From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx2.suse.de ([195.135.220.15]:56109 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751994AbdKIDZS (ORCPT ); Wed, 8 Nov 2017 22:25:18 -0500 From: NeilBrown To: Al Viro Date: Thu, 09 Nov 2017 14:22:07 +1100 Subject: [PATCH 1/3] VFS: use synchronize_rcu_expedited() in namespace_unlock() Cc: linux-fsdevel@vger.kernel.org, Linus Torvalds , linux-kernel@vger.kernel.org Message-ID: <151019772756.30101.64944229146330338.stgit@noble> In-Reply-To: <151019756744.30101.3832608128627682973.stgit@noble> References: <151019756744.30101.3832608128627682973.stgit@noble> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Sender: linux-fsdevel-owner@vger.kernel.org List-ID: The synchronize_rcu() in namespace_unlock() is called every time a filesystem is unmounted. If a great many filesystems are mounted, this can cause a noticable slow-down in, for example, system shutdown. The sequence: mkdir -p /tmp/Mtest/{0..5000} time for i in /tmp/Mtest/*; do mount -t tmpfs tmpfs $i ; done time umount /tmp/Mtest/* on a 4-cpu VM can report 8 seconds to mount the tmpfs filesystems, and 100 seconds to unmount them. Boot the same VM with 1 CPU and it takes 18 seconds to mount the tmpfs filesystems, but only 36 to unmount. If we change the synchronize_rcu() to synchronize_rcu_expedited() the umount time on a 4-cpu VM drop to 0.6 seconds I think this 200-fold speed up is worth the slightly higher system impact of using synchronize_rcu_expedited(). Acked-by: Paul E. McKenney (from general rcu perspective) Signed-off-by: NeilBrown --- fs/namespace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/namespace.c b/fs/namespace.c index d18deb4c410b..23cdf6c62895 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1420,7 +1420,7 @@ static void namespace_unlock(void) if (likely(hlist_empty(&head))) return; - synchronize_rcu(); + synchronize_rcu_expedited(); group_pin_kill(&head); }