summaryrefslogtreecommitdiff
path: root/libavcodec/x86
Commit message (Collapse)AuthorAgeFilesLines
* optimize h264_loop_filter_strength_mmx2Loren Merritt2010-01-261-74/+70
| | | | | | 244->160 cycles on core2 Originally committed as revision 21462 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Implement an sse version of scalarproduct_float().Alex Converse2010-01-222-0/+29
| | | | Originally committed as revision 21386 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Move array specifiers outside DECLARE_ALIGNED() invocationsMåns Rullgård2010-01-2215-51/+51
| | | | Originally committed as revision 21377 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Use two separate memory arguments since 8+() is invalid gas syntaxDavid Conrad2010-01-211-19/+19
| | | | Originally committed as revision 21360 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Attempt to fix asm compilation failure.Michael Niedermayer2010-01-201-21/+20
| | | | | | Only tested on gcc 4 & x86_64. Originally committed as revision 21355 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Move COPY3_IF_LT to lavc/mathops.hMåns Rullgård2010-01-201-0/+12
| | | | | | | | This obscure macro is only used in motion_est.c so having it in lavc makes more sense. See discussion here: http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/2008-November/056561.html Originally committed as revision 21346 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Use constant offsets for memory operands since gcc is unable toDavid Conrad2010-01-201-12/+8
| | | | | | This fixes gcc failing to fit 6 memory locations into 7 registers on x86-32 Originally committed as revision 21337 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Fix h264_loop_filter_strength_mmx2() so it works with b frames.Michael Niedermayer2010-01-191-0/+34
| | | | Originally committed as revision 21327 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Remove -2 -> -1 remapping, its not needed anymore as we must remap allMichael Niedermayer2010-01-191-4/+0
| | | | | | references per LUT anyway. Originally committed as revision 21323 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Fix XvMC. XvMCCreateBlocks() may not allocate 16-byte aligned blocks,Gwenole Beauchesne2010-01-041-1/+3
| | | | | | so we can't use SSE-optimized routines. Originally committed as revision 21011 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Reduce number of ASM constraints for ff_lpc_compute_autocorr_sse2 since itReimar Döffinger2010-01-021-9/+10
| | | | | | | causes no significant speed difference and can avoid compilation issues with --enable-pic. Originally committed as revision 21003 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Get rid of pointless CONFIG_ANY_H263 preprocessor definition.Diego Biurrun2009-12-301-4/+3
| | | | Originally committed as revision 20975 to svn://svn.ffmpeg.org/ffmpeg/trunk
* fix a crash in ape decoding on x86_32 sse2Loren Merritt2009-12-081-1/+1
| | | | Originally committed as revision 20777 to svn://svn.ffmpeg.org/ffmpeg/trunk
* slightly faster scalarproduct_and_madd_int16_ssse3 on penryn, no change on ↵Loren Merritt2009-12-051-5/+13
| | | | | | conroe Originally committed as revision 20743 to svn://svn.ffmpeg.org/ffmpeg/trunk
* r20739 broke compilation on systems without yasmLoren Merritt2009-12-051-7/+6
| | | | Originally committed as revision 20742 to svn://svn.ffmpeg.org/ffmpeg/trunk
* refactor and optimize scalarproductLoren Merritt2009-12-052-47/+136
| | | | | | | | 29-105% faster apply_filter, 6-90% faster ape decoding on core2 (Any x86 other than core2 probably gets much less, since this is mostly due to ssse3 cachesplit avoidance and I haven't written the full gamut of other cachesplit modes.) 9-123% faster ape decoding on G4. Originally committed as revision 20739 to svn://svn.ffmpeg.org/ffmpeg/trunk
* port ape dsp functions from sse2 to mmxLoren Merritt2009-12-032-75/+93
