From: William Lee Irwin III <wli@holomorphy.com>
To: Dave McCracken <dmccr@us.ibm.com>
Cc: Mika Penttil? <mika.penttila@kolumbus.fi>,
Linux Memory Management <linux-mm@kvack.org>,
Linux Kernel <linux-kernel@vger.kernel.org>
Subject: Re: Race between vmtruncate and mapped areas?
Date: Tue, 13 May 2003 16:16:07 -0700 [thread overview]
Message-ID: <20030513231607.GA8978@holomorphy.com> (raw)
In-Reply-To: <220550000.1052866808@baldur.austin.ibm.com>
On Tuesday, May 13, 2003 15:49:29 -0700 William Lee Irwin III <wli@holomorphy.com> wrote:
>> That doesn't sound like it's going to help, there isn't a unique
>> mmap_sem to be taken and so we just get caught between acquisitions
>> with the same problem.
On Tue, May 13, 2003 at 06:00:08PM -0500, Dave McCracken wrote:
> Actually it does fix it. I added code in vmtruncate_list() to do a
> down_write(&vma->vm_mm->mmap_sem) around the zap_page_range(), and the
> problem went away. It serializes against any outstanding page faults on a
> particular page table. New faults will see that the page is no longer in
> the file and fail with SIGBUS. Andrew's test case stopped failing.
> I've attached the patch so you can see what I did.
> Can anyone think of any gotchas to this solution?
How many processes are you doing this with? If it's a small number can
you try it with 1000 or so?
-- wli
WARNING: multiple messages have this Message-ID (diff)
From: William Lee Irwin III <wli@holomorphy.com>
To: Dave McCracken <dmccr@us.ibm.com>
Cc: Mika Penttil? <mika.penttila@kolumbus.fi>,
Linux Memory Management <linux-mm@kvack.org>,
Linux Kernel <linux-kernel@vger.kernel.org>
Subject: Re: Race between vmtruncate and mapped areas?
Date: Tue, 13 May 2003 16:16:07 -0700 [thread overview]
Message-ID: <20030513231607.GA8978@holomorphy.com> (raw)
In-Reply-To: <220550000.1052866808@baldur.austin.ibm.com>
On Tuesday, May 13, 2003 15:49:29 -0700 William Lee Irwin III <wli@holomorphy.com> wrote:
>> That doesn't sound like it's going to help, there isn't a unique
>> mmap_sem to be taken and so we just get caught between acquisitions
>> with the same problem.
On Tue, May 13, 2003 at 06:00:08PM -0500, Dave McCracken wrote:
> Actually it does fix it. I added code in vmtruncate_list() to do a
> down_write(&vma->vm_mm->mmap_sem) around the zap_page_range(), and the
> problem went away. It serializes against any outstanding page faults on a
> particular page table. New faults will see that the page is no longer in
> the file and fail with SIGBUS. Andrew's test case stopped failing.
> I've attached the patch so you can see what I did.
> Can anyone think of any gotchas to this solution?
How many processes are you doing this with? If it's a small number can
you try it with 1000 or so?
-- wli
--
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/ .
Don't email: <a href=mailto:"aart@kvack.org"> aart@kvack.org </a>
next prev parent reply other threads:[~2003-05-13 23:03 UTC|newest]
Thread overview: 100+ messages / expand[flat|nested] mbox.gz Atom feed top
2003-05-13 20:44 Race between vmtruncate and mapped areas? Dave McCracken
2003-05-13 20:44 ` Dave McCracken
2003-05-13 20:58 ` Mika Penttilä
2003-05-13 20:58 ` Mika Penttilä
2003-05-13 21:04 ` William Lee Irwin III
2003-05-13 21:04 ` William Lee Irwin III
2003-05-13 22:26 ` Dave McCracken
2003-05-13 22:26 ` Dave McCracken
2003-05-13 22:49 ` William Lee Irwin III
2003-05-13 22:49 ` William Lee Irwin III
2003-05-13 23:00 ` Dave McCracken
