From: Peter Zijlstra <peterz@infradead.org>
To: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>,
Ingo Molnar <mingo@elte.hu>, Paul Mackerras <paulus@samba.org>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 1/3] perf_event: fix race in perf_swevent_get_recursion_context()
Date: Tue, 19 Jan 2010 09:46:17 +0100 [thread overview]
Message-ID: <1263890777.4283.638.camel@laptop> (raw)
In-Reply-To: <4B5508AF.1080302@cn.fujitsu.com>
On Tue, 2010-01-19 at 09:19 +0800, Xiao Guangrong wrote:
>
>
> It's because hard-irq(we can handle interruption with interruption enabled)
> and NMI are nested, for example:
>
> int perf_swevent_get_recursion_context(void)
> {
> ......
> if (cpuctx->recursion[rctx]) {
> put_cpu_var(perf_cpu_context);
> return -1;
> }
>
> /*
> * Another interruption handler/NMI will re-enter there if it
> * happed, it make the recursion value chaotic
> */
> cpuctx->recursion[rctx]++;
> ......
> }
This doesn't make any sense at all, if IRQs are nested (bad bad bad)
then it will still end up with the same recursion context and detect
that recursion and bail on the nested one. So what's the problem?
NMIs are not allowed to nest, ever.
next prev parent reply other threads:[~2010-01-19 8:46 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-01-18 13:42 [PATCH 1/3] perf_event: fix race in perf_swevent_get_recursion_context() Xiao Guangrong
2010-01-18 13:44 ` [PATCH 2/3] perf_event: cleanup for event profile buffer operation Xiao Guangrong
2010-01-18 13:46 ` [PATCH 3/3] tracing/kprobe: cleanup unused return value of function Xiao Guangrong
2010-01-18 16:16 ` Masami Hiramatsu
2010-01-19 8:37 ` [PATCH 1/3 v2] perf_event: fix race in perf_swevent_get_recursion_context() Xiao Guangrong
2010-01-19 8:46 ` Peter Zijlstra
2010-01-19 9:06 ` Xiao Guangrong
2010-01-19 8:39 ` [PATCH 2/3 v2] perf_event: cleanup for event profile buffer operation Xiao Guangrong
2010-01-19 8:41 ` [PATCH 3/3 v2] tracing/kprobe: cleanup unused return value of function Xiao Guangrong
2010-01-18 16:21 ` [PATCH 2/3] perf_event: cleanup for event profile buffer operation Masami Hiramatsu
2010-01-18 17:20 ` Frederic Weisbecker
2010-01-18 17:48 ` Masami Hiramatsu
2010-01-18 18:02 ` Frederic Weisbecker
2010-01-19 1:26 ` Xiao Guangrong
2010-01-19 9:00 ` Frederic Weisbecker
2010-01-19 14:26 ` Masami Hiramatsu
2010-01-18 17:11 ` Frederic Weisbecker
2010-01-18 13:55 ` [PATCH 1/3] perf_event: fix race in perf_swevent_get_recursion_context() Peter Zijlstra
2010-01-19 7:36 ` Xiao Guangrong
2010-01-19 8:41 ` Peter Zijlstra
2010-01-18 16:41 ` Frederic Weisbecker
2010-01-19 1:19 ` Xiao Guangrong
2010-01-19 8:46 ` Peter Zijlstra [this message]
2010-01-19 8:58 ` Frederic Weisbecker
2010-01-19 9:09 ` Xiao Guangrong
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1263890777.4283.638.camel@laptop \
--to=peterz@infradead.org \
--cc=fweisbec@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=paulus@samba.org \
--cc=xiaoguangrong@cn.fujitsu.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox