summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Add missing libavutil/time.h includes.Anton Khirnov2012-07-2810-0/+10
|
* vp3: don't compile mmx IDCT functions on x86-64.Ronald S. Bultje2012-07-272-0/+5
| | | | | 64-bit CPUs always have SSE2, and a SSE2 version exists, thus the MMX version will never be used.
* h264_loopfilter: port x86 simd to cpuflags.Ronald S. Bultje2012-07-273-121/+120
|
* build: support non-standard replacements for -c flagMans Rullgard2012-07-282-4/+11
| | | | | | | | This allows non-standard replacements for the -c compiler flag. Some compilers use other flags or no flag at all in place of the usual one. Signed-off-by: Mans Rullgard <[email protected]>
* build: support non-standard replacements for -E flagMans Rullgard2012-07-281-2/+10
| | | | | | | | | | This allows using non-standard flags for running the C preprocessor. The -o flag must be included in this setting due to strange syntax required by some compilers. Set the correct flags for tms470. Signed-off-by: Mans Rullgard <[email protected]>
* h264_chromamc_10bit: port x86 simd to cpuflags.Ronald S. Bultje2012-07-272-28/+28
|
* vp3: port x86 SIMD to cpuflags.Ronald S. Bultje2012-07-271-47/+47
|
* build: Use portable compiler flag constructs in header compilation ruleDiego Biurrun2012-07-281-1/+1
|
* x11grab: cosmetics: consistent naming for x11grab-related thingsDiego Biurrun2012-07-285-17/+16
|
* build: Rename YASMDEP variable to DEPYASM for consistencyDiego Biurrun2012-07-283-3/+3
|
* hls: Proceed to the next segment at any error codeMartin Storsjö2012-07-281-2/+0
| | | | | | | | | | | | | Previously, we returned any error code except AVERROR_EOF to the caller - only if AVERROR_EOF or 0 was returned, we proceeded to the next segment. With some setups of web servers, using Connection: close in https and GnuTLS, we don't get a clean error code at the end of segments. In those cases, just proceed to the next segment. Tested-by: Antti Seppälä <[email protected]> Signed-off-by: Martin Storsjö <[email protected]>
* tls: Return AVERROR_EOF if the TLS_read/write functions return 0Martin Storsjö2012-07-281-2/+2
| | | | | | | | | | OpenSSL returns 0 when the peer has closed the connection. GnuTLS doesn't return that though, but returns GNUTLS_E_UNEXPECTED_PACKET_LENGTH if the connection simply is closed without a clean close notify packet. Tested-by: Antti Seppälä <[email protected]> Signed-off-by: Martin Storsjö <[email protected]>
* rv34: port x86 SIMD to cpuflags.Ronald S. Bultje2012-07-271-5/+6
|
* vp56: only compile MMX SIMD on x86-32.Ronald S. Bultje2012-07-272-0/+4
| | | | | All x86-64 CPUs have SSE2, so the MMX version will never be used. This leads to smaller binaries.
* vp56: port x86 simd to cpuflags.Ronald S. Bultje2012-07-271-19/+15
|
* proresdsp: port x86 assembly to cpuflags.Ronald S. Bultje2012-07-271-21/+18
|
* lavr: x86: improve non-SSE4 version of S16_TO_S32_SX macroJustin Ruggles2012-07-271-1/+2
| | | | | Removes a false dependency on existing contents of the 2nd dst register, giving better performance for OOE.
* lavfi: better channel layout negotiationJustin Ruggles2012-07-271-11/+72
| | | | | | | Allow substitution of channel pairs in the input for nearby channel pairs in the output in order to get a closer match. Also weigh LFE channel mismatch differently to favor matching the same layout without LFE over one less channel with LFE.
* alac: check for truncated packetsJustin Ruggles2012-07-271-4/+11
| | | | | This will give a clearer error message when the error is caused by a truncated packet.
* alac: reverse lpc coeff order, simplify filterJustin Ruggles2012-07-271-13/+12
| | | | Reversing the lpc coefficient order simplifies indexing in the filter.
* lavr: add x86-optimized mixing functionsJustin Ruggles2012-07-273-3/+427
| | | | | Adds optimized functions for mixing 3 through 8 input channels to 1 and 2 output channels in fltp or s16p format with flt coeffs.
* x86: add support for fmaddps fma4 instruction with abstraction to avx/sseJustin Ruggles2012-07-272-5/+16
|
* tscc2: fix typo in array indexKostya Shishkov2012-07-271-2/+2
|
* build: use COMPILE template for HOSTOBJSMans Rullgard2012-07-264-8/+3
| | | | Signed-off-by: Mans Rullgard <[email protected]>
* build: do full flag handling for all compiler-type toolsMans Rullgard2012-07-262-123/+182
| | | | | | | This adds a full identification probe of CC, AS, LD and HOSTCC, and sets up correct flags and dependency tracking for each. Signed-off-by: Mans Rullgard <[email protected]>
* eval: fix printing of NaN in eval fate test.Ronald S. Bultje2012-07-261-1/+4
| | | | | This fixes "make fate-eval" on MSVC builds. Without this, the test outputs "-1.#NaN" instead of "nan" on MSVS 2010.
