[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Pine.LNX.4.64.0708181451130.5186@fbirervta.pbzchgretzou.qr>
Date: Sat, 18 Aug 2007 14:53:29 +0200 (CEST)
From: Jan Engelhardt <jengelh@...putergmbh.de>
To: Måns Rullgård <mans@...sr.com>
cc: linux-kernel@...r.kernel.org
Subject: Re: Panic with XFS on RHEL5 (2.6.18-8.1.8.el5)
On Aug 18 2007 13:31, Måns Rullgård wrote:
>>
>> BUG: unable to handle kernel paging request at virtual address b8af9d60
>> printing eip:
>> c0415974
>> *pde = 00000000
>> Oops: 0000 [#1]
>> SMP last sysfs file: /block/loop7/dev
>> Modules linked in: loop nfsd exportfs lockd nfs_acl iscsi_trgt(U)
>> autofs4 hidp nls_utf8 cifs ppdev rfcomm l2cap bluetooth vmnet(U)
>> vmmon(U) sunrpc ipv6 xfs(U) video sbs i2c_ec button battery asus_acpi
>> ac lp st sg floppy serio_raw intel_rng pcspkr e100 mii e7xxx_edac
>> i2c_i801 edac_mc i2c_core e1000 r8169 ide_cd cdrom parport_pc parport
>> dm_snapshot dm_zero dm_mirror dm_mod cciss mptspi mptscsih
>> scsi_transport_spi sd_mod scsi_mod mptbase ext3 jbd ehci_hcd ohci_hcd
>> uhci_hcd
>> CPU: 1
>> EIP: 0060:[<c0415974>] Tainted: P VLI
>> EFLAGS: 00010046 (2.6.18-8.1.8.el5 #1) EIP is at
>> smp_send_reschedule+0x3/0x53
>> eax: c213f000 ebx: c213f000 ecx: eef84000 edx: c213f000
>> esi: 00001086 edi: f668c000 ebp: f4f2fce8 esp: f4f2fc8c
>> ds: 007b es: 007b ss: 0068
>> Process crond (pid: 3146, ti=f4f2f000 task=f51faaa0 task.ti=f4f2f000)
>> Stack: 66d66b89 c041dc23 00000000 a9afbb0e fffffea5 01904500 00000000
>> 0000000f 00000000 00000001 00000001 c200c6e0 00000100 00000000
>> 00000069 00000180 018fc500 c200d240 00000003 00000292 f601efc0
>> f6027e00 00000000 00000050 Call Trace:
>> [<c041dc23>] try_to_wake_up+0x351/0x37b
>> [<f936884e>] xfsbufd_wakeup+0x28/0x49 [xfs]
>> [<c04572f9>] shrink_slab+0x56/0x13c
[...]
>
>Your Redhat kernel is probably built with 4k stacks and XFS+loop+ext3
>seems to be enough to overflow it.
I think we should include the vermagic string in oopses too,
so that the flags SMP, PREEMPT, RT, 4KSTACKS, mod_unload, etc. are shown
and the situation is a bit more apparent.
Jan
--
Powered by blists - more mailing lists