Doing dd to xfs filesystem produced the following.
Rebuilding with debug pagealloc and a few more in case it will help
tracking it down.
The kernel also has the udm2 patches applied and the xfs filesystem is
on a dm_multipath:ed disk.
It so far looks repeatable.
Please CC me.
------------[ cut here ]------------
kernel BUG at mm/page_alloc.c:575!
invalid operand: 0000 [#1]
SMP
Modules linked in: xfs dm_round_robin md5 ipv6 uhci_hcd ohci_hcd ehci_hcd usbcore qla2300 qla2xxx scsi_transport_fc e1000 i2c_i801 i2c_core hw_random floppy rtc dm_multipath dm_mod autofs4 sg ext3 jbd sd_mod aic79xx scsi_mod
CPU: 15
EIP: 0060:[<c0138d63>] Not tainted VLI
EFLAGS: 00010002 (2.6.10-rc2-udm2-hpc2n-p4-smp)
EIP is at buffered_rmqueue+0x175/0x1f4
eax: 00000001 ebx: c029f480 ecx: 00001000 edx: 00049878
esi: c1930f00 edi: 00000000 ebp: f2568f04 esp: f2568ee0
ds: 007b es: 007b ss: 0068
Process (pid: 0, threadinfo=f2568000 task=ec889000)
Stack: c029ed00 c1930f00 00000000 00000000 c029f500 c1930f00 c029ed00 baadf00d
00000000 f2568f54 c01390f4 c029ed00 00000000 00000200 f256a000 f256a012
f256a012 00000001 00000001 00000001 ec889000 00000000 c029f600 00000000
Call Trace:
[<c0103437>] show_stack+0x80/0x96
[<c01035ce>] show_registers+0x15f/0x1c3
[<c01037c5>] die+0xfa/0x180
[<c0103ccf>] do_invalid_op+0x102/0x104
[<c01030cb>] error_code+0x2b/0x30
[<c01390f4>] __alloc_pages+0x312/0x333
[<c013913c>] __get_free_pages+0x27/0x40
[<c013c2f5>] kmem_getpages+0x20/0xce
[<c013cf9a>] cache_grow+0xab/0x14a
[<c013d1a7>] cache_alloc_refill+0x16e/0x20f
[<c013d411>] kmem_cache_alloc+0x4a/0x4c
=======================
Unable to handle kernel paging request at virtual address 00060030
printing eip:
c0103388
*pde = 00000000
Oops: 0000 [#2]
SMP
Modules linked in: xfs dm_round_robin md5 ipv6 uhci_hcd ohci_hcd ehci_hcd usbcore qla2300 qla2xxx scsi_transport_fc e1000 i2c_i801 i2c_core hw_random floppy rtc dm_multipath dm_mod autofs4 sg ext3 jbd sd_mod aic79xx scsi_mod
CPU: 15
EIP: 0060:[<c0103388>] Not tainted VLI
EFLAGS: 00010096 (2.6.10-rc2-udm2-hpc2n-p4-smp)
EIP is at show_trace+0x7b/0xaa
eax: 00060ffd ebx: f2569000 ecx: 00000000 edx: 00000086
esi: f2569000 edi: 00060000 ebp: f2568d50 esp: f2568d38
ds: 007b es: 007b ss: 0068
Process (pid: 0, threadinfo=f2568000 task=ec889000)
Stack: c0270c6f c013d411 00060ffd f2568f40 00000018 00000000 f2568d6c c0103437
00000000 f2568ee0 f2568000 f2568eac 00000000 f2568da4 c01035ce 00000000
f2568ee0 00000000 f2568000 ec889000 00010002 c029a242 00680d68 00000001
Call Trace:
[<c0103437>] show_stack+0x80/0x96
[<c01035ce>] show_registers+0x15f/0x1c3
[<c01037c5>] die+0xfa/0x180
[<c01115ba>] do_page_fault+0x316/0x63f
[<c01030cb>] error_code+0x2b/0x30
[<c0103437>] show_stack+0x80/0x96
[<c01035ce>] show_registers+0x15f/0x1c3
[<c01037c5>] die+0xfa/0x180
[<c0103ccf>] do_invalid_op+0x102/0x104
[<c01030cb>] error_code+0x2b/0x30
[<c01390f4>] __alloc_pages+0x312/0x333
[<c013913c>] __get_free_pages+0x27/0x40
[<c013c2f5>] kmem_getpages+0x20/0xce
[<c013cf9a>] cache_grow+0xab/0x14a
[<c013d1a7>] cache_alloc_refill+0x16e/0x20f
[<c013d411>] kmem_cache_alloc+0x4a/0x4c
=======================
Unable to handle kernel paging request at virtual address 00060030
printing eip:
c0103388
*pde = 00000000
Recursive die() failure, output suppressed
<1>Unable to handle kernel paging request at virtual address 21a9214f
printing eip:
c0113226
*pde = 00000000
Recursive die() failure, output suppressed
--
Ake Sandgren, HPC2N, Umea University, S-90187 Umea, Sweden
Internet: [email protected] Phone: +46 90 7866134 Fax: +46 90 7866126
Mobile: +46 70 7716134 WWW: http://www.hpc2n.umu.se