* build: Rename aandct component to more descriptive aandcttablesDiego Biurrun2012-07-262-14/+14
|
* mpegaudio: bury inline asm under HAVE_INLINE_ASM.Ronald S. Bultje2012-07-261-0/+5
|
* x86inc: automatically insert vzeroupper for YMM functions.Ronald S. Bultje2012-07-266-53/+12
|
* rtmp: Check the buffer length of ping packetsSamuel Pitoiset2012-07-261-0/+12
| | | | Signed-off-by: Martin Storsjö <[email protected]>
* rtmp: Allow having more unknown data at the end of a chunk size packet ↵Samuel Pitoiset2012-07-261-2/+2
| | | | | | without failing Signed-off-by: Martin Storsjö <[email protected]>
* rtmp: Prevent reading outside of an allocate buffer when receiving server ↵Samuel Pitoiset2012-07-261-0/+7
| | | | | | bandwidth packets Signed-off-by: Martin Storsjö <[email protected]>
* libopenjpeg: support YUV and deep RGB pixel formatsMichael Bradshaw2012-07-261-52/+247
| | | | | | | Based on FFmpeg version from commit 3275981207e30e140cffaea334ac390f1a04266a Signed-off-by: Luca Barbato <[email protected]>
* Fix typo in v410 decoder.Carl Eugen Hoyos2012-07-261-2/+2
| | | | | Signed-off-by: Derek Buitenhuis <[email protected]> Signed-off-by: Kostya Shishkov <[email protected]>
* vf_yadif: unset cur_buf on the input link.Anton Khirnov2012-07-261-0/+1
| | | | | | | The buffer is stored internally, so this prevents it from being unreffed automatically. Signed-off-by: Luca Barbato <[email protected]>
* vf_overlay: ensure the overlay frame does not get leaked.Anton Khirnov2012-07-261-0/+1
| | | | Signed-off-by: Luca Barbato <[email protected]>
* vf_overlay: prevent premature freeing of cur_bufMichael Niedermayer2012-07-261-0/+1
| | | | | Signed-off-by: Anton Khirnov <[email protected]> Signed-off-by: Luca Barbato <[email protected]>
* Support urlencoded http authentication credentialsAntti Seppälä2012-07-265-12/+146
| | | | | | | | | It should be possible to specify usernames in http requests containing urlencoded characters. This patch adds support for decoding the auth strings. Signed-off-by: Antti Seppälä <[email protected]> Signed-off-by: Martin Storsjö <[email protected]>
* rtmp: Return an error when the client bandwidth is incorrectSamuel Pitoiset2012-07-251-2/+10
| | | | Signed-off-by: Martin Storsjö <[email protected]>
* rtmp: Return proper error code in handle_server_bwSamuel Pitoiset2012-07-251-1/+1
| | | | Signed-off-by: Martin Storsjö <[email protected]>
* rtmp: Return proper error code in handle_client_bwSamuel Pitoiset2012-07-251-1/+1
| | | | Signed-off-by: Martin Storsjö <[email protected]>
* rtmp: Return proper error codes in handle_chunk_sizeSamuel Pitoiset2012-07-251-2/+2
| | | | Signed-off-by: Martin Storsjö <[email protected]>
* lavr: x86: add missing vzeroupper in ff_mix_1_to_2_fltp_flt()Justin Ruggles2012-07-251-0/+5
|
* vp8: Replace x*155/100 by x*101581>>16.Ronald S. Bultje2012-07-251-6/+7
| | | | | | Idea stolen from webp (by Pascal Massimino) - because it's Cool. Signed-off-by: Derek Buitenhuis <[email protected]>
* vp3: don't use calls to inline asm in yasm code.Ronald S. Bultje2012-07-251-41/+79
| | | | | | | | Mixing yasm and inline asm is a bad idea, since if either yasm or inline asm is not supported by your toolchain, all of the asm stops working. Thus, better to use either one or the other alone. Signed-off-by: Derek Buitenhuis <[email protected]>
* x86/dsputil: put inline asm under HAVE_INLINE_ASM.Ronald S. Bultje2012-07-258-19/+74
| | | | | | | This allows compiling with compilers that don't support gcc-style inline assembly. Signed-off-by: Derek Buitenhuis <[email protected]>
* dsputil_mmx: fix incorrect assembly codeYang Wang2012-07-251-9/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | In ff_put_pixels_clamped_mmx(), there are two assembly code blocks. In the first block (in the unrolled loop), the instructions "movq 8%3, %%mm1 \n\t", and so forth, have problems. From above instruction, it is clear what the programmer wants: a load from p + 8. But this assembly code doesn’t guarantee that. It only works if the compiler puts p in a register to produce an instruction like this: "movq 8(%edi), %mm1". During compiler optimization, it is possible that the compiler will be able to constant propagate into p. Suppose p = &x[10000]. Then operand 3 can become 10000(%edi), where %edi holds &x. And the instruction becomes "movq 810000(%edx)". That is, it will stride by 810000 instead of 8. This will cause a segmentation fault. This error was fixed in the second block of the assembly code, but not in the unrolled loop. How to reproduce: This error is exposed when we build using Intel C++ Compiler, with IPO+PGO optimization enabled. Crashed when decoding an MJPEG video. Signed-off-by: Michael Niedermayer <[email protected]> Signed-off-by: Derek Buitenhuis <[email protected]>
* rtmp: Factorize the code by adding handle_invokeSamuel Pitoiset2012-07-251-66/+77
| | | | Signed-off-by: Martin Storsjö <[email protected]>
* rtmp: Factorize the code by adding handle_chunk_sizeSamuel Pitoiset2012-07-251-15/+30
| | | | Signed-off-by: Martin Storsjö <[email protected]>
* rtmp: Factorize the code by adding handle_pingSamuel Pitoiset2012-07-251-4/+16
| | | | Signed-off-by: Martin Storsjö <[email protected]>
generated by cgit v1.2.3 (git 2.25.1) at 2025-09-23 13:40:08 +0000