From: Boaz Harrosh Subject: Re: [PATCH,RFC] ext4: add lazytime mount option Date: Thu, 13 Nov 2014 10:44:59 +0200 Message-ID: <54646F8B.8040000@plexistor.com> References: <1415765227-9561-1-git-send-email-tytso@mit.edu> <20141113064150.GI28565@dastard> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Cc: Ext4 Developers List , linux-fsdevel@vger.kernel.org To: Dave Chinner , Theodore Ts'o Return-path: Received: from mail-wi0-f174.google.com ([209.85.212.174]:61832 "EHLO mail-wi0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751013AbaKMIpB (ORCPT ); Thu, 13 Nov 2014 03:45:01 -0500 Received: by mail-wi0-f174.google.com with SMTP id d1so7498848wiv.1 for ; Thu, 13 Nov 2014 00:45:00 -0800 (PST) In-Reply-To: <20141113064150.GI28565@dastard> Sender: linux-ext4-owner@vger.kernel.org List-ID: On 11/13/2014 08:41 AM, Dave Chinner wrote: <> > > I think this needs to a VFS level inode timestamp update option. > The games ext4 is playing with reference counts inside .drop_inode are > pretty nasty and could be avoided if this is implemented at the VFs > level.. > > I think that the "lazy time update" status should really be tracked > in the inode->i_state field. Something like lazytime updates do not > call ->update_inode, nor do they mark the inode dirty, but they do > update the inode->i_[acm]time fields and set a TIMEDIRTY state flag. > > Then when the filesystem next logs or writes the inode it can > log those fields and clear the TIMEDIRTY flag, or if iput_final() > sees that flag it can call ->update_time directly to get the > filesystem to run a transaction to update the timestamps before the > inode is evicted from cache. > > That way the same update semantics can be easily implemented on all > filesystems, and filesystems that already implement update_time > should work without any modification at all... > +1 Like from Boaz. I think Dave is very much right! > Cheers, Indeed Boaz > Dave. <>