From mboxrd@z Thu Jan 1 00:00:00 1970 From: Hans Reiser Subject: Re: ext3 file system Date: Thu, 18 Dec 2003 13:41:18 +0300 Sender: linux-fsdevel-owner@vger.kernel.org Message-ID: <3FE1844E.9070700@namesys.com> References: <3FE23273@webmail.colostate.edu> <20031218083957.GA6438@matchmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Cc: jshankar , linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org Return-path: Received: from thebsh.namesys.com ([212.16.7.65]:33979 "HELO thebsh.namesys.com") by vger.kernel.org with SMTP id S265059AbTLRKlU (ORCPT ); Thu, 18 Dec 2003 05:41:20 -0500 To: Mike Fedyk In-Reply-To: <20031218083957.GA6438@matchmail.com> List-Id: linux-fsdevel.vger.kernel.org Mike Fedyk wrote: >On Wed, Dec 17, 2003 at 09:47:59PM -0700, jshankar wrote: > > >>Hello Hans, >> >> >> >>>Filesystems don't usually wait on the IO to complete before submitting >>>more IO in response to the next write() syscall. They can do this by >>>batching a whole bunch of operations into one committed transaction. >>> >>> >>> >>Is there a timeout mechanism for batching operations. >> At some point due to its age or size you decide the batch needs to commit. >> What if certain >>operation >>is done after the batch operation is executed. Does it mean that the new >>operation has to wait. >> >> > >You don't have to wait unless you run out of available non-dirty memory, or >issue a call to sync to the disks. > > > > -- Hans