| Commit message (Expand) | Author | Age | Files | Lines |
* | Simplify vsad16_mmx2(). | Michael Niedermayer | 2008-05-17 | 1 | -3/+2 |
* | Simplify vsad16_mmx(). | Michael Niedermayer | 2008-05-17 | 1 | -3/+2 |
* | Simplify vsad_intra16_mmx2() | Michael Niedermayer | 2008-05-17 | 1 | -3/+2 |
* | Simplify vsad_intra16_mmx() | Michael Niedermayer | 2008-05-17 | 1 | -3/+2 |
* | Add missing required header directly. | Diego Biurrun | 2008-05-09 | 1 | -0/+1 |
* | Add missing path to #include. | Diego Biurrun | 2008-05-09 | 1 | -1/+1 |
* | Use full path for #includes from another directory. | Diego Biurrun | 2008-05-09 | 22 | -45/+49 |
* | Do not misuse long as the size of a register in x86. | Ramiro Polla | 2008-05-08 | 20 | -175/+185 |
* | Rename i386/cputest.c --> i386/cpuid.c. | Diego Biurrun | 2008-04-26 | 1 | -0/+0 |
* | cosmetics: __asm__ __volatile__ --> asm volatile | Diego Biurrun | 2008-04-17 | 2 | -11/+11 |
* | cosmetics: Fix nonstandard indentation. | Diego Biurrun | 2008-04-16 | 1 | -21/+21 |
* | Cosmetics: | Jeff Downs | 2008-04-16 | 1 | -9/+21 |
* | Redo r12838, this time using svn copy to create h264_i386.h from cabac.h. | Jeff Downs | 2008-04-16 | 1 | -0/+144 |
* | Revert 12838 to redo it the right way (use svn copy to create new | Jeff Downs | 2008-04-16 | 1 | -144/+0 |
* | Add a new xvid-style IDCT using SSE2. | Alexander Strange | 2008-04-16 | 3 | -1/+402 |
* | Move decode_significance_x86() and decode_significance_8x8_x86() to | Jeff Downs | 2008-04-15 | 1 | -0/+144 |
* | Eliminate movdqu in vp3dsp_sse2, patch from Alexander Strange astrangeAtithin... | Luca Barbato | 2008-04-14 | 1 | -146/+146 |
* | Add a header file to declare Xvid IDCT functions. | Alexander Strange | 2008-04-12 | 2 | -3/+33 |
* | Fix H.264 interframe decoding when compiling with icc. Patch by Loren | Loren Merritt | 2008-04-08 | 1 | -2/+3 |
* | h264 chroma mc ssse3 | Loren Merritt | 2008-04-01 | 5 | -2/+251 |
* | cosmetics: typo fixes | Diego Biurrun | 2008-03-22 | 1 | -1/+1 |
* | Add missed call to ff_cavsdsp_init_3dnow() in dsputil_init_mmx() | Zuxy Meng | 2008-03-21 | 1 | -0/+4 |
* | Hardcode register to prevent aparent miscompilation. | Michael Niedermayer | 2008-03-20 | 1 | -2/+2 |
* | remove unused temp | Michael Niedermayer | 2008-03-20 | 1 | -5/+3 |
* | get register names from x86_cpu.h | Måns Rullgård | 2008-03-17 | 1 | -8/+1 |
* | move draw_edges() into dsputil | Aurelien Jacobs | 2008-03-04 | 2 | -90/+90 |
* | split encoding part of dsputil_mmx into its own file | Aurelien Jacobs | 2008-02-25 | 3 | -1409/+1434 |
* | __asm __volatile -> asm volatile part 2 | Reimar Döffinger | 2008-02-24 | 2 | -32/+32 |
* | __asm __volatile -> asm volatile, improves code consistency and works | Reimar Döffinger | 2008-02-24 | 2 | -19/+19 |
* | simd and unroll png_filter_row | Loren Merritt | 2008-02-21 | 1 | -0/+98 |
* | Disabling all SSE* code for old gcc to avoid alignment issues. | Michael Niedermayer | 2008-02-21 | 1 | -2/+6 |
* | Fix warnings: | Reimar Döffinger | 2008-02-19 | 1 | -2/+2 |
* | cosmetics: Replace // by /* */ comments. | Diego Biurrun | 2008-02-13 | 1 | -202/+202 |
* | asm argument that might be in memory needs a size | Loren Merritt | 2008-02-10 | 1 | -1/+1 |
* | use fewer registers in apply_welch_window_sse2 | Loren Merritt | 2008-02-09 | 1 | -8/+7 |
* | sse2 h264 motion compensation. not new code, just separate out the cases that... | Loren Merritt | 2008-02-06 | 2 | -120/+245 |
* | put loop counter in a register if possible. makes some of the qpel functions ... | Loren Merritt | 2008-02-06 | 2 | -6/+6 |
* | fix aliasing warnings. simpler too. | Loren Merritt | 2008-02-06 | 1 | -41/+34 |
* | constant was excessively aligned | Loren Merritt | 2008-02-06 | 1 | -1/+1 |
* | ssse3 h264 motion compensation. | Loren Merritt | 2008-02-05 | 3 | -60/+512 |
* | add qpel rounder once during hv rather than twice during hv and whatever it's... | Loren Merritt | 2008-02-05 | 1 | -23/+8 |
* | clean up an ugliness introduced in r11826. this syntax will require fewer cha... | Loren Merritt | 2008-02-05 | 4 | -9/+11 |
* | Deprecate old and inefficient per instruction asm(). | Michael Niedermayer | 2008-02-04 | 1 | -0/+2 |
* | reduce code duplication | Loren Merritt | 2008-02-04 | 1 | -175/+70 |
* | avg_pixels4_mmx2 | Loren Merritt | 2008-02-03 | 2 | -2/+25 |
* | use mmx2/3dnow avg functions in avg_qpel*_mc00 | Loren Merritt | 2008-02-03 | 2 | -3/+16 |
* | ff_h264_idct8_add_sse2. | Loren Merritt | 2008-02-03 | 4 | -8/+97 |
* | remove some movq in ff_h264_idct8_add_mmx. 225->217 cycles on core2. | Loren Merritt | 2008-02-03 | 1 | -29/+23 |
* | add parenthesis, fix warning: i386/dsputil_mmx.c:2618: warning: suggest paren... | Baptiste Coudurier | 2008-01-30 | 1 | -2/+2 |
* | fix prototypes, remove warning: i386/dsputil_mmx.c:3594: warning: assignment ... | Baptiste Coudurier | 2008-01-30 | 1 | -4/+4 |