summaryrefslogtreecommitdiffstats
path: root/libavcodec/x86/dcadsp_init.c
Commit message (Collapse)AuthorAgeFilesLines
* x86/dcadsp: add ff_dca_lfe_fir0_fma3James Almer2014-04-051-0/+5
| | | | | | | ~10% faster than the SSE version. Signed-off-by: James Almer <[email protected]> Signed-off-by: Michael Niedermayer <[email protected]>
* x86/synth_filter: compile avx and fma3 functions unconditionallyJames Almer2014-04-051-6/+0
| | | | | | | Fixes compilation failures with "--disable-{avx,fma3} --disable-optimizations" Signed-off-by: James Almer <[email protected]> Signed-off-by: Michael Niedermayer <[email protected]>
* avcodec/x86/dcadsp_init: fix compilation failure without FMA3Michael Niedermayer2014-04-051-0/+4
| | | | | | | alternatively the call could be put under #if or the #if over the function removed Signed-off-by: Michael Niedermayer <[email protected]>
* Merge commit 'c74b86699c86bdf62e8570f41d8a38be5710baa3'Michael Niedermayer2014-04-041-3/+5
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * commit 'c74b86699c86bdf62e8570f41d8a38be5710baa3': x86/synth_filter: add synth_filter_fma3 x86/synth_filter: add synth_filter_avx x86/synth_filter: add synth_filter_sse Conflicts: libavcodec/x86/dcadsp.asm libavcodec/x86/dcadsp_init.c See: 64672098361361cd15d37e36f747ab44de5b80ca See: 68c3ed936a76c3ff7738f602fa90237ac7e3ce08 See: 7fd64e3e36f79204c0eda7cacce6884c14ddc1fb See: aa1f38015cb0d04a5c50a8957dd7aba79f0d8882 See: dfd865e51b890d9be394804bccddf55198f4a251 Merged-by: Michael Niedermayer <[email protected]>
| * x86/synth_filter: add synth_filter_fma3James Almer2014-04-041-0/+6
| | | | | | | | | | Signed-off-by: James Almer <[email protected]> Signed-off-by: Anton Khirnov <[email protected]>
| * x86/synth_filter: add synth_filter_avxJames Almer2014-04-041-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | Sandy Bridge Win64: 180 cycles in ff_synth_filter_inner_sse2 150 cycles in ff_synth_filter_inner_avx Also switch some instructions to a three operand format to avoid assembly errors with Yasm 1.1.0 or older. Signed-off-by: James Almer <[email protected]> Signed-off-by: Anton Khirnov <[email protected]>
| * x86/synth_filter: add synth_filter_sseJames Almer2014-04-041-17/+28
| | | | | | | | | | | | | | Build only on x86_32 targets. Signed-off-by: James Almer <[email protected]> Signed-off-by: Anton Khirnov <[email protected]>
* | Merge commit '3bfdee00cd92ff07c364d4901c4aefda32780756'Michael Niedermayer2014-03-061-2/+2
|\| | | | | | | | | | | | | | | | | | | | | * commit '3bfdee00cd92ff07c364d4901c4aefda32780756': x86: dcadsp: Fix linking with yasm and optimizations disabled Conflicts: libavcodec/x86/dcadsp_init.c See: 206167a295a5c28cec3c38f7308835b0b7e0618f Merged-by: Michael Niedermayer <[email protected]>
| * x86: dcadsp: Fix linking with yasm and optimizations disabledDiego Biurrun2014-03-051-0/+4
| | | | | | | | | | Some optimized functions reference optimized symbols, so the functions must be explicitly disabled when those symbols are unavailable.
