Re: [FFmpeg-devel] [PATCH] swscale/arm: add ff_nv{12, 21}_to_{argb, rgba, abgr, bgra}_neon

2015-11-23 Thread Clément Bœsch
On Thu, Nov 19, 2015 at 11:48:53AM +0100, Clément Bœsch wrote: > From: Matthieu Bouron > > Signed-off-by: Matthieu Bouron > Signed-off-by: Clément Bœsch > > --- > The function takes about 29ms with a 1080p source (testsrc2) on a > cortex-a8. Though, 16ms (more than half the time!) is spend in

Re: [FFmpeg-devel] [PATCH] swscale/arm: add ff_nv{12, 21}_to_{argb, rgba, abgr, bgra}_neon

2015-11-20 Thread Michael Niedermayer
On Thu, Nov 19, 2015 at 06:29:23PM +0100, Clément Bœsch wrote: > On Thu, Nov 19, 2015 at 04:50:54PM +0100, Michael Niedermayer wrote: > > On Thu, Nov 19, 2015 at 11:48:53AM +0100, Clément Bœsch wrote: > > > From: Matthieu Bouron > > > > > > Signed-off-by: Matthieu Bouron > > > Signed-off-by: Clé

Re: [FFmpeg-devel] [PATCH] swscale/arm: add ff_nv{12, 21}_to_{argb, rgba, abgr, bgra}_neon

2015-11-19 Thread Clément Bœsch
On Thu, Nov 19, 2015 at 04:50:54PM +0100, Michael Niedermayer wrote: > On Thu, Nov 19, 2015 at 11:48:53AM +0100, Clément Bœsch wrote: > > From: Matthieu Bouron > > > > Signed-off-by: Matthieu Bouron > > Signed-off-by: Clément Bœsch > > > > --- > > The function takes about 29ms with a 1080p sou

Re: [FFmpeg-devel] [PATCH] swscale/arm: add ff_nv{12, 21}_to_{argb, rgba, abgr, bgra}_neon

2015-11-19 Thread Michael Niedermayer
On Thu, Nov 19, 2015 at 11:48:53AM +0100, Clément Bœsch wrote: > From: Matthieu Bouron > > Signed-off-by: Matthieu Bouron > Signed-off-by: Clément Bœsch > > --- > The function takes about 29ms with a 1080p source (testsrc2) on a > cortex-a8. Though, 16ms (more than half the time!) is spend in

Re: [FFmpeg-devel] [PATCH] swscale/arm: add ff_nv{12, 21}_to_{argb, rgba, abgr, bgra}_neon

2015-11-19 Thread Clément Bœsch
On Thu, Nov 19, 2015 at 11:48:53AM +0100, Clément Bœsch wrote: [...] > +vst2.32 {q6, q7}, [\dst,:128]! Sorry, this is supposed to be vst4.8 -- Clément B. signature.asc Description: PGP signature ___ ffmpeg-devel mailing list ffmpeg-de

[FFmpeg-devel] [PATCH] swscale/arm: add ff_nv{12, 21}_to_{argb, rgba, abgr, bgra}_neon

2015-11-19 Thread Clément Bœsch
From: Matthieu Bouron Signed-off-by: Matthieu Bouron Signed-off-by: Clément Bœsch --- The function takes about 29ms with a 1080p source (testsrc2) on a cortex-a8. Though, 16ms (more than half the time!) is spend in the vst2 call. Any suggestion on how to speed up this? Also, the reference cod