************************************* +* Advanced stuff (should actually be hosted elsewhere) +***************************************/ + +/* custom crash signal handler */ +void FIO_addAbortHandler(void); + + + +#if defined (__cplusplus) +} +#endif + +#endif /* FILEIO_H_23981798732 */ diff --git a/contrib/libs/zstd/programs/platform.h b/contrib/libs/zstd/programs/platform.h index b858e3b484..c4622c5823 100644 --- a/contrib/libs/zstd/programs/platform.h +++ b/contrib/libs/zstd/programs/platform.h @@ -1,215 +1,215 @@ -/* - * Copyright (c) Przemyslaw Skibinski, Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef PLATFORM_H_MODULE -#define PLATFORM_H_MODULE - -#if defined (__cplusplus) -extern "C" { -#endif - - - -/* ************************************** -* Compiler Options -****************************************/ -#if defined(_MSC_VER) -# define _CRT_SECURE_NO_WARNINGS /* Disable Visual Studio warning messages for fopen, strncpy, strerror */ -# define _CRT_NONSTDC_NO_WARNINGS /* Disable C4996 complaining about posix function names */ -# if (_MSC_VER <= 1800) /* 1800 == Visual Studio 2013 */ -# define _CRT_SECURE_NO_DEPRECATE /* VS2005 - must be declared before <io.h> and <windows.h> */ -# define snprintf sprintf_s /* snprintf unsupported by Visual <= 2013 */ -# endif -# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */ -#endif - - -/* ************************************** -* Detect 64-bit OS -* http://nadeausoftware.com/articles/2012/02/c_c_tip_how_detect_processor_type_using_compiler_predefined_macros -****************************************/ -#if defined __ia64 || defined _M_IA64 /* Intel Itanium */ \ - || defined __powerpc64__ || defined __ppc64__ || defined __PPC64__ /* POWER 64-bit */ \ - || (defined __sparc && (defined __sparcv9 || defined __sparc_v9__ || defined __arch64__)) || defined __sparc64__ /* SPARC 64-bit */ \ - || defined __x86_64__s || defined _M_X64 /* x86 64-bit */ \ - || defined __arm64__ || defined __aarch64__ || defined __ARM64_ARCH_8__ /* ARM 64-bit */ \ - || (defined __mips && (__mips == 64 || __mips == 4 || __mips == 3)) /* MIPS 64-bit */ \ - || defined _LP64 || defined __LP64__ /* NetBSD, OpenBSD */ || defined __64BIT__ /* AIX */ || defined _ADDR64 /* Cray */ \ - || (defined __SIZEOF_POINTER__ && __SIZEOF_POINTER__ == 8) /* gcc */ -# if !defined(__64BIT__) -# define __64BIT__ 1 -# endif -#endif - - -/* ********************************************************* -* Turn on Large Files support (>4GB) for 32-bit Linux/Unix -***********************************************************/ -#if !defined(__64BIT__) || defined(__MINGW32__) /* No point defining Large file for 64 bit but MinGW-w64 requires it */ -# if !defined(_FILE_OFFSET_BITS) -# define _FILE_OFFSET_BITS 64 /* turn off_t into a 64-bit type for ftello, fseeko */ -# endif -# if !defined(_LARGEFILE_SOURCE) /* obsolete macro, replaced with _FILE_OFFSET_BITS */ -# define _LARGEFILE_SOURCE 1 /* Large File Support extension (LFS) - fseeko, ftello */ -# endif -# if defined(_AIX) || defined(__hpux) -# define _LARGE_FILES /* Large file support on 32-bits AIX and HP-UX */ -# endif -#endif - - -/* ************************************************************ -* Detect POSIX version -* PLATFORM_POSIX_VERSION = 0 for non-Unix e.g. Windows -* PLATFORM_POSIX_VERSION = 1 for Unix-like but non-POSIX -* PLATFORM_POSIX_VERSION > 1 is equal to found _POSIX_VERSION -* Value of PLATFORM_POSIX_VERSION can be forced on command line -***************************************************************/ -#ifndef PLATFORM_POSIX_VERSION - -# if (defined(__APPLE__) && defined(__MACH__)) || defined(__SVR4) || defined(_AIX) || defined(__hpux) /* POSIX.1-2001 (SUSv3) conformant */ \ - || defined(__DragonFly__) || defined(__FreeBSD__) || defined(__NetBSD__) || defined(__OpenBSD__) /* BSD distros */ - /* exception rule : force posix version to 200112L, - * note: it's better to use unistd.h's _POSIX_VERSION whenever possible */ -# define PLATFORM_POSIX_VERSION 200112L - -/* try to determine posix version through official unistd.h's _POSIX_VERSION (http://pubs.opengroup.org/onlinepubs/7908799/xsh/unistd.h.html). - * note : there is no simple way to know in advance if <unistd.h> is present or not on target system, - * Posix specification mandates its presence and its content, but target system must respect this spec. - * It's necessary to _not_ #include <unistd.h> whenever target OS is not unix-like - * otherwise it will block preprocessing stage. - * The following list of build macros tries to "guess" if target OS is likely unix-like, and therefore can #include <unistd.h> - */ -# elif !defined(_WIN32) \ - && ( defined(__unix__) || defined(__unix) \ - || defined(__midipix__) || defined(__VMS) || defined(__HAIKU__) ) - -# if defined(__linux__) || defined(__linux) || defined(__CYGWIN__) -# ifndef _POSIX_C_SOURCE -# define _POSIX_C_SOURCE 200809L /* feature test macro : https://www.gnu.org/software/libc/manual/html_node/Feature-Test-Macros.html */ -# endif -# endif -# include <unistd.h> /* declares _POSIX_VERSION */ -# if defined(_POSIX_VERSION) /* POSIX compliant */ -# define PLATFORM_POSIX_VERSION _POSIX_VERSION -# else -# define PLATFORM_POSIX_VERSION 1 -# endif - -# ifdef __UCLIBC__ -# ifndef __USE_MISC -# define __USE_MISC /* enable st_mtim on uclibc */ -# endif -# endif - -# else /* non-unix target platform (like Windows) */ -# define PLATFORM_POSIX_VERSION 0 -# endif - -#endif /* PLATFORM_POSIX_VERSION */ - - -#if PLATFORM_POSIX_VERSION > 1 - /* glibc < 2.26 may not expose struct timespec def without this. - * See issue #1920. */ -# ifndef _ATFILE_SOURCE -# define _ATFILE_SOURCE -# endif -#endif - - -/*-********************************************* -* Detect if isatty() and fileno() are available -************************************************/ -#if (defined(__linux__) && (PLATFORM_POSIX_VERSION > 1)) \ - || (PLATFORM_POSIX_VERSION >= 200112L) \ - || defined(__DJGPP__) -# include <unistd.h> /* isatty */ -# include <stdio.h> /* fileno */ -# define IS_CONSOLE(stdStream) isatty(fileno(stdStream)) -#elif defined(MSDOS) || defined(OS2) -# include <io.h> /* _isatty */ -# define IS_CONSOLE(stdStream) _isatty(_fileno(stdStream)) -#elif defined(WIN32) || defined(_WIN32) -# include <io.h> /* _isatty */ -# include <windows.h> /* DeviceIoControl, HANDLE, FSCTL_SET_SPARSE */ -# include <stdio.h> /* FILE */ -static __inline int IS_CONSOLE(FILE* stdStream) { - DWORD dummy; - return _isatty(_fileno(stdStream)) && GetConsoleMode((HANDLE)_get_osfhandle(_fileno(stdStream)), &dummy); -} -#else -# define IS_CONSOLE(stdStream) 0 -#endif - - -/****************************** -* OS-specific IO behaviors -******************************/ -#if defined(MSDOS) || defined(OS2) || defined(WIN32) || defined(_WIN32) -# include <fcntl.h> /* _O_BINARY */ -# include <io.h> /* _setmode, _fileno, _get_osfhandle */ -# if !defined(__DJGPP__) -# include <windows.h> /* DeviceIoControl, HANDLE, FSCTL_SET_SPARSE */ -# include <winioctl.h> /* FSCTL_SET_SPARSE */ -# define SET_BINARY_MODE(file) { int const unused=_setmode(_fileno(file), _O_BINARY); (void)unused; } -# define SET_SPARSE_FILE_MODE(file) { DWORD dw; DeviceIoControl((HANDLE) _get_osfhandle(_fileno(file)), FSCTL_SET_SPARSE, 0, 0, 0, 0, &dw, 0); } -# else -# define SET_BINARY_MODE(file) setmode(fileno(file), O_BINARY) -# define SET_SPARSE_FILE_MODE(file) -# endif -#else -# define SET_BINARY_MODE(file) -# define SET_SPARSE_FILE_MODE(file) -#endif - - -#ifndef ZSTD_SPARSE_DEFAULT -# if (defined(__APPLE__) && defined(__MACH__)) -# define ZSTD_SPARSE_DEFAULT 0 -# else -# define ZSTD_SPARSE_DEFAULT 1 -# endif -#endif - - -#ifndef ZSTD_START_SYMBOLLIST_FRAME -# ifdef __linux__ -# define ZSTD_START_SYMBOLLIST_FRAME 2 -# elif defined __APPLE__ -# define ZSTD_START_SYMBOLLIST_FRAME 4 -# else -# define ZSTD_START_SYMBOLLIST_FRAME 0 -# endif -#endif - - -#ifndef ZSTD_SETPRIORITY_SUPPORT - /* mandates presence of <sys/resource.h> and support for setpriority() : http://man7.org/linux/man-pages/man2/setpriority.2.html */ -# define ZSTD_SETPRIORITY_SUPPORT (PLATFORM_POSIX_VERSION >= 200112L) -#endif - - -#ifndef ZSTD_NANOSLEEP_SUPPORT - /* mandates support of nanosleep() within <time.h> : http://man7.org/linux/man-pages/man2/nanosleep.2.html */ -# if (defined(__linux__) && (PLATFORM_POSIX_VERSION >= 199309L)) \ - || (PLATFORM_POSIX_VERSION >= 200112L) -# define ZSTD_NANOSLEEP_SUPPORT 1 -# else -# define ZSTD_NANOSLEEP_SUPPORT 0 -# endif -#endif - - -#if defined (__cplusplus) -} -#endif - -#endif /* PLATFORM_H_MODULE */ +/* + * Copyright (c) Przemyslaw Skibinski, Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + +#ifndef PLATFORM_H_MODULE +#define PLATFORM_H_MODULE + +#if defined (__cplusplus) +extern "C" { +#endif + + + +/* ************************************** +* Compiler Options +****************************************/ +#if defined(_MSC_VER) +# define _CRT_SECURE_NO_WARNINGS /* Disable Visual Studio warning messages for fopen, strncpy, strerror */ +# define _CRT_NONSTDC_NO_WARNINGS /* Disable C4996 complaining about posix function names */ +# if (_MSC_VER <= 1800) /* 1800 == Visual Studio 2013 */ +# define _CRT_SECURE_NO_DEPRECATE /* VS2005 - must be declared before <io.h> and <windows.h> */ +# define snprintf sprintf_s /* snprintf unsupported by Visual <= 2013 */ +# endif +# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */ +#endif + + +/* ************************************** +* Detect 64-bit OS +* http://nadeausoftware.com/articles/2012/02/c_c_tip_how_detect_processor_type_using_compiler_predefined_macros +****************************************/ +#if defined __ia64 || defined _M_IA64 /* Intel Itanium */ \ + || defined __powerpc64__ || defined __ppc64__ || defined __PPC64__ /* POWER 64-bit */ \ + || (defined __sparc && (defined __sparcv9 || defined __sparc_v9__ || defined __arch64__)) || defined __sparc64__ /* SPARC 64-bit */ \ + || defined __x86_64__s || defined _M_X64 /* x86 64-bit */ \ + || defined __arm64__ || defined __aarch64__ || defined __ARM64_ARCH_8__ /* ARM 64-bit */ \ + || (defined __mips && (__mips == 64 || __mips == 4 || __mips == 3)) /* MIPS 64-bit */ \ + || defined _LP64 || defined __LP64__ /* NetBSD, OpenBSD */ || defined __64BIT__ /* AIX */ || defined _ADDR64 /* Cray */ \ + || (defined __SIZEOF_POINTER__ && __SIZEOF_POINTER__ == 8) /* gcc */ +# if !defined(__64BIT__) +# define __64BIT__ 1 +# endif +#endif + + +/* ********************************************************* +* Turn on Large Files support (>4GB) for 32-bit Linux/Unix +***********************************************************/ +#if !defined(__64BIT__) || defined(__MINGW32__) /* No point defining Large file for 64 bit but MinGW-w64 requires it */ +# if !defined(_FILE_OFFSET_BITS) +# define _FILE_OFFSET_BITS 64 /* turn off_t into a 64-bit type for ftello, fseeko */ +# endif +# if !defined(_LARGEFILE_SOURCE) /* obsolete macro, replaced with _FILE_OFFSET_BITS */ +# define _LARGEFILE_SOURCE 1 /* Large File Support extension (LFS) - fseeko, ftello */ +# endif +# if defined(_AIX) || defined(__hpux) +# define _LARGE_FILES /* Large file support on 32-bits AIX and HP-UX */ +# endif +#endif + + +/* ************************************************************ +* Detect POSIX version +* PLATFORM_POSIX_VERSION = 0 for non-Unix e.g. Windows +* PLATFORM_POSIX_VERSION = 1 for Unix-like but non-POSIX +* PLATFORM_POSIX_VERSION > 1 is equal to found _POSIX_VERSION +* Value of PLATFORM_POSIX_VERSION can be forced on command line +***************************************************************/ +#ifndef PLATFORM_POSIX_VERSION + +# if (defined(__APPLE__) && defined(__MACH__)) || defined(__SVR4) || defined(_AIX) || defined(__hpux) /* POSIX.1-2001 (SUSv3) conformant */ \ + || defined(__DragonFly__) || defined(__FreeBSD__) || defined(__NetBSD__) || defined(__OpenBSD__) /* BSD distros */ + /* exception rule : force posix version to 200112L, + * note: it's better to use unistd.h's _POSIX_VERSION whenever possible */ +# define PLATFORM_POSIX_VERSION 200112L + +/* try to determine posix version through official unistd.h's _POSIX_VERSION (http://pubs.opengroup.org/onlinepubs/7908799/xsh/unistd.h.html). + * note : there is no simple way to know in advance if <unistd.h> is present or not on target system, + * Posix specification mandates its presence and its content, but target system must respect this spec. + * It's necessary to _not_ #include <unistd.h> whenever target OS is not unix-like + * otherwise it will block preprocessing stage. + * The following list of build macros tries to "guess" if target OS is likely unix-like, and therefore can #include <unistd.h> + */ +# elif !defined(_WIN32) \ + && ( defined(__unix__) || defined(__unix) \ + || defined(__midipix__) || defined(__VMS) || defined(__HAIKU__) ) + +# if defined(__linux__) || defined(__linux) || defined(__CYGWIN__) +# ifndef _POSIX_C_SOURCE +# define _POSIX_C_SOURCE 200809L /* feature test macro : https://www.gnu.org/software/libc/manual/html_node/Feature-Test-Macros.html */ +# endif +# endif +# include <unistd.h> /* declares _POSIX_VERSION */ +# if defined(_POSIX_VERSION) /* POSIX compliant */ +# define PLATFORM_POSIX_VERSION _POSIX_VERSION +# else +# define PLATFORM_POSIX_VERSION 1 +# endif + +# ifdef __UCLIBC__ +# ifndef __USE_MISC +# define __USE_MISC /* enable st_mtim on uclibc */ +# endif +# endif + +# else /* non-unix target platform (like Windows) */ +# define PLATFORM_POSIX_VERSION 0 +# endif + +#endif /* PLATFORM_POSIX_VERSION */ + + +#if PLATFORM_POSIX_VERSION > 1 + /* glibc < 2.26 may not expose struct timespec def without this. + * See issue #1920. */ +# ifndef _ATFILE_SOURCE +# define _ATFILE_SOURCE +# endif +#endif + + +/*-********************************************* +* Detect if isatty() and fileno() are available +************************************************/ +#if (defined(__linux__) && (PLATFORM_POSIX_VERSION > 1)) \ + || (PLATFORM_POSIX_VERSION >= 200112L) \ + || defined(__DJGPP__) +# include <unistd.h> /* isatty */ +# include <stdio.h> /* fileno */ +# define IS_CONSOLE(stdStream) isatty(fileno(stdStream)) +#elif defined(MSDOS) || defined(OS2) +# include <io.h> /* _isatty */ +# define IS_CONSOLE(stdStream) _isatty(_fileno(stdStream)) +#elif defined(WIN32) || defined(_WIN32) +# include <io.h> /* _isatty */ +# include <windows.h> /* DeviceIoControl, HANDLE, FSCTL_SET_SPARSE */ +# include <stdio.h> /* FILE */ +static __inline int IS_CONSOLE(FILE* stdStream) { + DWORD dummy; + return _isatty(_fileno(stdStream)) && GetConsoleMode((HANDLE)_get_osfhandle(_fileno(stdStream)), &dummy); +} +#else +# define IS_CONSOLE(stdStream) 0 +#endif + + +/****************************** +* OS-specific IO behaviors +******************************/ +#if defined(MSDOS) || defined(OS2) || defined(WIN32) || defined(_WIN32) +# include <fcntl.h> /* _O_BINARY */ +# include <io.h> /* _setmode, _fileno, _get_osfhandle */ +# if !defined(__DJGPP__) +# include <windows.h> /* DeviceIoControl, HANDLE, FSCTL_SET_SPARSE */ +# include <winioctl.h> /* FSCTL_SET_SPARSE */ +# define SET_BINARY_MODE(file) { int const unused=_setmode(_fileno(file), _O_BINARY); (void)unused; } +# define SET_SPARSE_FILE_MODE(file) { DWORD dw; DeviceIoControl((HANDLE) _get_osfhandle(_fileno(file)), FSCTL_SET_SPARSE, 0, 0, 0, 0, &dw, 0); } +# else +# define SET_BINARY_MODE(file) setmode(fileno(file), O_BINARY) +# define SET_SPARSE_FILE_MODE(file) +# endif +#else +# define SET_BINARY_MODE(file) +# define SET_SPARSE_FILE_MODE(file) +#endif + + +#ifndef ZSTD_SPARSE_DEFAULT +# if (defined(__APPLE__) && defined(__MACH__)) +# define ZSTD_SPARSE_DEFAULT 0 +# else +# define ZSTD_SPARSE_DEFAULT 1 +# endif +#endif + + +#ifndef ZSTD_START_SYMBOLLIST_FRAME +# ifdef __linux__ +# define ZSTD_START_SYMBOLLIST_FRAME 2 +# elif defined __APPLE__ +# define ZSTD_START_SYMBOLLIST_FRAME 4 +# else +# define ZSTD_START_SYMBOLLIST_FRAME 0 +# endif +#endif + + +#ifndef ZSTD_SETPRIORITY_SUPPORT + /* mandates presence of <sys/resource.h> and support for setpriority() : http://man7.org/linux/man-pages/man2/setpriority.2.html */ +# define ZSTD_SETPRIORITY_SUPPORT (PLATFORM_POSIX_VERSION >= 200112L) +#endif + + +#ifndef ZSTD_NANOSLEEP_SUPPORT + /* mandates support of nanosleep() within <time.h> : http://man7.org/linux/man-pages/man2/nanosleep.2.html */ +# if (defined(__linux__) && (PLATFORM_POSIX_VERSION >= 199309L)) \ + || (PLATFORM_POSIX_VERSION >= 200112L) +# define ZSTD_NANOSLEEP_SUPPORT 1 +# else +# define ZSTD_NANOSLEEP_SUPPORT 0 +# endif +#endif + + +#if defined (__cplusplus) +} +#endif + +#endif /* PLATFORM_H_MODULE */ diff --git a/contrib/libs/zstd/programs/timefn.c b/contrib/libs/zstd/programs/timefn.c index 64577b0e93..539ada006e 100644 --- a/contrib/libs/zstd/programs/timefn.c +++ b/contrib/libs/zstd/programs/timefn.c @@ -1,169 +1,169 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - -/* === Dependencies === */ - -#include "timefn.h" - - -/*-**************************************** -* Time functions -******************************************/ - -#if defined(_WIN32) /* Windows */ - -#include <stdlib.h> /* abort */ -#include <stdio.h> /* perror */ - -UTIL_time_t UTIL_getTime(void) { UTIL_time_t x; QueryPerformanceCounter(&x); return x; } - -PTime UTIL_getSpanTimeMicro(UTIL_time_t clockStart, UTIL_time_t clockEnd) -{ - static LARGE_INTEGER ticksPerSecond; - static int init = 0; - if (!init) { - if (!QueryPerformanceFrequency(&ticksPerSecond)) { - perror("timefn::QueryPerformanceFrequency"); - abort(); - } - init = 1; - } - return 1000000ULL*(clockEnd.QuadPart - clockStart.QuadPart)/ticksPerSecond.QuadPart; -} - -PTime UTIL_getSpanTimeNano(UTIL_time_t clockStart, UTIL_time_t clockEnd) -{ - static LARGE_INTEGER ticksPerSecond; - static int init = 0; - if (!init) { - if (!QueryPerformanceFrequency(&ticksPerSecond)) { - perror("timefn::QueryPerformanceFrequency"); - abort(); - } - init = 1; - } - return 1000000000ULL*(clockEnd.QuadPart - clockStart.QuadPart)/ticksPerSecond.QuadPart; -} - - - -#elif defined(__APPLE__) && defined(__MACH__) - -UTIL_time_t UTIL_getTime(void) { return mach_absolute_time(); } - -PTime UTIL_getSpanTimeMicro(UTIL_time_t clockStart, UTIL_time_t clockEnd) -{ - static mach_timebase_info_data_t rate; - static int init = 0; - if (!init) { - mach_timebase_info(&rate); - init = 1; - } - return (((clockEnd - clockStart) * (PTime)rate.numer) / ((PTime)rate.denom))/1000ULL; -} - -PTime UTIL_getSpanTimeNano(UTIL_time_t clockStart, UTIL_time_t clockEnd) -{ - static mach_timebase_info_data_t rate; - static int init = 0; - if (!init) { - mach_timebase_info(&rate); - init = 1; - } - return ((clockEnd - clockStart) * (PTime)rate.numer) / ((PTime)rate.denom); -} - - -/* C11 requires timespec_get, but FreeBSD 11 lacks it, while still claiming C11 compliance. - Android also lacks it but does define TIME_UTC. */ -#elif (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 201112L) /* C11 */) \ - && defined(TIME_UTC) && !defined(__ANDROID__) - -#include <stdlib.h> /* abort */ -#include <stdio.h> /* perror */ - -UTIL_time_t UTIL_getTime(void) -{ - /* time must be initialized, othersize it may fail msan test. - * No good reason, likely a limitation of timespec_get() for some target */ - UTIL_time_t time = UTIL_TIME_INITIALIZER; - if (timespec_get(&time, TIME_UTC) != TIME_UTC) { - perror("timefn::timespec_get"); - abort(); - } - return time; -} - -static UTIL_time_t UTIL_getSpanTime(UTIL_time_t begin, UTIL_time_t end) -{ - UTIL_time_t diff; - if (end.tv_nsec < begin.tv_nsec) { - diff.tv_sec = (end.tv_sec - 1) - begin.tv_sec; - diff.tv_nsec = (end.tv_nsec + 1000000000ULL) - begin.tv_nsec; - } else { - diff.tv_sec = end.tv_sec - begin.tv_sec; - diff.tv_nsec = end.tv_nsec - begin.tv_nsec; - } - return diff; -} - -PTime UTIL_getSpanTimeMicro(UTIL_time_t begin, UTIL_time_t end) -{ - UTIL_time_t const diff = UTIL_getSpanTime(begin, end); - PTime micro = 0; - micro += 1000000ULL * diff.tv_sec; - micro += diff.tv_nsec / 1000ULL; - return micro; -} - -PTime UTIL_getSpanTimeNano(UTIL_time_t begin, UTIL_time_t end) -{ - UTIL_time_t const diff = UTIL_getSpanTime(begin, end); - PTime nano = 0; - nano += 1000000000ULL * diff.tv_sec; - nano += diff.tv_nsec; - return nano; -} - - - -#else /* relies on standard C90 (note : clock_t measurements can be wrong when using multi-threading) */ - -UTIL_time_t UTIL_getTime(void) { return clock(); } -PTime UTIL_getSpanTimeMicro(UTIL_time_t clockStart, UTIL_time_t clockEnd) { return 1000000ULL * (clockEnd - clockStart) / CLOCKS_PER_SEC; } -PTime UTIL_getSpanTimeNano(UTIL_time_t clockStart, UTIL_time_t clockEnd) { return 1000000000ULL * (clockEnd - clockStart) / CLOCKS_PER_SEC; } - -#endif - - - -/* returns time span in microseconds */ -PTime UTIL_clockSpanMicro(UTIL_time_t clockStart ) -{ - UTIL_time_t const clockEnd = UTIL_getTime(); - return UTIL_getSpanTimeMicro(clockStart, clockEnd); -} - -/* returns time span in microseconds */ -PTime UTIL_clockSpanNano(UTIL_time_t clockStart ) -{ - UTIL_time_t const clockEnd = UTIL_getTime(); - return UTIL_getSpanTimeNano(clockStart, clockEnd); -} - -void UTIL_waitForNextTick(void) -{ - UTIL_time_t const clockStart = UTIL_getTime(); - UTIL_time_t clockEnd; - do { - clockEnd = UTIL_getTime(); - } while (UTIL_getSpanTimeNano(clockStart, clockEnd) == 0); -} +/* + * Copyright (c) Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + + +/* === Dependencies === */ + +#include "timefn.h" + + +/*-**************************************** +* Time functions +******************************************/ + +#if defined(_WIN32) /* Windows */ + +#include <stdlib.h> /* abort */ +#include <stdio.h> /* perror */ + +UTIL_time_t UTIL_getTime(void) { UTIL_time_t x; QueryPerformanceCounter(&x); return x; } + +PTime UTIL_getSpanTimeMicro(UTIL_time_t clockStart, UTIL_time_t clockEnd) +{ + static LARGE_INTEGER ticksPerSecond; + static int init = 0; + if (!init) { + if (!QueryPerformanceFrequency(&ticksPerSecond)) { + perror("timefn::QueryPerformanceFrequency"); + abort(); + } + init = 1; + } + return 1000000ULL*(clockEnd.QuadPart - clockStart.QuadPart)/ticksPerSecond.QuadPart; +} + +PTime UTIL_getSpanTimeNano(UTIL_time_t clockStart, UTIL_time_t clockEnd) +{ + static LARGE_INTEGER ticksPerSecond; + static int init = 0; + if (!init) { + if (!QueryPerformanceFrequency(&ticksPerSecond)) { + perror("timefn::QueryPerformanceFrequency"); + abort(); + } + init = 1; + } + return 1000000000ULL*(clockEnd.QuadPart - clockStart.QuadPart)/ticksPerSecond.QuadPart; +} + + + +#elif defined(__APPLE__) && defined(__MACH__) + +UTIL_time_t UTIL_getTime(void) { return mach_absolute_time(); } + +PTime UTIL_getSpanTimeMicro(UTIL_time_t clockStart, UTIL_time_t clockEnd) +{ + static mach_timebase_info_data_t rate; + static int init = 0; + if (!init) { + mach_timebase_info(&rate); + init = 1; + } + return (((clockEnd - clockStart) * (PTime)rate.numer) / ((PTime)rate.denom))/1000ULL; +} + +PTime UTIL_getSpanTimeNano(UTIL_time_t clockStart, UTIL_time_t clockEnd) +{ + static mach_timebase_info_data_t rate; + static int init = 0; + if (!init) { + mach_timebase_info(&rate); + init = 1; + } + return ((clockEnd - clockStart) * (PTime)rate.numer) / ((PTime)rate.denom); +} + + +/* C11 requires timespec_get, but FreeBSD 11 lacks it, while still claiming C11 compliance. + Android also lacks it but does define TIME_UTC. */ +#elif (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 201112L) /* C11 */) \ + && defined(TIME_UTC) && !defined(__ANDROID__) + +#include <stdlib.h> /* abort */ +#include <stdio.h> /* perror */ + +UTIL_time_t UTIL_getTime(void) +{ + /* time must be initialized, othersize it may fail msan test. + * No good reason, likely a limitation of timespec_get() for some target */ + UTIL_time_t time = UTIL_TIME_INITIALIZER; + if (timespec_get(&time, TIME_UTC) != TIME_UTC) { + perror("timefn::timespec_get"); + abort(); + } + return time; +} + +static UTIL_time_t UTIL_getSpanTime(UTIL_time_t begin, UTIL_time_t end) +{ + UTIL_time_t diff; + if (end.tv_nsec < begin.tv_nsec) { + diff.tv_sec = (end.tv_sec - 1) - begin.tv_sec; + diff.tv_nsec = (end.tv_nsec + 1000000000ULL) - begin.tv_nsec; + } else { + diff.tv_sec = end.tv_sec - begin.tv_sec; + diff.tv_nsec = end.tv_nsec - begin.tv_nsec; + } + return diff; +} + +PTime UTIL_getSpanTimeMicro(UTIL_time_t begin, UTIL_time_t end) +{ + UTIL_time_t const diff = UTIL_getSpanTime(begin, end); + PTime micro = 0; + micro += 1000000ULL * diff.tv_sec; + micro += diff.tv_nsec / 1000ULL; + return micro; +} + +PTime UTIL_getSpanTimeNano(UTIL_time_t begin, UTIL_time_t end) +{ + UTIL_time_t const diff = UTIL_getSpanTime(begin, end); + PTime nano = 0; + nano += 1000000000ULL * diff.tv_sec; + nano += diff.tv_nsec; + return nano; +} + + + +#else /* relies on standard C90 (note : clock_t measurements can be wrong when using multi-threading) */ + +UTIL_time_t UTIL_getTime(void) { return clock(); } +PTime UTIL_getSpanTimeMicro(UTIL_time_t clockStart, UTIL_time_t clockEnd) { return 1000000ULL * (clockEnd - clockStart) / CLOCKS_PER_SEC; } +PTime UTIL_getSpanTimeNano(UTIL_time_t clockStart, UTIL_time_t clockEnd) { return 1000000000ULL * (clockEnd - clockStart) / CLOCKS_PER_SEC; } + +#endif + + + +/* returns time span in microseconds */ +PTime UTIL_clockSpanMicro(UTIL_time_t clockStart ) +{ + UTIL_time_t const clockEnd = UTIL_getTime(); + return UTIL_getSpanTimeMicro(clockStart, clockEnd); +} + +/* returns time span in microseconds */ +PTime UTIL_clockSpanNano(UTIL_time_t clockStart ) +{ + UTIL_time_t const clockEnd = UTIL_getTime(); + return UTIL_getSpanTimeNano(clockStart, clockEnd); +} + +void UTIL_waitForNextTick(void) +{ + UTIL_time_t const clockStart = UTIL_getTime(); + UTIL_time_t clockEnd; + do { + clockEnd = UTIL_getTime(); + } while (UTIL_getSpanTimeNano(clockStart, clockEnd) == 0); +} diff --git a/contrib/libs/zstd/programs/timefn.h b/contrib/libs/zstd/programs/timefn.h index 3fcd78a28e..29de0f3b3e 100644 --- a/contrib/libs/zstd/programs/timefn.h +++ b/contrib/libs/zstd/programs/timefn.h @@ -1,89 +1,89 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef TIME_FN_H_MODULE_287987 -#define TIME_FN_H_MODULE_287987 - -#if defined (__cplusplus) -extern "C" { -#endif - - -/*-**************************************** -* Dependencies -******************************************/ -#include <time.h> /* clock_t, clock, CLOCKS_PER_SEC */ - - - -/*-**************************************** -* Local Types -******************************************/ - -#if !defined (__VMS) && (defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) ) -# if defined(_AIX) -# include <inttypes.h> -# else -# include <stdint.h> /* intptr_t */ -# endif - typedef uint64_t PTime; /* Precise Time */ -#else - typedef unsigned long long PTime; /* does not support compilers without long long support */ -#endif - - - -/*-**************************************** -* Time functions -******************************************/ -#if defined(_WIN32) /* Windows */ - - #include <windows.h> /* LARGE_INTEGER */ - typedef LARGE_INTEGER UTIL_time_t; - #define UTIL_TIME_INITIALIZER { { 0, 0 } } - -#elif defined(__APPLE__) && defined(__MACH__) - - #include <mach/mach_time.h> - typedef PTime UTIL_time_t; - #define UTIL_TIME_INITIALIZER 0 - -/* C11 requires timespec_get, but FreeBSD 11 lacks it, while still claiming C11 compliance. - Android also lacks it but does define TIME_UTC. */ -#elif (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 201112L) /* C11 */) \ - && defined(TIME_UTC) && !defined(__ANDROID__) - - typedef struct timespec UTIL_time_t; - #define UTIL_TIME_INITIALIZER { 0, 0 } - -#else /* relies on standard C90 (note : clock_t measurements can be wrong when using multi-threading) */ - - typedef clock_t UTIL_time_t; - #define UTIL_TIME_INITIALIZER 0 - -#endif - - -UTIL_time_t UTIL_getTime(void); -PTime UTIL_getSpanTimeMicro(UTIL_time_t clockStart, UTIL_time_t clockEnd); -PTime UTIL_getSpanTimeNano(UTIL_time_t clockStart, UTIL_time_t clockEnd); - -#define SEC_TO_MICRO ((PTime)1000000) -PTime UTIL_clockSpanMicro(UTIL_time_t clockStart); -PTime UTIL_clockSpanNano(UTIL_time_t clockStart); - -void UTIL_waitForNextTick(void); - - -#if defined (__cplusplus) -} -#endif - -#endif /* TIME_FN_H_MODULE_287987 */ +/* + * Copyright (c) Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + +#ifndef TIME_FN_H_MODULE_287987 +#define TIME_FN_H_MODULE_287987 + +#if defined (__cplusplus) +extern "C" { +#endif + + +/*-**************************************** +* Dependencies +******************************************/ +#include <time.h> /* clock_t, clock, CLOCKS_PER_SEC */ + + + +/*-**************************************** +* Local Types +******************************************/ + +#if !defined (__VMS) && (defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) ) +# if defined(_AIX) +# include <inttypes.h> +# else +# include <stdint.h> /* intptr_t */ +# endif + typedef uint64_t PTime; /* Precise Time */ +#else + typedef unsigned long long PTime; /* does not support compilers without long long support */ +#endif + + + +/*-**************************************** +* Time functions +******************************************/ +#if defined(_WIN32) /* Windows */ + + #include <windows.h> /* LARGE_INTEGER */ + typedef LARGE_INTEGER UTIL_time_t; + #define UTIL_TIME_INITIALIZER { { 0, 0 } } + +#elif defined(__APPLE__) && defined(__MACH__) + + #include <mach/mach_time.h> + typedef PTime UTIL_time_t; + #define UTIL_TIME_INITIALIZER 0 + +/* C11 requires timespec_get, but FreeBSD 11 lacks it, while still claiming C11 compliance. + Android also lacks it but does define TIME_UTC. */ +#elif (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 201112L) /* C11 */) \ + && defined(TIME_UTC) && !defined(__ANDROID__) + + typedef struct timespec UTIL_time_t; + #define UTIL_TIME_INITIALIZER { 0, 0 } + +#else /* relies on standard C90 (note : clock_t measurements can be wrong when using multi-threading) */ + + typedef clock_t UTIL_time_t; + #define UTIL_TIME_INITIALIZER 0 + +#endif + + +UTIL_time_t UTIL_getTime(void); +PTime UTIL_getSpanTimeMicro(UTIL_time_t clockStart, UTIL_time_t clockEnd); +PTime UTIL_getSpanTimeNano(UTIL_time_t clockStart, UTIL_time_t clockEnd); + +#define SEC_TO_MICRO ((PTime)1000000) +PTime UTIL_clockSpanMicro(UTIL_time_t clockStart); +PTime UTIL_clockSpanNano(UTIL_time_t clockStart); + +void UTIL_waitForNextTick(void); + + +#if defined (__cplusplus) +} +#endif + +#endif /* TIME_FN_H_MODULE_287987 */ diff --git a/contrib/libs/zstd/programs/util.c b/contrib/libs/zstd/programs/util.c index d69b72a37c..36140260ed 100644 --- a/contrib/libs/zstd/programs/util.c +++ b/contrib/libs/zstd/programs/util.c @@ -1,164 +1,164 @@ -/* - * Copyright (c) Przemyslaw Skibinski, Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#if defined (__cplusplus) -extern "C" { -#endif - - -/*-**************************************** -* Dependencies -******************************************/ -#include "util.h" /* note : ensure that platform.h is included first ! */ -#include <stdlib.h> /* malloc, realloc, free */ -#include <stdio.h> /* fprintf */ -#include <time.h> /* clock_t, clock, CLOCKS_PER_SEC, nanosleep */ -#include <errno.h> -#include <assert.h> - -#if defined(_WIN32) -# include <sys/utime.h> /* utime */ -# include <io.h> /* _chmod */ -#else -# include <unistd.h> /* chown, stat */ -# if PLATFORM_POSIX_VERSION < 200809L || !defined(st_mtime) -# include <utime.h> /* utime */ -# else -# include <fcntl.h> /* AT_FDCWD */ -# include <sys/stat.h> /* utimensat */ -# endif -#endif - -#if defined(_MSC_VER) || defined(__MINGW32__) || defined (__MSVCRT__) -#include <direct.h> /* needed for _mkdir in windows */ -#endif - -#if defined(__linux__) || (PLATFORM_POSIX_VERSION >= 200112L) /* opendir, readdir require POSIX.1-2001 */ -# include <dirent.h> /* opendir, readdir */ -# include <string.h> /* strerror, memcpy */ -#endif /* #ifdef _WIN32 */ - -/*-**************************************** -* Internal Macros -******************************************/ - -/* CONTROL is almost like an assert(), but is never disabled. - * It's designed for failures that may happen rarely, - * but we don't want to maintain a specific error code path for them, - * such as a malloc() returning NULL for example. - * Since it's always active, this macro can trigger side effects. - */ -#define CONTROL(c) { \ - if (!