PR #20951 opened by HecaiYuan
URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20951
Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20951.patch
This wrapper performs runtime dispatch to SIMD-optimized functions and
falls back to a C implementation for unsupported block heights.
>From 3d5
> On Nov 18, 2025, at 09:26, Xiang, Haihao via ffmpeg-devel
> wrote:
>
> From: Disclosure
>
> Reviewed-by: Zhong Li
> Reviewed-by: Haihao Xiang
> Cc: Michael Niedermayer
> Signed-off-by: Disclosure
> ---
> libavcodec/qsv.c | 2 ++
> libavfilter/qsvvpp.c | 2 ++
> libavutil/
From: Disclosure
Reviewed-by: Zhong Li
Reviewed-by: Haihao Xiang
Cc: Michael Niedermayer
Signed-off-by: Disclosure
---
libavcodec/qsv.c | 2 ++
libavfilter/qsvvpp.c | 2 ++
libavutil/hwcontext_qsv.c | 20 ++--
3 files changed, 18 insertions(+), 6 deletions(-)
Won't this busy-loop in filter.c if it keeps returning EAGAIN for longer
periods of time/forever?
The whole API really isn't prepared for non-blocking operations, so I'm
not sure this is a good idea.
OpenPGP_0xD85AAA6874B7507E.asc
Description: OpenPGP public key
OpenPGP_signature.asc
Descript
PR #20950 opened by Kacper Michajłow (kasper93)
URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20950
Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20950.patch
This is useful if one wants to inspect build artifacts after running
fate.sh script.
From b1132d722d2e681b87f48946461cb61b0540
PR #20949 opened by anders-mjoll
URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20949
Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20949.patch
During migration to the activation filter API the aselect filter was
accidentally turned into a no-op filter.
>From dd7f99e5b17acde5de9275872
From cdc2e439bab494ccfcef423ece1f066fab17683c Mon Sep 17 00:00:00 2001
From: Daniel Riehm
Date: Mon, 17 Nov 2025 15:25:42 -0600
Subject: [PATCH] avformat/aviobuf: ensure EAGAIN is not interpreted as EOF
When AVIO_FLAG_NONBLOCK is enabled, I/O operations may
return AVERROR(EAGAIN). avio_read() sh
PR #20948 opened by Kacper Michajłow (kasper93)
URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20948
Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20948.patch
While cl.exe supports -guard:signret, armasm64 complains about
unknown flag. Note that -guard:ehcont is accepted by armasm64.
F
PR #20947 opened by Kacper Michajłow (kasper93)
URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20947
Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20947.patch
From 3616b152ce8899437442b290d1a74bfbb4836961 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Kacper=20Michaj=C5=82ow?=
Date: Mon, 17
PR #20946 opened by anthonybajoua
URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20946
Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20946.patch
Currently, when there is a file that is muxed in fragmented mode with flags
`frag_keyframe+delay_moov+empty_moov+dash` the second track's len
PR #20944 opened by Jack Lau (JackLau)
URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20944
Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20944.patch
Fix #20940
This reverts commit 4440e499ba
It breaks some filters like feedback, and make
li->frame_blocked_in check invaild.
Signed-of
PR #20943 opened by Marvin Scholz (ePirat)
URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20943
Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20943.patch
Backports #20541 and #20542 to 7.1.
>From 2730ec798357856ba996ac856b14f8325168ebcd Mon Sep 17 00:00:00 2001
From: Marvin Scholz
Da
PR #20942 opened by mkver
URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20942
Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20942.patch
>From fd5fbb41ccaf846a3dc1742ba65dfe6714a543aa Mon Sep 17 00:00:00 2001
From: Andreas Rheinhardt
Date: Mon, 17 Nov 2025 15:35:38 +0100
Subject: [PAT
On Monday, November 17th, 2025 at 3:48 PM, Tobias Rapp via ffmpeg-devel
wrote:
>
>
> On 11/11/2025 03:33, Michael Niedermayer via ffmpeg-devel wrote:
>
> > Hi
> >
> > adding niklas to the CC so its not missed
> > but i agree the patch LGTM
> >
> > On Tue, Nov 04, 2025 at 08:52:36PM +0100, N
On 11/11/2025 03:33, Michael Niedermayer via ffmpeg-devel wrote:
Hi
adding niklas to the CC so its not missed
but i agree the patch LGTM
On Tue, Nov 04, 2025 at 08:52:36PM +0100, Nicolas George via ffmpeg-devel wrote:
Carl Hetherington via ffmpeg-devel (HE12025-11-03):
Since 3b26b782eeded9b9
Hi everyone
Iam working on setting up ffmpeg github sponsors support (this will be
connected to SPI)
thx
--
Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB
I am the wisest man alive, for I know one thing, and that is that I know
nothing. -- Socrates
signature.asc
Des
Hi
On Sun, Nov 16, 2025 at 09:16:03PM -0800, ff--- via ffmpeg-devel wrote:
> On 2025-11-16 07:59, Michael Niedermayer via ffmpeg-devel wrote:
> > On Sun, Nov 16, 2025 at 04:12:00PM +0100, Michael Niedermayer via
> > ffmpeg-devel wrote:
> > > On Sun, Nov 16, 2025 at 01:56:49PM +, Kieran Kunhya
PR #20941 opened by Lynne
URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20941
Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20941.patch
>From ac98c3e731bccf35480a3bddebc930627cd2d482 Mon Sep 17 00:00:00 2001
From: Lynne
Date: Sat, 20 May 2023 16:09:22 +0200
Subject: [PATCH] aacdec: a
I have been going through some code and found a couple of FIXME's that
mention the section of code to be "slow" and I was just curious since
I am relatively new to most kinds of profiling, how do you evaluate
this and if anybody were to come up with a patch for this, how can
they reliably show spee
PR #20939 opened by Niklas Haas (haasn)
URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20939
Patch URL: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20939.patch
The approach of this ASM routine is to process two channels at a time using
AVX instructions. Obviously, there is no point in doing this
20 matches
Mail list logo
