summaryrefslogtreecommitdiffstats
path: root/libavcodec/x86
Commit message (Expand)AuthorAgeFilesLines
* pixblockdsp: x86: Condense diff_pixels_* to a shared macroTimothy Gu2015-11-071-34/+30
* all: fix -Wextra-semi reported on clangGanesh Ajjanagadde2015-10-245-350/+345
* videodsp: don't overread edges in vfix3 emu_edge.Ronald S. Bultje2015-10-241-4/+11
* avcodec/x86/vc1dsp: Remove unused macroMichael Niedermayer2015-10-221-7/+0
* lavc/x86/vc1dsp_init: Fix compilation with --disable-yasm.Carl Eugen Hoyos2015-10-221-2/+2
* x86/Makefile: move decoder/encoder objects out of the subsystems sectionJames Almer2015-10-221-9/+8
* vc1dsp: Port ff_vc1_put_ver_16b_shift2_mmx to yasmTimothy Gu2015-10-212-53/+95
* huffyuvencdsp: Cherry pick changes left out in the last commitTimothy Gu2015-10-211-3/+3
* huffyuvencdsp: Add ff_diff_bytes_{sse2,avx2}Timothy Gu2015-10-212-19/+105
* huffyuvencdsp: Convert ff_diff_bytes_mmx to yasmTimothy Gu2015-10-203-30/+81
* huffyuvencdsp: Use intptr_t for widthTimothy Gu2015-10-191-2/+3
* x86: vc1dsp_mmx: Move yasm initiation steps to vc1dsp_initTimothy Gu2015-10-192-38/+25
* x86: fpel: Remove erroneous ff_put_pixels8_mmxext prototypeTimothy Gu2015-10-191-2/+0
* x86: fpel: Move prototypes for 4-px block functionsTimothy Gu2015-10-192-4/+6
* x86/vp9itxfm: fix register clobbering in ff_vp9_idct_idct_4x4_add_12_sse2James Almer2015-10-131-1/+1
* x86: simple_idct10_template: use constChristophe Gisquet2015-10-135-57/+39
* vp9: use registers for constant loading where possible.Ronald S. Bultje2015-10-131-122/+146
* vp9: refactor itx coefficients and share between 8 and 10/12bpp.Ronald S. Bultje2015-10-132-88/+91
* vp9: add itxfm_add eob shortcuts to 10/12bpp functions.Ronald S. Bultje2015-10-131-33/+200
* vp9: add 10/12bpp idct_idct_32x32 sse2 SIMD version.Ronald S. Bultje2015-10-132-44/+488
* vp9: 10/12bpp sse2 SIMD for iadst16.Ronald S. Bultje2015-10-132-2/+296
* vp9: refactor 10/12bpp dc-only code in 4x4/8x8 and add to 16x16.Ronald S. Bultje2015-10-131-56/+79
* vp9: add 10/12bpp sse2 SIMD version for idct_idct_16x16.Ronald S. Bultje2015-10-132-16/+223
* vp9: add 10/12bpp sse2 SIMD versions of iadst8x8.Ronald S. Bultje2015-10-132-13/+202
* vp9: add 10/12bpp sse2 SIMD for idct_idct_8x8.Ronald S. Bultje2015-10-132-69/+249
* vp9: add 12bpp sse2 versions of iadst4.Ronald S. Bultje2015-10-132-4/+151
* vp9: initial attempt at a idct_idct_4x4 12bpp x86 simd (sse2) impl.Ronald S. Bultje2015-10-132-0/+126
* vp9: add x86 simd (sse2/ssse3) for iadst4 10bpp functions.Ronald S. Bultje2015-10-135-82/+157
* vp9: add 10bpp simd (mmxext/ssse3) for idct_idct_4x4.Ronald S. Bultje2015-10-136-49/+159
* vp9: add 10/12bpp mmxext-optimized iwht_iwht_4x4 function.Ronald S. Bultje2015-10-137-22/+187
* x86: dct-test: add more idctsChristophe Gisquet2015-10-131-0/+12
* x86: simple_idct: 12bits versionsChristophe Gisquet2015-10-133-2/+42
* x86: simple_idct(_put): 10bits versionsChristophe Gisquet2015-10-136-33/+136
* x86: simple_idct10_template: fix overflow in passChristophe Gisquet2015-10-132-5/+16
* x86: prores: templatize 10 bits simple_idctChristophe Gisquet2015-10-134-260/+346
* x86/takdsp: use arithmetic shift instructionsJames Almer2015-10-091-4/+4
* avcodec/takdec: add x86 SIMD for rest of decorrelation modesPaul B Mahol2015-10-093-0/+163
* vp9: don't keep a stack pointer if we don't need it.Ronald S. Bultje2015-10-071-15/+5
* x86/alacdsp: add simd optimized functionsJames Almer2015-10-063-0/+179
* vp9: fix msvc build by using 6 GPRs on 32bit if stack!=aligned.Ronald S. Bultje2015-10-051-0/+10
* blockdsp: reindent after parameter removalChristophe Gisquet2015-10-031-8/+8
* vp9: 10/12bpp SIMD (sse2/ssse3/avx) for directional intra prediction.Ronald S. Bultje2015-10-036-2/+1562
* vp9: 16bpp tm/dc/h/v intra pred simd (mostly sse2) functions.Ronald S. Bultje2015-10-039-5/+669
* vp9: sse2/ssse3/avx 16bpp loopfilter x86 simd.Ronald S. Bultje2015-10-036-1/+918
* avcodec/x86/hpeldsp_rnd_template: silence -Wunused-function on --disable-mmxGanesh Ajjanagadde2015-10-031-2/+2
* blockdsp: remove high bitdepth parameterChristophe Gisquet2015-10-021-3/+1
* x86/hevc_sao: move 10/12bit functions into a separate fileJames Almer2015-09-303-340/+490
* avcodec/x86/rnd_template: silence -Wunused-function on --disable-mmxGanesh Ajjanagadde2015-09-291-2/+2
* avcodec/x86/sbrdsp: Fix using uninitialized upper 32bit of noiseMichael Niedermayer2015-09-291-0/+1
* avcodec/x86/cavsdsp: silence -Wunused-variable on --disable-mmxGanesh Ajjanagadde2015-09-241-1/+1
OpenPOWER on IntegriCloud