From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ted Ts'o Subject: Re: ext4_fallocate Date: Tue, 26 Jun 2012 14:59:49 -0400 Message-ID: <20120626185949.GB26669@thunk.org> References: <4FE8086F.4070506@zoho.com> <20120625085159.GA18931@gmail.com> <20120625191744.GB9688@thunk.org> <4FE9B57F.4030704@redhat.com> <4FE9F9F4.7010804@zoho.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Ric Wheeler , linux-ext4@vger.kernel.org, Andreas Dilger , wenqing.lz@taobao.com To: Fredrick Return-path: Received: from li9-11.members.linode.com ([67.18.176.11]:52859 "EHLO imap.thunk.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753432Ab2FZS7v (ORCPT ); Tue, 26 Jun 2012 14:59:51 -0400 Content-Disposition: inline In-Reply-To: <4FE9F9F4.7010804@zoho.com> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Tue, Jun 26, 2012 at 11:05:40AM -0700, Fredrick wrote: > > I had run perf stat on ext4 functions between two runs of our program > writing data to a file for the first time and writing data to the file > for the second time(where the extents are initialized). >>From your mballoc differences, it sounds like you were comparing fallocate with not using fallocate at all; is that right? The comparison you need to do is using normal fallocate versus fallocate with the no-hide-stale feature enabled. It's obvious that allocating blocks as you need will always be more expensive than using fallocate. - Ted