[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20171016221902.GM21978@ZenIV.linux.org.uk>
Date: Mon, 16 Oct 2017 23:19:02 +0100
From: Al Viro <viro@...IV.linux.org.uk>
To: Leon Yang <leon.gh.yang@...il.com>
Cc: "open list:FILESYSTEMS (VFS and infrastructure)"
<linux-fsdevel@...r.kernel.org>,
open list <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] Batch unmount cleanup
On Mon, Oct 16, 2017 at 01:52:16PM -0500, Leon Yang wrote:
> From: Leon Yang <leon.gh.yang@...il.com>
>
> Each time the unmounted list is cleanup, synchronize_rcu() is
> called, which is relatively costly. Scheduling the cleanup in a
> workqueue, similar to what is being done in
> net/core/net_namespace.c:cleanup_net, makes unmounting faster
> without adding too much overhead. This is useful especially for
> servers with many containers where mounting/unmounting happens a
> lot.
NAK. You really do _not_ want to return from umount(2) too early.
Powered by blists - more mailing lists