Re: [FFmpeg-devel] [PATCH v2] swscale/output: Altivec-optimize yuv2plane1_8

2018-11-26 Thread Carl Eugen Hoyos
2018-11-21 18:19 GMT+01:00, Lauri Kasanen : > First I ran with THREADS=3, baseline blew up in > fate-h264-conformance-frext-hpcafl_bcrm_c I opened ticket #7570, thank you for the report! Carl Eugen ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org

Re: [FFmpeg-devel] [PATCH v2] swscale/output: Altivec-optimize yuv2plane1_8

2018-11-26 Thread Carl Eugen Hoyos
2018-11-27 0:17 GMT+01:00, Carl Eugen Hoyos : > 2018-11-17 9:12 GMT+01:00, Lauri Kasanen : >> ./ffmpeg_g -f rawvideo -pix_fmt rgb24 -s hd1080 -i /dev/zero -pix_fmt >> yuv420p \ >> -f null -vframes 100 -v error -nostats - >> >> 1158 UNITS in planar1, 65528 runs, 8 skips >> >> -cpuflags 0 >> >

Re: [FFmpeg-devel] [PATCH v2] swscale/output: Altivec-optimize yuv2plane1_8

2018-11-26 Thread Carl Eugen Hoyos
2018-11-17 9:12 GMT+01:00, Lauri Kasanen : > ./ffmpeg_g -f rawvideo -pix_fmt rgb24 -s hd1080 -i /dev/zero -pix_fmt > yuv420p \ > -f null -vframes 100 -v error -nostats - > > 1158 UNITS in planar1, 65528 runs, 8 skips > > -cpuflags 0 > > 19082 UNITS in planar1, 65533 runs, 3 skips > >

Re: [FFmpeg-devel] [PATCH v2] swscale/output: Altivec-optimize yuv2plane1_8

2018-11-26 Thread Lauri Kasanen
On Mon, 26 Nov 2018 11:03:55 +0300 Michael Kostylev wrote: > > http://fate.xffm.org/?sort=arch > /ppc Yeah, mentioned in the commit message. Follow-up patch coming today. - Lauri ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org

Re: [FFmpeg-devel] [PATCH v2] swscale/output: Altivec-optimize yuv2plane1_8

2018-11-25 Thread Michael Niedermayer
On Sat, Nov 24, 2018 at 11:33:01AM +0200, Lauri Kasanen wrote: > On Fri, 23 Nov 2018 23:01:02 +0100 > Michael Niedermayer wrote: > > > On Fri, Nov 23, 2018 at 10:38:13AM +0200, Lauri Kasanen wrote: > > > I mean, if my patch adds no failures, is that enough to apply it? > > > > yes that and the t

Re: [FFmpeg-devel] [PATCH v2] swscale/output: Altivec-optimize yuv2plane1_8

2018-11-24 Thread Lauri Kasanen
On Fri, 23 Nov 2018 23:01:02 +0100 Michael Niedermayer wrote: > On Fri, Nov 23, 2018 at 10:38:13AM +0200, Lauri Kasanen wrote: > > I mean, if my patch adds no failures, is that enough to apply it? > > yes that and the tests failing should still fail the same way with the > same checksums > This

Re: [FFmpeg-devel] [PATCH v2] swscale/output: Altivec-optimize yuv2plane1_8

2018-11-23 Thread Michael Niedermayer
On Fri, Nov 23, 2018 at 10:38:13AM +0200, Lauri Kasanen wrote: > On Fri, 23 Nov 2018 03:26:50 +0100 > Michael Niedermayer wrote: > > > On Wed, Nov 21, 2018 at 07:19:45PM +0200, Lauri Kasanen wrote: > > > On Wed, 21 Nov 2018 17:22:36 +0100 > > > Michael Niedermayer wrote: > > > > the full fate te

Re: [FFmpeg-devel] [PATCH v2] swscale/output: Altivec-optimize yuv2plane1_8

2018-11-23 Thread Lauri Kasanen
On Fri, 23 Nov 2018 03:26:50 +0100 Michael Niedermayer wrote: > On Wed, Nov 21, 2018 at 07:19:45PM +0200, Lauri Kasanen wrote: > > On Wed, 21 Nov 2018 17:22:36 +0100 > > Michael Niedermayer wrote: > > > the full fate tests must be run, many of these tests use swscale without > > > having "scale"