(c)) { \ - UTIL_DISPLAYLEVEL(1, "Error : %s, %i : %s", \ - __FILE__, __LINE__, #c); \ - exit(1); \ -} } - -/* console log */ -#define UTIL_DISPLAY(...) fprintf(stderr, __VA_ARGS__) -#define UTIL_DISPLAYLEVEL(l, ...) { if (g_utilDisplayLevel>=l) { UTIL_DISPLAY(__VA_ARGS__); } } - -/* A modified version of realloc(). - * If UTIL_realloc() fails the original block is freed. - */ -UTIL_STATIC void* UTIL_realloc(void *ptr, size_t size) -{ - void *newptr = realloc(ptr, size); - if (newptr) return newptr; - free(ptr); - return NULL; -} - -#if defined(_MSC_VER) - #define chmod _chmod -#endif - - -/*-**************************************** -* Console log -******************************************/ -int g_utilDisplayLevel; - -int UTIL_requireUserConfirmation(const char* prompt, const char* abortMsg, - const char* acceptableLetters, int hasStdinInput) { - int ch, result; - - if (hasStdinInput) { - UTIL_DISPLAY("stdin is an input - not proceeding.\n"); - return 1; - } - - UTIL_DISPLAY("%s", prompt); - ch = getchar(); - result = 0; - if (strchr(acceptableLetters, ch) == NULL) { - UTIL_DISPLAY("%s", abortMsg); - result = 1; - } - /* flush the rest */ - while ((ch!=EOF) && (ch!='\n')) - ch = getchar(); - return result; -} - - -/*-************************************* -* Constants -***************************************/ -#define LIST_SIZE_INCREASE (8*1024) -#define MAX_FILE_OF_FILE_NAMES_SIZE (1<<20)*50 - - -/*-************************************* -* Functions -***************************************/ - -int UTIL_stat(const char* filename, stat_t* statbuf) -{ -#if defined(_MSC_VER) - return !_stat64(filename, statbuf); -#elif defined(__MINGW32__) && defined (__MSVCRT__) - return !_stati64(filename, statbuf); -#else - return !stat(filename, statbuf); -#endif -} - -int UTIL_isRegularFile(const char* infilename) -{ - stat_t statbuf; - return UTIL_stat(infilename, &statbuf) && UTIL_isRegularFileStat(&statbuf); -} - -int UTIL_isRegularFileStat(const stat_t* statbuf) -{ -#if defined(_MSC_VER) - return (statbuf->st_mode & S_IFREG) != 0; -#else - return S_ISREG(statbuf->st_mode) != 0; -#endif -} - -/* like chmod, but avoid changing permission of /dev/null */ -int UTIL_chmod(char const* filename, const stat_t* statbuf, mode_t permissions) -{ - stat_t localStatBuf; - if (statbuf == NULL) { - if (!UTIL_stat(filename, &localStatBuf)) return 0; - statbuf = &localStatBuf; - } - if (!UTIL_isRegularFileStat(statbuf)) return 0; /* pretend success, but don't change anything */ - return chmod(filename, permissions); -} - +/* + * Copyright (c) Przemyslaw Skibinski, Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + +#if defined (__cplusplus) +extern "C" { +#endif + + +/*-**************************************** +* Dependencies +******************************************/ +#include "util.h" /* note : ensure that platform.h is included first ! */ +#include <stdlib.h> /* malloc, realloc, free */ +#include <stdio.h> /* fprintf */ +#include <time.h> /* clock_t, clock, CLOCKS_PER_SEC, nanosleep */ +#include <errno.h> +#include <assert.h> + +#if defined(_WIN32) +# include <sys/utime.h> /* utime */ +# include <io.h> /* _chmod */ +#else +# include <unistd.h> /* chown, stat */ +# if PLATFORM_POSIX_VERSION < 200809L || !defined(st_mtime) +# include <utime.h> /* utime */ +# else +# include <fcntl.h> /* AT_FDCWD */ +# include <sys/stat.h> /* utimensat */ +# endif +#endif + +#if defined(_MSC_VER) || defined(__MINGW32__) || defined (__MSVCRT__) +#include <direct.h> /* needed for _mkdir in windows */ +#endif + +#if defined(__linux__) || (PLATFORM_POSIX_VERSION >= 200112L) /* opendir, readdir require POSIX.1-2001 */ +# include <dirent.h> /* opendir, readdir */ +# include <string.h> /* strerror, memcpy */ +#endif /* #ifdef _WIN32 */ + +/*-**************************************** +* Internal Macros +******************************************/ + +/* CONTROL is almost like an assert(), but is never disabled. + * It's designed for failures that may happen rarely, + * but we don't want to maintain a specific error code path for them, + * such as a malloc() returning NULL for example. + * Since it's always active, this macro can trigger side effects. + */ +#define CONTROL(c) { \ + if (!(c)) { \ + UTIL_DISPLAYLEVEL(1, "Error : %s, %i : %s", \ + __FILE__, __LINE__, #c); \ + exit(1); \ +} } + +/* console log */ +#define UTIL_DISPLAY(...) fprintf(stderr, __VA_ARGS__) +#define UTIL_DISPLAYLEVEL(l, ...) { if (g_utilDisplayLevel>=l) { UTIL_DISPLAY(__VA_ARGS__); } } + +/* A modified version of realloc(). + * If UTIL_realloc() fails the original block is freed. + */ +UTIL_STATIC void* UTIL_realloc(void *ptr, size_t size) +{ + void *newptr = realloc(ptr, size); + if (newptr) return newptr; + free(ptr); + return NULL; +} + +#if defined(_MSC_VER) + #define chmod _chmod +#endif + + +/*-**************************************** +* Console log +******************************************/ +int g_utilDisplayLevel; + +int UTIL_requireUserConfirmation(const char* prompt, const char* abortMsg, + const char* acceptableLetters, int hasStdinInput) { + int ch, result; + + if (hasStdinInput) { + UTIL_DISPLAY("stdin is an input - not proceeding.\n"); + return 1; + } + + UTIL_DISPLAY("%s", prompt); + ch = getchar(); + result = 0; + if (strchr(acceptableLetters, ch) == NULL) { + UTIL_DISPLAY("%s", abortMsg); + result = 1; + } + /* flush the rest */ + while ((ch!=EOF) && (ch!='\n')) + ch = getchar(); + return result; +} + + +/*-************************************* +* Constants +***************************************/ +#define LIST_SIZE_INCREASE (8*1024) +#define MAX_FILE_OF_FILE_NAMES_SIZE (1<<20)*50 + + +/*-************************************* +* Functions +***************************************/ + +int UTIL_stat(const char* filename, stat_t* statbuf) +{ +#if defined(_MSC_VER) + return !_stat64(filename, statbuf); +#elif defined(__MINGW32__) && defined (__MSVCRT__) + return !_stati64(filename, statbuf); +#else + return !stat(filename, statbuf); +#endif +} + +int UTIL_isRegularFile(const char* infilename) +{ + stat_t statbuf; + return UTIL_stat(infilename, &statbuf) && UTIL_isRegularFileStat(&statbuf); +} + +int UTIL_isRegularFileStat(const stat_t* statbuf) +{ +#if defined(_MSC_VER) + return (statbuf->st_mode & S_IFREG) != 0; +#else + return S_ISREG(statbuf->st_mode) != 0; +#endif +} + +/* like chmod, but avoid changing permission of /dev/null */ +int UTIL_chmod(char const* filename, const stat_t* statbuf, mode_t permissions) +{ + stat_t localStatBuf; + if (statbuf == NULL) { + if (!UTIL_stat(filename, &localStatBuf)) return 0; + statbuf = &localStatBuf; + } + if (!UTIL_isRegularFileStat(statbuf)) return 0; /* pretend success, but don't change anything */ + return chmod(filename, permissions); +} + /* set access and modification times */ int UTIL_utime(const char* filename, const stat_t *statbuf) { @@ -182,136 +182,136 @@ int UTIL_utime(const char* filename, const stat_t *statbuf) return ret; } -int UTIL_setFileStat(const char *filename, const stat_t *statbuf) -{ - int res = 0; - - stat_t curStatBuf; - if (!UTIL_stat(filename, &curStatBuf) || !UTIL_isRegularFileStat(&curStatBuf)) - return -1; - - /* set access and modification times */ +int UTIL_setFileStat(const char *filename, const stat_t *statbuf) +{ + int res = 0; + + stat_t curStatBuf; + if (!UTIL_stat(filename, &curStatBuf) || !UTIL_isRegularFileStat(&curStatBuf)) + return -1; + + /* set access and modification times */ res += UTIL_utime(filename, statbuf); - -#if !defined(_WIN32) - res += chown(filename, statbuf->st_uid, statbuf->st_gid); /* Copy ownership */ -#endif - - res += UTIL_chmod(filename, &curStatBuf, statbuf->st_mode & 07777); /* Copy file permissions */ - - errno = 0; - return -res; /* number of errors is returned */ -} - -int UTIL_isDirectory(const char* infilename) -{ - stat_t statbuf; - return UTIL_stat(infilename, &statbuf) && UTIL_isDirectoryStat(&statbuf); -} - -int UTIL_isDirectoryStat(const stat_t* statbuf) -{ -#if defined(_MSC_VER) - return (statbuf->st_mode & _S_IFDIR) != 0; -#else - return S_ISDIR(statbuf->st_mode) != 0; -#endif -} - -int UTIL_compareStr(const void *p1, const void *p2) { - return strcmp(* (char * const *) p1, * (char * const *) p2); -} - -int UTIL_isSameFile(const char* fName1, const char* fName2) -{ - assert(fName1 != NULL); assert(fName2 != NULL); -#if defined(_MSC_VER) || defined(_WIN32) - /* note : Visual does not support file identification by inode. - * inode does not work on Windows, even with a posix layer, like msys2. - * The following work-around is limited to detecting exact name repetition only, - * aka `filename` is considered different from `subdir/../filename` */ - return !strcmp(fName1, fName2); -#else - { stat_t file1Stat; - stat_t file2Stat; - return UTIL_stat(fName1, &file1Stat) - && UTIL_stat(fName2, &file2Stat) - && (file1Stat.st_dev == file2Stat.st_dev) - && (file1Stat.st_ino == file2Stat.st_ino); - } -#endif -} - -/* UTIL_isFIFO : distinguish named pipes */ -int UTIL_isFIFO(const char* infilename) -{ -/* macro guards, as defined in : https://linux.die.net/man/2/lstat */ -#if PLATFORM_POSIX_VERSION >= 200112L - stat_t statbuf; - if (UTIL_stat(infilename, &statbuf) && UTIL_isFIFOStat(&statbuf)) return 1; -#endif - (void)infilename; - return 0; -} - -/* UTIL_isFIFO : distinguish named pipes */ -int UTIL_isFIFOStat(const stat_t* statbuf) -{ -/* macro guards, as defined in : https://linux.die.net/man/2/lstat */ -#if PLATFORM_POSIX_VERSION >= 200112L - if (S_ISFIFO(statbuf->st_mode)) return 1; -#endif - (void)statbuf; - return 0; -} - -/* UTIL_isBlockDevStat : distinguish named pipes */ -int UTIL_isBlockDevStat(const stat_t* statbuf) -{ -/* macro guards, as defined in : https://linux.die.net/man/2/lstat */ -#if PLATFORM_POSIX_VERSION >= 200112L - if (S_ISBLK(statbuf->st_mode)) return 1; -#endif - (void)statbuf; - return 0; -} - -int UTIL_isLink(const char* infilename) -{ -/* macro guards, as defined in : https://linux.die.net/man/2/lstat */ -#if PLATFORM_POSIX_VERSION >= 200112L - stat_t statbuf; - int const r = lstat(infilename, &statbuf); - if (!r && S_ISLNK(statbuf.st_mode)) return 1; -#endif - (void)infilename; - return 0; -} - -U64 UTIL_getFileSize(const char* infilename) -{ - stat_t statbuf; - if (!UTIL_stat(infilename, &statbuf)) return UTIL_FILESIZE_UNKNOWN; - return UTIL_getFileSizeStat(&statbuf); -} - -U64 UTIL_getFileSizeStat(const stat_t* statbuf) -{ - if (!UTIL_isRegularFileStat(statbuf)) return UTIL_FILESIZE_UNKNOWN; -#if defined(_MSC_VER) - if (!(statbuf->st_mode & S_IFREG)) return UTIL_FILESIZE_UNKNOWN; -#elif defined(__MINGW32__) && defined (__MSVCRT__) - if (!(statbuf->st_mode & S_IFREG)) return UTIL_FILESIZE_UNKNOWN; -#else - if (!S_ISREG(statbuf->st_mode)) return UTIL_FILESIZE_UNKNOWN; -#endif - return (U64)statbuf->st_size; -} - + +#if !defined(_WIN32) + res += chown(filename, statbuf->st_uid, statbuf->st_gid); /* Copy ownership */ +#endif + + res += UTIL_chmod(filename, &curStatBuf, statbuf->st_mode & 07777); /* Copy file permissions */ + + errno = 0; + return -res; /* number of errors is returned */ +} + +int UTIL_isDirectory(const char* infilename) +{ + stat_t statbuf; + return UTIL_stat(infilename, &statbuf) && UTIL_isDirectoryStat(&statbuf); +} + +int UTIL_isDirectoryStat(const stat_t* statbuf) +{ +#if defined(_MSC_VER) + return (statbuf->st_mode & _S_IFDIR) != 0; +#else + return S_ISDIR(statbuf->st_mode) != 0; +#endif +} + +int UTIL_compareStr(const void *p1, const void *p2) { + return strcmp(* (char * const *) p1, * (char * const *) p2); +} + +int UTIL_isSameFile(const char* fName1, const char* fName2) +{ + assert(fName1 != NULL); assert(fName2 != NULL); +#if defined(_MSC_VER) || defined(_WIN32) + /* note : Visual does not support file identification by inode. + * inode does not work on Windows, even with a posix layer, like msys2. + * The following work-around is limited to detecting exact name repetition only, + * aka `filename` is considered different from `subdir/../filename` */ + return !strcmp(fName1, fName2); +#else + { stat_t file1Stat; + stat_t file2Stat; + return UTIL_stat(fName1, &file1Stat) + && UTIL_stat(fName2, &file2Stat) + && (file1Stat.st_dev == file2Stat.st_dev) + && (file1Stat.st_ino == file2Stat.st_ino); + } +#endif +} + +/* UTIL_isFIFO : distinguish named pipes */ +int UTIL_isFIFO(const char* infilename) +{ +/* macro guards, as defined in : https://linux.die.net/man/2/lstat */ +#if PLATFORM_POSIX_VERSION >= 200112L + stat_t statbuf; + if (UTIL_stat(infilename, &statbuf) && UTIL_isFIFOStat(&statbuf)) return 1; +#endif + (void)infilename; + return 0; +} + +/* UTIL_isFIFO : distinguish named pipes */ +int UTIL_isFIFOStat(const stat_t* statbuf) +{ +/* macro guards, as defined in : https://linux.die.net/man/2/lstat */ +#if PLATFORM_POSIX_VERSION >= 200112L + if (S_ISFIFO(statbuf->st_mode)) return 1; +#endif + (void)statbuf; + return 0; +} + +/* UTIL_isBlockDevStat : distinguish named pipes */ +int UTIL_isBlockDevStat(const stat_t* statbuf) +{ +/* macro guards, as defined in : https://linux.die.net/man/2/lstat */ +#if PLATFORM_POSIX_VERSION >= 200112L + if (S_ISBLK(statbuf->st_mode)) return 1; +#endif + (void)statbuf; + return 0; +} + +int UTIL_isLink(const char* infilename) +{ +/* macro guards, as defined in : https://linux.die.net/man/2/lstat */ +#if PLATFORM_POSIX_VERSION >= 200112L + stat_t statbuf; + int const r = lstat(infilename, &statbuf); + if (!r && S_ISLNK(statbuf.st_mode)) return 1; +#endif + (void)infilename; + return 0; +} + +U64 UTIL_getFileSize(const char* infilename) +{ + stat_t statbuf; + if (!UTIL_stat(infilename, &statbuf)) return UTIL_FILESIZE_UNKNOWN; + return UTIL_getFileSizeStat(&statbuf); +} + +U64 UTIL_getFileSizeStat(const stat_t* statbuf) +{ + if (!UTIL_isRegularFileStat(statbuf)) return UTIL_FILESIZE_UNKNOWN; +#if defined(_MSC_VER) + if (!(statbuf->st_mode & S_IFREG)) return UTIL_FILESIZE_UNKNOWN; +#elif defined(__MINGW32__) && defined (__MSVCRT__) + if (!(statbuf->st_mode & S_IFREG)) return UTIL_FILESIZE_UNKNOWN; +#else + if (!S_ISREG(statbuf->st_mode)) return UTIL_FILESIZE_UNKNOWN; +#endif + return (U64)statbuf->st_size; +} + UTIL_HumanReadableSize_t UTIL_makeHumanReadableSize(U64 size) { UTIL_HumanReadableSize_t hrs; - + if (g_utilDisplayLevel > 3) { /* In verbose mode, do not scale sizes down, except in the case of * values that exceed the integral precision of a double. */ @@ -365,766 +365,766 @@ UTIL_HumanReadableSize_t UTIL_makeHumanReadableSize(U64 size) return hrs; } -U64 UTIL_getTotalFileSize(const char* const * fileNamesTable, unsigned nbFiles) -{ - U64 total = 0; - unsigned n; - for (n=0; n<nbFiles; n++) { - U64 const size = UTIL_getFileSize(fileNamesTable[n]); - if (size == UTIL_FILESIZE_UNKNOWN) return UTIL_FILESIZE_UNKNOWN; - total += size; - } - return total; -} - - -/* condition : @file must be valid, and not have reached its end. - * @return : length of line written into @buf, ended with `\0` instead of '\n', - * or 0, if there is no new line */ -static size_t readLineFromFile(char* buf, size_t len, FILE* file) -{ - assert(!feof(file)); - if ( fgets(buf, (int) len, file) == NULL ) return 0; - { size_t linelen = strlen(buf); - if (strlen(buf)==0) return 0; - if (buf[linelen-1] == '\n') linelen--; - buf[linelen] = '\0'; - return linelen+1; - } -} - -/* Conditions : - * size of @inputFileName file must be < @dstCapacity - * @dst must be initialized - * @return : nb of lines - * or -1 if there's an error - */ -static int -readLinesFromFile(void* dst, size_t dstCapacity, - const char* inputFileName) -{ - int nbFiles = 0; - size_t pos = 0; - char* const buf = (char*)dst; - FILE* const inputFile = fopen(inputFileName, "r"); - - assert(dst != NULL); - - if(!inputFile) { - if (g_utilDisplayLevel >= 1) perror("zstd:util:readLinesFromFile"); - return -1; - } - - while ( !feof(inputFile) ) { - size_t const lineLength = readLineFromFile(buf+pos, dstCapacity-pos, inputFile); - if (lineLength == 0) break; - assert(pos + lineLength < dstCapacity); - pos += lineLength; - ++nbFiles; - } - - CONTROL( fclose(inputFile) == 0 ); - - return nbFiles; -} - -/*Note: buf is not freed in case function successfully created table because filesTable->fileNames[0] = buf*/ -FileNamesTable* -UTIL_createFileNamesTable_fromFileName(const char* inputFileName) -{ - size_t nbFiles = 0; - char* buf; - size_t bufSize; - size_t pos = 0; - stat_t statbuf; - - if (!UTIL_stat(inputFileName, &statbuf) || !UTIL_isRegularFileStat(&statbuf)) - return NULL; - - { U64 const inputFileSize = UTIL_getFileSizeStat(&statbuf); - if(inputFileSize > MAX_FILE_OF_FILE_NAMES_SIZE) - return NULL; - bufSize = (size_t)(inputFileSize + 1); /* (+1) to add '\0' at the end of last filename */ - } - - buf = (char*) malloc(bufSize); - CONTROL( buf != NULL ); - - { int const ret_nbFiles = readLinesFromFile(buf, bufSize, inputFileName); - - if (ret_nbFiles <= 0) { - free(buf); - return NULL; - } - nbFiles = (size_t)ret_nbFiles; - } - - { const char** filenamesTable = (const char**) malloc(nbFiles * sizeof(*filenamesTable)); - CONTROL(filenamesTable != NULL); - - { size_t fnb; - for (fnb = 0, pos = 0; fnb < nbFiles; fnb++) { - filenamesTable[fnb] = buf+pos; - pos += strlen(buf+pos)+1; /* +1 for the finishing `\0` */ - } } - assert(pos <= bufSize); - - return UTIL_assembleFileNamesTable(filenamesTable, nbFiles, buf); - } -} - -static FileNamesTable* -UTIL_assembleFileNamesTable2(const char** filenames, size_t tableSize, size_t tableCapacity, char* buf) -{ - FileNamesTable* const table = (FileNamesTable*) malloc(sizeof(*table)); - CONTROL(table != NULL); - table->fileNames = filenames; - table->buf = buf; - table->tableSize = tableSize; - table->tableCapacity = tableCapacity; - return table; -} - -FileNamesTable* -UTIL_assembleFileNamesTable(const char** filenames, size_t tableSize, char* buf) -{ - return UTIL_assembleFileNamesTable2(filenames, tableSize, tableSize, buf); -} - -void UTIL_freeFileNamesTable(FileNamesTable* table) -{ - if (table==NULL) return; - free((void*)table->fileNames); - free(table->buf); - free(table); -} - -FileNamesTable* UTIL_allocateFileNamesTable(size_t tableSize) -{ - const char** const fnTable = (const char**)malloc(tableSize * sizeof(*fnTable)); - FileNamesTable* fnt; - if (fnTable==NULL) return NULL; - fnt = UTIL_assembleFileNamesTable(fnTable, tableSize, NULL); - fnt->tableSize = 0; /* the table is empty */ - return fnt; -} - -void UTIL_refFilename(FileNamesTable* fnt, const char* filename) -{ - assert(fnt->tableSize < fnt->tableCapacity); - fnt->fileNames[fnt->tableSize] = filename; - fnt->tableSize++; -} - -static size_t getTotalTableSize(FileNamesTable* table) -{ - size_t fnb = 0, totalSize = 0; - for(fnb = 0 ; fnb < table->tableSize && table->fileNames[fnb] ; ++fnb) { - totalSize += strlen(table->fileNames[fnb]) + 1; /* +1 to add '\0' at the end of each fileName */ - } - return totalSize; -} - -FileNamesTable* -UTIL_mergeFileNamesTable(FileNamesTable* table1, FileNamesTable* table2) -{ - unsigned newTableIdx = 0; - size_t pos = 0; - size_t newTotalTableSize; - char* buf; - - FileNamesTable* const newTable = UTIL_assembleFileNamesTable(NULL, 0, NULL); - CONTROL( newTable != NULL ); - - newTotalTableSize = getTotalTableSize(table1) + getTotalTableSize(table2); - - buf = (char*) calloc(newTotalTableSize, sizeof(*buf)); - CONTROL ( buf != NULL ); - - newTable->buf = buf; - newTable->tableSize = table1->tableSize + table2->tableSize; - newTable->fileNames = (const char **) calloc(newTable->tableSize, sizeof(*(newTable->fileNames))); - CONTROL ( newTable->fileNames != NULL ); - - { unsigned idx1; - for( idx1=0 ; (idx1 < table1->tableSize) && table1->fileNames[idx1] && (pos < newTotalTableSize); ++idx1, ++newTableIdx) { - size_t const curLen = strlen(table1->fileNames[idx1]); - memcpy(buf+pos, table1->fileNames[idx1], curLen); - assert(newTableIdx <= newTable->tableSize); - newTable->fileNames[newTableIdx] = buf+pos; - pos += curLen+1; - } } - - { unsigned idx2; - for( idx2=0 ; (idx2 < table2->tableSize) && table2->fileNames[idx2] && (pos < newTotalTableSize) ; ++idx2, ++newTableIdx) { - size_t const curLen = strlen(table2->fileNames[idx2]); - memcpy(buf+pos, table2->fileNames[idx2], curLen); - assert(newTableIdx <= newTable->tableSize); - newTable->fileNames[newTableIdx] = buf+pos; - pos += curLen+1; - } } - assert(pos <= newTotalTableSize); - newTable->tableSize = newTableIdx; - - UTIL_freeFileNamesTable(table1); - UTIL_freeFileNamesTable(table2); - - return newTable; -} - -#ifdef _WIN32 -static int UTIL_prepareFileList(const char* dirName, - char** bufStart, size_t* pos, - char** bufEnd, int followLinks) -{ - char* path; - size_t dirLength, pathLength; - int nbFiles = 0; - WIN32_FIND_DATAA cFile; - HANDLE hFile; - - dirLength = strlen(dirName); - path = (char*) malloc(dirLength + 3); - if (!path) return 0; - - memcpy(path, dirName, dirLength); - path[dirLength] = '\\'; - path[dirLength+1] = '*'; - path[dirLength+2] = 0; - - hFile=FindFirstFileA(path, &cFile); - if (hFile == INVALID_HANDLE_VALUE) { - UTIL_DISPLAYLEVEL(1, "Cannot open directory '%s'\n", dirName); - return 0; - } - free(path); - - do { - size_t const fnameLength = strlen(cFile.cFileName); - path = (char*) malloc(dirLength + fnameLength + 2); - if (!path) { FindClose(hFile); return 0; } - memcpy(path, dirName, dirLength); - path[dirLength] = '\\'; - memcpy(path+dirLength+1, cFile.cFileName, fnameLength); - pathLength = dirLength+1+fnameLength; - path[pathLength] = 0; - if (cFile.dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY) { - if ( strcmp (cFile.cFileName, "..") == 0 - || strcmp (cFile.cFileName, ".") == 0 ) - continue; - /* Recursively call "UTIL_prepareFileList" with the new path. */ - nbFiles += UTIL_prepareFileList(path, bufStart, pos, bufEnd, followLinks); - if (*bufStart == NULL) { free(path); FindClose(hFile); return 0; } - } else if ( (cFile.dwFileAttributes & FILE_ATTRIBUTE_NORMAL) - || (cFile.dwFileAttributes & FILE_ATTRIBUTE_ARCHIVE) - || (cFile.dwFileAttributes & FILE_ATTRIBUTE_COMPRESSED) ) { - if (*bufStart + *pos + pathLength >= *bufEnd) { - ptrdiff_t const newListSize = (*bufEnd - *bufStart) + LIST_SIZE_INCREASE; - *bufStart = (char*)UTIL_realloc(*bufStart, newListSize); - if (*bufStart == NULL) { free(path); FindClose(hFile); return 0; } - *bufEnd = *bufStart + newListSize; - } - if (*bufStart + *pos + pathLength < *bufEnd) { - memcpy(*bufStart + *pos, path, pathLength+1 /* include final \0 */); - *pos += pathLength + 1; - nbFiles++; - } } - free(path); - } while (FindNextFileA(hFile, &cFile)); - - FindClose(hFile); - return nbFiles; -} - -#elif defined(__linux__) || (PLATFORM_POSIX_VERSION >= 200112L) /* opendir, readdir require POSIX.1-2001 */ - -static int UTIL_prepareFileList(const char *dirName, - char** bufStart, size_t* pos, - char** bufEnd, int followLinks) -{ - DIR* dir; - struct dirent * entry; - size_t dirLength; - int nbFiles = 0; - - if (!(dir = opendir(dirName))) { - UTIL_DISPLAYLEVEL(1, "Cannot open directory '%s': %s\n", dirName, strerror(errno)); - return 0; - } - - dirLength = strlen(dirName); - errno = 0; - while ((entry = readdir(dir)) != NULL) { - char* path; - size_t fnameLength, pathLength; - if (strcmp (entry->d_name, "..") == 0 || - strcmp (entry->d_name, ".") == 0) continue; - fnameLength = strlen(entry->d_name); - path = (char*) malloc(dirLength + fnameLength + 2); - if (!path) { closedir(dir); return 0; } - memcpy(path, dirName, dirLength); - - path[dirLength] = '/'; - memcpy(path+dirLength+1, entry->d_name, fnameLength); - pathLength = dirLength+1+fnameLength; - path[pathLength] = 0; - - if (!followLinks && UTIL_isLink(path)) { - UTIL_DISPLAYLEVEL(2, "Warning : %s is a symbolic link, ignoring\n", path); - free(path); - continue; - } - - if (UTIL_isDirectory(path)) { - nbFiles += UTIL_prepareFileList(path, bufStart, pos, bufEnd, followLinks); /* Recursively call "UTIL_prepareFileList" with the new path. */ - if (*bufStart == NULL) { free(path); closedir(dir); return 0; } - } else { - if (*bufStart + *pos + pathLength >= *bufEnd) { - ptrdiff_t newListSize = (*bufEnd - *bufStart) + LIST_SIZE_INCREASE; - assert(newListSize >= 0); - *bufStart = (char*)UTIL_realloc(*bufStart, (size_t)newListSize); - *bufEnd = *bufStart + newListSize; - if (*bufStart == NULL) { free(path); closedir(dir); return 0; } - } - if (*bufStart + *pos + pathLength < *bufEnd) { - memcpy(*bufStart + *pos, path, pathLength + 1); /* with final \0 */ - *pos += pathLength + 1; - nbFiles++; - } } - free(path); - errno = 0; /* clear errno after UTIL_isDirectory, UTIL_prepareFileList */ - } - - if (errno != 0) { - UTIL_DISPLAYLEVEL(1, "readdir(%s) error: %s \n", dirName, strerror(errno)); - free(*bufStart); - *bufStart = NULL; - } - closedir(dir); - return nbFiles; -} - -#else - -static int UTIL_prepareFileList(const char *dirName, - char** bufStart, size_t* pos, - char** bufEnd, int followLinks) -{ - (void)bufStart; (void)bufEnd; (void)pos; (void)followLinks; - UTIL_DISPLAYLEVEL(1, "Directory %s ignored (compiled without _WIN32 or _POSIX_C_SOURCE) \n", dirName); - return 0; -} - -#endif /* #ifdef _WIN32 */ - -int UTIL_isCompressedFile(const char *inputName, const char *extensionList[]) -{ - const char* ext = UTIL_getFileExtension(inputName); - while(*extensionList!