From: Andrea Arcangeli <aarcange@redhat.com>
To: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: LKML <linux-kernel@vger.kernel.org>,
Linus Torvalds <torvalds@osdl.org>, Andrew Morton <akpm@osdl.org>,
Nick Piggin <nickpiggin@yahoo.com.au>,
Jeff Moyer <jmoyer@redhat.com>,
linux-mm@kvack.org, linux-fsdevel@vger.kernel.org,
Hugh Dickins <hugh@veritas.com>
Subject: Re: [RFC][PATCH v3 1/6] mm: Don't unmap gup()ed page
Date: Tue, 14 Apr 2009 16:38:29 +0200 [thread overview]
Message-ID: <20090414143829.GG28265@random.random> (raw)
In-Reply-To: <20090414151554.C64A.A69D9226@jp.fujitsu.com>
On Tue, Apr 14, 2009 at 03:16:52PM +0900, KOSAKI Motohiro wrote:
> + if (PageSwapCache(page) &&
> + page_count(page) != page_mapcount(page) + 2) {
> + ret = SWAP_FAIL;
> + goto out_unmap;
> + }
> +
Besides the race pointed out by Nick, this also would break KVM
swapping with mmu notifier. mmu_notifier_invalidate_page must be
invoked before reading page_count for this to work. However the
invalidate has to be moved below the
mlock/ptep_clear_flush_young_notify, no point to get rid of sptes if
any of the spte or the pte is still young.
WARNING: multiple messages have this Message-ID (diff)
From: Andrea Arcangeli <aarcange@redhat.com>
To: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: LKML <linux-kernel@vger.kernel.org>,
Linus Torvalds <torvalds@osdl.org>, Andrew Morton <akpm@osdl.org>,
Nick Piggin <nickpiggin@yahoo.com.au>,
Jeff Moyer <jmoyer@redhat.com>,
linux-mm@kvack.org, linux-fsdevel@vger.kernel.org,
Hugh Dickins <hugh@veritas.com>
Subject: Re: [RFC][PATCH v3 1/6] mm: Don't unmap gup()ed page
Date: Tue, 14 Apr 2009 16:38:29 +0200 [thread overview]
Message-ID: <20090414143829.GG28265@random.random> (raw)
In-Reply-To: <20090414151554.C64A.A69D9226@jp.fujitsu.com>
On Tue, Apr 14, 2009 at 03:16:52PM +0900, KOSAKI Motohiro wrote:
> + if (PageSwapCache(page) &&
> + page_count(page) != page_mapcount(page) + 2) {
> + ret = SWAP_FAIL;
> + goto out_unmap;
> + }
> +
Besides the race pointed out by Nick, this also would break KVM
swapping with mmu notifier. mmu_notifier_invalidate_page must be
invoked before reading page_count for this to work. However the
invalidate has to be moved below the
mlock/ptep_clear_flush_young_notify, no point to get rid of sptes if
any of the spte or the pte is still young.
--
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:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2009-04-14 14:40 UTC|newest]
Thread overview: 93+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-04-14 6:15 [RFC][PATCH 0/6] IO pinning(get_user_pages()) vs fork race fix KOSAKI Motohiro
2009-04-14 6:15 ` KOSAKI Motohiro
2009-04-14 6:16 ` [RFC][PATCH v3 1/6] mm: Don't unmap gup()ed page KOSAKI Motohiro
2009-04-14 6:16 ` KOSAKI Motohiro
2009-04-14 9:25 ` Nick Piggin
2009-04-14 9:25 ` Nick Piggin
2009-04-14 12:02 ` KOSAKI Motohiro
2009-04-14 12:02 ` KOSAKI Motohiro
2009-04-14 12:25 ` Nick Piggin
2009-04-14 12:25 ` Nick Piggin
2009-04-14 13:39 ` KOSAKI Motohiro
2009-04-14 13:39 ` KOSAKI Motohiro
2009-04-14 13:39 ` KOSAKI Motohiro
2009-04-14 14:12 ` Andrea Arcangeli
2009-04-14 14:12 ` Andrea Arcangeli
2009-04-14 14:26 ` Nick Piggin
2009-04-14 14:26 ` Nick Piggin
2009-04-14 14:32 ` Andrea Arcangeli
2009-04-14 14:32 ` Andrea Arcangeli
