Quoting James Almer (2024-03-23 03:53:26)
> On 3/22/2024 5:28 PM, Anton Khirnov wrote:
> > lavfi does not require aligned buffers, so we can safely apply top/left
> > cropping by any amount, without passing any special flags to lavc.
> > Longer term, an even better solution would probably be auto-i
Quoting James Almer (2024-03-23 03:33:13)
> Why does this test change? And is it just the order or side data in the
> output frame?
Yes, just the order. And the test is being removed, so it doesn't matter
either way.
--
Anton Khirnov
___
ffmpeg-devel
On date Thursday 2024-03-21 10:46:30 +0800, wenbin.chen-at-intel@ffmpeg.org
wrote:
> From: Wenbin Chen
>
> Signed-off-by: Wenbin Chen
> ---
> Changelog | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/Changelog b/Changelog
> index e3ca52430c..4af55ff537 100644
> --- a/Changelog
>
On Tue, Mar 19, 2024 at 06:49:43PM +0100, Matthieu Bouron wrote:
> On Sun, Mar 17, 2024 at 11:28:31PM +0100, Matthieu Bouron wrote:
> > Diff with previous iteration:
> > - rebaed on latest master
> > - applied feedback from Zhao (removed HAVE_SETMODE block, use a typedef
> > struct,
> > check th
On date Wednesday 2024-03-20 17:47:31 +0100, Andreas Rheinhardt wrote:
> Stefano Sabatini:
> > ---
> > doc/muxers.texi | 10 ++
> > 1 file changed, 10 insertions(+)
> >
> > diff --git a/doc/muxers.texi b/doc/muxers.texi
> > index d9dd72e613..25d1a5907e 100644
> > --- a/doc/muxers.texi
> >
av_ts_make_time_string() used "%.6g" format, but this format was losing
precision even when the timestamp to be printed was not that large. For example
for 3 hours (10800) seconds, only 1 decimal digit was printed, which made this
format inaccurate when it was used in e.g. the silencedetect filter.
By calling the av_ts_make_time_string2() from the function we can fix the
precision issue.
Signed-off-by: Marton Balint
---
libavutil/timestamp.h| 4 +---
tests/ref/fate/filter-metadata-scdet | 12 ++--
tests/ref/fate/filter-metadata-silencedetect | 2 +-
On Mon, 18 Mar 2024, Marton Balint wrote:
On Mon, 18 Mar 2024, Andreas Rheinhardt wrote:
Marton Balint:
Previously we always assumed that the channels are in native order, even
if
they were not. The new channel layout API allows us to signal the proper
channel order, so let's do so.
On Tue, 19 Mar 2024, Marton Balint wrote:
FL channel ID is 0, so for an unset value we must check for ID < 0.
Regression since 1f96db959c1235bb7079d354e09914a0a2608f62.
Will apply the series soon.
Regards,
Marton
Signed-off-by: Marton Balint
---
libavfilter/af_channelmap.c | 4 ++--
1 f
Andreas Rheinhardt:
> C11 provides static assertions via _Static_assert and
> provides static_assert as a convenience define for this
> in assert.h. Our codebase uses the latter, as _Static_assert
> has actually already been deprecated in C23.
>
> Not all toolchains that declare support for C11 ac
On Fri, 22 Mar 2024 22:02:39 +0100 Damiano Galassi wrote:
> There two new fields were never sent down the filter chain, and no filter
> after the first had colorspace and color_range set, causing breakage in
> zscale and possible other filters.
> ---
> libavfilter/avfilter.c | 4
> libavf
Signed-off-by: Andreas Rheinhardt
---
libavcodec/mpegutils.c | 21 +
1 file changed, 5 insertions(+), 16 deletions(-)
diff --git a/libavcodec/mpegutils.c b/libavcodec/mpegutils.c
index fc3e270631..7f499b3d0f 100644
--- a/libavcodec/mpegutils.c
+++ b/libavcodec/mpegutils.c
@@
Signed-off-by: Andreas Rheinhardt
---
libavcodec/mpegutils.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/mpegutils.c b/libavcodec/mpegutils.c
index 7f499b3d0f..d94e8f422f 100644
--- a/libavcodec/mpegutils.c
+++ b/libavcodec/mpegutils.c
@@ -265,7 +265,7 @@ void f
(The discrepancy between the definition and the declaration
in protocols.c is actually UB.)
