From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Martin K. Petersen" Subject: Re: How to make kernel block layer generate bigger request in the request queue? Date: Mon, 12 Apr 2010 14:26:41 -0400 Message-ID: References: <4BBFBE3E.5070605@gmail.com> <1270911503.2806.194.camel@mulgrave.site> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from rcsinet12.oracle.com ([148.87.113.124]:63757 "EHLO rcsinet12.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752659Ab0DLS2o (ORCPT ); Mon, 12 Apr 2010 14:28:44 -0400 In-Reply-To: <1270911503.2806.194.camel@mulgrave.site> (James Bottomley's message of "Sat, 10 Apr 2010 09:58:23 -0500") Sender: linux-mmc-owner@vger.kernel.org List-Id: linux-mmc@vger.kernel.org To: James Bottomley Cc: "Martin K. Petersen" , Robert Hancock , "Gao, Yunpeng" , "linux-ide@vger.kernel.org" , "linux-mmc@vger.kernel.org" >>>>> "James" == James Bottomley writes: >> Correct. It's quite unlikely for pages to be contiguous so this is >> the best we can do. James> Actually, average servers do about 50% contiguous on average James> since we changed the mm layer to allocate in ascending physical James> page order ... this figure is highly sensitive to mm changes James> though, and can vary from release to release. Interesting. When did this happen? Last time I gathered data on segment merge efficiency (1 year+ ago) I found that adjacent pages were quite rare for a normal fs type workload. Certainly not in the 50% ballpark. I'll take another look when I have a moment... -- Martin K. Petersen Oracle Linux Engineering