diff options
| author | Michael Niedermayer <michaelni@gmx.at> | 2011-07-02 03:07:06 +0200 |
|---|---|---|
| committer | Michael Niedermayer <michaelni@gmx.at> | 2011-07-02 03:24:32 +0200 |
| commit | 3074f03a074de3aab79639d261cbd0ccc265b5b4 (patch) | |
| tree | 9710041e852ee69f6de6ef6e6333af82f6ca5931 /libavcodec/x86/dsputil_mmx.c | |
| parent | 392acaedcb052fa64386d5d0aea4931386f72d64 (diff) | |
| parent | 23ce6e72123a40895baaeefeb27c7c18748bd67e (diff) | |
| download | ffmpeg-3074f03a074de3aab79639d261cbd0ccc265b5b4.tar.gz | |
Merge remote-tracking branch 'qatar/master'
* qatar/master:
get_bits: remove x86 inline asm in A32 bitstream reader
doc: Remove outdated information about our issue tracker
avidec: Factor out the sync fucntionality.
fate-aac: Expand coverage.
ac3dsp: add x86-optimized versions of ac3dsp.extract_exponents().
ac3dsp: simplify extract_exponents() now that it does not need to do clipping.
ac3enc: clip coefficients after MDCT.
ac3enc: add int32_t array clipping function to DSPUtil, including x86 versions.
swscale: for >8bit scaling, read in native bit-depth.
matroskadec: matroska_read_seek after after EBML_STOP leads to failure.
doxygen: fix usage of @file directive in libavutil/{dict,file}.h
doxygen: Help doxygen parser to understand the DECLARE_ALIGNED and offsetof macros
Conflicts:
doc/issue_tracker.txt
libavformat/avidec.c
libavutil/dict.h
libswscale/swscale.c
libswscale/utils.c
tests/ref/lavfi/pixfmts_scale
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Diffstat (limited to 'libavcodec/x86/dsputil_mmx.c')
| -rw-r--r-- | libavcodec/x86/dsputil_mmx.c | 23 |
1 files changed, 23 insertions, 0 deletions
diff --git a/libavcodec/x86/dsputil_mmx.c b/libavcodec/x86/dsputil_mmx.c index 78cad4c13a..80bb6cd26c 100644 --- a/libavcodec/x86/dsputil_mmx.c +++ b/libavcodec/x86/dsputil_mmx.c @@ -2333,6 +2333,15 @@ int ff_add_hfyu_left_prediction_sse4(uint8_t *dst, const uint8_t *src, int w, i float ff_scalarproduct_float_sse(const float *v1, const float *v2, int order); +void ff_vector_clip_int32_mmx (int32_t *dst, const int32_t *src, int32_t min, + int32_t max, unsigned int len); +void ff_vector_clip_int32_sse2 (int32_t *dst, const int32_t *src, int32_t min, + int32_t max, unsigned int len); +void ff_vector_clip_int32_sse2_int(int32_t *dst, const int32_t *src, int32_t min, + int32_t max, unsigned int len); +void ff_vector_clip_int32_sse41 (int32_t *dst, const int32_t *src, int32_t min, + int32_t max, unsigned int len); + void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx) { int mm_flags = av_get_cpu_flags(); @@ -2473,6 +2482,8 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx) c->put_rv40_chroma_pixels_tab[0]= ff_put_rv40_chroma_mc8_mmx; c->put_rv40_chroma_pixels_tab[1]= ff_put_rv40_chroma_mc4_mmx; + + c->vector_clip_int32 = ff_vector_clip_int32_mmx; #endif if (mm_flags & AV_CPU_FLAG_MMX2) { @@ -2756,6 +2767,11 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx) #if HAVE_YASM c->scalarproduct_int16 = ff_scalarproduct_int16_sse2; c->scalarproduct_and_madd_int16 = ff_scalarproduct_and_madd_int16_sse2; + if (mm_flags & AV_CPU_FLAG_ATOM) { + c->vector_clip_int32 = ff_vector_clip_int32_sse2_int; + } else { + c->vector_clip_int32 = ff_vector_clip_int32_sse2; + } if (avctx->flags & CODEC_FLAG_BITEXACT) { c->apply_window_int16 = ff_apply_window_int16_sse2_ba; } else { @@ -2781,6 +2797,13 @@ void dsputil_init_mmx(DSPContext* c, AVCodecContext *avctx) } #endif } + + if (mm_flags & AV_CPU_FLAG_SSE4 && HAVE_SSE) { +#if HAVE_YASM + c->vector_clip_int32 = ff_vector_clip_int32_sse41; +#endif + } + #if HAVE_AVX && HAVE_YASM if (mm_flags & AV_CPU_FLAG_AVX) { if (bit_depth == 10) { |