All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jeff Layton <jeff.layton-7I+n7zu2hftEKMMhf/gKZA@public.gmane.org>
To: Christoph Hellwig <hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
Cc: linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	bfields-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org,
	cluster-devel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org,
	linux-cifs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [PATCH 09/10] locks: move freeing of leases outside of i_lock
Date: Sun, 24 Aug 2014 21:35:02 -0400	[thread overview]
Message-ID: <20140824213502.4cab1ba3@synchrony.poochiereds.net> (raw)
In-Reply-To: <20140824160804.GH15908-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>

On Sun, 24 Aug 2014 09:08:04 -0700
Christoph Hellwig <hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org> wrote:

> On Sat, Aug 23, 2014 at 10:41:17AM -0400, Jeff Layton wrote:
> > There was only one place where we still could free a file_lock while
> > holding the i_lock -- lease_modify. Add a new list_head argument to the
> > lm_change operation, pass in a private list when calling it, and fix
> > those callers to dispose of the list once the lock has been dropped.
> 
> Why do we care about locks held when simply freeing a piece of memory?
> 

That's all that's done now, but locks_free_lock can call
fl->fl_release_private. Currently, no filesystem sets that for leases,
but while we're in here we might as well make it possible for that to
block for all lock types since that's simpler than having different
rules for them.

It doesn't really cost us anything other than a bit of list
manipulation and reduces contention on the i_lock slightly.

-- 
Jeff Layton <jlayton-7I+n7zu2hftEKMMhf/gKZA@public.gmane.org>

WARNING: multiple messages have this Message-ID (diff)
From: Jeff Layton <jeff.layton@primarydata.com>
To: Christoph Hellwig <hch@infradead.org>
Cc: linux-fsdevel@vger.kernel.org, bfields@fieldses.org,
	cluster-devel@redhat.com, linux-cifs@vger.kernel.org,
	linux-nfs@vger.kernel.org
Subject: Re: [PATCH 09/10] locks: move freeing of leases outside of i_lock
Date: Sun, 24 Aug 2014 21:35:02 -0400	[thread overview]
Message-ID: <20140824213502.4cab1ba3@synchrony.poochiereds.net> (raw)
In-Reply-To: <20140824160804.GH15908@infradead.org>

On Sun, 24 Aug 2014 09:08:04 -0700
Christoph Hellwig <hch@infradead.org> wrote:

> On Sat, Aug 23, 2014 at 10:41:17AM -0400, Jeff Layton wrote:
> > There was only one place where we still could free a file_lock while
> > holding the i_lock -- lease_modify. Add a new list_head argument to the
> > lm_change operation, pass in a private list when calling it, and fix
> > those callers to dispose of the list once the lock has been dropped.
> 
> Why do we care about locks held when simply freeing a piece of memory?
> 

That's all that's done now, but locks_free_lock can call
fl->fl_release_private. Currently, no filesystem sets that for leases,
but while we're in here we might as well make it possible for that to
block for all lock types since that's simpler than having different
rules for them.

It doesn't really cost us anything other than a bit of list
manipulation and reduces contention on the i_lock slightly.

-- 
Jeff Layton <jlayton@primarydata.com>

WARNING: multiple messages have this Message-ID (diff)
From: Jeff Layton <jeff.layton@primarydata.com>
To: cluster-devel.redhat.com
Subject: [Cluster-devel] [PATCH 09/10] locks: move freeing of leases outside of i_lock
Date: Sun, 24 Aug 2014 21:35:02 -0400	[thread overview]
Message-ID: <20140824213502.4cab1ba3@synchrony.poochiereds.net> (raw)
In-Reply-To: <20140824160804.GH15908@infradead.org>

On Sun, 24 Aug 2014 09:08:04 -0700
Christoph Hellwig <hch@infradead.org> wrote:

> On Sat, Aug 23, 2014 at 10:41:17AM -0400, Jeff Layton wrote:
> > There was only one place where we still could free a file_lock while
> > holding the i_lock -- lease_modify. Add a new list_head argument to the
> > lm_change operation, pass in a private list when calling it, and fix
> > those callers to dispose of the list once the lock has been dropped.
> 
> Why do we care about locks held when simply freeing a piece of memory?
> 

That's all that's done now, but locks_free_lock can call
fl->fl_release_private. Currently, no filesystem sets that for leases,
but while we're in here we might as well make it possible for that to
block for all lock types since that's simpler than having different
rules for them.

It doesn't really cost us anything other than a bit of list
manipulation and reduces contention on the i_lock slightly.

