Same regression here on Ubuntu 24.04 with the HWE kernel
7.0.0-42.42~24.04.1 (linux-hwe-7.0): nine identical Dell PowerEdge nodes
with BCM57414 (firmware 233.0.195.0 / pkg 23.31.18.10, Intel IOMMU,
802.3ad bond with VLANs) all hit it within hours of a fresh deployment,
7.0.0-38.38~24.04.4 on the same hosts was fine. The faults start seconds
after the first VLAN traffic and leave both ports down; the pattern
matches CVE-2026-99081 ("bnxt_en: fix DMA mapping length for padded
small packets"), which is pending in 7.0.0-43 for resolute but not yet
built for linux-hwe-7.0 on noble. Workaround in use: pin 7.0.0-38.
kernel: DMAR: [DMA Read NO_PASID] Request device [0e:00.1] fault addr
0xf307b000 [fault reason 0x71] PTE Read access is not set
kernel: bnxt_en 0000:0e:00.0 eth0: Fatal firmware reset event, data1: 0x201,
data2: 0xda27
kernel: bond0: (slave eth0): link status definitely down, disabling slave
kernel: bnxt_en 0000:0e:00.0 eth0: hwrm req_type 0x23 seq id 0xad9 error 0xf
kernel: bnxt_en 0000:0e:00.0 eth0: NETDEV WATCHDOG: CPU: 2: transmit queue 11
timed out
kernel: bnxt_en 0000:0e:00.0 eth0: TX timeout detected, starting reset task!
kernel: bnxt_en 0000:0e:00.1 eth1: Abandoning msg {0x23 0xa6e} len: 0 due to
firmware status: 0x2000001
kernel: bnxt_en 0000:0e:00.1 eth1: TX timeout detected, starting reset task!
/proc/net/bonding/bond0: MII Status: down (both slaves)
** CVE added: https://cve.org/CVERecord?id=CVE-2026-99081
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2172450
Title:
bnxt_en: IOMMU DMA read faults followed by fatal firmware reset on
BCM57414 with 7.0.0-42-generic (7.0.0-29-generic is fine)
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2172450/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
