On Mon, Jul 18, 2022 at 08:56:32PM +0200, Michael Niedermayer wrote: > On Sat, Jul 09, 2022 at 08:45:31AM +0200, Steve Lhomme wrote: > > Patch attached in the email. > > > > In some cases, the submit packet can result in configurations changes of the > > hardware decoders. The previous HW context is then freed and a new one > > created. That context is supposed to move up to the main context and the > > other threads. > > > > It appears that when more than 2 frame threads are involved, the new context > > doesn't move in the right place (or rather at the right time). And it can > > create a crash when reusing the old HW context. This patch fixes the issue I > > could reproduce in VLC with DXVA decoding. > > > > I have no idea if this is correct or the side effects induced by this. It > > seems the right thing to do. Keeping the previous call exhibits the issue. > > It seems odd the other existing thread context are not updated with the > > current hwaccel_priv_data. But maybe they are updated later from the "main" > > thread context, in which case this patch seems solid. > > > pthread_frame.c | 4 ++-- > > 1 file changed, 2 insertions(+), 2 deletions(-) > > 2274e52382008f403c7bf52f76d608d2a56ef859 > > 0001-avcodec-pthread_frame-update-the-main-avctx-from-the.patch > > From e8abeeff92f5d7b3b553acdb7595d40153cbec1e Mon Sep 17 00:00:00 2001 > > From: Steve Lhomme <rob...@ycbcr.xyz> > > Date: Fri, 8 Jul 2022 11:49:27 +0200 > > Subject: [PATCH] avcodec/pthread_frame: update the main avctx from the > > current > > ThreadContext > > > > After a submit_decoder() the hwaccel_priv_data may have changed in that > > avctx. > > > > Doing it after the "next available frame" loop will likely point to the > > hwaccel_priv_data from another PerThreadContext which had the old value > > which > > might have been freed, if the submit_decoder() resulting in a format change. > > --- > > libavcodec/pthread_frame.c | 4 ++-- > > 1 file changed, 2 insertions(+), 2 deletions(-) > > It would be nice if a solution to this would make it in 5.1
noone ? i guess the fix will then be in 5.1.1 is there any note / warning that should be included in the release notes or something about this open issue ? thx [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB What does censorship reveal? It reveals fear. -- Julian Assange
signature.asc
Description: PGP signature
_______________________________________________ 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 "unsubscribe".
