diff options
| author | Michael Niedermayer <michaelni@gmx.at> | 2012-08-04 22:39:25 +0200 |
|---|---|---|
| committer | Michael Niedermayer <michaelni@gmx.at> | 2012-08-04 23:51:43 +0200 |
| commit | e776ee8f294984f7643a3c45db803c7266e1edfd (patch) | |
| tree | a1fd00ab7a0760ec0f2848aed9dc3f79d889e816 /libavcodec/x86/dsputil_yasm.asm | |
| parent | 88fc1438c693ffb7793aeb111d89775440491840 (diff) | |
| parent | 8821ae649e61097ec57ca58472c3e4239c82913c (diff) | |
| download | ffmpeg-e776ee8f294984f7643a3c45db803c7266e1edfd.tar.gz | |
Merge remote-tracking branch 'qatar/master'
* qatar/master:
lavr: fix handling of custom mix matrices
fate: force pix_fmt in lagarith-rgb32 test
fate: add tests for lagarith lossless video codec.
ARMv6: vp8: fix stack allocation with Apple's assembler
ARM: vp56: allow inline asm to build with clang
fft: 3dnow: fix register name typo in DECL_IMDCT macro
x86: dct32: port to cpuflags
x86: build: replace mmx2 by mmxext
Revert "wmapro: prevent division by zero when sample rate is unspecified"
wmapro: prevent division by zero when sample rate is unspecified
lagarith: fix color plane inversion for YUY2 output.
lagarith: pad RGB buffer by 1 byte.
dsputil: make add_hfyu_left_prediction_sse4() support unaligned src.
Conflicts:
doc/APIchanges
libavcodec/lagarith.c
libavfilter/x86/gradfun.c
libavutil/cpu.h
libavutil/version.h
libswscale/utils.c
libswscale/version.h
libswscale/x86/yuv2rgb.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Diffstat (limited to 'libavcodec/x86/dsputil_yasm.asm')
| -rw-r--r-- | libavcodec/x86/dsputil_yasm.asm | 20 |
1 files changed, 13 insertions, 7 deletions
diff --git a/libavcodec/x86/dsputil_yasm.asm b/libavcodec/x86/dsputil_yasm.asm index 5a6c3d1eae..06d2027c69 100644 --- a/libavcodec/x86/dsputil_yasm.asm +++ b/libavcodec/x86/dsputil_yasm.asm @@ -388,12 +388,16 @@ cglobal add_hfyu_median_prediction_mmx2, 6,6,0, dst, top, diff, w, left, left_to RET -%macro ADD_HFYU_LEFT_LOOP 1 ; %1 = is_aligned +%macro ADD_HFYU_LEFT_LOOP 2 ; %1 = dst_is_aligned, %2 = src_is_aligned add srcq, wq add dstq, wq neg wq %%.loop: +%if %2 mova m1, [srcq+wq] +%else + movu m1, [srcq+wq] +%endif mova m2, m1 psllw m1, 8 paddb m1, m2 @@ -435,7 +439,7 @@ cglobal add_hfyu_left_prediction_ssse3, 3,3,7, dst, src, w, left mova m3, [pb_zz11zz55zz99zzdd] movd m0, leftm psllq m0, 56 - ADD_HFYU_LEFT_LOOP 1 + ADD_HFYU_LEFT_LOOP 1, 1 INIT_XMM cglobal add_hfyu_left_prediction_sse4, 3,3,7, dst, src, w, left @@ -446,12 +450,14 @@ cglobal add_hfyu_left_prediction_sse4, 3,3,7, dst, src, w, left movd m0, leftm pslldq m0, 15 test srcq, 15 - jnz add_hfyu_left_prediction_ssse3.skip_prologue + jnz .src_unaligned test dstq, 15 - jnz .unaligned - ADD_HFYU_LEFT_LOOP 1 -.unaligned: - ADD_HFYU_LEFT_LOOP 0 + jnz .dst_unaligned + ADD_HFYU_LEFT_LOOP 1, 1 +.dst_unaligned: + ADD_HFYU_LEFT_LOOP 0, 1 +.src_unaligned: + ADD_HFYU_LEFT_LOOP 0, 0 ; float scalarproduct_float_sse(const float *v1, const float *v2, int len) |