From: Andi Kleen <andi@firstfloor.org>
To: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
Andi Kleen <andi@firstfloor.org>, Ingo Molnar <mingo@elte.hu>,
LKML <linux-kernel@vger.kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
Peter Zijlstra <peterz@infradead.org>,
Steven Rostedt <rostedt@goodmis.org>,
Steven Rostedt <rostedt@rostedt.homelinux.com>,
Thomas Gleixner <tglx@linutronix.de>,
Christoph Hellwig <hch@lst.de>, Li Zefan <lizf@cn.fujitsu.com>,
Lai Jiangshan <laijs@cn.fujitsu.com>,
Johannes Berg <johannes.berg@intel.com>,
Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>,
Arnaldo Carvalho de Melo <acme@infradead.org>,
Tom Zanussi <tzanussi@gmail.com>,
KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>,
"H. Peter Anvin" <hpa@zytor.com>,
Jeremy Fitzhardinge <jeremy@goop.org>,
"Frank Ch. Eigler" <fche@redhat.com>,
Tejun Heo <htejun@gmail.com>
Subject: Re: [patch 1/2] x86_64 page fault NMI-safe
Date: Thu, 15 Jul 2010 16:35:18 +0200 [thread overview]
Message-ID: <20100715143518.GA18038@basil.fritz.box> (raw)
In-Reply-To: <20100715141118.GA6417@nowhere>
> But then how did the previous tasks get this new mapping? You said
> we don't walk through every process page tables for vmalloc.
No because those are always shared for the kernel and have been
filled in for init_mm.
Also most updates only update the lower tables anyways, top level
updates are extremly rare. In fact on PAE36 they should only happen
at most once, if at all, and most likely at early boot anyways
where you only have a single task.
On x86-64 they will only happen once every 512GB of vmalloc.
So for most systems also at most once at early boot.
>
> I would understand this race if we were to walk on every processes page
> tables and add the new mapping on them, but we missed one new task that
> forked or so, because we didn't lock (or just rcu).
The new task will always get a copy of the reference init_mm, which
was already updated.
-Andi
next prev parent reply other threads:[~2010-07-15 14:35 UTC|newest]
Thread overview: 163+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-07-14 15:49 [patch 0/2] x86: NMI-safe trap handlers Mathieu Desnoyers
2010-07-14 15:49 ` [patch 1/2] x86_64 page fault NMI-safe Mathieu Desnoyers
2010-07-14 16:28 ` Linus Torvalds
2010-07-14 17:06 ` Mathieu Desnoyers
2010-07-14 18:10 ` Linus Torvalds
2010-07-14 18:46 ` Ingo Molnar
2010-07-14 19:14 ` Linus Torvalds
2010-07-14 19:36 ` Frederic Weisbecker
2010-07-14 19:54 ` Linus Torvalds
2010-07-14 20:17 ` Mathieu Desnoyers
2010-07-14 20:55 ` Linus Torvalds
2010-07-14 21:18 ` Ingo Molnar
2010-07-14 22:14 ` Frederic Weisbecker
2010-07-14 22:31 ` Mathieu Desnoyers
2010-07-14 22:48 ` Frederic Weisbecker
2010-07-14 23:11 ` Mathieu Desnoyers
2010-07-14 23:38 ` Frederic Weisbecker
2010-07-15 16:26 ` Mathieu Desnoyers
2010-08-03 17:18 ` Peter Zijlstra
2010-08-03 18:25 ` Mathieu Desnoyers
