diff options
| author | Michael Niedermayer <michaelni@gmx.at> | 2007-08-21 00:03:18 +0000 |
|---|---|---|
| committer | Michael Niedermayer <michaelni@gmx.at> | 2007-08-21 00:03:18 +0000 |
| commit | 72dee89b5d37b6b4c4dedce63ffdd28060186b75 (patch) | |
| tree | b3209063114c6f16a839f2285c26243e5b09f0a6 /libavcodec | |
| parent | d0dae46a9087874116316352815fac3bac9284d6 (diff) | |
| download | ffmpeg-72dee89b5d37b6b4c4dedce63ffdd28060186b75.tar.gz | |
Simplify and optimize the 4th vertical lifting step of the SSE2 code (untested)
This also reduces the needed headroom in that step by 1 bit
Originally committed as revision 10165 to svn://svn.ffmpeg.org/ffmpeg/trunk
Diffstat (limited to 'libavcodec')
| -rw-r--r-- | libavcodec/i386/snowdsp_mmx.c | 3 |
1 files changed, 1 insertions, 2 deletions
diff --git a/libavcodec/i386/snowdsp_mmx.c b/libavcodec/i386/snowdsp_mmx.c index cfdfb951fc..b6509aab6b 100644 --- a/libavcodec/i386/snowdsp_mmx.c +++ b/libavcodec/i386/snowdsp_mmx.c @@ -498,9 +498,8 @@ void ff_snow_vertical_compose97i_sse2(DWTELEM *b0, DWTELEM *b1, DWTELEM *b2, DWT snow_vertical_compose_sse2_store(REG_c,"xmm0","xmm2","xmm4","xmm6") snow_vertical_compose_sse2_add(REG_S,"xmm0","xmm2","xmm4","xmm6") snow_vertical_compose_sse2_move("xmm0","xmm2","xmm4","xmm6","xmm1","xmm3","xmm5","xmm7") - snow_vertical_compose_sse2_r2r_add("xmm0","xmm2","xmm4","xmm6","xmm0","xmm2","xmm4","xmm6") - snow_vertical_compose_sse2_r2r_add("xmm1","xmm3","xmm5","xmm7","xmm0","xmm2","xmm4","xmm6") snow_vertical_compose_sse2_sra("1","xmm0","xmm2","xmm4","xmm6") + snow_vertical_compose_sse2_r2r_add("xmm1","xmm3","xmm5","xmm7","xmm0","xmm2","xmm4","xmm6") snow_vertical_compose_sse2_add(REG_a,"xmm0","xmm2","xmm4","xmm6") snow_vertical_compose_sse2_store(REG_a,"xmm0","xmm2","xmm4","xmm6") |