[FFmpeg-devel] [PATCH] avfilter/vf_grayworld: Use the correct pointer for av_log()

Fixes: crash Found-by: Elias Myllymäki Signed-off-by: Michael Niedermayer --- libavfilter/vf_grayworld.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/libavfilter/vf_grayworld.c b/libavfilter/vf_grayworld.c index a6a51dcb7db..91b097a4b80 100644 --- a/libavfilter/vf_gra

[FFmpeg-devel] [PATCH 1/5] ffv1dec: use dedicated pix_fmt field and call ff_get_format

Adding support for hwaccels means that avctx->pix_fmt will indicate hardware formats. --- libavcodec/ffv1.h| 1 + libavcodec/ffv1dec.c | 145 +++ 2 files changed, 79 insertions(+), 67 deletions(-) diff --git a/libavcodec/ffv1.h b/libavcodec/ffv1.h ind

[FFmpeg-devel] [PATCH 2/5] ffv1dec: move slice start finding into a function

This also cleans up and gives the code some much needed comments. --- libavcodec/ffv1dec.c | 66 +++- 1 file changed, 47 insertions(+), 19 deletions(-) diff --git a/libavcodec/ffv1dec.c b/libavcodec/ffv1dec.c index b7f235c47f..aa988571bc 100644 --- a/libavc

Re: [FFmpeg-devel] [PATCH 1/3] ffv1dec: move slice start finding into a function

On 06/01/2025 12:40, Michael Niedermayer wrote: Hi Lynne On Sun, Jan 05, 2025 at 05:59:35PM +0900, Lynne wrote: This also cleans up and gives the code some much needed comments. --- libavcodec/ffv1dec.c | 66 +++- 1 file changed, 47 insertions(+), 19

[FFmpeg-devel] [PATCH 5/5] ffv1dec: add support for hwaccels

This commit adds support for hardware accelerated decoding to the decoder. The previous commits already refactored the decoder, this commit simply adds calls to hooks to decode. --- libavcodec/ffv1.h| 1 + libavcodec/ffv1dec.c | 62 +--- 2 files changed

[FFmpeg-devel] [PATCH 4/5] ffv1dec: move slice decoding into a separate function

This simply movies all slice decoding code from decode_frame to decode_slices; nothing more. --- libavcodec/ffv1dec.c | 106 --- 1 file changed, 60 insertions(+), 46 deletions(-) diff --git a/libavcodec/ffv1dec.c b/libavcodec/ffv1dec.c index 68a233932d..83c

[FFmpeg-devel] [PATCH 3/5] ffv1dec: move header parsing into a separate function

--- libavcodec/ffv1dec.c | 64 +++- 1 file changed, 40 insertions(+), 24 deletions(-) diff --git a/libavcodec/ffv1dec.c b/libavcodec/ffv1dec.c index aa988571bc..68a233932d 100644 --- a/libavcodec/ffv1dec.c +++ b/libavcodec/ffv1dec.c @@ -548,11 +548,10 @@ st

Re: [FFmpeg-devel] [PATCH v2] ffv1dec: add support for hwaccels

On 06/01/2025 12:29, Michael Niedermayer wrote: Hi On Sun, Jan 05, 2025 at 06:00:23PM +0900, Lynne wrote: This commit adds support for hardware accelerated decoding to the decoder. The previous commits already refactored the decoder, this commit simply adds calls to hooks to decode. --- lib

Re: [FFmpeg-devel] [PATCH v2] ffv1dec: add support for hwaccels

Hi On Sun, Jan 05, 2025 at 06:00:23PM +0900, Lynne wrote: > This commit adds support for hardware accelerated decoding to > the decoder. > The previous commits already refactored the decoder, this commit > simply adds calls to hooks to decode. > --- > libavcodec/ffv1.h| 1 + > libavcodec/ffv

Re: [FFmpeg-devel] [PATCH 2/3] ffv1dec: move header parsing into a separate function

Hi Lynne On Sun, Jan 05, 2025 at 05:59:36PM +0900, Lynne wrote: > --- > libavcodec/ffv1dec.c | 64 +++- > 1 file changed, 40 insertions(+), 24 deletions(-) patch 1 applied fine but this does not apply cleanly Applying: ffv1dec: move header parsing into a

Re: [FFmpeg-devel] [PATCH 1/3] ffv1dec: move slice start finding into a function

Hi Lynne On Sun, Jan 05, 2025 at 05:59:35PM +0900, Lynne wrote: > This also cleans up and gives the code some much needed comments. > --- > libavcodec/ffv1dec.c | 66 +++- > 1 file changed, 47 insertions(+), 19 deletions(-) > > diff --git a/libavcodec/ffv1

Re: [FFmpeg-devel] [PATCH] avcodec.h: document padding requirements for get_buffer() callback.

