Re: [FFmpeg-devel] How to preserve @subheading directives, and styling, when compiling doc/developer.texi ?

2017-11-26 Thread Jim DeLaHunt
On 2017-11-21 01:20, Jim DeLaHunt wrote: Hello, doc maintainers: Could I get some help with the Texinfo compilation which produces /developer.html, please? Page is a document which has some section headings displayed in green text, and some subheadings

[FFmpeg-devel] [PATCH] hls demuxer: add option to defer parsing of variants

2017-11-26 Thread Rainer Hochecker
fixed mem leak poined out by Steven --- doc/demuxers.texi | 5 + libavformat/hls.c | 304 -- 2 files changed, 209 insertions(+), 100 deletions(-) diff --git a/doc/demuxers.texi b/doc/demuxers.texi index 73dc0feec1..634b122e10 100644 ---

Re: [FFmpeg-devel] [PATCH] Refactor Developer Docs, update dev list section (v2)

2017-11-26 Thread Nicolas George
Paul B Mahol (2017-11-26): > Your opinions are irrelevant. # Be friendly and respectful towards others and third parties. # Treat others the way you yourself want to be treated. Please stop trampling the code of conduct. -- Nicolas George signature.asc Description: Digital signature

Re: [FFmpeg-devel] [PATCH] Refactor Developer Docs, update dev list section (v2)

2017-11-26 Thread Paul B Mahol
On 11/26/17, Carl Eugen Hoyos wrote: > 2017-11-26 9:31 GMT+01:00 Jim DeLaHunt : > >> -@subsection Documentation/Other >> +@section Documentation/Other >> +@subheading Subscribe to the ffmpeg-devel mailing list. >> +It is important to be subscribed to

[FFmpeg-devel] [PATCH] Refactor Developer Docs, update dev list section (v2)

2017-11-26 Thread Jim DeLaHunt
Previously, the Developer Documentation contained a single chapter, 1. Developer Guide, with all content under that single chapter. Thus the document structure was one level deeper and more complicated than it needed to be. It differed from similar documents such as /faq.html, which have

Re: [FFmpeg-devel] [PATCH] hls demuxer: add option to defer parsing of variants

2017-11-26 Thread Steven Liu
2017-11-26 18:46 GMT+08:00 Rainer Hochecker : > fixed mem leak poined out by Steven Hi Rainer, I'm not sure that is memleak, but looks like memleak when reading the code, i see the code always in hls.c before this patch, but no people report it memleak. If that is

Re: [FFmpeg-devel] [PATCH] Refactor Developer Docs, update dev list section (v2)

2017-11-26 Thread Carl Eugen Hoyos
2017-11-26 9:31 GMT+01:00 Jim DeLaHunt : > -@subsection Documentation/Other > +@section Documentation/Other > +@subheading Subscribe to the ffmpeg-devel mailing list. > +It is important to be subscribed to the Of course it is important but I would much, much prefer if

Re: [FFmpeg-devel] [PATCH 00/15] OpenCL infrastructure, filters

2017-11-26 Thread hydra3333
OK, the experimental -opencl_bench option has been removed by recent commits. Thank you for the updates to OpenCL. It cross-compiles OK, however I'm not sure about the replacement functionality. So, 2 things. 1. may I enquire how one ascertains the numbers to use like this ?

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Carl Eugen Hoyos
2017-11-26 14:40 GMT+01:00 Derek Buitenhuis : > On 11/26/2017 1:35 PM, Carl Eugen Hoyos wrote: >> Attached patch adds a missing dependency to libvmaf, I don't know if >> other threads also work. > > This should also be filed as a bug against libvmaf, since its

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Derek Buitenhuis
On 11/26/2017 2:06 PM, Carl Eugen Hoyos wrote: > In this case the library and our interface both depend on pthreads > but configure ignores this. Sorry, I wasn't aware our own wrapper code use pthreads too. Patch should be OK then. Upstream pkg-config file is still broken, though. :) - Derek

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Nicolas George
Derek Buitenhuis (2017-11-26): > The problem is that libvmaf's .pc file put all of its deps in > Libs instead of splitting them out into Libs.private, which > is used only when static linking. Stuff like -lpthreads is > only needed if static linking, and stuff like -lstdc++ is > just wrong on any

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Derek Buitenhuis
On 11/26/2017 2:09 PM, Carl Eugen Hoyos wrote: > As said before, it is non-trivial to find such a system > (it worked fine here on osx when I tested last). OS X does not ship with libstdc++, IIRC. You must have been using a ports-build toolchain? Using clang-cl or MSVC on windows will also not

Re: [FFmpeg-devel] [PATCH] lavc: reset codec on receiving packet after EOF in compat_decode

2017-11-26 Thread Michael Niedermayer
On Tue, Nov 21, 2017 at 10:48:19PM +0100, Marton Balint wrote: > > > On Thu, 9 Nov 2017, James Cowgill wrote: > > >Hi, > > > >On 09/11/17 14:02, Hendrik Leppkes wrote: > >>On Thu, Nov 9, 2017 at 1:21 PM, James Cowgill wrote: > >>>In commit 061a0c14bb57 ("decode:

Re: [FFmpeg-devel] [PATCH 0/1][TOOL][HACK] Allocation NULL check fuzzer

2017-11-26 Thread Derek Buitenhuis
On 11/25/2017 12:07 AM, Michael Niedermayer wrote: > I do not know that but i would be surprised if null dereferences tests > where unwelcome > > oss-fuzz will already report null derferences and OOM conditions, as > well as undefined behavior. So in some sense various points on the map >

