From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail02.iobjects.de ([188.40.134.68]:44968 "EHLO mail02.iobjects.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752338AbcFTSQj (ORCPT ); Mon, 20 Jun 2016 14:16:39 -0400 Subject: Re: What happened to the sb writeback list (aka sync efficiency) fix? To: Brian Foster References: <20160620164334.GB9474@laptop.bfoster> Cc: linux-fsdevel@vger.kernel.org From: =?UTF-8?Q?Holger_Hoffst=c3=a4tte?= Message-ID: <57682D18.6030005@applied-asynchrony.com> Date: Mon, 20 Jun 2016 19:51:20 +0200 MIME-Version: 1.0 In-Reply-To: <20160620164334.GB9474@laptop.bfoster> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On 06/20/16 18:43, Brian Foster wrote: > On Mon, Jun 20, 2016 at 01:46:38PM +0000, Holger Hoffstätte wrote: >> >> Once upon a time there was this fine patch set called >> "improve sync efficiency with sb inode wb list" [1] >> by Brian Foster, who fixed up the original version by Josef Bacik. >> >> I've been running with this since then and it seems to work flawlessly, >> yet it doesn't seem that this ever got merged..does anybody know why? >> >> Waiman Long has been working on something similar with his per-CPU >> lists, but those patches naturally collide a bit, so I'm wondering >> what's what. >> >> Fwiw the effect of the wb list on systems with many cached inodes is >> phenomal; it would be a shame if this went unmerged. >> > > FWIW, the latest version posted was v7: > > http://thread.gmane.org/gmane.linux.file-systems/104078 Ah yes..that was a comment update so I didn't merge it. > Thanks for the testing feedback. Unfortunately, I've not really heard > any feedback on getting this merged. I'm not sure the previous > version(s) by Josef and Dave got much traction either. :( The early versions by Josef had a lockdep issue, which your version apparently fixed. I've not had a single problem with it..but that admittedly doesn't mean anything. At some point I tried merging it with the per-CPU list patches (for fun :) but that turned out a bit too hairy for me. cheers, Holger