2010-08-04 6:46 ` Peter Zijlstra
2010-08-04 7:14 ` Ingo Molnar
2010-08-04 14:45 ` Mathieu Desnoyers
2010-08-04 14:56 ` Peter Zijlstra
2010-08-06 1:49 ` Mathieu Desnoyers
2010-08-06 9:51 ` Peter Zijlstra
2010-08-06 13:46 ` Mathieu Desnoyers
2010-08-06 6:18 ` Masami Hiramatsu
2010-08-06 9:50 ` Peter Zijlstra
2010-08-06 13:37 ` Mathieu Desnoyers
2010-08-07 9:51 ` Masami Hiramatsu
2010-08-09 16:53 ` Frederic Weisbecker
2010-08-03 18:56 ` Linus Torvalds
2010-08-03 19:45 ` Mathieu Desnoyers
2010-08-03 20:02 ` Linus Torvalds
2010-08-03 20:10 ` Ingo Molnar
2010-08-03 20:21 ` Ingo Molnar
2010-08-03 21:16 ` Mathieu Desnoyers
2010-08-03 20:54 ` Mathieu Desnoyers
2010-08-04 6:27 ` Peter Zijlstra
2010-08-04 14:06 ` Mathieu Desnoyers
2010-08-04 14:50 ` Peter Zijlstra
2010-08-06 1:42 ` Mathieu Desnoyers
2010-08-06 10:11 ` Peter Zijlstra
2010-08-06 11:14 ` Peter Zijlstra
2010-08-06 14:15 ` Mathieu Desnoyers
2010-08-06 14:13 ` Mathieu Desnoyers
2010-08-11 14:44 ` Steven Rostedt
2010-08-11 14:34 ` Steven Rostedt
2010-08-15 13:35 ` Mathieu Desnoyers
2010-08-15 16:33 ` Avi Kivity
2010-08-15 16:44 ` Mathieu Desnoyers
2010-08-15 16:51 ` Avi Kivity
2010-08-15 18:31 ` Mathieu Desnoyers
2010-08-16 10:49 ` Avi Kivity
2010-08-16 11:29 ` Avi Kivity
2010-08-04 6:46 ` Dave Chinner
2010-08-04 7:21 ` Ingo Molnar
2010-07-14 23:40 ` Steven Rostedt
2010-07-14 19:41 ` Linus Torvalds
2010-07-14 19:56 ` Andi Kleen
2010-07-14 20:05 ` Mathieu Desnoyers
2010-07-14 20:07 ` Andi Kleen
2010-07-14 20:08 ` H. Peter Anvin
2010-07-14 23:32 ` Tejun Heo
2010-07-14 22:31 ` Frederic Weisbecker
2010-07-14 22:56 ` Linus Torvalds
2010-07-14 23:09 ` Andi Kleen
2010-07-14 23:22 ` Linus Torvalds
2010-07-15 14:11 ` Frederic Weisbecker
2010-07-15 14:35 ` Andi Kleen [this message]
2010-07-16 11:21 ` Frederic Weisbecker
2010-07-15 14:46 ` Steven Rostedt
2010-07-16 10:47 ` Frederic Weisbecker
2010-07-16 11:43 ` Steven Rostedt
2010-07-15 14:51 ` Linus Torvalds
2010-07-15 15:38 ` Linus Torvalds
2010-07-16 12:00 ` Frederic Weisbecker
2010-07-16 12:54 ` Steven Rostedt
2010-07-14 20:39 ` Mathieu Desnoyers
2010-07-14 21:23 ` Linus Torvalds
2010-07-14 21:45 ` Maciej W. Rozycki
2010-07-14 21:52 ` Linus Torvalds
2010-07-14 22:31 ` Maciej W. Rozycki
2010-07-14 22:21 ` Mathieu Desnoyers
2010-07-14 22:37 ` Linus Torvalds
2010-07-14 22:51 ` Jeremy Fitzhardinge
2010-07-14 23:02 ` Linus Torvalds
2010-07-14 23:54 ` Jeremy Fitzhardinge
2010-07-15 1:23 ` Linus Torvalds
2010-07-15 1:45 ` Linus Torvalds
2010-07-15 18:31 ` Mathieu Desnoyers
2010-07-15 18:43 ` Linus Torvalds
2010-07-15 18:48 ` Linus Torvalds
2010-07-15 22:01 ` Mathieu Desnoyers
2010-07-15 22:16 ` Linus Torvalds
2010-07-15 22:24 ` H. Peter Anvin
2010-07-15 22:26 ` Linus Torvalds
2010-07-15 22:46 ` H. Peter Anvin
2010-07-15 22:58 ` Andi Kleen
2010-07-15 23:20 ` H. Peter Anvin
2010-07-15 23:23 ` Linus Torvalds
2010-07-15 23:41 ` H. Peter Anvin