=NULL) - { - const int isCompressedExtension = strcmp(ext,*extensionList); - if(isCompressedExtension==0) - return 1; - ++extensionList; - } - return 0; -} - -/*Utility function to get file extension from file */ -const char* UTIL_getFileExtension(const char* infilename) -{ - const char* extension = strrchr(infilename, '.'); - if(!extension || extension==infilename) return ""; - return extension; -} - -static int pathnameHas2Dots(const char *pathname) -{ - /* We need to figure out whether any ".." present in the path is a whole - * path token, which is the case if it is bordered on both sides by either - * the beginning/end of the path or by a directory separator. - */ - const char *needle = pathname; - while (1) { - needle = strstr(needle, ".."); - - if (needle == NULL) { - return 0; - } - - if ((needle == pathname || needle[-1] == PATH_SEP) - && (needle[2] == '\0' || needle[2] == PATH_SEP)) { - return 1; - } - - /* increment so we search for the next match */ - needle++; - }; - return 0; -} - -static int isFileNameValidForMirroredOutput(const char *filename) -{ - return !pathnameHas2Dots(filename); -} - - -#define DIR_DEFAULT_MODE 0755 -static mode_t getDirMode(const char *dirName) -{ - stat_t st; - if (!UTIL_stat(dirName, &st)) { - UTIL_DISPLAY("zstd: failed to get DIR stats %s: %s\n", dirName, strerror(errno)); - return DIR_DEFAULT_MODE; - } - if (!UTIL_isDirectoryStat(&st)) { - UTIL_DISPLAY("zstd: expected directory: %s\n", dirName); - return DIR_DEFAULT_MODE; - } - return st.st_mode; -} - -static int makeDir(const char *dir, mode_t mode) -{ -#if defined(_MSC_VER) || defined(__MINGW32__) || defined (__MSVCRT__) - int ret = _mkdir(dir); - (void) mode; -#else - int ret = mkdir(dir, mode); -#endif - if (ret != 0) { - if (errno == EEXIST) - return 0; - UTIL_DISPLAY("zstd: failed to create DIR %s: %s\n", dir, strerror(errno)); - } - return ret; -} - -/* this function requires a mutable input string */ -static void convertPathnameToDirName(char *pathname) -{ - size_t len = 0; - char* pos = NULL; - /* get dir name from pathname similar to 'dirname()' */ - assert(pathname != NULL); - - /* remove trailing '/' chars */ - len = strlen(pathname); - assert(len > 0); - while (pathname[len] == PATH_SEP) { - pathname[len] = '\0'; - len--; - } - if (len == 0) return; - - /* if input is a single file, return '.' instead. i.e. - * "xyz/abc/file.txt" => "xyz/abc" - "./file.txt" => "." - "file.txt" => "." - */ - pos = strrchr(pathname, PATH_SEP); - if (pos == NULL) { - pathname[0] = '.'; - pathname[1] = '\0'; - } else { - *pos = '\0'; - } -} - -/* pathname must be valid */ -static const char* trimLeadingRootChar(const char *pathname) -{ - assert(pathname != NULL); - if (pathname[0] == PATH_SEP) - return pathname + 1; - return pathname; -} - -/* pathname must be valid */ -static const char* trimLeadingCurrentDirConst(const char *pathname) -{ - assert(pathname != NULL); - if ((pathname[0] == '.') && (pathname[1] == PATH_SEP)) - return pathname + 2; - return pathname; -} - -static char* -trimLeadingCurrentDir(char *pathname) -{ - /* 'union charunion' can do const-cast without compiler warning */ - union charunion { - char *chr; - const char* cchr; - } ptr; - ptr.cchr = trimLeadingCurrentDirConst(pathname); - return ptr.chr; -} - -/* remove leading './' or '/' chars here */ -static const char * trimPath(const char *pathname) -{ - return trimLeadingRootChar( - trimLeadingCurrentDirConst(pathname)); -} - -static char* mallocAndJoin2Dir(const char *dir1, const char *dir2) -{ - const size_t dir1Size = strlen(dir1); - const size_t dir2Size = strlen(dir2); - char *outDirBuffer, *buffer, trailingChar; - - assert(dir1 != NULL && dir2 != NULL); - outDirBuffer = (char *) malloc(dir1Size + dir2Size + 2); - CONTROL(outDirBuffer != NULL); - - memcpy(outDirBuffer, dir1, dir1Size); - outDirBuffer[dir1Size] = '\0'; - - if (dir2[0] == '.') - return outDirBuffer; - - buffer = outDirBuffer + dir1Size; - trailingChar = *(buffer - 1); - if (trailingChar != PATH_SEP) { - *buffer = PATH_SEP; - buffer++; - } - memcpy(buffer, dir2, dir2Size); - buffer[dir2Size] = '\0'; - - return outDirBuffer; -} - -/* this function will return NULL if input srcFileName is not valid name for mirrored output path */ -char* UTIL_createMirroredDestDirName(const char* srcFileName, const char* outDirRootName) -{ - char* pathname = NULL; - if (!isFileNameValidForMirroredOutput(srcFileName)) - return NULL; - - pathname = mallocAndJoin2Dir(outDirRootName, trimPath(srcFileName)); - - convertPathnameToDirName(pathname); - return pathname; -} - -static int -mirrorSrcDir(char* srcDirName, const char* outDirName) -{ - mode_t srcMode; - int status = 0; - char* newDir = mallocAndJoin2Dir(outDirName, trimPath(srcDirName)); - if (!newDir) - return -ENOMEM; - - srcMode = getDirMode(srcDirName); - status = makeDir(newDir, srcMode); - free(newDir); - return status; -} - -static int -mirrorSrcDirRecursive(char* srcDirName, const char* outDirName) -{ - int status = 0; - char* pp = trimLeadingCurrentDir(srcDirName); - char* sp = NULL; - - while ((sp = strchr(pp, PATH_SEP)) != NULL) { - if (sp != pp) { - *sp = '\0'; - status = mirrorSrcDir(srcDirName, outDirName); - if (status != 0) - return status; - *sp = PATH_SEP; - } - pp = sp + 1; - } - status = mirrorSrcDir(srcDirName, outDirName); - return status; -} - -static void -makeMirroredDestDirsWithSameSrcDirMode(char** srcDirNames, unsigned nbFile, const char* outDirName) -{ - unsigned int i = 0; - for (i = 0; i < nbFile; i++) - mirrorSrcDirRecursive(srcDirNames[i], outDirName); -} - -static int -firstIsParentOrSameDirOfSecond(const char* firstDir, const char* secondDir) -{ - size_t firstDirLen = strlen(firstDir), - secondDirLen = strlen(secondDir); - return firstDirLen <= secondDirLen && - (secondDir[firstDirLen] == PATH_SEP || secondDir[firstDirLen] == '\0') && - 0 == strncmp(firstDir, secondDir, firstDirLen); -} - -static int compareDir(const void* pathname1, const void* pathname2) { - /* sort it after remove the leading '/' or './'*/ - const char* s1 = trimPath(*(char * const *) pathname1); - const char* s2 = trimPath(*(char * const *) pathname2); - return strcmp(s1, s2); -} - -static void -makeUniqueMirroredDestDirs(char** srcDirNames, unsigned nbFile, const char* outDirName) -{ - unsigned int i = 0, uniqueDirNr = 0; - char** uniqueDirNames = NULL; - - if (nbFile == 0) - return; - - uniqueDirNames = (char** ) malloc(nbFile * sizeof (char *)); - CONTROL(uniqueDirNames != NULL); - - /* if dirs is "a/b/c" and "a/b/c/d", we only need call: - * we just need "a/b/c/d" */ - qsort((void *)srcDirNames, nbFile, sizeof(char*), compareDir); - - uniqueDirNr = 1; - uniqueDirNames[uniqueDirNr - 1] = srcDirNames[0]; - for (i = 1; i < nbFile; i++) { - char* prevDirName = srcDirNames[i - 1]; - char* currDirName = srcDirNames[i]; - +U64 UTIL_getTotalFileSize(const char* const * fileNamesTable, unsigned nbFiles) +{ + U64 total = 0; + unsigned n; + for (n=0; n<nbFiles; n++) { + U64 const size = UTIL_getFileSize(fileNamesTable[n]); + if (size == UTIL_FILESIZE_UNKNOWN) return UTIL_FILESIZE_UNKNOWN; + total += size; + } + return total; +} + + +/* condition : @file must be valid, and not have reached its end. + * @return : length of line written into @buf, ended with `\0` instead of '\n', + * or 0, if there is no new line */ +static size_t readLineFromFile(char* buf, size_t len, FILE* file) +{ + assert(!feof(file)); + if ( fgets(buf, (int) len, file) == NULL ) return 0; + { size_t linelen = strlen(buf); + if (strlen(buf)==0) return 0; + if (buf[linelen-1] == '\n') linelen--; + buf[linelen] = '\0'; + return linelen+1; + } +} + +/* Conditions : + * size of @inputFileName file must be < @dstCapacity + * @dst must be initialized + * @return : nb of lines + * or -1 if there's an error + */ +static int +readLinesFromFile(void* dst, size_t dstCapacity, + const char* inputFileName) +{ + int nbFiles = 0; + size_t pos = 0; + char* const buf = (char*)dst; + FILE* const inputFile = fopen(inputFileName, "r"); + + assert(dst != NULL); + + if(!inputFile) { + if (g_utilDisplayLevel >= 1) perror("zstd:util:readLinesFromFile"); + return -1; + } + + while ( !feof(inputFile) ) { + size_t const lineLength = readLineFromFile(buf+pos, dstCapacity-pos, inputFile); + if (lineLength == 0) break; + assert(pos + lineLength < dstCapacity); + pos += lineLength; + ++nbFiles; + } + + CONTROL( fclose(inputFile) == 0 ); + + return nbFiles; +} + +/*Note: buf is not freed in case function successfully created table because filesTable->fileNames[0] = buf*/ +FileNamesTable* +UTIL_createFileNamesTable_fromFileName(const char* inputFileName) +{ + size_t nbFiles = 0; + char* buf; + size_t bufSize; + size_t pos = 0; + stat_t statbuf; + + if (!UTIL_stat(inputFileName, &statbuf) || !UTIL_isRegularFileStat(&statbuf)) + return NULL; + + { U64 const inputFileSize = UTIL_getFileSizeStat(&statbuf); + if(inputFileSize > MAX_FILE_OF_FILE_NAMES_SIZE) + return NULL; + bufSize = (size_t)(inputFileSize + 1); /* (+1) to add '\0' at the end of last filename */ + } + + buf = (char*) malloc(bufSize); + CONTROL( buf != NULL ); + + { int const ret_nbFiles = readLinesFromFile(buf, bufSize, inputFileName); + + if (ret_nbFiles <= 0) { + free(buf); + return NULL; + } + nbFiles = (size_t)ret_nbFiles; + } + + { const char** filenamesTable = (const char**) malloc(nbFiles * sizeof(*filenamesTable)); + CONTROL(filenamesTable != NULL); + + { size_t fnb; + for (fnb = 0, pos = 0; fnb < nbFiles; fnb++) { + filenamesTable[fnb] = buf+pos; + pos += strlen(buf+pos)+1; /* +1 for the finishing `\0` */ + } } + assert(pos <= bufSize); + + return UTIL_assembleFileNamesTable(filenamesTable, nbFiles, buf); + } +} + +static FileNamesTable* +UTIL_assembleFileNamesTable2(const char** filenames, size_t tableSize, size_t tableCapacity, char* buf) +{ + FileNamesTable* const table = (FileNamesTable*) malloc(sizeof(*table)); + CONTROL(table != NULL); + table->fileNames = filenames; + table->buf = buf; + table->tableSize = tableSize; + table->tableCapacity = tableCapacity; + return table; +} + +FileNamesTable* +UTIL_assembleFileNamesTable(const char** filenames, size_t tableSize, char* buf) +{ + return UTIL_assembleFileNamesTable2(filenames, tableSize, tableSize, buf); +} + +void UTIL_freeFileNamesTable(FileNamesTable* table) +{ + if (table==NULL) return; + free((void*)table->fileNames); + free(table->buf); + free(table); +} + +FileNamesTable* UTIL_allocateFileNamesTable(size_t tableSize) +{ + const char** const fnTable = (const char**)malloc(tableSize * sizeof(*fnTable)); + FileNamesTable* fnt; + if (fnTable==NULL) return NULL; + fnt = UTIL_assembleFileNamesTable(fnTable, tableSize, NULL); + fnt->tableSize = 0; /* the table is empty */ + return fnt; +} + +void UTIL_refFilename(FileNamesTable* fnt, const char* filename) +{ + assert(fnt->tableSize < fnt->tableCapacity); + fnt->fileNames[fnt->tableSize] = filename; + fnt->tableSize++; +} + +static size_t getTotalTableSize(FileNamesTable* table) +{ + size_t fnb = 0, totalSize = 0; + for(fnb = 0 ; fnb < table->tableSize && table->fileNames[fnb] ; ++fnb) { + totalSize += strlen(table->fileNames[fnb]) + 1; /* +1 to add '\0' at the end of each fileName */ + } + return totalSize; +} + +FileNamesTable* +UTIL_mergeFileNamesTable(FileNamesTable* table1, FileNamesTable* table2) +{ + unsigned newTableIdx = 0; + size_t pos = 0; + size_t newTotalTableSize; + char* buf; + + FileNamesTable* const newTable = UTIL_assembleFileNamesTable(NULL, 0, NULL); + CONTROL( newTable != NULL ); + + newTotalTableSize = getTotalTableSize(table1) + getTotalTableSize(table2); + + buf = (char*) calloc(newTotalTableSize, sizeof(*buf)); + CONTROL ( buf != NULL ); + + newTable->buf = buf; + newTable->tableSize = table1->tableSize + table2->tableSize; + newTable->fileNames = (const char **) calloc(newTable->tableSize, sizeof(*(newTable->fileNames))); + CONTROL ( newTable->fileNames != NULL ); + + { unsigned idx1; + for( idx1=0 ; (idx1 < table1->tableSize) && table1->fileNames[idx1] && (pos < newTotalTableSize); ++idx1, ++newTableIdx) { + size_t const curLen = strlen(table1->fileNames[idx1]); + memcpy(buf+pos, table1->fileNames[idx1], curLen); + assert(newTableIdx <= newTable->tableSize); + newTable->fileNames[newTableIdx] = buf+pos; + pos += curLen+1; + } } + + { unsigned idx2; + for( idx2=0 ; (idx2 < table2->tableSize) && table2->fileNames[idx2] && (pos < newTotalTableSize) ; ++idx2, ++newTableIdx) { + size_t const curLen = strlen(table2->fileNames[idx2]); + memcpy(buf+pos, table2->fileNames[idx2], curLen); + assert(newTableIdx <= newTable->tableSize); + newTable->fileNames[newTableIdx] = buf+pos; + pos += curLen+1; + } } + assert(pos <= newTotalTableSize); + newTable->tableSize = newTableIdx; + + UTIL_freeFileNamesTable(table1); + UTIL_freeFileNamesTable(table2); + + return newTable; +} + +#ifdef _WIN32 +static int UTIL_prepareFileList(const char* dirName, + char** bufStart, size_t* pos, + char** bufEnd, int followLinks) +{ + char* path; + size_t dirLength, pathLength; + int nbFiles = 0; + WIN32_FIND_DATAA cFile; + HANDLE hFile; + + dirLength = strlen(dirName); + path = (char*) malloc(dirLength + 3); + if (!path) return 0; + + memcpy(path, dirName, dirLength); + path[dirLength] = '\\'; + path[dirLength+1] = '*'; + path[dirLength+2] = 0; + + hFile=FindFirstFileA(path, &cFile); + if (hFile == INVALID_HANDLE_VALUE) { + UTIL_DISPLAYLEVEL(1, "Cannot open directory '%s'\n", dirName); + return 0; + } + free(path); + + do { + size_t const fnameLength = strlen(cFile.cFileName); + path = (char*) malloc(dirLength + fnameLength + 2); + if (!path) { FindClose(hFile); return 0; } + memcpy(path, dirName, dirLength); + path[dirLength] = '\\'; + memcpy(path+dirLength+1, cFile.cFileName, fnameLength); + pathLength = dirLength+1+fnameLength; + path[pathLength] = 0; + if (cFile.dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY) { + if ( strcmp (cFile.cFileName, "..") == 0 + || strcmp (cFile.cFileName, ".") == 0 ) + continue; + /* Recursively call "UTIL_prepareFileList" with the new path. */ + nbFiles += UTIL_prepareFileList(path, bufStart, pos, bufEnd, followLinks); + if (*bufStart == NULL) { free(path); FindClose(hFile); return 0; } + } else if ( (cFile.dwFileAttributes & FILE_ATTRIBUTE_NORMAL) + || (cFile.dwFileAttributes & FILE_ATTRIBUTE_ARCHIVE) + || (cFile.dwFileAttributes & FILE_ATTRIBUTE_COMPRESSED) ) { + if (*bufStart + *pos + pathLength >= *bufEnd) { + ptrdiff_t const newListSize = (*bufEnd - *bufStart) + LIST_SIZE_INCREASE; + *bufStart = (char*)UTIL_realloc(*bufStart, newListSize); + if (*bufStart == NULL) { free(path); FindClose(hFile); return 0; } + *bufEnd = *bufStart + newListSize; + } + if (*bufStart + *pos + pathLength < *bufEnd) { + memcpy(*bufStart + *pos, path, pathLength+1 /* include final \0 */); + *pos += pathLength + 1; + nbFiles++; + } } + free(path); + } while (FindNextFileA(hFile, &cFile)); + + FindClose(hFile); + return nbFiles; +} + +#elif defined(__linux__) || (PLATFORM_POSIX_VERSION >= 200112L) /* opendir, readdir require POSIX.1-2001 */ + +static int UTIL_prepareFileList(const char *dirName, + char** bufStart, size_t* pos, + char** bufEnd, int followLinks) +{ + DIR* dir; + struct dirent * entry; + size_t dirLength; + int nbFiles = 0; + + if (!(dir = opendir(dirName))) { + UTIL_DISPLAYLEVEL(1, "Cannot open directory '%s': %s\n", dirName, strerror(errno)); + return 0; + } + + dirLength = strlen(dirName); + errno = 0; + while ((entry = readdir(dir)) != NULL) { + char* path; + size_t fnameLength, pathLength; + if (strcmp (entry->d_name, "..") == 0 || + strcmp (entry->d_name, ".") == 0) continue; + fnameLength = strlen(entry->d_name); + path = (char*) malloc(dirLength + fnameLength + 2); + if (!path) { closedir(dir); return 0; } + memcpy(path, dirName, dirLength); + + path[dirLength] = '/'; + memcpy(path+dirLength+1, entry->d_name, fnameLength); + pathLength = dirLength+1+fnameLength; + path[pathLength] = 0; + + if (!followLinks && UTIL_isLink(path)) { + UTIL_DISPLAYLEVEL(2, "Warning : %s is a symbolic link, ignoring\n", path); + free(path); + continue; + } + + if (UTIL_isDirectory(path)) { + nbFiles += UTIL_prepareFileList(path, bufStart, pos, bufEnd, followLinks); /* Recursively call "UTIL_prepareFileList" with the new path. */ + if (*bufStart == NULL) { free(path); closedir(dir); return 0; } + } else { + if (*bufStart + *pos + pathLength >= *bufEnd) { + ptrdiff_t newListSize = (*bufEnd - *bufStart) + LIST_SIZE_INCREASE; + assert(newListSize >= 0); + *bufStart = (char*)UTIL_realloc(*bufStart, (size_t)newListSize); + *bufEnd = *bufStart + newListSize; + if (*bufStart == NULL) { free(path); closedir(dir); return 0; } + } + if (*bufStart + *pos + pathLength < *bufEnd) { + memcpy(*bufStart + *pos, path, pathLength + 1); /* with final \0 */ + *pos += pathLength + 1; + nbFiles++; + } } + free(path); + errno = 0; /* clear errno after UTIL_isDirectory, UTIL_prepareFileList */ + } + + if (errno != 0) { + UTIL_DISPLAYLEVEL(1, "readdir(%s) error: %s \n", dirName, strerror(errno)); + free(*bufStart); + *bufStart = NULL; + } + closedir(dir); + return nbFiles; +} + +#else + +static int UTIL_prepareFileList(const char *dirName, + char** bufStart, size_t* pos, + char** bufEnd, int followLinks) +{ + (void)bufStart; (void)bufEnd; (void)pos; (void)followLinks; + UTIL_DISPLAYLEVEL(1, "Directory %s ignored (compiled without _WIN32 or _POSIX_C_SOURCE) \n", dirName); + return 0; +} + +#endif /* #ifdef _WIN32 */ + +int UTIL_isCompressedFile(const char *inputName, const char *extensionList[]) +{ + const char* ext = UTIL_getFileExtension(inputName); + while(*extensionList!=NULL) + { + const int isCompressedExtension = strcmp(ext,*extensionList); + if(isCompressedExtension==0) + return 1; + ++extensionList; + } + return 0; +} + +/*Utility function to get file extension from file */ +const char* UTIL_getFileExtension(const char* infilename) +{ + const char* extension = strrchr(infilename, '.'); + if(!extension || extension==infilename) return ""; + return extension; +} + +static int pathnameHas2Dots(const char *pathname) +{ + /* We need to figure out whether any ".." present in the path is a whole + * path token, which is the case if it is bordered on both sides by either + * the beginning/end of the path or by a directory separator. + */ + const char *needle = pathname; + while (1) { + needle = strstr(needle, ".."); + + if (needle == NULL) { + return 0; + } + + if ((needle == pathname || needle[-1] == PATH_SEP) + && (needle[2] == '\0' || needle[2] == PATH_SEP)) { + return 1; + } + + /* increment so we search for the next match */ + needle++; + }; + return 0; +} + +static int isFileNameValidForMirroredOutput(const char *filename) +{ + return !pathnameHas2Dots(filename); +} + + +#define DIR_DEFAULT_MODE 0755 +static mode_t getDirMode(const char *dirName) +{ + stat_t st; + if (!UTIL_stat(dirName, &st)) { + UTIL_DISPLAY("zstd: failed to get DIR stats %s: %s\n", dirName, strerror(errno)); + return DIR_DEFAULT_MODE; + } + if (!UTIL_isDirectoryStat(&st)) { + UTIL_DISPLAY("zstd: expected directory: %s\n", dirName); + return DIR_DEFAULT_MODE; + } + return st.st_mode; +} + +static int makeDir(const char *dir, mode_t mode) +{ +#if defined(_MSC_VER) || defined(__MINGW32__) || defined (__MSVCRT__) + int ret = _mkdir(dir); + (void) mode; +#else + int ret = mkdir(dir, mode); +#endif + if (ret != 0) { + if (errno == EEXIST) + return 0; + UTIL_DISPLAY("zstd: failed to create DIR %s: %s\n", dir, strerror(errno)); + } + return ret; +} + +/* this function requires a mutable input string */ +static void convertPathnameToDirName(char *pathname) +{ + size_t len = 0; + char* pos = NULL; + /* get dir name from pathname similar to 'dirname()' */ + assert(pathname != NULL); + + /* remove trailing '/' chars */ + len = strlen(pathname); + assert(len > 0); + while (pathname[len] == PATH_SEP) { + pathname[len] = '\0'; + len--; + } + if (len == 0) return; + + /* if input is a single file, return '.' instead. i.e. + * "xyz/abc/file.txt" => "xyz/abc" + "./file.txt" => "." + "file.txt" => "." + */ + pos = strrchr(pathname, PATH_SEP); + if (pos == NULL) { + pathname[0] = '.'; + pathname[1] = '\0'; + } else { + *pos = '\0'; + } +} + +/* pathname must be valid */ +static const char* trimLeadingRootChar(const char *pathname) +{ + assert(pathname != NULL); + if (pathname[0] == PATH_SEP) + return pathname + 1; + return pathname; +} + +/* pathname must be valid */ +static const char* trimLeadingCurrentDirConst(const char *pathname) +{ + assert(pathname != NULL); + if ((pathname[0] == '.') && (pathname[1] == PATH_SEP)) + return pathname + 2; + return pathname; +} + +static char* +trimLeadingCurrentDir(char *pathname) +{ + /* 'union charunion' can do const-cast without compiler warning */ + union charunion { + char *chr; + const char* cchr; + } ptr; + ptr.cchr = trimLeadingCurrentDirConst(pathname); + return ptr.chr; +} + +/* remove leading './' or '/' chars here */ +static const char * trimPath(const char *pathname) +{ + return trimLeadingRootChar( + trimLeadingCurrentDirConst(pathname)); +} + +static char* mallocAndJoin2Dir(const char *dir1, const char *dir2) +{ + const size_t dir1Size = strlen(dir1); + const size_t dir2Size = strlen(dir2); + char *outDirBuffer, *buffer, trailingChar; + + assert(dir1 != NULL && dir2 != NULL); + outDirBuffer = (char *) malloc(dir1Size + dir2Size + 2); + CONTROL(outDirBuffer != NULL); + + memcpy(outDirBuffer, dir1, dir1Size); + outDirBuffer[dir1Size] = '\0'; + + if (dir2[0] == '.') + return outDirBuffer; + + buffer = outDirBuffer + dir1Size; + trailingChar = *(buffer - 1); + if (trailingChar != PATH_SEP) { + *buffer = PATH_SEP; + buffer++; + } + memcpy(buffer, dir2, dir2Size); + buffer[dir2Size] = '\0'; + + return outDirBuffer; +} + +/* this function will return NULL if input srcFileName is not valid name for mirrored output path */ +char* UTIL_createMirroredDestDirName(const char* srcFileName, const char* outDirRootName) +{ + char* pathname = NULL; + if (!isFileNameValidForMirroredOutput(srcFileName)) + return NULL; + + pathname = mallocAndJoin2Dir(outDirRootName, trimPath(srcFileName)); + + convertPathnameToDirName(pathname); + return pathname; +} + +static int +mirrorSrcDir(char* srcDirName, const char* outDirName) +{ + mode_t srcMode; + int status = 0; + char* newDir = mallocAndJoin2Dir(outDirName, trimPath(srcDirName)); + if (!newDir) + return -ENOMEM; + + srcMode = getDirMode(srcDirName); + status = makeDir(newDir, srcMode); + free(newDir); + return status; +} + +static int +mirrorSrcDirRecursive(char* srcDirName, const char* outDirName) +{ + int status = 0; + char* pp = trimLeadingCurrentDir(srcDirName); + char* sp = NULL; + + while ((sp = strchr(pp, PATH_SEP)) != NULL) { + if (sp != pp) { + *sp = '\0'; + status = mirrorSrcDir(srcDirName, outDirName); + if (status != 0) + return status; + *sp = PATH_SEP; + } + pp = sp + 1; + } + status = mirrorSrcDir(srcDirName, outDirName); + return status; +} + +static void +makeMirroredDestDirsWithSameSrcDirMode(char** srcDirNames, unsigned nbFile, const char* outDirName) +{ + unsigned int i = 0; + for (i = 0; i < nbFile; i++) + mirrorSrcDirRecursive(srcDirNames[i], outDirName); +} + +static int +firstIsParentOrSameDirOfSecond(const char* firstDir, const char* secondDir) +{ + size_t firstDirLen = strlen(firstDir), + secondDirLen = strlen(secondDir); + return firstDirLen <= secondDirLen && + (secondDir[firstDirLen] == PATH_SEP || secondDir[firstDirLen] == '\0') && + 0 == strncmp(firstDir, secondDir, firstDirLen); +} + +static int compareDir(const void* pathname1, const void* pathname2) { + /* sort it after remove the leading '/' or './'*/ + const char* s1 = trimPath(*(char * const *) pathname1); + const char* s2 = trimPath(*(char * const *) pathname2); + return strcmp(s1, s2); +} + +static void +makeUniqueMirroredDestDirs(char** srcDirNames, unsigned nbFile, const char* outDirName) +{ + unsigned int i = 0, uniqueDirNr = 0; + char** uniqueDirNames = NULL; + + if (nbFile == 0) + return; + + uniqueDirNames = (char** ) malloc(nbFile * sizeof (char *)); + CONTROL(uniqueDirNames != NULL); + + /* if dirs is "a/b/c" and "a/b/c/d", we only need call: + * we just need "a/b/c/d" */ + qsort((void *)srcDirNames, nbFile, sizeof(char*), compareDir); + + uniqueDirNr = 1; + uniqueDirNames[uniqueDirNr - 1] = srcDirNames[0]; + for (i = 1; i < nbFile; i++) { + char* prevDirName = srcDirNames[i - 1]; + char* currDirName = srcDirNames[i]; + /* note: we always compare trimmed path, i.e.: - * src dir of "./foo" and "/foo" will be both saved into: - * "outDirName/foo/" */ - if (!firstIsParentOrSameDirOfSecond(trimPath(prevDirName), - trimPath(currDirName))) - uniqueDirNr++; - - /* we need maintain original src dir name instead of trimmed + * src dir of "./foo" and "/foo" will be both saved into: + * "outDirName/foo/" */ + if (!firstIsParentOrSameDirOfSecond(trimPath(prevDirName), + trimPath(currDirName))) + uniqueDirNr++; + + /* we need maintain original src dir name instead of trimmed * dir, so we can retrieve the original src dir's mode_t */ - uniqueDirNames[uniqueDirNr - 1] = currDirName; - } - - makeMirroredDestDirsWithSameSrcDirMode(uniqueDirNames, uniqueDirNr, outDirName); - - free(uniqueDirNames); -} - -static void -makeMirroredDestDirs(char** srcFileNames, unsigned nbFile, const char* outDirName) -{ - unsigned int i = 0; - for (i = 0; i < nbFile; ++i) - convertPathnameToDirName(srcFileNames[i]); - makeUniqueMirroredDestDirs(srcFileNames, nbFile, outDirName); -} - -void UTIL_mirrorSourceFilesDirectories(const char** inFileNames, unsigned int nbFile, const char* outDirName) -{ - unsigned int i = 0, validFilenamesNr = 0; - char** srcFileNames = (char **) malloc(nbFile * sizeof (char *)); - CONTROL(srcFileNames != NULL); - - /* check input filenames is valid */ - for (i = 0; i < nbFile; ++i) { - if (isFileNameValidForMirroredOutput(inFileNames[i])) { - char* fname = STRDUP(inFileNames[i]); - CONTROL(fname != NULL); - srcFileNames[validFilenamesNr++] = fname; - } - } - - if (validFilenamesNr > 0) { - makeDir(outDirName, DIR_DEFAULT_MODE); - makeMirroredDestDirs(srcFileNames, validFilenamesNr, outDirName); - } - - for (i = 0; i < validFilenamesNr; i++) - free(srcFileNames[i]); - free(srcFileNames); -} - -FileNamesTable* -UTIL_createExpandedFNT(const char* const* inputNames, size_t nbIfns, int followLinks) -{ - unsigned nbFiles; - char* buf = (char*)malloc(LIST_SIZE_INCREASE); - char* bufend = buf + LIST_SIZE_INCREASE; - - if (!buf) return NULL; - - { size_t ifnNb, pos; - for (ifnNb=0, pos=0, nbFiles=0; ifnNb<nbIfns; ifnNb++) { - if (!UTIL_isDirectory(inputNames[ifnNb])) { - size_t const len = strlen(inputNames[ifnNb]); - if (buf + pos + len >= bufend) { - ptrdiff_t newListSize = (bufend - buf) + LIST_SIZE_INCREASE; - assert(newListSize >= 0); - buf = (char*)UTIL_realloc(buf, (size_t)newListSize); - if (!buf) return NULL; - bufend = buf + newListSize; - } - if (buf + pos + len < bufend) { - memcpy(buf+pos, inputNames[ifnNb], len+1); /* including final \0 */ - pos += len + 1; - nbFiles++; - } - } else { - nbFiles += (unsigned)UTIL_prepareFileList(inputNames[ifnNb], &buf, &pos, &bufend, followLinks); - if (buf == NULL) return NULL; - } } } - - /* note : even if nbFiles==0, function returns a valid, though empty, FileNamesTable* object */ - - { size_t ifnNb, pos; - size_t const fntCapacity = nbFiles + 1; /* minimum 1, allows adding one reference, typically stdin */ - const char** const fileNamesTable = (const char**)malloc(fntCapacity * sizeof(*fileNamesTable)); - if (!fileNamesTable) { free(buf); return NULL; } - - for (ifnNb = 0, pos = 0; ifnNb < nbFiles; ifnNb++) { - fileNamesTable[ifnNb] = buf + pos; - if (buf + pos > bufend) { free(buf); free((void*)fileNamesTable); return NULL; } - pos += strlen(fileNamesTable[ifnNb]) + 1; - } - return UTIL_assembleFileNamesTable2(fileNamesTable, nbFiles, fntCapacity, buf); - } -} - - -void UTIL_expandFNT(FileNamesTable** fnt, int followLinks) -{ - FileNamesTable* const newFNT = UTIL_createExpandedFNT((*fnt)->fileNames, (*fnt)->tableSize, followLinks); - CONTROL(newFNT != NULL); - UTIL_freeFileNamesTable(*fnt); - *fnt = newFNT; -} - -FileNamesTable* UTIL_createFNT_fromROTable(const char** filenames, size_t nbFilenames) -{ - size_t const sizeof_FNTable = nbFilenames * sizeof(*filenames); - const char** const newFNTable = (const char**)malloc(sizeof_FNTable); - if (newFNTable==NULL) return NULL; - memcpy((void*)newFNTable, filenames, sizeof_FNTable); /* void* : mitigate a Visual compiler bug or limitation */ - return UTIL_assembleFileNamesTable(newFNTable, nbFilenames, NULL); -} - - -/*-**************************************** + uniqueDirNames[uniqueDirNr - 1] = currDirName; + } + + makeMirroredDestDirsWithSameSrcDirMode(uniqueDirNames, uniqueDirNr, outDirName); + + free(uniqueDirNames); +} + +static void +makeMirroredDestDirs(char** srcFileNames, unsigned nbFile, const char* outDirName) +{ + unsigned int i = 0; + for (i = 0; i < nbFile; ++i) + convertPathnameToDirName(srcFileNames[i]); + makeUniqueMirroredDestDirs(srcFileNames, nbFile, outDirName); +} + +void UTIL_mirrorSourceFilesDirectories(const char** inFileNames, unsigned int nbFile, const char* outDirName) +{ + unsigned int i = 0, validFilenamesNr = 0; + char** srcFileNames = (char **) malloc(nbFile * sizeof (char *)); + CONTROL(srcFileNames != NULL); + + /* check input filenames is valid */ + for (i = 0; i < nbFile; ++i) { + if (isFileNameValidForMirroredOutput(inFileNames[i])) { + char* fname = STRDUP(inFileNames[i]); + CONTROL(fname != NULL); + srcFileNames[validFilenamesNr++] = fname; + } + } + + if (validFilenamesNr > 0) { + makeDir(outDirName, DIR_DEFAULT_MODE); + makeMirroredDestDirs(srcFileNames, validFilenamesNr, outDirName); + } + + for (i = 0; i < validFilenamesNr; i++) + free(srcFileNames[i]); + free(srcFileNames); +} + +FileNamesTable* +UTIL_createExpandedFNT(const char* const* inputNames, size_t nbIfns, int followLinks) +{ + unsigned nbFiles; + char* buf = (char*)malloc(LIST_SIZE_INCREASE); + char* bufend = buf + LIST_SIZE_INCREASE; + + if (!buf) return NULL; + + { size_t ifnNb, pos; + for (ifnNb=0, pos=0, nbFiles=0; ifnNb<nbIfns; ifnNb++) { + if (!UTIL_isDirectory(inputNames[ifnNb])) { + size_t const len = strlen(inputNames[ifnNb]); + if (buf + pos + len >= bufend) { + ptrdiff_t newListSize = (bufend - buf) + LIST_SIZE_INCREASE; + assert(newListSize >= 0); + buf = (char*)UTIL_realloc(buf, (size_t)newListSize); + if (!buf) return NULL; + bufend = buf + newListSize; + } + if (buf + pos + len < bufend) { + memcpy(buf+pos, inputNames[ifnNb], len+1); /* including final \0 */ + pos += len + 1; + nbFiles++; + } + } else { + nbFiles += (unsigned)UTIL_prepareFileList(inputNames[ifnNb], &buf, &pos, &bufend, followLinks); + if (buf == NULL) return NULL; + } } } + + /* note : even if nbFiles==0, function returns a valid, though empty, FileNamesTable* object */ + + { size_t ifnNb, pos; + size_t const fntCapacity = nbFiles + 1; /* minimum 1, allows adding one reference, typically stdin */ + const char** const fileNamesTable = (const char**)malloc(fntCapacity * sizeof(*fileNamesTable)); + if (!fileNamesTable) { free(buf); return NULL; } + + for (ifnNb = 0, pos = 0; ifnNb < nbFiles; ifnNb++) { + fileNamesTable[ifnNb] = buf + pos; + if (buf + pos > bufend) { free(buf); free((void*)fileNamesTable); return NULL; } + pos += strlen(fileNamesTable[ifnNb]) + 1; + } + return UTIL_assembleFileNamesTable2(fileNamesTable, nbFiles, fntCapacity, buf); + } +} + + +void UTIL_expandFNT(FileNamesTable** fnt, int followLinks) +{ + FileNamesTable* const newFNT = UTIL_createExpandedFNT((*fnt)->fileNames, (*fnt)->tableSize, followLinks); + CONTROL(newFNT != NULL); + UTIL_freeFileNamesTable(*fnt); + *fnt = newFNT; +} + +FileNamesTable* UTIL_createFNT_fromROTable(const char** filenames, size_t nbFilenames) +{ + size_t const sizeof_FNTable = nbFilenames * sizeof(*filenames); + const char** const newFNTable = (const char**)malloc(sizeof_FNTable); + if (newFNTable==NULL) return NULL; + memcpy((void*)newFNTable, filenames, sizeof_FNTable); /* void* : mitigate a Visual compiler bug or limitation */ + return UTIL_assembleFileNamesTable(newFNTable, nbFilenames, NULL); +} + + +/*-**************************************** * count the number of cores -******************************************/ - -#if defined(_WIN32) || defined(WIN32) - -#include <windows.h> - -typedef BOOL(WINAPI* LPFN_GLPI)(PSYSTEM_LOGICAL_PROCESSOR_INFORMATION, PDWORD); - +******************************************/ + +#if defined(_WIN32) || defined(WIN32) + +#include <windows.h> + +typedef BOOL(WINAPI* LPFN_GLPI)(PSYSTEM_LOGICAL_PROCESSOR_INFORMATION, PDWORD); + DWORD CountSetBits(ULONG_PTR bitMask) -{ +{ DWORD LSHIFT = sizeof(ULONG_PTR)*8 - 1; DWORD bitSetCount = 0; ULONG_PTR bitTest = (ULONG_PTR)1 << LSHIFT; DWORD i; - + for (i = 0; i <= LSHIFT; ++i) { bitSetCount += ((bitMask & bitTest)?1:0); @@ -1139,189 +1139,189 @@ int UTIL_countCores(int logical) static int numCores = 0; if (numCores != 0) return numCores; - { LPFN_GLPI glpi; - BOOL done = FALSE; - PSYSTEM_LOGICAL_PROCESSOR_INFORMATION buffer = NULL; - PSYSTEM_LOGICAL_PROCESSOR_INFORMATION ptr = NULL; - DWORD returnLength = 0; - size_t byteOffset = 0; - -#if defined(_MSC_VER) -/* Visual Studio does not like the following cast */ -# pragma warning( disable : 4054 ) /* conversion from function ptr to data ptr */ -# pragma warning( disable : 4055 ) /* conversion from data ptr to function ptr */ -#endif - glpi = (LPFN_GLPI)(void*)GetProcAddress(GetModuleHandle(TEXT("kernel32")), - "GetLogicalProcessorInformation"); - - if (glpi == NULL) { - goto failed; - } - - while(!done) { - DWORD rc = glpi(buffer, &returnLength); - if (FALSE == rc) { - if (GetLastError() == ERROR_INSUFFICIENT_BUFFER) { - if (buffer) - free(buffer); - buffer = (PSYSTEM_LOGICAL_PROCESSOR_INFORMATION)malloc(returnLength); - - if (buffer == NULL) { - perror("zstd"); - exit(1); - } - } else { - /* some other error */ - goto failed; - } - } else { - done = TRUE; - } } - - ptr = buffer; - - while (byteOffset + sizeof(SYSTEM_LOGICAL_PROCESSOR_INFORMATION) <= returnLength) { - - if (ptr->Relationship == RelationProcessorCore) { + { LPFN_GLPI glpi; + BOOL done = FALSE; + PSYSTEM_LOGICAL_PROCESSOR_INFORMATION buffer = NULL; + PSYSTEM_LOGICAL_PROCESSOR_INFORMATION ptr = NULL; + DWORD returnLength = 0; + size_t byteOffset = 0; + +#if defined(_MSC_VER) +/* Visual Studio does not like the following cast */ +# pragma warning( disable : 4054 ) /* conversion from function ptr to data ptr */ +# pragma warning( disable : 4055 ) /* conversion from data ptr to function ptr */ +#endif + glpi = (LPFN_GLPI)(void*)GetProcAddress(GetModuleHandle(TEXT("kernel32")), + "GetLogicalProcessorInformation"); + + if (glpi == NULL) { + goto