On Thu, Aug 20, 2009 at 10:30:46PM -0400, Trond Myklebust wrote:
> On Fri, 2009-08-21 at 09:42 +1000, Stephen Rothwell wrote:
> > Hi Trond,
> >
> > Booting next-20090820 on three different PowerPC machines get the
> > following OOPS:
> >
> > calling .init_nfs_fs+0x0/0x184 @ 1
> > Unable to handle kernel paging request for data at address 0x00000000
> > Faulting instruction address: 0xc00000000013be00
> > Oops: Kernel access of bad area, sig: 11 [#1]
> > SMP NR_CPUS=128 NUMA pSeries
> > Modules linked in:
> > NIP: c00000000013be00 LR: c00000000013bd00 CTR: c00000000056f098
> > REGS: c00000007d2db5c0 TRAP: 0300 Not tainted (2.6.31-rc6-autokern1)
> > MSR: 8000000000009032 <EE,ME,IR,DR> CR: 48000028 XER: 00000005
> > DAR: 0000000000000000, DSISR: 0000000040000000
> > TASK = c0000000410ca000[1] 'swapper' THREAD: c00000007d2d8000 CPU: 1
> > GPR00: c00000000013bd00 c00000007d2db840 c000000000b84e98 0000000000000001
> > GPR04: c000000000a831e8 c0000000410ca948 0000000000000002 c0000000410ca948
> > GPR08: 0000000000000025 0000000000000000 ef7bdef7bdef7bdf 0000000009ac4000
> > GPR12: 0000000088000084 c000000000bd4400 0000000000000000 0000000003000000
> > GPR16: c000000000720608 c00000000071ed80 0000000000000000 00000000003e7800
> > GPR20: 000000000382de28 c00000000082de28 000000000382e098 c00000000082e098
> > GPR24: 0000000000000000 c000000000b25c58 c000000000b25c40 c000000000ac9d18
> > GPR28: c000000000b7ba40 fffffffffffffe10 c000000000ae5e70 0000000000000000
> > NIP [c00000000013be00] .sget+0x14c/0x418
> > LR [c00000000013bd00] .sget+0x4c/0x418
> > Call Trace:
> > [c00000007d2db840] [c00000000013bd00] .sget+0x4c/0x418 (unreliable)
> > [c00000007d2db8f0] [c00000000013cca8] .get_sb_single+0x4c/0x114
> > [c00000007d2db9a0] [c00000000056f0b8] .rpc_get_sb+0x20/0x38
> > [c00000007d2dba20] [c00000000013c54c] .vfs_kern_mount+0x80/0xf8
> > [c00000007d2dbac0] [c00000000015d434] .simple_pin_fs+0x74/0x130
> > [c00000007d2dbb60] [c000000000570734] .rpc_get_mount+0x2c/0x54
> > [c00000007d2dbbe0] [c00000000023ffec] .nfs_cache_register+0x28/0xc0
> > [c00000007d2dbd10] [c00000000023fa78] .nfs_dns_resolver_init+0x1c/0x34
> > [c00000007d2dbd90] [c000000000813fac] .init_nfs_fs+0x1c/0x184
> > [c00000007d2dbe10] [c0000000000094bc] .do_one_initcall+0x90/0x1b0
> > [c00000007d2dbf00] [c0000000007f3c98] .kernel_init+0x1f4/0x270
> > [c00000007d2dbf90] [c0000000000268f0] .kernel_thread+0x54/0x70
> > Instruction dump:
> > 48445fad 60000000 387d0070 4bf4f7a9 60000000 7fa3eb78 4bfff911 48442e89
> > 60000000 4bffff04 e93d01f0 3ba9fe10 <e81d01f0> 2fa00000 419e0008 7c00022c
> > ---[ end trace 561bb236c800851f ]---
> > Kernel panic - not syncing: Attempted to kill init!
> > Call Trace:
> > [c00000007d2db220] [c000000000010228] .show_stack+0x70/0x184 (unreliable)
> > [c00000007d2db2d0] [c000000000067c40] .panic+0x80/0x1b4
> > [c00000007d2db370] [c00000000006c3cc] .do_exit+0x84/0x6fc
> > [c00000007d2db430] [c000000000024950] .die+0x24c/0x27c
> > [c00000007d2db4d0] [c0000000000328e0] .bad_page_fault+0xb8/0xd4
> > [c00000007d2db550] [c0000000000051dc] handle_page_fault+0x3c/0x74
> > --- Exception: 300 at .sget+0x14c/0x418
> > LR = .sget+0x4c/0x418
> > [c00000007d2db8f0] [c00000000013cca8] .get_sb_single+0x4c/0x114
> > [c00000007d2db9a0] [c00000000056f0b8] .rpc_get_sb+0x20/0x38
> > [c00000007d2dba20] [c00000000013c54c] .vfs_kern_mount+0x80/0xf8
> > [c00000007d2dbac0] [c00000000015d434] .simple_pin_fs+0x74/0x130
> > [c00000007d2dbb60] [c000000000570734] .rpc_get_mount+0x2c/0x54
> > [c00000007d2dbbe0] [c00000000023ffec] .nfs_cache_register+0x28/0xc0
> > [c00000007d2dbd10] [c00000000023fa78] .nfs_dns_resolver_init+0x1c/0x34
> > [c00000007d2dbd90] [c000000000813fac] .init_nfs_fs+0x1c/0x184
> > [c00000007d2dbe10] [c0000000000094bc] .do_one_initcall+0x90/0x1b0
> > [c00000007d2dbf00] [c0000000007f3c98] .kernel_init+0x1f4/0x270
> > [c00000007d2dbf90] [c0000000000268f0] .kernel_thread+0x54/0x70
> > Rebooting in 180 seconds..-- 0:conmux-control -- time-stamp -- Aug/20/09 19:25:14 --
> >
> > It may not be NFS changes ... there were just a few changes in the nfs
> > tree between next-20090819 and next-20090820.
> >
> Hi Stephen,
>
> Yes, that sounds like the bug that Bruce hit earlier today. I strongly
> suspect that it is due to the fact that you both compiled NFS+sunrpc
> into the main kernel, and that the NFS init routine is being called
> before the sunrpc init routine.
>
> Could both you and Bruce check if the following patch fixes the problem?
Yep, that boots for me, thanks.
--b.
>
> Cheers
> Trond
> ----------------------------------------------------------------
> From: Trond Myklebust <Trond.Myklebust@xxxxxxxxxx>
> SUNRPC: Ensure that sunrpc gets initialised before nfs, lockd, etc...
>
> We can oops if rpc_pipefs isn't properly initialised before we start to set
> up objects that depend upon it.
>
> Signed-off-by: Trond Myklebust <Trond.Myklebust@xxxxxxxxxx>
> ---
>
> net/sunrpc/sunrpc_syms.c | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
>
> diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c
> index adaa819..8cce921 100644
> --- a/net/sunrpc/sunrpc_syms.c
> +++ b/net/sunrpc/sunrpc_syms.c
> @@ -69,5 +69,5 @@ cleanup_sunrpc(void)
> rcu_barrier(); /* Wait for completion of call_rcu()'s */
> }
> MODULE_LICENSE("GPL");
> -module_init(init_sunrpc);
> +fs_initcall(init_sunrpc); /* Ensure we're initialised before nfs */
> module_exit(cleanup_sunrpc);
>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-next" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
