| 2006-03-25 |
Loren Merritt | simplified and slightly faster h264_chroma_mc8_mmx |
tree | commitdiff |
| 2006-03-23 |
Loren Merritt | prefetch pixels for future motion compensation. 2-5... |
tree | commitdiff |
| 2006-03-22 |
Loren Merritt | 10l |
tree | commitdiff |
| 2006-03-22 |
Loren Merritt | added mmx implementation of h264_chroma_mc2 |
tree | commitdiff |
| 2006-03-22 |
Robert Edele | add MMX and SSE versions of ff_snow_inner_add_yblock |
tree | commitdiff |
| 2006-03-21 |
Robert Edele | snow mmx+sse2 optimizations, part 4 |
tree | commitdiff |
| 2006-03-20 |
Robert Edele | Add the mmx and sse2 implementations of ff_snow_vertica... |
tree | commitdiff |
| 2006-03-16 |
Robert Edele | First part of a series of speed-enchancing patches. |
tree | commitdiff |
| 2006-03-08 |
Zuxy Meng | 3DNow! & Extended 3DNow! versions of FFT |
tree | commitdiff |
| 2006-03-07 |
Loren Merritt | h264_idct8_add_mmx |
tree | commitdiff |
| 2006-03-07 |
Loren Merritt | h264_idct_add only needs mmx1 |
tree | commitdiff |
| 2006-03-05 |
Zuxy Meng | use xorps instead of mulps to toggle the sign of a... |
tree | commitdiff |
| 2006-03-05 |
Loren Merritt | gcc2.95 workaround |
tree | commitdiff |
| 2006-03-04 |
Loren Merritt | remove some useless instructions |
tree | commitdiff |
| 2006-03-02 |
Loren Merritt | 4% faster h264_qpel_mc |
tree | commitdiff |
| 2006-02-10 |
Loren Merritt | h264: special case dc-only idct. ~1% faster overall |
tree | commitdiff |
| 2006-02-09 |
Loren Merritt | 10l in 1.12 |
tree | commitdiff |
| 2006-02-07 |
Loren Merritt | 10l (x86_64) |
tree | commitdiff |
| 2006-02-07 |
Loren Merritt | 18% faster put_h264_qpel16_mc[13]2_mmx2 |
tree | commitdiff |
| 2006-02-07 |
Loren Merritt | 11% faster put_h264_qpel16_v_lowpass_mmx2 |
tree | commitdiff |
| 2006-02-07 |
Loren Merritt | 15% faster put_h264_qpel16_hv_lowpass_mmx2 |
tree | commitdiff |
| 2006-02-05 |
Steve L'Homme | MSVC-compatible __align8/__align16 declaration |
tree | commitdiff |
| 2006-01-12 |
Diego Biurrun | Update licensing information: The FSF changed postal... |
tree | commitdiff |
| 2006-01-09 |
Loren Merritt | tweak h264_biweight |
tree | commitdiff |
| 2005-12-30 |
Loren Merritt | fix some potential arithmetic overflows in pred_direct_... |
tree | commitdiff |
| 2005-12-22 |
Diego Biurrun | COSMETICS: tabs --> spaces, some prettyprinting |
tree | commitdiff |
| 2005-12-17 |
Diego Biurrun | COSMETICS: Remove all trailing whitespace. |
tree | commitdiff |
| 2005-12-10 |
Guillaume Poirier | Add the rest of missing Reg_* macros to support both... |
tree | commitdiff |
| 2005-11-12 |
Loren Merritt | use sse16_sse2() in nsse |
tree | commitdiff |
| 2005-10-27 |
Loren Merritt | faster h264_chroma_mc8_mmx, added h264_chroma_mc4_mmx. |
tree | commitdiff |
| 2005-10-09 |
Loren Merritt | h264 mmx weighted prediction. up to 3% overall speedup. |
tree | commitdiff |
| 2005-09-30 |
Loren Merritt | sse2 16x16 sum squared diff (306=>268 cycles on a K8) |
tree | commitdiff |
| 2005-09-21 |
Michael Niedermayer | replace a few mov + psrlq with pshufw, there are more... |
tree | commitdiff |
| 2005-09-10 |
Reimar Döffinger | Fix compile without CONFIG_GPL, misplaced #endif caused... |
tree | commitdiff |
| 2005-09-10 |