Re: [FFmpeg-devel] [PATCH] lavc: reset codec on receiving packet after EOF in compat_decode

2017-11-26 Thread James Almer
On 11/21/2017 6:48 PM, Marton Balint wrote: > > > On Thu, 9 Nov 2017, James Cowgill wrote: > >> Hi, >> >> On 09/11/17 14:02, Hendrik Leppkes wrote: >>> On Thu, Nov 9, 2017 at 1:21 PM, James Cowgill >>> wrote: In commit 061a0c14bb57 ("decode: restructure the core

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Derek Buitenhuis
On 11/26/2017 1:35 PM, Carl Eugen Hoyos wrote: > Attached patch adds a missing dependency to libvmaf, I don't know if > other threads also work. This should also be filed as a bug against libvmaf, since its pkg-config file isn't complete, then. - Derek

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Derek Buitenhuis
On 11/26/2017 2:15 PM, Carl Eugen Hoyos wrote: > I believe I have explained before that I always only use > vanilla toolchains because that's the only thing users > typically have access to. It's possible the OS X version was outdated then, since the system clang has used libc++ as default since

Re: [FFmpeg-devel] [PATCH 2/2] vorbisenc: Check the return value of av_frame_clone

2017-11-26 Thread Derek Buitenhuis
On 11/24/2017 7:27 PM, Derek Buitenhuis wrote: > Prevents a segfault when alloc fails. > > Signed-off-by: Derek Buitenhuis > --- > libavcodec/vorbisenc.c | 6 +- > 1 file changed, 5 insertions(+), 1 deletion(-) If there are no objections, I'll push this today

Re: [FFmpeg-devel] [PATCH 3/3] udp: Actually fail when we're missing required options, like the "warning" says.

2017-11-26 Thread Derek Buitenhuis
On 11/22/2017 3:28 PM, Derek Buitenhuis wrote: > Signed-off-by: Derek Buitenhuis > --- > There was no reasoning in the commit that added this, so maybe someone on > the list has some insights. > --- > libavformat/udp.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-)

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Carl Eugen Hoyos
2017-11-26 14:53 GMT+01:00 Derek Buitenhuis : > On 11/26/2017 1:50 PM, Carl Eugen Hoyos wrote: >> Sorry, I don't understand how pkg-config is related to the missing >> dependency of our configure script: Please explain. > > Because pthreads is a dependency of libvmaf. >

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Nicolas George
Carl Eugen Hoyos (2017-11-26): > The following produces an ffmpeg binary that loads pthreads > at start-time: > $ ./configure --enable-libvmaf --disable-pthreads That is not a problem. The fact that it produces a binary with pthreads symbols in it is. Regards, -- Nicolas George

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Carl Eugen Hoyos
2017-11-26 15:11 GMT+01:00 Derek Buitenhuis : > On 11/26/2017 2:09 PM, Carl Eugen Hoyos wrote: >> As said before, it is non-trivial to find such a system >> (it worked fine here on osx when I tested last). > > OS X does not ship with libstdc++, IIRC. You must > have

Re: [FFmpeg-devel] [PATCH] lavc: reset codec on receiving packet after EOF in compat_decode

2017-11-26 Thread Marton Balint
On Sun, 26 Nov 2017, James Almer wrote: On 11/26/2017 12:19 PM, Nicolas George wrote: James Almer (2017-11-26): The old decode API is not scheduled for removal right now probably because 99% of decoders need to be ported. I think this statement contains some confusion that is harmful to

[FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Carl Eugen Hoyos
Hi! Attached patch adds a missing dependency to libvmaf, I don't know if other threads also work. Please comment, Carl Eugen From 4ee0fe8778c67a0f623b352a257e68485dd1d559 Mon Sep 17 00:00:00 2001 From: Carl Eugen Hoyos Date: Sun, 26 Nov 2017 14:29:27 +0100 Subject: [PATCH]

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Derek Buitenhuis
On 11/26/2017 1:50 PM, Carl Eugen Hoyos wrote: > Sorry, I don't understand how pkg-config is related to the missing > dependency of our configure script: Please explain. Because pthreads is a dependency of libvmaf. Looking at libvmaf, it does list pthreads as a dependency:

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Carl Eugen Hoyos
2017-11-26 14:59 GMT+01:00 Nicolas George : > Carl Eugen Hoyos (2017-11-26): >> The following produces an ffmpeg binary that loads pthreads >> at start-time: >> $ ./configure --enable-libvmaf --disable-pthreads > > That is not a problem. The fact that it produces a binary with >

Re: [FFmpeg-devel] [PATCH] lavc: reset codec on receiving packet after EOF in compat_decode

2017-11-26 Thread Nicolas George
James Almer (2017-11-26): > The old decode API is not scheduled for removal right now probably > because 99% of decoders need to be ported. I think this statement contains some confusion that is harmful to the discussion. There are two interfaces worth considering in this discussion: the

[FFmpeg-devel] [PATCH] lavf/mov: fix huge alloc in mov_read_ctts

2017-11-26 Thread John Stebbins
An invalid file may cause huge alloc. Delay expansion of ctts entries until the number of samples is known in mov_build_index. Found-by: zhao dongzhuo, AD-lab of Venustech --- libavformat/mov.c | 31 +++ 1 file changed, 27 insertions(+), 4 deletions(-) diff --git

[FFmpeg-devel] avcodec/huffyuvenc : try to call dsp with aligned data, and remove code duplication

2017-11-26 Thread Martin Vignali
Hello, in attach patchs 0001-avcodec-huffyuvenc-increase-scalar-loop-count and 0003-avcodec-huffyuvenc-sub_left_prediction_bgr32-call-ds like diff_bytes and diff_bytes16, have AVX2 version, increase the scalar loop to call the aligned version in most case

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Derek Buitenhuis
On 11/26/2017 2:02 PM, Carl Eugen Hoyos wrote: > The way I understand (Linux) dynamic libraries, one implies > the other. The problem is that libvmaf's .pc file put all of its deps in Libs instead of splitting them out into Libs.private, which is used only when static linking. Stuff like

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Carl Eugen Hoyos
2017-11-26 15:05 GMT+01:00 Nicolas George : > Carl Eugen Hoyos (2017-11-26): >> The way I understand (Linux) dynamic libraries, one implies >> the other. > > Yes, but not the other way around. > > If a library uses threads internally but provides an interface that can > be used

Re: [FFmpeg-devel] [PATCH 00/15] OpenCL infrastructure, filters

2017-11-26 Thread Mark Thompson
On 26/11/17 13:07, hydra3...@gmail.com wrote: > OK, the experimental -opencl_bench option has been removed by recent commits. > Thank you for the updates to OpenCL. > > It cross-compiles OK, however I'm not sure about the replacement > functionality.  So, 2 things. > > 1. may I enquire how one

Re: [FFmpeg-devel] [PATCH 00/15] OpenCL infrastructure, filters

2017-11-26 Thread hydra3333
-Original Message- From: hydra3...@gmail.com Sent: Sunday, November 26, 2017 11:37 PM To: ffmpeg-devel@ffmpeg.org Cc: hydra Subject: Re: [FFmpeg-devel] [PATCH 00/15] OpenCL infrastructure, filters OK, the experimental -opencl_bench option has been removed by recent commits. Thank

[FFmpeg-devel] [PATCH] avformat/matroskaenc: actually enforce the stream limit

2017-11-26 Thread James Almer
Prevents out of array accesses. Adressess ticket #6873 Signed-off-by: James Almer --- libavformat/matroskaenc.c | 7 +++ 1 file changed, 7 insertions(+) diff --git a/libavformat/matroskaenc.c b/libavformat/matroskaenc.c index dad6d6c93f..06126781f8 100644 ---

Re: [FFmpeg-devel] [PATCH 00/15] OpenCL infrastructure, filters

2017-11-26 Thread Mark Thompson
On 26/11/17 16:14, hydra3...@gmail.com wrote: > Most systems have a "clinfo" program which will list all the queryable > properties of the available devices. > >> From docs: > > -init_hw_device type[=name][:device[,key=value...]] >    Initialise a new hardware device of type type called name,

[FFmpeg-devel] fate/hap : add test for hap encoding

2017-11-26 Thread Martin Vignali
Hello, Patch in attach, add test for hap encoding (currently not cover) (patch 002) and move decoding tests to a separate file (patch 001) decoding can be test with make fate-hap SAMPLES=fate-suite/ and encoding can be test with make fate-hapenc SAMPLES=fate-suite/ Hap encoding need ffmpeg

Re: [FFmpeg-devel] [PATCH 0/1][TOOL][HACK] Allocation NULL check fuzzer

2017-11-26 Thread Carl Eugen Hoyos
2017-11-26 2:05 GMT+01:00 Derek Buitenhuis : > On 11/26/2017 12:14 AM, Carl Eugen Hoyos wrote: >> I am of course in favour of such checks but is there an allocator we support >> that actually returns NULL on oom? > > Anything that doesn't use overcommit. Windows is the

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Derek Buitenhuis
On 11/26/2017 2:05 PM, Nicolas George wrote: > If a library does not use threads but our calls to the library require > threads, then we must consider it a dependency. Netflix made their pkg-config file incorrectly, which causes this. It should be fixed there, but do we want to work around that

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Carl Eugen Hoyos
2017-11-26 15:06 GMT+01:00 Derek Buitenhuis : > On 11/26/2017 2:02 PM, Carl Eugen Hoyos wrote: >> The way I understand (Linux) dynamic libraries, one implies >> the other. > > The problem is that libvmaf's .pc file put all of its deps in > Libs instead of splitting them

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Derek Buitenhuis
On 11/26/2017 2:10 PM, Nicolas George wrote: > -> our code depends on pthreads for this filters, it must be expressed > in configure: Carl Eugen's patch is right, there is no need to bugreport > anything. His explanations later were wrong, but it may only be caused > by the confusion you brought.

Re: [FFmpeg-devel] [PATCH]configure: libvmaf depends on pthreads

2017-11-26 Thread Carl Eugen Hoyos
2017-11-26 15:10 GMT+01:00 Nicolas George : > $ grep -c pthread libavfilter/*vmaf* > libavfilter/vf_libvmaf.c:22 > libavfilter/vf_vmafmotion.c:0 > libavfilter/vmaf_motion.h:0 > > -> our code depends on pthreads for this filters, it must be expressed > in configure: Carl Eugen's

Re: [FFmpeg-devel] [PATCH] lavf/mov: fix huge alloc in mov_read_ctts

2017-11-26 Thread John Stebbins
On 11/25/2017 05:03 PM, James Almer wrote: > On 11/25/2017 10:00 PM, John Stebbins wrote: >> On 11/25/2017 04:03 PM, Carl Eugen Hoyos wrote: >>> 2017-11-25 21:11 GMT+01:00 John Stebbins : An invalid file may cause huge alloc. Delay expansion of ctts entries

Re: [FFmpeg-devel] [PATCH] lavf/mov: fix huge alloc in mov_read_ctts

2017-11-26 Thread Derek Buitenhuis
On 11/26/2017 3:10 PM, John Stebbins wrote: > Is there some git magic for this, or is this just something you add manually > to the bottom of the commit message? It's just manual. - Derek ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org

Re: [FFmpeg-devel] [PATCH] lavc: reset codec on receiving packet after EOF in compat_decode

2017-11-26 Thread James Almer
On 11/26/2017 12:19 PM, Nicolas George wrote: > James Almer (2017-11-26): >> The old decode API is not scheduled for removal right now probably >> because 99% of decoders need to be ported. > > I think this statement contains some confusion that is harmful to the > discussion. > > There are two

Re: [FFmpeg-devel] [PATCH 00/15] OpenCL infrastructure, filters

2017-11-26 Thread hydra3333
Most systems have a "clinfo" program which will list all the queryable properties of the available devices. From docs: -init_hw_device type[=name][:device[,key=value...]] Initialise a new hardware device of type type called name, using the given device parameters. If no name is specified

Re: [FFmpeg-devel] [PATCH] avfilter/vf_tile: add init_padding option

2017-11-26 Thread Nicolas George
Paul B Mahol (2017-11-24): > Signed-off-by: Paul B Mahol > --- > doc/filters.texi | 4 > libavfilter/vf_tile.c | 12 +++- > 2 files changed, 15 insertions(+), 1 deletion(-) > > diff --git a/doc/filters.texi b/doc/filters.texi > index 76929e4db5..11ce0482c2

Re: [FFmpeg-devel] [PATCH] lavc: reset codec on receiving packet after EOF in compat_decode

2017-11-26 Thread Nicolas George
Marton Balint (2017-11-26): > Okay, I am exagarating a bit, but unconditionally returning AVERROR(ENOSYS) > would be an even better incentive, no? :) For invalid uses of the API that can be easily avoided by the application (like not explicitly passing NULL to a function), a hard crash is even

Re: [FFmpeg-devel] [PATCH] Refactor Developer Docs, update dev list section (v2)

2017-11-26 Thread Jim DeLaHunt
On 2017-11-26 04:38, Paul B Mahol wrote: On 11/26/17, Nicolas George wrote: Paul B Mahol (2017-11-26): Your opinions are irrelevant. # Be friendly and respectful towards others and third parties. # Treat others the way you yourself want to be treated. Please stop trampling

[FFmpeg-devel] Accurately describing ffmpeg-cvslog list [was: Re: [PATCH] Refactor Developer Docs, update dev list section (v2)]

2017-11-26 Thread Jim DeLaHunt
On 2017-11-26 03:42, Carl Eugen Hoyos wrote: 2017-11-26 9:31 GMT+01:00 Jim DeLaHunt : [...] + @subheading Subscribe to the ffmpeg-cvslog mailing list. -It is important to do this as the diffs of all commits are sent there and -reviewed by all the other developers.

Re: [FFmpeg-devel] [PATCH v3 3/3] error_resilience: remove avpriv_atomic usage

2017-11-26 Thread Michael Niedermayer
On Sat, Nov 25, 2017 at 05:01:57PM +, Rostislav Pehlivanov wrote: > Signed-off-by: Rostislav Pehlivanov > --- > libavcodec/error_resilience.c | 20 ++-- > libavcodec/error_resilience.h | 3 ++- > 2 files changed, 12 insertions(+), 11 deletions(-) > >

[FFmpeg-devel] [PATCH] avcodec/error_resilience: Use atomic set on writing error_occurred per slice

2017-11-26 Thread Michael Niedermayer
This is more correct if multiple slices are handled in parallel Signed-off-by: Michael Niedermayer --- libavcodec/error_resilience.c | 6 +++--- libavcodec/error_resilience.h | 2 +- libavcodec/h263dec.c | 2 +- libavcodec/h264_slice.c | 4 ++--

Re: [FFmpeg-devel] [PATCH] avformat/matroskaenc: actually enforce the stream limit

2017-11-26 Thread Michael Niedermayer
On Sun, Nov 26, 2017 at 02:03:09PM -0300, James Almer wrote: > Prevents out of array accesses. Adressess ticket #6873 > > Signed-off-by: James Almer > --- > libavformat/matroskaenc.c | 7 +++ > 1 file changed, 7 insertions(+) LGTM thx [...] -- Michael GnuPG

Re: [FFmpeg-devel] [PATCH] avfilter: add normalize filter

2017-11-26 Thread Richard Ling
Thanks Paul. Thanks also to all reviewers for your comments! It's very helpful to have extra sets of eyes to find my bugs. Moritz is right, there is an unused #define, I will try to find time to patch. Or maybe Paul can remove it Regards R. ___

Re: [FFmpeg-devel] [PATCH] hls demuxer: add option to defer parsing of variants

2017-11-26 Thread Rainer Hochecker
2017-11-26 12:04 GMT+01:00 Steven Liu : > 2017-11-26 18:46 GMT+08:00 Rainer Hochecker : >> fixed mem leak poined out by Steven > Hi Rainer, > > I'm not sure that is memleak, but looks like memleak when reading > the code, i see the code always in