2010-07-15 23:44 ` Linus Torvalds
2010-07-15 23:46 ` H. Peter Anvin
2010-07-15 23:48 ` Andi Kleen
2010-07-15 22:30 ` Mathieu Desnoyers
2010-07-16 19:13 ` Mathieu Desnoyers
2010-07-15 16:44 ` Mathieu Desnoyers
2010-07-15 16:49 ` Linus Torvalds
2010-07-15 17:38 ` Mathieu Desnoyers
2010-07-15 20:44 ` H. Peter Anvin
2010-07-18 11:03 ` Avi Kivity
2010-07-18 17:36 ` Linus Torvalds
2010-07-18 18:04 ` Avi Kivity
2010-07-18 18:22 ` Linus Torvalds
2010-07-19 7:32 ` Avi Kivity
2010-07-18 18:17 ` Linus Torvalds
2010-07-18 18:43 ` Steven Rostedt
2010-07-18 19:26 ` Linus Torvalds
2010-07-14 15:49 ` [patch 2/2] x86 NMI-safe INT3 and Page Fault Mathieu Desnoyers
2010-07-14 16:42 ` Maciej W. Rozycki
2010-07-14 18:12 ` Mathieu Desnoyers
2010-07-14 19:21 ` Maciej W. Rozycki
2010-07-14 19:58 ` Mathieu Desnoyers
2010-07-14 20:36 ` Maciej W. Rozycki
2010-07-16 12:28 ` Avi Kivity
2010-07-16 14:49 ` Mathieu Desnoyers
2010-07-16 15:34 ` Andi Kleen
2010-07-16 15:40 ` Mathieu Desnoyers
2010-07-16 16:47 ` Avi Kivity
2010-07-16 16:58 ` Mathieu Desnoyers
2010-07-16 17:54 ` Avi Kivity
2010-07-16 18:05 ` H. Peter Anvin
2010-07-16 18:15 ` Avi Kivity
2010-07-16 18:17 ` H. Peter Anvin
2010-07-16 18:28 ` Avi Kivity
2010-07-16 18:37 ` Linus Torvalds
2010-07-16 19:26 ` Avi Kivity
2010-07-16 21:39 ` Linus Torvalds
2010-07-16 22:07 ` Andi Kleen
2010-07-16 22:26 ` Linus Torvalds
2010-07-16 22:41 ` Andi Kleen
2010-07-17 1:15 ` Linus Torvalds
2010-07-16 22:40 ` Mathieu Desnoyers
2010-07-18 9:23 ` Avi Kivity
2010-07-16 18:22 ` Mathieu Desnoyers
2010-07-16 18:32 ` Avi Kivity
2010-07-16 19:29 ` H. Peter Anvin
2010-07-16 19:39 ` Avi Kivity
2010-07-16 19:32 ` Andi Kleen
2010-07-16 18:25 ` Linus Torvalds
2010-07-16 19:30 ` Andi Kleen
2010-07-18 9:26 ` Avi Kivity
2010-07-16 19:28 ` Andi Kleen
2010-07-16 19:32 ` Avi Kivity
2010-07-16 19:34 ` Andi Kleen
2010-08-04 9:46 ` Peter Zijlstra
2010-08-04 20:23 ` H. Peter Anvin
2010-07-14 17:06 ` [patch 0/2] x86: NMI-safe trap handlers Andi Kleen
2010-07-14 17:08 ` Mathieu Desnoyers
2010-07-14 18:56 ` Andi Kleen
2010-07-14 23:29 ` Tejun Heo
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=20100715143518.GA18038@basil.fritz.box \
--to=andi@firstfloor.org \
--cc=acme@infradead.org \
--cc=akpm@linux-foundation.org \
--cc=fche@redhat.com \
--cc=fweisbec@gmail.com \
--cc=hch@lst.de \
--cc=hpa@zytor.com \
--cc=htejun@gmail.com \
--cc=jeremy@goop.org \
--cc=johannes.berg@intel.com \
--cc=kosaki.motohiro@jp.fujitsu.com \
--cc=laijs@cn.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=lizf@cn.fujitsu.com \
--cc=masami.hiramatsu.pt@hitachi.com \
--cc=mathieu.desnoyers@efficios.com \
--cc=mingo@elte.hu \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=rostedt@rostedt.homelinux.com \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=tzanussi@gmail.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;
as well as URLs for NNTP newsgroup(s).