Michael Niedermayer | remove unused table entries |
tree | commitdiff |
| 2005-09-10 |
Michael Niedermayer | xvids mmx&mmx2 idcts |
tree | commitdiff |
| 2005-09-06 |
Måns Rullgård | Kill some compiler warnings. Compiled code verified... |
tree | commitdiff |
| 2005-09-03 |
Michael Niedermayer | simplify (d&a) and (d&~a) calculation, hint by skal |
tree | commitdiff |
| 2005-09-02 |
Michael Niedermayer | add consts (this was in my local tree, dunno where... |
tree | commitdiff |
| 2005-08-14 |
Måns Rullgård | kill a bunch of compiler warnings |
tree | commitdiff |
| 2005-08-01 |
Alexander Strasser | libavutil: Utility code from libavcodec moved to a... |
tree | commitdiff |
| 2005-06-02 |
Loren Merritt | sort H.264 mmx dsp functions into their own file |
tree | commitdiff |
| 2005-06-01 |
Michael Niedermayer | adding a few const |
tree | commitdiff |
| 2005-06-01 |
Michael Niedermayer | 100l for myself (breaking amd64) |
tree | commitdiff |
| 2005-06-01 |
Michael Niedermayer | merge a few asm blocks so gcc cant unoptimize it (658... |
tree | commitdiff |
| 2005-06-01 |
Michael Niedermayer | get rid of 2 movq (680 -> 658 dezicycles on duron) |
tree | commitdiff |
| 2005-06-01 |
Michael Niedermayer | avoid one transpose (730->680 dezicycles on duron) |
tree | commitdiff |
| 2005-06-01 |
Loren Merritt | 10l (symbol mangling) |
tree | commitdiff |
| 2005-06-01 |
Michael Niedermayer | add rounding bias before the horizontal idct (765-... |
tree | commitdiff |
| 2005-05-31 |
Loren Merritt | MMX for H.264 iDCT (adapted from x264) |
tree | commitdiff |
| 2005-05-25 |
Zoltán Hidvégi | MMX code for (put|avg)_h264_chroma_mc8 |
tree | commitdiff |
| 2005-05-18 |
Loren Merritt | H.264 deblocking optimizations (mmx for chroma_bS4... |
tree | commitdiff |
| 2005-05-17 |
Michael Niedermayer | porting the mmx&sse2 (sse2 untested) vp3 idcts to the... |
tree | commitdiff |
| 2005-05-17 |
Michael Niedermayer | disabling vp3 mmx&mmx2 idcts, they must be ported over... |
tree | commitdiff |
| 2005-05-08 |
Michael Niedermayer | fix PIC |
tree | commitdiff |
| 2005-04-25 |
Loren Merritt | MMX for H.264 deblocking filter |
tree | commitdiff |
| 2005-04-17 |
Michael Niedermayer | read 32bit instead of 64bit to avoid overreading and... |
tree | commitdiff |
| 2005-02-07 |
Martin Drab | optimization and gcc 4.0 bug workaround patch by (Marti... |
tree | commitdiff |
| 2004-12-13 |
Michael Niedermayer | x86 cpu capabilities detection rewrite / cleanup |
tree | commitdiff |
| 2004-12-05 |
Matthieu Castet | detect sse on athlon-xp patch by (matthieu castet ... |
tree | commitdiff |
| 2004-10-28 |
Michael Niedermayer | H.261 encoder by (Maarten Daniels <maarten dot daniels... |
tree | commitdiff |
| 2004-10-17 |
Aurelien Jacobs | gcc 3.4.3 preversions do not appreciate invalid instruc... |
tree | commitdiff |
| 2004-10-11 |
Aurelien Jacobs | adapting existing mmx/mmx2/sse/3dnow optimizations... |
tree | commitdiff |
| 2004-09-25 |
Michael Niedermayer | 1/2 resolution decoding |
tree | commitdiff |
| 2004-09-08 |
Michael Niedermayer | h264_qpel8_hv_lowpass_mmx2/3dnow |
tree | commitdiff |
| 2004-09-08 |
Michael Niedermayer | h264_qpel4_hv_lowpass_mmx2/3dnow |
tree | commitdiff |
| 2004-09-08 |
Michael Niedermayer | optimization |
