diff options
author | Michael Niedermayer <michaelni@gmx.at> | 2012-06-24 02:09:53 +0200 |
---|---|---|
committer | Michael Niedermayer <michaelni@gmx.at> | 2012-06-24 02:09:53 +0200 |
commit | 1c600888857544986d6576bc164e0dc8f0f4b6c7 (patch) | |
tree | f4e77c53d1050223a9170f75e4ea719f29cd04eb /libavcodec/x86/dsputil_mmx.h | |
parent | da728d5d2e7993911c7ed92c212ab900b7be180c (diff) | |
parent | fe07c9c6b5a870b8f2ffcfac649228b4d76e9505 (diff) | |
download | ffmpeg-1c600888857544986d6576bc164e0dc8f0f4b6c7.tar.gz |
Merge remote-tracking branch 'qatar/master'
* qatar/master:
x86: Only use optimizations with cmov if the CPU supports the instruction
x86: Add CPU flag for the i686 cmov instruction
x86: remove unused inline asm macros from dsputil_mmx.h
x86: move some inline asm macros to the only places they are used
lavfi: Add the af_channelmap audio channel mapping filter.
lavfi: add join audio filter.
lavfi: allow audio filters to request a given number of samples.
lavfi: support automatically inserting the fifo filter when needed.
lavfi/audio: eliminate ff_default_filter_samples().
Conflicts:
Changelog
libavcodec/x86/h264dsp_mmx.c
libavfilter/Makefile
libavfilter/allfilters.c
libavfilter/avfilter.h
libavfilter/avfiltergraph.c
libavfilter/version.h
libavutil/x86/cpu.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Diffstat (limited to 'libavcodec/x86/dsputil_mmx.h')
-rw-r--r-- | libavcodec/x86/dsputil_mmx.h | 102 |
1 files changed, 0 insertions, 102 deletions
diff --git a/libavcodec/x86/dsputil_mmx.h b/libavcodec/x86/dsputil_mmx.h index 91940f6ee8..d0b0344917 100644 --- a/libavcodec/x86/dsputil_mmx.h +++ b/libavcodec/x86/dsputil_mmx.h @@ -66,24 +66,6 @@ extern const xmm_reg ff_pb_FE; extern const double ff_pd_1[2]; extern const double ff_pd_2[2]; -#define LOAD4(stride,in,a,b,c,d)\ - "movq 0*"#stride"+"#in", "#a"\n\t"\ - "movq 1*"#stride"+"#in", "#b"\n\t"\ - "movq 2*"#stride"+"#in", "#c"\n\t"\ - "movq 3*"#stride"+"#in", "#d"\n\t" - -#define STORE4(stride,out,a,b,c,d)\ - "movq "#a", 0*"#stride"+"#out"\n\t"\ - "movq "#b", 1*"#stride"+"#out"\n\t"\ - "movq "#c", 2*"#stride"+"#out"\n\t"\ - "movq "#d", 3*"#stride"+"#out"\n\t" - -/* in/out: mma=mma+mmb, mmb=mmb-mma */ -#define SUMSUB_BA( a, b ) \ - "paddw "#b", "#a" \n\t"\ - "paddw "#b", "#b" \n\t"\ - "psubw "#a", "#b" \n\t" - #define SBUTTERFLY(a,b,t,n,m)\ "mov" #m " " #a ", " #t " \n\t" /* abcd */\ "punpckl" #n " " #b ", " #a " \n\t" /* aebf */\ @@ -95,90 +77,6 @@ extern const double ff_pd_2[2]; SBUTTERFLY(a,c,d,dq,q) /* a=aeim d=bfjn */\ SBUTTERFLY(t,b,c,dq,q) /* t=cgko c=dhlp */ -static inline void transpose4x4(uint8_t *dst, uint8_t *src, x86_reg dst_stride, x86_reg src_stride){ - __asm__ volatile( //FIXME could save 1 instruction if done as 8x4 ... - "movd (%1), %%mm0 \n\t" - "add %3, %1 \n\t" - "movd (%1), %%mm1 \n\t" - "movd (%1,%3,1), %%mm2 \n\t" - "movd (%1,%3,2), %%mm3 \n\t" - "punpcklbw %%mm1, %%mm0 \n\t" - "punpcklbw %%mm3, %%mm2 \n\t" - "movq %%mm0, %%mm1 \n\t" - "punpcklwd %%mm2, %%mm0 \n\t" - "punpckhwd %%mm2, %%mm1 \n\t" - "movd %%mm0, (%0) \n\t" - "add %2, %0 \n\t" - "punpckhdq %%mm0, %%mm0 \n\t" - "movd %%mm0, (%0) \n\t" - "movd %%mm1, (%0,%2,1) \n\t" - "punpckhdq %%mm1, %%mm1 \n\t" - "movd %%mm1, (%0,%2,2) \n\t" - - : "+&r" (dst), - "+&r" (src) - : "r" (dst_stride), - "r" (src_stride) - : "memory" - ); -} - -// e,f,g,h can be memory -// out: a,d,t,c -#define TRANSPOSE8x4(a,b,c,d,e,f,g,h,t)\ - "punpcklbw " #e ", " #a " \n\t" /* a0 e0 a1 e1 a2 e2 a3 e3 */\ - "punpcklbw " #f ", " #b " \n\t" /* b0 f0 b1 f1 b2 f2 b3 f3 */\ - "punpcklbw " #g ", " #c " \n\t" /* c0 g0 c1 g1 c2 g2 d3 g3 */\ - "punpcklbw " #h ", " #d " \n\t" /* d0 h0 d1 h1 d2 h2 d3 h3 */\ - SBUTTERFLY(a, b, t, bw, q) /* a= a0 b0 e0 f0 a1 b1 e1 f1 */\ - /* t= a2 b2 e2 f2 a3 b3 e3 f3 */\ - SBUTTERFLY(c, d, b, bw, q) /* c= c0 d0 g0 h0 c1 d1 g1 h1 */\ - /* b= c2 d2 g2 h2 c3 d3 g3 h3 */\ - SBUTTERFLY(a, c, d, wd, q) /* a= a0 b0 c0 d0 e0 f0 g0 h0 */\ - /* d= a1 b1 c1 d1 e1 f1 g1 h1 */\ - SBUTTERFLY(t, b, c, wd, q) /* t= a2 b2 c2 d2 e2 f2 g2 h2 */\ - /* c= a3 b3 c3 d3 e3 f3 g3 h3 */ - -#if ARCH_X86_64 -// permutes 01234567 -> 05736421 -#define TRANSPOSE8(a,b,c,d,e,f,g,h,t)\ - SBUTTERFLY(a,b,%%xmm8,wd,dqa)\ - SBUTTERFLY(c,d,b,wd,dqa)\ - SBUTTERFLY(e,f,d,wd,dqa)\ - SBUTTERFLY(g,h,f,wd,dqa)\ - SBUTTERFLY(a,c,h,dq,dqa)\ - SBUTTERFLY(%%xmm8,b,c,dq,dqa)\ - SBUTTERFLY(e,g,b,dq,dqa)\ - SBUTTERFLY(d,f,g,dq,dqa)\ - SBUTTERFLY(a,e,f,qdq,dqa)\ - SBUTTERFLY(%%xmm8,d,e,qdq,dqa)\ - SBUTTERFLY(h,b,d,qdq,dqa)\ - SBUTTERFLY(c,g,b,qdq,dqa)\ - "movdqa %%xmm8, "#g" \n\t" -#else -#define TRANSPOSE8(a,b,c,d,e,f,g,h,t)\ - "movdqa "#h", "#t" \n\t"\ - SBUTTERFLY(a,b,h,wd,dqa)\ - "movdqa "#h", 16"#t" \n\t"\ - "movdqa "#t", "#h" \n\t"\ - SBUTTERFLY(c,d,b,wd,dqa)\ - SBUTTERFLY(e,f,d,wd,dqa)\ - SBUTTERFLY(g,h,f,wd,dqa)\ - SBUTTERFLY(a,c,h,dq,dqa)\ - "movdqa "#h", "#t" \n\t"\ - "movdqa 16"#t", "#h" \n\t"\ - SBUTTERFLY(h,b,c,dq,dqa)\ - SBUTTERFLY(e,g,b,dq,dqa)\ - SBUTTERFLY(d,f,g,dq,dqa)\ - SBUTTERFLY(a,e,f,qdq,dqa)\ - SBUTTERFLY(h,d,e,qdq,dqa)\ - "movdqa "#h", 16"#t" \n\t"\ - "movdqa "#t", "#h" \n\t"\ - SBUTTERFLY(h,b,d,qdq,dqa)\ - SBUTTERFLY(c,g,b,qdq,dqa)\ - "movdqa 16"#t", "#g" \n\t" -#endif - #define MOVQ_WONE(regd) \ __asm__ volatile ( \ "pcmpeqd %%" #regd ", %%" #regd " \n\t" \ |