From mboxrd@z Thu Jan 1 00:00:00 1970 From: Martin Jambor Subject: Re: Simulated Ordered Mode Date: Sun, 31 Jul 2005 21:27:19 +0200 Message-ID: <8e70aacf050731122777773017@mail.gmail.com> References: <8e70aacf0507310702ad09c99@mail.gmail.com> <200507311907.WAA07896@raad.intranet> Reply-To: Martin Jambor Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Cc: linux-fsdevel@vger.kernel.org Return-path: Received: from zproxy.gmail.com ([64.233.162.207]:15878 "EHLO zproxy.gmail.com") by vger.kernel.org with ESMTP id S261943AbVGaT10 convert rfc822-to-8bit (ORCPT ); Sun, 31 Jul 2005 15:27:26 -0400 Received: by zproxy.gmail.com with SMTP id r28so688185nza for ; Sun, 31 Jul 2005 12:27:19 -0700 (PDT) To: Al Boldi In-Reply-To: <200507311907.WAA07896@raad.intranet> Content-Disposition: inline Sender: linux-fsdevel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On 7/31/05, Al Boldi wrote: > Martin Jambor wrote: { > On 7/31/05, Al Boldi wrote: > > Is it possible to instruct the FS to delay metadata update until after > > a filedata sync? > > If you delayed any update until after a sync it wouldn't be a sync anymore, > would it? > } > > True, but what about an implied MetaData sync _after_ a FileData sync? What do you mean by "sync?" If you mean an operation that brings an on disk image of a filesystem into a consistent state then you cannot delay any writeout until after it. If you mean an ordinary update of the file (and directory) contents then such capability must be coded at the filesystem level because only the specific filesystem knows what are metadata and what are data and what files different metadata belong to. HTH Martin