[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20100422225209.GT2524@linux.vnet.ibm.com>
Date: Thu, 22 Apr 2010 15:52:09 -0700
From: "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>
To: Mathieu Desnoyers <mathieu.desnoyers@...icios.com>
Cc: akpm@...ux-foundation.org, mingo@...e.hu,
linux-kernel@...r.kernel.org, laijs@...fujitsu.com,
dipankar@...ibm.com, josh@...htriplett.org, dvhltc@...ibm.com,
niv@...ibm.com, tglx@...utronix.de, peterz@...radead.org,
rostedt@...dmis.org, Valdis.Kletnieks@...edu, dhowells@...hat.com,
eric.dumazet@...il.com, adobriyan@...il.com, davem@...emloft.net
Subject: Re: [PATCH] rcu fix rcu head init on stack in rcutree_plugin.h
On Wed, Apr 21, 2010 at 09:23:09PM -0400, Mathieu Desnoyers wrote:
> [Paul]
> [...]
> And testing got me the following debugobjects splat
> [...]
> ODEBUG: object is on stack, but not annotated
> ------------[ cut here ]------------
> Badness at lib/debugobjects.c:294
> NIP: c0000000002c76f0 LR: c0000000002c76ec CTR: c00000000041ecd8
> REGS: c0000001de71b280 TRAP: 0700 Tainted: G W (2.6.34-rc3-autokern1)
> MSR: 8000000000029032 <EE,ME,CE,IR,DR> CR: 24000424 XER: 0000000f
> TASK = c0000001de7dca00[3695] 'arping' THREAD: c0000001de718000 CPU: 1
> GPR00: c0000000002c76ec c0000001de71b500 c00000000096c048 0000000000000034
> GPR04: 0000000000000001 c000000000063918 0000000000000000 0000000000000002
> GPR08: 0000000000000003 0000000000000000 c000000000086f68 c0000001de7dca00
> GPR12: 000000000000256d c0000000074e4200 0000000000000000 0000000000000000
> GPR16: 0000000000000000 0000000000000000 0000000000000000 00000000201b8f60
> GPR20: 00000000201b8f70 00000000201b8f48 0000000000000000 c0000000008766b8
> GPR24: c0000001de71b800 0000000000000001 c0000000008ad400 c000000001247478
> GPR28: c0000000e6abb8c0 c0000000e6abb8c0 c000000000904570 c000000001247470
> NIP [c0000000002c76f0] .__debug_object_init+0x314/0x40c
> LR [c0000000002c76ec] .__debug_object_init+0x310/0x40c
> Call Trace:
> [c0000001de71b500] [c0000000002c76ec] .__debug_object_init+0x310/0x40c (unreliable)
> [c0000001de71b5d0] [c00000000007d990] .rcuhead_fixup_activate+0x40/0xdc
> [c0000001de71b660] [c0000000002c6a7c] .debug_object_fixup+0x4c/0x74
> [c0000001de71b6f0] [c0000000000c5e54] .__call_rcu+0x3c/0x1d4
> [c0000001de71b790] [c0000000000c6050] .synchronize_rcu+0x4c/0x6c
> [c0000001de71b870] [c0000000004be218] .synchronize_net+0x10/0x24
> [c0000001de71b8e0] [c0000000005498c8] .packet_release+0x1d4/0x274
> [c0000001de71b990] [c0000000004ac1f0] .sock_release+0x54/0x124
> [c0000001de71ba20] [c0000000004ac9e4] .sock_close+0x34/0x4c
> [c0000001de71baa0] [c00000000012469c] .__fput+0x174/0x264
> [c0000001de71bb40] [c000000000120c54] .filp_close+0xb0/0xd8
> [c0000001de71bbd0] [c000000000065e70] .put_files_struct+0x1a8/0x314
> [c0000001de71bc70] [c000000000067e04] .do_exit+0x234/0x6f0
> [c0000001de71bd30] [c000000000068354] .do_group_exit+0x94/0xc8
> [c0000001de71bdc0] [c00000000006839c] .SyS_exit_group+0x14/0x28
> [c0000001de71be30] [c000000000008554] syscall_exit+0x0/0x40
> Instruction dump:
> 7f80b000 419e0030 2fa00000 e93e8140 380b0001 90090000 419e000c e87e8148
> 48000008 e87e8150 4bd9cb89 60000000 <0fe00000> 801c0010 2f800003 419e0024
>
> [Mathieu]
>
> Here is the fix.
Thank you very much, Mathieu!!! Queued for 2.6.35.
> Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@...icios.com>
> CC: "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>
> ---
> kernel/rcutree_plugin.h | 2 ++
> 1 file changed, 2 insertions(+)
>
> Index: linux.trees.git/kernel/rcutree_plugin.h
> ===================================================================
> --- linux.trees.git.orig/kernel/rcutree_plugin.h 2010-04-21 21:15:45.000000000 -0400
> +++ linux.trees.git/kernel/rcutree_plugin.h 2010-04-21 21:16:57.000000000 -0400
> @@ -515,11 +515,13 @@ void synchronize_rcu(void)
> if (!rcu_scheduler_active)
> return;
>
> + init_rcu_head_on_stack(&rcu.head);
> init_completion(&rcu.completion);
> /* Will wake me after RCU finished. */
> call_rcu(&rcu.head, wakeme_after_rcu);
> /* Wait for it. */
> wait_for_completion(&rcu.completion);
> + destroy_rcu_head_on_stack(&rcu.head);
> }
> EXPORT_SYMBOL_GPL(synchronize_rcu);
>
> --
> Mathieu Desnoyers
> Operating System Efficiency R&D Consultant
> EfficiOS Inc.
> http://www.efficios.com
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@...r.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists