Just now I did the following experiment: I booted Linux 6.10 without
"pcie_aspm=off", then I booted it with "pcie_aspm=off", then again
without, etc. In total I did nearly 13 boots with "pcie_aspm=off" and
nearly 13 boots without "pcie_aspm=off". All these boots were performed
automatically by script using kexec. No hibernation was involved.

System was booted, then "sleep 60" was executed, then kexec booted
again.

Results are so:
- All boots with "pcie_aspm=off" display no errors
- 3 out of 13 boots without "pcie_aspm=off" display no errors. Other 10 boots 
show these "corrected error received" errors (at least one time). But number of 
these messages is small (say, 5 messages in one boot, i. e. in 60 seconds). (My 
practice shows that if you really want to see a LOT of these messages, you 
should do hibernate and then resume.)

So, it seems that:
- Debugging using "kexec" works
- You don't need hibernate to reproduce this bug (but number of error messages 
will be small)
- "pcie_aspm=off" makes messages disappear
- There is a stable way to reproduce bug: boot 13 times using kexec. 10 times 
out of 13 you will see at least one message

Ask me any questions. I'm ready to do other experiments, send any kind
of logs, etc

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1521173

Title:
  AER: Corrected error received: id=00e0

To manage notifications about this bug go to:
https://bugs.launchpad.net/linux/+bug/1521173/+subscriptions


-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to