I get this crash running the chelsio cxgb3 module on 2.6.22-rc5. This
is regression. I believe Divy from Chelsio has already posted a fix for
this. This needs to be in 2.6.22...
Jeff do you have this fix queued for 2.6.22?
I think Divy's patches are here:
http://lkml.org/lkml/2007/5/30/291
and here:
http://lkml.org/lkml/2007/6/5/211
Thanks,
Steve.
----
Unable to handle kernel NULL pointer dereference at 0000000000000220 RIP:
[<ffffffff881f34f9>] :cxgb3:process_responses+0x6a3/0xa06
PGD 73d98067 PUD 0
Oops: 0002 [1] SMP
CPU 0
Modules linked in: iw_cxgb3 cxgb3 parport_pc lp parport autofs4 sunrpc
rdma_ucm rdma_cm iw_cm ib_addr dm_multipath button battery ac i2c_i801
i2c_core ib_mthca ib_ipoib ib_umad ib_ucm ib_uverbs ib_cm ib_sa ib_mad
ib_core e1000 floppy dm_snapshot dm_zero dm_mirror ext3 jbd dm_mod
ata_piix libata sd_mod scsi_mod
Pid: 0, comm: swapper Not tainted 2.6.22-rc5-g802c7fc4 #1
RIP: 0010:[<ffffffff881f34f9>] [<ffffffff881f34f9>]
:cxgb3:process_responses+0x6a3/0xa06
RSP: 0018:ffffffff8059be00 EFLAGS: 00010206
RAX: 0000000100012ecf RBX: ffff81006f9c77c0 RCX: ffff81006f9c8a00
RDX: 0000000000000000 RSI: ffff8100702b9046 RDI: ffff81006f9c77c0
RBP: ffff81006f9c8a02 R08: 0000000000000000 R09: ffff81006f9c77c0
R10: 0000000000000000 R11: ffffffff804d2080 R12: ffff81006f98c480
R13: ffff810070350000 R14: 0000000000000046 R15: 0000000000000000
FS: 0000000000000000(0000) GS:ffffffff80516000(0000) knlGS:0000000000000000
CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b
CR2: 0000000000000220 CR3: 0000000073d97000 CR4: 00000000000006e0
Process swapper (pid: 0, threadinfo ffffffff8053e000, task ffffffff804d2080)
Stack: 0000000000000000 ffff8100026374b0 0000000000000000 0000002b00000000
0000000283000000 0000000000000001 0000004000000000 ffff8100701c0100
0000004002637440 ffff81006f98c000 ffff810002637440 0000000000000000
Call Trace:
<IRQ> [<ffffffff8041482e>] __lock_text_start+0x6/0x10
[<ffffffff80223c02>] run_rebalance_domains+0x3f4/0x434
[<ffffffff881f3892>] :cxgb3:napi_rx_handler+0x36/0xb6
[<ffffffff803b3559>] net_rx_action+0xb1/0x167
[<ffffffff8022e839>] __do_softirq+0x50/0xbb
[<ffffffff8020a42c>] call_softirq+0x1c/0x28
[<ffffffff8020b5e9>] do_softirq+0x2e/0x95
[<ffffffff8020b5a3>] do_IRQ+0xb7/0xcf
[<ffffffff80207bc5>] mwait_idle+0x0/0x42
Steve Wise wrote:
>
> I get this crash running the chelsio cxgb3 module on 2.6.22-rc5. This
> is regression. I believe Divy from Chelsio has already posted a fix for
> this. This needs to be in 2.6.22...
>
> Jeff do you have this fix queued for 2.6.22?
>
> I think Divy's patches are here:
>
> http://lkml.org/lkml/2007/5/30/291
>
> and here:
>
> http://lkml.org/lkml/2007/6/5/211
hmmm, no idea where these got to. They are now in my queue, rescued.
Jeff