Hi On Thu, Jan 02, 2025 at 06:09:19PM -0500, Ronald S. Bultje wrote: > Hi, > > On Thu, Jan 2, 2025 at 5:48 PM James Almer wrote: > > > On 1/2/2025 7:39 PM, Ronald S. Bultje wrote: [...] > > And for that matter, see fuzz_video_get_buffer() in > > tools/target_dec_fuzzer.c, where no such thing

[FFmpeg-devel] [PATCH 2/3] ffv1dec: move header parsing into a separate function

--- libavcodec/ffv1dec.c | 64 +++- 1 file changed, 40 insertions(+), 24 deletions(-) diff --git a/libavcodec/ffv1dec.c b/libavcodec/ffv1dec.c index aa988571bc..68a233932d 100644 --- a/libavcodec/ffv1dec.c +++ b/libavcodec/ffv1dec.c @@ -548,11 +548,10 @@ st

Re: [FFmpeg-devel] [PATCH] avcodec/jpeg2000dec: clear array length when freeing it

Hi On Wed, Jan 01, 2025 at 11:58:39PM -0300, James Almer wrote: > Fixes NULL pointer dereferences. > Fixes ticket #11393. > > Signed-off-by: James Almer > --- > libavcodec/jpeg2000dec.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/libavcodec/jpeg2000dec.c b/libavcodec/jpeg2000dec.c

[FFmpeg-devel] [PATCH] lavfi/src_movie: set pkt_timebase

Fix “Could not update timestamps for skipped samples” warning and associated misfeature. Signed-off-by: Nicolas George --- libavfilter/src_movie.c | 1 + 1 file changed, 1 insertion(+) diff --git a/libavfilter/src_movie.c b/libavfilter/src_movie.c index d2aa572d12..a1b3d026a6 100644 --- a/libav

[FFmpeg-devel] [PATCH] avcodec/hevc: Add support for output_corrupt/showall flags

From: Zhao Zhili hevc decoder output corrupted frames when the reference frames are missing. The output_corrupt flag has no effect. ffplay shows artifacting after seek when playback TS stream. This patch mark those frames with AV_FRAME_FLAG_CORRUPT flag when output_corrupt flag is set, and drop

[FFmpeg-devel] [PATCH v2] ffv1dec: add support for hwaccels

This commit adds support for hardware accelerated decoding to the decoder. The previous commits already refactored the decoder, this commit simply adds calls to hooks to decode. --- libavcodec/ffv1.h| 1 + libavcodec/ffv1dec.c | 62 +--- 2 files changed

[FFmpeg-devel] [PATCH 3/3] ffv1dec: move slice decoding into a separate function

This simply movies all slice decoding code from decode_frame to decode_slices; nothing more. --- libavcodec/ffv1dec.c | 106 --- 1 file changed, 60 insertions(+), 46 deletions(-) diff --git a/libavcodec/ffv1dec.c b/libavcodec/ffv1dec.c index 68a233932d..83c

[FFmpeg-devel] [PATCH 1/3] ffv1dec: move slice start finding into a function

This also cleans up and gives the code some much needed comments. --- libavcodec/ffv1dec.c | 66 +++- 1 file changed, 47 insertions(+), 19 deletions(-) diff --git a/libavcodec/ffv1dec.c b/libavcodec/ffv1dec.c index b7f235c47f..aa988571bc 100644 --- a/libavc

[FFmpeg-devel] [PATCH] avformat/flvenc: fix missing sequence start with MP3 tracks

When muxing to FLV/RTMP a MP3 track with an ID greater than zero, enhanced RTMP has to be used, and a sequence start should preceed track data. This is already implemented (see line 823 of flvenc.c) but the code is never reached due to a too-strict condition before it. This patch fixes the issue.

Re: [FFmpeg-devel] [PATCH 1/2] avfilter/vf_addroi: Add missing NULL termination to addroi_var_names[]()

On 2025-01-05 05:44 +0100, Michael Niedermayer wrote: > Fixes: out of array read > > Found-by: Elias Myllymäki > Signed-off-by: Michael Niedermayer > --- > libavfilter/vf_addroi.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/libavfilter/vf_addroi.c b/libavfilter/vf_addroi.c > index d

Re: [FFmpeg-devel] [PATCH] avcodec/vvc decode: ALF filtering without CC-ALF

On Fri, Jan 3, 2025 at 1:56 AM Frank Plowman wrote: > Hi both, > > Can we push either of these? Someone else has reported the same issue > on Trac (#11325). The difference between the two proposed > implementations seems rather superficial to me. > Seems my suggestion would complicate ff_vvc_al

Re: [FFmpeg-devel] [PATCH v2] lavc/vvc: Fix race condition for MVs cropped to subpic

On Sat, Jan 4, 2025 at 9:51 PM Nuo Mi wrote: > > > On Fri, Jan 3, 2025 at 2:01 AM Frank Plowman > wrote: > >> Thank you for your review. >> >> On 01/01/2025 04:30, Nuo Mi wrote: >> > 👍 >> > >> > On Tue, Dec 31, 2024 at 2:02 AM Frank Plowman >> wrote: >> > >> >> When the current subpicture has s

Re: [FFmpeg-devel] [RFC PATCH] avformat/rtpdec: fix another int overflow in start_time_realtime

On Tue, Nov 26, 2024, at 1:35 AM, j...@jonb.org wrote: > > This patch replaces av_rescale, which operates on int64_t, with > ff_parse_ntp_time, which operates on uint65_t. This will give the correct > values for timestamps back around the NTP epoch and present day timestamps. > Any feedback