- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
In Linux, when decoding a HEVC stream with missing references, MFXVideoDECODE_DecodeFrameAsync eventually fails with MFX_ERR_GPU_HANG, after having previously stalled for ~7 seconds with (MFX_WRN_DEVICE_BUSY). The media-driver call vaSyncSurface is responsible for this abnormal delay.
CPU model: 12th Gen Intel(R) Core(TM) i5-1235U (also reproduced on a i7 TigerLake Cpu/Gpu)
OS: Ubuntu 22.04 LTS
Kernel: 5.18.0-051800-generic
OneVPL runtime 2022Q3: 22.5.4
i915 kernel messages:
[1298307.045332] i915 0000:00:02.0: [drm] Resetting vcs0 for preemption time out
[1298307.045393] i915 0000:00:02.0: [drm] sample_decode[141421] context reset due to GPU hang
[1298307.303391] i915 0000:00:02.0: [drm] GPU HANG: ecode 12:4:28fffffd, in sample_decode [141421]
Repro command line using the sample_decode app:
sample_decode h265 -i ./missing_refs.h265 -o /dev/null
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi,
Thank you for posting in Intel communities.
Thanks for reporting this issue. We were able to reproduce it and we have informed the development team about it.
Thank You!
Regards
Varad
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi,
Good day to you!
Thank you for your patience.
After investigating with the development team, we found that this behavior is expected because the 40th frame of this clip has missed reference frames. Intel HW decoder cannot decode it rightly.
Since this is not a bug we are closing this thread. If you need any additional information, please post a new question as this thread will no longer be monitored by Intel.
Thank You!
Regards
Varad
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page