Re: [FFmpeg-devel] [PATCH] hls demuxer: add option to defer parsing of variants

2017-11-26 Thread Rainer Hochecker
Variants are presented as programs and can be loaded later by setting discard flags on the program. Currently Kodi chooses the program that best matches the desired bit rate. Rainer ___ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org

[FFmpeg-devel] avcodec/utvideodec : add x86 SIMD (SSSE3) for gradient prediction

2017-11-26 Thread Martin Vignali
Hello, Patch in attach add SIMD (SSSE 3) for gradient prediction and a checkasm test Checkasm result (width = 1024) (kaby lake, macos 10.12) add_gradient_pred_c: 1708.8 add_gradient_pred_ssse3: 533.0 Benchmark on a 3 min HD File in gradient (422) without SIMD : bench: utime=102.695s bench:

Re: [FFmpeg-devel] [PATCH 3/3] avformat: deprecate AVFMT_FLAG_AUTO_BSFr

2017-11-26 Thread Clément Bœsch
On Sun, Nov 26, 2017 at 05:51:04PM -0300, James Almer wrote: > The bitstream filters inserted by this option should not be optional. Will ffmpeg error out if it's built without the required bsf? (or is there a hard dep in the configure?) -- Clément B. signature.asc Description: PGP signature

Re: [FFmpeg-devel] [PATCH] Refactor Developer Docs, update dev list section (v2)

2017-11-26 Thread Paul B Mahol
On 11/26/17, Jim DeLaHunt wrote: > On 2017-11-26 04:38, Paul B Mahol wrote: >> On 11/26/17, Nicolas George wrote: >>> Paul B Mahol (2017-11-26): Your opinions are irrelevant. >>> # Be friendly and respectful towards others and third parties. >>> #

Re: [FFmpeg-devel] [PATCH] avcodec: Implement vp8 nvdec hwaccel

2017-11-26 Thread Mark Thompson
On 26/11/17 22:04, Philip Langdale wrote: > Signed-off-by: Philip Langdale > --- > Changelog | 2 +- > configure | 2 ++ > libavcodec/Makefile| 1 + > libavcodec/hwaccels.h | 1 + > libavcodec/nvdec.c | 1 + > libavcodec/nvdec_vp8.c | 97

[FFmpeg-devel] [PATCH 7/8] lavc/flacenc: add AVX2 version of the 32-bit LPC encoder

2017-11-26 Thread James Darnley
When compared to the SSE4.2 version runtime, is reduced by 1 to 26%. The function itself is around 2 times faster. --- libavcodec/x86/flac_dsp_gpl.asm | 56 +++-- libavcodec/x86/flacdsp_init.c | 5 +++- 2 files changed, 47 insertions(+), 14 deletions(-)

Re: [FFmpeg-devel] [PATCH 12/17] vaapi_decode: Ignore the profile when not useful

2017-11-26 Thread Mark Thompson
On 24/11/17 16:50, Philip Langdale wrote: > On Fri, 24 Nov 2017 00:51:29 + > Mark Thompson wrote: > >> Enables VP8 decoding - the decoder places the the bitstream version >> in the profile field, which we want to ignore. >> --- >> libavcodec/vaapi_decode.c | 3 ++- >> 1 file

[FFmpeg-devel] [PATCH 5/8] lavc/x86/flac_dsp_gpl: cosmetic whitespace alignment

2017-11-26 Thread James Darnley
--- libavcodec/x86/flac_dsp_gpl.asm | 40 1 file changed, 20 insertions(+), 20 deletions(-) diff --git a/libavcodec/x86/flac_dsp_gpl.asm b/libavcodec/x86/flac_dsp_gpl.asm index 4d212ed212..952fc8b86b 100644 --- a/libavcodec/x86/flac_dsp_gpl.asm +++

Re: [FFmpeg-devel] [PATCH 3/8] avcodec/flac: add SSE4.2 version of the 32-bit lpc encoder

2017-11-26 Thread Carl Eugen Hoyos
2017-11-26 23:51 GMT+01:00 James Darnley : > +if (EXTERNAL_SSE42(cpu_flags)) { > +if (CONFIG_GPL) > +c->lpc32_encode = ff_flac_enc_lpc_32_sse42; > +} Any objections over "if (CONFIG_GPL && EXTERNAL_..)"? Carl Eugen

[FFmpeg-devel] Policy on ffmpeg-devel list and contributions [was: Re: [PATCH] Refactor Developer Docs, update dev list section (v2)]

2017-11-26 Thread Jim DeLaHunt
On 2017-11-26 03:42, Carl Eugen Hoyos wrote: 2017-11-26 9:31 GMT+01:00 Jim DeLaHunt : -@subsection Documentation/Other +@section Documentation/Other +@subheading Subscribe to the ffmpeg-devel mailing list. +It is important to be subscribed to the Of course it is

[FFmpeg-devel] [PATCH] avcodec: Implement vp8 nvdec hwaccel

2017-11-26 Thread Philip Langdale
Signed-off-by: Philip Langdale --- Changelog | 2 +- configure | 2 ++ libavcodec/Makefile| 1 + libavcodec/hwaccels.h | 1 + libavcodec/nvdec.c | 1 + libavcodec/nvdec_vp8.c | 97 ++