Signed-off-by: Andreas Rheinhardt
---
This and the next patch have not been tested (apart from "it compiles
when the android content protocol is disabled").
libavformat/file.c | 2 +-
1 file changed, 1 i
Besides improving readability this also ensures that
a developer who has the android content protocol enabled
and works on the other parts of the file will not
forget to add necessary inclusions just because of
(indirect) inclusions from the files included only
when said protocol is enabled.
Signe
Please add coverage for regression cases in FATE.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe"
On Sat, Mar 23, 2024 at 1:11 PM Niklas Haas wrote:
> On Fri, 22 Mar 2024 22:02:39 +0100 Damiano Galassi
> wrote:
> > There two new fields were never sent down the filter chain, and no
> filter after the first had colorspace and color_range set, causing breakage
> in zscale and possible other fil
There is no need to free the already-added items, they will be freed
alongside the codec context. There is also little point in an error
message, as the only reason this can fail is malloc failure.
---
fftools/ffmpeg_enc.c | 9 +
1 file changed, 1 insertion(+), 8 deletions(-)
diff --git a
Nothing about this code is video-specific.
---
fftools/ffmpeg_enc.c | 41 +
1 file changed, 21 insertions(+), 20 deletions(-)
diff --git a/fftools/ffmpeg_enc.c b/fftools/ffmpeg_enc.c
index 6a91fd0398..61310c67f1 100644
--- a/fftools/ffmpeg_enc.c
+++ b/fftoo
They allow exporting extended information about side data types.
---
doc/APIchanges | 4 +++
libavutil/frame.c | 70 ++---
libavutil/frame.h | 27 +
libavutil/version.h | 2 +-
4 files changed, 72 insertions(+), 31 deletions(-)
di
Skip those side data types that do not make sense as global side data.
---
fftools/ffmpeg_enc.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/fftools/ffmpeg_enc.c b/fftools/ffmpeg_enc.c
index f01be1c22f..6a91fd0398 100644
--- a/fftools/ffmpeg_enc.c
+++ b/fftools/ffmpeg_enc.c
@@ -247,6 +
Anton Khirnov:
> Skip those side data types that do not make sense as global side data.
> ---
> fftools/ffmpeg_enc.c | 5 +
> 1 file changed, 5 insertions(+)
>
> diff --git a/fftools/ffmpeg_enc.c b/fftools/ffmpeg_enc.c
> index f01be1c22f..6a91fd0398 100644
> --- a/fftools/ffmpeg_enc.c
> +++ b
Quoting Andreas Rheinhardt (2024-03-23 14:15:06)
> Anton Khirnov:
> > Skip those side data types that do not make sense as global side data.
> > ---
> > fftools/ffmpeg_enc.c | 5 +
> > 1 file changed, 5 insertions(+)
> >
> > diff --git a/fftools/ffmpeg_enc.c b/fftools/ffmpeg_enc.c
> > index f
Anton Khirnov:
> Quoting Andreas Rheinhardt (2024-03-23 14:15:06)
>> Anton Khirnov:
>>> Skip those side data types that do not make sense as global side data.
>>> ---
>>> fftools/ffmpeg_enc.c | 5 +
>>> 1 file changed, 5 insertions(+)
>>>
>>> diff --git a/fftools/ffmpeg_enc.c b/fftools/ffmpeg_
It is highly unsafe, as AVCodecContext contains many allocated fields.
Everything needed by worked threads should be covered by
* routing through AVCodecParameters
* av_opt_copy()
* copying quantisation matrices manually
avcodec_free_context() can now be used for per-thread contexts.
---
libavcod
Anton Khirnov:
> It is highly unsafe, as AVCodecContext contains many allocated fields.
> Everything needed by worked threads should be covered by
> * routing through AVCodecParameters
> * av_opt_copy()
> * copying quantisation matrices manually
>
> avcodec_free_context() can now be used for per-t
Hi,
According to your advices, I made some changes of mine last patch. I feel like
this one way more better. I removed trivial calculations but I want to say I am
not proud of how I handled histogram in float despite 16 bit integer histogram.
I am storing dB values instead of storing samples. I
C11 required to use ATOMIC_VAR_INIT to statically initialize
atomic objects with static storage duration. Yet this macro
was unsuitable for initializing structures [1] and was actually
unneeded for all known implementations (this includes our
compatibility fallback implementations which simply wrap
On Fri, Mar 22, 2024 at 6:36 PM Jun Zhao wrote:
> From: Jun Zhao
>
> small cleanup for style, redundant semicolons, goto labels,
> in FFmpeg, we put goto labels at brace level.
>
Pushed this patch.
Thanks, Jun
> Signed-off-by: Jun Zhao
> ---
> libavcodec/vvc_parser.c | 8
> 1 file c
On Thu, Mar 21, 2024 at 9:15 AM Michael Niedermayer
wrote:
> Fixes: out of array access
> Fixes:
> 67021/clusterfuzz-testcase-minimized-ffmpeg_DEMUXER_fuzzer-4883576579489792
>
> Found-by: continuous fuzzing process
> https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
> Signed-off-by
On Sat, Mar 23, 2024 at 3:28 PM Yiğithan Yiğit
wrote:
> Hi,
>
> According to your advices, I made some changes of mine last patch. I feel
> like this one way more better. I removed trivial calculations but I want to
> say I am not proud of how I handled histogram in float despite 16 bit
> integer
Fixes ffplay playback of unknown layouts, when SDL directly supports the audio
format, such as:
ffplay -f lavfi anullsrc=cl=2C,aformat=s16
Without the patch, "Channel layout change is not supported" errors are
generated because buffersrc (unknown 2 channel) and buffersink (stereo)
negotiated a st
Following situation:
Someone wants to stack two yuv420p frames of the size 3x3 pixel each.
Now the various AV_CEIL_RSHIFT() calculations calculate a height of 2
pixel for each items subsampled planes.
However, for example in case of a vstack, the output frames will have a
height of 6, so the subsam
Fixes #9613
---
libavfilter/af_volumedetect.c | 234 +-
1 file changed, 172 insertions(+), 62 deletions(-)
diff --git a/libavfilter/af_volumedetect.c b/libavfilter/af_volumedetect.c
index 8b001d1cf2..d13d043f88 100644
--- a/libavfilter/af_volumedetect.c
+++ b/lib
From: Niklas Haas
The Dolby Vision RPU contains a CRC32 to validate the payload against.
The implementation is CRC32/MPEG-2.
The CRC is only verified with the AV_EF_CRCCHECK flag.
Co-authored-by: quietvoid
---
libavcodec/av1dec.c | 3 ++-
libavcodec/dovi_rpu.c | 18 --
liba
From: Niklas Haas
This ensures that `gb` in the following section is fully byte-aligned,
points at the start of the actual RPU, and ends on the CRC terminator.
This is important for both calculation of the CRC, as well as dovi
extension block parsing (which aligns to byte boundaries in various
p
From: Niklas Haas
As well as accessors plus a function for allocating this struct with
extension blocks,
Definitions generously taken from quietvoid/dovi_tool, which is
assembled as a collection of various patent fragments, as well as output
by the official Dolby Vision bitstream verifier tool.
From: quietvoid
The NLQ pivots are not documented but should be present in the header
for profile 7 RPU format. It has been verified using Dolby's
verification toolkit.
Signed-off-by: quietvoid
Signed-off-by: Niklas Haas
---
doc/APIchanges| 3 +++
libavcodec/dovi_rpu.c | 9 -
From: Niklas Haas
---
libavcodec/dovi_rpu.c | 1 +
libavcodec/dovi_rpu.h | 7 +++
2 files changed, 8 insertions(+)
diff --git a/libavcodec/dovi_rpu.c b/libavcodec/dovi_rpu.c
index cd54c8716dc..21cb1850e3e 100644
--- a/libavcodec/dovi_rpu.c
+++ b/libavcodec/dovi_rpu.c
@@ -48,6 +48,7 @@ void
From: Niklas Haas
We split the inner loop between v1 and v2 extension blocks to print
a warning where an extension block was encountered in an unexpected
context.
