Open dseomn opened 2 years ago
@dseomn as per our documentation of mPCIe modules: https://github.com/pcengines/apu2-documentation/blob/master/docs/mpcie_modules.md you may want to:
amd_iommu=off
to the kernel cmdline ORThen see if it helps
Thanks for the link, I just added amd_iommu=off
. I'll update here if it panics again.
It panicked again, this time with amd_iommu=off
.
Null pointer dereference: looks like a kernel bug in ath driver - please report to the corresponding kernel maintainer. In your case - Debian kernel - the right place is the Debain Kernel Team.
Note that HW vendors usually neither have the capacity nor the knowledge to give SW or OS support.
PCEngines firmware version v4.14.0.5
APU variant apu2e4
OS and OS Version Debian GNU/Linux 10 (buster)
Affected component(s), peripheral(s) or functionality The entire system crashes.
Brief summary The kernel panics occasionally. The watchdog still works though, so it gets rebooted and then works.
How reproducible I think it happens a bit less than once per week.
How to reproduce I haven't noticed any pattern of when it happens.
Expected behavior The kernel shouldn't panic.
Actual behavior The kernel panics. Here's a log from the serial console the most recent time it happened:
Solutions you've tried The watchdog timer turns this from a major issue into a tolerable one.