| | | | | | now requires yasm Originally committed as revision 20722 to svn://svn.ffmpeg.org/ffmpeg/trunk
* s/movdqa/movaps/ in sse1 fft. (regression in r20293)Loren Merritt2009-10-251-0/+2
| | | | Originally committed as revision 20371 to svn://svn.ffmpeg.org/ffmpeg/trunk
* fix linking on systems with a function name prefix (10l in r20287)Loren Merritt2009-10-181-1/+1
| | | | Originally committed as revision 20294 to svn://svn.ffmpeg.org/ffmpeg/trunk
* sync yasm macros to x264Loren Merritt2009-10-181-5/+3
| | | | Originally committed as revision 20293 to svn://svn.ffmpeg.org/ffmpeg/trunk
* huffyuv: add some const qualifiersLoren Merritt2009-10-183-7/+7
| | | | Originally committed as revision 20290 to svn://svn.ffmpeg.org/ffmpeg/trunk
* simd add_hfyu_left_predictionLoren Merritt2009-10-183-0/+83
| | | | | | | 2.2x faster than C on conroe, 3.6x on penryn. 4-6% faster huffyuv decoding if using left or plane mode and yuv Originally committed as revision 20287 to svn://svn.ffmpeg.org/ffmpeg/trunk
* add CONFIG_LPC to the build system for lpc dsputil functions. fixes buildJustin Ruggles2009-10-181-0/+2
| | | | | | problems when lpc.c is not compiled. Originally committed as revision 20285 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Move autocorrelation function from flacenc.c to lpc.c. Also rename theJustin Ruggles2009-10-172-6/+4
| | | | | | | | corresponding dsputil functions and remove their dependency on the FLAC encoder. Fixes Issue1486. Originally committed as revision 20266 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Use MANGLE in cavsdsp, the current version using "m" constraints will notReimar Döffinger2009-10-011-23/+23
| | | | | | compile on e.g. OpenBSD due to running out of registers. Originally committed as revision 20123 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Replace several #ifdef PIC with the more obvious and correctReimar Döffinger2009-09-302-10/+10
| | | | | | #if !HAVE_EBX_AVAILABLE, since all it does is avoid using ebx. Originally committed as revision 20094 to svn://svn.ffmpeg.org/ffmpeg/trunk
* cosmetics: fix indentation after previous commitMåns Rullgård2009-09-271-32/+32
| | | | Originally committed as revision 20062 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Drop unused args from vector_fmul_add_add, simpify code, and renameMåns Rullgård2009-09-271-70/+6
| | | | | | | | | The src3 and step arguments to vector_fmul_add_add() are always zero and one, respectively. This removes these arguments from the function, simplifies the code accordingly, and renames the function to better match the new operation. Originally committed as revision 20061 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Merge FFTContext and MDCTContextMåns Rullgård2009-09-203-18/+18
| | | | Originally committed as revision 19931 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Move per-arch fft init bits into the corresponding subdirsMåns Rullgård2009-09-154-0/+82
| | | | Originally committed as revision 19864 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Move declarations of some mmx functions to dsputil_mmx.hMåns Rullgård2009-08-294-0/+8
| | | | Originally committed as revision 19739 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Mark "i" parameter of vector_clipf_sse() as early-clobberVitor Sessak2009-08-271-1/+1
| | | | Originally committed as revision 19731 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Mark parameter src of vector_clipf() as constVitor Sessak2009-08-271-1/+1
| | | | Originally committed as revision 19729 to svn://svn.ffmpeg.org/ffmpeg/trunk
* SSE optimized vector_clipf(). 10% faster TwinVQ decoding.Vitor Sessak2009-08-271-0/+35
| | | | Originally committed as revision 19728 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Update x264 asm code to latest to add support for 64-bit Windows.John Adcock2009-08-045-142/+507