failed; + } + + while(!done) { + DWORD rc = glpi(buffer, &returnLength); + if (FALSE == rc) { + if (GetLastError() == ERROR_INSUFFICIENT_BUFFER) { + if (buffer) + free(buffer); + buffer = (PSYSTEM_LOGICAL_PROCESSOR_INFORMATION)malloc(returnLength); + + if (buffer == NULL) { + perror("zstd"); + exit(1); + } + } else { + /* some other error */ + goto failed; + } + } else { + done = TRUE; + } } + + ptr = buffer; + + while (byteOffset + sizeof(SYSTEM_LOGICAL_PROCESSOR_INFORMATION) <= returnLength) { + + if (ptr->Relationship == RelationProcessorCore) { if (logical) numCores += CountSetBits(ptr->ProcessorMask); else numCores++; - } - - ptr++; - byteOffset += sizeof(SYSTEM_LOGICAL_PROCESSOR_INFORMATION); - } - - free(buffer); - + } + + ptr++; + byteOffset += sizeof(SYSTEM_LOGICAL_PROCESSOR_INFORMATION); + } + + free(buffer); + return numCores; - } - -failed: - /* try to fall back on GetSystemInfo */ - { SYSTEM_INFO sysinfo; - GetSystemInfo(&sysinfo); + } + +failed: + /* try to fall back on GetSystemInfo */ + { SYSTEM_INFO sysinfo; + GetSystemInfo(&sysinfo); numCores = sysinfo.dwNumberOfProcessors; if (numCores == 0) numCores = 1; /* just in case */ - } + } return numCores; -} - -#elif defined(__APPLE__) - -#include <sys/sysctl.h> - -/* Use apple-provided syscall - * see: man 3 sysctl */ +} + +#elif defined(__APPLE__) + +#include <sys/sysctl.h> + +/* Use apple-provided syscall + * see: man 3 sysctl */ int UTIL_countCores(int logical) -{ +{ static S32 numCores = 0; /* apple specifies int32_t */ if (numCores != 0) return numCores; - - { size_t size = sizeof(S32); + + { size_t size = sizeof(S32); int const ret = sysctlbyname(logical ? "hw.logicalcpu" : "hw.physicalcpu", &numCores, &size, NULL, 0); - if (ret != 0) { - if (errno == ENOENT) { - /* entry not present, fall back on 1 */ + if (ret != 0) { + if (errno == ENOENT) { + /* entry not present, fall back on 1 */ numCores = 1; - } else { + } else { perror("zstd: can't get number of cpus"); - exit(1); - } - } - + exit(1); + } + } + return numCores; - } -} - -#elif defined(__linux__) - -/* parse /proc/cpuinfo - * siblings / cpu cores should give hyperthreading ratio - * otherwise fall back on sysconf */ + } +} + +#elif defined(__linux__) + +/* parse /proc/cpuinfo + * siblings / cpu cores should give hyperthreading ratio + * otherwise fall back on sysconf */ int UTIL_countCores(int logical) -{ +{ static int numCores = 0; - + if (numCores != 0) return numCores; - + numCores = (int)sysconf(_SC_NPROCESSORS_ONLN); if (numCores == -1) { - /* value not queryable, fall back on 1 */ + /* value not queryable, fall back on 1 */ return numCores = 1; - } - - /* try to determine if there's hyperthreading */ - { FILE* const cpuinfo = fopen("/proc/cpuinfo", "r"); -#define BUF_SIZE 80 - char buff[BUF_SIZE]; - - int siblings = 0; - int cpu_cores = 0; - int ratio = 1; - - if (cpuinfo == NULL) { - /* fall back on the sysconf value */ + } + + /* try to determine if there's hyperthreading */ + { FILE* const cpuinfo = fopen("/proc/cpuinfo", "r"); +#define BUF_SIZE 80 + char buff[BUF_SIZE]; + + int siblings = 0; + int cpu_cores = 0; + int ratio = 1; + + if (cpuinfo == NULL) { + /* fall back on the sysconf value */ return numCores; - } - - /* assume the cpu cores/siblings values will be constant across all - * present processors */ - while (!feof(cpuinfo)) { - if (fgets(buff, BUF_SIZE, cpuinfo) != NULL) { - if (strncmp(buff, "siblings", 8) == 0) { - const char* const sep = strchr(buff, ':'); - if (sep == NULL || *sep == '\0') { - /* formatting was broken? */ - goto failed; - } - - siblings = atoi(sep + 1); - } - if (strncmp(buff, "cpu cores", 9) == 0) { - const char* const sep = strchr(buff, ':'); - if (sep == NULL || *sep == '\0') { - /* formatting was broken? */ - goto failed; - } - - cpu_cores = atoi(sep + 1); - } - } else if (ferror(cpuinfo)) { - /* fall back on the sysconf value */ - goto failed; - } } - if (siblings && cpu_cores && siblings > cpu_cores) { - ratio = siblings / cpu_cores; - } - + } + + /* assume the cpu cores/siblings values will be constant across all + * present processors */ + while (!feof(cpuinfo)) { + if (fgets(buff, BUF_SIZE, cpuinfo) != NULL) { + if (strncmp(buff, "siblings", 8) == 0) { + const char* const sep = strchr(buff, ':'); + if (sep == NULL || *sep == '\0') { + /* formatting was broken? */ + goto failed; + } + + siblings = atoi(sep + 1); + } + if (strncmp(buff, "cpu cores", 9) == 0) { + const char* const sep = strchr(buff, ':'); + if (sep == NULL || *sep == '\0') { + /* formatting was broken? */ + goto failed; + } + + cpu_cores = atoi(sep + 1); + } + } else if (ferror(cpuinfo)) { + /* fall back on the sysconf value */ + goto failed; + } } + if (siblings && cpu_cores && siblings > cpu_cores) { + ratio = siblings / cpu_cores; + } + if (ratio && numCores > ratio && !logical) { numCores = numCores / ratio; - } - -failed: - fclose(cpuinfo); + } + +failed: + fclose(cpuinfo); return numCores; - } -} - -#elif defined(__FreeBSD__) - -#include <sys/param.h> -#include <sys/sysctl.h> - -/* Use physical core sysctl when available - * see: man 4 smp, man 3 sysctl */ + } +} + +#elif defined(__FreeBSD__) + +#include <sys/param.h> +#include <sys/sysctl.h> + +/* Use physical core sysctl when available + * see: man 4 smp, man 3 sysctl */ int UTIL_countCores(int logical) -{ +{ static int numCores = 0; /* freebsd sysctl is native int sized */ #if __FreeBSD_version >= 1300008 static int perCore = 1; #endif if (numCores != 0) return numCores; - -#if __FreeBSD_version >= 1300008 + +#if __FreeBSD_version >= 1300008 { size_t size = sizeof(numCores); int ret = sysctlbyname("kern.smp.cores", &numCores, &size, NULL, 0); if (ret == 0) { @@ -1334,56 +1334,56 @@ int UTIL_countCores(int logical) return numCores; } - if (errno != ENOENT) { + if (errno != ENOENT) { perror("zstd: can't get number of cpus"); - exit(1); - } - /* sysctl not present, fall through to older sysconf method */ - } + exit(1); + } + /* sysctl not present, fall through to older sysconf method */ + } #else /* suppress unused parameter warning */ (void) logical; -#endif - +#endif + numCores = (int)sysconf(_SC_NPROCESSORS_ONLN); if (numCores == -1) { - /* value not queryable, fall back on 1 */ + /* value not queryable, fall back on 1 */ numCores = 1; - } + } return numCores; -} - -#elif defined(__NetBSD__) || defined(__OpenBSD__) || defined(__DragonFly__) || defined(__CYGWIN__) - -/* Use POSIX sysconf - * see: man 3 sysconf */ +} + +#elif defined(__NetBSD__) || defined(__OpenBSD__) || defined(__DragonFly__) || defined(__CYGWIN__) + +/* Use POSIX sysconf + * see: man 3 sysconf */ int UTIL_countCores(int logical) -{ +{ static int numCores = 0; - + /* suppress unused parameter warning */ (void)logical; - + if (numCores != 0) return numCores; numCores = (int)sysconf(_SC_NPROCESSORS_ONLN); if (numCores == -1) { - /* value not queryable, fall back on 1 */ + /* value not queryable, fall back on 1 */ return numCores = 1; - } + } return numCores; -} - -#else - +} + +#else + int UTIL_countCores(int logical) -{ - /* assume 1 */ - return 1; -} - -#endif - +{ + /* assume 1 */ + return 1; +} + +#endif + int UTIL_countPhysicalCores(void) { return UTIL_countCores(0); @@ -1394,6 +1394,6 @@ int UTIL_countLogicalCores(void) return UTIL_countCores(1); } -#if defined (__cplusplus) -} -#endif +#if defined (__cplusplus) +} +#endif diff --git a/contrib/libs/zstd/programs/util.h b/contrib/libs/zstd/programs/util.h index add165d57c..227bcf3e13 100644 --- a/contrib/libs/zstd/programs/util.h +++ b/contrib/libs/zstd/programs/util.h @@ -1,141 +1,141 @@ -/* - * Copyright (c) Przemyslaw Skibinski, Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef UTIL_H_MODULE -#define UTIL_H_MODULE - -#if defined (__cplusplus) -extern "C" { -#endif - - -/*-**************************************** -* Dependencies -******************************************/ -#include "platform.h" /* PLATFORM_POSIX_VERSION, ZSTD_NANOSLEEP_SUPPORT, ZSTD_SETPRIORITY_SUPPORT */ -#include <stddef.h> /* size_t, ptrdiff_t */ -#include <sys/types.h> /* stat, utime */ -#include <sys/stat.h> /* stat, chmod */ -#include "../lib/common/mem.h" /* U64 */ - - -/*-************************************************************ -* Avoid fseek()'s 2GiB barrier with MSVC, macOS, *BSD, MinGW -***************************************************************/ -#if defined(_MSC_VER) && (_MSC_VER >= 1400) -# define UTIL_fseek _fseeki64 -#elif !defined(__64BIT__) && (PLATFORM_POSIX_VERSION >= 200112L) /* No point defining Large file for 64 bit */ -# define UTIL_fseek fseeko -#elif defined(__MINGW32__) && defined(__MSVCRT__) && !defined(__STRICT_ANSI__) && !defined(__NO_MINGW_LFS) -# define UTIL_fseek fseeko64 -#else -# define UTIL_fseek fseek -#endif - - -/*-************************************************* -* Sleep & priority functions: Windows - Posix - others -***************************************************/ -#if defined(_WIN32) -# include <windows.h> -# define SET_REALTIME_PRIORITY SetPriorityClass(GetCurrentProcess(), REALTIME_PRIORITY_CLASS) -# define UTIL_sleep(s) Sleep(1000*s) -# define UTIL_sleepMilli(milli) Sleep(milli) - -#elif PLATFORM_POSIX_VERSION > 0 /* Unix-like operating system */ -# include <unistd.h> /* sleep */ -# define UTIL_sleep(s) sleep(s) -# if ZSTD_NANOSLEEP_SUPPORT /* necessarily defined in platform.h */ -# define UTIL_sleepMilli(milli) { struct timespec t; t.tv_sec=0; t.tv_nsec=milli*1000000ULL; nanosleep(&t, NULL); } -# else -# define UTIL_sleepMilli(milli) /* disabled */ -# endif -# if ZSTD_SETPRIORITY_SUPPORT -# include <sys/resource.h> /* setpriority */ -# define SET_REALTIME_PRIORITY setpriority(PRIO_PROCESS, 0, -20) -# else -# define SET_REALTIME_PRIORITY /* disabled */ -# endif - +/* + * Copyright (c) Przemyslaw Skibinski, Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + +#ifndef UTIL_H_MODULE +#define UTIL_H_MODULE + +#if defined (__cplusplus) +extern "C" { +#endif + + +/*-**************************************** +* Dependencies +******************************************/ +#include "platform.h" /* PLATFORM_POSIX_VERSION, ZSTD_NANOSLEEP_SUPPORT, ZSTD_SETPRIORITY_SUPPORT */ +#include <stddef.h> /* size_t, ptrdiff_t */ +#include <sys/types.h> /* stat, utime */ +#include <sys/stat.h> /* stat, chmod */ +#include "../lib/common/mem.h" /* U64 */ + + +/*-************************************************************ +* Avoid fseek()'s 2GiB barrier with MSVC, macOS, *BSD, MinGW +***************************************************************/ +#if defined(_MSC_VER) && (_MSC_VER >= 1400) +# define UTIL_fseek _fseeki64 +#elif !defined(__64BIT__) && (PLATFORM_POSIX_VERSION >= 200112L) /* No point defining Large file for 64 bit */ +# define UTIL_fseek fseeko +#elif defined(__MINGW32__) && defined(__MSVCRT__) && !defined(__STRICT_ANSI__) && !defined(__NO_MINGW_LFS) +# define UTIL_fseek fseeko64 +#else +# define UTIL_fseek fseek +#endif + + +/*-************************************************* +* Sleep & priority functions: Windows - Posix - others +***************************************************/ +#if defined(_WIN32) +# include <windows.h> +# define SET_REALTIME_PRIORITY SetPriorityClass(GetCurrentProcess(), REALTIME_PRIORITY_CLASS) +# define UTIL_sleep(s) Sleep(1000*s) +# define UTIL_sleepMilli(milli) Sleep(milli) + +#elif PLATFORM_POSIX_VERSION > 0 /* Unix-like operating system */ +# include <unistd.h> /* sleep */ +# define UTIL_sleep(s) sleep(s) +# if ZSTD_NANOSLEEP_SUPPORT /* necessarily defined in platform.h */ +# define UTIL_sleepMilli(milli) { struct timespec t; t.tv_sec=0; t.tv_nsec=milli*1000000ULL; nanosleep(&t, NULL); } +# else +# define UTIL_sleepMilli(milli) /* disabled */ +# endif +# if ZSTD_SETPRIORITY_SUPPORT +# include <sys/resource.h> /* setpriority */ +# define SET_REALTIME_PRIORITY setpriority(PRIO_PROCESS, 0, -20) +# else +# define SET_REALTIME_PRIORITY /* disabled */ +# endif + #else /* unknown non-unix operating system */ -# define UTIL_sleep(s) /* disabled */ -# define UTIL_sleepMilli(milli) /* disabled */ -# define SET_REALTIME_PRIORITY /* disabled */ -#endif - - -/*-**************************************** -* Compiler specifics -******************************************/ -#if defined(__INTEL_COMPILER) -# pragma warning(disable : 177) /* disable: message #177: function was declared but never referenced, useful with UTIL_STATIC */ -#endif -#if defined(__GNUC__) -# define UTIL_STATIC static __attribute__((unused)) -#elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) -# define UTIL_STATIC static inline -#elif defined(_MSC_VER) -# define UTIL_STATIC static __inline -#else -# define UTIL_STATIC static /* this version may generate warnings for unused static functions; disable the relevant warning */ -#endif - - -/*-**************************************** -* Console log -******************************************/ -extern int g_utilDisplayLevel; - -/** - * Displays a message prompt and returns success (0) if first character from stdin - * matches any from acceptableLetters. Otherwise, returns failure (1) and displays abortMsg. - * If any of the inputs are stdin itself, then automatically return failure (1). - */ -int UTIL_requireUserConfirmation(const char* prompt, const char* abortMsg, const char* acceptableLetters, int hasStdinInput); - - -/*-**************************************** -* File functions -******************************************/ -#if defined(_MSC_VER) - typedef struct __stat64 stat_t; - typedef int mode_t; -#elif defined(__MINGW32__) && defined (__MSVCRT__) - typedef struct _stati64 stat_t; -#else - typedef struct stat stat_t; -#endif - -#if defined(_MSC_VER) || defined(__MINGW32__) || defined (__MSVCRT__) /* windows support */ -#define PATH_SEP '\\' -#define STRDUP(s) _strdup(s) -#else -#define PATH_SEP '/' -#include <libgen.h> -#define STRDUP(s) strdup(s) -#endif - - -/** - * Calls platform's equivalent of stat() on filename and writes info to statbuf. - * Returns success (1) or failure (0). - */ -int UTIL_stat(const char* filename, stat_t* statbuf); - -/** - * Instead of getting a file's stats, this updates them with the info in the - * provided stat_t. Currently sets owner, group, atime, and mtime. Will only - * update this info for regular files. - */ -int UTIL_setFileStat(const char* filename, const stat_t* statbuf); - +# define UTIL_sleep(s) /* disabled */ +# define UTIL_sleepMilli(milli) /* disabled */ +# define SET_REALTIME_PRIORITY /* disabled */ +#endif + + +/*-**************************************** +* Compiler specifics +******************************************/ +#if defined(__INTEL_COMPILER) +# pragma warning(disable : 177) /* disable: message #177: function was declared but never referenced, useful with UTIL_STATIC */ +#endif +#if defined(__GNUC__) +# define UTIL_STATIC static __attribute__((unused)) +#elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) +# define UTIL_STATIC static inline +#elif defined(_MSC_VER) +# define UTIL_STATIC static __inline +#else +# define UTIL_STATIC static /* this version may generate warnings for unused static functions; disable the relevant warning */ +#endif + + +/*-**************************************** +* Console log +******************************************/ +extern int g_utilDisplayLevel; + +/** + * Displays a message prompt and returns success (0) if first character from stdin + * matches any from acceptableLetters. Otherwise, returns failure (1) and displays abortMsg. + * If any of the inputs are stdin itself, then automatically return failure (1). + */ +int UTIL_requireUserConfirmation(const char* prompt, const char* abortMsg, const char* acceptableLetters, int hasStdinInput); + + +/*-**************************************** +* File functions +******************************************/ +#if defined(_MSC_VER) + typedef struct __stat64 stat_t; + typedef int mode_t; +#elif defined(__MINGW32__) && defined (__MSVCRT__) + typedef struct _stati64 stat_t; +#else + typedef struct stat stat_t; +#endif + +#if defined(_MSC_VER) || defined(__MINGW32__) || defined (__MSVCRT__) /* windows support */ +#define PATH_SEP '\\' +#define STRDUP(s) _strdup(s) +#else +#define PATH_SEP '/' +#include <libgen.h> +#define STRDUP(s) strdup(s) +#endif + + +/** + * Calls platform's equivalent of stat() on filename and writes info to statbuf. + * Returns success (1) or failure (0). + */ +int UTIL_stat(const char* filename, stat_t* statbuf); + +/** + * Instead of getting a file's stats, this updates them with the info in the + * provided stat_t. Currently sets owner, group, atime, and mtime. Will only + * update this info for regular files. + */ +int UTIL_setFileStat(const char* filename, const stat_t* statbuf); + /** * Set atime to now and mtime to the st_mtim in statbuf. * @@ -144,41 +144,41 @@ int UTIL_setFileStat(const char* filename, const stat_t* statbuf); */ int UTIL_utime(const char* filename, const stat_t *statbuf); -/* - * These helpers operate on a pre-populated stat_t, i.e., the result of - * calling one of the above functions. - */ - -int UTIL_isRegularFileStat(const stat_t* statbuf); -int UTIL_isDirectoryStat(const stat_t* statbuf); -int UTIL_isFIFOStat(const stat_t* statbuf); -int UTIL_isBlockDevStat(const stat_t* statbuf); -U64 UTIL_getFileSizeStat(const stat_t* statbuf); - -/** - * Like chmod(), but only modifies regular files. Provided statbuf may be NULL, - * in which case this function will stat() the file internally, in order to - * check whether it should be modified. - */ -int UTIL_chmod(char const* filename, const stat_t* statbuf, mode_t permissions); - -/* - * In the absence of a pre-existing stat result on the file in question, these - * functions will do a stat() call internally and then use that result to - * compute the needed information. - */ - -int UTIL_isRegularFile(const char* infilename); -int UTIL_isDirectory(const char* infilename); -int UTIL_isSameFile(const char* file1, const char* file2); -int UTIL_isCompressedFile(const char* infilename, const char *extensionList[]); -int UTIL_isLink(const char* infilename); -int UTIL_isFIFO(const char* infilename); - -#define UTIL_FILESIZE_UNKNOWN ((U64)(-1)) -U64 UTIL_getFileSize(const char* infilename); -U64 UTIL_getTotalFileSize(const char* const * fileNamesTable, unsigned nbFiles); - +/* + * These helpers operate on a pre-populated stat_t, i.e., the result of + * calling one of the above functions. + */ + +int UTIL_isRegularFileStat(const stat_t* statbuf); +int UTIL_isDirectoryStat(const stat_t* statbuf); +int UTIL_isFIFOStat(const stat_t* statbuf); +int UTIL_isBlockDevStat(const stat_t* statbuf); +U64 UTIL_getFileSizeStat(const stat_t* statbuf); + +/** + * Like chmod(), but only modifies regular files. Provided statbuf may be NULL, + * in which case this function will stat() the file internally, in order to + * check whether it should be modified. + */ +int UTIL_chmod(char const* filename, const stat_t* statbuf, mode_t permissions); + +/* + * In the absence of a pre-existing stat result on the file in question, these + * functions will do a stat() call internally and then use that result to + * compute the needed information. + */ + +int UTIL_isRegularFile(const char* infilename); +int UTIL_isDirectory(const char* infilename); +int UTIL_isSameFile(const char* file1, const char* file2); +int UTIL_isCompressedFile(const char* infilename, const char *extensionList[]); +int UTIL_isLink(const char* infilename); +int UTIL_isFIFO(const char* infilename); + +#define UTIL_FILESIZE_UNKNOWN ((U64)(-1)) +U64 UTIL_getFileSize(const char* infilename); +U64 UTIL_getTotalFileSize(const char* const * fileNamesTable, unsigned nbFiles); + /** * Take @size in bytes, * prepare the components to pretty-print it in a scaled way. @@ -196,127 +196,127 @@ typedef struct { UTIL_HumanReadableSize_t UTIL_makeHumanReadableSize(U64 size); -int UTIL_compareStr(const void *p1, const void *p2); -const char* UTIL_getFileExtension(const char* infilename); -void UTIL_mirrorSourceFilesDirectories(const char** fileNamesTable, unsigned int nbFiles, const char *outDirName); -char* UTIL_createMirroredDestDirName(const char* srcFileName, const char* outDirRootName); - - - -/*-**************************************** - * Lists of Filenames - ******************************************/ - -typedef struct -{ const char** fileNames; - char* buf; /* fileNames are stored in this buffer (or are read-only) */ - size_t tableSize; /* nb of fileNames */ - size_t tableCapacity; -} FileNamesTable; - -/*! UTIL_createFileNamesTable_fromFileName() : - * read filenames from @inputFileName, and store them into returned object. - * @return : a FileNamesTable*, or NULL in case of error (ex: @inputFileName doesn't exist). - * Note: inputFileSize must be less than 50MB - */ -FileNamesTable* -UTIL_createFileNamesTable_fromFileName(const char* inputFileName); - -/*! UTIL_assembleFileNamesTable() : - * This function takes ownership of its arguments, @filenames and @buf, - * and store them inside the created object. - * note : this function never fails, - * it will rather exit() the program if internal allocation fails. - * @return : resulting FileNamesTable* object. - */ -FileNamesTable* -UTIL_assembleFileNamesTable(const char** filenames, size_t tableSize, char* buf); - -/*! UTIL_freeFileNamesTable() : - * This function is compatible with NULL argument and never fails. - */ -void UTIL_freeFileNamesTable(FileNamesTable* table); - -/*! UTIL_mergeFileNamesTable(): - * @return : FileNamesTable*, concatenation of @table1 and @table2 - * note: @table1 and @table2 are consumed (freed) by this operation - */ -FileNamesTable* -UTIL_mergeFileNamesTable(FileNamesTable* table1, FileNamesTable* table2); - - -/*! UTIL_expandFNT() : - * read names from @fnt, and expand those corresponding to directories - * update @fnt, now containing only file names, - * @return : 0 in case of success, 1 if error - * note : in case of error, @fnt[0] is NULL - */ -void UTIL_expandFNT(FileNamesTable** fnt, int followLinks); - -/*! UTIL_createFNT_fromROTable() : - * copy the @filenames pointer table inside the returned object. - * The names themselves are still stored in their original buffer, which must outlive the object. - * @return : a FileNamesTable* object, - * or NULL in case of error - */ -FileNamesTable* -UTIL_createFNT_fromROTable(const char** filenames, size_t nbFilenames); - -/*! UTIL_allocateFileNamesTable() : - * Allocates a table of const char*, to insert read-only names later on. - * The created FileNamesTable* doesn't hold a buffer. - * @return : FileNamesTable*, or NULL, if allocation fails. - */ -FileNamesTable* UTIL_allocateFileNamesTable(size_t tableSize); - - -/*! UTIL_refFilename() : - * Add a reference to read-only name into @fnt table. - * As @filename is only referenced, its lifetime must outlive @fnt. - * Internal table must be large enough to reference a new member, - * otherwise its UB (protected by an `assert()`). - */ -void UTIL_refFilename(FileNamesTable* fnt, const char* filename); - - -/* UTIL_createExpandedFNT() is only active if UTIL_HAS_CREATEFILELIST is defined. - * Otherwise, UTIL_createExpandedFNT() is a shell function which does nothing - * apart from displaying a warning message. - */ -#ifdef _WIN32 -# define UTIL_HAS_CREATEFILELIST -#elif defined(__linux__) || (PLATFORM_POSIX_VERSION >= 200112L) /* opendir, readdir require POSIX.1-2001 */ -# define UTIL_HAS_CREATEFILELIST -# define UTIL_HAS_MIRRORFILELIST -#else - /* do not define UTIL_HAS_CREATEFILELIST */ -#endif - -/*! UTIL_createExpandedFNT() : - * read names from @filenames, and expand those corresponding to directories. - * links are followed or not depending on @followLinks directive. - * @return : an expanded FileNamesTable*, where each name is a file - * or NULL in case of error - */ -FileNamesTable* -UTIL_createExpandedFNT(const char* const* filenames, size_t nbFilenames, int followLinks); - +int UTIL_compareStr(const void *p1, const void *p2); +const char* UTIL_getFileExtension(const char* infilename); +void UTIL_mirrorSourceFilesDirectories(const char** fileNamesTable, unsigned int nbFiles, const char *outDirName); +char* UTIL_createMirroredDestDirName(const char* srcFileName, const char* outDirRootName); + + + +/*-**************************************** + * Lists of Filenames + ******************************************/ + +typedef struct +{ const char** fileNames; + char* buf; /* fileNames are stored in this buffer (or are read-only) */ + size_t tableSize; /* nb of fileNames */ + size_t tableCapacity; +} FileNamesTable; + +/*! UTIL_createFileNamesTable_fromFileName() : + * read filenames from @inputFileName, and store them into returned object. + * @return : a FileNamesTable*, or NULL in case of error (ex: @inputFileName doesn't exist). + * Note: inputFileSize must be less than 50MB + */ +FileNamesTable* +UTIL_createFileNamesTable_fromFileName(const char* inputFileName); + +/*! UTIL_assembleFileNamesTable() : + * This function takes ownership of its arguments, @filenames and @buf, + * and store them inside the created object. + * note : this function never fails, + * it will rather exit() the program if internal allocation fails. + * @return : resulting FileNamesTable* object. + */ +FileNamesTable* +UTIL_assembleFileNamesTable(const char** filenames, size_t tableSize, char* buf); + +/*! UTIL_freeFileNamesTable() : + * This function is compatible with NULL argument and never fails. + */ +void UTIL_freeFileNamesTable(FileNamesTable* table); + +/*! UTIL_mergeFileNamesTable(): + * @return : FileNamesTable*, concatenation of @table1 and @table2 + * note: @table1 and @table2 are consumed (freed) by this operation + */ +FileNamesTable* +UTIL_mergeFileNamesTable(FileNamesTable* table1, FileNamesTable* table2); + + +/*! UTIL_expandFNT() : + * read names from @fnt, and expand those corresponding to directories + * update @fnt, now containing only file names, + * @return : 0 in case of success, 1 if error + * note : in case of error, @fnt[0] is NULL + */ +void UTIL_expandFNT(FileNamesTable** fnt, int followLinks); + +/*! UTIL_createFNT_fromROTable() : + * copy the @filenames pointer table inside the returned object. + * The names themselves are still stored in their original buffer, which must outlive the object. + * @return : a FileNamesTable* object, + * or NULL in case of error + */ +FileNamesTable* +UTIL_createFNT_fromROTable(const char** filenames, size_t nbFilenames); + +/*! UTIL_allocateFileNamesTable() : + * Allocates a table of const char*, to insert read-only names later on. + * The created FileNamesTable* doesn't hold a buffer. + * @return : FileNamesTable*, or NULL, if allocation fails. + */ +FileNamesTable* UTIL_allocateFileNamesTable(size_t tableSize); + + +/*! UTIL_refFilename() : + * Add a reference to read-only name into @fnt table. + * As @filename is only referenced, its lifetime must outlive @fnt. + * Internal table must be large enough to reference a new member, + * otherwise its UB (protected by an `assert()`). + */ +void UTIL_refFilename(FileNamesTable* fnt, const char* filename); + + +/* UTIL_createExpandedFNT() is only active if UTIL_HAS_CREATEFILELIST is defined. + * Otherwise, UTIL_createExpandedFNT() is a shell function which does nothing + * apart from displaying a warning message. + */ +#ifdef _WIN32 +# define UTIL_HAS_CREATEFILELIST +#elif defined(__linux__) || (PLATFORM_POSIX_VERSION >= 200112L) /* opendir, readdir require POSIX.1-2001 */ +# define UTIL_HAS_CREATEFILELIST +# define UTIL_HAS_MIRRORFILELIST +#else + /* do not define UTIL_HAS_CREATEFILELIST */ +#endif + +/*! UTIL_createExpandedFNT() : + * read names from @filenames, and expand those corresponding to directories. + * links are followed or not depending on @followLinks directive. + * @return : an expanded FileNamesTable*, where each name is a file + * or NULL in case of error + */ +FileNamesTable* +UTIL_createExpandedFNT(const char* const* filenames, size_t nbFilenames, int followLinks); + #if defined(_WIN32) || defined(WIN32) DWORD CountSetBits(ULONG_PTR bitMask); #endif - -/*-**************************************** - * System - ******************************************/ - + +/*-**************************************** + * System + ******************************************/ + int UTIL_countCores(int logical); -int UTIL_countPhysicalCores(void); - +int UTIL_countPhysicalCores(void); + int UTIL_countLogicalCores(void); - -#if defined (__cplusplus) -} -#endif - -#endif /* UTIL_H_MODULE */ + +#if defined (__cplusplus) +} +#endif + +#endif /* UTIL_H_MODULE */ diff --git a/contrib/libs/zstd/programs/zstd/ya.make b/contrib/libs/zstd/programs/zstd/ya.make index 684f19e513..288a6dce95 100644 --- a/contrib/libs/zstd/programs/zstd/ya.make +++ b/contrib/libs/zstd/programs/zstd/ya.make @@ -1,39 +1,39 @@ -# Generated by devtools/yamaker. - -PROGRAM() - +# Generated by devtools/yamaker. + +PROGRAM() + WITHOUT_LICENSE_TEXTS() - + OWNER( orivej velavokr g:cpp-contrib ) -LICENSE(BSD-3-Clause) - -PEERDIR( - contrib/libs/zstd -) - +LICENSE(BSD-3-Clause) + +PEERDIR( + contrib/libs/zstd +) + ADDINCL( contrib/libs/zstd/lib contrib/libs/zstd/lib/common contrib/libs/zstd/programs ) -NO_COMPILER_WARNINGS() - -NO_RUNTIME() - -CFLAGS( - -DZSTD_LEGACY_SUPPORT=1 - -DZSTD_MULTITHREAD -) - +NO_COMPILER_WARNINGS() + +NO_RUNTIME() + +CFLAGS( + -DZSTD_LEGACY_SUPPORT=1 + -DZSTD_MULTITHREAD +) + SRCDIR(contrib/libs/zstd/programs) - -SRCS( + +SRCS( benchfn.c benchzstd.c datagen.c @@ -43,6 +43,6 @@ SRCS( util.c zstdcli.c zstdcli_trace.c -) - -END() +) + +END() diff --git a/contrib/libs/zstd/programs/zstdcli.c b/contrib/libs/zstd/programs/zstdcli.c index bfe18c0c1b..a33466a372 100644 --- a/contrib/libs/zstd/programs/zstdcli.c +++ b/contrib/libs/zstd/programs/zstdcli.c @@ -1,360 +1,360 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - -/*-************************************ -* Tuning parameters -**************************************/ -#ifndef ZSTDCLI_CLEVEL_DEFAULT -# define ZSTDCLI_CLEVEL_DEFAULT 3 -#endif - -#ifndef ZSTDCLI_CLEVEL_MAX -# define ZSTDCLI_CLEVEL_MAX 19 /* without using --ultra */ -#endif - -#ifndef ZSTDCLI_NBTHREADS_DEFAULT -# define ZSTDCLI_NBTHREADS_DEFAULT 1 -#endif - -/*-************************************ -* Dependencies -**************************************/ -#include "platform.h" /* IS_CONSOLE, PLATFORM_POSIX_VERSION */ -#include "util.h" /* UTIL_HAS_CREATEFILELIST, UTIL_createFileList */ -#include <stdlib.h> /* getenv */ -#include <string.h> /* strcmp, strlen */ -#include <stdio.h> /* fprintf(), stdin, stdout, stderr */ -#include <errno.h> /* errno */ -#include <assert.h> /* assert */ - -#include "fileio.h" /* stdinmark, stdoutmark, ZSTD_EXTENSION */ -#ifndef ZSTD_NOBENCH -# include "benchzstd.h" /* BMK_benchFiles */ -#endif -#ifndef ZSTD_NODICT -# include "dibio.h" /* ZDICT_cover_params_t, DiB_trainFromFiles() */ -#endif -#ifndef ZSTD_NOTRACE -# include "zstdcli_trace.h" -#endif -#include "../lib/zstd.h" /* ZSTD_VERSION_STRING, ZSTD_minCLevel, ZSTD_maxCLevel */ - - -/*-************************************ -* Constants -**************************************/ -#define COMPRESSOR_NAME "zstd command line interface" -#ifndef ZSTD_VERSION -# define ZSTD_VERSION "v" ZSTD_VERSION_STRING -#endif -#define AUTHOR "Yann Collet" -#define WELCOME_MESSAGE "*** %s %i-bits %s, by %s ***\n", COMPRESSOR_NAME, (int)(sizeof(size_t)*8), ZSTD_VERSION, AUTHOR - -#define ZSTD_ZSTDMT "zstdmt" -#define ZSTD_UNZSTD "unzstd" -#define ZSTD_CAT "zstdcat" -#define ZSTD_ZCAT "zcat" -#define ZSTD_GZ "gzip" -#define ZSTD_GUNZIP "gunzip" -#define ZSTD_GZCAT "gzcat" -#define ZSTD_LZMA "lzma" -#define ZSTD_UNLZMA "unlzma" -#define ZSTD_XZ "xz" -#define ZSTD_UNXZ "unxz" -#define ZSTD_LZ4 "lz4" -#define ZSTD_UNLZ4 "unlz4" - -#define KB *(1 <<10) -#define MB *(1 <<20) -#define GB *(1U<<30) - -#define DISPLAY_LEVEL_DEFAULT 2 - -static const char* g_defaultDictName = "dictionary"; -static const unsigned g_defaultMaxDictSize = 110 KB; -static const int g_defaultDictCLevel = 3; -static const unsigned g_defaultSelectivityLevel = 9; -static const unsigned g_defaultMaxWindowLog = 27; -#define OVERLAP_LOG_DEFAULT 9999 -#define LDM_PARAM_DEFAULT 9999 /* Default for parameters where 0 is valid */ -static U32 g_overlapLog = OVERLAP_LOG_DEFAULT; -static U32 g_ldmHashLog = 0; -static U32 g_ldmMinMatch = 0; -static U32 g_ldmHashRateLog = LDM_PARAM_DEFAULT; -static U32 g_ldmBucketSizeLog = LDM_PARAM_DEFAULT; - - -#define DEFAULT_ACCEL 1 - -typedef enum { cover, fastCover, legacy } dictType; - -/*-************************************ -* Display Macros -**************************************/ -#define DISPLAY_F(f, ...) fprintf((f), __VA_ARGS__) -#define DISPLAYOUT(...) DISPLAY_F(stdout, __VA_ARGS__) -#define DISPLAY(...) DISPLAY_F(stderr, __VA_ARGS__) -#define DISPLAYLEVEL(l, ...) { if (g_displayLevel>=l) { DISPLAY(__VA_ARGS__); } } -static int g_displayLevel = DISPLAY_LEVEL_DEFAULT; /* 0 : no display, 1: errors, 2 : + result + interaction + warnings, 3 : + progression, 4 : + information */ - - -/*-************************************ -* Check Version (when CLI linked to dynamic library) -**************************************/ - -/* Due to usage of experimental symbols and capabilities by the CLI, - * the CLI must be linked against a dynamic library of same version */ -static void checkLibVersion(void) -{ - if (strcmp(ZSTD_VERSION_STRING, ZSTD_versionString())) { - DISPLAYLEVEL(1, "Error : incorrect library version (expecting : %s ; actual : %s ) \n", - ZSTD_VERSION_STRING, ZSTD_versionString()); - DISPLAYLEVEL(1, "Please update library to version %s, or use stand-alone zstd binary \n", - ZSTD_VERSION_STRING); - exit(1); - } -} - - -/*-************************************ -* Command Line -**************************************/ -/* print help either in `stderr` or `stdout` depending on originating request - * error (badusage) => stderr - * help (usage_advanced) => stdout - */ -static void usage(FILE* f, const char* programName) -{ - DISPLAY_F(f, "Usage : \n"); - DISPLAY_F(f, " %s [args] [FILE(s)] [-o file] \n", programName); - DISPLAY_F(f, "\n"); - DISPLAY_F(f, "FILE : a filename \n"); - DISPLAY_F(f, " with no FILE, or when FILE is - , read standard input\n"); - DISPLAY_F(f, "Arguments : \n"); -#ifndef ZSTD_NOCOMPRESS - DISPLAY_F(f, " -# : # compression level (1-%d, default: %d) \n", ZSTDCLI_CLEVEL_MAX, ZSTDCLI_CLEVEL_DEFAULT); -#endif -#ifndef ZSTD_NODECOMPRESS - DISPLAY_F(f, " -d : decompression \n"); -#endif - DISPLAY_F(f, " -D DICT: use DICT as Dictionary for compression or decompression \n"); - DISPLAY_F(f, " -o file: result stored into `file` (only 1 output file) \n"); - DISPLAY_F(f, " -f : disable input and output checks. Allows overwriting existing files,\n"); - DISPLAY_F(f, " input from console, output to stdout, operating on links,\n"); - DISPLAY_F(f, " block devices, etc.