aboutsummaryrefslogtreecommitdiffstats
path: root/libavcodec/i386
Commit message (Expand)AuthorAgeFilesLines
* sort H.264 mmx dsp functions into their own fileLoren Merritt2005-06-024-886/+917
* adding a few constMichael Niedermayer2005-06-013-5/+5
* 100l for myself (breaking amd64)Michael Niedermayer2005-06-011-4/+4
* merge a few asm blocks so gcc cant unoptimize it (658->631 dezicycles on duron)Michael Niedermayer2005-06-011-7/+13
* get rid of 2 movq (680 -> 658 dezicycles on duron)Michael Niedermayer2005-06-011-17/+15
* avoid one transpose (730->680 dezicycles on duron)Michael Niedermayer2005-06-011-7/+4
* 10l (symbol mangling)Loren Merritt2005-06-011-7/+7
* add rounding bias before the horizontal idct (765->730 dezicyles on duron)Michael Niedermayer2005-06-011-2/+3
* MMX for H.264 iDCT (adapted from x264)Loren Merritt2005-05-312-0/+90
* MMX code for (put|avg)_h264_chroma_mc8Zoltán Hidvégi2005-05-252-0/+168
* H.264 deblocking optimizations (mmx for chroma_bS4 case, convert existing cas...Loren Merritt2005-05-181-171/+235
* porting the mmx&sse2 (sse2 untested) vp3 idcts to the lavc idct APIMichael Niedermayer2005-05-173-266/+43
* disabling vp3 mmx&mmx2 idcts, they must be ported over to the lavc idct API, ...Michael Niedermayer2005-05-171-7/+7
* fix PICMichael Niedermayer2005-05-081-1/+1
* MMX for H.264 deblocking filterLoren Merritt2005-04-251-0/+204
* read 32bit instead of 64bit to avoid overreading and missalignmentsMichael Niedermayer2005-04-171-4/+8
* optimization and gcc 4.0 bug workaround patch by (Martin Drab >drab kepler.fj...Martin Drab2005-02-071-16/+12
* x86 cpu capabilities detection rewrite / cleanupMichael Niedermayer2004-12-131-61/+40
* detect sse on athlon-xp patch by (matthieu castet <castet >.< matthieu >at< f...Matthieu Castet2004-12-051-3/+3
* H.261 encoder by (Maarten Daniels <maarten dot daniels at luc dot ac dot be>)Michael Niedermayer2004-10-281-1/+1
* gcc 3.4.3 preversions do not appreciate invalid instruction and operand combi...Aurelien Jacobs2004-10-171-4/+5
* adapting existing mmx/mmx2/sse/3dnow optimizations so they work on x86_64 pat...Aurelien Jacobs2004-10-119-586/+603
* 1/2 resolution decodingMichael Niedermayer2004-09-251-16/+17
* h264_qpel8_hv_lowpass_mmx2/3dnowMichael Niedermayer2004-09-081-6/+79
* h264_qpel4_hv_lowpass_mmx2/3dnowMichael Niedermayer2004-09-081-49/+96
* optimizationMichael Niedermayer2004-09-081-54/+20
* optimizationMichael Niedermayer2004-09-071-80/+49
* optimizationMichael Niedermayer2004-09-071-17/+31
* h264 luma motion compensation in mmx2/3dnowMichael Niedermayer2004-09-073-1/+641
* 10000l fix and use more mmx2/3dnow code for mpeg4 qpel which has been written...Michael Niedermayer2004-09-062-39/+329
* 10l and better MMX/SSE detection for VIA1000Michael Niedermayer2004-07-161-1/+3
* dc_precission=11 fixMichael Niedermayer2004-07-061-1/+1
* transmeta crusoe mmx detection by "Nobody/Anonymous - nobody" / sf feature re...Michael Niedermayer2004-06-241-0/+11
* sse8 and nsse in mmxMichael Niedermayer2004-06-081-0/+302
* some of the warning fixes by (Michael Roitzsch <mroi at users dot sourceforge...Michael Niedermayer2004-05-187-6/+11
* mmx dct optimizationMichael Niedermayer2004-04-301-138/+135
* move the 0x80 vector outside of the function, thus saving the compilerMike Melanson2004-04-271-2/+3
* separate out put_signed_pixels_clamped() into its own function andMike Melanson2004-04-271-0/+19
* yes, that's nice...Mike Melanson2004-04-261-1/+0
* reorganize and simplify the VP3 IDCT stuffMike Melanson2004-04-263-141/+8
* hush...Mike Melanson2004-04-251-1/+1
* hook up support for SSE2-optimized VP3 IDCTMike Melanson2004-04-251-3/+9
* add selected SSE2 ASM macrosMike Melanson2004-04-251-0/+24
* SSE2-optimized variant of VP3 IDCTMike Melanson2004-04-251-0/+890
* correct MMX-optimized variant of VP3 IDCT, with comments (thank youMike Melanson2004-04-251-407/+396
* fixing alignment problems -> SSE2 support enabled again in libavcodec (from f...Michael Niedermayer2004-04-091-4/+17
* fft_*() renamed into ff_fft_*() patch by (Gildas Bazin <gbazin at altern dot ...Gildas Bazin2004-03-131-1/+1
* use optimized VP3 functions where appropriateMike Melanson2004-03-081-7/+12
* seperated out the C-based VP3 DSP functions into a different file; alsoMike Melanson2004-03-081-0/+652
* attribute used patch by (mitya at school dot ioffe dot ru (Dmitry Baryshkov))Dmitry Baryshkov2004-02-293-11/+11