* | x86/synth_filter: add synth_filter_fma3James Almer2014-03-051-0/+4
| | | | | | | | | | Signed-off-by: James Almer <[email protected]> Signed-off-by: Michael Niedermayer <[email protected]>
* | x86/synth_filter: add missing HAVE_YASM guardJames Almer2014-03-041-0/+4
| | | | | | | | | | | | | | Should fix compilation failures with --disable-yasm on some compilers Signed-off-by: James Almer <[email protected]> Signed-off-by: Michael Niedermayer <[email protected]>
* | x86/synth_filter: add synth_filter_avxJames Almer2014-03-021-0/+4
| | | | | | | | | | | | | | | | | | Sandy Bridge Win64: 180 cycles on ff_synth_filter_inner_sse2 150 cycles on ff_synth_filter_inner_avx Signed-off-by: James Almer <[email protected]> Signed-off-by: Michael Niedermayer <[email protected]>
* | x86/synth_filter: add synth_filter_sseJames Almer2014-03-011-16/+27
| | | | | | | | | | | | | | Build only on x86_32 targets. Signed-off-by: James Almer <[email protected]> Signed-off-by: Michael Niedermayer <[email protected]>
* | Merge commit '4cb6964244fd6c099383d8b7e99731e72cc844b9'Michael Niedermayer2014-02-281-6/+12
|\| | | | | | | | | | | | | | | | | | | | | * commit '4cb6964244fd6c099383d8b7e99731e72cc844b9': dcadec: simplify decoding of VQ high frequencies Conflicts: configure libavcodec/dcadec.c Merged-by: Michael Niedermayer <[email protected]>
| * dcadec: simplify decoding of VQ high frequenciesChristophe Gisquet2014-02-281-6/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The vector dequantization has a test in a loop preventing effective SIMD implementation. By moving it out of the loop, this loop can be DSPized. Therefore, modify the current DSP implementation. In particular, the DSP implementation no longer has to handle null loop sizes. The decode_hf implementations have following timings: For x86 Arrandale: C SSE SSE2 SSE4 win32: 260 162 119 104 win64: 242 N/A 89 72 The arm NEON optimizations follow in a later patch as external asm. The now unused check for the y modifier in arm inline asm is removed from configure.
* | Merge commit '08e3ea60ff4059341b74be04a428a38f7c3630b0'Michael Niedermayer2014-02-281-1/+1
|\| | | | | | | | | | | | | | | | | | | | | | | * commit '08e3ea60ff4059341b74be04a428a38f7c3630b0': x86: synth filter float: implement SSE2 version Conflicts: libavcodec/x86/dcadsp.asm libavcodec/x86/dcadsp_init.c See: 2cdbcc004837ce092a14f326f24d97a29512a2c3 Merged-by: Michael Niedermayer <[email protected]>
| * x86: synth filter float: implement SSE2 versionChristophe Gisquet2014-02-281-0/+28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Timings for Arrandale: C SSE win32: 2108 334 win64: 1152 322 Factorizing the inner loop with a call/jmp is a >15 cycles cost, even with the jmp destination being aligned. Unrolling for ARCH_X86_64 is a 20 cycles gain. Signed-off-by: Janne Grunau <[email protected]>
| * x86: dcadsp: implement SSE lfe_dirChristophe Gisquet2014-02-281-0/+4
| | | | | | | | | | | | | | | | Results for Arrandale/Windows: 32: 1670 -> 316 64: 728 -> 298 Signed-off-by: Janne Grunau <[email protected]>
* | x86: synth filter float: implement SSE2 versionChristophe Gisquet2014-02-281-0/+28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Timings for Arrandale: C SSE win32: 2108 334 win64: 1152 322 Factorizing the inner loop with a call/jmp is a >15 cycles cost, even with the jmp destination being aligned. Unrolling for ARCH_X86_64 is a 20 cycles gain. Signed-off-by: Michael Niedermayer <[email protected]>
* | x86: dcadsp: implement SSE lfe_dirChristophe Gisquet2014-02-281-0/+4
| | | | | | | | | | | | | | | | Results for Arrandale/Windows: 32: 1670 -> 316 64: 728 -> 298 Signed-off-by: Michael Niedermayer <[email protected]>
* | Merge commit '5b59a9fc6152169599561f04b4f66370edda5c9c'Michael Niedermayer2014-02-081-4/+4
|/ | | | | | | * commit '5b59a9fc6152169599561f04b4f66370edda5c9c': x86: dcadsp: implement int8x8_fmul_int32 Merged-by: Michael Niedermayer <[email protected]>
* x86: dcadsp: implement int8x8_fmul_int32Christophe Gisquet2014-02-071-0/+47
For the callable function (as opposed to the inline one): C SSE SSE2 SSE4 Win32: 47 42 29 26 Win64: 30 33 25 23 The SSE version is neither compiled nor set for ARCH_X86_64, as the inlinable function takes over. Signed-off-by: Janne Grunau <[email protected]>
generated by cgit v1.2.3 (git 2.25.1) at 2025-09-23 14:01:37 +0000