[FFmpeg-devel] [PATCH] avfilter: add alphamerge_cuda filter

2025-08-06 Thread Jorge Estrada
This patch adds the alphamerge_cuda video filter Example usage: ffmpeg -f lavfi -i "color=c=red:s=1280x720:d=5,format=yuva420p" \ -f lavfi -i "color=c=black:s=1280x720:d=5,geq=lum='255*gt(W/4,hypot(X-W/2,Y-H/2))'" \ -f lavfi -i "color=c=blue:s=1280x720:d=5" \ -filter_complex \ "[0:v]hwupload_cuda

[FFmpeg-devel] [PATCH] libavfilter/avf_showcqt: Add check for av_calloc()

2025-08-06 Thread Jiasheng Jiang
Add check for the return value of av_calloc() to avoid potential NULL pointer dereference. Fixes: 0068b3d0f0 ("avfilter/avf_showcqt: switch to TX FFT from avutil") Signed-off-by: Jiasheng Jiang --- libavfilter/avf_showcqt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib

[FFmpeg-devel] [PATCH] avutil/avassert: use __builtin_assume if available (PR #20151)

2025-08-06 Thread Kacper Michajłow
PR #20151 opened by Kacper Michajłow (kasper93) URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20151 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20151.patch From 888d807e9cb95419cac7d272bee7cc3ca00c2157 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Kacper=20Michaj=C5=82ow?= Date: Thu, 7

Re: [FFmpeg-devel] [PATCH v3] examples: Add check and replace av_free() to avoid potential memory errors

2025-08-06 Thread Michael Niedermayer
On Wed, Aug 06, 2025 at 04:56:42PM +0200, Nicolas George wrote: > Jiasheng Jiang (HE12025-08-06): > > Add check for the return value of av_packet_alloc() to avoid potential NULL > > pointer dereference. > > Moreover, replace redundant av_free() with fprintf(). > > > > Fixes: 9a38184a14 ("examples

Re: [FFmpeg-devel] [PATCH] libavcodec/tests/snowenc: Add av_free() to avoid memory leak

2025-08-06 Thread Michael Niedermayer
On Sun, Aug 03, 2025 at 11:31:27PM +, Jiasheng Jiang wrote: > Add av_free() to free s.temp_dwt_buffer and s.temp_idwt_buffer at the end of > the function to avoid memory leak. > > Fixes: 5d48e4eafa ("Merge commit 'a6a750c7ef240b72ce01e9653343a0ddf247d196'") > Signed-off-by: Jiasheng Jiang >

Re: [FFmpeg-devel] [PATCH] libavfilter/af_firequalizer: Add check for av_malloc_array()

2025-08-06 Thread Michael Niedermayer
On Wed, Aug 06, 2025 at 04:39:47PM +, Jiasheng Jiang wrote: > Add check for the return value of av_malloc_array() to avoid potential NULL > pointer dereference. > > Fixes: d3be186ed1 ("avfilter/firequalizer: add dumpfile and dumpscale option") > Signed-off-by: Jiasheng Jiang > --- > libavfi

Re: [FFmpeg-devel] [PATCH] examples: Add proper deallocations to avoid potential memory leaks

2025-08-06 Thread Michael Niedermayer
On Wed, Aug 06, 2025 at 03:07:21PM +, Jiasheng Jiang wrote: > Add proper deallocations in the error paths to avoid potential memory leaks. > > Fixes: e7dfaf16a4 ("libavfilter: example audio filtering program") > Signed-off-by: Jiasheng Jiang > --- > doc/examples/filter_audio.c | 9 -

Re: [FFmpeg-devel] [PATCH] libavcodec/videotoolbox_vp9: Move av_malloc() to avoid memory leak

2025-08-06 Thread Michael Niedermayer
On Tue, Aug 05, 2025 at 07:31:15PM +, Jiasheng Jiang wrote: > Move av_malloc() after the check for subsampling to avoid memory leak if > subsampling < 0 and av_malloc() succeeds. > > Fixes: a41a2efc85 ("lavc/videotoolbox: add VP9 hardware acceleration") > Signed-off-by: Jiasheng Jiang > ---

