Following up on Alex Shi's suggestions...

hardlockup_panic=1 is now set. ath10k firmware version is
WLAN.RM.4.4.1-00309-QCARMSWPZ-1 - haven't yet tested 6.17.0-29.

I have attempted to reproduce the issue by running Docker container
crash loops alongside direct veth pair create/destroy (both via ip link
and raw rtnetlink) and WiFi ping stress (both custom bash scripts and C
programs). Despite generating hundreds of veth lifecycle events per
second with WiFi active, no ath10k_pci messages appeared in dmesg at all
- veth churn alone doesn't appear to trigger the peer stats request
path. The original crash required a container restarting under Docker's
restart policy for ~10 hours (608 restarts total).

The ath10k mailing list search yielded a directly related thread by Paul
Menzel (August 2025):
https://lists.infradead.org/pipermail/ath10k/2025-August/016464.html —
same firmware (WLAN.RM.4.4.1-00309), same chip (QCA6174 hw3.2), same
error sequence starting with timed out waiting peer stats info → wmi
command timeout, restarting hardware, across kernels 6.14–6.17-rc3. In
his case the driver eventually recovered; in my case the timeout handler
spin-waited on CPU6 with interrupts disabled and triggered a hard lockup
instead. Baochen Qiang responded that Qualcomm was working on obtaining
firmware dumps from the community, but the issue appeared unresolved as
of October 2025.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2152582

Title:
  ath10k firmware timeout causes hard lockup on CPU6 when veth
  interfaces cycle repeatedly (Docker restart loop)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux-hwe-6.17/+bug/2152582/+subscriptions


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

Reply via email to