\n"); - DISPLAY_F(f, "--rm : remove source file(s) after successful de/compression \n"); - DISPLAY_F(f, " -k : preserve source file(s) (default) \n"); - DISPLAY_F(f, " -h/-H : display help/long help and exit \n"); -} - -static void usage_advanced(const char* programName) -{ - DISPLAYOUT(WELCOME_MESSAGE); - usage(stdout, programName); - DISPLAYOUT( "\n"); - DISPLAYOUT( "Advanced arguments : \n"); - DISPLAYOUT( " -V : display Version number and exit \n"); - +/* + * Copyright (c) Yann Collet, Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + + +/*-************************************ +* Tuning parameters +**************************************/ +#ifndef ZSTDCLI_CLEVEL_DEFAULT +# define ZSTDCLI_CLEVEL_DEFAULT 3 +#endif + +#ifndef ZSTDCLI_CLEVEL_MAX +# define ZSTDCLI_CLEVEL_MAX 19 /* without using --ultra */ +#endif + +#ifndef ZSTDCLI_NBTHREADS_DEFAULT +# define ZSTDCLI_NBTHREADS_DEFAULT 1 +#endif + +/*-************************************ +* Dependencies +**************************************/ +#include "platform.h" /* IS_CONSOLE, PLATFORM_POSIX_VERSION */ +#include "util.h" /* UTIL_HAS_CREATEFILELIST, UTIL_createFileList */ +#include <stdlib.h> /* getenv */ +#include <string.h> /* strcmp, strlen */ +#include <stdio.h> /* fprintf(), stdin, stdout, stderr */ +#include <errno.h> /* errno */ +#include <assert.h> /* assert */ + +#include "fileio.h" /* stdinmark, stdoutmark, ZSTD_EXTENSION */ +#ifndef ZSTD_NOBENCH +# include "benchzstd.h" /* BMK_benchFiles */ +#endif +#ifndef ZSTD_NODICT +# include "dibio.h" /* ZDICT_cover_params_t, DiB_trainFromFiles() */ +#endif +#ifndef ZSTD_NOTRACE +# include "zstdcli_trace.h" +#endif +#include "../lib/zstd.h" /* ZSTD_VERSION_STRING, ZSTD_minCLevel, ZSTD_maxCLevel */ + + +/*-************************************ +* Constants +**************************************/ +#define COMPRESSOR_NAME "zstd command line interface" +#ifndef ZSTD_VERSION +# define ZSTD_VERSION "v" ZSTD_VERSION_STRING +#endif +#define AUTHOR "Yann Collet" +#define WELCOME_MESSAGE "*** %s %i-bits %s, by %s ***\n", COMPRESSOR_NAME, (int)(sizeof(size_t)*8), ZSTD_VERSION, AUTHOR + +#define ZSTD_ZSTDMT "zstdmt" +#define ZSTD_UNZSTD "unzstd" +#define ZSTD_CAT "zstdcat" +#define ZSTD_ZCAT "zcat" +#define ZSTD_GZ "gzip" +#define ZSTD_GUNZIP "gunzip" +#define ZSTD_GZCAT "gzcat" +#define ZSTD_LZMA "lzma" +#define ZSTD_UNLZMA "unlzma" +#define ZSTD_XZ "xz" +#define ZSTD_UNXZ "unxz" +#define ZSTD_LZ4 "lz4" +#define ZSTD_UNLZ4 "unlz4" + +#define KB *(1 <<10) +#define MB *(1 <<20) +#define GB *(1U<<30) + +#define DISPLAY_LEVEL_DEFAULT 2 + +static const char* g_defaultDictName = "dictionary"; +static const unsigned g_defaultMaxDictSize = 110 KB; +static const int g_defaultDictCLevel = 3; +static const unsigned g_defaultSelectivityLevel = 9; +static const unsigned g_defaultMaxWindowLog = 27; +#define OVERLAP_LOG_DEFAULT 9999 +#define LDM_PARAM_DEFAULT 9999 /* Default for parameters where 0 is valid */ +static U32 g_overlapLog = OVERLAP_LOG_DEFAULT; +static U32 g_ldmHashLog = 0; +static U32 g_ldmMinMatch = 0; +static U32 g_ldmHashRateLog = LDM_PARAM_DEFAULT; +static U32 g_ldmBucketSizeLog = LDM_PARAM_DEFAULT; + + +#define DEFAULT_ACCEL 1 + +typedef enum { cover, fastCover, legacy } dictType; + +/*-************************************ +* Display Macros +**************************************/ +#define DISPLAY_F(f, ...) fprintf((f), __VA_ARGS__) +#define DISPLAYOUT(...) DISPLAY_F(stdout, __VA_ARGS__) +#define DISPLAY(...) DISPLAY_F(stderr, __VA_ARGS__) +#define DISPLAYLEVEL(l, ...) { if (g_displayLevel>=l) { DISPLAY(__VA_ARGS__); } } +static int g_displayLevel = DISPLAY_LEVEL_DEFAULT; /* 0 : no display, 1: errors, 2 : + result + interaction + warnings, 3 : + progression, 4 : + information */ + + +/*-************************************ +* Check Version (when CLI linked to dynamic library) +**************************************/ + +/* Due to usage of experimental symbols and capabilities by the CLI, + * the CLI must be linked against a dynamic library of same version */ +static void checkLibVersion(void) +{ + if (strcmp(ZSTD_VERSION_STRING, ZSTD_versionString())) { + DISPLAYLEVEL(1, "Error : incorrect library version (expecting : %s ; actual : %s ) \n", + ZSTD_VERSION_STRING, ZSTD_versionString()); + DISPLAYLEVEL(1, "Please update library to version %s, or use stand-alone zstd binary \n", + ZSTD_VERSION_STRING); + exit(1); + } +} + + +/*-************************************ +* Command Line +**************************************/ +/* print help either in `stderr` or `stdout` depending on originating request + * error (badusage) => stderr + * help (usage_advanced) => stdout + */ +static void usage(FILE* f, const char* programName) +{ + DISPLAY_F(f, "Usage : \n"); + DISPLAY_F(f, " %s [args] [FILE(s)] [-o file] \n", programName); + DISPLAY_F(f, "\n"); + DISPLAY_F(f, "FILE : a filename \n"); + DISPLAY_F(f, " with no FILE, or when FILE is - , read standard input\n"); + DISPLAY_F(f, "Arguments : \n"); +#ifndef ZSTD_NOCOMPRESS + DISPLAY_F(f, " -# : # compression level (1-%d, default: %d) \n", ZSTDCLI_CLEVEL_MAX, ZSTDCLI_CLEVEL_DEFAULT); +#endif +#ifndef ZSTD_NODECOMPRESS + DISPLAY_F(f, " -d : decompression \n"); +#endif + DISPLAY_F(f, " -D DICT: use DICT as Dictionary for compression or decompression \n"); + DISPLAY_F(f, " -o file: result stored into `file` (only 1 output file) \n"); + DISPLAY_F(f, " -f : disable input and output checks. Allows overwriting existing files,\n"); + DISPLAY_F(f, " input from console, output to stdout, operating on links,\n"); + DISPLAY_F(f, " block devices, etc.\n"); + DISPLAY_F(f, "--rm : remove source file(s) after successful de/compression \n"); + DISPLAY_F(f, " -k : preserve source file(s) (default) \n"); + DISPLAY_F(f, " -h/-H : display help/long help and exit \n"); +} + +static void usage_advanced(const char* programName) +{ + DISPLAYOUT(WELCOME_MESSAGE); + usage(stdout, programName); + DISPLAYOUT( "\n"); + DISPLAYOUT( "Advanced arguments : \n"); + DISPLAYOUT( " -V : display Version number and exit \n"); + DISPLAYOUT( " -c : write to standard output (even if it is the console) \n"); - - DISPLAYOUT( " -v : verbose mode; specify multiple times to increase verbosity \n"); - DISPLAYOUT( " -q : suppress warnings; specify twice to suppress errors too \n"); - DISPLAYOUT( "--[no-]progress : forcibly display, or never display the progress counter.\n"); - DISPLAYOUT( " note: any (de)compressed output to terminal will mix with progress counter text. \n"); - -#ifdef UTIL_HAS_CREATEFILELIST - DISPLAYOUT( " -r : operate recursively on directories \n"); - DISPLAYOUT( "--filelist FILE : read list of files to operate upon from FILE \n"); - DISPLAYOUT( "--output-dir-flat DIR : processed files are stored into DIR \n"); -#endif - -#ifdef UTIL_HAS_MIRRORFILELIST - DISPLAYOUT( "--output-dir-mirror DIR : processed files are stored into DIR respecting original directory structure \n"); -#endif - - -#ifndef ZSTD_NOCOMPRESS - DISPLAYOUT( "--[no-]check : during compression, add XXH64 integrity checksum to frame (default: enabled)"); -#ifndef ZSTD_NODECOMPRESS - DISPLAYOUT( ". If specified with -d, decompressor will ignore/validate checksums in compressed frame (default: validate)."); -#endif -#else -#ifdef ZSTD_NOCOMPRESS - DISPLAYOUT( "--[no-]check : during decompression, ignore/validate checksums in compressed frame (default: validate)."); -#endif -#endif /* ZSTD_NOCOMPRESS */ - -#ifndef ZSTD_NOTRACE - DISPLAYOUT( "\n"); - DISPLAYOUT( "--trace FILE : log tracing information to FILE."); -#endif - DISPLAYOUT( "\n"); - - DISPLAYOUT( "-- : All arguments after \"--\" are treated as files \n"); - -#ifndef ZSTD_NOCOMPRESS - DISPLAYOUT( "\n"); - DISPLAYOUT( "Advanced compression arguments : \n"); - DISPLAYOUT( "--ultra : enable levels beyond %i, up to %i (requires more memory) \n", ZSTDCLI_CLEVEL_MAX, ZSTD_maxCLevel()); - DISPLAYOUT( "--long[=#]: enable long distance matching with given window log (default: %u) \n", g_defaultMaxWindowLog); - DISPLAYOUT( "--fast[=#]: switch to very fast compression levels (default: %u) \n", 1); - DISPLAYOUT( "--adapt : dynamically adapt compression level to I/O conditions \n"); - DISPLAYOUT( "--[no-]row-match-finder : force enable/disable usage of fast row-based matchfinder for greedy, lazy, and lazy2 strategies \n"); + + DISPLAYOUT( " -v : verbose mode; specify multiple times to increase verbosity \n"); + DISPLAYOUT( " -q : suppress warnings; specify twice to suppress errors too \n"); + DISPLAYOUT( "--[no-]progress : forcibly display, or never display the progress counter.\n"); + DISPLAYOUT( " note: any (de)compressed output to terminal will mix with progress counter text. \n"); + +#ifdef UTIL_HAS_CREATEFILELIST + DISPLAYOUT( " -r : operate recursively on directories \n"); + DISPLAYOUT( "--filelist FILE : read list of files to operate upon from FILE \n"); + DISPLAYOUT( "--output-dir-flat DIR : processed files are stored into DIR \n"); +#endif + +#ifdef UTIL_HAS_MIRRORFILELIST + DISPLAYOUT( "--output-dir-mirror DIR : processed files are stored into DIR respecting original directory structure \n"); +#endif + + +#ifndef ZSTD_NOCOMPRESS + DISPLAYOUT( "--[no-]check : during compression, add XXH64 integrity checksum to frame (default: enabled)"); +#ifndef ZSTD_NODECOMPRESS + DISPLAYOUT( ". If specified with -d, decompressor will ignore/validate checksums in compressed frame (default: validate)."); +#endif +#else +#ifdef ZSTD_NOCOMPRESS + DISPLAYOUT( "--[no-]check : during decompression, ignore/validate checksums in compressed frame (default: validate)."); +#endif +#endif /* ZSTD_NOCOMPRESS */ + +#ifndef ZSTD_NOTRACE + DISPLAYOUT( "\n"); + DISPLAYOUT( "--trace FILE : log tracing information to FILE."); +#endif + DISPLAYOUT( "\n"); + + DISPLAYOUT( "-- : All arguments after \"--\" are treated as files \n"); + +#ifndef ZSTD_NOCOMPRESS + DISPLAYOUT( "\n"); + DISPLAYOUT( "Advanced compression arguments : \n"); + DISPLAYOUT( "--ultra : enable levels beyond %i, up to %i (requires more memory) \n", ZSTDCLI_CLEVEL_MAX, ZSTD_maxCLevel()); + DISPLAYOUT( "--long[=#]: enable long distance matching with given window log (default: %u) \n", g_defaultMaxWindowLog); + DISPLAYOUT( "--fast[=#]: switch to very fast compression levels (default: %u) \n", 1); + DISPLAYOUT( "--adapt : dynamically adapt compression level to I/O conditions \n"); + DISPLAYOUT( "--[no-]row-match-finder : force enable/disable usage of fast row-based matchfinder for greedy, lazy, and lazy2 strategies \n"); DISPLAYOUT( "--patch-from=FILE : specify the file to be used as a reference point for zstd's diff engine. \n"); -# ifdef ZSTD_MULTITHREAD - DISPLAYOUT( " -T# : spawns # compression threads (default: 1, 0==# cores) \n"); - DISPLAYOUT( " -B# : select size of each job (default: 0==automatic) \n"); - DISPLAYOUT( "--single-thread : use a single thread for both I/O and compression (result slightly different than -T1) \n"); +# ifdef ZSTD_MULTITHREAD + DISPLAYOUT( " -T# : spawns # compression threads (default: 1, 0==# cores) \n"); + DISPLAYOUT( " -B# : select size of each job (default: 0==automatic) \n"); + DISPLAYOUT( "--single-thread : use a single thread for both I/O and compression (result slightly different than -T1) \n"); DISPLAYOUT( "--auto-threads={physical,logical} (default: physical} : use either physical cores or logical cores as default when specifying -T0 \n"); - DISPLAYOUT( "--rsyncable : compress using a rsync-friendly method (-B sets block size) \n"); -# endif - DISPLAYOUT( "--exclude-compressed: only compress files that are not already compressed \n"); - DISPLAYOUT( "--stream-size=# : specify size of streaming input from `stdin` \n"); - DISPLAYOUT( "--size-hint=# optimize compression parameters for streaming input of approximately this size \n"); - DISPLAYOUT( "--target-compressed-block-size=# : generate compressed block of approximately targeted size \n"); - DISPLAYOUT( "--no-dictID : don't write dictID into header (dictionary compression only) \n"); - DISPLAYOUT( "--[no-]compress-literals : force (un)compressed literals \n"); - - DISPLAYOUT( "--format=zstd : compress files to the .zst format (default) \n"); -#ifdef ZSTD_GZCOMPRESS - DISPLAYOUT( "--format=gzip : compress files to the .gz format \n"); -#endif -#ifdef ZSTD_LZMACOMPRESS - DISPLAYOUT( "--format=xz : compress files to the .xz format \n"); - DISPLAYOUT( "--format=lzma : compress files to the .lzma format \n"); -#endif -#ifdef ZSTD_LZ4COMPRESS - DISPLAYOUT( "--format=lz4 : compress files to the .lz4 format \n"); -#endif -#endif /* !ZSTD_NOCOMPRESS */ - -#ifndef ZSTD_NODECOMPRESS - DISPLAYOUT( "\n"); - DISPLAYOUT( "Advanced decompression arguments : \n"); - DISPLAYOUT( " -l : print information about zstd compressed files \n"); - DISPLAYOUT( "--test : test compressed file integrity \n"); - DISPLAYOUT( " -M# : Set a memory usage limit for decompression \n"); -# if ZSTD_SPARSE_DEFAULT - DISPLAYOUT( "--[no-]sparse : sparse mode (default: enabled on file, disabled on stdout) \n"); -# else - DISPLAYOUT( "--[no-]sparse : sparse mode (default: disabled) \n"); -# endif -#endif /* ZSTD_NODECOMPRESS */ - -#ifndef ZSTD_NODICT - DISPLAYOUT( "\n"); - DISPLAYOUT( "Dictionary builder : \n"); - DISPLAYOUT( "--train ## : create a dictionary from a training set of files \n"); - DISPLAYOUT( "--train-cover[=k=#,d=#,steps=#,split=#,shrink[=#]] : use the cover algorithm with optional args \n"); - DISPLAYOUT( "--train-fastcover[=k=#,d=#,f=#,steps=#,split=#,accel=#,shrink[=#]] : use the fast cover algorithm with optional args \n"); - DISPLAYOUT( "--train-legacy[=s=#] : use the legacy algorithm with selectivity (default: %u) \n", g_defaultSelectivityLevel); - DISPLAYOUT( " -o DICT : DICT is dictionary name (default: %s) \n", g_defaultDictName); - DISPLAYOUT( "--maxdict=# : limit dictionary to specified size (default: %u) \n", g_defaultMaxDictSize); - DISPLAYOUT( "--dictID=# : force dictionary ID to specified value (default: random) \n"); -#endif - -#ifndef ZSTD_NOBENCH - DISPLAYOUT( "\n"); - DISPLAYOUT( "Benchmark arguments : \n"); - DISPLAYOUT( " -b# : benchmark file(s), using # compression level (default: %d) \n", ZSTDCLI_CLEVEL_DEFAULT); - DISPLAYOUT( " -e# : test all compression levels successively from -b# to -e# (default: 1) \n"); - DISPLAYOUT( " -i# : minimum evaluation time in seconds (default: 3s) \n"); - DISPLAYOUT( " -B# : cut file into independent blocks of size # (default: no block) \n"); - DISPLAYOUT( " -S : output one benchmark result per input file (default: consolidated result) \n"); - DISPLAYOUT( "--priority=rt : set process priority to real-time \n"); -#endif - -} - -static void badusage(const char* programName) -{ - DISPLAYLEVEL(1, "Incorrect parameters \n"); - if (g_displayLevel >= 2) usage(stderr, programName); -} - -static void waitEnter(void) -{ - int unused; - DISPLAY("Press enter to continue... \n"); - unused = getchar(); - (void)unused; -} - -static const char* lastNameFromPath(const char* path) -{ - const char* name = path; - if (strrchr(name, '/')) name = strrchr(name, '/') + 1; - if (strrchr(name, '\\')) name = strrchr(name, '\\') + 1; /* windows */ - return name; -} - -/*! exeNameMatch() : - @return : a non-zero value if exeName matches test, excluding the extension - */ -static int exeNameMatch(const char* exeName, const char* test) -{ - return !strncmp(exeName, test, strlen(test)) && - (exeName[strlen(test)] == '\0' || exeName[strlen(test)] == '.'); -} - -static void errorOut(const char* msg) -{ - DISPLAY("%s \n", msg); exit(1); -} - -/*! readU32FromCharChecked() : - * @return 0 if success, and store the result in *value. - * allows and interprets K, KB, KiB, M, MB and MiB suffix. - * Will also modify `*stringPtr`, advancing it to position where it stopped reading. - * @return 1 if an overflow error occurs */ -static int readU32FromCharChecked(const char** stringPtr, unsigned* value) -{ - unsigned result = 0; - while ((**stringPtr >='0') && (**stringPtr <='9')) { - unsigned const max = ((unsigned)(-1)) / 10; - unsigned last = result; - if (result > max) return 1; /* overflow error */ - result *= 10; - result += (unsigned)(**stringPtr - '0'); - if (result < last) return 1; /* overflow error */ - (*stringPtr)++ ; - } - if ((**stringPtr=='K') || (**stringPtr=='M')) { - unsigned const maxK = ((unsigned)(-1)) >> 10; - if (result > maxK) return 1; /* overflow error */ - result <<= 10; - if (**stringPtr=='M') { - if (result > maxK) return 1; /* overflow error */ - result <<= 10; - } - (*stringPtr)++; /* skip `K` or `M` */ - if (**stringPtr=='i') (*stringPtr)++; - if (**stringPtr=='B') (*stringPtr)++; - } - *value = result; - return 0; -} - -/*! readU32FromChar() : - * @return : unsigned integer value read from input in `char` format. - * allows and interprets K, KB, KiB, M, MB and MiB suffix. - * Will also modify `*stringPtr`, advancing it to position where it stopped reading. - * Note : function will exit() program if digit sequence overflows */ -static unsigned readU32FromChar(const char** stringPtr) { - static const char errorMsg[] = "error: numeric value overflows 32-bit unsigned int"; - unsigned result; - if (readU32FromCharChecked(stringPtr, &result)) { errorOut(errorMsg); } - return result; -} - + DISPLAYOUT( "--rsyncable : compress using a rsync-friendly method (-B sets block size) \n"); +# endif + DISPLAYOUT( "--exclude-compressed: only compress files that are not already compressed \n"); + DISPLAYOUT( "--stream-size=# : specify size of streaming input from `stdin` \n"); + DISPLAYOUT( "--size-hint=# optimize compression parameters for streaming input of approximately this size \n"); + DISPLAYOUT( "--target-compressed-block-size=# : generate compressed block of approximately targeted size \n"); + DISPLAYOUT( "--no-dictID : don't write dictID into header (dictionary compression only) \n"); + DISPLAYOUT( "--[no-]compress-literals : force (un)compressed literals \n"); + + DISPLAYOUT( "--format=zstd : compress files to the .zst format (default) \n"); +#ifdef ZSTD_GZCOMPRESS + DISPLAYOUT( "--format=gzip : compress files to the .gz format \n"); +#endif +#ifdef ZSTD_LZMACOMPRESS + DISPLAYOUT( "--format=xz : compress files to the .xz format \n"); + DISPLAYOUT( "--format=lzma : compress files to the .lzma format \n"); +#endif +#ifdef ZSTD_LZ4COMPRESS + DISPLAYOUT( "--format=lz4 : compress files to the .lz4 format \n"); +#endif +#endif /* !ZSTD_NOCOMPRESS */ + +#ifndef ZSTD_NODECOMPRESS + DISPLAYOUT( "\n"); + DISPLAYOUT( "Advanced decompression arguments : \n"); + DISPLAYOUT( " -l : print information about zstd compressed files \n"); + DISPLAYOUT( "--test : test compressed file integrity \n"); + DISPLAYOUT( " -M# : Set a memory usage limit for decompression \n"); +# if ZSTD_SPARSE_DEFAULT + DISPLAYOUT( "--[no-]sparse : sparse mode (default: enabled on file, disabled on stdout) \n"); +# else + DISPLAYOUT( "--[no-]sparse : sparse mode (default: disabled) \n"); +# endif +#endif /* ZSTD_NODECOMPRESS */ + +#ifndef ZSTD_NODICT + DISPLAYOUT( "\n"); + DISPLAYOUT( "Dictionary builder : \n"); + DISPLAYOUT( "--train ## : create a dictionary from a training set of files \n"); + DISPLAYOUT( "--train-cover[=k=#,d=#,steps=#,split=#,shrink[=#]] : use the cover algorithm with optional args \n"); + DISPLAYOUT( "--train-fastcover[=k=#,d=#,f=#,steps=#,split=#,accel=#,shrink[=#]] : use the fast cover algorithm with optional args \n"); + DISPLAYOUT( "--train-legacy[=s=#] : use the legacy algorithm with selectivity (default: %u) \n", g_defaultSelectivityLevel); + DISPLAYOUT( " -o DICT : DICT is dictionary name (default: %s) \n", g_defaultDictName); + DISPLAYOUT( "--maxdict=# : limit dictionary to specified size (default: %u) \n", g_defaultMaxDictSize); + DISPLAYOUT( "--dictID=# : force dictionary ID to specified value (default: random) \n"); +#endif + +#ifndef ZSTD_NOBENCH + DISPLAYOUT( "\n"); + DISPLAYOUT( "Benchmark arguments : \n"); + DISPLAYOUT( " -b# : benchmark file(s), using # compression level (default: %d) \n", ZSTDCLI_CLEVEL_DEFAULT); + DISPLAYOUT( " -e# : test all compression levels successively from -b# to -e# (default: 1) \n"); + DISPLAYOUT( " -i# : minimum evaluation time in seconds (default: 3s) \n"); + DISPLAYOUT( " -B# : cut file into independent blocks of size # (default: no block) \n"); + DISPLAYOUT( " -S : output one benchmark result per input file (default: consolidated result) \n"); + DISPLAYOUT( "--priority=rt : set process priority to real-time \n"); +#endif + +} + +static void badusage(const char* programName) +{ + DISPLAYLEVEL(1, "Incorrect parameters \n"); + if (g_displayLevel >= 2) usage(stderr, programName); +} + +static void waitEnter(void) +{ + int unused; + DISPLAY("Press enter to continue... \n"); + unused = getchar(); + (void)unused; +} + +static const char* lastNameFromPath(const char* path) +{ + const char* name = path; + if (strrchr(name, '/')) name = strrchr(name, '/') + 1; + if (strrchr(name, '\\')) name = strrchr(name, '\\') + 1; /* windows */ + return name; +} + +/*! exeNameMatch() : + @return : a non-zero value if exeName matches test, excluding the extension + */ +static int exeNameMatch(const char* exeName, const char* test) +{ + return !strncmp(exeName, test, strlen(test)) && + (exeName[strlen(test)] == '\0' || exeName[strlen(test)] == '.'); +} + +static void errorOut(const char* msg) +{ + DISPLAY("%s \n", msg); exit(1); +} + +/*! readU32FromCharChecked() : + * @return 0 if success, and store the result in *value. + * allows and interprets K, KB, KiB, M, MB and MiB suffix. + * Will also modify `*stringPtr`, advancing it to position where it stopped reading. + * @return 1 if an overflow error occurs */ +static int readU32FromCharChecked(const char** stringPtr, unsigned* value) +{ + unsigned result = 0; + while ((**stringPtr >='0') && (**stringPtr <='9')) { + unsigned const max = ((unsigned)(-1)) / 10; + unsigned last = result; + if (result > max) return 1; /* overflow error */ + result *= 10; + result += (unsigned)(**stringPtr - '0'); + if (result < last) return 1; /* overflow error */ + (*stringPtr)++ ; + } + if ((**stringPtr=='K') || (**stringPtr=='M')) { + unsigned const maxK = ((unsigned)(-1)) >> 10; + if (result > maxK) return 1; /* overflow error */ + result <<= 10; + if (**stringPtr=='M') { + if (result > maxK) return 1; /* overflow error */ + result <<= 10; + } + (*stringPtr)++; /* skip `K` or `M` */ + if (**stringPtr=='i') (*stringPtr)++; + if (**stringPtr=='B') (*stringPtr)++; + } + *value = result; + return 0; +} + +/*! readU32FromChar() : + * @return : unsigned integer value read from input in `char` format. + * allows and interprets K, KB, KiB, M, MB and MiB suffix. + * Will also modify `*stringPtr`, advancing it to position where it stopped reading. + * Note : function will exit() program if digit sequence overflows */ +static unsigned readU32FromChar(const char** stringPtr) { + static const char errorMsg[] = "error: numeric value overflows 32-bit unsigned int"; + unsigned result; + if (readU32FromCharChecked(stringPtr, &result)) { errorOut(errorMsg); } + return result; +} + /*! readIntFromChar() : * @return : signed integer value read from input in `char` format. * allows and interprets K, KB, KiB, M, MB and MiB suffix. @@ -372,281 +372,281 @@ static int readIntFromChar(const char** stringPtr) { return (int) result * sign; } -/*! readSizeTFromCharChecked() : - * @return 0 if success, and store the result in *value. - * allows and interprets K, KB, KiB, M, MB and MiB suffix. - * Will also modify `*stringPtr`, advancing it to position where it stopped reading. - * @return 1 if an overflow error occurs */ -static int readSizeTFromCharChecked(const char** stringPtr, size_t* value) -{ - size_t result = 0; - while ((**stringPtr >='0') && (**stringPtr <='9')) { - size_t const max = ((size_t)(-1)) / 10; - size_t last = result; - if (result > max) return 1; /* overflow error */ - result *= 10; - result += (size_t)(**stringPtr - '0'); - if (result < last) return 1; /* overflow error */ - (*stringPtr)++ ; - } - if ((**stringPtr=='K') || (**stringPtr=='M')) { - size_t const maxK = ((size_t)(-1)) >> 10; - if (result > maxK) return 1; /* overflow error */ - result <<= 10; - if (**stringPtr=='M') { - if (result > maxK) return 1; /* overflow error */ - result <<= 10; - } - (*stringPtr)++; /* skip `K` or `M` */ - if (**stringPtr=='i') (*stringPtr)++; - if (**stringPtr=='B') (*stringPtr)++; - } - *value = result; - return 0; -} - -/*! readSizeTFromChar() : - * @return : size_t value read from input in `char` format. - * allows and interprets K, KB, KiB, M, MB and MiB suffix. - * Will also modify `*stringPtr`, advancing it to position where it stopped reading. - * Note : function will exit() program if digit sequence overflows */ -static size_t readSizeTFromChar(const char** stringPtr) { - static const char errorMsg[] = "error: numeric value overflows size_t"; - size_t result; - if (readSizeTFromCharChecked(stringPtr, &result)) { errorOut(errorMsg); } - return result; -} - -/** longCommandWArg() : - * check if *stringPtr is the same as longCommand. - * If yes, @return 1 and advances *stringPtr to the position which immediately follows longCommand. - * @return 0 and doesn't modify *stringPtr otherwise. - */ -static int longCommandWArg(const char** stringPtr, const char* longCommand) -{ - size_t const comSize = strlen(longCommand); - int const result = !strncmp(*stringPtr, longCommand, comSize); - if (result) *stringPtr += comSize; - return result; -} - - -#ifndef ZSTD_NODICT - -static const unsigned kDefaultRegression = 1; -/** - * parseCoverParameters() : - * reads cover parameters from *stringPtr (e.g. "--train-cover=k=48,d=8,steps=32") into *params - * @return 1 means that cover parameters were correct - * @return 0 in case of malformed parameters - */ -static unsigned parseCoverParameters(const char* stringPtr, ZDICT_cover_params_t* params) -{ - memset(params, 0, sizeof(*params)); - for (; ;) { - if (longCommandWArg(&stringPtr, "k=")) { params->k = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "d=")) { params->d = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "steps=")) { params->steps = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "split=")) { - unsigned splitPercentage = readU32FromChar(&stringPtr); - params->splitPoint = (double)splitPercentage / 100.0; - if (stringPtr[0]==',') { stringPtr++; continue; } else break; - } - if (longCommandWArg(&stringPtr, "shrink")) { - params->shrinkDictMaxRegression = kDefaultRegression; - params->shrinkDict = 1; - if (stringPtr[0]=='=') { - stringPtr++; - params->shrinkDictMaxRegression = readU32FromChar(&stringPtr); - } - if (stringPtr[0]==',') { - stringPtr++; - continue; - } - else break; - } - return 0; - } - if (stringPtr[0] != 0) return 