Crash in [@ shmget]
Categories
(Core :: Security: Process Sandboxing, defect, P5)
Tracking
()
People
(Reporter: gsvelto, Unassigned)
References
Details
(Keywords: crash)
Crash Data
This bug is for crash report bp-eb80de76-2e3d-4056-917a-70e5a0200615.
Top 4 frames of crashing thread:
0 libc.so.6 shmget /usr/src/debug/glibc-2.31/sysvipc/../sysdeps/unix/sysv/linux/shmget.c:31
1 iHD_drv_video.so OsContextSpecific::CreateIPC /usr/src/debug/intel-media-driver-20.1.1-1.fc32.x86_64/media_driver/linux/common/os/mos_context_specific.cpp:281
2 iHD_drv_video.so OsContextSpecific::Init /usr/src/debug/intel-media-driver-20.1.1-1.fc32.x86_64/media_driver/linux/common/os/mos_context_specific.cpp:497
3 iHD_drv_video.so DdiMedia__Initialize /usr/src/debug/intel-media-driver-20.1.1-1.fc32.x86_64/media_driver/linux/common/ddi/media_libva.cpp:1697
This appears to be the Intel video acceleration drivers acting funny. The crash is a SIGSYS with the 0x40 value which should correspond to semget but somehow the stack trace shows it crashing in shmget. It's possible that being simple syscall funnels the functions have identical implementations in glibc and thus the debuginfo was merged so the stack walker mistakes one for the other.
|
||
Comment 1•4 years ago
|
||
We think this is from a non-default config.
|
Reporter | |
Comment 2•4 years ago
|
||
Yeah, the video library comes from the RPMFusion non-free repositories.
|
||
Comment 3•4 years ago
|
||
(In reply to Gian-Carlo Pascutto [:gcp] from comment #1)
We think this is from a non-default config.
Specifically, this feature is preffed off.
|
||
Comment 4•4 years ago
|
||
Closing because no crashes reported for 12 weeks.
Description
•