Re: [FFmpeg-devel] [PATCH 1/3] avcodec/libdav1d: move itut-t35 parsing to a separate function

2025-08-06 Thread Michael Niedermayer
On Tue, Aug 05, 2025 at 12:12:39PM +0200, Maryla Ustarroz-Calonge via ffmpeg-devel wrote: > This is in preparation to change the switch statement to if/else. > > Signed-off-by: Maryla Ustarroz-Calonge > --- > libavcodec/libdav1d.c | 165 ++ > 1 file chang

[FFmpeg-devel] [PATCH] Mark C globals with small code model (PR #20150)

2025-08-06 Thread pranavk
PR #20150 opened by pranavk URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20150 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20150.patch By default, all globals in C/C++ compiled by clang are allocated in non-large data sections. See [1] for background on code models. For PIC (Positio

[FFmpeg-devel] [PATCH] avdevice/gdigrab: fix -show_region 1 overlay window crashing FFmpeg when clicked (PR #20148)

2025-08-06 Thread Hunter Kvalevog
PR #20148 opened by Hunter Kvalevog (k-vog) URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20148 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20148.patch Tested on Windows 7, 10, 11, and Debian (Xorg + Wine) Fixes: #11539 >From b1012d3796f98a8f1eb6543fcd62be80ca963fa1 Mon Sep 17 00:

[FFmpeg-devel] [PATCH] Revert "forgejo/lint_commit_msg: add script for commit message linting" (PR #20147)

2025-08-06 Thread michaelni
PR #20147 opened by michaelni URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20147 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20147.patch This should be done by fate. This script already blocks security fixes (https://code.ffmpeg.org/FFmpeg/FFmpeg/actions/runs/1046/jobs/0) IMHO its

Re: [FFmpeg-devel] [h264] Make ff_h264_build_ref_list stricter with AV_EF_EXPLODE

2025-08-06 Thread Michael Niedermayer
Hi On Tue, Aug 05, 2025 at 02:52:28PM -0700, Dale Curtis wrote: > Don't silently skip errors when AV_EF_EXPLODE is specified. This can > lead to out-of-bound reads with ff_put_h264_chroma_mc4_ssse3() when > small padding is used with the checked bitstream reader. > > Signed-off-by: Dale Curtis

[FFmpeg-devel] [PATCH] .forgejo/CODEOWNERS: Add myself for a more libavutil bits which i know well (PR #20146)

2025-08-06 Thread michaelni
PR #20146 opened by michaelni URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20146 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20146.patch Signed-off-by: Michael Niedermayer >From fb8afa5f018f406250e58d801b84f7af73683480 Mon Sep 17 00:00:00 2001 From: Michael Niedermayer Date: Wed

[FFmpeg-devel] [PATCH] avcodec/mpc8: init avctx->sample_rate (PR #20144)

2025-08-06 Thread Kacper Michajłow
PR #20144 opened by Kacper Michajłow (kasper93) URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20144 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20144.patch Fixes frame validation. Fixes: OSS-Fuzz/416134551 From b3f94d3da8a60fde592bd76bc61ba394c1f18aef Mon Sep 17 00:00:00 2001 Fro

[FFmpeg-devel] [PATCH] avformat/Makefile: fix error unresolved external symbol ff_dtls_protocol (PR #20143)

2025-08-06 Thread jianhuaw
PR #20143 opened by jianhuaw URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20143 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20143.patch Introduced by 307983b292adb60c82e75c8e5e41deedc89efbf0 Use the following command line to reproduce the issue: ./configure --toolchain=msvc --disab

[FFmpeg-devel] [PATCH] configure: add POWER9 & 10 cpus (PR #20137)

2025-08-06 Thread Sean McGovern
PR #20137 opened by Sean McGovern (sean_mcg) URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20137 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20137.patch IBM introduced POWER9 in 2017 and likewise POWER10 in 2020. >From 37eebd7fff8dc28bdc3652b941c7e11be54dfc69 Mon Sep 17 00:00:00 20

[FFmpeg-devel] [PATCH] avcodec/vvc/ctu: check coeff before multiply (PR #20142)

