Re: [FFmpeg-devel] [PATCH 2/2] x86/vf_w3fdif: simplify w3fdif_simple_high
On Thu, Jan 7, 2016 at 3:13 AM, Hendrik Leppkes wrote: > On Mon, Oct 12, 2015 at 1:21 AM, James Almer wrote: >> On 10/11/2015 3:11 PM, Ronald S. Bultje wrote: >>> Hi, >>> >>> On Sun, Oct 11, 2015 at 1:17 PM, James Almer wrote: >>> On 10/11/2015 4:31 AM, Paul B Mahol wrote: > On 10/11/15, James Almer wrote: >> Signed-off-by: James Almer >> --- >> libavfilter/x86/vf_w3fdif.asm | 16 +++- >> 1 file changed, 7 insertions(+), 9 deletions(-) >> >> diff --git a/libavfilter/x86/vf_w3fdif.asm b/libavfilter/x86/vf_w3fdif.asm >> index f02319b..f2001a4 100644 >> --- a/libavfilter/x86/vf_w3fdif.asm >> +++ b/libavfilter/x86/vf_w3fdif.asm >> @@ -103,13 +103,11 @@ REP_RET >> >> %if ARCH_X86_64 >> >> -cglobal w3fdif_simple_high, 5, 9, 9, 0, work_line, in_lines_cur0, >> in_lines_adj0, coef, linesize >> +cglobal w3fdif_simple_high, 5, 9, 8, 0, work_line, in_lines_cur0, >> in_lines_adj0, coef, linesize >> movq m2, [coefq] >> DEFINE_ARGSwork_line, in_lines_cur0, in_lines_adj0, in_lines_cur1, >> linesize, offset, in_lines_cur2, in_lines_adj1, in_lines_adj2 >> -SPLATWm0, m2, 0 >> -SPLATWm1, m2, 1 >> +pshufdm0, m2, q >> SPLATWm2, m2, 2 >> -SBUTTERFLYwd, 0, 1, 7 >> pxor m7, m7 >> mov offsetq, 0 >> mov in_lines_cur2q, [in_lines_cur0q+gprsize*2] >> @@ -124,23 +122,23 @@ cglobal w3fdif_simple_high, 5, 9, 9, 0, work_line, >> in_lines_cur0, in_lines_adj0, >> movh m4, [in_lines_cur1q+offsetq] >> punpcklbw m3, m7 >> punpcklbw m4, m7 >> -SBUTTERFLY wd, 3, 4, 8 >> +SBUTTERFLY wd, 3, 4, 1 >> pmaddwdm3, m0 >> -pmaddwdm4, m1 >> +pmaddwdm4, m0 >> movh m5, [in_lines_adj0q+offsetq] >> movh m6, [in_lines_adj1q+offsetq] >> punpcklbw m5, m7 >> punpcklbw m6, m7 >> -SBUTTERFLY wd, 5, 6, 8 >> +SBUTTERFLY wd, 5, 6, 1 >> pmaddwdm5, m0 >> -pmaddwdm6, m1 >> +pmaddwdm6, m0 >> paddd m3, m5 >> paddd m4, m6 >> movh m5, [in_lines_cur2q+offsetq] >> movh m6, [in_lines_adj2q+offsetq] >> punpcklbw m5, m7 >> punpcklbw m6, m7 >> -SBUTTERFLY wd, 5, 6, 8 >> +SBUTTERFLY wd, 5, 6, 1 >> pmaddwdm5, m2 >> pmaddwdm6, m2 >> paddd m3, m5 >> -- >> 2.6.0 >> >> ___ >> ffmpeg-devel mailing list >> ffmpeg-devel@ffmpeg.org >> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel >> > > Cant this now be used on x32? >>> >>> Add to the data pointers directly (in_lines_cur0q and work_lineq). Then sub >>> all other curXq/adjXq from cur0q (on 32bit only) before the loop and you >>> have to adds (on 32bit) instead of one (on 64bit), but one reg less >>> (offset), making it 7, which means it works. >>> >>> Ronald >> >> Ah, like it's being done in PACK_6CH from swr's audio_convert.asm >> For complex_high some stack ab/use will be needed (see PACK_8CH), but it >> should >> be doable. >> This way w3fdif will be able to fully dethrone yadif :P > > Are you still working on w3fdif_simple_high for 32bit? > I would be interested in that. Otherwise I might try to do it myself, > although it sounds like a lot of #if'ery I was bored and it was easy, so patch coming up anyway! ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 2/2] x86/vf_w3fdif: simplify w3fdif_simple_high