Co-authored-by: quietvoid
---
libavcodec/dovi_rpu.c | 178 ++
1 file changed, 178 insertion
On Tue, 19 Mar 2024 20:16:39 +0100 Niklas Haas wrote:
> From: Niklas Haas
>
> AV1 streams don't use configuration records, so delete them when
> encoding to AV1. Ideally this would be, as the comment suggests, handled
> at the frame-level (and stripped by the av1 encoder), but given the
> status
On Wed, 20 Mar 2024 20:55:26 +0100 Michael Niedermayer
wrote:
> On Wed, Mar 20, 2024 at 03:23:53PM +0100, Niklas Haas wrote:
> > On Tue, 19 Mar 2024 22:55:56 +0100 Michael Niedermayer
> > wrote:
> > > is it neccessary to drop compatibility to the old syntax ?
> >
> > Only if we want to use fra
On Fri, 09 Feb 2024 13:11:38 +0100 Niklas Haas wrote:
> On Mon, 05 Feb 2024 19:04:30 +0100 Andreas Rheinhardt
> wrote:
> > This presumes the relevant states to be a cartesian product. Which need
> > not be true. A callback would be better; this would also allow to base
> > the list on other valu
On Sat, 23 Mar 2024 19:02:26 +0100 Andreas Rheinhardt
wrote:
> Niklas Haas:
> > From: Niklas Haas
> >
> > This ensures that `gb` in the following section is fully byte-aligned,
> > points at the start of the actual RPU, and ends on the CRC terminator.
> >
> > This is important for both calcula
On Sat, 23 Mar 2024 18:58:51 +0100 Andreas Rheinhardt
wrote:
> Niklas Haas:
> > From: Niklas Haas
> >
> > As well as accessors plus a function for allocating this struct with
> > extension blocks,
> >
> > Definitions generously taken from quietvoid/dovi_tool, which is
> > assembled as a collec
On Sat, 23 Mar 2024 19:00:59 +0100 Andreas Rheinhardt
wrote:
> Niklas Haas:
> > From: Niklas Haas
> >
> > We split the inner loop between v1 and v2 extension blocks to print
> > a warning where an extension block was encountered in an unexpected
> > context.
> >
> > Co-authored-by: quietvoid
On Sat, 23 Mar 2024 19:08:04 +0100 Niklas Haas wrote:
> On Sat, 23 Mar 2024 18:58:51 +0100 Andreas Rheinhardt
> wrote:
> > Niklas Haas:
> > > From: Niklas Haas
> > >
> > > As well as accessors plus a function for allocating this struct with
> > > extension blocks,
> > >
> > > Definitions gene
From: 联盟少侠
The errors indicate that there are multiple definitions of several JNI (Java
Native Interface) symbols in the FFmpeg library. Specifically, the linker
ld.lld has found duplicate definitions for the following symbols:
- ff_jni_get_env
- ff_jni_jstring_to_utf_chars
- ff_jni_utf_chars_
Changes since v1:
- Rebased onto master
- Dropped av_dovi_metadata_alloc_ext(), instead statically allocate
space for 32 extension blocks (AV_DOVI_MAX_EXT_BLOCKS)
- Removed misleading comment on nlq_pivots
- Fix memory leak on ff_dovi_ctx_flush()
- Switch to using ff_refstruct to properly manage
From: quietvoid
The NLQ pivots are not documented but should be present in the header
for profile 7 RPU format. It has been verified using Dolby's
verification toolkit.
Signed-off-by: quietvoid
Signed-off-by: Niklas Haas
---
doc/APIchanges| 3 +++
libavcodec/dovi_rpu.c | 9 -
From: Niklas Haas
As well as accessors plus a function for allocating this struct with
extension blocks,
Definitions generously taken from quietvoid/dovi_tool, which is
assembled as a collection of various patent fragments, as well as output
by the official Dolby Vision bitstream verifier tool.
