---
libavcodec/hw_base_encode.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/libavcodec/hw_base_encode.c b/libavcodec/hw_base_encode.c
index 33a30c8d10..927aeb4bad 100644
--- a/libavcodec/hw_base_encode.c
+++ b/libavcodec/hw_base_encode.c
@@ -666,6 +666,11 @@ int ff_hw_base_init_gop_st
> >>> On Jun 24, 2025, at 11:09, Steven Zhou wrote:
> >>>
> >>> fftools/ffmpeg_filter: at EOF, unref frame used for FPS conversion
> >>> logic
> >>>
> >>> Ever since commit 64f3feb added
> >>> `avcodec_free_context(&enc->enc_ctx);`
> >>> to `enc_free()` the encoder context is closed before filterg
--- Begin Message ---
address sanitizer showed some leaks of drmVersion structs.
`vaapi_device_create` did not call drmFreeVersion in all possible code
paths.
Signed-off-by: Tim Blechmann
---
libavutil/hwcontext_vaapi.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/libavutil/hwcontext_
> -Original Message-
> From: ffmpeg-devel On Behalf Of compn
> Sent: Dienstag, 24. Juni 2025 08:00
> To: ffmpeg-devel@ffmpeg.org
> Subject: Re: [FFmpeg-devel] [PATCH] Revert "Add FUNDING.json"
>
> On Mon, 23 Jun 2025 21:20:56 +0100, Derek Buitenhuis wrote:
>
> > This was pushed with out
This commit introduces a new hardware-accelerated video filter, scale_d3d11,
which performs scaling and format conversion using Direct3D 11. The filter
enables
efficient GPU-based scaling and pixel format conversion (p010 to nv12), reducing
CPU overhead and latency in video pipelines.
> +
Michael Niedermayer writes:
> doesnt the failure path here need the free too ?
> int ret = fir_to_phase(s, &h[longer], &n, &post_peak, s->phase);
> if (ret < 0)
> return ret;
>
I checked fir_to_phase() and found that the only place it allocates memory
in
h[longer] is
--- Begin Message ---
On Wed, 25 Jun 2025, 02:09 compn, wrote:
> On Tue, 24 Jun 2025 10:22:26 +0200, Kieran Kunhya via ffmpeg-devel
> wrote:
>
> > On Tue, 24 Jun 2025, 09:28 compn, wrote:
> >
> > > On Tue, 24 Jun 2025 07:48:30 +0100, Kieran Kunhya via ffmpeg-devel
> > > wrote:
> > >
> > > > On T
On Tue, 24 Jun 2025 10:22:26 +0200, Kieran Kunhya via ffmpeg-devel
wrote:
> On Tue, 24 Jun 2025, 09:28 compn, wrote:
>
> > On Tue, 24 Jun 2025 07:48:30 +0100, Kieran Kunhya via ffmpeg-devel
> > wrote:
> >
> > > On Tue, 24 Jun 2025, 07:00 compn, wrote:
> > >
> > > > On Mon, 23 Jun 2025 21:20
If I don't hear any objections in the next 72 hours I will apply this at
that time, or later if I forget.
Pavel.
On Mon, Jun 23, 2025 at 2:37 PM Pavel Koshevoy wrote:
> ping ... as this is a segfault fix for `ffplay/ffprobe -i 1_poc.mp4` I
> think it deserves some review/approval priority.
>
>
Even if all inputs are blocked an activate callback should request a frame on
some if its inputs if a frame is requested on any of its outputs.
Signed-off-by: Marton Balint
---
libavfilter/avfilter.c | 11 +++
1 file changed, 11 insertions(+)
diff --git a/libavfilter/avfilter.c b/libavf
Since 155508c6e925f4f2f5e77087a7e1925b3de735ff specifying multiple
bsfs for different streams was broken:
"[bsfs/a=h264_metadata:bsfs/v=h264_metadata]out.mp4|..."
This incorrectly only parsed the first bsfs specification. The reason
for this is that the dictionary is modified in the iterator, hen
> On Jun 24, 2025, at 10:53 AM, Michael Niedermayer
> wrote:
>
> Hi Derek
>
> so many different subjects you raise here, ill try to reply seperately
> to keep it clean and terse
>
>> On Tue, Jun 24, 2025 at 06:24:32AM +0100, Derek Buitenhuis wrote:
>> Hi,
>>
>> On Monday, June 23, 2025, Mic
--- Begin Message ---
On Tue, Jun 24, 2025 at 8:48 AM Kieran Kunhya wrote:
>
>
>
> On Tue, 24 Jun 2025, 07:00 compn, wrote:
>>
>> On Mon, 23 Jun 2025 21:20:56 +0100, Derek Buitenhuis wrote:
>>
>> > This was pushed with outstanding reservations, with no positive
>> > affirmation or review from any
Instead of the convoluted nested macros, use a convenience
helper function. While this makes the code slightly longer, it
is now much clearer what is happening without running the file
through a preprocessor first.
