loading Chelsio cxgb4 driver during boot leads to kernel oops
Issue
A kernel oops happens when loading Chelsio cxgb4 driver on RHEL7
The call trace is :
[ 91.636713] Call Trace:
[ 91.639176] [<ffffffff812c2e3b>] string.isra.5+0x3b/0xf0
[ 91.644576] [<ffffffff812c4371>] vsnprintf+0x201/0x6a0
[ 91.649812] [<ffffffff814fdfff>] ? nlmsg_notify+0x4f/0xb0
[ 91.655301] [<ffffffff812c4996>] sprintf+0x56/0x80
[ 91.660196] [<ffffffffa0495354>] init_one+0xb64/0x1310 [cxgb4]
[ 91.666123] [<ffffffff8101a603>] ? native_sched_clock+0x13/0x80
[ 91.672146] [<ffffffff812eaf45>] local_pci_probe+0x45/0xa0
[ 91.677714] [<ffffffff8107af14>] work_for_cpu_fn+0x14/0x20
[ 91.683287] [<ffffffff8107e02b>] process_one_work+0x17b/0x460
[ 91.689115] [<ffffffff8107ef73>] worker_thread+0x293/0x400
[ 91.694680] [<ffffffff8107ece0>] ? rescuer_thread+0x400/0x400
[ 91.700529] [<ffffffff81085aef>] kthread+0xcf/0xe0
[ 91.705406] [<ffffffff81085a20>] ? kthread_create_on_node+0x140/0x140
[ 91.711949] [<ffffffff815f206c>] ret_from_fork+0x7c/0xb0
[ 91.717351] [<ffffffff81085a20>] ? kthread_create_on_node+0x140/0x140
[ 91.723871] Code: c0 01 80 38 00 75 f7 48 29 f8 5d c3 31 c0 5d c3 66
66 66 66 66 2e 0f 1f 84 00 00 00 00 00 55 48 85 f6 48 8d 4e ff 48 89 e5
74 2a <80> 3f 00 74 25 48 89 f8 31 d2 eb 10 0f 1f 80 00 00 00 00 48 83
[ 91.743846] RIP [<ffffffff812c0a3d>] strnlen+0xd/0x40
[ 91.748997] RSP <ffff8808534d3bf0>
[ 91.752490] CR2: 0000123c000011fc
The machine boots when blacklisting modules cxbg4, csiostor
(modprobe.blacklist=cxgb4 modprobe.blacklist=csiostor).
The same card works OK and the machine boots when a newer version of the
Chelsio drivers is used (2.9.1.0).
Environment
- Red Hat Entreprise Linux (RHEL) 7.0
- kernel 3.10.0-123.el7.x86_64
- NIC: Chelsio T580-LP-CR
- driver : cxgb4 version : 2.0.0-ko
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.