Subject: Re: [PATCH 1/3] perf_event: fix race in
 perf_swevent_get_recursion_context()
From: Peter Zijlstra <peterz@infradead.org>
To: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>, Frederic Weisbecker <fweisbec@gmail.com>,
       Paul Mackerras <paulus@samba.org>, LKML <linux-kernel@vger.kernel.org>
In-Reply-To: <4B556118.7020502@cn.fujitsu.com>
References: <4B54654A.4090601@cn.fujitsu.com>
	 <1263822958.4283.560.camel@laptop>  <4B556118.7020502@cn.fujitsu.com>
Content-Type: text/plain; charset="UTF-8"
Date: Tue, 19 Jan 2010 09:41:03 +0100
Message-ID: <1263890463.4283.636.camel@laptop>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1020
Lines: 28

On Tue, 2010-01-19 at 15:36 +0800, Xiao Guangrong wrote:
> Hi Peter,
> 
> Peter Zijlstra wrote:
> > On Mon, 2010-01-18 at 21:42 +0800, Xiao Guangrong wrote:
> >> It only disable preemption in perf_swevent_get_recursion_context()
> >> it can't avoid race of hard-irq and NMI
> >>
> >> In this patch, we use atomic operation to avoid it and reduce
> >> cpu_ctx->recursion size, it also make this patch no need diable
> >> preemption
> > 
> > Uhm why?
> > 
> > This patch looks terminally broken
> 
> Please see my explanation in another mail:
> http://lkml.org/lkml/2010/1/18/501

Still its not going to happen, we need those 4 recursion contexts.
Otherwise we could not receive a software event from an IRQ while we are
processing a software event from process context, etc.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/