From: Andi Kleen <andi@firstfloor.org>
To: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: cl@linux.com, linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [RFC][PATCH] fix move/migrate_pages() race on task struct
Date: Thu, 23 Feb 2012 10:45:00 -0800 [thread overview]
Message-ID: <m2zkc9pexf.fsf@firstfloor.org> (raw)
In-Reply-To: <20120223180740.C4EC4156@kernel> (Dave Hansen's message of "Thu, 23 Feb 2012 10:07:40 -0800")
Dave Hansen <dave@linux.vnet.ibm.com> writes:
> sys_move_pages() and sys_migrate_pages() are a pretty nice copy
> and paste job of each other. They both take a pid, find the task
> struct, and then grab a ref on the mm. They both also do an
> rcu_read_unlock() after they've taken the mm and then proceed to
> access 'task'. I think this is a bug in both cases.
Can we share code?
>
> This patch takes the pid-to-task code along with the credential
> and security checks in sys_move_pages() and sys_migrate_pages()
> and consolidates them. It now takes a task reference in
> the new function and requires the caller to drop it. I
> believe this resolves the race.
Looks good to me.
Reviewed-by: Andi Kleen <ak@linux.intel.com>
BTW looks like we really need a better stress test for these
syscalls.
-Andi
--
ak@linux.intel.com -- Speaking for myself only
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: Andi Kleen <andi@firstfloor.org>
To: Dave Hansen <dave@linux.vnet.ibm.com>
Cc: cl@linux.com, linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [RFC][PATCH] fix move/migrate_pages() race on task struct
Date: Thu, 23 Feb 2012 10:45:00 -0800 [thread overview]
Message-ID: <m2zkc9pexf.fsf@firstfloor.org> (raw)
In-Reply-To: <20120223180740.C4EC4156@kernel> (Dave Hansen's message of "Thu, 23 Feb 2012 10:07:40 -0800")
Dave Hansen <dave@linux.vnet.ibm.com> writes:
> sys_move_pages() and sys_migrate_pages() are a pretty nice copy
> and paste job of each other. They both take a pid, find the task
> struct, and then grab a ref on the mm. They both also do an
> rcu_read_unlock() after they've taken the mm and then proceed to
> access 'task'. I think this is a bug in both cases.
Can we share code?
>
> This patch takes the pid-to-task code along with the credential
> and security checks in sys_move_pages() and sys_migrate_pages()
> and consolidates them. It now takes a task reference in
> the new function and requires the caller to drop it. I
> believe this resolves the race.
Looks good to me.
Reviewed-by: Andi Kleen <ak@linux.intel.com>
BTW looks like we really need a better stress test for these
syscalls.
-Andi
--
ak@linux.intel.com -- Speaking for myself only
next prev parent reply other threads:[~2012-02-23 18:44 UTC|newest]
Thread overview: 56+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-02-23 18:07 [RFC][PATCH] fix move/migrate_pages() race on task struct Dave Hansen
2012-02-23 18:07 ` Dave Hansen
2012-02-23 18:45 ` Andi Kleen [this message]
2012-02-23 18:45 ` Andi Kleen
2012-02-23 18:45 ` Christoph Lameter
2012-02-23 18:45 ` Christoph Lameter
2012-02-23 19:10 ` Dave Hansen
2012-02-23 19:10 ` Dave Hansen
2012-02-23 19:40 ` Christoph Lameter
2012-02-23 19:40 ` Christoph Lameter
2012-02-23 20:04 ` Dave Hansen
2012-02-23 20:04 ` Dave Hansen
2012-02-23 21:41 ` Christoph Lameter
2012-02-23 21:41 ` Christoph Lameter
2012-02-24 3:14 ` Eric W. Biederman
2012-02-24 3:14 ` Eric W. Biederman
2012-02-24 15:20 ` Christoph Lameter
2012-02-24 15:20 ` Christoph Lameter
2012-02-24 15:41 ` Eric W. Biederman
2012-02-24 15:41 ` Eric W. Biederman
2012-02-24 16:48 ` Dave Hansen
2012-02-24 16:48 ` Dave Hansen
2012-02-24 16:54 ` Christoph Lameter
2012-02-24 16:54 ` Christoph Lameter
2012-02-24 17:04 ` Dave Hansen
2012-02-24 17:04 ` Dave Hansen
2012-02-24 17:08 ` Christoph Lameter
2012-02-24 17:08 ` Christoph Lameter
2012-02-24 17:25 ` Dave Hansen
2012-02-24 17:25 ` Dave Hansen
2012-02-24 17:32 ` Christoph Lameter
2012-02-24 17:32 ` Christoph Lameter
2012-02-24 21:37 ` Dave Hansen
2012-02-24 21:37 ` Dave Hansen
2012-02-24 23:12 ` Eric W. Biederman
2012-02-24 23:12 ` Eric W. Biederman
2012-02-27 16:43 ` Christoph Lameter
2012-02-27 16:43 ` Christoph Lameter
2012-02-25 12:13 ` Eric W. Biederman
2012-02-25 12:13 ` Eric W. Biederman
2012-02-27 19:01 ` Christoph Lameter
2012-02-27 19:01 ` Christoph Lameter
2012-02-27 20:15 ` Eric W. Biederman
2012-02-27 20:15 ` Eric W. Biederman
2012-02-27 22:39 ` Christoph Lameter
2012-02-27 22:39 ` Christoph Lameter
2012-02-28 19:30 ` Christoph Lameter
2012-02-28 19:30 ` Christoph Lameter
2012-02-29 20:31 ` Andrew Morton
2012-02-29 20:31 ` Andrew Morton
2012-02-29 20:33 ` Christoph Lameter
2012-02-29 20:33 ` Christoph Lameter
2012-02-29 20:36 ` Dave Hansen
2012-02-29 20:36 ` Dave Hansen
2012-02-24 17:07 ` KOSAKI Motohiro
2012-02-24 17:07 ` KOSAKI Motohiro
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=m2zkc9pexf.fsf@firstfloor.org \
--to=andi@firstfloor.org \
--cc=cl@linux.com \
--cc=dave@linux.vnet.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.