On Mon, Oct 12, 2015 at 1:21 AM, James Almer wrote: > On 10/11/2015 3:11 PM, Ronald S. Bultje wrote: >> Hi, >> >> On Sun, Oct 11, 2015 at 1:17 PM, James Almer wrote: >> >>> On 10/11/2015 4:31 AM, Paul B Mahol wrote: On 10/11/15, James Almer wrote: > Signed-off-by: James Almer > --- > libavfilter/x86/vf_w3fdif.asm | 16 +++- > 1 file changed, 7 insertions(+), 9 deletions(-) > > diff --git a/libavfilter/x86/vf_w3fdif.asm >>> b/libavfilter/x86/vf_w3fdif.asm > index f02319b..f2001a4 100644 > --- a/libavfilter/x86/vf_w3fdif.asm > +++ b/libavfilter/x86/vf_w3fdif.asm > @@ -103,13 +103,11 @@ REP_RET > > %if ARCH_X86_64 > > -cglobal w3fdif_simple_high, 5, 9, 9, 0, work_line, in_lines_cur0, > in_lines_adj0, coef, linesize > +cglobal w3fdif_simple_high, 5, 9, 8, 0, work_line, in_lines_cur0, > in_lines_adj0, coef, linesize > movq m2, [coefq] > DEFINE_ARGSwork_line, in_lines_cur0, in_lines_adj0, >>> in_lines_cur1, > linesize, offset, in_lines_cur2, in_lines_adj1, in_lines_adj2 > -SPLATWm0, m2, 0 > -SPLATWm1, m2, 1 > +pshufdm0, m2, q > SPLATWm2, m2, 2 > -SBUTTERFLYwd, 0, 1, 7 > pxor m7, m7 > mov offsetq, 0 > mov in_lines_cur2q, [in_lines_cur0q+gprsize*2] > @@ -124,23 +122,23 @@ cglobal w3fdif_simple_high, 5, 9, 9, 0, work_line, > in_lines_cur0, in_lines_adj0, > movh m4, [in_lines_cur1q+offsetq] > punpcklbw m3, m7 > punpcklbw m4, m7 > -SBUTTERFLY wd, 3, 4, 8 > +SBUTTERFLY wd, 3, 4, 1 > pmaddwdm3, m0 > -pmaddwdm4, m1 > +pmaddwdm4, m0 > movh m5, [in_lines_adj0q+offsetq] > movh m6, [in_lines_adj1q+offsetq] > punpcklbw m5, m7 > punpcklbw m6, m7 > -SBUTTERFLY wd, 5, 6, 8 > +SBUTTERFLY wd, 5, 6, 1 > pmaddwdm5, m0 > -pmaddwdm6, m1 > +pmaddwdm6, m0 > paddd m3, m5 > paddd m4, m6 > movh m5, [in_lines_cur2q+offsetq] > movh m6, [in_lines_adj2q+offsetq] > punpcklbw m5, m7 > punpcklbw m6, m7 > -SBUTTERFLY wd, 5, 6, 8 > +SBUTTERFLY wd, 5, 6, 1 > pmaddwdm5, m2 > pmaddwdm6, m2 > paddd m3, m5 > -- > 2.6.0 > > ___ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > http://ffmpeg.org/mailman/listinfo/ffmpeg-devel > Cant this now be used on x32? >>> >> >> Add to the data pointers directly (in_lines_cur0q and work_lineq). Then sub >> all other curXq/adjXq from cur0q (on 32bit only) before the loop and you >> have to adds (on 32bit) instead of one (on 64bit), but one reg less >> (offset), making it 7, which means it works. >> >> Ronald > > Ah, like it's being done in PACK_6CH from swr's audio_convert.asm > For complex_high some stack ab/use will be needed (see PACK_8CH), but it > should > be doable. > This way w3fdif will be able to fully dethrone yadif :P Are you still working on w3fdif_simple_high for 32bit? I would be interested in that. Otherwise I might try to do it myself, although it sounds like a lot of #if'ery ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 2/2] x86/vf_w3fdif: simplify w3fdif_simple_high
> > > Ah, like it's being done in PACK_6CH from swr's audio_convert.asm > For complex_high some stack ab/use will be needed (see PACK_8CH), but it > should > be doable. > This way w3fdif will be able to fully dethrone yadif :P Once it supports +8bit, that is ;-) ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 2/2] x86/vf_w3fdif: simplify w3fdif_simple_high
On 10/11/2015 3:11 PM, Ronald S. Bultje wrote: > Hi, > > On Sun, Oct 11, 2015 at 1:17 PM, James Almer wrote: > >> On 10/11/2015 4:31 AM, Paul B Mahol wrote: >>> On 10/11/15, James Almer wrote: Signed-off-by: James Almer --- libavfilter/x86/vf_w3fdif.asm | 16 +++- 1 file changed, 7 insertions(+), 9 deletions(-) diff --git a/libavfilter/x86/vf_w3fdif.asm >> b/libavfilter/x86/vf_w3fdif.asm index f02319b..f2001a4 100644 --- a/libavfilter/x86/vf_w3fdif.asm +++ b/libavfilter/x86/vf_w3fdif.asm @@ -103,13 +103,11 @@ REP_RET %if ARCH_X86_64 -cglobal w3fdif_simple_high, 5, 9, 9, 0, work_line, in_lines_cur0, in_lines_adj0, coef, linesize +cglobal w3fdif_simple_high, 5, 9, 8, 0, work_line, in_lines_cur0, in_lines_adj0, coef, linesize movq m2, [coefq] DEFINE_ARGSwork_line, in_lines_cur0, in_lines_adj0, >> in_lines_cur1, linesize, offset, in_lines_cur2, in_lines_adj1, in_lines_adj2 -SPLATWm0, m2, 0 -SPLATWm1, m2, 1 +pshufdm0, m2, q SPLATWm2, m2, 2 -SBUTTERFLYwd, 0, 1, 7 pxor m7, m7 mov offsetq, 0 mov in_lines_cur2q, [in_lines_cur0q+gprsize*2] @@ -124,23 +122,23 @@ cglobal w3fdif_simple_high, 5, 9, 9, 0, work_line, in_lines_cur0, in_lines_adj0, movh m4, [in_lines_cur1q+offsetq] punpcklbw m3, m7 punpcklbw m4, m7 -SBUTTERFLY wd, 3, 4, 8 +SBUTTERFLY wd, 3, 4, 1 pmaddwdm3, m0 -pmaddwdm4, m1 +pmaddwdm4, m0 movh m5, [in_lines_adj0q+offsetq] movh m6, [in_lines_adj1q+offsetq] punpcklbw m5, m7 punpcklbw m6, m7 -SBUTTERFLY wd, 5, 6, 8 +SBUTTERFLY wd, 5, 6, 1 pmaddwdm5, m0 -pmaddwdm6, m1 +pmaddwdm6, m0 paddd m3, m5 paddd m4, m6 movh m5, [in_lines_cur2q+offsetq] movh m6, [in_lines_adj2q+offsetq] punpcklbw m5, m7 punpcklbw m6, m7 -SBUTTERFLY wd, 5, 6, 8 +SBUTTERFLY wd, 5, 6, 1 pmaddwdm5, m2 pmaddwdm6, m2 paddd m3, m5 -- 2.6.0 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel >>> >>> Cant this now be used on x32? >> > > Add to the data pointers directly (in_lines_cur0q and work_lineq). Then sub > all other curXq/adjXq from cur0q (on 32bit only) before the loop and you > have to adds (on 32bit) instead of one (on 64bit), but one reg less > (offset), making it 7, which means it works. > > Ronald Ah, like it's being done in PACK_6CH from swr's audio_convert.asm For complex_high some stack ab/use will be needed (see PACK_8CH), but it should be doable. This way w3fdif will be able to fully dethrone yadif :P ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 2/2] x86/vf_w3fdif: simplify w3fdif_simple_high
Hi, On Sun, Oct 11, 2015 at 1:17 PM, James Almer wrote: > On 10/11/2015 4:31 AM, Paul B Mahol wrote: > > On 10/11/15, James Almer wrote: > >> Signed-off-by: James Almer > >> --- > >> libavfilter/x86/vf_w3fdif.asm | 16 +++- > >> 1 file changed, 7 insertions(+), 9 deletions(-) > >> > >> diff --git a/libavfilter/x86/vf_w3fdif.asm > b/libavfilter/x86/vf_w3fdif.asm > >> index f02319b..f2001a4 100644 > >> --- a/libavfilter/x86/vf_w3fdif.asm > >> +++ b/libavfilter/x86/vf_w3fdif.asm > >> @@ -103,13 +103,11 @@ REP_RET > >> > >> %if ARCH_X86_64 > >> > >> -cglobal w3fdif_simple_high, 5, 9, 9, 0, work_line, in_lines_cur0, > >> in_lines_adj0, coef, linesize > >> +cglobal w3fdif_simple_high, 5, 9, 8, 0, work_line, in_lines_cur0, > >> in_lines_adj0, coef, linesize > >> movq m2, [coefq] > >> DEFINE_ARGSwork_line, in_lines_cur0, in_lines_adj0, > in_lines_cur1, > >> linesize, offset, in_lines_cur2, in_lines_adj1, in_lines_adj2 > >> -SPLATWm0, m2, 0 > >> -SPLATWm1, m2, 1 > >> +pshufdm0, m2, q > >> SPLATWm2, m2, 2 > >> -SBUTTERFLYwd, 0, 1, 7 > >> pxor m7, m7 > >> mov offsetq, 0 > >> mov in_lines_cur2q, [in_lines_cur0q+gprsize*2] > >> @@ -124,23 +122,23 @@ cglobal w3fdif_simple_high, 5, 9, 9, 0, work_line, > >> in_lines_cur0, in_lines_adj0, > >> movh m4, [in_lines_cur1q+offsetq] > >> punpcklbw m3, m7 > >> punpcklbw m4, m7 > >> -SBUTTERFLY wd, 3, 4, 8 > >> +SBUTTERFLY wd, 3, 4, 1 > >> pmaddwdm3, m0 > >> -pmaddwdm4, m1 > >> +pmaddwdm4, m0 > >> movh m5, [in_lines_adj0q+offsetq] > >> movh m6, [in_lines_adj1q+offsetq] > >> punpcklbw m5, m7 > >> punpcklbw m6, m7 > >> -SBUTTERFLY wd, 5, 6, 8 > >> +SBUTTERFLY wd, 5, 6, 1 > >> pmaddwdm5, m0 > >> -pmaddwdm6, m1 > >> +pmaddwdm6, m0 > >> paddd m3, m5 > >> paddd m4, m6 > >> movh m5, [in_lines_cur2q+offsetq] > >> movh m6, [in_lines_adj2q+offsetq] > >> punpcklbw m5, m7 > >> punpcklbw m6, m7 > >> -SBUTTERFLY wd, 5, 6, 8 > >> +SBUTTERFLY wd, 5, 6, 1 > >> pmaddwdm5, m2 > >> pmaddwdm6, m2 > >> paddd m3, m5 > >> -- > >> 2.6.0 > >> > >> ___ > >> ffmpeg-devel mailing list > >> ffmpeg-devel@ffmpeg.org > >> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel > >> > > > > Cant this now be used on x32? > Add to the data pointers directly (in_lines_cur0q and work_lineq). Then sub all other curXq/adjXq from cur0q (on 32bit only) before the loop and you have to adds (on 32bit) instead of one (on 64bit), but one reg less (offset), making it 7, which means it works. Ronald ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 2/2] x86/vf_w3fdif: simplify w3fdif_simple_high
