From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261786AbUKPUDZ (ORCPT ); Tue, 16 Nov 2004 15:03:25 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261770AbUKPTvL (ORCPT ); Tue, 16 Nov 2004 14:51:11 -0500 Received: from omx1-ext.sgi.com ([192.48.179.11]:31886 "EHLO omx1.americas.sgi.com") by vger.kernel.org with ESMTP id S261786AbUKPTt6 (ORCPT ); Tue, 16 Nov 2004 14:49:58 -0500 Date: Tue, 16 Nov 2004 13:48:58 -0600 From: Robin Holt To: Andrew Morton Cc: Robin Holt , linux-kernel@vger.kernel.org, dev@sw.ru, wli@holomorphy.com, steiner@sgi.com, sandeen@sgi.com Subject: Re: 21 million inodes is causing severe pauses. Message-ID: <20041116194858.GA2549@lnx-holt.americas.sgi.com> References: <20041115195551.GA15380@lnx-holt.americas.sgi.com> <20041115145714.3f757012.akpm@osdl.org> <20041116162859.GA5594@lnx-holt.americas.sgi.com> <20041116111321.36a6cd06.akpm@osdl.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20041116111321.36a6cd06.akpm@osdl.org> User-Agent: Mutt/1.4.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Nov 16, 2004 at 11:13:21AM -0800, Andrew Morton wrote: > Robin Holt wrote: > > > > I guess I am very concerned at this point. If I can do a > > release/reacquire, why not just change generic_shutdown_super() so the > > lock_kernel() does not happen until the first pass has occurred. ie: > > > > --- super.c.orig 2004-11-16 10:22:17 -06:00 > > +++ super.c 2004-11-16 10:22:41 -06:00 > > @@ -232,10 +232,10 @@ > > dput(root); > > fsync_super(sb); > > lock_super(sb); > > - lock_kernel(); > > sb->s_flags &= ~MS_ACTIVE; > > /* bad name - it should be evict_inodes() */ > > invalidate_inodes(sb); > > + lock_kernel(); > > > > if (sop->write_super && sb->s_dirt) > > sop->write_super(sb); > > > > This at least makes the lock_kernel time much smaller than it is right > > now. It also does not affect any callers that may really need the BKL. > > lock_kernel() is also taken way up in do_umount(), hence the need for > release_kernel_lock()/reacquire_kernel_lock(). It looks like it is only held very briefly during the early parts of do_umount. I have moved lock_kernel() as above in addition to the two patches you pointed to earlier. This has left me with a system which has 21M inodes and undetectable delays during heavy mount/umount activity. I am starting one last test which attempts a umount of a filesystem which has many inodes associated with it. At this point, I have checked the entire code path and see no reason the BKL is held for the first call to invalidate_inodes.