From: Niklas Haas
Instead of AVERROR(EINVAL)
---
libavcodec/dovi_rpu.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/dovi_rpu.c b/libavcodec/dovi_rpu.c
index c84a942f476..53f8c288db0 100644
--- a/libavcodec/dovi_rpu.c
+++ b/libavcodec/dovi_rpu.c
@@ -483,5 +483,5
From: Niklas Haas
This ensures that `gb` in the following section is fully byte-aligned,
points at the start of the actual RPU, and ends on the CRC terminator.
This is important for both calculation of the CRC, as well as dovi
extension block parsing (which aligns to byte boundaries in various
p
From: Niklas Haas
---
libavcodec/dovi_rpu.c | 3 +++
libavcodec/dovi_rpu.h | 6 ++
2 files changed, 9 insertions(+)
diff --git a/libavcodec/dovi_rpu.c b/libavcodec/dovi_rpu.c
index d584d99590a..6097bd5422f 100644
--- a/libavcodec/dovi_rpu.c
+++ b/libavcodec/dovi_rpu.c
@@ -47,6 +47,7 @@ void
From: Niklas Haas
The Dolby Vision RPU contains a CRC32 to validate the payload against.
The implementation is CRC32/MPEG-2.
The CRC is only verified with the AV_EF_CRCCHECK flag.
Co-authored-by: quietvoid
---
libavcodec/av1dec.c | 3 ++-
libavcodec/dovi_rpu.c | 18 --
liba
From: Niklas Haas
We split the inner loop between v1 and v2 extension blocks to print
a warning where an extension block was encountered in an unexpected
context.
Co-authored-by: quietvoid
---
libavcodec/dovi_rpu.c | 178 ++
1 file changed, 178 insertion
From: Niklas Haas
---
libavcodec/dovi_rpu.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/libavcodec/dovi_rpu.c b/libavcodec/dovi_rpu.c
index b3defd87bda..2b4d2b470c2 100644
--- a/libavcodec/dovi_rpu.c
+++ b/libavcodec/dovi_rpu.c
@@ -94,7 +94,7 @@ int ff_dovi_attach_sid
On Fri, Mar 22, 2024 at 05:01:00PM +, Antoine Soulier via ffmpeg-devel
wrote:
> The LC3 audio codec is the default codec of Bluetooth LE audio.
> This is a wrapper over the liblc3 library (https://github.com/google/liblc3).
>
> Signed-off-by: Antoine Soulier
> Signed-off-by: Antoine SOULIER
A few minor tweaks to libavutil/film_grain_params.c.
Leo Izen (2):
avutil/film_grain_params: remove unused variables
avutil/film_grain_params: remove do loop in CHECK macro
libavutil/film_grain_params.c | 7 +--
1 file changed, 1 insertion(+), 6 deletions(-)
--
2.44.0
These variables are never read from, so they trigger -Wunused-variables
Signed-off-by: Leo Izen
---
libavutil/film_grain_params.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/libavutil/film_grain_params.c b/libavutil/film_grain_params.c
index fff7252f2f..b3fa37f527 100644
--- a/libavuti
The continue statement will break out of the do/while loop, not the
outer loop as intended. This is one (compound) statement anyway, so we
can remove the do/while entirely.
Signed-off-by: Leo Izen
---
libavutil/film_grain_params.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --
On Fri, Mar 22, 2024 at 09:28:37PM +0100, Anton Khirnov wrote:
> It is highly unsafe, as AVCodecContext contains many allocated fields.
> Copying information via AVCodecParameters and with av_opt_copy() should
> handle everything needed by thread workers.
> ---
> libavcodec/frame_thread_encoder.c
On 2024/3/24 03:05, ad...@shaoxia.xyz wrote:
From: 联盟少侠
Could you configure git to avoid Chinese character?
