From: Trond Myklebust <Trond.Myklebust@netapp.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Nick Bowler <nbowler@elliptictech.com>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
linux-nfs@vger.kernel.org
Subject: Re: [PATCH 2/3] NFS: lock the readdir page while it is in use
Date: Tue, 30 Nov 2010 23:29:13 -0500 [thread overview]
Message-ID: <1291177753.7694.17.camel@heimdal.trondhjem.org> (raw)
In-Reply-To: <AANLkTin8FSeM8cQ8BWYWjiyGy33EwzpMfFwauK+4j0uT@mail.gmail.com>
On Tue, 2010-11-30 at 20:10 -0800, Linus Torvalds wrote:
> On Tue, Nov 30, 2010 at 7:47 PM, Trond Myklebust
> <Trond.Myklebust@netapp.com> wrote:
> > Otherwise, the VM may end up removing it while we're reading from it.
>
> I don't think this is valid.
>
> Maybe it fixes a bug, but the commit description is misleading at
> best. Since you have a reference count to the page, the page is not
> going away. Locking may hide some other bug (due to serializing with
> other code you care about), but it is _not_ about the "VM may end up
> removing it".
>
> Even from a serialization angle, I think this patch is a bit suspect,
> since readdir() will always be called under the inode semaphore, so I
> think you'll always be serialized wrt other readdir users. Of course,
> you may have invalidation events etc that are outside of readdir, so
> ...
I'm not worried about other readdir calls invalidating the page. My
concern is rather about the VM memory reclaimers ejecting the page from
the page cache, and calling nfs_readdir_clear_array while we're
referencing the page. This wasn't a problem with the previous readdir
code, but it will be with the new incarnation because the actual
filenames are stored outside the page itself.
As far as I can see, the only way to protect against that is to lock the
page, perform the usual tests and then release the page lock when we're
done...
> Anyway if this patch matters, there's something else going on, and you
> need to describe that.
No problem. I just wanted to get the patches out so that the people who
are reporting regressions can start testing.
--
Trond Myklebust
Linux NFS client maintainer
NetApp
Trond.Myklebust@netapp.com
www.netapp.com
next prev parent reply other threads:[~2010-12-01 4:29 UTC|newest]
Thread overview: 70+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-11-30 17:42 [PATCH] NFS: Fix a readdirplus bug Trond Myklebust
2010-11-30 22:10 ` Linus Torvalds
2010-11-30 22:13 ` Trond Myklebust
2010-12-01 3:47 ` [PATCH 0/3] Fix more NFS readdir regressions Trond Myklebust
2010-12-01 3:47 ` [PATCH 1/3] NFS: Ensure we use the correct cookie in nfs_readdir_xdr_filler Trond Myklebust
2010-12-01 15:04 ` Nick Bowler
2010-12-01 15:36 ` [PATCH v2 0/3] Fix more NFS readdir regressions Trond Myklebust
2010-12-01 15:36 ` [PATCH v2 1/3] NFS: Ensure we use the correct cookie in nfs_readdir_xdr_filler Trond Myklebust
2010-12-01 15:36 ` [PATCH v2 2/3] NFS: lock the readdir page while it is in use Trond Myklebust
2010-12-01 15:36 ` [PATCH v2 3/3] NFS: Fix a memory leak in nfs_readdir Trond Myklebust
2010-12-01 16:17 ` Linus Torvalds
2010-12-01 16:35 ` Rik van Riel
2010-12-01 16:45 ` Benny Halevy
2010-12-01 16:47 ` Linus Torvalds
2010-12-01 17:02 ` Rik van Riel
2010-12-01 17:58 ` Trond Myklebust
2010-12-01 18:29 ` Miklos Szeredi
2010-12-01 18:54 ` Trond Myklebust
2010-12-01 19:23 ` Hugh Dickins
2010-12-01 19:52 ` Linus Torvalds
2010-12-01 20:05 ` Trond Myklebust
2010-12-01 20:39 ` Andrew Morton
2010-12-01 21:29 ` Neil Brown
2010-12-01 22:43 ` Andrew Morton
2010-12-01 23:01 ` Neil Brown
2010-12-01 19:47 ` Linus Torvalds
2010-12-01 20:10 ` Trond Myklebust
2010-12-01 20:18 ` Linus Torvalds
2010-12-01 20:31 ` Hugh Dickins
2010-12-01 20:33 ` Andrew Morton
2010-12-01 21:02 ` Hugh Dickins
2010-12-01 21:15 ` Hugh Dickins
2010-12-01 21:38 ` Andrew Morton
2010-12-01 21:51 ` Trond Myklebust
2010-12-01 22:13 ` Andrew Morton
2010-12-01 22:24 ` Linus Torvalds
2010-12-01 22:38 ` Andrew Morton
2010-12-01 22:47 ` Trond Myklebust
2010-12-01 23:21 ` Trond Myklebust
2010-12-01 23:46 ` Andrew Morton
2010-12-01 23:56 ` Trond Myklebust
2010-12-01 23:31 ` Linus Torvalds
2010-12-01 23:36 ` Andrew Morton
2010-12-02 1:05 ` Linus Torvalds
2010-12-02 1:22 ` Andrew Morton
2010-12-02 1:42 ` Linus Torvalds
2010-12-02 2:05 ` Andrew Morton
2010-12-02 3:08 ` [PATCH v3 0/3] Fix more NFS readdir regressions Trond Myklebust
2010-12-02 3:08 ` [PATCH v3 1/3] NFS: Ensure we use the correct cookie in nfs_readdir_xdr_filler Trond Myklebust
2010-12-02 3:08 ` [PATCH v3 2/3] Call the filesystem back whenever a page is removed from the page cache Trond Myklebust
2010-12-02 3:34 ` Hugh Dickins
2010-12-02 3:53 ` Trond Myklebust
2010-12-02 3:58 ` Linus Torvalds
2010-12-06 16:59 ` [PATCH v4 0/3] Fix more NFS readdir regressions Trond Myklebust
2010-12-06 16:59 ` [PATCH v4 1/3] NFS: Ensure we use the correct cookie in nfs_readdir_xdr_filler Trond Myklebust
2010-12-06 16:59 ` [PATCH v4 2/3] Call the filesystem back whenever a page is removed from the page cache Trond Myklebust
2010-12-07 7:08 ` Nick Piggin
2010-12-06 16:59 ` [PATCH v4 3/3] NFS: Fix a memory leak in nfs_readdir Trond Myklebust
2010-12-02 3:08 ` [PATCH v3 " Trond Myklebust
2010-12-03 9:12 ` [PATCH v2 " Nick Piggin
2010-12-01 23:43 ` Trond Myklebust
2010-12-01 22:43 ` Hugh Dickins
2010-12-01 3:47 ` [PATCH 2/3] NFS: lock the readdir page while it is in use Trond Myklebust
2010-12-01 4:10 ` Linus Torvalds
2010-12-01 4:29 ` Trond Myklebust [this message]
2010-12-01 5:06 ` Linus Torvalds
2010-12-01 14:49 ` Trond Myklebust
2010-12-01 13:14 ` Rik van Riel
2010-12-01 14:55 ` Trond Myklebust
2010-12-01 3:47 ` [PATCH 3/3] NFS: Fix a memory leak in nfs_readdir Trond Myklebust
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=1291177753.7694.17.camel@heimdal.trondhjem.org \
--to=trond.myklebust@netapp.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nfs@vger.kernel.org \
--cc=nbowler@elliptictech.com \
--cc=torvalds@linux-foundation.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 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®