Re: [FFmpeg-devel] [PATCH v2] swscale/output: Altivec-optimize yuv2plane1_8

2018-11-22 Thread Michael Niedermayer
On Wed, Nov 21, 2018 at 07:19:45PM +0200, Lauri Kasanen wrote: > On Wed, 21 Nov 2018 17:22:36 +0100 > Michael Niedermayer wrote: > > > the full fate tests must be run, many of these tests use swscale without > > having "scale" in their name > > and yes on lower end hardware 20min and longer is po

Re: [FFmpeg-devel] [PATCH v2] swscale/output: Altivec-optimize yuv2plane1_8

2018-11-21 Thread Lauri Kasanen
On Wed, 21 Nov 2018 17:22:36 +0100 Michael Niedermayer wrote: > the full fate tests must be run, many of these tests use swscale without > having "scale" in their name > and yes on lower end hardware 20min and longer is possible I get failures on the baseline, without my patch. What is the proce

Re: [FFmpeg-devel] [PATCH v2] swscale/output: Altivec-optimize yuv2plane1_8

2018-11-21 Thread Michael Niedermayer
On Wed, Nov 21, 2018 at 02:35:32PM +0200, Lauri Kasanen wrote: > On Wed, 21 Nov 2018 13:21:58 +0100 > Michael Niedermayer wrote: > > > On Wed, Nov 21, 2018 at 10:12:48AM +0200, Lauri Kasanen wrote: > > > > ./ffmpeg_g -f rawvideo -pix_fmt rgb24 -s hd1080 -i /dev/zero -pix_fmt > > > > yuv420p \ >

Re: [FFmpeg-devel] [PATCH v2] swscale/output: Altivec-optimize yuv2plane1_8

2018-11-21 Thread Lauri Kasanen
On Wed, 21 Nov 2018 13:21:58 +0100 Michael Niedermayer wrote: > On Wed, Nov 21, 2018 at 10:12:48AM +0200, Lauri Kasanen wrote: > > > ./ffmpeg_g -f rawvideo -pix_fmt rgb24 -s hd1080 -i /dev/zero -pix_fmt > > > yuv420p \ > > > -f null -vframes 100 -v error -nostats - > > > > > > 1158 UNITS in pla

Re: [FFmpeg-devel] [PATCH v2] swscale/output: Altivec-optimize yuv2plane1_8

2018-11-21 Thread Michael Niedermayer
On Wed, Nov 21, 2018 at 10:12:48AM +0200, Lauri Kasanen wrote: > > ./ffmpeg_g -f rawvideo -pix_fmt rgb24 -s hd1080 -i /dev/zero -pix_fmt > > yuv420p \ > > -f null -vframes 100 -v error -nostats - > > > > 1158 UNITS in planar1, 65528 runs, 8 skips > > > > -cpuflags 0 > > > > 19082 UNITS i

Re: [FFmpeg-devel] [PATCH v2] swscale/output: Altivec-optimize yuv2plane1_8

2018-11-21 Thread Lauri Kasanen
> ./ffmpeg_g -f rawvideo -pix_fmt rgb24 -s hd1080 -i /dev/zero -pix_fmt yuv420p > \ > -f null -vframes 100 -v error -nostats - > > 1158 UNITS in planar1, 65528 runs, 8 skips > > -cpuflags 0 > > 19082 UNITS in planar1, 65533 runs, 3 skips > > 16.48 speedup ratio. On x86, SSE2 is ~

Re: [FFmpeg-devel] [PATCH v2] swscale/output: Altivec-optimize yuv2plane1_8

2018-11-17 Thread Michael Niedermayer
On Sat, Nov 17, 2018 at 10:12:14AM +0200, Lauri Kasanen wrote: > ./ffmpeg_g -f rawvideo -pix_fmt rgb24 -s hd1080 -i /dev/zero -pix_fmt yuv420p > \ > -f null -vframes 100 -v error -nostats - > > 1158 UNITS in planar1, 65528 runs, 8 skips > > -cpuflags 0 > > 19082 UNITS in planar1, 65533