Xen devel - Jul 2005 - kernel panic: fatal exception in interrupt

Hi everyone,

I''m tracking xen-unstable (hg) -- this is using a checkout from 2 days
back. Its a P4, dual processor (but we run nosmp), has a Broadcom 5782
NetXtreme Gigabit ethernet card.

A VM boots up fine, I can ping it, SSH into it. I got this exception
while trying to run some benchmarks using httperf (the VM is running
Apache). I can consistent crash the kernel, but it doesn''t always
happen at the same time -- sometimes after 2 seconds, sometimes after
5 seconds. The webserver is not loaded (this particular dump is from a
run where I was sending a measly 50 req/sec).

Has anyone seen something similar recently?

Diwaker

=======================================Unable to handle kernel paging request at
virtual address 74736f48
 printing eip:
c03a92ff
*pde = ma 00000000 pa 55555000
Oops: 0000 [#1]
PREEMPT
Modules linked in: iptable_filter ip_tables video thermal processor
fan button battery ac
CPU:    0
EIP:    0061:[<c03a92ff>]    Not tainted VLI
EFLAGS: 00010293   (2.6.12-xen0)
EIP is at skb_release_data+0x4f/0xec
eax: 65707474   ebx: 00000000   ecx: ce497080   edx: 74736f48
esi: ce468280   edi: ce468280   ebp: c059a800   esp: c051fc74
ds: 007b   es: 007b   ss: 0069
Process swapper (pid: 0, threadinfo=c051e000 task=c0488680)
Stack: c04d7180 ce468280 ce468280 c03a9445 ce468280 06be6012 c03a93ac ce468280
       ced87620 ced87620 c02c1009 ce468280 0000001d 00000000 06be6012 00000042
       00000001 00000013 00000042 0000e497 ced87620 ced87400 99ebd6a5 000000e1
Call Trace:
 [<c03a9445>] __kfree_skb+0x83/0x136
 [<c03a93ac>] kfree_skbmem+0x10/0x26
 [<c02c1009>] net_rx_action+0x234/0x4e9
 [<c03fcd71>] br_nf_pre_routing_finish+0x139/0x39a
 [<c03f9240>] br_handle_frame_finish+0x0/0x127
 [<c03f8657>] br_forward_finish+0x0/0x66
 [<c03f92d2>] br_handle_frame_finish+0x92/0x127
 [<c03fcd71>] br_nf_pre_routing_finish+0x139/0x39a
 [<c03f9240>] br_handle_frame_finish+0x0/0x127
 [<c03b9c06>] nf_iterate+0x58/0x84
 [<c03fcc38>] br_nf_pre_routing_finish+0x0/0x39a
 [<c03fcc38>] br_nf_pre_routing_finish+0x0/0x39a
 [<c03ba020>] nf_hook_slow+0x76/0x12a
 [<c03fcc38>] br_nf_pre_routing_finish+0x0/0x39a
 [<c03f9240>] br_handle_frame_finish+0x0/0x127
 [<c03fd616>] br_nf_pre_routing+0x302/0x471
 [<c03fcc38>] br_nf_pre_routing_finish+0x0/0x39a
 [<c03b9c06>] nf_iterate+0x58/0x84
 [<c03f9240>] br_handle_frame_finish+0x0/0x127
 [<c03f9240>] br_handle_frame_finish+0x0/0x127
 [<c03ba020>] nf_hook_slow+0x76/0x12a
 [<c03f9240>] br_handle_frame_finish+0x0/0x127
 [<c03f9517>] br_handle_frame+0x1b0/0x249
 [<c03f9240>] br_handle_frame_finish+0x0/0x127
 [<c03af95d>] netif_receive_skb+0x118/0x235
 [<c029f338>] tg3_rx+0x2c5/0x48a
 [<c029b7ab>] tg3_restart_ints+0x25/0x97
 [<c029f6b7>] tg3_poll+0x1ba/0x2c8
 [<c0122cac>] tasklet_action+0x74/0x10d
 [<c012286c>] __do_softirq+0x6c/0xf5
 [<c0122959>] do_softirq+0x64/0x77
 [<c0122a25>] irq_exit+0x36/0x38
 [<c010e332>] do_IRQ+0x22/0x28
 [<c01063fd>] evtchn_do_upcall+0x60/0x86
 [<c0109d7c>] hypervisor_callback+0x2c/0x34
 [<c0107ab0>] xen_idle+0x32/0x6b
 [<c0107b1b>] cpu_idle+0x32/0x47
 [<c0105024>] rest_init+0x24/0x37
 [<c0105068>] init+0x0/0x110
 [<c05207cf>] start_kernel+0x1b0/0x206
 [<c0520327>] unknown_bootoption+0x0/0x1f6
Code: 00 f7 d8 89 c2 0f c1 01 01 d0 85 c0 74 0c 83 c4 10 5b 5e c3 8b
8e a8 00 00 00 8b 41 04 85 c0 74 32 31 db 39 c3 eb 2a 8b 54 d9 10 <8b>
02 f6 c4 08 75 19 8b 42 04 83 c0 01 74 59 83 42 04 ff 0f 98
 <0>Kernel panic - not syncing: Fatal exception in interrupt
=======================================-- 
Web/Blog/Gallery: http://floatingsun.net

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

On 21 Jul 2005, at 01:55, Diwaker Gupta wrote:
> A VM boots up fine, I can ping it, SSH into it. I got this exception
> while trying to run some benchmarks using httperf (the VM is running
> Apache). I can consistent crash the kernel, but it doesn''t always
> happen at the same time -- sometimes after 2 seconds, sometimes after
> 5 seconds. The webserver is not loaded (this particular dump is from a
> run where I was sending a measly 50 req/sec).
>
> Has anyone seen something similar recently?
This got fixed yesterday. :-)

  -- Keir


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

With a huge performance boost too!  Good job guys.

David


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel