When using slice decoding vp9_free_entries() is called before
vp9_alloc_entries() is ever called. It should destroy properly
initialized variables (or check it was never called before).
It usually works undetected as pthread implementations allows NULL as a
special value (and should return EINVAL
v2: shorter commit lines and removed an extra space, now I realize it's
the wrong one, the original being also wrong...
On 2021-09-02 11:19, Steve Lhomme wrote:
When using slice decoding vp9_free_entries() is called before
vp9_alloc_entries() is ever called. It should destroy pro
When using slice decoding vp9_free_entries() is called before
vp9_alloc_entries() is ever called. It should destroy properly
initialized variables (or check it was never called before).
It usually works undetected as pthread implementations allows NULL as a
special value (and should return EINVAL
On 2021-09-02 11:27, Andreas Rheinhardt wrote:
Steve Lhomme:
When using slice decoding vp9_free_entries is called before vp9_alloc_entries
is ever called. It should destroy properly initialized variables (or check it
was never called before).
It usually works undetected as pthread
initialization.
This is undefined behaviour and certain threading implementations
like pthreadGC2 crash when it is attempted.
Fix this by initializing these objects once during init and freeing
them in close.
Works for me.
Reported-by: Steve Lhomme
Signed-off-by: Andreas Rheinhardt
On 2021-09-02 17:41, Andreas Rheinhardt wrote:
The earlier code did not properly check these initializations:
It only recorded whether the part of init where these initializations
are has been reached, but it did not check whether the initializations
were successful, although destroying them woul
We should return 0 and assert there's something wrong. This is how it's done
with the other DXVA variants.
It fixes bad reference usage in http://www.gbbsoft.pl/!download/!/Film1.mp4
It's using a frame that doesn't have any data[] fields set (yet?).
Signed-off-by: Steve Lho
/26738
Signed-off-by: Steve Lhomme
---
libavcodec/dxva2_hevc.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/dxva2_hevc.c b/libavcodec/dxva2_hevc.c
index c91bcf3eeb..6b239d9917 100644
--- a/libavcodec/dxva2_hevc.c
+++ b/libavcodec/dxva2_hevc.c
@@ -184,7 +184,7
ample file, and
playback seems correct.
On 2022-03-23 14:54, Steve Lhomme wrote:
Similar to how a frame is considered for referencing for the RefPicList array.
This will do the same for RefPicSetStCurrBefore, RefPicSetStCurrAfter and
RefPicSetLtCurr.
Fixes playback of http://www.gbbsoft.pl/!download/
The callback passed to CreateThread is not the same as with _beginthreadex().
This WinRT check could be removed if Win8 WinRT is not maintained
as _beginthreadex() is now available [1]
[1]
https://learn.microsoft.com/en-us/cpp/cppcx/crt-functions-not-supported-in-universal-windows-platform-apps?
his patch seems solid.From e8abeeff92f5d7b3b553acdb7595d40153cbec1e Mon Sep 17 00:00:00 2001
From: Steve Lhomme
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
Hi,
On 2022-08-02 16:19, Anton Khirnov wrote:
Why are you not resubmitting your original patch that stops copying
hwaccel_priv_data to the user-facing context?
It seemed more correct to me, since the user-facing context should never
see any hwaccel data. Or does it not fix the issue fully?
Th
Hi ffmmpeg developers,
As you may know, we are working hard on the Matroska specifications at
the IETF. We already got EBML as an RFC [1]. We are in the process of
finalizing the main Matroska document. Before we submit the document for
formal review before "final" publishing, we would like pe
On 2022-04-01 14:33, Steve Lhomme wrote:
Hi ffmmpeg developers,
As you may know, we are working hard on the Matroska specifications at
the IETF. We already got EBML as an RFC [1]. We are in the process of
finalizing the main Matroska document. Before we submit the document for
formal review
Hi,
For the record we have been running this in VLC for quite some time,
only for Intel hardware.
https://code.videolan.org/videolan/vlc/-/blob/3.0.x/contrib/src/ffmpeg/0001-avcodec-dxva2_hevc-add-support-for-parsing-HEVC-Rang.patch?ref_type=heads
https://code.videolan.org/videolan/vlc/-/blob/3
201 - 215 of 215 matches
Mail list logo