On 10/11/2015 4:31 AM, Paul B Mahol wrote: > On 10/11/15, James Almer wrote: >> Signed-off-by: James Almer >> --- >> libavfilter/x86/vf_w3fdif.asm | 16 +++- >> 1 file changed, 7 insertions(+), 9 deletions(-) >> >> diff --git a/libavfilter/x86/vf_w3fdif.asm b/libavfilter/x86/vf_w3fdif.asm >> index f02319b..f2001a4 100644 >> --- a/libavfilter/x86/vf_w3fdif.asm >> +++ b/libavfilter/x86/vf_w3fdif.asm >> @@ -103,13 +103,11 @@ REP_RET >> >> %if ARCH_X86_64 >> >> -cglobal w3fdif_simple_high, 5, 9, 9, 0, work_line, in_lines_cur0, >> in_lines_adj0, coef, linesize >> +cglobal w3fdif_simple_high, 5, 9, 8, 0, work_line, in_lines_cur0, >> in_lines_adj0, coef, linesize >> movq m2, [coefq] >> DEFINE_ARGSwork_line, in_lines_cur0, in_lines_adj0, in_lines_cur1, >> linesize, offset, in_lines_cur2, in_lines_adj1, in_lines_adj2 >> -SPLATWm0, m2, 0 >> -SPLATWm1, m2, 1 >> +pshufdm0, m2, q >> SPLATWm2, m2, 2 >> -SBUTTERFLYwd, 0, 1, 7 >> pxor m7, m7 >> mov offsetq, 0 >> mov in_lines_cur2q, [in_lines_cur0q+gprsize*2] >> @@ -124,23 +122,23 @@ cglobal w3fdif_simple_high, 5, 9, 9, 0, work_line, >> in_lines_cur0, in_lines_adj0, >> movh m4, [in_lines_cur1q+offsetq] >> punpcklbw m3, m7 >> punpcklbw m4, m7 >> -SBUTTERFLY wd, 3, 4, 8 >> +SBUTTERFLY wd, 3, 4, 1 >> pmaddwdm3, m0 >> -pmaddwdm4, m1 >> +pmaddwdm4, m0 >> movh m5, [in_lines_adj0q+offsetq] >> movh m6, [in_lines_adj1q+offsetq] >> punpcklbw m5, m7 >> punpcklbw m6, m7 >> -SBUTTERFLY wd, 5, 6, 8 >> +SBUTTERFLY wd, 5, 6, 1 >> pmaddwdm5, m0 >> -pmaddwdm6, m1 >> +pmaddwdm6, m0 >> paddd m3, m5 >> paddd m4, m6 >> movh m5, [in_lines_cur2q+offsetq] >> movh m6, [in_lines_adj2q+offsetq] >> punpcklbw m5, m7 >> punpcklbw m6, m7 >> -SBUTTERFLY wd, 5, 6, 8 >> +SBUTTERFLY wd, 5, 6, 1 >> pmaddwdm5, m2 >> pmaddwdm6, m2 >> paddd m3, m5 >> -- >> 2.6.0 >> >> ___ >> ffmpeg-devel mailing list >> ffmpeg-devel@ffmpeg.org >> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel >> > > Cant this now be used on x32? Even though i got it down to eight xmm regs, we still have only seven grps to work with on x86_32. The function has seven pointers plus the offset variable used as part of effective addresses, which means they can't be accessed directly from stack for this purpose, something that can be done with linesize. So it will need some changes, like constant movs of grps to and from stack to get it working. ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
Re: [FFmpeg-devel] [PATCH 2/2] x86/vf_w3fdif: simplify w3fdif_simple_high
On 10/11/15, James Almer wrote: > Signed-off-by: James Almer > --- > libavfilter/x86/vf_w3fdif.asm | 16 +++- > 1 file changed, 7 insertions(+), 9 deletions(-) > > diff --git a/libavfilter/x86/vf_w3fdif.asm b/libavfilter/x86/vf_w3fdif.asm > index f02319b..f2001a4 100644 > --- a/libavfilter/x86/vf_w3fdif.asm > +++ b/libavfilter/x86/vf_w3fdif.asm > @@ -103,13 +103,11 @@ REP_RET > > %if ARCH_X86_64 > > -cglobal w3fdif_simple_high, 5, 9, 9, 0, work_line, in_lines_cur0, > in_lines_adj0, coef, linesize > +cglobal w3fdif_simple_high, 5, 9, 8, 0, work_line, in_lines_cur0, > in_lines_adj0, coef, linesize > movq