From: Steven Rostedt <rostedt@goodmis.org>
To: paulmck@linux.vnet.ibm.com
Cc: Peter Zijlstra <peterz@infradead.org>,
Josh Triplett <josh@joshtriplett.org>,
Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>,
linux-kernel@vger.kernel.org, Ingo Molnar <mingo@elte.hu>,
akpm@linux-foundation.org, tglx@linutronix.de,
Valdis.Kletnieks@vt.edu, dhowells@redhat.com,
laijs@cn.fujitsu.com, dipankar@in.ibm.com
Subject: Re: [RFC PATCH] introduce sys_membarrier(): process-wide memory barrier
Date: Thu, 07 Jan 2010 12:44:37 -0500 [thread overview]
Message-ID: <1262886277.28171.3734.camel@gandalf.stny.rr.com> (raw)
In-Reply-To: <20100107173118.GG6764@linux.vnet.ibm.com>
On Thu, 2010-01-07 at 09:31 -0800, Paul E. McKenney wrote:
> Something like the following for sys_membarrier(), then?
>
> smp_mb();
> for_each_cpu(cpu, current->mm->cpu_vm_mask) {
> if (cpu_curr(cpu)->mm == current->mm)
> smp_call_function_single(cpu, func, NULL, 1);
> }
>
> Then the code changing ->mm on the other CPU also needs to have a
> full smp_mb() somewhere after the change to ->mm, but before starting
> user-space execution. Which it might well just due to overhead, but
> we need to make sure that someone doesn't optimize us out of existence.
To change the mm requires things like flushing the TLB. I'd be surprised
if the change of the mm does not already do a smp_mb() somewhere.
-- Steve
next prev parent reply other threads:[~2010-01-07 17:44 UTC|newest]
Thread overview: 107+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-01-07 4:40 Mathieu Desnoyers
2010-01-07 5:02 ` Paul E. McKenney
2010-01-07 5:39 ` Mathieu Desnoyers
2010-01-07 8:32 ` Peter Zijlstra
2010-01-07 16:39 ` Paul E. McKenney
2010-01-07 5:28 ` Josh Triplett
2010-01-07 6:04 ` Mathieu Desnoyers
2010-01-07 6:32 ` Josh Triplett
2010-01-07 17:45 ` Mathieu Desnoyers
2010-01-07 16:46 ` Paul E. McKenney
2010-01-07 5:40 ` Steven Rostedt
2010-01-07 6:19 ` Mathieu Desnoyers
2010-01-07 6:35 ` Josh Triplett
2010-01-07 8:44 ` Peter Zijlstra
2010-01-07 13:15 ` Steven Rostedt
2010-01-07 15:07 ` Mathieu Desnoyers
2010-01-07 16:52 ` Paul E. McKenney
2010-01-07 17:18 ` Peter Zijlstra
2010-01-07 17:31 ` Paul E. McKenney
2010-01-07 17:44 ` Mathieu Desnoyers
2010-01-07 17:55 ` Paul E. McKenney
2010-01-07 17:44 ` Steven Rostedt [this message]
2010-01-07 17:56 ` Paul E. McKenney
2010-01-07 18:04 ` Steven Rostedt
2010-01-07 18:40 ` Paul E. McKenney
2010-01-07 17:36 ` Mathieu Desnoyers
2010-01-07 14:27 ` Steven Rostedt
2010-01-07 15:10 ` Mathieu Desnoyers
2010-01-07 16:49 ` Paul E. McKenney
2010-01-07 17:00 ` Steven Rostedt
2010-01-07 8:27 ` Peter Zijlstra
2010-01-07 18:30 ` Oleg Nesterov
2010-01-07 18:39 ` Paul E. McKenney
2010-01-07 18:59 ` Steven Rostedt
2010-01-07 19:16 ` Paul E. McKenney
2010-01-07 19:40 ` Steven Rostedt
2010-01-07 20:58 ` Paul E. McKenney
2010-01-07 21:35 ` Steven Rostedt
2010-01-07 22:34 ` Paul E. McKenney
2010-01-08 22:28 ` Mathieu Desnoyers
2010-01-08 23:53 ` Mathieu Desnoyers
2010-01-09 0:20 ` Paul E. McKenney
2010-01-09 1:02 ` Mathieu Desnoyers
