On Wed, Jun 14, 2023 at 7:33 PM Nicolas George wrote:
> Paul B Mahol (12023-06-14):
> > To flag filters that can work with variable frame size changes
> > all the time in graph. So no rescalers are need to be inserted or
> > filtergraph reset.
> >
> > Once all such filters are flaged with such ca
On Sat, Jun 3, 2023 at 9:44 PM Michael Niedermayer
wrote:
> Fixes: out of array read
> Fixes: Ticket 10308
>
Did you forgot to apply this?
>
> Signed-off-by: Michael Niedermayer
> ---
> libavcodec/h263dec.c | 2 +-
> libavcodec/mpeg4videodec.c | 4 ++--
> 2 files changed, 3 insertions(
On Sat, Jun 17, 2023 at 4:54 AM Anton Khirnov wrote:
> Make sure we don't repeatedly try to decode the same packet, making no
> progress and possibly causing an infinite loop.
>
Doesn't all error paths, bellow goto label in function, needs this (reset
of ch_offset to 0 and unref of pkt) ?
> --
On Thu, Jun 15, 2023 at 12:34 PM Iskandar Safarov
wrote:
> Thanks for looking into this,
>
> The idea is to page-align the entire software-backed video pool allocation
> where video frames are being taken from – both for decoding and/or encoding
> (when done in software only).
>
> The default get
Please no more pseudo demuxers.
Use libavfilter API.
___
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 "unsubscr
Quoting Michael Niedermayer (2023-06-16 22:58:21)
> On Wed, Jun 14, 2023 at 06:49:03PM +0200, Anton Khirnov wrote:
> > This is only a preparatory step to a fully threaded architecture and
> > does not yet make decoding truly parallel - the main thread will
> > currently submit a packet and wait unt
Make sure we don't repeatedly try to decode the same packet, making no
progress and possibly causing an infinite loop.
---
libavcodec/binkaudio.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/libavcodec/binkaudio.c b/libavcodec/binkaudio.c
index f28ecba760..265f93a822 1
asivery:
> Hello, apologies for bothering you all with this, but if anyone could take a
> look at this patch it would be great. The only other container that supports
> ATRAC1 is AEA, and it's not that good of a format since it has no magic
> number, and therefore is easily detected incorrectly.
Hello, apologies for bothering you all with this, but if anyone could take a
look at this patch it would be great. The only other container that supports
ATRAC1 is AEA, and it's not that good of a format since it has no magic number,
and therefore is easily detected incorrectly.
Thank you in ad
---
configure | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/configure b/configure
index 0c77d8e1fe..653aafb1ea 100755
--- a/configure
+++ b/configure
@@ -7009,8 +7009,7 @@ enabled alsa && { check_pkg_config alsa alsa
"alsa/asoundlib.h" snd_pcm_htimesta
enabled libjack &
Signed-off-by: Michael Niedermayer
---
configure|5 +
doc/demuxers.texi| 71 ++
libavformat/Makefile |2 +
libavformat/allformats.c |2 +
libavformat/sdrdemux.c | 1710 ++
5 files changed, 1790 insertions(+)
create m
This patchset adds some basic SDR support to FFmpeg
ATM, this supports AM demodulation, i will add FM later
I have no immedeate plans to add DAB, DVB, ...
but these should be added too and would be very welcome
if someone else wants to work on them.
The only dependancy this has is libsoapy which
Signed-off-by: Michael Niedermayer
---
libavcodec/aacdec_template.c | 8
libavcodec/aactab.c | 4 ++--
libavcodec/ac3dec.c | 2 +-
libavcodec/ac3enc_fixed.c| 2 +-
libavcodec/ac3enc_float.c| 2 +-
libavcodec/dolby_e.c | 4 ++--
libavcodec/kbdwin.c
Signed-off-by: Michael Niedermayer
---
libavcodec/kbdwin.c | 22 ++
libavcodec/kbdwin.h | 8 +---
2 files changed, 19 insertions(+), 11 deletions(-)
diff --git a/libavcodec/kbdwin.c b/libavcodec/kbdwin.c
index 163f9d5251..82755874d4 100644
--- a/libavcodec/kbdwin.c
+++ b
On 6/16/23 15:00, Andreas Rheinhardt wrote:
Leo Izen:
This switches the jpegxl_collect_codestream_header function to use
avcodec/bytestream2, which better enforces barriers, and should avoid
overrunning buffers with jxlp boxes if the size is zero or if the size
is so small the box is invalid.
