Re: [FFmpeg-devel] [PATCH] avfilter/vf_deshake: replace qsort with AV_QSORT

Le nonidi 9 brumaire, an CCXXIV, Ganesh Ajjanagadde a écrit : > Sample benchmark (x86-64, Haswell, GNU/Linux): > File: original from https://trac.ffmpeg.org/ticket/1430 > command: ffmpeg -stream_loop 8 -i file.webm -vf deshake=rx=64:ry=64 -f null - > > Timer truncated at 1024 runs. > new: > 39676

Re: [FFmpeg-devel] [PATCH] avfilter/vf_deshake: replace qsort with AV_QSORT

On Sat, Oct 31, 2015 at 11:19:04AM +0100, Nicolas George wrote: > Le nonidi 9 brumaire, an CCXXIV, Ganesh Ajjanagadde a écrit : > > Sample benchmark (x86-64, Haswell, GNU/Linux): > > File: original from https://trac.ffmpeg.org/ticket/1430 > > command: ffmpeg -stream_loop 8 -i file.webm -vf

Re: [FFmpeg-devel] [PATCH] avfilter/vf_deshake: replace qsort with AV_QSORT

On Sat, Oct 31, 2015 at 6:19 AM, Nicolas George wrote: > Le nonidi 9 brumaire, an CCXXIV, Ganesh Ajjanagadde a écrit : >> Sample benchmark (x86-64, Haswell, GNU/Linux): >> File: original from https://trac.ffmpeg.org/ticket/1430 >> command: ffmpeg -stream_loop 8 -i file.webm -vf

Re: [FFmpeg-devel] [PATCH] avfilter/vf_deshake: replace qsort with AV_QSORT

On Sat, Oct 31, 2015 at 12:43 PM, Nicolas George wrote: > Le decadi 10 brumaire, an CCXXIV, Ganesh Ajjanagadde a écrit : >> Entirely possible. I just did this as a low hanging fruit. I highly >> suspect that the algorithm can be improved (see Michael's comment). >> The start/stop

Re: [FFmpeg-devel] [PATCH] avfilter/vf_deshake: replace qsort with AV_QSORT

Le decadi 10 brumaire, an CCXXIV, Ganesh Ajjanagadde a écrit : > For the benefit of the thread, here is a benchmark for the find_motion > function to address this point: > new: > 1815531917 decicycles in find_motion,1024 runs, 0 skips > > old: > 1900282483 decicycles in find_motion,

Re: [FFmpeg-devel] [PATCH] avfilter/vf_deshake: replace qsort with AV_QSORT

On Sat, Oct 31, 2015 at 1:27 PM, Ganesh Ajjanagadde wrote: > On Sat, Oct 31, 2015 at 12:43 PM, Nicolas George wrote: >> Le decadi 10 brumaire, an CCXXIV, Ganesh Ajjanagadde a écrit : >>> Entirely possible. I just did this as a low hanging fruit. I highly >>>

Re: [FFmpeg-devel] possible farewell; dev policy clarification

On Sat, Oct 31, 2015 at 1:01 AM, Rostislav Pehlivanov wrote: > I agree with the others. There's quite a lot of noise on the ML and having > some of > your trivial patches (e.g. replacing functions with FFmpeg defined ones) > merged into > one would help to decrease it. Could

[FFmpeg-devel] [PATCH 1/4] avfilter/all: check and propagate the return code of av_expr_parse_and_eval

This function can return ENOMEM that needs to be propagated. Signed-off-by: Ganesh Ajjanagadde --- libavfilter/vf_pad.c| 10 ++ libavfilter/vf_rotate.c | 3 +-- libavfilter/vf_scale.c | 5 +++-- libavfilter/vf_zscale.c | 5 +++-- 4 files changed, 13

Re: [FFmpeg-devel] [PATCH] avfilter/vf_deshake: replace qsort with AV_QSORT

On Sat, Oct 31, 2015 at 2:30 PM, Nicolas George wrote: > Le decadi 10 brumaire, an CCXXIV, Ganesh Ajjanagadde a écrit : >> For the benefit of the thread, here is a benchmark for the find_motion >> function to address this point: >> new: >> 1815531917 decicycles in find_motion,

[FFmpeg-devel] [PATCH 1/2] avcodec: add Interplay ACM decoder

Signed-off-by: Paul B Mahol --- libavcodec/Makefile | 1 + libavcodec/allcodecs.c| 1 + libavcodec/avcodec.h | 1 + libavcodec/codec_desc.c | 7 + libavcodec/interplayacm.c | 590 ++ 5 files changed, 600

[FFmpeg-devel] [PATCH 2/2] avformat: add acm demuxer

Signed-off-by: Paul B Mahol --- libavformat/Makefile | 1 + libavformat/acm.c| 71 libavformat/allformats.c | 1 + 3 files changed, 73 insertions(+) create mode 100644 libavformat/acm.c diff --git

Re: [FFmpeg-devel] [PATCH] avutil/opencl_internal: add av_warn_unused_result

On Thu, Oct 29, 2015 at 4:26 PM, Michael Niedermayer wrote: > On Wed, Oct 28, 2015 at 10:53:59PM -0400, Ganesh Ajjanagadde wrote: >> On Thu, Oct 15, 2015 at 6:24 PM, Ganesh Ajjanagadde >> wrote: >> > Signed-off-by: Ganesh Ajjanagadde