The errors indicate that there are multiple definitions of several JNI (Java
Native Interface) symbols in the FFmpeg library. Specifically, the linker
ld.lld has found duplicate defi
On Fri, Mar 22, 2024 at 01:52:55PM +0100, Anton Khirnov wrote:
> Michael,
> following up on the previous discussion in this thread - if you,
> personally, would like to vote for a different option than those
> suggested so far, please propose one.
The goal of the vote is to find the option which
Since log statements printing int64 were made portable in
4464b7eeb194e98ac115f18d2b6be67361407c8a, let us include
inttypes.h explicitly (as it is unclear where PRId64 and
such are coming from now).
Reported-by: Traneptora
Signed-off-by: Marth64
---
libavformat/dvdvideodec.c | 1 +
1 file change
Since v7/8:
* Simplified demuxer error handling logic
* Fixes buggy change to ff_subtitles_queue_insert()
* Documentation improvements
Signed-off-by: Marth64
--
2.34.1
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/li
If ff_subtitles_queue_insert() were given a NULL buffer
with 0 length, it would still attempt to grow the packet
or memcpy depending on if merge option is enabled.
In this commit, allow passing a NULL buffer with 0 length
without attempting to do such operations. This way, if a
subtitle demuxer ha
RCWT (Raw Captions With Time) is a format native to ccextractor,
a commonly used OSS tool for processing 608/708 Closed Captions (CC).
RCWT can be used to archive the original extracted CC bitstream.
The muxer was added in January 2024. In this commit, add the demuxer.
One can now demux RCWT files
The high level summary of RCWT can be delegated doc/muxers, which
makes it easier to maintain and more consistent with the documentation
of the demuxer.
Signed-off-by: Marth64
---
libavformat/rcwtenc.c | 5 -
1 file changed, 5 deletions(-)
diff --git a/libavformat/rcwtenc.c b/libavformat/rc
Signed-off-by: Marth64
---
doc/muxers.texi | 29 +++--
1 file changed, 19 insertions(+), 10 deletions(-)
diff --git a/doc/muxers.texi b/doc/muxers.texi
index a10a8e216f..4161401059 100644
--- a/doc/muxers.texi
+++ b/doc/muxers.texi
@@ -3038,19 +3038,18 @@ ogg files can be
Signed-off-by: Marth64
---
doc/indevs.texi | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/doc/indevs.texi b/doc/indevs.texi
index d1b2bacf8b..fc14737181 100644
--- a/doc/indevs.texi
+++ b/doc/indevs.texi
@@ -1069,9 +1069,9 @@ ffplay -f lavfi
"movie=test.avi[out0];amovie=
Since v3:
* Added trace logging of coordinates (requested in feedback)
* Improved calculation of coordinates by using NTSC base resolution
Signed-off-by: Marth64
--
2.34.1
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailma
Closed Captions decoder prints multiple \pos ASS tags per cue,
and preceding the \pos tag is a fixed \an7 that fixes rendering
position from the top left corner and enforces justification.
Printing multiple \an and \pos tags in this way is invalid behavior,
because only the first \pos tag in a cue
Closed Captions are an NTSC concept, but the base resolution
used for x/y calculation is the default PAL base in ASS encoder.
This causes subtle positioning issues especially in 16:9
content.
Use NTSC 240 TVL (360x240) as the base instead, and compensate
the font size of the style. Update tests ac
From: Zhao Zhili
---
libavformat/Makefile | 1 +
libavformat/ffjni.c | 1 +
libavformat/file.c | 3 +--
3 files changed, 3 insertions(+), 2 deletions(-)
create mode 100644 libavformat/ffjni.c
diff --git a/libavformat/Makefile b/libavformat/Makefile
index 44aa485029..e10ad436a8 100644
--- a/
On 2024/3/24 10:16, Zhao Zhili wrote:
diff --git a/libavformat/file.c b/libavformat/file.c
index dd5819c..fa13ae9 100644
--- a/libavformat/file.c
+++ b/libavformat/file.c
@@ -43,7 +43,7 @@
#if CONFIG_ANDROID_CONTENT_PROTOCOL
#include
#include "libavcodec/jni.h"
-#include "libavcodec/ffjn
76 matches
Mail list logo