tree | commitdiff |
| 2004-09-07 |
Michael Niedermayer | optimization |
tree | commitdiff |
| 2004-09-07 |
Michael Niedermayer | optimization |
tree | commitdiff |
| 2004-09-07 |
Michael Niedermayer | h264 luma motion compensation in mmx2/3dnow |
tree | commitdiff |
| 2004-09-06 |
Michael Niedermayer | 10000l fix and use more mmx2/3dnow code for mpeg4 qpel... |
tree | commitdiff |
| 2004-07-16 |
Michael Niedermayer | 10l and better MMX/SSE detection for VIA1000 |
tree | commitdiff |
| 2004-07-06 |
Michael Niedermayer | dc_precission=11 fix |
tree | commitdiff |
| 2004-06-24 |
Michael Niedermayer | transmeta crusoe mmx detection by "Nobody/Anonymous... |
tree | commitdiff |
| 2004-06-08 |
Michael Niedermayer | sse8 and nsse in mmx |
tree | commitdiff |
| 2004-05-18 |
Michael Niedermayer | some of the warning fixes by (Michael Roitzsch <mroi... |
tree | commitdiff |
| 2004-04-30 |
Michael Niedermayer | mmx dct optimization |
tree | commitdiff |
| 2004-04-27 |
Mike Melanson | move the 0x80 vector outside of the function, thus... |
tree | commitdiff |
| 2004-04-27 |
Mike Melanson | separate out put_signed_pixels_clamped() into its own... |
tree | commitdiff |
| 2004-04-26 |
Mike Melanson | yes, that's nice... |
tree | commitdiff |
| 2004-04-26 |
Mike Melanson | reorganize and simplify the VP3 IDCT stuff |
tree | commitdiff |
| 2004-04-25 |
Mike Melanson | hush... |
tree | commitdiff |
| 2004-04-25 |
Mike Melanson | hook up support for SSE2-optimized VP3 IDCT |
tree | commitdiff |
| 2004-04-25 |
Mike Melanson | add selected SSE2 ASM macros |
tree | commitdiff |
| 2004-04-25 |
Mike Melanson | SSE2-optimized variant of VP3 IDCT |
tree | commitdiff |
| 2004-04-25 |
Mike Melanson | correct MMX-optimized variant of VP3 IDCT, with comment... |
tree | commitdiff |
| 2004-04-09 |
Michael Niedermayer | fixing alignment problems -> SSE2 support enabled again... |
tree | commitdiff |
| 2004-03-13 |
Gildas Bazin | fft_*() renamed into ff_fft_*() patch by (Gildas Bazin... |
tree | commitdiff |
| 2004-03-08 |
Mike Melanson | use optimized VP3 functions where appropriate |
tree | commitdiff |
| 2004-03-08 |
Mike Melanson | seperated out the C-based VP3 DSP functions into a... |
tree | commitdiff |
| 2004-02-29 |
Dmitry Baryshkov | attribute used patch by (mitya at school dot ioffe... |
tree | commitdiff |
| 2004-02-06 |
Michael Niedermayer | quantizer noise shaping optimization |
tree | commitdiff |
| 2004-01-31 |
Mike Melanson | vsad16_mmx2 only applies if encoders are turned on |
tree | commitdiff |
| 2004-01-29 |
Michael Niedermayer | SSE2 fdct by (Balatoni Denes <pnis at coder dot hu>) |
tree | commitdiff |
| 2004-01-10 |
Michael Niedermayer | copyright year update of the files i touched and rememb... |
tree | commitdiff |
| 2004-01-05 |
Michael Niedermayer | interlaced dct decision cleanup |
tree | commitdiff |
| 2004-01-03 |
Denes Balatoni | denoise_dct_sse2() patch by (Balatoni Denes <pnis at... |
tree | commitdiff |
| 2004-01-02 |
Michael Niedermayer | denoise_dct_mmx() |
tree | commitdiff |
| 2003-12-30 |
Michael Niedermayer | interlaced motion estimation |
tree | commitdiff |
| 2003-12-14 |
Michael Niedermayer | split intra / inter dequantization |
tree | commitdiff |
| next |
projects / ffmpeg / history