| Commit message (Expand) | Author | Age | Files | Lines |
* | x86: move function prototypes to header files | Måns Rullgård | 2010-03-06 | 6 | -35/+27 |
* | Add some missing #includes | Måns Rullgård | 2010-03-06 | 4 | -0/+4 |
* | Move FFT parts from dsputil.h to fft.h | Måns Rullgård | 2010-03-06 | 1 | -1/+1 |
* | Remove DECLARE_ALIGNED_{8,16} macros | Måns Rullgård | 2010-03-06 | 11 | -54/+54 |
* | Move NEG_[US]SR32 macros to mathops.h | Måns Rullgård | 2010-02-17 | 1 | -0/+19 |
* | Enable SSE2 (put|avg)_pixels_16_sse2 | David Conrad | 2010-02-10 | 1 | -2/+0 |
* | Make the jump-table section-relative for x86_64 with PIC enabled. | Reimar Döffinger | 2010-01-30 | 1 | -7/+12 |
* | optimize h264_loop_filter_strength_mmx2 | Loren Merritt | 2010-01-26 | 1 | -74/+70 |
* | Implement an sse version of scalarproduct_float(). | Alex Converse | 2010-01-22 | 2 | -0/+29 |
* | Move array specifiers outside DECLARE_ALIGNED() invocations | Måns Rullgård | 2010-01-22 | 15 | -51/+51 |
* | Use two separate memory arguments since 8+() is invalid gas syntax | David Conrad | 2010-01-21 | 1 | -19/+19 |
* | Attempt to fix asm compilation failure. | Michael Niedermayer | 2010-01-20 | 1 | -21/+20 |
* | Move COPY3_IF_LT to lavc/mathops.h | Måns Rullgård | 2010-01-20 | 1 | -0/+12 |
* | Use constant offsets for memory operands since gcc is unable to | David Conrad | 2010-01-20 | 1 | -12/+8 |
* | Fix h264_loop_filter_strength_mmx2() so it works with b frames. | Michael Niedermayer | 2010-01-19 | 1 | -0/+34 |
* | Remove -2 -> -1 remapping, its not needed anymore as we must remap all | Michael Niedermayer | 2010-01-19 | 1 | -4/+0 |
* | Fix XvMC. XvMCCreateBlocks() may not allocate 16-byte aligned blocks, | Gwenole Beauchesne | 2010-01-04 | 1 | -1/+3 |
* | Reduce number of ASM constraints for ff_lpc_compute_autocorr_sse2 since it | Reimar Döffinger | 2010-01-02 | 1 | -9/+10 |
* | Get rid of pointless CONFIG_ANY_H263 preprocessor definition. | Diego Biurrun | 2009-12-30 | 1 | -4/+3 |
* | fix a crash in ape decoding on x86_32 sse2 | Loren Merritt | 2009-12-08 | 1 | -1/+1 |
* | slightly faster scalarproduct_and_madd_int16_ssse3 on penryn, no change on co... | Loren Merritt | 2009-12-05 | 1 | -5/+13 |
* | r20739 broke compilation on systems without yasm | Loren Merritt | 2009-12-05 | 1 | -7/+6 |
* | refactor and optimize scalarproduct | Loren Merritt | 2009-12-05 | 2 | -47/+136 |
* | port ape dsp functions from sse2 to mmx | Loren Merritt | 2009-12-03 | 2 | -75/+93 |
* | s/movdqa/movaps/ in sse1 fft. (regression in r20293) | Loren Merritt | 2009-10-25 | 1 | -0/+2 |
* | fix linking on systems with a function name prefix (10l in r20287) | Loren Merritt | 2009-10-18 | 1 | -1/+1 |
* | sync yasm macros to x264 | Loren Merritt | 2009-10-18 | 1 | -5/+3 |
* | huffyuv: add some const qualifiers | Loren Merritt | 2009-10-18 | 3 | -7/+7 |
* | simd add_hfyu_left_prediction | Loren Merritt | 2009-10-18 | 3 | -0/+83 |
* | add CONFIG_LPC to the build system for lpc dsputil functions. fixes build | Justin Ruggles | 2009-10-18 | 1 | -0/+2 |
* | Move autocorrelation function from flacenc.c to lpc.c. Also rename the | Justin Ruggles | 2009-10-17 | 2 | -6/+4 |
* | Use MANGLE in cavsdsp, the current version using "m" constraints will not | Reimar Döffinger | 2009-10-01 | 1 | -23/+23 |
* | Replace several #ifdef PIC with the more obvious and correct | Reimar Döffinger | 2009-09-30 | 2 | -10/+10 |
* | cosmetics: fix indentation after previous commit | Måns Rullgård | 2009-09-27 | 1 | -32/+32 |
* | Drop unused args from vector_fmul_add_add, simpify code, and rename | Måns Rullgård | 2009-09-27 | 1 | -70/+6 |
* | Merge FFTContext and MDCTContext | Måns Rullgård | 2009-09-20 | 3 | -18/+18 |
* | Move per-arch fft init bits into the corresponding subdirs | Måns Rullgård | 2009-09-15 | 4 | -0/+82 |
* | Move declarations of some mmx functions to dsputil_mmx.h | Måns Rullgård | 2009-08-29 | 4 | -0/+8 |
* | Mark "i" parameter of vector_clipf_sse() as early-clobber | Vitor Sessak | 2009-08-27 | 1 | -1/+1 |
* | Mark parameter src of vector_clipf() as const | Vitor Sessak | 2009-08-27 | 1 | -1/+1 |
* | SSE optimized vector_clipf(). 10% faster TwinVQ decoding. | Vitor Sessak | 2009-08-27 | 1 | -0/+35 |
* | Update x264 asm code to latest to add support for 64-bit Windows. | John Adcock | 2009-08-04 | 5 | -142/+507 |
* | Do not check for both CONFIG_VC1_DECODER and CONFIG_WMV3_DECODER, | Diego Biurrun | 2009-07-29 | 1 | -1/+1 |
* | Do not redundantly check for both CONFIG_THEORA_DECODER and CONFIG_VP3_DECODER. | Diego Biurrun | 2009-07-22 | 1 | -2/+2 |
* | Icc 11.1 still does not align the stack pointer, disable some x264 functions. | Carl Eugen Hoyos | 2009-07-17 | 1 | -1/+1 |
* | SSE version of clear_blocks | Jason Garrett-Glaser | 2009-06-16 | 1 | -2/+26 |
* | idct_dc for VC-1/WMV3 decoder; ~11% faster decoding overall. | Jason Garrett-Glaser | 2009-06-16 | 1 | -0/+203 |
* | Replace more uses of __attribute__((aligned)) by DECLARE_ALIGNED. | Ramiro Polla | 2009-06-04 | 2 | -2/+2 |
* | Use fewer macros in x86-optimized mlpdsp. | Ramiro Polla | 2009-06-03 | 1 | -19/+10 |
* | H264: Fix out of bounds reads in SSSE3 MC | Alexander Strange | 2009-05-30 | 1 | -38/+38 |