| | | | | | | | | | Use the new x86inc features to support 64-bit Windows on all non-x264 nasm assembly code as well. Patch by John Adcock, dscaler.johnad AT googlemail DOT com. Win64 changes originally by Anton Mitrofanov. x86util changes mostly by Holger Lubitz. Originally committed as revision 19580 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Do not check for both CONFIG_VC1_DECODER and CONFIG_WMV3_DECODER,Diego Biurrun2009-07-291-1/+1
| | | | | | the former depends upon the latter. Originally committed as revision 19533 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Do not redundantly check for both CONFIG_THEORA_DECODER and CONFIG_VP3_DECODER.Diego Biurrun2009-07-221-2/+2
| | | | | | The Theora decoder depends on the VP3 decoder. Originally committed as revision 19492 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Icc 11.1 still does not align the stack pointer, disable some x264 functions.Carl Eugen Hoyos2009-07-171-1/+1
| | | | Originally committed as revision 19454 to svn://svn.ffmpeg.org/ffmpeg/trunk
* SSE version of clear_blocksJason Garrett-Glaser2009-06-161-2/+26
| | | | Originally committed as revision 19206 to svn://svn.ffmpeg.org/ffmpeg/trunk
* idct_dc for VC-1/WMV3 decoder; ~11% faster decoding overall.Jason Garrett-Glaser2009-06-161-0/+203
| | | | | | | | | Includes mmx2 asm for the various functions. Note that the actual idct still does not have an x86 SIMD implemtation. For wmv3 files using regular idct, the decoder just falls back to simple_idct, since simple_idct_dc doesn't exist (yet). Originally committed as revision 19204 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Replace more uses of __attribute__((aligned)) by DECLARE_ALIGNED.Ramiro Polla2009-06-042-2/+2
| | | | Originally committed as revision 19089 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Use fewer macros in x86-optimized mlpdsp.Ramiro Polla2009-06-031-19/+10
| | | | | | Fixes compilation on 32-bit llvm which didn't allow a cast in an m operand. Originally committed as revision 19086 to svn://svn.ffmpeg.org/ffmpeg/trunk
* H264: Fix out of bounds reads in SSSE3 MCAlexander Strange2009-05-301-38/+38
| | | | | | | | | Reading above src[-2] isn't safe, so move loads and palignr ahead 3 pixels to load starting at the first pixel actually used. Fixes issue941. Originally committed as revision 18999 to svn://svn.ffmpeg.org/ffmpeg/trunk
* mlp: Use LABEL_MANGLE() to export label symbols from inside asm block.Ramiro Polla2009-05-251-5/+5
| | | | Originally committed as revision 18935 to svn://svn.ffmpeg.org/ffmpeg/trunk
* MLP DSP functions x86-optimized.Ramiro Polla2009-05-231-0/+190
| | | | | | | | 12.59% overall speedup in x86_32 9.98% overall speedup in x86_64 compared to gcc 4.3.3 Originally committed as revision 18903 to svn://svn.ffmpeg.org/ffmpeg/trunk
* avg_ pixel functions need to use (dst+pix+1)>>1 to average with existingDavid Conrad2009-04-152-16/+22
| | | | | | | pixels, not (dst+pix)>>1. This makes the mmx functions bitexact with the C functions. Originally committed as revision 18527 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VC1: extend MMX qpel MC to include MMX2 avg qpelDavid Conrad2009-04-152-133/+187
| | | | Originally committed as revision 18519 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VC1: add and use avg_no_rnd chroma MC functionsDavid Conrad2009-04-142-0/+11
| | | | Originally committed as revision 18518 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Rename put_no_rnd_h264_chroma* to reflect its usage in VC1 onlyDavid Conrad2009-04-142-4/+4
| | | | Originally committed as revision 18517 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Do not use SSE2 SAD for snow as it requires more alignment than can beMichael Niedermayer2009-04-091-1/+1
| | | | | | | easily provided. Fixes issue315. Originally committed as revision 18404 to svn://svn.ffmpeg.org/ffmpeg/trunk
View Lorry Controller Status