S
SDR needs this for switching between mono and stereo stations
Signed-off-by: Michael Niedermayer
---
libavcodec/pcm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/pcm.c b/libavcodec/pcm.c
index 23955ba2dd..467ecb4fe0 100644
--- a/libavcodec/pcm.c
+++ b/libavcode
Signed-off-by: Michael Niedermayer
---
libavutil/tx_template.c | 12
1 file changed, 12 insertions(+)
diff --git a/libavutil/tx_template.c b/libavutil/tx_template.c
index 983de75a47..c4ec9502e0 100644
--- a/libavutil/tx_template.c
+++ b/libavutil/tx_template.c
@@ -43,6 +43,10 @@
Devin Heitmueller:
> In order to properly process SCTE-35 packets, we need the original
> PTS value from the demux (i.e. not mangled by the application or
> reclocked for the output). This allows us to set the pts_adjustment
> field in an BSF on the output side.
>
> Introduce a new side data type
Devin Heitmueller:
> Because SCTE-35 messages are represented in TS streams as sections
> rather than PES packets, we cannot rely on ffmpeg's standard
> mechanisms to adjust PTS values if reclocking the stream.
> This filter will leverage the SCTE-35 pts_adjust field to
> compensate for any change
Changes were made between in the last two years to periodically
send PCR-only packets on all PIDs, but for cases where the stream
may send packets very infrequently (like SCTE-35), this results in
extra TR101290 errors because it fails the PCR interval test.
I am not quite sure what the "right" fi
Because SCTE-35 messages are represented in TS streams as sections
rather than PES packets, we cannot rely on ffmpeg's standard
mechanisms to adjust PTS values if reclocking the stream.
This filter will leverage the SCTE-35 pts_adjust field to
compensate for any change in the PTS values in the stre
Introduce the ability to pass through SCTE-35 packets when creating
MPEG transport streams. Note that this patch makes no effort to
change the PTS values in the SCTE-35 packets, and thus only works
when not reclocking the stream (i.e. using -copyts). A subsequent
patch includes a BSF to recompute
We need the original PTS value in order to do subsequent processing,
so set it as packet side data.
Signed-off-by: Devin Heitmueller
---
libavformat/mpegts.c | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/libavformat/mpegts.c b/libavformat/mpegts.c
index 385d78b..b8f
In order to properly process SCTE-35 packets, we need the original
PTS value from the demux (i.e. not mangled by the application or
reclocked for the output). This allows us to set the pts_adjustment
field in an BSF on the output side.
Introduce a new side data type to store the original PTS.
Si
Properly set up the MPEG-TS mux and recalculate the pts_adjust field
in SCTE_35 packets, such that a user can transparently pass through
SCTE-35 streams when both the input and output are MPEG-TS.
Devin Heitmueller (5):
avcodec: Add new side data type to contain original PTS value
mpegts: Stas
On Wed, Jun 14, 2023 at 06:49:03PM +0200, Anton Khirnov wrote:
> This is only a preparatory step to a fully threaded architecture and
> does not yet make decoding truly parallel - the main thread will
> currently submit a packet and wait until it has been fully processed by
> the decoding thread be
Leo Izen:
> This switches the jpegxl_collect_codestream_header function to use
> avcodec/bytestream2, which better enforces barriers, and should avoid
> overrunning buffers with jxlp boxes if the size is zero or if the size
> is so small the box is invalid.
>
> Signed-off-by: Leo Izen
> ---
> li
On 6/12/23 09:22, Leo Izen wrote:
This switches the jpegxl_collect_codestream_header function to use
avcodec/bytestream2, which better enforces barriers, and should avoid
overrunning buffers with jxlp boxes if the size is zero or if the size
is so small the box is invalid.
Signed-off-by: Leo Ize
On 6/15/2023 12:18 PM, James Almer wrote:
Signed-off-by: James Almer
---
libavformat/evcdec.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavformat/evcdec.c b/libavformat/evcdec.c
index 89eda0f53e..807406885a 100644
--- a/libavformat/evcdec.c
+++ b/libavformat/evcdec
On Thu, 1 Jun 2023 at 21:44, Kacper Michajłow wrote:
>
> Content-Type can include charset and boundary which is not a part of
> mime type and shouldn't be copied as such.
>
> Fixes HLS playback when the Content-Type includes additional fields.
>
> Signed-off-by: Kacper Michajłow
> ---
> libavfor
As we are introducing two new formats and supporting conversions
between them, and also with the existing 0rgb32/0bgr32 formats, we get
a combinatorial explosion of kernels. I introduced a few new macros to
keep the things mostly managable.
The conversions are all simple, following existing patter
Bumping with a forgotten Changelog entry
From 33045f32da50390e6d58c34e95b3d344a53b6968 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tomas=20H=C3=A4rdin?=
Date: Fri, 16 Jun 2023 15:19:24 +0200
Subject: [PATCH 3/3] Add Changelog entry
---
Changelog | 1 +
1 file changed, 1 insertion(+)
diff --git a/C
On Fri, 16 Jun 2023 14:09:56 +0200 Andreas Rheinhardt
wrote:
> Niklas Haas:
> > From: Niklas Haas
> >
> > For now, hard-coded to 1 element.
> > ---
> > libavfilter/vf_libplacebo.c | 18 +-
> > 1 file changed, 13 insertions(+), 5 deletions(-)
> >
> > diff --git a/libavfilter/vf
Signed-off-by: Tong Wu
---
libavutil/hwcontext_qsv.c | 12 ++--
1 file changed, 10 insertions(+), 2 deletions(-)
diff --git a/libavutil/hwcontext_qsv.c b/libavutil/hwcontext_qsv.c
index 931f905a51..4cbe8cc494 100644
--- a/libavutil/hwcontext_qsv.c
+++ b/libavutil/hwcontext_qsv.c
@@ -756,
Signed-off-by: Tong Wu
---
libavutil/hwcontext_qsv.c | 12
1 file changed, 8 insertions(+), 4 deletions(-)
diff --git a/libavutil/hwcontext_qsv.c b/libavutil/hwcontext_qsv.c
index 713dc90ffd..931f905a51 100644
--- a/libavutil/hwcontext_qsv.c
+++ b/libavutil/hwcontext_qsv.c
@@ -665,6
Jun 13, 2023, 19:18 by d...@lynne.ee:
> Jun 13, 2023, 18:56 by andreas.rheinha...@outlook.com:
>
>> Lynne:
>>
>>> +dst->opaque= av_buffer_ref(src->opaque);
>>> +if (!dst->opaque) {
>>> +ret = AVERROR(ENOMEM);
>>> +goto fail;
>>> +}
>>>
>>
>> According to the doxy, o
Hei,
> -原始邮件-发件人:"Rémi Denis-Courmont" 发送时间:2023-06-16
> 03:25:07 (星期五)收件人:ffmpeg-devel@ffmpeg.org抄送:"Shen Peiting"
> 主题:Re: [FFmpeg-devel] [PATCH 3/6]
> lavc/ac3dsp: RISC-V V ac3_sum_square_butterfly_int32
>
> Le torstaina 15. kesäkuuta 2023, 13.36.42 EEST Peiting Shen a écrit :
> >
From: Niklas Haas
It may be the case that we want to skip the single frame cache for some
inputs but not others.
---
libavfilter/vf_libplacebo.c | 7 ++-
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/libavfilter/vf_libplacebo.c b/libavfilter/vf_libplacebo.c
index a30b244843..
From: Niklas Haas
---
libavfilter/vf_libplacebo.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/libavfilter/vf_libplacebo.c b/libavfilter/vf_libplacebo.c
index 427a15dc47..a30b244843 100644
--- a/libavfilter/vf_libplacebo.c
+++ b/libavfilter/vf_libplacebo.c
@@ -1129,7 +11
From: Niklas Haas
When combining multiple inputs with different PTS and durations, in
input-timed mode, we emit one output frame for every input frame PTS,
from *any* input. So when combining a low FPS stream with a high FPS
stream, the output framerate would match the higher FPS, independent of
From: Niklas Haas
To control the number of inputs.
---
doc/filters.texi| 5 +
libavfilter/vf_libplacebo.c | 25 ++---
2 files changed, 19 insertions(+), 11 deletions(-)
diff --git a/doc/filters.texi b/doc/filters.texi
index 2ea4083c3e..41cf4976cb 100644
---
From: Niklas Haas
This commit still relies on a single input for PTS determination, to be
changed in the next commit.
---
libavfilter/vf_libplacebo.c | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/libavfilter/vf_libplacebo.c b/libavfilter/vf_libplacebo.c
index 3d8125
From: Niklas Haas
This field will effectively hold the most recent status set by any
input. Currently functionally equivalent to input->status, but will
change soon.
---
libavfilter/vf_libplacebo.c | 11 +--
1 file changed, 9 insertions(+), 2 deletions(-)
diff --git a/libavfilter/vf_lib
From: Niklas Haas
Subsequent inputs required frame blending to be enable, in order to not
overwrite the existing frame contents.
For output metadata, we implicitly copy the metadata of the *first*
available stream (falling back to the second stream if the first has
already reached EOF, and so on
From: Niklas Haas
Use the gcd of all input timebases to ensure PTS accuracy. For the
framerate, just pick the highest of all the inputs, under the assumption
that we will render frames with approximately this frequency. Of course,
this is not 100% accurate, in particular if the input frames are b
From: Niklas Haas
Fixes an oversight in the previous code which should have been >=, not >.