-- 
Jeff Layton <jlayton@primarydata.com>



  parent reply	other threads:[~2014-08-25  1:35 UTC|newest]

Thread overview: 101+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-08-23 14:41 [PATCH 00/10] locks/nfsd: internal lease API overhaul Jeff Layton
2014-08-23 14:41 ` [Cluster-devel] " Jeff Layton
2014-08-23 14:41 ` Jeff Layton
2014-08-23 14:41 ` [PATCH 04/10] locks: clean up vfs_setlease kerneldoc comments Jeff Layton
2014-08-23 14:41   ` [Cluster-devel] " Jeff Layton
2014-08-24 15:51   ` Christoph Hellwig
2014-08-24 15:51     ` [Cluster-devel] " Christoph Hellwig
     [not found]     ` <20140824155117.GC15908-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2014-08-25 20:11       ` J. Bruce Fields
2014-08-25 20:11         ` [Cluster-devel] " J. Bruce Fields
2014-08-25 20:11         ` J. Bruce Fields
2014-08-23 14:41 ` [PATCH 07/10] locks: define a lm_setup handler for leases Jeff Layton
2014-08-23 14:41   ` [Cluster-devel] " Jeff Layton
     [not found]   ` <1408804878-1331-8-git-send-email-jlayton-7I+n7zu2hftEKMMhf/gKZA@public.gmane.org>
2014-08-24 15:58     ` Christoph Hellwig
2014-08-24 15:58       ` [Cluster-devel] " Christoph Hellwig
2014-08-24 15:58       ` Christoph Hellwig
2014-08-25  1:19       ` Jeff Layton
2014-08-25  1:19         ` [Cluster-devel] " Jeff Layton
2014-08-26 13:58         ` Christoph Hellwig
2014-08-26 13:58           ` [Cluster-devel] " Christoph Hellwig
2014-08-23 14:41 ` [PATCH 08/10] locks: move i_lock acquisition into generic_*_lease handlers Jeff Layton
2014-08-23 14:41   ` [Cluster-devel] " Jeff Layton
     [not found]   ` <1408804878-1331-9-git-send-email-jlayton-7I+n7zu2hftEKMMhf/gKZA@public.gmane.org>
2014-08-24 16:06     ` Christoph Hellwig
2014-08-24 16:06       ` [Cluster-devel] " Christoph Hellwig
2014-08-24 16:06       ` Christoph Hellwig
     [not found]       ` <20140824160634.GG15908-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2014-08-24 16:11         ` Christoph Hellwig
2014-08-24 16:11           ` [Cluster-devel] " Christoph Hellwig
2014-08-24 16:11           ` Christoph Hellwig
     [not found]           ` <20140824161134.GJ15908-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2014-08-31 14:51             ` Jeff Layton
2014-08-31 14:51               ` [Cluster-devel] " Jeff Layton
2014-08-31 14:51               ` Jeff Layton
2014-08-25  1:36         ` Jeff Layton
2014-08-25  1:36           ` [Cluster-devel] " Jeff Layton
2014-08-25  1:36           ` Jeff Layton
     [not found] ` <1408804878-1331-1-git-send-email-jlayton-7I+n7zu2hftEKMMhf/gKZA@public.gmane.org>
2014-08-23 14:41   ` [PATCH 01/10] locks: close potential race in lease_get_mtime Jeff Layton
2014-08-23 14:41     ` [Cluster-devel] " Jeff Layton
2014-08-23 14:41     ` Jeff Layton
     [not found]     ` <1408804878-1331-2-git-send-email-jlayton-7I+n7zu2hftEKMMhf/gKZA@public.gmane.org>
2014-08-24 15:48       ` Christoph Hellwig
2014-08-24 15:48         ` [Cluster-devel] " Christoph Hellwig
2014-08-24 15:48         ` Christoph Hellwig
2014-08-25 20:01     ` J. Bruce Fields
2014-08-25 20:01       ` [Cluster-devel] " J. Bruce Fields
2014-08-23 14:41   ` [PATCH 02/10] nfsd: fix potential lease memory leak in nfs4_setlease Jeff Layton
2014-08-23 14:41     ` [Cluster-devel] " Jeff Layton
2014-08-23 14:41     ` Jeff Layton
     [not found]     ` <1408804878-1331-3-git-send-email-jlayton-7I+n7zu2hftEKMMhf/gKZA@public.gmane.org>
2014-08-24 15:48       ` Christoph Hellwig
2014-08-24 15:48         ` [Cluster-devel] " Christoph Hellwig
2014-08-24 15:48         ` Christoph Hellwig
2014-08-23 14:41   ` [PATCH 03/10] locks: generic_delete_lease doesn't need a file_lock at all Jeff Layton
2014-08-23 14:41     ` [Cluster-devel] " Jeff Layton
2014-08-23 14:41     ` Jeff Layton
     [not found]     ` <1408804878-1331-4-git-send-email-jlayton-7I+n7zu2hftEKMMhf/gKZA@public.gmane.org>
2014-08-24  1:27       ` Christoph Hellwig
2014-08-24  1:27         ` [Cluster-devel] " Christoph Hellwig
2014-08-24  1:27         ` Christoph Hellwig
     [not found]         ` <20140824012757.GA21609-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2014-08-24 10:09           ` Jeff Layton
2014-08-24 10:09             ` [Cluster-devel] " Jeff Layton
2014-08-24 10:09             ` Jeff Layton
2014-08-23 14:41   ` [PATCH 05/10] nfsd: don't keep a pointer to the lease in nfs4_file Jeff Layton
2014-08-23 14:41     ` [Cluster-devel] " Jeff Layton
2014-08-23 14:41     ` Jeff Layton
     [not found]     ` <1408804878-1331-6-git-send-email-jlayton-7I+n7zu2hftEKMMhf/gKZA@public.gmane.org>
2014-08-24 15:51       ` Christoph Hellwig
2014-08-24 15:51         ` [Cluster-devel] " Christoph Hellwig
2014-08-24 15:51         ` Christoph Hellwig
2014-08-23 14:41   ` [PATCH 06/10] locks: plumb an "aux" pointer into the setlease routines Jeff Layton
2014-08-23 14:41     ` [Cluster-devel] " Jeff Layton
2014-08-23 14:41     ` Jeff Layton
     [not found]     ` <1408804878-1331-7-git-send-email-jlayton-7I+n7zu2hftEKMMhf/gKZA@public.gmane.org>
2014-08-24  1:33       ` Christoph Hellwig
2014-08-24  1:33         ` [Cluster-devel] " Christoph Hellwig
2014-08-24  1:33         ` Christoph Hellwig
     [not found]         ` <20140824013305.GB21609-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2014-08-24 10:08           ` Jeff Layton
2014-08-24 10:08             ` [Cluster-devel] " Jeff Layton
2014-08-24 10:08             ` Jeff Layton
     [not found]             ` <20140824060801.5402880c-08S845evdOaAjSkqwZiSMmfYqLom42DlXqFh9Ls21Oc@public.gmane.org>
2014-08-24 15:54               ` Christoph Hellwig
2014-08-24 15:54                 ` [Cluster-devel] " Christoph Hellwig
2014-08-24 15:54                 ` Christoph Hellwig
2014-08-25 20:28       ` J. Bruce Fields
2014-08-25 20:28         ` [Cluster-devel] " J. Bruce Fields
2014-08-25 20:28         ` J. Bruce Fields
     [not found]         ` <20140825202852.GD21957-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org>
2014-08-26 10:53           ` Jeff Layton
2014-08-26 10:53             ` [Cluster-devel] " Jeff Layton
2014-08-26 10:53             ` Jeff Layton
2014-08-23 14:41   ` [PATCH 09/10] locks: move freeing of leases outside of i_lock Jeff Layton
2014-08-23 14:41     ` [Cluster-devel] " Jeff Layton
2014-08-23 14:41     ` Jeff Layton
     [not found]     ` <1408804878-1331-10-git-send-email-jlayton-7I+n7zu2hftEKMMhf/gKZA@public.gmane.org>
2014-08-24 16:08       ` Christoph Hellwig
2014-08-24 16:08         ` [Cluster-devel] " Christoph Hellwig
2014-08-24 16:08         ` Christoph Hellwig
     [not found]         ` <20140824160804.GH15908-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2014-08-25  1:35           ` Jeff Layton [this message]
2014-08-25  1:35             ` [Cluster-devel] " Jeff Layton
2014-08-25  1:35             ` Jeff Layton
2014-08-23 14:41   ` [PATCH 10/10] locks: update Documentation/filesystems with lease API changes Jeff Layton
2014-08-23 14:41     ` [Cluster-devel] " Jeff Layton
2014-08-23 14:41     ` Jeff Layton
2014-08-24 16:10   ` [PATCH 00/10] locks/nfsd: internal lease API overhaul Christoph Hellwig
2014-08-24 16:10     ` [Cluster-devel] " Christoph Hellwig
2014-08-24 16:10     ` Christoph Hellwig
     [not found]     ` <20140824161046.GI15908-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2014-08-25  1:43       ` Jeff Layton
2014-08-25  1:43         ` [Cluster-devel] " Jeff Layton
2014-08-25  1:43         ` Jeff Layton
     [not found]         ` <20140824214301.61019123-08S845evdOaAjSkqwZiSMmfYqLom42DlXqFh9Ls21Oc@public.gmane.org>
2014-08-26 13:59           ` Christoph Hellwig
2014-08-26 13:59             ` [Cluster-devel] " Christoph Hellwig
2014-08-26 13:59             ` Christoph Hellwig

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=20140824213502.4cab1ba3@synchrony.poochiereds.net \
    --to=jeff.layton-7i+n7zu2hftekmmhf/gkza@public.gmane.org \
    --cc=bfields-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org \
    --cc=cluster-devel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
    --cc=hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org \
    --cc=linux-cifs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.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.