linux-kernel - Re: [PATCH 4/6] Use raw_cpu_write for initialization of per cpu refcount.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20131016084322.GR10651@twins.programming.kicks-ass.net>
Date:	Wed, 16 Oct 2013 10:43:22 +0200
From:	Peter Zijlstra <peterz@...radead.org>
To:	Christoph Lameter <cl@...ux.com>
Cc:	Tejun Heo <tj@...nel.org>, akpm@...uxfoundation.org,
	rostedt@...dmis.org, linux-kernel@...r.kernel.org,
	Ingo Molnar <mingo@...nel.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	Rusty Russell <rusty@...tcorp.com.au>
Subject: Re: [PATCH 4/6] Use raw_cpu_write for initialization of per cpu
 refcount.

On Tue, Oct 15, 2013 at 12:47:26PM -0500, Christoph Lameter wrote:
> The initialization of a structure is not subject to synchronization.
> The use of __this_cpu would trigger a false positive with the
> additional preemption checks for __this_cpu ops.
> 
> So simply disable the check through the use of raw_cpu ops.
> 
> Trace:
> 
> [    0.668066] __this_cpu_write operation in preemptible [00000000] code: modprobe/286
> [    0.668108] caller is __this_cpu_preempt_check+0x38/0x60
> [    0.668111] CPU: 3 PID: 286 Comm: modprobe Tainted: GF            3.12.0-rc4+ #187
> [    0.668112] Hardware name: FUJITSU CELSIUS W530 Power/D3227-A1, BIOS V4.6.5.4 R1.10.0 for D3227-A1x 09/16/2013
> [    0.668113]  0000000000000003 ffff8807edda1d18 ffffffff816d5a57 ffff8807edda1fd8
> [    0.668117]  ffff8807edda1d48 ffffffff8137359c ffff8807edda1ef8 ffffffffa0002178
> [    0.668121]  ffffc90000067730 ffff8807edda1e48 ffff8807edda1d88 ffffffff813735f8
> [    0.668124] Call Trace:
> [    0.668129]  [<ffffffff816d5a57>] dump_stack+0x4e/0x82
> [    0.668132]  [<ffffffff8137359c>] check_preemption_disabled+0xec/0x110
> [    0.668135]  [<ffffffff813735f8>] __this_cpu_preempt_check+0x38/0x60
> [    0.668139]  [<ffffffff810c24fd>] load_module+0xcfd/0x2650
> [    0.668143]  [<ffffffff816dd922>] ? page_fault+0x22/0x30
> [    0.668146]  [<ffffffff810c3ef6>] SyS_init_module+0xa6/0xd0
> [    0.668150]  [<ffffffff816e4fd3>] tracesys+0xe1/0xe6
> 
> 
> 
> Cc: Rusty Russell <rusty@...tcorp.com.au>
> Signed-off-by: Christoph Lameter <cl@...ux.com>
> 
> Index: linux/kernel/module.c
> ===================================================================
> --- linux.orig/kernel/module.c	2013-09-05 13:43:30.557687773 -0500
> +++ linux/kernel/module.c	2013-10-07 12:33:43.732059759 -0500
> @@ -643,7 +643,7 @@ static int module_unload_init(struct mod
>  	INIT_LIST_HEAD(&mod->target_list);
>  
>  	/* Hold reference count during initialization. */
> -	__this_cpu_write(mod->refptr->incs, 1);
> +	raw_cpu_write(mod->refptr->incs, 1);

Just add: Safe to use raw because we're the only user. So we all
instantly know.

>  	/* Backwards compatibility macros put refcount during init. */
>  	mod->waiter = current;
>  
> 
--
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/