---
libavfilter/vf_libplacebo.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavfilter/vf_libplacebo.c b/libavfilter/vf_libplacebo.c
index b78391441a..9b1526f19e 100644
--- a/liba
From: Niklas Haas
Instead of hard-coding input 0, pass the per-input structure and use the
link contained inside it.
---
libavfilter/vf_libplacebo.c | 15 +++
1 file changed, 7 insertions(+), 8 deletions(-)
diff --git a/libavfilter/vf_libplacebo.c b/libavfilter/vf_libplacebo.c
index
From: Niklas Haas
For now, hard-coded to 1 element.
---
libavfilter/vf_libplacebo.c | 18 +-
1 file changed, 13 insertions(+), 5 deletions(-)
diff --git a/libavfilter/vf_libplacebo.c b/libavfilter/vf_libplacebo.c
index 408fb3918a..fbac1b0354 100644
--- a/libavfilter/vf_libplaceb
From: Niklas Haas
Instead of finding the ref frame in output_frame() and then passing its
signature to update_crops(), pull out the logic and invoke it a second
time inside update_crops().
This may seem wasteful at present, but will actually become required in
the future, since update_crops() ru
From: Niklas Haas
Instead, rename this function to `output_frame` and make it pull the
`pl_frame_mix` from the input structure. Step towards handling multiple
inputs.
---
libavfilter/vf_libplacebo.c | 23 +++
1 file changed, 11 insertions(+), 12 deletions(-)
diff --git a/lib
From: Niklas Haas
To allow placing an input dynamically, as a function of the input index.
---
doc/filters.texi| 2 ++
libavfilter/vf_libplacebo.c | 9 +++--
2 files changed, 9 insertions(+), 2 deletions(-)
diff --git a/doc/filters.texi b/doc/filters.texi
index a46357bfd8..2ea40
From: Niklas Haas
Because these can differ based on the input, for multiple inputs.
---
libavfilter/vf_libplacebo.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/libavfilter/vf_libplacebo.c b/libavfilter/vf_libplacebo.c
index 5af8167bb3..f86493f4e2 100644
--- a/l
From: Niklas Haas
In the event that some frame mixes are OK while others are not, the
priority goes:
1. Errors in updating any frame -> return error
2. Any input incomplete -> request frames and return
3. Any inputs OK -> ignore EOF streams and render remaining inputs
4. No inputs OK -> set outp
From: Niklas Haas
When combining multiple inputs, the output PTS may be less than the PTS
of the input. In this case, the current's code assumption of always
draining one value from the FIFO is incorrect. Replace by a smarter
function which drains only those PTS values that were actually consumed
From: Niklas Haas
Including the queue status, because these will need to be re-queried
inside output_frame_mix when that function is refactored to handle
multiple inputs.
---
libavfilter/vf_libplacebo.c | 11 +--
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/libavfilter/v
From: Niklas Haas
Assign local variable 'in' for 's->input' and replace 'inlink' by
'in->link' to avoid hard-coding ID 0 in more than one place.
---
libavfilter/vf_libplacebo.c | 24
1 file changed, 12 insertions(+), 12 deletions(-)
diff --git a/libavfilter/vf_libplaceb
From: Niklas Haas
To be re-used once we support more than one input.
---
libavfilter/vf_libplacebo.c | 45 -
1 file changed, 29 insertions(+), 16 deletions(-)
diff --git a/libavfilter/vf_libplacebo.c b/libavfilter/vf_libplacebo.c
index 10fd432745..879a3a0508
From: Niklas Haas
In anticipation of a refactor which will enable multiple input support.
Note: the renderer is also input-specific because it maintains a frame
cache, HDR peak detection state and mixing cache, all of which are tied
to a specific input stream.
---
libavfilter/vf_libplacebo.c |
From: Niklas Haas
If the input queue is EOF, then the s->status check should already have
covered it, and prevented the code from getting this far.
If we still hit this case for some reason, it's probably a bug. Better
to hit the AVERROR_BUG branch.
---
libavfilter/vf_libplacebo.c | 3 ---
1 fi
Quoting Armin Hasitzka (2023-06-08 13:37:12)
> Hi all,
>
> just checking; do you consider this for merge into the upstream? It does
> feel like a lot of upside without any downside.
Looks ok, will push soonish if nobody objects.
--
Anton Khirnov
___
f
When qsv device is created by device_derive, the ctx->free function is
not registered, causing potential memory leak because of not properly
closing the MFX session.
Signed-off-by: Tong Wu
Signed-off-by: Wenbin Chen
---
libavutil/hwcontext_qsv.c | 3 +++
1 file changed, 3 insertions(+)
diff --
61 matches
Mail list logo
