Re: [FFmpeg-devel] [PATCH] avformat/matroskaenc: Regression fix for invalid MKV headers

On Tue, Jan 03, 2017 at 11:49:07PM +, Soft Works wrote: > The following three commits created a regression by writing initially > invalid mkv headers: > > 650e17d88b63b5aca6e0a43483e89e64b0f7d2dd avformat/matroskaenc: write a > CRC32 element on Tags > 3bcadf822711720ff0f8d14db71ae47cdf97e652

Re: [FFmpeg-devel] [PATCH 2/2] imdct15: replace the FFT with a faster PFA FFT algorithm

On 4 January 2017 at 10:16, Rostislav Pehlivanov wrote: > > +ff_fft_init(>ptwo_fft, N - 1, 1); > > Forgot to check the return value here, changed locally to: if (ff_fft_init(>ptwo_fft, N - 1, 1) < 0); goto fail; ___

[FFmpeg-devel] [PATCH 4/4] dxva2: allow an empty array of ID3D11VideoDecoderOutputView

We can pick the correct slice index directly from the ID3D11VideoDecoderOutputView casted from data[3]. --- libavcodec/dxva2_internal.h | 4 ++-- libavcodec/version.h| 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/libavcodec/dxva2_internal.h

[FFmpeg-devel] [PATCH 3/4] dxva2: get the slice number directly from the surface in D3D11VA

No need to loop through the known surfaces, we'll use the requested surface anyway. The loop is only done for DXVA2. --- libavcodec/dxva2.c | 15 +++ 1 file changed, 7 insertions(+), 8 deletions(-) diff --git a/libavcodec/dxva2.c b/libavcodec/dxva2.c index 2ac3f3e..c782f1c 100644

[FFmpeg-devel] [PATCH 2/6] ffmpeg: dxva2: factorize some code that can be common with d3d11va

ffmpeg_dxva_template.c has to be included and misc. typedefs have to be set to use the proper DXVA2 structures. initguid.h is included in ffmpeg_dxva.h so any includes after that will also define GUIDs locally. --- ffmpeg_dxva2.c | 282 --

[FFmpeg-devel] [PATCH 5/6] ffmpeg: add ffmpeg_d3d11va

The code is similar to ffmpeg_dxva2. The decoded output needs to be copied into a staging texture that can be accessed by the CPU as the decoder texture can't be accessed by the CPU. --- Changelog| 1 + Makefile | 1 + configure| 14 ffmpeg.h | 2 +

[FFmpeg-devel] [PATCH 0/6] D3D11VA in ffmpeg

Update on this patchset with a lot of fixes that was discussed during the last weeks. The patches have been split more. The public structures in libavutil are more documented and require less things set by the caller. The GUIDs declared in the template are defined in a separate ffmpeg_guid.c

[FFmpeg-devel] [PATCH 4/6] ffmpeg_dxva2: remove unused initial values

--- ffmpeg_dxva2.c | 12 ++-- 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/ffmpeg_dxva2.c b/ffmpeg_dxva2.c index dd19fec..68e0365 100644 --- a/ffmpeg_dxva2.c +++ b/ffmpeg_dxva2.c @@ -125,8 +125,8 @@ static int dxva2_get_decoder_configuration(AVCodecContext *s, const GUID

[FFmpeg-devel] Maintainer responsibilities

Hello, I am the maintainer of the dejudder filter. I haven't done anything to it since it contributed it almost two years ago. I follow this list, never seeing anything specifically mentioning it. Is there something else I should be doing as the maintainer? -Nick

[FFmpeg-devel] [PATCH 2/4] dxva2: use a single macro to test if the DXVA context is valid

--- libavcodec/dxva2_h264.c | 4 +--- libavcodec/dxva2_hevc.c | 4 +--- libavcodec/dxva2_internal.h | 5 + libavcodec/dxva2_mpeg2.c| 4 +--- libavcodec/dxva2_vc1.c | 4 +--- libavcodec/dxva2_vp9.c | 4 +--- 6 files changed, 10 insertions(+), 15 deletions(-) diff --git

[FFmpeg-devel] [PATCH 0/4] DXVA2 cleaning and D3D11VA simplification

This patchset was already sent but various cleaning has been done and it has been rebased to the latest master. The main goal is to have patch 3/4 that allows more flexibility on the decoder pool in D3D11VA, which we need in VLC for better efficiency. Patch 4/4 is the one that enables this

[FFmpeg-devel] [PATCH 1/4] dxva2: make ff_dxva2_get_surface() static and rename it