2009-04-14 14:42 ` Nick Piggin
2009-04-14 14:42 ` Nick Piggin
2009-04-14 15:21 ` Andrea Arcangeli
2009-04-14 15:21 ` Andrea Arcangeli
2009-04-15 8:05 ` KOSAKI Motohiro
2009-04-15 8:05 ` KOSAKI Motohiro
2009-04-15 8:22 ` Nick Piggin
2009-04-15 8:22 ` Nick Piggin
2009-04-15 9:22 ` Nick Piggin
2009-04-15 9:22 ` Nick Piggin
2009-04-15 10:46 ` Andrea Arcangeli
2009-04-15 10:46 ` Andrea Arcangeli
2009-04-15 11:39 ` KOSAKI Motohiro
2009-04-15 11:39 ` KOSAKI Motohiro
2009-04-15 11:39 ` KOSAKI Motohiro
2009-04-15 11:41 ` Andrea Arcangeli
2009-04-15 11:41 ` Andrea Arcangeli
2009-04-15 11:41 ` Andrea Arcangeli
2009-04-15 11:53 ` KOSAKI Motohiro
2009-04-15 11:53 ` KOSAKI Motohiro
2009-04-19 12:37 ` KOSAKI Motohiro
2009-04-19 12:37 ` KOSAKI Motohiro
2009-04-14 14:38 ` Andrea Arcangeli [this message]
2009-04-14 14:38 ` Andrea Arcangeli
2009-04-14 6:18 ` [RFC][PATCH v3 2/6] mm, directio: fix fork vs direct-io race (read(2) side IOW gup(write) side) KOSAKI Motohiro
2009-04-14 6:18 ` KOSAKI Motohiro
2009-04-14 6:25 ` KOSAKI Motohiro
2009-04-14 6:25 ` KOSAKI Motohiro
2009-04-14 16:45 ` Jeff Moyer
2009-04-14 16:45 ` Jeff Moyer
2009-04-14 17:51 ` Andrea Arcangeli
2009-04-14 17:51 ` Andrea Arcangeli
2009-04-14 18:10 ` Jeff Moyer
2009-04-14 18:10 ` Jeff Moyer
2009-04-14 19:48 ` Andrea Arcangeli
2009-04-14 19:48 ` Andrea Arcangeli
2009-04-14 6:19 ` [RFC][PATCH v3 3/6] nfs, direct-io: fix fork vs direct-io race on nfs KOSAKI Motohiro
2009-04-14 6:19 ` KOSAKI Motohiro
2009-04-14 16:48 ` Jeff Moyer
2009-04-14 16:48 ` Jeff Moyer
2009-04-14 6:20 ` [RFC][PATCH v3 4/6] aio: Don't inherit aio ring memory at fork KOSAKI Motohiro
2009-04-14 6:20 ` KOSAKI Motohiro
2009-04-14 13:41 ` Andrea Arcangeli
2009-04-14 13:41 ` Andrea Arcangeli
2009-04-14 16:01 ` Jeff Moyer
2009-04-14 16:01 ` Jeff Moyer
2009-04-15 0:56 ` KOSAKI Motohiro
2009-04-15 0:56 ` KOSAKI Motohiro
2009-04-15 2:44 ` Jeff Moyer
2009-04-15 2:44 ` Jeff Moyer
2009-04-15 3:00 ` KOSAKI Motohiro
2009-04-15 3:00 ` KOSAKI Motohiro
2009-04-14 6:21 ` [RFC][PATCH v3 5/6] don't use bio-map in read() path KOSAKI Motohiro
2009-04-14 6:21 ` KOSAKI Motohiro
2009-04-14 6:23 ` [RFC][PATCH v3 6/6] fix wrong get_user_pages usage in iovlock.c KOSAKI Motohiro
2009-04-14 6:23 ` KOSAKI Motohiro
2009-04-14 6:56 ` Nick Piggin
2009-04-14 6:56 ` Nick Piggin
2009-04-14 6:58 ` KOSAKI Motohiro
2009-04-14 6:58 ` KOSAKI Motohiro
2009-04-15 8:48 ` KOSAKI Motohiro
2009-04-15 8:48 ` KOSAKI Motohiro
2009-04-17 15:07 ` Sosnowski, Maciej
2009-04-17 15:07 ` Sosnowski, Maciej
2009-04-19 12:37 ` KOSAKI Motohiro
2009-04-19 12:37 ` KOSAKI Motohiro
2009-04-23 12:48 ` Sosnowski, Maciej
2009-04-23 12:48 ` Sosnowski, Maciej
2009-04-14 8:41 ` [RFC][PATCH 0/6] IO pinning(get_user_pages()) vs fork race fix Nick Piggin
2009-04-14 8:41 ` Nick Piggin
2009-04-14 9:19 ` KOSAKI Motohiro
2009-04-14 9:19 ` KOSAKI Motohiro
2009-04-14 9:37 ` Nick Piggin
2009-04-14 9:37 ` Nick Piggin
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=20090414143829.GG28265@random.random \
--to=aarcange@redhat.com \
--cc=akpm@osdl.org \
--cc=hugh@veritas.com \
--cc=jmoyer@redhat.com \
--cc=kosaki.motohiro@jp.fujitsu.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=nickpiggin@yahoo.com.au \
--cc=torvalds@osdl.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.