0; - DISPLAYLEVEL(4, "cover: k=%u\nd=%u\nsteps=%u\nsplit=%u\nshrink%u\n", params->k, params->d, params->steps, (unsigned)(params->splitPoint * 100), params->shrinkDictMaxRegression); - return 1; -} - -/** - * parseFastCoverParameters() : - * reads fastcover parameters from *stringPtr (e.g. "--train-fastcover=k=48,d=8,f=20,steps=32,accel=2") into *params - * @return 1 means that fastcover parameters were correct - * @return 0 in case of malformed parameters - */ -static unsigned parseFastCoverParameters(const char* stringPtr, ZDICT_fastCover_params_t* params) -{ - memset(params, 0, sizeof(*params)); - for (; ;) { - if (longCommandWArg(&stringPtr, "k=")) { params->k = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "d=")) { params->d = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "f=")) { params->f = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "steps=")) { params->steps = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "accel=")) { params->accel = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "split=")) { - unsigned splitPercentage = readU32FromChar(&stringPtr); - params->splitPoint = (double)splitPercentage / 100.0; - if (stringPtr[0]==',') { stringPtr++; continue; } else break; - } - if (longCommandWArg(&stringPtr, "shrink")) { - params->shrinkDictMaxRegression = kDefaultRegression; - params->shrinkDict = 1; - if (stringPtr[0]=='=') { - stringPtr++; - params->shrinkDictMaxRegression = readU32FromChar(&stringPtr); - } - if (stringPtr[0]==',') { - stringPtr++; - continue; - } - else break; - } - return 0; - } - if (stringPtr[0] != 0) return 0; - DISPLAYLEVEL(4, "cover: k=%u\nd=%u\nf=%u\nsteps=%u\nsplit=%u\naccel=%u\nshrink=%u\n", params->k, params->d, params->f, params->steps, (unsigned)(params->splitPoint * 100), params->accel, params->shrinkDictMaxRegression); - return 1; -} - -/** - * parseLegacyParameters() : - * reads legacy dictionary builder parameters from *stringPtr (e.g. "--train-legacy=selectivity=8") into *selectivity - * @return 1 means that legacy dictionary builder parameters were correct - * @return 0 in case of malformed parameters - */ -static unsigned parseLegacyParameters(const char* stringPtr, unsigned* selectivity) -{ - if (!longCommandWArg(&stringPtr, "s=") && !longCommandWArg(&stringPtr, "selectivity=")) { return 0; } - *selectivity = readU32FromChar(&stringPtr); - if (stringPtr[0] != 0) return 0; - DISPLAYLEVEL(4, "legacy: selectivity=%u\n", *selectivity); - return 1; -} - -static ZDICT_cover_params_t defaultCoverParams(void) -{ - ZDICT_cover_params_t params; - memset(¶ms, 0, sizeof(params)); - params.d = 8; - params.steps = 4; - params.splitPoint = 1.0; - params.shrinkDict = 0; - params.shrinkDictMaxRegression = kDefaultRegression; - return params; -} - -static ZDICT_fastCover_params_t defaultFastCoverParams(void) -{ - ZDICT_fastCover_params_t params; - memset(¶ms, 0, sizeof(params)); - params.d = 8; - params.f = 20; - params.steps = 4; - params.splitPoint = 0.75; /* different from default splitPoint of cover */ - params.accel = DEFAULT_ACCEL; - params.shrinkDict = 0; - params.shrinkDictMaxRegression = kDefaultRegression; - return params; -} -#endif - - -/** parseAdaptParameters() : - * reads adapt parameters from *stringPtr (e.g. "--zstd=min=1,max=19) and store them into adaptMinPtr and adaptMaxPtr. - * Both adaptMinPtr and adaptMaxPtr must be already allocated and correctly initialized. - * There is no guarantee that any of these values will be updated. - * @return 1 means that parsing was successful, - * @return 0 in case of malformed parameters - */ -static unsigned parseAdaptParameters(const char* stringPtr, int* adaptMinPtr, int* adaptMaxPtr) -{ - for ( ; ;) { +/*! readSizeTFromCharChecked() : + * @return 0 if success, and store the result in *value. + * allows and interprets K, KB, KiB, M, MB and MiB suffix. + * Will also modify `*stringPtr`, advancing it to position where it stopped reading. + * @return 1 if an overflow error occurs */ +static int readSizeTFromCharChecked(const char** stringPtr, size_t* value) +{ + size_t result = 0; + while ((**stringPtr >='0') && (**stringPtr <='9')) { + size_t const max = ((size_t)(-1)) / 10; + size_t last = result; + if (result > max) return 1; /* overflow error */ + result *= 10; + result += (size_t)(**stringPtr - '0'); + if (result < last) return 1; /* overflow error */ + (*stringPtr)++ ; + } + if ((**stringPtr=='K') || (**stringPtr=='M')) { + size_t const maxK = ((size_t)(-1)) >> 10; + if (result > maxK) return 1; /* overflow error */ + result <<= 10; + if (**stringPtr=='M') { + if (result > maxK) return 1; /* overflow error */ + result <<= 10; + } + (*stringPtr)++; /* skip `K` or `M` */ + if (**stringPtr=='i') (*stringPtr)++; + if (**stringPtr=='B') (*stringPtr)++; + } + *value = result; + return 0; +} + +/*! readSizeTFromChar() : + * @return : size_t value read from input in `char` format. + * allows and interprets K, KB, KiB, M, MB and MiB suffix. + * Will also modify `*stringPtr`, advancing it to position where it stopped reading. + * Note : function will exit() program if digit sequence overflows */ +static size_t readSizeTFromChar(const char** stringPtr) { + static const char errorMsg[] = "error: numeric value overflows size_t"; + size_t result; + if (readSizeTFromCharChecked(stringPtr, &result)) { errorOut(errorMsg); } + return result; +} + +/** longCommandWArg() : + * check if *stringPtr is the same as longCommand. + * If yes, @return 1 and advances *stringPtr to the position which immediately follows longCommand. + * @return 0 and doesn't modify *stringPtr otherwise. + */ +static int longCommandWArg(const char** stringPtr, const char* longCommand) +{ + size_t const comSize = strlen(longCommand); + int const result = !strncmp(*stringPtr, longCommand, comSize); + if (result) *stringPtr += comSize; + return result; +} + + +#ifndef ZSTD_NODICT + +static const unsigned kDefaultRegression = 1; +/** + * parseCoverParameters() : + * reads cover parameters from *stringPtr (e.g. "--train-cover=k=48,d=8,steps=32") into *params + * @return 1 means that cover parameters were correct + * @return 0 in case of malformed parameters + */ +static unsigned parseCoverParameters(const char* stringPtr, ZDICT_cover_params_t* params) +{ + memset(params, 0, sizeof(*params)); + for (; ;) { + if (longCommandWArg(&stringPtr, "k=")) { params->k = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "d=")) { params->d = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "steps=")) { params->steps = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "split=")) { + unsigned splitPercentage = readU32FromChar(&stringPtr); + params->splitPoint = (double)splitPercentage / 100.0; + if (stringPtr[0]==',') { stringPtr++; continue; } else break; + } + if (longCommandWArg(&stringPtr, "shrink")) { + params->shrinkDictMaxRegression = kDefaultRegression; + params->shrinkDict = 1; + if (stringPtr[0]=='=') { + stringPtr++; + params->shrinkDictMaxRegression = readU32FromChar(&stringPtr); + } + if (stringPtr[0]==',') { + stringPtr++; + continue; + } + else break; + } + return 0; + } + if (stringPtr[0] != 0) return 0; + DISPLAYLEVEL(4, "cover: k=%u\nd=%u\nsteps=%u\nsplit=%u\nshrink%u\n", params->k, params->d, params->steps, (unsigned)(params->splitPoint * 100), params->shrinkDictMaxRegression); + return 1; +} + +/** + * parseFastCoverParameters() : + * reads fastcover parameters from *stringPtr (e.g. "--train-fastcover=k=48,d=8,f=20,steps=32,accel=2") into *params + * @return 1 means that fastcover parameters were correct + * @return 0 in case of malformed parameters + */ +static unsigned parseFastCoverParameters(const char* stringPtr, ZDICT_fastCover_params_t* params) +{ + memset(params, 0, sizeof(*params)); + for (; ;) { + if (longCommandWArg(&stringPtr, "k=")) { params->k = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "d=")) { params->d = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "f=")) { params->f = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "steps=")) { params->steps = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "accel=")) { params->accel = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "split=")) { + unsigned splitPercentage = readU32FromChar(&stringPtr); + params->splitPoint = (double)splitPercentage / 100.0; + if (stringPtr[0]==',') { stringPtr++; continue; } else break; + } + if (longCommandWArg(&stringPtr, "shrink")) { + params->shrinkDictMaxRegression = kDefaultRegression; + params->shrinkDict = 1; + if (stringPtr[0]=='=') { + stringPtr++; + params->shrinkDictMaxRegression = readU32FromChar(&stringPtr); + } + if (stringPtr[0]==',') { + stringPtr++; + continue; + } + else break; + } + return 0; + } + if (stringPtr[0] != 0) return 0; + DISPLAYLEVEL(4, "cover: k=%u\nd=%u\nf=%u\nsteps=%u\nsplit=%u\naccel=%u\nshrink=%u\n", params->k, params->d, params->f, params->steps, (unsigned)(params->splitPoint * 100), params->accel, params->shrinkDictMaxRegression); + return 1; +} + +/** + * parseLegacyParameters() : + * reads legacy dictionary builder parameters from *stringPtr (e.g. "--train-legacy=selectivity=8") into *selectivity + * @return 1 means that legacy dictionary builder parameters were correct + * @return 0 in case of malformed parameters + */ +static unsigned parseLegacyParameters(const char* stringPtr, unsigned* selectivity) +{ + if (!longCommandWArg(&stringPtr, "s=") && !longCommandWArg(&stringPtr, "selectivity=")) { return 0; } + *selectivity = readU32FromChar(&stringPtr); + if (stringPtr[0] != 0) return 0; + DISPLAYLEVEL(4, "legacy: selectivity=%u\n", *selectivity); + return 1; +} + +static ZDICT_cover_params_t defaultCoverParams(void) +{ + ZDICT_cover_params_t params; + memset(¶ms, 0, sizeof(params)); + params.d = 8; + params.steps = 4; + params.splitPoint = 1.0; + params.shrinkDict = 0; + params.shrinkDictMaxRegression = kDefaultRegression; + return params; +} + +static ZDICT_fastCover_params_t defaultFastCoverParams(void) +{ + ZDICT_fastCover_params_t params; + memset(¶ms, 0, sizeof(params)); + params.d = 8; + params.f = 20; + params.steps = 4; + params.splitPoint = 0.75; /* different from default splitPoint of cover */ + params.accel = DEFAULT_ACCEL; + params.shrinkDict = 0; + params.shrinkDictMaxRegression = kDefaultRegression; + return params; +} +#endif + + +/** parseAdaptParameters() : + * reads adapt parameters from *stringPtr (e.g. "--zstd=min=1,max=19) and store them into adaptMinPtr and adaptMaxPtr. + * Both adaptMinPtr and adaptMaxPtr must be already allocated and correctly initialized. + * There is no guarantee that any of these values will be updated. + * @return 1 means that parsing was successful, + * @return 0 in case of malformed parameters + */ +static unsigned parseAdaptParameters(const char* stringPtr, int* adaptMinPtr, int* adaptMaxPtr) +{ + for ( ; ;) { if (longCommandWArg(&stringPtr, "min=")) { *adaptMinPtr = readIntFromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } if (longCommandWArg(&stringPtr, "max=")) { *adaptMaxPtr = readIntFromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - DISPLAYLEVEL(4, "invalid compression parameter \n"); - return 0; - } - if (stringPtr[0] != 0) return 0; /* check the end of string */ - if (*adaptMinPtr > *adaptMaxPtr) { - DISPLAYLEVEL(4, "incoherent adaptation limits \n"); - return 0; - } - return 1; -} - - -/** parseCompressionParameters() : - * reads compression parameters from *stringPtr (e.g. "--zstd=wlog=23,clog=23,hlog=22,slog=6,mml=3,tlen=48,strat=6") into *params - * @return 1 means that compression parameters were correct - * @return 0 in case of malformed parameters - */ -static unsigned parseCompressionParameters(const char* stringPtr, ZSTD_compressionParameters* params) -{ - for ( ; ;) { - if (longCommandWArg(&stringPtr, "windowLog=") || longCommandWArg(&stringPtr, "wlog=")) { params->windowLog = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "chainLog=") || longCommandWArg(&stringPtr, "clog=")) { params->chainLog = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "hashLog=") || longCommandWArg(&stringPtr, "hlog=")) { params->hashLog = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "searchLog=") || longCommandWArg(&stringPtr, "slog=")) { params->searchLog = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "minMatch=") || longCommandWArg(&stringPtr, "mml=")) { params->minMatch = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "targetLength=") || longCommandWArg(&stringPtr, "tlen=")) { params->targetLength = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "strategy=") || longCommandWArg(&stringPtr, "strat=")) { params->strategy = (ZSTD_strategy)(readU32FromChar(&stringPtr)); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "overlapLog=") || longCommandWArg(&stringPtr, "ovlog=")) { g_overlapLog = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "ldmHashLog=") || longCommandWArg(&stringPtr, "lhlog=")) { g_ldmHashLog = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "ldmMinMatch=") || longCommandWArg(&stringPtr, "lmml=")) { g_ldmMinMatch = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "ldmBucketSizeLog=") || longCommandWArg(&stringPtr, "lblog=")) { g_ldmBucketSizeLog = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - if (longCommandWArg(&stringPtr, "ldmHashRateLog=") || longCommandWArg(&stringPtr, "lhrlog=")) { g_ldmHashRateLog = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } - DISPLAYLEVEL(4, "invalid compression parameter \n"); - return 0; - } - - DISPLAYLEVEL(4, "windowLog=%d, chainLog=%d, hashLog=%d, searchLog=%d \n", params->windowLog, params->chainLog, params->hashLog, params->searchLog); - DISPLAYLEVEL(4, "minMatch=%d, targetLength=%d, strategy=%d \n", params->minMatch, params->targetLength, params->strategy); - if (stringPtr[0] != 0) return 0; /* check the end of string */ - return 1; -} - -static void printVersion(void) -{ - if (g_displayLevel < DISPLAY_LEVEL_DEFAULT) { - DISPLAYOUT("%s\n", ZSTD_VERSION_STRING); - return; - } - - DISPLAYOUT(WELCOME_MESSAGE); - if (g_displayLevel >= 3) { - /* format support */ - DISPLAYOUT("*** supports: zstd"); - #if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>0) && (ZSTD_LEGACY_SUPPORT<8) - DISPLAYOUT(", zstd legacy v0.%d+", ZSTD_LEGACY_SUPPORT); - #endif - #ifdef ZSTD_GZCOMPRESS - DISPLAYOUT(", gzip"); - #endif - #ifdef ZSTD_LZ4COMPRESS - DISPLAYOUT(", lz4"); - #endif - #ifdef ZSTD_LZMACOMPRESS - DISPLAYOUT(", lzma, xz "); - #endif - DISPLAYOUT("\n"); - if (g_displayLevel >= 4) { - /* posix support */ - #ifdef _POSIX_C_SOURCE - DISPLAYOUT("_POSIX_C_SOURCE defined: %ldL\n", (long) _POSIX_C_SOURCE); - #endif - #ifdef _POSIX_VERSION - DISPLAYOUT("_POSIX_VERSION defined: %ldL \n", (long) _POSIX_VERSION); - #endif - #ifdef PLATFORM_POSIX_VERSION - DISPLAYOUT("PLATFORM_POSIX_VERSION defined: %ldL\n", (long) PLATFORM_POSIX_VERSION); - #endif - } } -} - + DISPLAYLEVEL(4, "invalid compression parameter \n"); + return 0; + } + if (stringPtr[0] != 0) return 0; /* check the end of string */ + if (*adaptMinPtr > *adaptMaxPtr) { + DISPLAYLEVEL(4, "incoherent adaptation limits \n"); + return 0; + } + return 1; +} + + +/** parseCompressionParameters() : + * reads compression parameters from *stringPtr (e.g. "--zstd=wlog=23,clog=23,hlog=22,slog=6,mml=3,tlen=48,strat=6") into *params + * @return 1 means that compression parameters were correct + * @return 0 in case of malformed parameters + */ +static unsigned parseCompressionParameters(const char* stringPtr, ZSTD_compressionParameters* params) +{ + for ( ; ;) { + if (longCommandWArg(&stringPtr, "windowLog=") || longCommandWArg(&stringPtr, "wlog=")) { params->windowLog = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "chainLog=") || longCommandWArg(&stringPtr, "clog=")) { params->chainLog = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "hashLog=") || longCommandWArg(&stringPtr, "hlog=")) { params->hashLog = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "searchLog=") || longCommandWArg(&stringPtr, "slog=")) { params->searchLog = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "minMatch=") || longCommandWArg(&stringPtr, "mml=")) { params->minMatch = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "targetLength=") || longCommandWArg(&stringPtr, "tlen=")) { params->targetLength = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "strategy=") || longCommandWArg(&stringPtr, "strat=")) { params->strategy = (ZSTD_strategy)(readU32FromChar(&stringPtr)); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "overlapLog=") || longCommandWArg(&stringPtr, "ovlog=")) { g_overlapLog = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "ldmHashLog=") || longCommandWArg(&stringPtr, "lhlog=")) { g_ldmHashLog = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "ldmMinMatch=") || longCommandWArg(&stringPtr, "lmml=")) { g_ldmMinMatch = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "ldmBucketSizeLog=") || longCommandWArg(&stringPtr, "lblog=")) { g_ldmBucketSizeLog = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + if (longCommandWArg(&stringPtr, "ldmHashRateLog=") || longCommandWArg(&stringPtr, "lhrlog=")) { g_ldmHashRateLog = readU32FromChar(&stringPtr); if (stringPtr[0]==',') { stringPtr++; continue; } else break; } + DISPLAYLEVEL(4, "invalid compression parameter \n"); + return 0; + } + + DISPLAYLEVEL(4, "windowLog=%d, chainLog=%d, hashLog=%d, searchLog=%d \n", params->windowLog, params->chainLog, params->hashLog, params->searchLog); + DISPLAYLEVEL(4, "minMatch=%d, targetLength=%d, strategy=%d \n", params->minMatch, params->targetLength, params->strategy); + if (stringPtr[0] != 0) return 0; /* check the end of string */ + return 1; +} + +static void printVersion(void) +{ + if (g_displayLevel < DISPLAY_LEVEL_DEFAULT) { + DISPLAYOUT("%s\n", ZSTD_VERSION_STRING); + return; + } + + DISPLAYOUT(WELCOME_MESSAGE); + if (g_displayLevel >= 3) { + /* format support */ + DISPLAYOUT("*** supports: zstd"); + #if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>0) && (ZSTD_LEGACY_SUPPORT<8) + DISPLAYOUT(", zstd legacy v0.%d+", ZSTD_LEGACY_SUPPORT); + #endif + #ifdef ZSTD_GZCOMPRESS + DISPLAYOUT(", gzip"); + #endif + #ifdef ZSTD_LZ4COMPRESS + DISPLAYOUT(", lz4"); + #endif + #ifdef ZSTD_LZMACOMPRESS + DISPLAYOUT(", lzma, xz "); + #endif + DISPLAYOUT("\n"); + if (g_displayLevel >= 4) { + /* posix support */ + #ifdef _POSIX_C_SOURCE + DISPLAYOUT("_POSIX_C_SOURCE defined: %ldL\n", (long) _POSIX_C_SOURCE); + #endif + #ifdef _POSIX_VERSION + DISPLAYOUT("_POSIX_VERSION defined: %ldL \n", (long) _POSIX_VERSION); + #endif + #ifdef PLATFORM_POSIX_VERSION + DISPLAYOUT("PLATFORM_POSIX_VERSION defined: %ldL\n", (long) PLATFORM_POSIX_VERSION); + #endif + } } +} + #define ZSTD_NB_STRATEGIES 9 static const char* ZSTD_strategyMap[ZSTD_NB_STRATEGIES + 1] = { "", "ZSTD_fast", "ZSTD_dfast", "ZSTD_greedy", "ZSTD_lazy", "ZSTD_lazy2", "ZSTD_btlazy2", @@ -689,308 +689,308 @@ static void printActualCParams(const char* filename, const char* dictFileName, i #endif -/* Environment variables for parameter setting */ -#define ENV_CLEVEL "ZSTD_CLEVEL" -#define ENV_NBTHREADS "ZSTD_NBTHREADS" /* takes lower precedence than directly specifying -T# in the CLI */ - -/* pick up environment variable */ -static int init_cLevel(void) { - const char* const env = getenv(ENV_CLEVEL); - if (env != NULL) { - const char* ptr = env; - int sign = 1; - if (*ptr == '-') { - sign = -1; - ptr++; - } else if (*ptr == '+') { - ptr++; - } - - if ((*ptr>='0') && (*ptr<='9')) { - unsigned absLevel; - if (readU32FromCharChecked(&ptr, &absLevel)) { - DISPLAYLEVEL(2, "Ignore environment variable setting %s=%s: numeric value too large \n", ENV_CLEVEL, env); - return ZSTDCLI_CLEVEL_DEFAULT; - } else if (*ptr == 0) { - return sign * (int)absLevel; - } } - - DISPLAYLEVEL(2, "Ignore environment variable setting %s=%s: not a valid integer value \n", ENV_CLEVEL, env); - } - - return ZSTDCLI_CLEVEL_DEFAULT; -} - -#ifdef ZSTD_MULTITHREAD -static unsigned init_nbThreads(void) { - const char* const env = getenv(ENV_NBTHREADS); - if (env != NULL) { - const char* ptr = env; - if ((*ptr>='0') && (*ptr<='9')) { - unsigned nbThreads; - if (readU32FromCharChecked(&ptr, &nbThreads)) { - DISPLAYLEVEL(2, "Ignore environment variable setting %s=%s: numeric value too large \n", ENV_NBTHREADS, env); - return ZSTDCLI_NBTHREADS_DEFAULT; - } else if (*ptr == 0) { - return nbThreads; - } - } - DISPLAYLEVEL(2, "Ignore environment variable setting %s=%s: not a valid unsigned value \n", ENV_NBTHREADS, env); - } - - return ZSTDCLI_NBTHREADS_DEFAULT; -} -#endif - -#define NEXT_FIELD(ptr) { \ - if (*argument == '=') { \ - ptr = ++argument; \ - argument += strlen(ptr); \ - } else { \ - argNb++; \ - if (argNb >= argCount) { \ - DISPLAY("error: missing command argument \n"); \ - CLEAN_RETURN(1); \ - } \ - ptr = argv[argNb]; \ - assert(ptr != NULL); \ - if (ptr[0]=='-') { \ - DISPLAY("error: command cannot be separated from its argument by another command \n"); \ - CLEAN_RETURN(1); \ -} } } - -#define NEXT_UINT32(val32) { \ - const char* __nb; \ - NEXT_FIELD(__nb); \ - val32 = readU32FromChar(&__nb); \ -} - -typedef enum { zom_compress, zom_decompress, zom_test, zom_bench, zom_train, zom_list } zstd_operation_mode; - -#define CLEAN_RETURN(i) { operationResult = (i); goto _end; } - -#ifdef ZSTD_NOCOMPRESS -/* symbols from compression library are not defined and should not be invoked */ -# define MINCLEVEL -99 -# define MAXCLEVEL 22 -#else -# define MINCLEVEL ZSTD_minCLevel() -# define MAXCLEVEL ZSTD_maxCLevel() -#endif - +/* Environment variables for parameter setting */ +#define ENV_CLEVEL "ZSTD_CLEVEL" +#define ENV_NBTHREADS "ZSTD_NBTHREADS" /* takes lower precedence than directly specifying -T# in the CLI */ + +/* pick up environment variable */ +static int init_cLevel(void) { + const char* const env = getenv(ENV_CLEVEL); + if (env != NULL) { + const char* ptr = env; + int sign = 1; + if (*ptr == '-') { + sign = -1; + ptr++; + } else if (*ptr == '+') { + ptr++; + } + + if ((*ptr>='0') && (*ptr<='9')) { + unsigned absLevel; + if (readU32FromCharChecked(&ptr, &absLevel)) { + DISPLAYLEVEL(2, "Ignore environment variable setting %s=%s: numeric value too large \n", ENV_CLEVEL, env); + return ZSTDCLI_CLEVEL_DEFAULT; + } else if (*ptr == 0) { + return sign * (int)absLevel; + } } + + DISPLAYLEVEL(2, "Ignore environment variable setting %s=%s: not a valid integer value \n", ENV_CLEVEL, env); + } + + return ZSTDCLI_CLEVEL_DEFAULT; +} + +#ifdef ZSTD_MULTITHREAD +static unsigned init_nbThreads(void) { + const char* const env = getenv(ENV_NBTHREADS); + if (env != NULL) { + const char* ptr = env; + if ((*ptr>='0') && (*ptr<='9')) { + unsigned nbThreads; + if (readU32FromCharChecked(&ptr, &nbThreads)) { + DISPLAYLEVEL(2, "Ignore environment variable setting %s=%s: numeric value too large \n", ENV_NBTHREADS, env); + return ZSTDCLI_NBTHREADS_DEFAULT; + } else if (*ptr == 0) { + return nbThreads; + } + } + DISPLAYLEVEL(2, "Ignore environment variable setting %s=%s: not a valid unsigned value \n", ENV_NBTHREADS, env); + } + + return ZSTDCLI_NBTHREADS_DEFAULT; +} +#endif + +#define NEXT_FIELD(ptr) { \ + if (*argument == '=') { \ + ptr = ++argument; \ + argument += strlen(ptr); \ + } else { \ + argNb++; \ + if (argNb >= argCount) { \ + DISPLAY("error: missing command argument \n"); \ + CLEAN_RETURN(1); \ + } \ + ptr = argv[argNb]; \ + assert(ptr != NULL); \ + if (ptr[0]=='-') { \ + DISPLAY("error: command cannot be separated from its argument by another command \n"); \ + CLEAN_RETURN(1); \ +} } } + +#define NEXT_UINT32(val32) { \ + const char* __nb; \ + NEXT_FIELD(__nb); \ + val32 = readU32FromChar(&__nb); \ +} + +typedef enum { zom_compress, zom_decompress, zom_test, zom_bench, zom_train, zom_list } zstd_operation_mode; + +#define CLEAN_RETURN(i) { operationResult = (i); goto _end; } + +#ifdef ZSTD_NOCOMPRESS +/* symbols from compression library are not defined and should not be invoked */ +# define MINCLEVEL -99 +# define MAXCLEVEL 22 +#else +# define MINCLEVEL ZSTD_minCLevel() +# define MAXCLEVEL ZSTD_maxCLevel() +#endif + int main(int argCount, const char* argv[]) -{ - int argNb, - followLinks = 0, - allowBlockDevices = 0, - forceStdin = 0, - forceStdout = 0, - hasStdout = 0, - ldmFlag = 0, - main_pause = 0, - nbWorkers = 0, - adapt = 0, - useRowMatchFinder = 0, - adaptMin = MINCLEVEL, - adaptMax = MAXCLEVEL, - rsyncable = 0, - nextArgumentsAreFiles = 0, - operationResult = 0, - separateFiles = 0, - setRealTimePrio = 0, - singleThread = 0, +{ + int argNb, + followLinks = 0, + allowBlockDevices = 0, + forceStdin = 0, + forceStdout = 0, + hasStdout = 0, + ldmFlag = 0, + main_pause = 0, + nbWorkers = 0, + adapt = 0, + useRowMatchFinder = 0, + adaptMin = MINCLEVEL, + adaptMax = MAXCLEVEL, + rsyncable = 0, + nextArgumentsAreFiles = 0, + operationResult = 0, + separateFiles = 0, + setRealTimePrio = 0, + singleThread = 0, #ifdef ZSTD_MULTITHREAD defaultLogicalCores = 0, #endif - showDefaultCParams = 0, - ultra=0, - contentSize=1; - double compressibility = 0.5; - unsigned bench_nbSeconds = 3; /* would be better if this value was synchronized from bench */ - size_t blockSize = 0; - - FIO_prefs_t* const prefs = FIO_createPreferences(); - FIO_ctx_t* const fCtx = FIO_createContext(); - zstd_operation_mode operation = zom_compress; - ZSTD_compressionParameters compressionParams; - int cLevel = init_cLevel(); - int cLevelLast = MINCLEVEL - 1; /* lower than minimum */ - unsigned recursive = 0; - unsigned memLimit = 0; - FileNamesTable* filenames = UTIL_allocateFileNamesTable((size_t)argCount); /* argCount >= 1 */ - FileNamesTable* file_of_names = UTIL_allocateFileNamesTable((size_t)argCount); /* argCount >= 1 */ - const char* programName = argv[0]; - const char* outFileName = NULL; - const char* outDirName = NULL; - const char* outMirroredDirName = NULL; - const char* dictFileName = NULL; - const char* patchFromDictFileName = NULL; - const char* suffix = ZSTD_EXTENSION; - unsigned maxDictSize = g_defaultMaxDictSize; - unsigned dictID = 0; - size_t streamSrcSize = 0; - size_t targetCBlockSize = 0; - size_t srcSizeHint = 0; - int dictCLevel = g_defaultDictCLevel; - unsigned dictSelect = g_defaultSelectivityLevel; -#ifndef ZSTD_NODICT - ZDICT_cover_params_t coverParams = defaultCoverParams(); - ZDICT_fastCover_params_t fastCoverParams = defaultFastCoverParams(); - dictType dict = fastCover; -#endif -#ifndef ZSTD_NOBENCH - BMK_advancedParams_t benchParams = BMK_initAdvancedParams(); -#endif + showDefaultCParams = 0, + ultra=0, + contentSize=1; + double compressibility = 0.5; + unsigned bench_nbSeconds = 3; /* would be better if this value was synchronized from bench */ + size_t blockSize = 0; + + FIO_prefs_t* const prefs = FIO_createPreferences(); + FIO_ctx_t* const fCtx = FIO_createContext(); + zstd_operation_mode operation = zom_compress; + ZSTD_compressionParameters compressionParams; + int cLevel = init_cLevel(); + int cLevelLast = MINCLEVEL - 1; /* lower than minimum */ + unsigned recursive = 0; + unsigned memLimit = 0; + FileNamesTable* filenames = UTIL_allocateFileNamesTable((size_t)argCount); /* argCount >= 1 */ + FileNamesTable* file_of_names = UTIL_allocateFileNamesTable((size_t)argCount); /* argCount >= 1 */ + const char* programName = argv[0]; + const char* outFileName = NULL; + const char* outDirName = NULL; + const char* outMirroredDirName = NULL; + const char* dictFileName = NULL; + const char* patchFromDictFileName = NULL; + const char* suffix = ZSTD_EXTENSION; + unsigned maxDictSize = g_defaultMaxDictSize; + unsigned dictID = 0; + size_t streamSrcSize = 0; + size_t targetCBlockSize = 0; + size_t srcSizeHint = 0; + int dictCLevel = g_defaultDictCLevel; + unsigned dictSelect = g_defaultSelectivityLevel; +#ifndef ZSTD_NODICT + ZDICT_cover_params_t coverParams = defaultCoverParams(); + ZDICT_fastCover_params_t fastCoverParams = defaultFastCoverParams(); + dictType dict = fastCover; +#endif +#ifndef ZSTD_NOBENCH + BMK_advancedParams_t benchParams = BMK_initAdvancedParams(); +#endif ZSTD_paramSwitch_e literalCompressionMode = ZSTD_ps_auto; - - - /* init */ - checkLibVersion(); - (void)recursive; (void)cLevelLast; /* not used when ZSTD_NOBENCH set */ - (void)memLimit; - assert(argCount >= 1); - if ((filenames==NULL) || (file_of_names==NULL)) { DISPLAY("zstd: allocation error \n"); exit(1); } - programName = lastNameFromPath(programName); -#ifdef ZSTD_MULTITHREAD - nbWorkers = init_nbThreads(); -#endif - - /* preset behaviors */ - if (exeNameMatch(programName, ZSTD_ZSTDMT)) nbWorkers=0, singleThread=0; - if (exeNameMatch(programName, ZSTD_UNZSTD)) operation=zom_decompress; - if (exeNameMatch(programName, ZSTD_CAT)) { operation=zom_decompress; FIO_overwriteMode(prefs); forceStdout=1; followLinks=1; outFileName=stdoutmark; g_displayLevel=1; } /* supports multiple formats */ - if (exeNameMatch(programName, ZSTD_ZCAT)) { operation=zom_decompress; FIO_overwriteMode(prefs); forceStdout=1; followLinks=1; outFileName=stdoutmark; g_displayLevel=1; } /* behave like zcat, also supports multiple formats */ - if (exeNameMatch(programName, ZSTD_GZ)) { suffix = GZ_EXTENSION; FIO_setCompressionType(prefs, FIO_gzipCompression); FIO_setRemoveSrcFile(prefs, 1); } /* behave like gzip */ - if (exeNameMatch(programName, ZSTD_GUNZIP)) { operation=zom_decompress; FIO_setRemoveSrcFile(prefs, 1); } /* behave like gunzip, also supports multiple formats */ - if (exeNameMatch(programName, ZSTD_GZCAT)) { operation=zom_decompress; FIO_overwriteMode(prefs); forceStdout=1; followLinks=1; outFileName=stdoutmark; g_displayLevel=1; } /* behave like gzcat, also supports multiple formats */ - if (exeNameMatch(programName, ZSTD_LZMA)) { suffix = LZMA_EXTENSION; FIO_setCompressionType(prefs, FIO_lzmaCompression); FIO_setRemoveSrcFile(prefs, 1); } /* behave like lzma */ - if (exeNameMatch(programName, ZSTD_UNLZMA)) { operation=zom_decompress; FIO_setCompressionType(prefs, FIO_lzmaCompression); FIO_setRemoveSrcFile(prefs, 1); } /* behave like unlzma, also supports multiple formats */ - if (exeNameMatch(programName, ZSTD_XZ)) { suffix = XZ_EXTENSION; FIO_setCompressionType(prefs, FIO_xzCompression); FIO_setRemoveSrcFile(prefs, 1); } /* behave like xz */ - if (exeNameMatch(programName, ZSTD_UNXZ)) { operation=zom_decompress; FIO_setCompressionType(prefs, FIO_xzCompression); FIO_setRemoveSrcFile(prefs, 1); } /* behave like unxz, also supports multiple formats */ - if (exeNameMatch(programName, ZSTD_LZ4)) { suffix = LZ4_EXTENSION; FIO_setCompressionType(prefs, FIO_lz4Compression); } /* behave like lz4 */ - if (exeNameMatch(programName, ZSTD_UNLZ4)) { operation=zom_decompress; FIO_setCompressionType(prefs, FIO_lz4Compression); } /* behave like unlz4, also supports multiple formats */ - memset(&compressionParams, 0, sizeof(compressionParams)); - - /* init crash handler */ - FIO_addAbortHandler(); - - /* command switches */ - for (argNb=1; argNb<argCount; argNb++) { - const char* argument = argv[argNb]; - if (!argument) continue; /* Protection if argument empty */ - - if (nextArgumentsAreFiles) { - UTIL_refFilename(filenames, argument); - continue; - } - - /* "-" means stdin/stdout */ - if (!strcmp(argument, "-")){ - UTIL_refFilename(filenames, stdinmark); - continue; - } - - /* Decode commands (note : aggregated commands are allowed) */ - if (argument[0]=='-') { - - if (argument[1]=='-') { - /* long commands (--long-word) */ - if (!strcmp(argument, "--")) { nextArgumentsAreFiles=1; continue; } /* only file names allowed from now on */ - if (!strcmp(argument, "--list")) { operation=zom_list; continue; } - if (!strcmp(argument, "--compress")) { operation=zom_compress; continue; } - if (!strcmp(argument, "--decompress")) { operation=zom_decompress; continue; } - if (!strcmp(argument, "--uncompress")) { operation=zom_decompress; continue; } - if (!strcmp(argument, "--force")) { FIO_overwriteMode(prefs); forceStdin=1; forceStdout=1; followLinks=1; allowBlockDevices=1; continue; } - if (!strcmp(argument, "--version")) { printVersion(); CLEAN_RETURN(0); } - if (!strcmp(argument, "--help")) { usage_advanced(programName); CLEAN_RETURN(0); } - if (!strcmp(argument, "--verbose")) { g_displayLevel++; continue; } - if (!strcmp(argument, "--quiet")) { g_displayLevel--; continue; } - if (!strcmp(argument, "--stdout")) { forceStdout=1; outFileName=stdoutmark; g_displayLevel-=(g_displayLevel==2); continue; } - if (!strcmp(argument, "--ultra")) { ultra=1; continue; } - if (!strcmp(argument, "--check")) { FIO_setChecksumFlag(prefs, 2); continue; } - if (!strcmp(argument, "--no-check")) { FIO_setChecksumFlag(prefs, 0); continue; } - if (!strcmp(argument, "--sparse")) { FIO_setSparseWrite(prefs, 2); continue; } - if (!strcmp(argument, "--no-sparse")) { FIO_setSparseWrite(prefs, 0); continue; } - if (!strcmp(argument, "--test")) { operation=zom_test; continue; } - if (!strcmp(argument, "--train")) { operation=zom_train; if (outFileName==NULL) outFileName=g_defaultDictName; continue; } - if (!strcmp(argument, "--no-dictID")) { FIO_setDictIDFlag(prefs, 0); continue; } - if (!strcmp(argument, "--keep")) { FIO_setRemoveSrcFile(prefs, 0); continue; } - if (!strcmp(argument, "--rm")) { FIO_setRemoveSrcFile(prefs, 1); continue; } - if (!strcmp(argument, "--priority=rt")) { setRealTimePrio = 1; continue; } - if (!strcmp(argument, "--show-default-cparams")) { showDefaultCParams = 1; continue; } - if (!strcmp(argument, "--content-size")) { contentSize = 1; continue; } - if (!strcmp(argument, "--no-content-size")) { contentSize = 0; continue; } - if (!strcmp(argument, "--adapt")) { adapt = 1; continue; } - if (!strcmp(argument, "--no-row-match-finder")) { useRowMatchFinder = 1; continue; } - if (!strcmp(argument, "--row-match-finder")) { useRowMatchFinder = 2; continue; } - if (longCommandWArg(&argument, "--adapt=")) { adapt = 1; if (!parseAdaptParameters(argument, &adaptMin, &adaptMax)) { badusage(programName); CLEAN_RETURN(1); } continue; } - if (!strcmp(argument, "--single-thread")) { nbWorkers = 0; singleThread = 1; continue; } - if (!strcmp(argument, "--format=zstd")) { suffix = ZSTD_EXTENSION; FIO_setCompressionType(prefs, FIO_zstdCompression); continue; } -#ifdef ZSTD_GZCOMPRESS - if (!strcmp(argument, "--format=gzip")) { suffix = GZ_EXTENSION; FIO_setCompressionType(prefs, FIO_gzipCompression); continue; } -#endif -#ifdef ZSTD_LZMACOMPRESS - if (!strcmp(argument, "--format=lzma")) { suffix = LZMA_EXTENSION; FIO_setCompressionType(prefs, FIO_lzmaCompression); continue; } - if (!strcmp(argument, "--format=xz")) { suffix = XZ_EXTENSION; FIO_setCompressionType(prefs, FIO_xzCompression); continue; } -#endif -#ifdef ZSTD_LZ4COMPRESS - if (!strcmp(argument, "--format=lz4")) { suffix = LZ4_EXTENSION; FIO_setCompressionType(prefs, FIO_lz4Compression); continue; } -#endif - if (!strcmp(argument, "--rsyncable")) { rsyncable = 1; continue; } + + + /* init */ + checkLibVersion(); + (void)recursive; (void)cLevelLast; /* not used when ZSTD_NOBENCH set */ + (void)memLimit; + assert(argCount >= 1); + if ((filenames==NULL) || (file_of_names==NULL)) { DISPLAY("zstd: allocation error \n"); exit(1); } + programName = lastNameFromPath(programName); +#ifdef ZSTD_MULTITHREAD + nbWorkers = init_nbThreads(); +#endif + + /* preset behaviors */ + if (exeNameMatch(programName, ZSTD_ZSTDMT)) nbWorkers=0, singleThread=0; + if (exeNameMatch(programName, ZSTD_UNZSTD)) operation=zom_decompress; + if (exeNameMatch(programName, ZSTD_CAT)) { operation=zom_decompress; FIO_overwriteMode(prefs); forceStdout=1; followLinks=1; outFileName=stdoutmark; g_displayLevel=1; } /* supports multiple formats */ + if (exeNameMatch(programName, ZSTD_ZCAT)) { operation=zom_decompress; FIO_overwriteMode(prefs); forceStdout=1; followLinks=1; outFileName=stdoutmark; g_displayLevel=1; } /* behave like zcat, also supports multiple formats */ + if (exeNameMatch(programName, ZSTD_GZ)) { suffix = GZ_EXTENSION; FIO_setCompressionType(prefs, FIO_gzipCompression); FIO_setRemoveSrcFile(prefs, 1); } /* behave like gzip */ + if (exeNameMatch(programName, ZSTD_GUNZIP)) { operation=zom_decompress; FIO_setRemoveSrcFile(prefs, 1); } /* behave like gunzip, also supports multiple formats */ + if (exeNameMatch(programName, ZSTD_GZCAT)) { operation=zom_decompress; FIO_overwriteMode(prefs); forceStdout=1; followLinks=1; outFileName=stdoutmark; g_displayLevel=1; } /* behave like gzcat, also supports multiple formats */ + if (exeNameMatch(programName, ZSTD_LZMA)) { suffix = LZMA_EXTENSION; FIO_setCompressionType(prefs, FIO_lzmaCompression); FIO_setRemoveSrcFile(prefs, 1); } /* behave like lzma */ + if (exeNameMatch(programName, ZSTD_UNLZMA)) { operation=zom_decompress; FIO_setCompressionType(prefs, FIO_lzmaCompression); FIO_setRemoveSrcFile(prefs, 1); } /* behave like unlzma, also supports multiple formats */ + if (exeNameMatch(programName, ZSTD_XZ)) { suffix = XZ_EXTENSION; FIO_setCompressionType(prefs, FIO_xzCompression); FIO_setRemoveSrcFile(prefs, 1); } /* behave like xz */ + if (exeNameMatch(programName, ZSTD_UNXZ)) { operation=zom_decompress; FIO_setCompressionType(prefs, FIO_xzCompression); FIO_setRemoveSrcFile(prefs, 1); } /* behave like unxz, also supports multiple formats */ + if (exeNameMatch(programName, ZSTD_LZ4)) { suffix = LZ4_EXTENSION; FIO_setCompressionType(prefs, FIO_lz4Compression); } /* behave like lz4 */ + if (exeNameMatch(programName, ZSTD_UNLZ4)) { operation=zom_decompress; FIO_setCompressionType(prefs, FIO_lz4Compression); } /* behave like unlz4, also supports multiple formats */ + memset(&compressionParams, 0, sizeof(compressionParams)); + + /* init crash handler */ + FIO_addAbortHandler(); + + /* command switches */ + for (argNb=1; argNb<argCount; argNb++) { + const char* argument = argv[argNb]; + if (!argument) continue; /* Protection if argument empty */ + + if (nextArgumentsAreFiles) { + UTIL_refFilename(filenames, argument); + continue; + } + + /* "-" means stdin/stdout */ + if (!strcmp(argument, "-")){ + UTIL_refFilename(filenames, stdinmark); + continue; + } + + /* Decode commands (note : aggregated commands are allowed) */ + if (argument[0]=='-') { + + if (argument[1]=='-') { + /* long commands (--long-word) */ + if (!strcmp(argument, "--")) { nextArgumentsAreFiles=1; continue; } /* only file names allowed from now on */ + if (!strcmp(argument, "--list")) { operation=zom_list; continue; } + if (!strcmp(argument, "--compress")) { operation=zom_compress; continue; } + if (!strcmp(argument, "--decompress")) { operation=zom_decompress; continue; } + if (!strcmp(argument, "--uncompress")) { operation=zom_decompress; continue; } + if (!strcmp(argument, "--force")) { FIO_overwriteMode(prefs); forceStdin=1; forceStdout=1; followLinks=1; allowBlockDevices=1; continue; } + if (!strcmp(argument, "--version")) { printVersion(); CLEAN_RETURN(0); } + if (!strcmp(argument, "--help")) { usage_advanced(programName); CLEAN_RETURN(0); } + if (!strcmp(argument, "--verbose")) { g_displayLevel++; continue; } + if (!strcmp(argument, "--quiet")) { g_displayLevel--; continue; } + if (!strcmp(argument, "--stdout")) { forceStdout=1; outFileName=stdoutmark; g_displayLevel-=(g_displayLevel==2); continue; } + if (!strcmp(argument, "--ultra")) { ultra=1; continue; } + if (!strcmp(argument, "--check")) { FIO_setChecksumFlag(prefs, 2); continue; } + if (!strcmp(argument, "--no-check")) { FIO_setChecksumFlag(prefs, 0); continue; } + if (!strcmp(argument, "--sparse")) { FIO_setSparseWrite(prefs, 2); continue; } + if (!strcmp(argument, "--no-sparse")) { FIO_setSparseWrite(prefs, 0); continue; } + if (!strcmp(argument, "--test")) { operation=zom_test; continue; } + if (!strcmp(argument, "--train")) { operation=zom_train; if (outFileName==NULL) outFileName=g_defaultDictName; continue; } + if (!strcmp(argument, "--no-dictID")) { FIO_setDictIDFlag(prefs, 0); continue; } + if (!strcmp(argument, "--keep")) { FIO_setRemoveSrcFile(prefs, 0); continue; } + if (!strcmp(argument, "--rm")) { FIO_setRemoveSrcFile(prefs, 1); continue; } + if (!strcmp(argument, "--priority=rt")) { setRealTimePrio = 1; continue; } + if (!strcmp(argument, "--show-default-cparams")) { showDefaultCParams = 1; continue; } + if (!strcmp(argument, "--content-size")) { contentSize = 1; continue; } + if (!strcmp(argument, "--no-content-size")) { contentSize = 0; continue; } + if (!strcmp(argument, "--adapt")) { adapt = 1; continue; } + if (!strcmp(argument, "--no-row-match-finder")) { useRowMatchFinder = 1; continue; } + if (!strcmp(argument, "--row-match-finder")) { useRowMatchFinder = 2; continue; } + if (longCommandWArg(&argument, "--adapt=")) { adapt = 1; if (!parseAdaptParameters(argument, &adaptMin, &adaptMax)) { badusage(programName); CLEAN_RETURN(1); } continue; } + if (!strcmp(argument, "--single-thread")) { nbWorkers = 0; singleThread = 1; continue; } + if (!strcmp(argument, "--format=zstd")) { suffix = ZSTD_EXTENSION; FIO_setCompressionType(prefs, FIO_zstdCompression); continue; } +#ifdef ZSTD_GZCOMPRESS + if (!strcmp(argument, "--format=gzip")) { suffix = GZ_EXTENSION; FIO_setCompressionType(prefs, FIO_gzipCompression); continue; } +#endif +#ifdef ZSTD_LZMACOMPRESS + if (!strcmp(argument, "--format=lzma")) { suffix = LZMA_EXTENSION; FIO_setCompressionType(prefs, FIO_lzmaCompression); continue; } + if (!strcmp(argument, "--format=xz")) { suffix = XZ_EXTENSION; FIO_setCompressionType(prefs, FIO_xzCompression); continue; } +#endif +#ifdef ZSTD_LZ4COMPRESS + if (!strcmp(argument, "--format=lz4")) { suffix = LZ4_EXTENSION; FIO_setCompressionType(prefs, FIO_lz4Compression); continue; } +#endif + if (!strcmp(argument, "--rsyncable")) { rsyncable = 1; continue; } if (!strcmp(argument, "--compress-literals")) { literalCompressionMode = ZSTD_ps_enable; continue; } if (!strcmp(argument, "--no-compress-literals")) { literalCompressionMode = ZSTD_ps_disable; continue; } - if (!strcmp(argument, "--no-progress")) { FIO_setProgressSetting(FIO_ps_never); continue; } - if (!strcmp(argument, "--progress")) { FIO_setProgressSetting(FIO_ps_always); continue; } - if (!strcmp(argument, "--exclude-compressed")) { FIO_setExcludeCompressedFile(prefs, 1); continue; } - - /* long commands with arguments */ -#ifndef ZSTD_NODICT - if (longCommandWArg(&argument, "--train-cover")) { - operation = zom_train; - if (outFileName == NULL) - outFileName = g_defaultDictName; - dict = cover; - /* Allow optional arguments following an = */ - if (*argument == 0) { memset(&coverParams, 0, sizeof(coverParams)); } - else if (*argument++ != '=') { badusage(programName); CLEAN_RETURN(1); } - else if (!parseCoverParameters(argument, &coverParams)) { badusage(programName); CLEAN_RETURN(1); } - continue; - } - if (longCommandWArg(&argument, "--train-fastcover")) { - operation = zom_train; - if (outFileName == NULL) - outFileName = g_defaultDictName; - dict = fastCover; - /* Allow optional arguments following an = */ - if (*argument == 0) { memset(&fastCoverParams, 0, sizeof(fastCoverParams)); } - else if (*argument++ != '=') { badusage(programName); CLEAN_RETURN(1); } - else if (!parseFastCoverParameters(argument, &fastCoverParams)) { badusage(programName); CLEAN_RETURN(1); } - continue; - } - if (longCommandWArg(&argument, "--train-legacy")) { - operation = zom_train; - if (outFileName == NULL) - outFileName = g_defaultDictName; - dict = legacy; - /* Allow optional arguments following an = */ - if (*argument == 0) { continue; } - else if (*argument++ != '=') { badusage(programName); CLEAN_RETURN(1); } - else if (!parseLegacyParameters(argument, &dictSelect)) { badusage(programName); CLEAN_RETURN(1); } - continue; - } -#endif - if (longCommandWArg(&argument, "--threads")) { NEXT_UINT32(nbWorkers); continue; } - if (longCommandWArg(&argument, "--memlimit")) { NEXT_UINT32(memLimit); continue; } - if (longCommandWArg(&argument, "--memory")) { NEXT_UINT32(memLimit); continue; } - if (longCommandWArg(&argument, "--memlimit-decompress")) { NEXT_UINT32(memLimit); continue; } - if (longCommandWArg(&argument, "--block-size=")) { blockSize = readSizeTFromChar(&argument); continue; } - if (longCommandWArg(&argument, "--maxdict")) { NEXT_UINT32(maxDictSize); continue; } - if (longCommandWArg(&argument, "--dictID")) { NEXT_UINT32(dictID); continue; } - if (longCommandWArg(&argument, "--zstd=")) { if (!parseCompressionParameters(argument, &compressionParams)) { badusage(programName); CLEAN_RETURN(1); } continue; } - if (longCommandWArg(&argument, "--stream-size=")) { streamSrcSize = readSizeTFromChar(&argument); continue; } - if (longCommandWArg(&argument, "--target-compressed-block-size=")) { targetCBlockSize = readSizeTFromChar(&argument); continue; } - if (longCommandWArg(&argument, "--size-hint=")) { srcSizeHint = readSizeTFromChar(&argument); continue; } - if (longCommandWArg(&argument, "--output-dir-flat")) { NEXT_FIELD(outDirName); continue; } + if (!strcmp(argument, "--no-progress")) { FIO_setProgressSetting(FIO_ps_never); continue; } + if (!strcmp(argument, "--progress")) { FIO_setProgressSetting(FIO_ps_always); continue; } + if (!strcmp(argument, "--exclude-compressed")) { FIO_setExcludeCompressedFile(prefs, 1); continue; } + + /* long commands with arguments */ +#ifndef ZSTD_NODICT + if (longCommandWArg(&argument, "--train-cover")) { + operation = zom_train; + if (outFileName == NULL) + outFileName = g_defaultDictName; + dict = cover; + /* Allow optional arguments following an = */ + if (*argument == 0) { memset(&coverParams, 0, sizeof(coverParams)); } + else if (*argument++ != '=') { badusage(programName); CLEAN_RETURN(1); } + else if (!parseCoverParameters(argument, &coverParams)) { badusage(programName); CLEAN_RETURN(1); } + continue; + } + if (longCommandWArg(&argument, "--train-fastcover")) { + operation = zom_train; + if (outFileName == NULL) + outFileName = g_defaultDictName; + dict = fastCover; + /* Allow optional arguments following an = */ + if (*argument == 0) { memset(&fastCoverParams, 0, sizeof(fastCoverParams)); } + else if (*argument++ != '=') { badusage(programName); CLEAN_RETURN(1); } + else if (!parseFastCoverParameters(argument, &fastCoverParams)) { badusage(programName); CLEAN_RETURN(1); } + continue; + } + if (longCommandWArg(&argument, "--train-legacy")) { + operation = zom_train; + if (outFileName == NULL) + outFileName = g_defaultDictName; + dict = legacy; + /* Allow optional arguments following an = */ + if (*argument == 0) { continue; } + else if (*argument++ != '=') { badusage(programName); CLEAN_RETURN(1); } + else if (!parseLegacyParameters(argument, &dictSelect)) { badusage(programName); CLEAN_RETURN(1); } + continue; + } +#endif + if (longCommandWArg(&argument, "--threads")) { NEXT_UINT32(nbWorkers); continue; } + if (longCommandWArg(&argument, "--memlimit")) { NEXT_UINT32(memLimit); continue; } + if (longCommandWArg(&argument, "--memory")) { NEXT_UINT32(memLimit); continue; } + if (longCommandWArg(&argument, "--memlimit-decompress")) { NEXT_UINT32(memLimit); continue; } + if (longCommandWArg(&argument, "--block-size=")) { blockSize = readSizeTFromChar(&argument); continue; } + if (longCommandWArg(&argument, "--maxdict")) { NEXT_UINT32(maxDictSize); continue; } + if (longCommandWArg(&argument, "--dictID")) { NEXT_UINT32(dictID); continue; } + if (longCommandWArg(&argument, "--zstd=")) { if (!parseCompressionParameters(argument, &compressionParams)) { badusage(programName); CLEAN_RETURN(1); } continue; } + if (longCommandWArg(&argument, "--stream-size=")) { streamSrcSize = readSizeTFromChar(&argument); continue; } + if (longCommandWArg(&argument, "--target-compressed-block-size=")) { targetCBlockSize = readSizeTFromChar(&argument); continue; } + if (longCommandWArg(&argument, "--size-hint=")) { srcSizeHint = readSizeTFromChar(&argument); continue; } + if (longCommandWArg(&argument, "--output-dir-flat")) { NEXT_FIELD(outDirName); continue; } #ifdef ZSTD_MULTITHREAD if (longCommandWArg(&argument, "--auto-threads")) { const char* threadDefault = NULL; @@ -1000,212 +1000,212 @@ int main(int argCount, const char* argv[]) continue; } #endif -#ifdef UTIL_HAS_MIRRORFILELIST - if (longCommandWArg(&argument, "--output-dir-mirror")) { NEXT_FIELD(outMirroredDirName); continue; } -#endif -#ifndef ZSTD_NOTRACE - if (longCommandWArg(&argument, "--trace")) { char const* traceFile; NEXT_FIELD(traceFile); TRACE_enable(traceFile); continue; } -#endif - if (longCommandWArg(&argument, "--patch-from")) { NEXT_FIELD(patchFromDictFileName); continue; } - if (longCommandWArg(&argument, "--long")) { - unsigned ldmWindowLog = 0; - ldmFlag = 1; - /* Parse optional window log */ - if (*argument == '=') { - ++argument; - ldmWindowLog = readU32FromChar(&argument); - } else if (*argument != 0) { - /* Invalid character following --long */ - badusage(programName); - CLEAN_RETURN(1); - } - /* Only set windowLog if not already set by --zstd */ - if (compressionParams.windowLog == 0) - compressionParams.windowLog = ldmWindowLog; - continue; - } -#ifndef ZSTD_NOCOMPRESS /* linking ZSTD_minCLevel() requires compression support */ - if (longCommandWArg(&argument, "--fast")) { - /* Parse optional acceleration factor */ - if (*argument == '=') { - U32 const maxFast = (U32)-ZSTD_minCLevel(); - U32 fastLevel; - ++argument; - fastLevel = readU32FromChar(&argument); - if (fastLevel > maxFast) fastLevel = maxFast; - if (fastLevel) { - dictCLevel = cLevel = -(int)fastLevel; - } else { - badusage(programName); - CLEAN_RETURN(1); - } - } else if (*argument != 0) { - /* Invalid character following --fast */ - badusage(programName); - CLEAN_RETURN(1); - } else { - cLevel = -1; /* default for --fast */ - } - continue; - } -#endif - - if (longCommandWArg(&argument, "--filelist")) { - const char* listName; - NEXT_FIELD(listName); - UTIL_refFilename(file_of_names, listName); - continue; - } - - /* fall-through, will trigger bad_usage() later on */ - } - - argument++; - while (argument[0]!=0) { - -#ifndef ZSTD_NOCOMPRESS - /* compression Level */ - if ((*argument>='0') && (*argument<='9')) { - dictCLevel = cLevel = (int)readU32FromChar(&argument); - continue; - } -#endif - - switch(argument[0]) - { - /* Display help */ - case 'V': printVersion(); CLEAN_RETURN(0); /* Version Only */ - case 'H': - case 'h': usage_advanced(programName); CLEAN_RETURN(0); - - /* Compress */ - case 'z': operation=zom_compress; argument++; break; - - /* Decoding */ - case 'd': -#ifndef ZSTD_NOBENCH - benchParams.mode = BMK_decodeOnly; - if (operation==zom_bench) { argument++; break; } /* benchmark decode (hidden option) */ -#endif - operation=zom_decompress; argument++; break; - - /* Force stdout, even if stdout==console */ - case 'c': forceStdout=1; outFileName=stdoutmark; argument++; break; - - /* Use file content as dictionary */ - case 'D': argument++; NEXT_FIELD(dictFileName); break; - - /* Overwrite */ - case 'f': FIO_overwriteMode(prefs); forceStdin=1; forceStdout=1; followLinks=1; allowBlockDevices=1; argument++; break; - - /* Verbose mode */ - case 'v': g_displayLevel++; argument++; break; - - /* Quiet mode */ - case 'q': g_displayLevel--; argument++; break; - - /* keep source file (default) */ - case 'k': FIO_setRemoveSrcFile(prefs, 0); argument++; break; - - /* Checksum */ - case 'C': FIO_setChecksumFlag(prefs, 2); argument++; break; - - /* test compressed file */ - case 't': operation=zom_test; argument++; break; - - /* destination file name */ - case 'o': argument++; NEXT_FIELD(outFileName); break; - - /* limit memory */ - case 'M': - argument++; - memLimit = readU32FromChar(&argument); - break; - case 'l': operation=zom_list; argument++; break; -#ifdef UTIL_HAS_CREATEFILELIST - /* recursive */ - case 'r': recursive=1; argument++; break; -#endif - -#ifndef ZSTD_NOBENCH - /* Benchmark */ - case 'b': - operation=zom_bench; - argument++; - break; - - /* range bench (benchmark only) */ - case 'e': - /* compression Level */ - argument++; - cLevelLast = (int)readU32FromChar(&argument); - break; - - /* Modify Nb Iterations (benchmark only) */ - case 'i': - argument++; - bench_nbSeconds = readU32FromChar(&argument); - break; - - /* cut input into blocks (benchmark only) */ - case 'B': - argument++; - blockSize = readU32FromChar(&argument); - break; - - /* benchmark files separately (hidden option) */ - case 'S': - argument++; - separateFiles = 1; - break; - -#endif /* ZSTD_NOBENCH */ - - /* nb of threads (hidden option) */ - case 'T': - argument++; - nbWorkers = (int)readU32FromChar(&argument); - break; - - /* Dictionary Selection level */ - case 's': - argument++; - dictSelect = readU32FromChar(&argument); - break; - - /* Pause at the end (-p) or set an additional param (-p#) (hidden option) */ - case 'p': argument++; -#ifndef ZSTD_NOBENCH - if ((*argument>='0') && (*argument<='9')) { - benchParams.additionalParam = (int)readU32FromChar(&argument); - } else -#endif - main_pause=1; - break; - - /* Select compressibility of synthetic sample */ - case 'P': - argument++; - compressibility = (double)readU32FromChar(&argument) / 100; - break; - - /* unknown command */ - default : badusage(programName); CLEAN_RETURN(1); - } - } - continue; - } /* if (argument[0]=='-') */ - - /* none of the above : add filename to list */ - UTIL_refFilename(filenames, argument); - } - - /* Welcome message (if verbose) */ - DISPLAYLEVEL(3, WELCOME_MESSAGE); - -#ifdef ZSTD_MULTITHREAD - if ((nbWorkers==0) && (!singleThread)) { - /* automatically set # workers based on # of reported cpus */ +#ifdef UTIL_HAS_MIRRORFILELIST + if (longCommandWArg(&argument, "--output-dir-mirror")) { NEXT_FIELD(outMirroredDirName); continue; } +#endif +#ifndef ZSTD_NOTRACE + if (longCommandWArg(&argument, "--trace")) { char const* traceFile; NEXT_FIELD(traceFile); TRACE_enable(traceFile); continue; } +#endif + if (longCommandWArg(&argument, "--patch-from")) { NEXT_FIELD(patchFromDictFileName); continue; } + if (longCommandWArg(&argument, "--long")) { + unsigned ldmWindowLog = 0; + ldmFlag = 1; + /* Parse optional window log */ + if (*argument == '=') { + ++argument; + ldmWindowLog = readU32FromChar(&argument); + } else if (*argument != 0) { + /* Invalid character following --long */ + badusage(programName); + CLEAN_RETURN(1); + } + /* Only set windowLog if not already set by --zstd */ + if (compressionParams.windowLog == 0) + compressionParams.windowLog = ldmWindowLog; + continue; + } +#ifndef ZSTD_NOCOMPRESS /* linking ZSTD_minCLevel() requires compression support */ + if (longCommandWArg(&argument, "--fast")) { + /* Parse optional acceleration factor */ + if (*argument == '=') { + U32 const maxFast = (U32)-ZSTD_minCLevel(); + U32 fastLevel; + ++argument; + fastLevel = readU32FromChar(&argument); + if (fastLevel > maxFast) fastLevel = maxFast; + if (fastLevel) { + dictCLevel = cLevel = -(int)fastLevel; + } else { + badusage(programName); + CLEAN_RETURN(1); + } + } else if (*argument != 0) { + /* Invalid character following --fast */ + badusage(programName); + CLEAN_RETURN(1); + } else { + cLevel = -1; /* default for --fast */ + } + continue; + } +#endif + + if (longCommandWArg(&argument, "--filelist")) { + const char* listName; + NEXT_FIELD(listName); + UTIL_refFilename(file_of_names, listName); + continue; + } + + /* fall-through, will trigger bad_usage() later on */ + } + + argument++; + while (argument[0]!=0) { + +#ifndef ZSTD_NOCOMPRESS + /* compression Level */ + if ((*argument>='0') && (*argument<='9')) { + dictCLevel = cLevel = (int)readU32FromChar(&argument); + continue; + } +#endif + + switch(argument[0]) + { + /* Display help */ + case 'V': printVersion(); CLEAN_RETURN(0); /* Version Only */ + case 'H': + case 'h': usage_advanced(programName); CLEAN_RETURN(0); + + /* Compress */ + case 'z': operation=zom_compress; argument++; break; + + /* Decoding */ + case 'd': +#ifndef ZSTD_NOBENCH + benchParams.mode = BMK_decodeOnly; + if (operation==zom_bench) { argument++; break; } /* benchmark decode (hidden option) */ +#endif + operation=zom_decompress; argument++; break; + + /* Force stdout, even if stdout==console */ + case 'c': forceStdout=1; outFileName=stdoutmark; argument++; break; + + /* Use file content as dictionary */ + case 'D': argument++; NEXT_FIELD(dictFileName); break; + + /* Overwrite */ + case 'f': FIO_overwriteMode(prefs); forceStdin=1; forceStdout=1; followLinks=1; allowBlockDevices=1; argument++; break; + + /* Verbose mode */ + case 'v': g_displayLevel++; argument++; break; + + /* Quiet mode */ + case 'q': g_displayLevel--; argument++; break; + + /* keep source file (default) */ + case 'k': FIO_setRemoveSrcFile(prefs, 0); argument++; break; + + /* Checksum */ + case 'C': FIO_setChecksumFlag(prefs, 2); argument++; break; + + /* test compressed file */ + case 't': operation=zom_test; argument++; break; + + /* destination file name */ + case 'o': argument++; NEXT_FIELD(outFileName); break; + + /* limit memory */ + case 'M': + argument++; + memLimit = readU32FromChar(&argument); + break; + case 'l': operation=zom_list; argument++; break; +#ifdef UTIL_HAS_CREATEFILELIST + /* recursive */ + case 'r': recursive=1; argument++; break; +#endif + +#ifndef ZSTD_NOBENCH + /* Benchmark */ + case 'b': + operation=zom_bench; + argument++; + break; + + /* range bench (benchmark only) */ + case 'e': + /* compression Level */ + argument++; + cLevelLast = (int)readU32FromChar(&argument); + break; + + /* Modify Nb Iterations (benchmark only) */ + case 'i': + argument++; + bench_nbSeconds = readU32FromChar(&argument); + break; + + /* cut input into blocks (benchmark only) */ + case 'B': + argument++; + blockSize = readU32FromChar(&argument); + break; + + /* benchmark files separately (hidden option) */ + case 'S': + argument++; + separateFiles = 1; + break; + +#endif /* ZSTD_NOBENCH */ + + /* nb of threads (hidden option) */ + case 'T': + argument++; + nbWorkers = (int)readU32FromChar(&argument); + break; + + /* Dictionary Selection level */ + case 's': + argument++; + dictSelect = readU32FromChar(&argument); + break; + + /* Pause at the end (-p) or set an additional param (-p#) (hidden option) */ + case 'p': argument++; +#ifndef ZSTD_NOBENCH + if ((*argument>='0') && (*argument<='9')) { + benchParams.additionalParam = (int)readU32FromChar(&argument); + } else +#endif + main_pause=1; + break; + + /* Select compressibility of synthetic sample */ + case 'P': + argument++; + compressibility = (double)readU32FromChar(&argument) / 100; + break; + + /* unknown command */ + default : badusage(programName); CLEAN_RETURN(1); + } + } + continue; + } /* if (argument[0]=='-') */ + + /* none of the above : add filename to list */ + UTIL_refFilename(filenames, argument); + } + + /* Welcome message (if verbose) */ + DISPLAYLEVEL(3, WELCOME_MESSAGE); + +#ifdef ZSTD_MULTITHREAD + if ((nbWorkers==0) && (!singleThread)) { + /* automatically set # workers based on # of reported cpus */ if (defaultLogicalCores) { nbWorkers = UTIL_countLogicalCores(); DISPLAYLEVEL(3, "Note: %d logical core(s) detected \n", nbWorkers); @@ -1213,282 +1213,282 @@ int main(int argCount, const char* argv[]) nbWorkers = UTIL_countPhysicalCores(); DISPLAYLEVEL(3, "Note: %d physical core(s) detected \n", nbWorkers); } - } -#else - (void)singleThread; (void)nbWorkers; -#endif - + } +#else + (void)singleThread; (void)nbWorkers; +#endif + g_utilDisplayLevel = g_displayLevel; -#ifdef UTIL_HAS_CREATEFILELIST - if (!followLinks) { - unsigned u, fileNamesNb; - unsigned const nbFilenames = (unsigned)filenames->tableSize; - for (u=0, fileNamesNb=0; u<nbFilenames; u++) { - if ( UTIL_isLink(filenames->fileNames[u]) - && !UTIL_isFIFO(filenames->fileNames[u]) - ) { - DISPLAYLEVEL(2, "Warning : %s is a symbolic link, ignoring \n", filenames->fileNames[u]); - } else { - filenames->fileNames[fileNamesNb++] = filenames->fileNames[u]; - } } - if (fileNamesNb == 0 && nbFilenames > 0) /* all names are eliminated */ - CLEAN_RETURN(1); - filenames->tableSize = fileNamesNb; - } /* if (!followLinks) */ - - /* read names from a file */ - if (file_of_names->tableSize) { - size_t const nbFileLists = file_of_names->tableSize; - size_t flNb; - for (flNb=0; flNb < nbFileLists; flNb++) { - FileNamesTable* const fnt = UTIL_createFileNamesTable_fromFileName(file_of_names->fileNames[flNb]); - if (fnt==NULL) { - DISPLAYLEVEL(1, "zstd: error reading %s \n", file_of_names->fileNames[flNb]); - CLEAN_RETURN(1); - } - filenames = UTIL_mergeFileNamesTable(filenames, fnt); - } - } - - if (recursive) { /* at this stage, filenameTable is a list of paths, which can contain both files and directories */ - UTIL_expandFNT(&filenames, followLinks); - } -#else - (void)followLinks; -#endif - - if (operation == zom_list) { -#ifndef ZSTD_NODECOMPRESS - int const ret = FIO_listMultipleFiles((unsigned)filenames->tableSize, filenames->fileNames, g_displayLevel); - CLEAN_RETURN(ret); -#else - DISPLAY("file information is not supported \n"); - CLEAN_RETURN(1); -#endif - } - - /* Check if benchmark is selected */ - if (operation==zom_bench) { -#ifndef ZSTD_NOBENCH - benchParams.blockSize = blockSize; - benchParams.nbWorkers = nbWorkers; - benchParams.realTime = (unsigned)setRealTimePrio; - benchParams.nbSeconds = bench_nbSeconds; - benchParams.ldmFlag = ldmFlag; - benchParams.ldmMinMatch = (int)g_ldmMinMatch; - benchParams.ldmHashLog = (int)g_ldmHashLog; - benchParams.useRowMatchFinder = useRowMatchFinder; - if (g_ldmBucketSizeLog != LDM_PARAM_DEFAULT) { - benchParams.ldmBucketSizeLog = (int)g_ldmBucketSizeLog; - } - if (g_ldmHashRateLog != LDM_PARAM_DEFAULT) { - benchParams.ldmHashRateLog = (int)g_ldmHashRateLog; - } - benchParams.literalCompressionMode = literalCompressionMode; - - if (cLevel > ZSTD_maxCLevel()) cLevel = ZSTD_maxCLevel(); - if (cLevelLast > ZSTD_maxCLevel()) cLevelLast = ZSTD_maxCLevel(); - if (cLevelLast < cLevel) cLevelLast = cLevel; - if (cLevelLast > cLevel) - DISPLAYLEVEL(3, "Benchmarking levels from %d to %d\n", cLevel, cLevelLast); - if (filenames->tableSize > 0) { - if(separateFiles) { - unsigned i; - for(i = 0; i < filenames->tableSize; i++) { - int c; - DISPLAYLEVEL(3, "Benchmarking %s \n", filenames->fileNames[i]); - for(c = cLevel; c <= cLevelLast; c++) { - BMK_benchFilesAdvanced(&filenames->fileNames[i], 1, dictFileName, c, &compressionParams, g_displayLevel, &benchParams); - } } - } else { - for(; cLevel <= cLevelLast; cLevel++) { - BMK_benchFilesAdvanced(filenames->fileNames, (unsigned)filenames->tableSize, dictFileName, cLevel, &compressionParams, g_displayLevel, &benchParams); - } } - } else { - for(; cLevel <= cLevelLast; cLevel++) { - BMK_syntheticTest(cLevel, compressibility, &compressionParams, g_displayLevel, &benchParams); - } } - -#else - (void)bench_nbSeconds; (void)blockSize; (void)setRealTimePrio; (void)separateFiles; (void)compressibility; -#endif - goto _end; - } - - /* Check if dictionary builder is selected */ - if (operation==zom_train) { -#ifndef ZSTD_NODICT - ZDICT_params_t zParams; - zParams.compressionLevel = dictCLevel; - zParams.notificationLevel = (unsigned)g_displayLevel; - zParams.dictID = dictID; - if (dict == cover) { - int const optimize = !coverParams.k || !coverParams.d; - coverParams.nbThreads = (unsigned)nbWorkers; - coverParams.zParams = zParams; +#ifdef UTIL_HAS_CREATEFILELIST + if (!followLinks) { + unsigned u, fileNamesNb; + unsigned const nbFilenames = (unsigned)filenames->tableSize; + for (u=0, fileNamesNb=0; u<nbFilenames; u++) { + if ( UTIL_isLink(filenames->fileNames[u]) + && !UTIL_isFIFO(filenames->fileNames[u]) + ) { + DISPLAYLEVEL(2, "Warning : %s is a symbolic link, ignoring \n", filenames->fileNames[u]); + } else { + filenames->fileNames[fileNamesNb++] = filenames->fileNames[u]; + } } + if (fileNamesNb == 0 && nbFilenames > 0) /* all names are eliminated */ + CLEAN_RETURN(1); + filenames->tableSize = fileNamesNb; + } /* if (!followLinks) */ + + /* read names from a file */ + if (file_of_names->tableSize) { + size_t const nbFileLists = file_of_names->tableSize; + size_t flNb; + for (flNb=0; flNb < nbFileLists; flNb++) { + FileNamesTable* const fnt = UTIL_createFileNamesTable_fromFileName(file_of_names->fileNames[flNb]); + if (fnt==NULL) { + DISPLAYLEVEL(1, "zstd: error reading %s \n", file_of_names->fileNames[flNb]); + CLEAN_RETURN(1); + } + filenames = UTIL_mergeFileNamesTable(filenames, fnt); + } + } + + if (recursive) { /* at this stage, filenameTable is a list of paths, which can contain both files and directories */ + UTIL_expandFNT(&filenames, followLinks); + } +#else + (void)followLinks; +#endif + + if (operation == zom_list) { +#ifndef ZSTD_NODECOMPRESS + int const ret = FIO_listMultipleFiles((unsigned)filenames->tableSize, filenames->fileNames, g_displayLevel); + CLEAN_RETURN(ret); +#else + DISPLAY("file information is not supported \n"); + CLEAN_RETURN(1); +#endif + } + + /* Check if benchmark is selected */ + if (operation==zom_bench) { +#ifndef ZSTD_NOBENCH + benchParams.blockSize = blockSize; + benchParams.nbWorkers = nbWorkers; + benchParams.realTime = (unsigned)setRealTimePrio; + benchParams.nbSeconds = bench_nbSeconds; + benchParams.ldmFlag = ldmFlag; + benchParams.ldmMinMatch = (int)g_ldmMinMatch; + benchParams.ldmHashLog = (int)g_ldmHashLog; + benchParams.useRowMatchFinder = useRowMatchFinder; + if (g_ldmBucketSizeLog != LDM_PARAM_DEFAULT) { + benchParams.ldmBucketSizeLog = (int)g_ldmBucketSizeLog; + } + if (g_ldmHashRateLog != LDM_PARAM_DEFAULT) { + benchParams.ldmHashRateLog = (int)g_ldmHashRateLog; + } + benchParams.literalCompressionMode = literalCompressionMode; + + if (cLevel > ZSTD_maxCLevel()) cLevel = ZSTD_maxCLevel(); + if (cLevelLast > ZSTD_maxCLevel()) cLevelLast = ZSTD_maxCLevel(); + if (cLevelLast < cLevel) cLevelLast = cLevel; + if (cLevelLast > cLevel) + DISPLAYLEVEL(3, "Benchmarking levels from %d to %d\n", cLevel, cLevelLast); + if (filenames->tableSize > 0) { + if(separateFiles) { + unsigned i; + for(i = 0; i < filenames->tableSize; i++) { + int c; + DISPLAYLEVEL(3, "Benchmarking %s \n", filenames->fileNames[i]); + for(c = cLevel; c <= cLevelLast; c++) { + BMK_benchFilesAdvanced(&filenames->fileNames[i], 1, dictFileName, c, &compressionParams, g_displayLevel, &benchParams); + } } + } else { + for(; cLevel <= cLevelLast; cLevel++) { + BMK_benchFilesAdvanced(filenames->fileNames, (unsigned)filenames->tableSize, dictFileName, cLevel, &compressionParams, g_displayLevel, &benchParams); + } } + } else { + for(; cLevel <= cLevelLast; cLevel++) { + BMK_syntheticTest(cLevel, compressibility, &compressionParams, g_displayLevel, &benchParams); + } } + +#else + (void)bench_nbSeconds; (void)blockSize; (void)setRealTimePrio; (void)separateFiles; (void)compressibility; +#endif + goto _end; + } + + /* Check if dictionary builder is selected */ + if (operation==zom_train) { +#ifndef ZSTD_NODICT + ZDICT_params_t zParams; + zParams.compressionLevel = dictCLevel; + zParams.notificationLevel = (unsigned)g_displayLevel; + zParams.dictID = dictID; + if (dict == cover) { + int const optimize = !coverParams.k || !coverParams.d; + coverParams.nbThreads = (unsigned)nbWorkers; + coverParams.zParams = zParams; operationResult = DiB_trainFromFiles(outFileName, maxDictSize, filenames->fileNames, (int)filenames->tableSize, blockSize, NULL, &coverParams, NULL, optimize, memLimit); - } else if (dict == fastCover) { - int const optimize = !fastCoverParams.k || !fastCoverParams.d; - fastCoverParams.nbThreads = (unsigned)nbWorkers; - fastCoverParams.zParams = zParams; + } else if (dict == fastCover) { + int const optimize = !fastCoverParams.k || !fastCoverParams.d; + fastCoverParams.nbThreads = (unsigned)nbWorkers; + fastCoverParams.zParams = zParams; operationResult = DiB_trainFromFiles(outFileName, maxDictSize, filenames->fileNames, (int)filenames->tableSize, blockSize, NULL, NULL, &fastCoverParams, optimize, memLimit); - } else { - ZDICT_legacy_params_t dictParams; - memset(&dictParams, 0, sizeof(dictParams)); - dictParams.selectivityLevel = dictSelect; - dictParams.zParams = zParams; + } else { + ZDICT_legacy_params_t dictParams; + memset(&dictParams, 0, sizeof(dictParams)); + dictParams.selectivityLevel = dictSelect; + dictParams.zParams = zParams; operationResult = DiB_trainFromFiles(outFileName, maxDictSize, filenames->fileNames, (int)filenames->tableSize, blockSize, &dictParams, NULL, NULL, 0, memLimit); - } -#else - (void)dictCLevel; (void)dictSelect; (void)dictID; (void)maxDictSize; /* not used when ZSTD_NODICT set */ - DISPLAYLEVEL(1, "training mode not available \n"); - operationResult = 1; -#endif - goto _end; - } - -#ifndef ZSTD_NODECOMPRESS - if (operation==zom_test) { FIO_setTestMode(prefs, 1); outFileName=nulmark; FIO_setRemoveSrcFile(prefs, 0); } /* test mode */ -#endif - - /* No input filename ==> use stdin and stdout */ - if (filenames->tableSize == 0) UTIL_refFilename(filenames, stdinmark); - if (!strcmp(filenames->fileNames[0], stdinmark) && !outFileName) - outFileName = stdoutmark; /* when input is stdin, default output is stdout */ - - /* Check if input/output defined as console; trigger an error in this case */ - if (!forceStdin - && !strcmp(filenames->fileNames[0], stdinmark) - && IS_CONSOLE(stdin) ) { - DISPLAYLEVEL(1, "stdin is a console, aborting\n"); - CLEAN_RETURN(1); - } - if ( outFileName && !strcmp(outFileName, stdoutmark) - && IS_CONSOLE(stdout) - && !strcmp(filenames->fileNames[0], stdinmark) - && !forceStdout - && operation!=zom_decompress ) { - DISPLAYLEVEL(1, "stdout is a console, aborting\n"); - CLEAN_RETURN(1); - } - -#ifndef ZSTD_NOCOMPRESS - /* check compression level limits */ - { int const maxCLevel = ultra ? ZSTD_maxCLevel() : ZSTDCLI_CLEVEL_MAX; - if (cLevel > maxCLevel) { - DISPLAYLEVEL(2, "Warning : compression level higher than max, reduced to %i \n", maxCLevel); - cLevel = maxCLevel; - } } -#endif - - if (showDefaultCParams) { - if (operation == zom_decompress) { - DISPLAY("error : can't use --show-default-cparams in decomrpession mode \n"); - CLEAN_RETURN(1); - } - } - - if (dictFileName != NULL && patchFromDictFileName != NULL) { - DISPLAY("error : can't use -D and --patch-from=# at the same time \n"); - CLEAN_RETURN(1); - } - - if (patchFromDictFileName != NULL && filenames->tableSize > 1) { - DISPLAY("error : can't use --patch-from=# on multiple files \n"); - CLEAN_RETURN(1); - } - - /* No status message in pipe mode (stdin - stdout) */ - hasStdout = outFileName && !strcmp(outFileName,stdoutmark); - + } +#else + (void)dictCLevel; (void)dictSelect; (void)dictID; (void)maxDictSize; /* not used when ZSTD_NODICT set */ + DISPLAYLEVEL(1, "training mode not available \n"); + operationResult = 1; +#endif + goto _end; + } + +#ifndef ZSTD_NODECOMPRESS + if (operation==zom_test) { FIO_setTestMode(prefs, 1); outFileName=nulmark; FIO_setRemoveSrcFile(prefs, 0); } /* test mode */ +#endif + + /* No input filename ==> use stdin and stdout */ + if (filenames->tableSize == 0) UTIL_refFilename(filenames, stdinmark); + if (!strcmp(filenames->fileNames[0], stdinmark) && !outFileName) + outFileName = stdoutmark; /* when input is stdin, default output is stdout */ + + /* Check if input/output defined as console; trigger an error in this case */ + if (!forceStdin + && !strcmp(filenames->fileNames[0], stdinmark) + && IS_CONSOLE(stdin) ) { + DISPLAYLEVEL(1, "stdin is a console, aborting\n"); + CLEAN_RETURN(1); + } + if ( outFileName && !strcmp(outFileName, stdoutmark) + && IS_CONSOLE(stdout) + && !strcmp(filenames->fileNames[0], stdinmark) + && !forceStdout + && operation!=zom_decompress ) { + DISPLAYLEVEL(1, "stdout is a console, aborting\n"); + CLEAN_RETURN(1); + } + +#ifndef ZSTD_NOCOMPRESS + /* check compression level limits */ + { int const maxCLevel = ultra ? ZSTD_maxCLevel() : ZSTDCLI_CLEVEL_MAX; + if (cLevel > maxCLevel) { + DISPLAYLEVEL(2, "Warning : compression level higher than max, reduced to %i \n", maxCLevel); + cLevel = maxCLevel; + } } +#endif + + if (showDefaultCParams) { + if (operation == zom_decompress) { + DISPLAY("error : can't use --show-default-cparams in decomrpession mode \n"); + CLEAN_RETURN(1); + } + } + + if (dictFileName != NULL && patchFromDictFileName != NULL) { + DISPLAY("error : can't use -D and --patch-from=# at the same time \n"); + CLEAN_RETURN(1); + } + + if (patchFromDictFileName != NULL && filenames->tableSize > 1) { + DISPLAY("error : can't use --patch-from=# on multiple files \n"); + CLEAN_RETURN(1); + } + + /* No status message in pipe mode (stdin - stdout) */ + hasStdout = outFileName && !strcmp(outFileName,stdoutmark); + if ((hasStdout || !IS_CONSOLE(stderr)) && (g_displayLevel==2)) g_displayLevel=1; - - /* IO Stream/File */ - FIO_setHasStdoutOutput(fCtx, hasStdout); - FIO_setNbFilesTotal(fCtx, (int)filenames->tableSize); - FIO_determineHasStdinInput(fCtx, filenames); - FIO_setNotificationLevel(g_displayLevel); - FIO_setAllowBlockDevices(prefs, allowBlockDevices); - FIO_setPatchFromMode(prefs, patchFromDictFileName != NULL); - if (memLimit == 0) { - if (compressionParams.windowLog == 0) { - memLimit = (U32)1 << g_defaultMaxWindowLog; - } else { - memLimit = (U32)1 << (compressionParams.windowLog & 31); - } } - if (patchFromDictFileName != NULL) - dictFileName = patchFromDictFileName; - FIO_setMemLimit(prefs, memLimit); - if (operation==zom_compress) { -#ifndef ZSTD_NOCOMPRESS - FIO_setContentSize(prefs, contentSize); - FIO_setNbWorkers(prefs, nbWorkers); - FIO_setBlockSize(prefs, (int)blockSize); - if (g_overlapLog!=OVERLAP_LOG_DEFAULT) FIO_setOverlapLog(prefs, (int)g_overlapLog); - FIO_setLdmFlag(prefs, (unsigned)ldmFlag); - FIO_setLdmHashLog(prefs, (int)g_ldmHashLog); - FIO_setLdmMinMatch(prefs, (int)g_ldmMinMatch); - if (g_ldmBucketSizeLog != LDM_PARAM_DEFAULT) FIO_setLdmBucketSizeLog(prefs, (int)g_ldmBucketSizeLog); - if (g_ldmHashRateLog != LDM_PARAM_DEFAULT) FIO_setLdmHashRateLog(prefs, (int)g_ldmHashRateLog); - FIO_setAdaptiveMode(prefs, (unsigned)adapt); - FIO_setUseRowMatchFinder(prefs, useRowMatchFinder); - FIO_setAdaptMin(prefs, adaptMin); - FIO_setAdaptMax(prefs, adaptMax); - FIO_setRsyncable(prefs, rsyncable); - FIO_setStreamSrcSize(prefs, streamSrcSize); - FIO_setTargetCBlockSize(prefs, targetCBlockSize); - FIO_setSrcSizeHint(prefs, srcSizeHint); - FIO_setLiteralCompressionMode(prefs, literalCompressionMode); - if (adaptMin > cLevel) cLevel = adaptMin; - if (adaptMax < cLevel) cLevel = adaptMax; - - /* Compare strategies constant with the ground truth */ - { ZSTD_bounds strategyBounds = ZSTD_cParam_getBounds(ZSTD_c_strategy); - assert(ZSTD_NB_STRATEGIES == strategyBounds.upperBound); - (void)strategyBounds; } - + + /* IO Stream/File */ + FIO_setHasStdoutOutput(fCtx, hasStdout); + FIO_setNbFilesTotal(fCtx, (int)filenames->tableSize); + FIO_determineHasStdinInput(fCtx, filenames); + FIO_setNotificationLevel(g_displayLevel); + FIO_setAllowBlockDevices(prefs, allowBlockDevices); + FIO_setPatchFromMode(prefs, patchFromDictFileName != NULL); + if (memLimit == 0) { + if (compressionParams.windowLog == 0) { + memLimit = (U32)1 << g_defaultMaxWindowLog; + } else { + memLimit = (U32)1 << (compressionParams.windowLog & 31); + } } + if (patchFromDictFileName != NULL) + dictFileName = patchFromDictFileName; + FIO_setMemLimit(prefs, memLimit); + if (operation==zom_compress) { +#ifndef ZSTD_NOCOMPRESS + FIO_setContentSize(prefs, contentSize); + FIO_setNbWorkers(prefs, nbWorkers); + FIO_setBlockSize(prefs, (int)blockSize); + if (g_overlapLog!=OVERLAP_LOG_DEFAULT) FIO_setOverlapLog(prefs, (int)g_overlapLog); + FIO_setLdmFlag(prefs, (unsigned)ldmFlag); + FIO_setLdmHashLog(prefs, (int)g_ldmHashLog); + FIO_setLdmMinMatch(prefs, (int)g_ldmMinMatch); + if (g_ldmBucketSizeLog != LDM_PARAM_DEFAULT) FIO_setLdmBucketSizeLog(prefs, (int)g_ldmBucketSizeLog); + if (g_ldmHashRateLog != LDM_PARAM_DEFAULT) FIO_setLdmHashRateLog(prefs, (int)g_ldmHashRateLog); + FIO_setAdaptiveMode(prefs, (unsigned)adapt); + FIO_setUseRowMatchFinder(prefs, useRowMatchFinder); + FIO_setAdaptMin(prefs, adaptMin); + FIO_setAdaptMax(prefs, adaptMax); + FIO_setRsyncable(prefs, rsyncable); + FIO_setStreamSrcSize(prefs, streamSrcSize); + FIO_setTargetCBlockSize(prefs, targetCBlockSize); + FIO_setSrcSizeHint(prefs, srcSizeHint); + FIO_setLiteralCompressionMode(prefs, literalCompressionMode); + if (adaptMin > cLevel) cLevel = adaptMin; + if (adaptMax < cLevel) cLevel = adaptMax; + + /* Compare strategies constant with the ground truth */ + { ZSTD_bounds strategyBounds = ZSTD_cParam_getBounds(ZSTD_c_strategy); + assert(ZSTD_NB_STRATEGIES == strategyBounds.upperBound); + (void)strategyBounds; } + if (showDefaultCParams || g_displayLevel >= 4) { - size_t fileNb; - for (fileNb = 0; fileNb < (size_t)filenames->tableSize; fileNb++) { + size_t fileNb; + for (fileNb = 0; fileNb < (size_t)filenames->tableSize; fileNb++) { if (showDefaultCParams) printDefaultCParams(filenames->fileNames[fileNb], dictFileName, cLevel); if (g_displayLevel >= 4) printActualCParams(filenames->fileNames[fileNb], dictFileName, cLevel, &compressionParams); - } - } - + } + } + if (g_displayLevel >= 4) FIO_displayCompressionParameters(prefs); - if ((filenames->tableSize==1) && outFileName) - operationResult = FIO_compressFilename(fCtx, prefs, outFileName, filenames->fileNames[0], dictFileName, cLevel, compressionParams); - else - operationResult = FIO_compressMultipleFilenames(fCtx, prefs, filenames->fileNames, outMirroredDirName, outDirName, outFileName, suffix, dictFileName, cLevel, compressionParams); -#else - (void)contentSize; (void)suffix; (void)adapt; (void)rsyncable; (void)ultra; (void)cLevel; (void)ldmFlag; (void)literalCompressionMode; (void)targetCBlockSize; (void)streamSrcSize; (void)srcSizeHint; (void)ZSTD_strategyMap; (void)useRowMatchFinder; /* not used when ZSTD_NOCOMPRESS set */ - DISPLAY("Compression not supported \n"); -#endif - } else { /* decompression or test */ -#ifndef ZSTD_NODECOMPRESS - if (filenames->tableSize == 1 && outFileName) { - operationResult = FIO_decompressFilename(fCtx, prefs, outFileName, filenames->fileNames[0], dictFileName); - } else { - operationResult = FIO_decompressMultipleFilenames(fCtx, prefs, filenames->fileNames, outMirroredDirName, outDirName, outFileName, dictFileName); - } -#else - DISPLAY("Decompression not supported \n"); -#endif - } - -_end: - FIO_freePreferences(prefs); - FIO_freeContext(fCtx); - if (main_pause) waitEnter(); - UTIL_freeFileNamesTable(filenames); - UTIL_freeFileNamesTable(file_of_names); -#ifndef ZSTD_NOTRACE - TRACE_finish(); -#endif - - return operationResult; -} + if ((filenames->tableSize==1) && outFileName) + operationResult = FIO_compressFilename(fCtx, prefs, outFileName, filenames->fileNames[0], dictFileName, cLevel, compressionParams); + else + operationResult = FIO_compressMultipleFilenames(fCtx, prefs, filenames->fileNames, outMirroredDirName, outDirName, outFileName, suffix, dictFileName, cLevel, compressionParams); +#else + (void)contentSize; (void)suffix; (void)adapt; (void)rsyncable; (void)ultra; (void)cLevel; (void)ldmFlag; (void)literalCompressionMode; (void)targetCBlockSize; (void)streamSrcSize; (void)srcSizeHint; (void)ZSTD_strategyMap; (void)useRowMatchFinder; /* not used when ZSTD_NOCOMPRESS set */ + DISPLAY("Compression not supported \n"); +#endif + } else { /* decompression or test */ +#ifndef ZSTD_NODECOMPRESS + if (filenames->tableSize == 1 && outFileName) { + operationResult = FIO_decompressFilename(fCtx, prefs, outFileName, filenames->fileNames[0], dictFileName); + } else { + operationResult = FIO_decompressMultipleFilenames(fCtx, prefs, filenames->fileNames, outMirroredDirName, outDirName, outFileName, dictFileName); + } +#else + DISPLAY("Decompression not supported \n"); +#endif + } + +_end: + FIO_freePreferences(prefs); + FIO_freeContext(fCtx); + if (main_pause) waitEnter(); + UTIL_freeFileNamesTable(filenames); + UTIL_freeFileNamesTable(file_of_names); +#ifndef ZSTD_NOTRACE + TRACE_finish(); +#endif + + return operationResult; +} diff --git a/contrib/libs/zstd/programs/zstdcli_trace.c b/contrib/libs/zstd/programs/zstdcli_trace.c index b3b977feb5..8847839949 100644 --- a/contrib/libs/zstd/programs/zstdcli_trace.c +++ b/contrib/libs/zstd/programs/zstdcli_trace.c @@ -1,172 +1,172 @@ -/* - * Copyright (c) Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#include "zstdcli_trace.h" - -#include <stdio.h> -#include <stdlib.h> - -#include "timefn.h" -#include "util.h" - -#define ZSTD_STATIC_LINKING_ONLY -#include "../lib/zstd.h" -/* We depend on the trace header to avoid duplicating the ZSTD_trace struct. - * But, we check the version so it is compatible with dynamic linking. - */ -#include "../lib/common/zstd_trace.h" -/* We only use macros from threading.h so it is compatible with dynamic linking */ -#include "../lib/common/threading.h" - -#if ZSTD_TRACE - -static FILE* g_traceFile = NULL; -static int g_mutexInit = 0; -static ZSTD_pthread_mutex_t g_mutex; -static UTIL_time_t g_enableTime = UTIL_TIME_INITIALIZER; - -void TRACE_enable(char const* filename) -{ - int const writeHeader = !UTIL_isRegularFile(filename); - if (g_traceFile) - fclose(g_traceFile); - g_traceFile = fopen(filename, "a"); - if (g_traceFile && writeHeader) { - /* Fields: - * algorithm - * version - * method - * streaming - * level - * workers - * dictionary size - * uncompressed size - * compressed size - * duration nanos - * compression ratio - * speed MB/s - */ - fprintf(g_traceFile, "Algorithm, Version, Method, Mode, Level, Workers, Dictionary Size, Uncompressed Size, Compressed Size, Duration Nanos, Compression Ratio, Speed MB/s\n"); - } - g_enableTime = UTIL_getTime(); - if (!g_mutexInit) { - if (!ZSTD_pthread_mutex_init(&g_mutex, NULL)) { - g_mutexInit = 1; - } else { - TRACE_finish(); - } - } -} - -void TRACE_finish(void) -{ - if (g_traceFile) { - fclose(g_traceFile); - } - g_traceFile = NULL; - if (g_mutexInit) { - ZSTD_pthread_mutex_destroy(&g_mutex); - g_mutexInit = 0; - } -} - -static void TRACE_log(char const* method, PTime duration, ZSTD_Trace const* trace) -{ - int level = 0; - int workers = 0; - double const ratio = (double)trace->uncompressedSize / (double)trace->compressedSize; - double const speed = ((double)trace->uncompressedSize * 1000) / (double)duration; - if (trace->params) { - ZSTD_CCtxParams_getParameter(trace->params, ZSTD_c_compressionLevel, &level); - ZSTD_CCtxParams_getParameter(trace->params, ZSTD_c_nbWorkers, &workers); - } - assert(g_traceFile != NULL); - - ZSTD_pthread_mutex_lock(&g_mutex); - /* Fields: - * algorithm - * version - * method - * streaming - * level - * workers - * dictionary size - * uncompressed size - * compressed size - * duration nanos - * compression ratio - * speed MB/s - */ - fprintf(g_traceFile, - "zstd, %u, %s, %s, %d, %d, %llu, %llu, %llu, %llu, %.2f, %.2f\n", - trace->version, - method, - trace->streaming ? "streaming" : "single-pass", - level, - workers, - (unsigned long long)trace->dictionarySize, - (unsigned long long)trace->uncompressedSize, - (unsigned long long)trace->compressedSize, - (unsigned long long)duration, - ratio, - speed); - ZSTD_pthread_mutex_unlock(&g_mutex); -} - -/** - * These symbols override the weak symbols provided by the library. - */ - -ZSTD_TraceCtx ZSTD_trace_compress_begin(ZSTD_CCtx const* cctx) -{ - (void)cctx; - if (g_traceFile == NULL) - return 0; - return (ZSTD_TraceCtx)UTIL_clockSpanNano(g_enableTime); -} - -void ZSTD_trace_compress_end(ZSTD_TraceCtx ctx, ZSTD_Trace const* trace) -{ - PTime const beginNanos = (PTime)ctx; - PTime const endNanos = UTIL_clockSpanNano(g_enableTime); - PTime const durationNanos = endNanos > beginNanos ? endNanos - beginNanos : 0; - assert(g_traceFile != NULL); - assert(trace->version == ZSTD_VERSION_NUMBER); /* CLI version must match. */ - TRACE_log("compress", durationNanos, trace); -} - -ZSTD_TraceCtx ZSTD_trace_decompress_begin(ZSTD_DCtx const* dctx) -{ - (void)dctx; - if (g_traceFile == NULL) - return 0; - return (ZSTD_TraceCtx)UTIL_clockSpanNano(g_enableTime); -} - -void ZSTD_trace_decompress_end(ZSTD_TraceCtx ctx, ZSTD_Trace const* trace) -{ - PTime const beginNanos = (PTime)ctx; - PTime const endNanos = UTIL_clockSpanNano(g_enableTime); - PTime const durationNanos = endNanos > beginNanos ? endNanos - beginNanos : 0; - assert(g_traceFile != NULL); - assert(trace->version == ZSTD_VERSION_NUMBER); /* CLI version must match. */ - TRACE_log("decompress", durationNanos, trace); -} - -#else /* ZSTD_TRACE */ - -void TRACE_enable(char const* filename) -{ - (void)filename; -} - -void TRACE_finish(void) {} - -#endif /* ZSTD_TRACE */ +/* + * Copyright (c) Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + +#include "zstdcli_trace.h" + +#include <stdio.h> +#include <stdlib.h> + +#include "timefn.h" +#include "util.h" + +#define ZSTD_STATIC_LINKING_ONLY +#include "../lib/zstd.h" +/* We depend on the trace header to avoid duplicating the ZSTD_trace struct. + * But, we check the version so it is compatible with dynamic linking. + */ +#include "../lib/common/zstd_trace.h" +/* We only use macros from threading.h so it is compatible with dynamic linking */ +#include "../lib/common/threading.h" + +#if ZSTD_TRACE + +static FILE* g_traceFile = NULL; +static int g_mutexInit = 0; +static ZSTD_pthread_mutex_t g_mutex; +static UTIL_time_t g_enableTime = UTIL_TIME_INITIALIZER; + +void TRACE_enable(char const* filename) +{ + int const writeHeader = !UTIL_isRegularFile(filename); + if (g_traceFile) + fclose(g_traceFile); + g_traceFile = fopen(filename, "a"); + if (g_traceFile && writeHeader) { + /* Fields: + * algorithm + * version + * method + * streaming + * level + * workers + * dictionary size + * uncompressed size + * compressed size + * duration nanos + * compression ratio + * speed MB/s + */ + fprintf(g_traceFile, "Algorithm, Version, Method, Mode, Level, Workers, Dictionary Size, Uncompressed Size, Compressed Size, Duration Nanos, Compression Ratio, Speed MB/s\n"); + } + g_enableTime = UTIL_getTime(); + if (!g_mutexInit) { + if (!ZSTD_pthread_mutex_init(&g_mutex, NULL)) { + g_mutexInit = 1; + } else { + TRACE_finish(); + } + } +} + +void TRACE_finish(void) +{ + if (g_traceFile) { + fclose(g_traceFile); + } + g_traceFile = NULL; + if (g_mutexInit) { + ZSTD_pthread_mutex_destroy(&g_mutex); + g_mutexInit = 0; + } +} + +static void TRACE_log(char const* method, PTime duration, ZSTD_Trace const* trace) +{ + int level = 0; + int workers = 0; + double const ratio = (double)trace->uncompressedSize / (double)trace->compressedSize; + double const speed = ((double)trace->uncompressedSize * 1000) / (double)duration; + if (trace->params) { + ZSTD_CCtxParams_getParameter(trace->params, ZSTD_c_compressionLevel, &level); + ZSTD_CCtxParams_getParameter(trace->params, ZSTD_c_nbWorkers, &workers); + } + assert(g_traceFile != NULL); + + ZSTD_pthread_mutex_lock(&g_mutex); + /* Fields: + * algorithm + * version + * method + * streaming + * level + * workers + * dictionary size + * uncompressed size + * compressed size + * duration nanos + * compression ratio + * speed MB/s + */ + fprintf(g_traceFile, + "zstd, %u, %s, %s, %d, %d, %llu, %llu, %llu, %llu, %.2f, %.2f\n", + trace->version, + method, + trace->streaming ? "streaming" : "single-pass", + level, + workers, + (unsigned long long)trace->dictionarySize, + (unsigned long long)trace->uncompressedSize, + (unsigned long long)trace->compressedSize, + (unsigned long long)duration, + ratio, + speed); + ZSTD_pthread_mutex_unlock(&g_mutex); +} + +/** + * These symbols override the weak symbols provided by the library. + */ + +ZSTD_TraceCtx ZSTD_trace_compress_begin(ZSTD_CCtx const* cctx) +{ + (void)cctx; + if (g_traceFile == NULL) + return 0; + return (ZSTD_TraceCtx)UTIL_clockSpanNano(g_enableTime); +} + +void ZSTD_trace_compress_end(ZSTD_TraceCtx ctx, ZSTD_Trace const* trace) +{ + PTime const beginNanos = (PTime)ctx; + PTime const endNanos = UTIL_clockSpanNano(g_enableTime); + PTime const durationNanos = endNanos > beginNanos ? endNanos - beginNanos : 0; + assert(g_traceFile != NULL); + assert(trace->version == ZSTD_VERSION_NUMBER); /* CLI version must match. */ + TRACE_log("compress", durationNanos, trace); +} + +ZSTD_TraceCtx ZSTD_trace_decompress_begin(ZSTD_DCtx const* dctx) +{ + (void)dctx; + if (g_traceFile == NULL) + return 0; + return (ZSTD_TraceCtx)UTIL_clockSpanNano(g_enableTime); +} + +void ZSTD_trace_decompress_end(ZSTD_TraceCtx ctx, ZSTD_Trace const* trace) +{ + PTime const beginNanos = (PTime)ctx; + PTime const endNanos = UTIL_clockSpanNano(g_enableTime); + PTime const durationNanos = endNanos > beginNanos ? endNanos - beginNanos : 0; + assert(g_traceFile != NULL); + assert(trace->version == ZSTD_VERSION_NUMBER); /* CLI version must match. */ + TRACE_log("decompress", durationNanos, trace); +} + +#else /* ZSTD_TRACE */ + +void TRACE_enable(char const* filename) +{ + (void)filename; +} + +void TRACE_finish(void) {} + +#endif /* ZSTD_TRACE */ diff --git a/contrib/libs/zstd/programs/zstdcli_trace.h b/contrib/libs/zstd/programs/zstdcli_trace.h index 38c27dc04c..597d6a588e 100644 --- a/contrib/libs/zstd/programs/zstdcli_trace.h +++ b/contrib/libs/zstd/programs/zstdcli_trace.h @@ -1,24 +1,24 @@ -/* - * Copyright (c) Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTDCLI_TRACE_H -#define ZSTDCLI_TRACE_H - -/** - * Enable tracing - log to filename. - */ -void TRACE_enable(char const* filename); - -/** - * Shut down the tracing library. - */ -void TRACE_finish(void); - -#endif /* ZSTDCLI_TRACE_H */ +/* + * Copyright (c) Facebook, Inc. + * All rights reserved. + * + * This source code is licensed under both the BSD-style license (found in the + * LICENSE file in the root directory of this source tree) and the GPLv2 (found + * in the COPYING file in the root directory of this source tree). + * You may select, at your option, one of the above-listed licenses. + */ + +#ifndef ZSTDCLI_TRACE_H +#define ZSTDCLI_TRACE_H + +/** + * Enable tracing - log to filename. + */ +void TRACE_enable(char const* filename); + +/** + * Shut down the tracing library. + */ +void TRACE_finish(void); + +#endif /* ZSTDCLI_TRACE_H */ diff --git a/contrib/libs/zstd/ya.make b/contrib/libs/zstd/ya.make index 70d6705d1e..737296157c 100644 --- a/contrib/libs/zstd/ya.make +++ b/contrib/libs/zstd/ya.make @@ -19,7 +19,7 @@ LICENSE( BSD-3-Clause AND MIT ) - + LICENSE_TEXTS(.yandex_meta/licenses.list.txt) PEERDIR( @@ -38,7 +38,7 @@ NO_RUNTIME() CFLAGS( -DZSTD_LEGACY_SUPPORT=1 - -DZSTD_MULTITHREAD + -DZSTD_MULTITHREAD ) IF (ARCH_X86_64 AND NOT MSVC) @@ -51,45 +51,45 @@ IF (ARCH_X86_64 AND NOT MSVC) ENDIF() SRCS( - lib/common/debug.c - lib/common/entropy_common.c - lib/common/error_private.c - lib/common/fse_decompress.c - lib/common/pool.c - lib/common/threading.c - lib/common/zstd_common.c - lib/compress/fse_compress.c - lib/compress/hist.c - lib/compress/huf_compress.c - lib/compress/zstd_compress.c - lib/compress/zstd_compress_literals.c - lib/compress/zstd_compress_sequences.c - lib/compress/zstd_compress_superblock.c - lib/compress/zstd_double_fast.c - lib/compress/zstd_fast.c - lib/compress/zstd_lazy.c - lib/compress/zstd_ldm.c - lib/compress/zstd_opt.c - lib/compress/zstdmt_compress.c - lib/decompress/huf_decompress.c - lib/decompress/zstd_ddict.c - lib/decompress/zstd_decompress.c - lib/decompress/zstd_decompress_block.c - lib/dictBuilder/cover.c - lib/dictBuilder/divsufsort.c - lib/dictBuilder/fastcover.c - lib/dictBuilder/zdict.c - lib/legacy/zstd_v01.c - lib/legacy/zstd_v02.c - lib/legacy/zstd_v03.c - lib/legacy/zstd_v04.c - lib/legacy/zstd_v05.c - lib/legacy/zstd_v06.c - lib/legacy/zstd_v07.c + lib/common/debug.c + lib/common/entropy_common.c + lib/common/error_private.c + lib/common/fse_decompress.c + lib/common/pool.c + lib/common/threading.c + lib/common/zstd_common.c + lib/compress/fse_compress.c + lib/compress/hist.c + lib/compress/huf_compress.c + lib/compress/zstd_compress.c + lib/compress/zstd_compress_literals.c + lib/compress/zstd_compress_sequences.c + lib/compress/zstd_compress_superblock.c + lib/compress/zstd_double_fast.c + lib/compress/zstd_fast.c + lib/compress/zstd_lazy.c + lib/compress/zstd_ldm.c + lib/compress/zstd_opt.c + lib/compress/zstdmt_compress.c + lib/decompress/huf_decompress.c + lib/decompress/zstd_ddict.c + lib/decompress/zstd_decompress.c + lib/decompress/zstd_decompress_block.c + lib/dictBuilder/cover.c + lib/dictBuilder/divsufsort.c + lib/dictBuilder/fastcover.c + lib/dictBuilder/zdict.c + lib/legacy/zstd_v01.c + lib/legacy/zstd_v02.c + lib/legacy/zstd_v03.c + lib/legacy/zstd_v04.c + lib/legacy/zstd_v05.c + lib/legacy/zstd_v06.c + lib/legacy/zstd_v07.c ) END() - -RECURSE( - programs/zstd -) + +RECURSE( + programs/zstd +) |