--- libavcodec/dxva2.c | 8 libavcodec/dxva2_internal.h | 2 -- 2 files changed, 4 insertions(+), 6 deletions(-) diff --git a/libavcodec/dxva2.c b/libavcodec/dxva2.c index 04a9c11..2ac3f3e 100644 --- a/libavcodec/dxva2.c +++ b/libavcodec/dxva2.c @@ -29,7 +29,7 @@ #include

[FFmpeg-devel] [PATCH 6/6] ffmpeg: dxva: use the typedefs more to make comparison between dxva2 and d3d11va

--- ffmpeg_d3d11va.c | 9 + ffmpeg_dxva2.c | 19 ++- 2 files changed, 15 insertions(+), 13 deletions(-) diff --git a/ffmpeg_d3d11va.c b/ffmpeg_d3d11va.c index b8ca9d8..3fd12f2 100644 --- a/ffmpeg_d3d11va.c +++ b/ffmpeg_d3d11va.c @@ -58,8 +58,8 @@ typedef

[FFmpeg-devel] [PATCH 1/6] libavutil: add support for AV_HWDEVICE_TYPE_D3D11VA

--- doc/APIchanges | 3 + libavutil/Makefile | 3 + libavutil/hwcontext.c | 3 + libavutil/hwcontext.h | 1 + libavutil/hwcontext_d3d11va.c | 461 + libavutil/hwcontext_d3d11va.h | 70 +++

[FFmpeg-devel] [PATCH 3/6] ffmpeg: dxva2: move the DXVA GUID definitions outside of the template

--- Makefile | 2 +- ffmpeg_dxva2.c | 1 + ffmpeg_dxva_template.c | 7 ++- ffmpeg_guid.c | 20 4 files changed, 28 insertions(+), 2 deletions(-) create mode 100644 ffmpeg_guid.c diff --git a/Makefile b/Makefile index 8aa72fd..25e1d54

Re: [FFmpeg-devel] [PATCH 2/2] imdct15: replace the FFT with a faster PFA FFT algorithm

> +/* 15-point FFT exptab */ > +for (i = 0; i < 19; i++) { > +if (i < 15) { > +double theta = 2 * M_PI * i / 15; > +s->exptab[i].re = cos(theta); > +s->exptab[i].im = sin(theta); > +} else { /* Wrap around to simplify fft[1]5 */ > +

Re: [FFmpeg-devel] Maintainer responsibilities

On 1/4/2017 11:13 AM, Nicholas Robbins wrote: > Hello, > > I am the maintainer of the dejudder filter. I haven't done anything to it > since it contributed it almost two years ago. I follow this list, never > seeing anything specifically mentioning it. Is there something else I should > be

Re: [FFmpeg-devel] [PATCH 2/2] imdct15: replace the FFT with a faster PFA FFT algorithm

On 2017-01-04 13:17, Rostislav Pehlivanov wrote: > Forgot to check the return value here, changed locally to: > > if (ff_fft_init(>ptwo_fft, N - 1, 1) < 0); > goto fail; I hope you have not changed it to that, with that semicolon at the end of the line. signature.asc Description:

Re: [FFmpeg-devel] [PATCH 2/2] imdct15: replace the FFT with a faster PFA FFT algorithm

First off, many thanks. > +const int inv_1 = l_ptwo << ((4 - b_ptwo) & 3); > +const int inv_2 = 0xeeef & ((1U << b_ptwo) - 1); It would be nice to add a comment here that the expression for inv_1 is (2^b_ptwo)^-1 mod 15 and inv_2 is 15^-1 mod 2^b_ptwo. (A general PFA FFT would need

Re: [FFmpeg-devel] Maintainer responsibilities

On 1/4/17, Nicholas Robbins wrote: > Hello, > > I am the maintainer of the dejudder filter. I haven't done anything to it > since it contributed it almost two years ago. I follow this list, never > seeing anything specifically mentioning it. Is there something

Re: [FFmpeg-devel] [PATCHv3] add signature filter for MPEG7 video signature

On Wed, Jan 04, 2017 at 05:05:41PM +0100, Gerion Entrup wrote: > On Dienstag, 3. Januar 2017 16:58:32 CET Moritz Barsnick wrote: > > > > The English opposite of "fine" is "coarse", not "course". :) > > > Oops. > > > > You still have a few "courses". (The actual variables, not the types. I > >

Re: [FFmpeg-devel] [PATCH] doc/hlsenc: hls_segment_filename, use_localtime, , use_localtime_mkdir, hls_flags

On Mon, Jan 02, 2017 at 02:24:22PM +0100, Bodecs Bela wrote: > > > 2017.01.02. 13:54 keltezéssel, Moritz Barsnick írta: > >On Mon, Jan 02, 2017 at 12:49:50 +0100, Bodecs Bela wrote: > >>Subject: [FFmpeg-devel] [PATCH] doc/hlsenc: hls_segment_filename, > >>use_localtime, , use_localtime_mkdir,

[FFmpeg-devel] [PATCH] avformat/hlsenc: fix Explicit null dereferenced

CID: 1323076 Passing null pointer loc to avformat_new_stream, which dereferences it fix: because the vtt_oc maybe have not value, so fix it. Signed-off-by: Steven Liu --- libavformat/hlsenc.c | 12 +--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git

Re: [FFmpeg-devel] [PATCH 2/2] imdct15: replace the FFT with a faster PFA FFT algorithm

On 4 January 2017 at 14:14, Peter Barfuss wrote: > First off, many thanks. > > > +const int inv_1 = l_ptwo << ((4 - b_ptwo) & 3); > > +const int inv_2 = 0xeeef & ((1U << b_ptwo) - 1); > > It would be nice to add a comment here that the expression for inv_1 > is

Re: [FFmpeg-devel] [PATCH] avformat/hlsenc: bugfix in duplicate filename detection

2017-01-05 3:17 GMT+08:00 Bodecs Bela : > Dear All, > > in hlsenc.c at detecting duplicate segment filenames a wrong, unitialized > variable is used for testing. This patch fixes this > typo. Please apply my patch. > applied! > > Thank you in advance. > > Bela Bodecs > > >

Re: [FFmpeg-devel] [PATCH] doc/hlsenc: hls_segment_filename, use_localtime, , use_localtime_mkdir, hls_flags

On Thu, Jan 05, 2017 at 01:24:00AM +0100, Michael Niedermayer wrote: > On Mon, Jan 02, 2017 at 02:24:22PM +0100, Bodecs Bela wrote: > > > > > > 2017.01.02. 13:54 keltezéssel, Moritz Barsnick írta: > > >On Mon, Jan 02, 2017 at 12:49:50 +0100, Bodecs Bela wrote: > > >>Subject: [FFmpeg-devel]

[FFmpeg-devel] [PATCH] avformat/test/fifo_muxer: add check for FailingMuxerPacketData alloc

CID: 1396257 Signed-off-by: Steven Liu --- libavformat/tests/fifo_muxer.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/libavformat/tests/fifo_muxer.c b/libavformat/tests/fifo_muxer.c index dc62965..e20bd6e 100644 --- a/libavformat/tests/fifo_muxer.c +++

[FFmpeg-devel] [PATCH] avformat/hlsenc: fix Explicit null dereferenced in hlsenc

CID: 1398228 Passing null pointer dirname to strlen, which dereferences it. Signed-off-by: Steven Liu --- libavformat/hlsenc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/libavformat/hlsenc.c b/libavformat/hlsenc.c index 808a797..3c6490a 100644

Re: [FFmpeg-devel] [PATCH] libavutil/hwcontext_qsv: Fix bug that the QSV encoded frames'width and height are 32-aligned

On 04/01/17 09:49, Huang, Zhengxu wrote: > 在 2017/1/3 20:40, Mark Thompson 写道: >> On 03/01/17 06:35, Huang, Zhengxu wrote: >>> From 8b1bcc0634f6ce36acfbd2bfdd26690a6323d09c Mon Sep 17 00:00:00 2001 >>> From: Zhengxu >>> Date: Fri, 16 Dec 2016 11:10:34 +0800 >>>

Re: [FFmpeg-devel] [PATCH] avformat/hlsenc: fix memleak in hlsenc

2017.01.05. 1:20 keltezéssel, Steven Liu írta: fix CID: 1398364 Resource leak refine the code of the new options Signed-off-by: Steven Liu --- libavformat/hlsenc.c | 21 + 1 file changed, 13 insertions(+), 8 deletions(-) diff --git

Re: [FFmpeg-devel] [PATCHv3] add signature filter for MPEG7 video signature

On Dienstag, 3. Januar 2017 16:58:32 CET Moritz Barsnick wrote: > > > The English opposite of "fine" is "coarse", not "course". :) > > Oops. > > You still have a few "courses". (The actual variables, not the types. I > don't care *too* much, but might be better for consistency.) You're right.