Additionally do not mess with the internals of the dictionary,
just to save two str
On Mon, Jun 23, 2025 at 11:43:04AM -0300, James Almer wrote:
> On 6/21/2025 6:15 PM, Michael Niedermayer wrote:
> > Fixes: NULL pointer dereference
> > Fixes:
> > 416811958/clusterfuzz-testcase-minimized-ffmpeg_dem_MOV_fuzzer-5425269114732544
> >
> > Found-by: continuous fuzzing process
> > http
No change in functionality.
Signed-off-by: Marton Balint
---
libavfilter/avfiltergraph.c | 2 ++
libavfilter/buffersink.c| 5 -
libavfilter/buffersrc.c | 4 ++--
libavfilter/filters.h | 1 +
4 files changed, 9 insertions(+), 3 deletions(-)
diff --git a/libavfilter/avfiltergrap
alfilter_graph_request_oldest() might return EAGAIN and produce a frame on not
the oldest sink.
Fixes ticket #11597.
Fixes excessive frame buffering in #10959.
Fixes excessive frame buffering in #11366.
Signed-off-by: Marton Balint
---
fftools/ffmpeg_filter.c | 31 --
Signed-off-by: Marton Balint
---
tests/fate-run.sh| 4
tests/fate/filter-video.mak | 3 +++
tests/ref/fate/filter-select-ffprobe | 25 +
3 files changed, 32 insertions(+)
create mode 100644 tests/ref/fate/filter-select-ffprobe
diff --g
The minimum valid packet length is 28, given that the length includes
the packet header.
This didn't cause any issues so far as the code did not care about the
last two fields in the SR section, but will be relevant in a future
commit.
---
libavformat/rtpdec.c | 2 +-
1 file changed, 1 insertion(
Hi
On Mon, Jun 23, 2025 at 08:48:05PM +0800, ld y wrote:
> Michael Niedermayer writes：
>
> > does the other return in config_output() not need this free too?
> >
>
> I believe all the other return behind fir_to_phase() has called
> av_free(h[longer])
> already. I would love to add this message
On Mon, Jun 23, 2025 at 08:54:24PM +0800, ld y wrote:
> Michael Niedermayer writes:
>
> > this can be factored like:
> >
> > like
> > +ret = 0;
> > fail:
> > +for (i = 0; i < s->nb_streams; i++)
> > +if (contexts[i])
> > +av_free(contexts[i]->url);
> > av_free
This will be used in a future commit to expose the SR as side-data.
---
libavcodec/defs.h | 14 ++
libavformat/rtpdec.c | 34 --
libavformat/rtpdec.h | 4 ++--
libavformat/rtspdec.c | 2 +-
4 files changed, 37 insertions(+), 17 deletions(-)
diff
From: Daniel N Pettersson
Add TLS options to RTSP for when TLS is used for the lower protocol.
Signed-off-by: Marvin Scholz
Co-authored-by: Marvin Scholz
---
libavformat/rtsp.c | 26 +-
libavformat/rtsp.h | 11 +++
2 files changed, 36 insertions(+), 1 deletion(
Hi Derek
so many different subjects you raise here, ill try to reply seperately
to keep it clean and terse
On Tue, Jun 24, 2025 at 06:24:32AM +0100, Derek Buitenhuis wrote:
> Hi,
>
> On Monday, June 23, 2025, Michael Niedermayer
[...]
> > On Mon, Jun 23, 2025 at 09:20:56PM +0100, Derek Buitenhu
It was never reliable to detect if a filtergraph have sources, because a filter
can act as a source only after some time, for example the loop filter.
So it is better to remove the source detection entirely and always give the
scheduler an oppurtunity to stop processing.
Fixes ticket #11604.
Sig
Multi-input or multi-output filters should use activate now.
Signed-off-by: Marton Balint
---
libavfilter/f_select.c | 45 --
1 file changed, 26 insertions(+), 19 deletions(-)
diff --git a/libavfilter/f_select.c b/libavfilter/f_select.c
index aa374c6ad0..
Used in a future commit to not duplicate options.
---
libavformat/tls.h | 10 +++---
1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/libavformat/tls.h b/libavformat/tls.h
index 2f381acc04..c1d2e21508 100644
--- a/libavformat/tls.h
+++ b/libavformat/tls.h
@@ -80,20 +80,24 @@ typed
A frame graph activation might not produce a frame in the requested sink, so
keep on requesting a frame there unless we encounter a filter activation with
buffersrc empty error.
This makes av_buffersink_get_frame(_flags) work according to its documentation
which claims that EAGAIN is only returned
Tests ticket #11624 with a slight modification.