[FFmpeg-devel] [PATCH 0/4] av_warn_unused_result for avutil/eval

This is an attempt on my end to be more systematic and avoid sloppiness and noise while sending patches. As can be seen, I have made use of the cover letter option and have organized an av_warn_unused_result change into one cohesive unit: 1/4 fixes issues of unchecked error codes from

Re: [FFmpeg-devel] [PATCH] avfilter/vf_deshake: replace qsort with AV_QSORT

On Sat, Oct 31, 2015 at 9:59 AM, Clément Bœsch wrote: > On Sat, Oct 31, 2015 at 08:47:37AM -0400, Ganesh Ajjanagadde wrote: >> On Sat, Oct 31, 2015 at 6:19 AM, Nicolas George wrote: >> > Le nonidi 9 brumaire, an CCXXIV, Ganesh Ajjanagadde a écrit : >> >> Sample

Re: [FFmpeg-devel] [PATCH] avdevice/dshow_enummediatypes: check return of av_malloc

On Wed, Oct 28, 2015 at 10:05 PM, Ganesh Ajjanagadde wrote: > On Wed, Oct 28, 2015 at 10:00 PM, Michael Niedermayer > wrote: >> On Tue, Oct 27, 2015 at 08:09:03PM -0400, Ganesh Ajjanagadde wrote: >>> Untested. >>> >>> Signed-off-by: Ganesh Ajjanagadde

[FFmpeg-devel] [PATCH 4/4] avutil/eval: add av_warn_unused_result

These functions do mallocs that can fail and return AVERROR(ENOMEM) that needs to be checked. Signed-off-by: Ganesh Ajjanagadde --- libavutil/eval.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/libavutil/eval.h b/libavutil/eval.h index 6159b0f..d65fef1 100644

Re: [FFmpeg-devel] [PATCH] avfilter/vf_deshake: replace qsort with AV_QSORT

Le decadi 10 brumaire, an CCXXIV, Ganesh Ajjanagadde a écrit : > Entirely possible. I just did this as a low hanging fruit. I highly > suspect that the algorithm can be improved (see Michael's comment). > The start/stop was just around the qsort statement. Since you asked So the speedup observed

Re: [FFmpeg-devel] [FFmpeg-cvslog] avfilter/vf_deshake: replace qsort with AV_QSORT

Le decadi 10 brumaire, an CCXXIV, Ganesh Ajjanagadde a écrit : > Reviewed-by: Nicolas George That is not true. Regards, -- Nicolas George signature.asc Description: Digital signature ___ ffmpeg-devel mailing list

Re: [FFmpeg-devel] [PATCH] avfilter/vf_deshake: replace qsort with AV_QSORT

On Sat, Oct 31, 2015 at 08:47:37AM -0400, Ganesh Ajjanagadde wrote: > On Sat, Oct 31, 2015 at 6:19 AM, Nicolas George wrote: > > Le nonidi 9 brumaire, an CCXXIV, Ganesh Ajjanagadde a écrit : > >> Sample benchmark (x86-64, Haswell, GNU/Linux): > >> File: original from

[FFmpeg-devel] [PATCH 2/4] avfilter/vf_rotate: correct log message

There seems to be some typos in the log messages that are fixed by this. Signed-off-by: Ganesh Ajjanagadde --- libavfilter/vf_rotate.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/libavfilter/vf_rotate.c b/libavfilter/vf_rotate.c index

[FFmpeg-devel] [PATCH 3/4] avutil/eval: check av_expr_parse_and_eval error code in the test build

This returns the error code from main in the test, in this case ENOMEM. This should not matter, and will ensure that no warnings are triggered when av_warn_unused_result is added to avutil/eval.h. Signed-off-by: Ganesh Ajjanagadde --- libavutil/eval.c | 22

[FFmpeg-devel] IETF Standardization: RFC by November 9

Dear all, We are excited to announce that the our CELLAR working group charter for the standardization of FFV1, Matroska, and FLAC has been made available for external review. What this means is that it is now under review and available for comment to the wider IETF community. Please send any

Re: [FFmpeg-devel] [PATCH 06/11] avfilter/af_volumedetect: use log10 instead of hardcoded constant

On Wed, Oct 28, 2015 at 9:20 PM, Ganesh Ajjanagadde wrote: > This is likely more precise and conveys the intent better. > > Signed-off-by: Ganesh Ajjanagadde > --- > libavfilter/af_volumedetect.c | 2 +- > 1 file changed, 1 insertion(+), 1

Re: [FFmpeg-devel] [PATCH 06/11] avfilter/af_volumedetect: use log10 instead of hardcoded constant

On Sat, Oct 31, 2015 at 6:02 PM, Mark Harris wrote: > On Wed, Oct 28, 2015 at 9:20 PM, Ganesh Ajjanagadde > wrote: >> This is likely more precise and conveys the intent better. >> >> Signed-off-by: Ganesh Ajjanagadde >> --- >>

Re: [FFmpeg-devel] [PATCH] pixblockdsp: Use memcpy for get_pixels_16_c

On Tue, Oct 20, 2015 at 11:29 PM Michael Niedermayer wrote: > > diff --git a/libavcodec/pixblockdsp.c b/libavcodec/pixblockdsp.c > > index 322e1dd..0f23d8a 100644 > > --- a/libavcodec/pixblockdsp.c > > +++ b/libavcodec/pixblockdsp.c > > @@ -23,12 +23,38 @@ > > #include