2025-08-06 Thread Kacper Michajłow
PR #20142 opened by Kacper Michajłow (kasper93) URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20142 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20142.patch ff_vvc_palette_escape_val() can return AVERROR in which case the coeff*scale will overflow. Fixes: runtime error: signed intege

[FFmpeg-devel] [PATCH] MAINTAINERS: add myself as hwcontext_d3d12va maintainer (PR #20141)

2025-08-06 Thread toqsxw
PR #20141 opened by toqsxw URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20141 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20141.patch >From 02b103af6008235afc9811ec9a2f3402239f79ce Mon Sep 17 00:00:00 2001 From: Wu Jianhua Date: Thu, 7 Aug 2025 00:28:22 +0800 Subject: [PATCH 1/2]

[FFmpeg-devel] [PATCH v2] examples: Add check and replace av_free() to avoid potential memory errors

2025-08-06 Thread Jiasheng Jiang
Add check for the return value of av_packet_alloc() to avoid potential NULL pointer dereference. Moreover, replace redundant av_free() with fprintf(). Fixes: 9a38184a14 ("examples/decode_audio: allocate the packet dynamically") Fixes: f5df897c4b ("examples/avcodec: split audio decoding into a sep

[FFmpeg-devel] [PATCH] libavfilter/af_firequalizer: Add check for av_malloc_array()

2025-08-06 Thread Jiasheng Jiang
Add check for the return value of av_malloc_array() to avoid potential NULL pointer dereference. Fixes: d3be186ed1 ("avfilter/firequalizer: add dumpfile and dumpscale option") Signed-off-by: Jiasheng Jiang --- libavfilter/af_firequalizer.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/l

[FFmpeg-devel] [PATCH] [PATCH] avformat/apngdec: allow other chunks between fcTL and fdAT/IDAT (PR #20140)

2025-08-06 Thread Park Jeonghwan
PR #20140 opened by Park Jeonghwan (devjeonghwan) URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20140 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20140.patch The APNG demuxer incorrectly assumed that fcTL chunks must be immediately followed by fdAT or IDAT chunks. Per PNG specificat

[FFmpeg-devel] [PATCH] avcodec/cbs_h266_syntax_template: fix out of bounds access (PR #20139)

2025-08-06 Thread Kacper Michajłow
PR #20139 opened by Kacper Michajłow (kasper93) URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20139 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20139.patch current->num_tile_columns is not updated in the loop, so the existing check was not working. Check current index instead and bre

[FFmpeg-devel] 回复: [PATCH v3] avutil/hwcontext_d3d12va: added resource and heap flags to DeviceContext

2025-08-06 Thread Wu Jianhua
Dmitrii Ovchinnikov: > --- > libavutil/hwcontext_d3d12va.c | 17 - > libavutil/hwcontext_d3d12va.h | 11 +++ > 2 files changed, 27 insertions(+), 1 deletion(-) > > diff --git a/libavutil/hwcontext_d3d12va.c b/libavutil/hwcontext_d3d12va.c > index 6507cf69c1..dca8ddc0c3 1006

[FFmpeg-devel] [PATCH v1] avformat/Makefile: fix error unresolved external symbol ff_dtls_protocol

2025-08-06 Thread toqsxw
From: Wu Jianhua Introduced by 307983b292adb60c82e75c8e5e41deedc89efbf0 Use the following command line to reproduce the issue: ./configure --toolchain=msvc --as=nasm --enable-ffmpeg --disable-everything --enable-decoder=vvc --enable-parser=vvc --enable-demuxer=vvc,mpegts --enable-protocol=file

[FFmpeg-devel] [PATCH] doc/examples/hw_decode: Add proper deallocations to avoid potential memory leaks

2025-08-06 Thread Jiasheng Jiang
Add proper deallocations in the error paths to avoid potential memory leaks. Fixes: 1e0c75ea16 ("examples/hw_decode: Add a HWAccel decoding example.") Signed-off-by: Jiasheng Jiang --- doc/examples/hw_decode.c | 48 +++- 1 file changed, 33 insertions(+), 15 de

