Public bug reported:
We have a set of servers, equipped with Broadcom BCM57414 NetXtreme-E 10Gb/25Gb
RDMA Ethernet Controllers.
Booting Ubuntu 24.04 (on 6.8 kernel) on these machines leads to the
bnxt_re driver stalling during boot, and outputting the following kernel
log:
bnxt_en 0000:ab:00.0: QPLIB: bnxt_re_is_fw_stalled: FW STALL Detected.
cmdq[0xf]=0x3 waited (102721 > 100000) msec active 1
bnxt_en 0000:ab:00.0 bnxt_re0: Failed to modify HW QP
infiniband bnxt_re0: Couldn't change QP1 state to INIT: -110
infiniband bnxt_re0: Couldn't start port
bnxt_en 0000:ab:00.0 bnxt_re0: Failed to destroy HW QP
This causes systemd-udev-settle.service to fail:
udevadm[1212]: Timed out for waiting the udev queue being empty.
After this point, if the machine is PXE booting and/ or provisioning via MaaS
(which is the case), the provisioning basically fails.
The current workaround is to disable RDMA in the BIOS, thus avoiding loading
bnxt_en, I believe.
This behavior doesn't seem to affect Ubuntu 22.04 with 5.15 kernel.
The following Blog seems to explain this issue in great detail:
https://utcc.utoronto.ca/~cks/space/blog/linux/BroadcomNetworkDriverAndRDMA
** Affects: linux (Ubuntu)
Importance: Undecided
Status: New
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2099708
Title:
Broadcom RDMA over Converged Ethernet driver bnxt_re stalling on 24.04
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2099708/+subscriptions
--
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