Signed-off-by: Marton Balint
---
tests/fate/filter-audio.mak | 3 +++
tests/ref/fate/filter-anullsink | 6 ++
2 files changed, 9 insertions(+)
create mode 100644 tests/ref/fate/filter-anullsink
diff --git a/tests/fate/filter-audio.mak b/t
Sinks without an activate callback have no means to request frames in their
input, therefore the default activate callback should do it for them.
Fixes ticket #11624.
Fixes ticket #10988.
Fixes ticket #10990.
Signed-off-by: Marton Balint
---
libavfilter/avfilter.c | 9 +
1 file changed,
EOF only need to be forwarded back if all outputs have reached EOF.
Fixes infinte loop with ffprobe -f lavfi -i
"smptebars=d=1,select=n=2:e=1[out0][out1]"
Regression since d9e41ead82263e96ebd14d4d88d6e7f858dd944c.
Fixes ticket #10959.
Fixes ticket #11366.
Signed-off-by: Marton Balint
---
liba
The status check is unneeded because an outlink with a nonzero status should
always return 0 for ff_outlink_frame_wanted(). Also use unsigned for index
because nb_outputs is unsigned as well.
Signed-off-by: Marton Balint
---
libavfilter/filters.h | 4 +---
1 file changed, 1 insertion(+), 3 delet
Signed-off-by: Marton Balint
---
doc/filter_design.txt | 1 +
libavfilter/af_acrossover.c| 10 +-
libavfilter/af_channelsplit.c | 10 +-
libavfilter/filters.h | 15 +++
libavfilter/split.c| 10 +-
libavfilter/vf_extractplanes
---
libavcodec/packet.c | 1 +
libavcodec/packet.h | 6 ++
libavformat/rtpdec.c | 18 ++
3 files changed, 25 insertions(+)
diff --git a/libavcodec/packet.c b/libavcodec/packet.c
index 5104eb98b1..2aace71c7d 100644
--- a/libavcodec/packet.c
+++ b/libavcodec/packet.c
@@ -308
Hi Derek
On Tue, Jun 24, 2025 at 06:24:32AM +0100, Derek Buitenhuis wrote:
> On Monday, June 23, 2025, Michael Niedermayer
[...]
> > The wallet can be handed over to the SPI treassurer, if the SPI treassurer
> > accepts it.
>
>
> Did you consult SPI?
There are long discussions between me and S
--- Begin Message ---
On Tue, Jun 24, 2025 at 4:53 PM Michael Niedermayer
wrote:
>
> Hi Derek
>
> so many different subjects you raise here, ill try to reply seperately
> to keep it clean and terse
>
> On Tue, Jun 24, 2025 at 06:24:32AM +0100, Derek Buitenhuis wrote:
> > Hi,
> >
> > On Monday, Jun
Hi,
On Monday, June 23, 2025, Michael Niedermayer
wrote:
> Hi Derek
>
> You said
> "The only mails you'll see from me in the future are code or reviews."
Perhaps I should have written "patches or reviews". Nice way of inserting a
"fuck you", though. Appreciate it.
>
> On Mon, Jun 23, 2025 at
On Mon, Jun 23, 2025 at 9:12 PM Michael Niedermayer
wrote:
> The General Assembly was informed about the donation on the 9th june
> and on the 18th about the solution with the json file
> as well as my intend to implement this within 48h
> that would be the 20th. i waited 3 more days both for mem
On Tue, 24 Jun 2025 07:48:30 +0100, Kieran Kunhya via ffmpeg-devel
wrote:
> On Tue, 24 Jun 2025, 07:00 compn, wrote:
>
> > On Mon, 23 Jun 2025 21:20:56 +0100, Derek Buitenhuis wrote:
> >
> > > This was pushed with outstanding reservations, with no positive
> > > affirmation or review from any
> -Original Message-
> From: ffmpeg-devel On Behalf Of Zhao
> Zhili
> Sent: Monday, June 23, 2025 8:34 PM
> To: FFmpeg development discussions and patches de...@ffmpeg.org>
> Subject: Re: [FFmpeg-devel] [PATCH] fftools/ffmpeg_filter: at EOF, unref frame
> used for FPS conversion logic
>
On Mon, 23 Jun 2025 21:20:56 +0100, Derek Buitenhuis wrote:
> This was pushed with outstanding reservations, with no positive
> affirmation or review from any other devs, and without a push noticed
> to say it was going to be pushed.
>
> Further, the wallet is controlled by a single individual, a
--- Begin Message ---
On Tue, 24 Jun 2025, 07:00 compn, wrote:
> On Mon, 23 Jun 2025 21:20:56 +0100, Derek Buitenhuis wrote:
>
> > This was pushed with outstanding reservations, with no positive
> > affirmation or review from any other devs, and without a push noticed
> > to say it was going to b
42 matches
Mail list logo