2010-01-09 1:21 ` Paul E. McKenney
2010-01-09 1:22 ` Paul E. McKenney
2010-01-09 2:38 ` Mathieu Desnoyers
2010-01-09 5:42 ` Paul E. McKenney
2010-01-09 19:20 ` Mathieu Desnoyers
2010-01-09 23:05 ` Steven Rostedt
2010-01-09 23:16 ` Steven Rostedt
2010-01-10 0:03 ` Paul E. McKenney
2010-01-10 0:41 ` Steven Rostedt
2010-01-10 1:14 ` Mathieu Desnoyers
2010-01-10 1:44 ` Mathieu Desnoyers
2010-01-10 2:12 ` Steven Rostedt
2010-01-10 5:25 ` Paul E. McKenney
2010-01-10 11:50 ` Steven Rostedt
2010-01-10 16:03 ` Mathieu Desnoyers
2010-01-10 16:21 ` Steven Rostedt
2010-01-10 17:10 ` Mathieu Desnoyers
2010-01-10 21:02 ` Steven Rostedt
2010-01-10 21:41 ` Mathieu Desnoyers
2010-01-11 1:21 ` Paul E. McKenney
2010-01-10 17:45 ` Paul E. McKenney
2010-01-10 18:24 ` Mathieu Desnoyers
2010-01-11 1:17 ` Paul E. McKenney
2010-01-11 4:25 ` Mathieu Desnoyers
2010-01-11 4:29 ` [RFC PATCH] introduce sys_membarrier(): process-wide memory barrier (v3a) Mathieu Desnoyers
2010-01-11 17:27 ` Paul E. McKenney
2010-01-11 17:35 ` Mathieu Desnoyers
2010-01-11 17:50 ` Peter Zijlstra
2010-01-11 20:52 ` Mathieu Desnoyers
2010-01-11 21:19 ` Peter Zijlstra
2010-01-11 22:04 ` Mathieu Desnoyers
2010-01-11 22:20 ` Peter Zijlstra
2010-01-11 22:48 ` Paul E. McKenney
2010-01-11 22:48 ` Mathieu Desnoyers
2010-01-11 21:19 ` Peter Zijlstra
2010-01-11 21:31 ` Peter Zijlstra
2010-01-11 4:30 ` [RFC PATCH] introduce sys_membarrier(): process-wide memory barrier (v3b) Mathieu Desnoyers
2010-01-11 22:43 ` Paul E. McKenney
2010-01-12 15:38 ` Mathieu Desnoyers
2010-01-12 16:27 ` Steven Rostedt
2010-01-12 16:38 ` Mathieu Desnoyers
2010-01-12 16:54 ` Paul E. McKenney
2010-01-12 18:12 ` Paul E. McKenney
2010-01-12 18:56 ` Mathieu Desnoyers
2010-01-13 0:23 ` Paul E. McKenney
2010-01-11 16:25 ` [RFC PATCH] introduce sys_membarrier(): process-wide memory barrier Paul E. McKenney
2010-01-11 20:21 ` Mathieu Desnoyers
2010-01-11 21:48 ` Paul E. McKenney
2010-01-14 2:56 ` Lai Jiangshan
2010-01-14 5:13 ` Paul E. McKenney
2010-01-14 5:39 ` Mathieu Desnoyers
2010-01-10 5:18 ` Paul E. McKenney
2010-01-10 1:12 ` Mathieu Desnoyers
2010-01-10 5:19 ` Paul E. McKenney
2010-01-10 1:04 ` Mathieu Desnoyers
2010-01-10 1:01 ` Mathieu Desnoyers
2010-01-09 23:59 ` Paul E. McKenney
2010-01-10 1:11 ` Mathieu Desnoyers
2010-01-07 9:50 ` Andi Kleen
2010-01-07 15:12 ` Mathieu Desnoyers
2010-01-07 16:56 ` Paul E. McKenney
2010-01-07 11:04 ` David Howells
2010-01-07 15:15 ` Mathieu Desnoyers
2010-01-07 15:47 ` David Howells
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=1262886277.28171.3734.camel@gandalf.stny.rr.com \
--to=rostedt@goodmis.org \
--cc=Valdis.Kletnieks@vt.edu \
--cc=akpm@linux-foundation.org \
--cc=dhowells@redhat.com \
--cc=dipankar@in.ibm.com \
--cc=josh@joshtriplett.org \
--cc=laijs@cn.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@polymtl.ca \
--cc=mingo@elte.hu \
--cc=paulmck@linux.vnet.ibm.com \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
/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
all inboxes | Powered by JetHome®