Re: [FFmpeg-devel] [PATCH 5/8] lavc/x86/flac_dsp_gpl: cosmetic whitespace alignment

2017-11-26 Thread Rostislav Pehlivanov
On 26 November 2017 at 22:51, James Darnley wrote: > --- > libavcodec/x86/flac_dsp_gpl.asm | 40 -- > -- > 1 file changed, 20 insertions(+), 20 deletions(-) > > diff --git a/libavcodec/x86/flac_dsp_gpl.asm >

Re: [FFmpeg-devel] [PATCH] hls demuxer: add option to defer parsing of variants

2017-11-26 Thread Rainer Hochecker
>> +/* >> + * If this is a live stream and this playlist looks like it is one >> segment >> + * behind, try to sync it up so that every substream starts at the same >> + * time position (so e.g. avformat_find_stream_info() will see packets >> from >> + * all active streams

[FFmpeg-devel] [PATCH 3/3] avformat: deprecate AVFMT_FLAG_AUTO_BSF

2017-11-26 Thread James Almer
The bitstream filters inserted by this option should not be optional. They are needed to succesfully mux files in some cases, and to prevent muxing broken files in others. This is more in line with AVCodec.bsfs() Signed-off-by: James Almer --- libavformat/avformat.h | 4

[FFmpeg-devel] [PATCH 2/3] avformat/mux: stop delaying writing the header

2017-11-26 Thread James Almer
Every bitstream filter behaves as intended now, so there's no need to wait for the first packet of every stream. Signed-off-by: James Almer --- libavformat/avformat.h | 2 +- libavformat/internal.h | 6 - libavformat/mux.c | 52

[FFmpeg-devel] [PATCH 1/3] ffmpeg: use avformat_init_output to initialize output files

2017-11-26 Thread James Almer
Postpone writing the header until the first output packet is ready to be written. This makes sure any stream parameter change that could take place while processing an input frame will be taken into account when writing the output file header. Signed-off-by: James Almer ---

Re: [FFmpeg-devel] [PATCH 3/3] avformat: deprecate AVFMT_FLAG_AUTO_BSFr

2017-11-26 Thread James Almer
On 11/26/2017 6:16 PM, Clément Bœsch wrote: > On Sun, Nov 26, 2017 at 05:51:04PM -0300, James Almer wrote: >> The bitstream filters inserted by this option should not be optional. > > Will ffmpeg error out if it's built without the required bsf? (or is there > a hard dep in the configure?) Mmh,

[FFmpeg-devel] [PATCH 0/8] left-overs of an ancient patch set for the flac encoder

2017-11-26 Thread James Darnley
Three years ago I was writing some assembly to speed-up the flac encoder. I got part of the set committed at that time. Since then the encoder had a small overhaul and a major bugfix. That all meant this set needed a little work to bring it back on top of master. I did most of that work in

[FFmpeg-devel] [PATCH 4/8] avcodec/flac: partially unroll loop in flac_enc_lpc_32

2017-11-26 Thread James Darnley
Now does 6 samples per iteration, up from 2. From 1.6 to 2.1 times faster again. 2.5 to 3.9 times faster overall. Runtime is reduced by a further 4 to 17%. Reduced by 9 to 65% overall. Same conditions as previously. --- libavcodec/x86/flac_dsp_gpl.asm | 30 +- 1

[FFmpeg-devel] [PATCH 2/8] avcodec/flac: add AVX2 version of the 16-bit LPC encoder

2017-11-26 Thread James Darnley
When compared to the SSE4 version, runtime is reduced by 0.5 to 20%. After a bug fix log, long ago in e609cfd697 the 16-bit lpc encoder is used so little that the runtime reduction is no longer correct. The function itself is around 2 times faster. (As one might expect for doing twice as many

[FFmpeg-devel] [PATCH 3/8] avcodec/flac: add SSE4.2 version of the 32-bit lpc encoder

2017-11-26 Thread James Darnley
From 1.3 to 2.5 times faster. Runtime reduced by 4 to 58%. As with the 16-bit version the speed-up generally increases with compression_level. Also like the 16-bit version, it is not used with levels less than 3. After this bug fix in long, long ago in e609cfd697 this 32-bit lpc encoder is

[FFmpeg-devel] [PATCH 1/8] avcodec/flac: document limitations of the LPC encoder

2017-11-26 Thread James Darnley
State that the maximum value of order is 32. This limit is used in both C and x86 assebly code. --- libavcodec/flacdsp.h | 8 1 file changed, 8 insertions(+) diff --git a/libavcodec/flacdsp.h b/libavcodec/flacdsp.h index 7bb0dd0e9a..90fd3f04b5 100644 --- a/libavcodec/flacdsp.h +++

[FFmpeg-devel] [PATCH 8/8] checkasm: add tests for flacenc lpc coder

2017-11-26 Thread James Darnley
--- tests/checkasm/flacdsp.c | 72 1 file changed, 72 insertions(+) diff --git a/tests/checkasm/flacdsp.c b/tests/checkasm/flacdsp.c index dccb54d672..08e5e264ea 100644 --- a/tests/checkasm/flacdsp.c +++ b/tests/checkasm/flacdsp.c @@ -20,13 +20,16

[FFmpeg-devel] [PATCH 6/8] lavc/x86/flac_dsp_gpl: partially unroll 32-bit LPC encoder

2017-11-26 Thread James Darnley
Around 1.1 times faster and reduces runtime by up to 6%. --- libavcodec/x86/flac_dsp_gpl.asm | 91 - 1 file changed, 72 insertions(+), 19 deletions(-) diff --git a/libavcodec/x86/flac_dsp_gpl.asm b/libavcodec/x86/flac_dsp_gpl.asm index

Re: [FFmpeg-devel] [PATCH] tests/checkasm/float_dsp: Increase allowed difference for float_dsp.vector_dmul

2017-11-26 Thread Michael Niedermayer
On Sun, Nov 26, 2017 at 12:47:22AM +0100, Michael Niedermayer wrote: > On Sun, Nov 26, 2017 at 12:10:38AM +0100, Michael Niedermayer wrote: > > On Fri, Nov 24, 2017 at 11:37:36PM -0300, James Almer wrote: > > > On 10/29/2017 11:57 AM, Michael Niedermayer wrote: > > > > The choosen value is the

Re: [FFmpeg-devel] [PATCH 6/8] lavc/x86/flac_dsp_gpl: partially unroll 32-bit LPC encoder

2017-11-26 Thread Rostislav Pehlivanov
On 26 November 2017 at 22:51, James Darnley wrote: > Around 1.1 times faster and reduces runtime by up to 6%. > --- > libavcodec/x86/flac_dsp_gpl.asm | 91 ++ > ++- > 1 file changed, 72 insertions(+), 19 deletions(-) > > diff --git

Re: [FFmpeg-devel] [PATCH 1/8] avcodec/flac: document limitations of the LPC encoder

2017-11-26 Thread Rostislav Pehlivanov
On 26 November 2017 at 22:51, James Darnley wrote: > State that the maximum value of order is 32. This limit is used in both > C and x86 assebly code. > --- > libavcodec/flacdsp.h | 8 > 1 file changed, 8 insertions(+) > > diff --git a/libavcodec/flacdsp.h

Re: [FFmpeg-devel] [PATCH 2/8] avcodec/flac: add AVX2 version of the 16-bit LPC encoder

2017-11-26 Thread Rostislav Pehlivanov
On 26 November 2017 at 22:51, James Darnley wrote: > When compared to the SSE4 version, runtime is reduced by 0.5 to 20%. > After a bug fix log, long ago in e609cfd697 the 16-bit lpc encoder is > used so little that the runtime reduction is no longer correct. The >

Re: [FFmpeg-devel] [PATCH 4/8] avcodec/flac: partially unroll loop in flac_enc_lpc_32

2017-11-26 Thread Rostislav Pehlivanov
On 26 November 2017 at 22:51, James Darnley wrote: > Now does 6 samples per iteration, up from 2. > > From 1.6 to 2.1 times faster again. 2.5 to 3.9 times faster overall. > Runtime is reduced by a further 4 to 17%. Reduced by 9 to 65% overall. > > Same conditions as

Re: [FFmpeg-devel] [PATCH 3/8] avcodec/flac: add SSE4.2 version of the 32-bit lpc encoder

2017-11-26 Thread James Almer
On 11/26/2017 8:07 PM, Carl Eugen Hoyos wrote: > 2017-11-26 23:51 GMT+01:00 James Darnley : > >> +if (EXTERNAL_SSE42(cpu_flags)) { >> +if (CONFIG_GPL) >> +c->lpc32_encode = ff_flac_enc_lpc_32_sse42; >> +} > > Any objections over "if

Re: [FFmpeg-devel] [PATCH 7/8] lavc/flacenc: add AVX2 version of the 32-bit LPC encoder

2017-11-26 Thread James Darnley
On 2017-11-27 00:13, Rostislav Pehlivanov wrote: > On 26 November 2017 at 22:51, James Darnley wrote: >> @@ -123,7 +123,10 @@ RET >> %endmacro >> >> %macro PMINSQ 3 >> -pcmpgtq %3, %2, %1 >> +mova%3, %2 >> +; We cannot use the 3-operand format because

Re: [FFmpeg-devel] [PATCH 7/8] lavc/flacenc: add AVX2 version of the 32-bit LPC encoder

2017-11-26 Thread James Almer
On 11/26/2017 7:51 PM, James Darnley wrote: > When compared to the SSE4.2 version runtime, is reduced by 1 to 26%. The > function itself is around 2 times faster. > --- > libavcodec/x86/flac_dsp_gpl.asm | 56 > +++-- > libavcodec/x86/flacdsp_init.c | 5

Re: [FFmpeg-devel] [PATCH 6/8] lavc/x86/flac_dsp_gpl: partially unroll 32-bit LPC encoder

2017-11-26 Thread James Darnley
On 2017-11-27 00:17, Rostislav Pehlivanov wrote: > On 26 November 2017 at 22:51, James Darnley wrote: >> @@ -152,13 +152,13 @@ RET >> %macro FUNCTION_BODY_32 0 >> >> %if ARCH_X86_64 >> -cglobal flac_enc_lpc_32, 5, 7, 8, mmsize, res, smp, len, order, coefs >> +

Re: [FFmpeg-devel] [PATCH 7/8] lavc/flacenc: add AVX2 version of the 32-bit LPC encoder

2017-11-26 Thread James Almer
On 11/26/2017 8:13 PM, Rostislav Pehlivanov wrote: > On 26 November 2017 at 22:51, James Darnley wrote: > >> When compared to the SSE4.2 version runtime, is reduced by 1 to 26%. The >> function itself is around 2 times faster. >> --- >> libavcodec/x86/flac_dsp_gpl.asm

Re: [FFmpeg-devel] [PATCH] avcodec: Implement vp8 nvdec hwaccel

2017-11-26 Thread Philip Langdale
On Sun, 26 Nov 2017 22:35:58 + Mark Thompson wrote: > On 26/11/17 22:04, Philip Langdale wrote: > > Signed-off-by: Philip Langdale > > --- > > Changelog | 2 +- > > configure | 2 ++ > > libavcodec/Makefile| 1 + > >

Re: [FFmpeg-devel] [PATCH 1/2] avcodec/kgv1dec: Check that there is enough input for maximum RLE compression

2017-11-26 Thread Michael Niedermayer
On Wed, Nov 22, 2017 at 09:00:57PM +0100, Michael Niedermayer wrote: > Fixes: Timeout > Fixes: 4271/clusterfuzz-testcase-4676667768307712 > > Found-by: continuous fuzzing process > https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg > Signed-off-by: Michael Niedermayer

Re: [FFmpeg-devel] [PATCH] libavformat/mov: Replace duplicate stream_nb check by assert

2017-11-26 Thread Michael Niedermayer
On Wed, Nov 22, 2017 at 08:19:45PM +, Derek Buitenhuis wrote: > On 11/22/2017 8:09 PM, Michael Niedermayer wrote: > > not much, no > > its a non static function tough > > i can remove the check completely if thats preferred ? > > I guess leave it since it's non-static. > > LGTM. ok, will

Re: [FFmpeg-devel] [PATCH] avcodec/h264idct_template: Fix integer overflow in ff_h264_idct8_add

2017-11-26 Thread Michael Niedermayer
On Mon, Nov 20, 2017 at 02:58:15PM +0100, Michael Niedermayer wrote: > Fixes: signed integer overflow: 452986184 - -2113885312 cannot be represented > in type 'int' > Fixes: 4196/clusterfuzz-testcase-minimized-5580648594014208 > > Found-by: continuous fuzzing process >

Re: [FFmpeg-devel] [PATCH] avcodec/mlpdsp: Fix signed integer overflow, 2nd try

2017-11-26 Thread Michael Niedermayer
On Mon, Nov 20, 2017 at 09:26:48PM +0100, Michael Niedermayer wrote: > The outputted bits should match what is used in the lossless check > > Fixes: runtime error: signed integer overflow: -538697856 * 256 cannot be > represented in type 'int' > Fixes:

Re: [FFmpeg-devel] [PATCH 3/3] tests/fate-run: Use -bitexact

2017-11-26 Thread Michael Niedermayer
On Sun, Oct 22, 2017 at 01:41:58AM +0200, Michael Niedermayer wrote: > Signed-off-by: Michael Niedermayer > --- > tests/fate-run.sh | 24 > 1 file changed, 12 insertions(+), 12 deletions(-) will apply [...] -- Michael GnuPG fingerprint:

Re: [FFmpeg-devel] Added HW H.264 and HEVC encoding for AMD GPUs based on AMF SDK

2017-11-26 Thread Mark Thompson
On 22/11/17 23:28, mmironov wrote: > From c669277afd764903d3da09d92a263d0fb58e24b1 Mon Sep 17 00:00:00 2001 > From: mmironov > Date: Tue, 14 Nov 2017 17:54:24 -0500 > Subject: [PATCH] Added HW H.264 and HEVC encoding for AMD GPUs based on AMF > SDK > > Signed-off-by:

[FFmpeg-devel] avcodec/samidec: check av_strdup() return value

2017-11-26 Thread Pan Bian
From: Pan Bian In function sami_paragraph_to_ass(), the return value of av_strdup() is not checked. To avoid potential NULL dereference, the return value should be checked against NULL. Signed-off-by: Pan Bian --- libavcodec/samidec.c | 3 +++ 1 file changed,

[FFmpeg-devel] ffmpeg: Check read_ffserver_streams() return value

2017-11-26 Thread Pan Bian
The function avformat_alloc_context() will return a NULL pointer on failure. However, in function read_ffserver_streams(), its return value is not validated and the subsequent dereference may result in a bad memory access bug. Check its return value against NULL and avoid potential NULL

Re: [FFmpeg-devel] ffmpeg: Check read_ffserver_streams() return value

2017-11-26 Thread Rostislav Pehlivanov
On 27 November 2017 at 01:12, Pan Bian wrote: > The function avformat_alloc_context() will return a NULL pointer on > failure. However, in function read_ffserver_streams(), its return value > is not validated and the subsequent dereference may result in a bad > memory access

[FFmpeg-devel] libavformat/rtpenc_mpegts: check avformat_new_stream() return value

2017-11-26 Thread Pan Bian
The function avformat_new_stream() returns a NULL pointer on failure. However, in function rtp_mpegts_write_header(), its return value is not validated before it is dereferenced. Check the return value against NULL to avoid potential NULL dereference. Signed-off-by: Pan Bian

[FFmpeg-devel] avcodec/nvenc: set correct error code

2017-11-26 Thread Pan Bian
In function process_output_surface(), the return value is 0 on the path that av_mallocz() returns a NULL pointer. 0 indicates success, which deviates from the fact. Return "AVERROR(ENOMEM)" instead of "0". Signed-off-by: Pan Bian --- libavcodec/nvenc.c | 4 +++- 1 file

  1   2   >