2003-05-13 23:11 ` William Lee Irwin III
2003-05-13 23:11 ` William Lee Irwin III
2003-05-13 23:16 ` Dave McCracken
2003-05-13 23:16 ` Dave McCracken
2003-05-13 23:20 ` William Lee Irwin III
2003-05-13 23:20 ` William Lee Irwin III
2003-05-13 23:28 ` Dave McCracken
2003-05-13 23:28 ` Dave McCracken
2003-05-13 23:29 ` William Lee Irwin III
2003-05-13 23:29 ` William Lee Irwin III
2003-05-13 23:16 ` William Lee Irwin III [this message]
2003-05-13 23:16 ` William Lee Irwin III
2003-05-14 1:10 ` Andrew Morton
2003-05-14 1:10 ` Andrew Morton
2003-05-14 15:02 ` Dave McCracken
2003-05-14 15:02 ` Dave McCracken
2003-05-14 1:10 ` Andrew Morton
2003-05-14 1:10 ` Andrew Morton
2003-05-14 15:02 ` Dave McCracken
2003-05-14 15:02 ` Dave McCracken
2003-05-14 15:06 ` William Lee Irwin III
2003-05-14 15:06 ` William Lee Irwin III
2003-05-14 15:25 ` Dave McCracken
2003-05-14 15:25 ` Dave McCracken
2003-05-14 16:42 ` Gerrit Huizenga
2003-05-14 16:42 ` Gerrit Huizenga
2003-05-14 17:34 ` Andrew Morton
2003-05-14 17:34 ` Andrew Morton
2003-05-14 17:42 ` Dave McCracken
2003-05-14 17:42 ` Dave McCracken
2003-05-14 17:57 ` Andrew Morton
2003-05-14 17:57 ` Andrew Morton
2003-05-14 18:05 ` Dave McCracken
2003-05-14 18:05 ` Dave McCracken
2003-05-14 18:17 ` Andrew Morton
2003-05-14 18:17 ` Andrew Morton
2003-05-14 18:24 ` Dave McCracken
2003-05-14 18:24 ` Dave McCracken
2003-05-14 18:53 ` Andrew Morton
2003-05-14 18:53 ` Andrew Morton
2003-05-15 8:50 ` Andrea Arcangeli
2003-05-15 8:50 ` Andrea Arcangeli
2003-05-14 19:02 ` Rik van Riel
2003-05-14 19:02 ` Rik van Riel
2003-05-14 19:04 ` Rik van Riel
2003-05-14 19:04 ` Rik van Riel
2003-05-14 19:07 ` Dave McCracken
2003-05-14 19:07 ` Dave McCracken
2003-05-14 19:11 ` Rik van Riel
2003-05-14 19:11 ` Rik van Riel
2003-05-15 0:49 ` Andrea Arcangeli
2003-05-15 0:49 ` Andrea Arcangeli
2003-05-15 2:36 ` Rik van Riel
2003-05-15 2:36 ` Rik van Riel
2003-05-15 9:46 ` Andrea Arcangeli
2003-05-15 9:46 ` Andrea Arcangeli
2003-05-15 9:55 ` Andrew Morton
2003-05-15 9:55 ` Andrew Morton
2003-05-15 8:32 ` Andrew Morton
2003-05-15 8:32 ` Andrew Morton
2003-05-15 8:42 ` Andrew Morton
2003-05-15 8:42 ` Andrew Morton
2003-05-15 8:55 ` Andrea Arcangeli
2003-05-15 8:55 ` Andrea Arcangeli
2003-05-15 9:20 ` Andrew Morton
2003-05-15 9:20 ` Andrew Morton
2003-05-15 9:40 ` Andrea Arcangeli
2003-05-15 9:40 ` Andrea Arcangeli
2003-05-15 9:58 ` Andrew Morton
2003-05-15 9:58 ` Andrew Morton
2003-05-15 16:38 ` Daniel McNeil
2003-05-15 19:19 ` Andrea Arcangeli
2003-05-15 19:19 ` Andrea Arcangeli
2003-05-15 22:04 ` Daniel McNeil
2003-05-15 22:04 ` Daniel McNeil
2003-05-15 23:17 ` Andrea Arcangeli
2003-05-15 23:17 ` Andrea Arcangeli
2003-05-17 0:27 ` Daniel McNeil
2003-05-17 0:27 ` Daniel McNeil
2003-05-17 17:29 ` Andrea Arcangeli
2003-05-17 17:29 ` Andrea Arcangeli
2003-05-13 21:00 ` William Lee Irwin III
2003-05-13 21:00 ` William Lee Irwin III
-- strict thread matches above, loose matches on Subject: below --
2003-05-17 18:19 Paul McKenney
2003-05-17 18:19 ` Paul McKenney
2003-05-17 18:42 ` Andrea Arcangeli
2003-05-17 18:42 ` Andrea Arcangeli
2003-05-19 18:11 Paul McKenney
2003-05-19 18:11 ` Paul McKenney
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=20030513231607.GA8978@holomorphy.com \
--to=wli@holomorphy.com \
--cc=dmccr@us.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mika.penttila@kolumbus.fi \
/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.