summaryrefslogtreecommitdiffstats
path: root/libavcodec/i386
Commit message (Expand)AuthorAgeFilesLines
* rename always_inline to av_always_inline and move to common.hMåns Rullgård2006-12-081-4/+4
* revert bad checkinMåns Rullgård2006-11-144-32/+32
* move some CFLAGS settings away from config.* writing sectionMåns Rullgård2006-11-144-32/+32
* zigzag_direct_noperm doesn't exist, remove declarationMåns Rullgård2006-11-121-1/+0
* rename inverse -> ff_inverseMåns Rullgård2006-11-121-1/+1
* adding more static keywordsMåns Rullgård2006-11-122-5/+5
* ensure alignment (no speed change)Michael Niedermayer2006-11-031-6/+6
* merging mov & and (no speedchange)Michael Niedermayer2006-11-031-3/+2
* 2 instructions less (same speed)Michael Niedermayer2006-11-031-6/+13
* comment about failed optimizationMichael Niedermayer2006-11-031-2/+2
* move luma tc0 related init into asmMichael Niedermayer2006-11-031-12/+17
* 2 instructions less in h264_loop_filter_luma_mmx2()Michael Niedermayer2006-11-031-5/+3
* preempt possible overflowMichael Niedermayer2006-11-031-1/+1
* 1 instruction lessMichael Niedermayer2006-11-031-7/+6
* optimize H264_DEBLOCK_P0_Q0Michael Niedermayer2006-11-032-40/+23
* Put libmpeg2 IDCT functions under CONFIG_GPL, fixes link failureDiego Biurrun2006-10-141-0/+2
* Rename ABS macro to FFABS.Diego Biurrun2006-10-111-6/+6
* Change license headers to say 'FFmpeg' instead of 'this program/this library'Diego Biurrun2006-10-0721-84/+126
* Compilation fix, printf gets redefined to please_use_av_log.Diego Biurrun2006-10-071-0/+2
* Switch idct_mmx_xvid.c from GPL to LGPL as permitted by theDiego Biurrun2006-10-052-16/+11
* approximate qpel functions: sacrifice some quality for some decoding speed. e...Loren Merritt2006-10-012-0/+105
* add some #ifdef CONFIG_ENCODERS/DECODERSMåns Rullgård2006-09-271-0/+2
* cosmetics (indentation)Loren Merritt2006-09-211-19/+19
* tweak ff_imdct_calc_3dn2Loren Merritt2006-09-211-17/+17
* sse implementation of imdct.Loren Merritt2006-09-211-0/+125
* New single instruction math operation headerLuca Barbato2006-09-191-0/+39
* disable vp3 mmx idct for theora files to avoid artifactsAurelien Jacobs2006-09-141-0/+1
* Remove the LGPL exception clause as discussed on ffmpeg-develDiego Biurrun2006-09-142-3/+2
* Enables back the mmx/sse optimized version of the vp3 idct.Aurelien Jacobs2006-09-121-3/+2
* Add official LGPL license headers to the files that were missing them.Diego Biurrun2006-09-103-2/+47
* remove redundant declarationsMåns Rullgård2006-09-021-1/+0
* mmx implementation of deblocking strength decision.Loren Merritt2006-08-282-0/+99
* ff_fft_calc_3dn/3dn2/sse: convert intrinsics to inline asm.Loren Merritt2006-08-183-280/+232
* simplifyMichael Niedermayer2006-08-181-21/+9
* insufficient alignmentMichael Niedermayer2006-08-171-1/+1
* Fix building with --disable-opts but MMX enabled.Marco Manfredini2006-08-131-3/+3
* Support for MacIntel, last part: balign directivesJohn Dalgliesh2006-08-127-36/+36
* vorbis simd tweaksLoren Merritt2006-08-111-51/+67
* convert vector_fmul_reverse_sse2 and vector_fmul_add_add_sse2 to sseMichael Niedermayer2006-08-101-22/+22
* sse and 3dnow implementations of float->int conversion and mdct windowing.Loren Merritt2006-08-101-3/+210
* Fix x86 SIMD asm and pic, patch from Martin von Gagern <Martin.vGagern@gmx.net>Luca Barbato2006-08-101-30/+32
* Support for MacIntel, take xx: '/nop' illegal for old versions of GASJohn Dalgliesh2006-08-101-76/+54
* Add support for Mac OS X Intel part 2: Assembler macros in fdct_mmx.cJohn Dalgliesh2006-08-101-53/+53
* Support for Mac OS X Intel, part 3: binary integer constants:John Dalgliesh2006-08-101-5/+5
* emms -> femmsLoren Merritt2006-08-092-2/+2
* gcc 2.95 and 3.4.x on x86 32bit without fomit-frame-pointer can't even find 5...Loren Merritt2006-08-091-10/+16
* slightly faster ff_imdct_calc_3dn2() on amd64. (gcc added a bunch of useless ...Loren Merritt2006-08-081-1/+1
* change vorbis_inverse_coupling_sse2() so it works on sse1 cpusMichael Niedermayer2006-08-081-9/+17
* 3dnow2 implementation of imdct.Loren Merritt2006-08-081-1/+82
* 10l, vorbis_inverse_coupling_sse() was really 3dnowLoren Merritt2006-08-031-3/+3
OpenPOWER on IntegriCloud