Re: [FFmpeg-devel] [PATCH] lavfi/af_ebur128: update filter to use new ebur128 API

On Thu, Nov 17, 2016 at 5:04 PM, Kyle Swanson wrote: > Hi, > > Here's a couple of patches which update the ebur128 filter to use the > recently added ebur128 API. This updated filter allows fine-tuned > control over which EBU R128 parameters are measured, and provides > modest speed

Re: [FFmpeg-devel] [PATCH] avcodec/texturedspenc: Fix indexing in color distribution determination

On Wed, Jan 4, 2017 at 1:06 AM, Michael Niedermayer wrote: > On Tue, Jan 03, 2017 at 02:15:20PM +0100, Vittorio Giovara wrote: >> On Mon, Jan 2, 2017 at 2:00 AM, James Almer wrote: >> > On 1/1/2017 8:28 PM, Michael Niedermayer wrote: >> >> Fixes

[FFmpeg-devel] [PATCH] avformat/hlsenc: fix memleak in hlsenc

fix CID: 1398364 Resource leak refine the code of the new options Signed-off-by: Steven Liu --- libavformat/hlsenc.c | 21 + 1 file changed, 13 insertions(+), 8 deletions(-) diff --git a/libavformat/hlsenc.c b/libavformat/hlsenc.c index

Re: [FFmpeg-devel] [PATCH] libavutil/hwcontext_qsv: Command line using hwaccel 'QSV' doesn't work

在 2017/1/3 21:14, Mark Thompson 写道: On 03/01/17 07:13, Huang, Zhengxu wrote: From 687ce9c804b2618f021100235c46a33b48fa522c Mon Sep 17 00:00:00 2001 From: Zhengxu Date: Wed, 14 Dec 2016 11:55:31 +0800 Subject: [PATCH] libavutil/hwcontext_qsv: Command line using

Re: [FFmpeg-devel] [PATCH] libavutil/hwcontext_qsv: Fix bug that the QSV encoded frames'width and height are 32-aligned

Hi, If fixing this issue by adjusting the cropping parameters after copying the FrameInfo, maybe it will still have some problem. This fix will only modify the encoder's surface information but other modules' surface information(CropW and CropH of the FrameInfo) also needs to be modified.

Re: [FFmpeg-devel] [PATCH] avformat/matroskaenc: Regression fix for invalid MKV headers

==> Revised patch in plain text (the attachment from my previous reply wasn't recognized by patchwork) The following three commits created a regression by writing initially invalid mkv headers: 650e17d88b63b5aca6e0a43483e89e64b0f7d2dd avformat/matroskaenc: write a CRC32 element on Tags

[FFmpeg-devel] [PATCH] avformat/hlsenc: bugfix in duplicate filename detection

Dear All, in hlsenc.c at detecting duplicate segment filenames a wrong, unitialized variable is used for testing. This patch fixes this typo. Please apply my patch. Thank you in advance. Bela Bodecs >From 2459227d3c304b9e45ae52071e8bcea36e287dac Mon Sep 17 00:00:00 2001 From: Bela Bodecs

Re: [FFmpeg-devel] [PATCH]lavf/matroska: Fix the A_MPEG/L1 codec_id

On Tue, Jan 03, 2017 at 03:46:06AM +0100, Carl Eugen Hoyos wrote: > Hi! > > Attached patch fixes MP1 muxing in mkv. > > Please comment, Carl Eugen > matroska.c |2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > 9ed063b5f9851ad17c82eb4896d9731e59b04d26 >

Re: [FFmpeg-devel] [PATCH] avformat/matroskaenc: Regression fix for invalid MKV headers

> From: ffmpeg-devel on behalf of Michael > Niedermayer > > this patch breaks fate > make fate I apologize, I had run fate but for some reason I hadn't got an error. But now I ran again, found and fixed the error. Please find attached

[FFmpeg-devel] FATE Sub-CC Test

I submitted a patch that failed a FATE test a while back. The patch changed the time position of subtitles. I have finally been able to take a look at the test (fate-sub-cc), and I believe the test is incorrect. It is looking for the following cues: Dialogue:

[FFmpeg-devel] [PATCH 1/2] imdct15: remove the AArch64 assembly

Prep work for the next commit, which will add a new FFT algorithm which makes the iMDCT over 3x faster than it is currently (standalone, the FFT is with some framesizes over 10x faster). The new FFT algorithm uses the already thouroughly SIMD'd power of two FFT which already has SIMD for AArch64,

[FFmpeg-devel] [PATCH 2/2] imdct15: replace the FFT with a faster PFA FFT algorithm

This commit replaces the current inefficient non-power-of-two FFT with a much faster FFT based on the Prime Factor Algorithm. Although it is already much faster than the old algorithm without SIMD, the new algorithm makes use of the already very throughouly SIMD'd power of two FFT, which improves