[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20080317222846.7B58.KOSAKI.MOTOHIRO@jp.fujitsu.com>
Date: Mon, 17 Mar 2008 22:32:15 +0900
From: KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>
To: linux-ia64@...r.kernel.org, Fenghua Yu <fenghua.yu@...el.com>,
Tony Luck <tony.luck@...el.com>
Cc: kosaki.motohiro@...fujitsu.com,
LKML <linux-kernel@...r.kernel.org>,
Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [2.6.25-rc5-mm1][regression?] BUG: spinlock wrong CPU, modprobe
Hi
if revert following patch
[IA64] Multiple outstanding ptc.g instruction support
then this problem doesn't reproduce.
I attached that revert patch.
Thanks!
> Hi
>
> I found strange message at 2.6.25-rc5-mm1 booting.
> and it is always reproduce.
>
> and, Eric Piel's spinlock related patch(below URL) can't fixed my problem.
> http://marc.info/?l=linux-kernel&m=120561027011749&w=2
>
> Is this known bug?
> or should i do bisect?
>
> -------------------------------------------------------------------
> BUG: spinlock wrong CPU on CPU#3, udevd/2495
> lock: a040000058246510, .magic: dead4ead, .owner: udevd/2495, .owner_cpu: 2
>
> Call Trace:
> [<a000000100015f00>] show_stack+0x80/0xa0
> sp=e00001609262fbd0 bsp=e000016092620fd8
> [<a000000100015f50>] dump_stack+0x30/0x60
> sp=e00001609262fda0 bsp=e000016092620fc0
> [<a0000001004674a0>] spin_bug+0x1a0/0x1c0
> sp=e00001609262fda0 bsp=e000016092620f90
> [<a000000100467590>] _raw_spin_unlock+0xd0/0x120
> sp=e00001609262fda0 bsp=e000016092620f60
> [<a00000010082bfe0>] _spin_unlock+0x20/0x40
> sp=e00001609262fda0 bsp=e000016092620f40
> [<a000000100138100>] do_wp_page+0x500/0xfe0
> sp=e00001609262fda0 bsp=e000016092620ed8
> [<a00000010013c560>] handle_mm_fault+0xe80/0x10e0
> sp=e00001609262fda0 bsp=e000016092620e50
> [<a00000010006c140>] ia64_do_page_fault+0x600/0xa40
> sp=e00001609262fda0 bsp=e000016092620df0
> [<a00000010000abe0>] ia64_leave_kernel+0x0/0x270
> sp=e00001609262fe30 bsp=e000016092620df0
Download attachment "revert-PALO.patch" of type "application/octet-stream" (8977 bytes)
Powered by blists - more mailing lists