m2, [coefq] > DEFINE_ARGSwork_line, in_lines_cur0, in_lines_adj0, in_lines_cur1, > linesize, offset, in_lines_cur2, in_lines_adj1, in_lines_adj2 > -SPLATWm0, m2, 0 > -SPLATWm1, m2, 1 > +pshufdm0, m2, q > SPLATWm2, m2, 2 > -SBUTTERFLYwd, 0, 1, 7 > pxor m7, m7 > mov offsetq, 0 > mov in_lines_cur2q, [in_lines_cur0q+gprsize*2] > @@ -124,23 +122,23 @@ cglobal w3fdif_simple_high, 5, 9, 9, 0, work_line, > in_lines_cur0, in_lines_adj0, > movh m4, [in_lines_cur1q+offsetq] > punpcklbw m3, m7 > punpcklbw m4, m7 > -SBUTTERFLY wd, 3, 4, 8 > +SBUTTERFLY wd, 3, 4, 1 > pmaddwdm3, m0 > -pmaddwdm4, m1 > +pmaddwdm4, m0 > movh m5, [in_lines_adj0q+offsetq] > movh m6, [in_lines_adj1q+offsetq] > punpcklbw m5, m7 > punpcklbw m6, m7 > -SBUTTERFLY wd, 5, 6, 8 > +SBUTTERFLY wd, 5, 6, 1 > pmaddwdm5, m0 > -pmaddwdm6, m1 > +pmaddwdm6, m0 > paddd m3, m5 > paddd m4, m6 > movh m5, [in_lines_cur2q+offsetq] > movh m6, [in_lines_adj2q+offsetq] > punpcklbw m5, m7 > punpcklbw m6, m7 > -SBUTTERFLY wd, 5, 6, 8 > +SBUTTERFLY wd, 5, 6, 1 > pmaddwdm5, m2 > pmaddwdm6, m2 > paddd m3, m5 > -- > 2.6.0 > > ___ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > http://ffmpeg.org/mailman/listinfo/ffmpeg-devel > Cant this now be used on x32? ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
[FFmpeg-devel] [PATCH 2/2] x86/vf_w3fdif: simplify w3fdif_simple_high
Signed-off-by: James Almer --- libavfilter/x86/vf_w3fdif.asm | 16 +++- 1 file changed, 7 insertions(+), 9 deletions(-) diff --git a/libavfilter/x86/vf_w3fdif.asm b/libavfilter/x86/vf_w3fdif.asm index f02319b..f2001a4 100644 --- a/libavfilter/x86/vf_w3fdif.asm +++ b/libavfilter/x86/vf_w3fdif.asm @@ -103,13 +103,11 @@ REP_RET %if ARCH_X86_64 -cglobal w3fdif_simple_high, 5, 9, 9, 0, work_line, in_lines_cur0, in_lines_adj0, coef, linesize +cglobal w3fdif_simple_high, 5, 9, 8, 0, work_line, in_lines_cur0, in_lines_adj0, coef, linesize movq m2, [coefq] DEFINE_ARGSwork_line, in_lines_cur0, in_lines_adj0, in_lines_cur1, linesize, offset, in_lines_cur2, in_lines_adj1, in_lines_adj2 -SPLATWm0, m2, 0 -SPLATWm1, m2, 1 +pshufdm0, m2, q SPLATWm2, m2, 2 -SBUTTERFLYwd, 0, 1, 7 pxor m7, m7 mov offsetq, 0 mov in_lines_cur2q, [in_lines_cur0q+gprsize*2] @@ -124,23 +122,23 @@ cglobal w3fdif_simple_high, 5, 9, 9, 0, work_line, in_lines_cur0, in_lines_adj0, movh m4, [in_lines_cur1q+offsetq] punpcklbw m3, m7 punpcklbw m4, m7 -SBUTTERFLY wd, 3, 4, 8 +SBUTTERFLY wd, 3, 4, 1 pmaddwdm3, m0 -pmaddwdm4, m1 +pmaddwdm4, m0 movh m5, [in_lines_adj0q+offsetq] movh m6, [in_lines_adj1q+offsetq] punpcklbw m5, m7 punpcklbw m6, m7 -SBUTTERFLY wd, 5, 6, 8 +SBUTTERFLY wd, 5, 6, 1 pmaddwdm5, m0 -pmaddwdm6, m1 +pmaddwdm6, m0 paddd m3, m5 paddd m4, m6 movh m5, [in_lines_cur2q+offsetq] movh m6, [in_lines_adj2q+offsetq] punpcklbw m5, m7 punpcklbw m6, m7 -SBUTTERFLY wd, 5, 6, 8 +SBUTTERFLY wd, 5, 6, 1 pmaddwdm5, m2 pmaddwdm6, m2 paddd m3, m5 -- 2.6.0 ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