[FFmpeg-devel] [PATCH] avfilter/vf_colordetect: optimize C functions a bit (PR #20138)

2025-08-06 Thread Kacper Michajłow
PR #20138 opened by Kacper Michajłow (kasper93) URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20138 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20138.patch They are used to process tail, so it's still good to have them faster. Even if AVX version are used. GCC 14.2.0 | x86_64 (defau

Re: [FFmpeg-devel] [PATCH] avformat/udp: Select output interfaces for ipv6 multicast

2025-08-06 Thread Rémi Denis-Courmont
Le 5 août 2025 14:56:18 GMT+07:00, Peter Enderborg a écrit : >This fixes two old TODO's in ipv6 multicast handling. >A helper function to iterate over all interfaces added to >help join the multicast group on interface if approperite. > >The default value is IN6ADDR_ANY (::) and it then joins al

[FFmpeg-devel] [PATCH] examples: Add proper deallocations to avoid potential memory leaks

2025-08-06 Thread Jiasheng Jiang
Add proper deallocations in the error paths to avoid potential memory leaks. Fixes: e7dfaf16a4 ("libavfilter: example audio filtering program") Signed-off-by: Jiasheng Jiang --- doc/examples/filter_audio.c | 9 - 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/doc/examples/f

[FFmpeg-devel] [PATCH v3] examples: Add check and replace av_free() to avoid potential memory errors

2025-08-06 Thread Jiasheng Jiang
Add check for the return value of av_packet_alloc() to avoid potential NULL pointer dereference. Moreover, replace redundant av_free() with fprintf(). Fixes: 9a38184a14 ("examples/decode_audio: allocate the packet dynamically") Signed-off-by: Jiasheng Jiang --- Changelog: v2 -> v3: 1. Remove w

Re: [FFmpeg-devel] [PATCH v3] examples: Add check and replace av_free() to avoid potential memory errors

2025-08-06 Thread Nicolas George
Jiasheng Jiang (HE12025-08-06): > Add check for the return value of av_packet_alloc() to avoid potential NULL > pointer dereference. > Moreover, replace redundant av_free() with fprintf(). > > Fixes: 9a38184a14 ("examples/decode_audio: allocate the packet dynamically") > Signed-off-by: Jiasheng J

Re: [FFmpeg-devel] [PATCH v2] examples: Add check and replace av_free() to avoid potential memory errors

2025-08-06 Thread Nicolas George
Jiasheng Jiang (HE12025-08-06): > Add check for the return value of av_packet_alloc() to avoid potential NULL > pointer dereference. > Moreover, replace redundant av_free() with fprintf(). > > Fixes: 9a38184a14 ("examples/decode_audio: allocate the packet dynamically") This one looks good to me,

[FFmpeg-devel] [PATCH] forgejo/CODEOWNERS: add myself to docs (PR #20135)

2025-08-06 Thread Gyan Doshi
PR #20135 opened by Gyan Doshi (GyanD) URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20135 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20135.patch >From 193f389793c9e7996a1a60d93fc89cde1acddf8b Mon Sep 17 00:00:00 2001 From: Gyan Doshi Date: Wed, 6 Aug 2025 18:28:16 +0530 Subject:

[FFmpeg-devel] [PATCH] avformat/movenc: add support for fragmented TTML muxing (PR #20136)

2025-08-06 Thread Jan Ekström
PR #20136 opened by Jan Ekström (jeeb) URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20136 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20136.patch This enables pushing TTML together with another track (usually video) as part of CMAF Ingest, as defined by the DASH-IF Live Media Ingest

[FFmpeg-devel] [PATCH] FFV1 / utvideo / magicyuv uninitilaized memory use (PR #20134)

2025-08-06 Thread michaelni
PR #20134 opened by michaelni URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20134 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20134.patch >From 74eb5aad285ed8174c70248153fca84f0b4f2248 Mon Sep 17 00:00:00 2001 From: Michael Niedermayer Date: Wed, 6 Aug 2025 12:49:49 +0200 Subject:

Re: [FFmpeg-devel] rebasing security

2025-08-06 Thread Michael Niedermayer
Hi On Wed, Aug 06, 2025 at 08:51:01AM +0200, Alexander Strasser via ffmpeg-devel wrote: > On 2025-08-06 00:37 +0200, Michael Niedermayer wrote: > > > > On Mon, Aug 04, 2025 at 10:15:53PM +0200, Alexander Strasser via > > ffmpeg-devel wrote: > [...] > > > > > > If I understand the original poin

[FFmpeg-devel] [PATCH] avcodec/oapv_decoder: Provided support for libopenapv APV decoder (PR #20133)

2025-08-06 Thread dkozinski
PR #20133 opened by dkozinski URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20133 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20133.patch - Added APV decoder wrapper - Changes in project configuration file and libavcodec Makefile - Added documentation for APV decoder wrapper Signed-

[FFmpeg-devel] fuzzer and security issues

2025-08-06 Thread Michael Niedermayer
Hi all theres been a surge of new issues being reported in recent days. And its affecting release schedule a bit People having access to ffmpeg-security or ossfuzz, are welcome to help with fixing them. (general rule is check ML for the testcase filename, before working on a case or just ask me

[FFmpeg-devel] [PATCH] avfilter/vf_thumbnail: fix buffer overflow for odd sized HBD inputs (PR #20132)

2025-08-06 Thread Niklas Haas
PR #20132 opened by Niklas Haas (haasn) URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20132 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20132.patch This histogram index was not correctly downshifted to 8-bit. >From 651ce7dbc9cb88ab01002867ddb5b44a6f17a647 Mon Sep 17 00:00:00 2001 F

[FFmpeg-devel] [PATCH] fix j2k cdef out of array (PR #20131)

2025-08-06 Thread michaelni
PR #20131 opened by michaelni URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20131 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20131.patch >From c7c16002ea3f9d6f284b61c7e5d4dc3fc9d1410a Mon Sep 17 00:00:00 2001 From: Michael Niedermayer Date: Tue, 5 Aug 2025 23:18:47 +0200 Subject:

Re: [FFmpeg-devel] [PATCH] examples: Replace with proper error handling to avoid potential memory errors

2025-08-06 Thread Nicolas George
Alexander Strasser via ffmpeg-devel (HE12025-08-06): > I'm a bit undecided about this; especially because these are example files. > > Best would be to have them correct, small, and beautiful while still > promoting good C practices. Good C practices are verbose, and verbosity makes for poor exam

[FFmpeg-devel] [PATCH] out of array fixes in exr (PR #20130)

2025-08-06 Thread michaelni
PR #20130 opened by michaelni URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20130 Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20130.patch >From 92a9948ae44ac4cee869c239b2a3667e2d36ac63 Mon Sep 17 00:00:00 2001 From: Michael Niedermayer Date: Wed, 6 Aug 2025 10:08:14 +0200 Subject:

Re: [FFmpeg-devel] [PATCH] examples: Replace with proper error handling to avoid potential memory errors

2025-08-06 Thread Alexander Strasser via ffmpeg-devel
On 2025-08-05 22:18 +0200, Nicolas George wrote: > Jiasheng Jiang (HE12025-08-05): > > Moreover, replace with proper error handling to avoid potential memory leak. > > There is no potential memory leak with exit(). > > And it does not have to be a lesson in good C practices. I'm a bit undecided

Re: [FFmpeg-devel] Forgejo entry threshold

2025-08-06 Thread Nicolas George
Rémi Denis-Courmont (HE12025-08-06): > But it is exclusionary, as not everyone can get a working reputable > outbound SMTP server. It has already been said time and again that it is not a requirement to participate. The only requirement is the ability to send and receive mail. -- Nicolas Geor

Re: [FFmpeg-devel] [PATCH] gcc: Don't disable '-ftree-vectorize' on GCC >= 13 on major architectures (PR #20120)

2025-08-06 Thread Martin Storsjö
On Wed, 6 Aug 2025, Rémi Denis-Courmont wrote: FWIW I don't really care personally but judging by the original author and their affiliation, the intent was most probably to enable this for RISC-V / RVA23. On that platform it is *probably* safe to enable vectorisation on all GCC versions but