From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757339AbaDVT6f (ORCPT ); Tue, 22 Apr 2014 15:58:35 -0400 Received: from casper.infradead.org ([85.118.1.10]:56008 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756528AbaDVT62 (ORCPT ); Tue, 22 Apr 2014 15:58:28 -0400 Date: Tue, 22 Apr 2014 13:56:29 +0200 From: Peter Zijlstra To: Alexander Gordeev Cc: linux-kernel@vger.kernel.org, Kent Overstreet , Jens Axboe , Shaohua Li , Nicholas Bellinger , Ingo Molnar Subject: Re: [PATCH RFC 0/2] percpu_ida: Take into account CPU topology when stealing tags Message-ID: <20140422115629.GI11182@twins.programming.kicks-ass.net> References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.21 (2012-12-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Mar 26, 2014 at 02:34:22PM +0100, Alexander Gordeev wrote: > Hello, > > This series is against 3.14.0-rc7. > > It is amied to further improve 'percpu_ida' tags locality by taking > into account system's CPU topology when stealing tags. That is try > to steal from a CPU which is 'closest' to the stealing one. > > I would not bother to post this, since on several system the change > did not show any improvement, i.e. on such one: There's is much lower level fruit to be had before doing something like this. https://lkml.org/lkml/2014/1/23/257 https://lkml.org/lkml/2014/1/23/329 https://lkml.org/lkml/2014/1/23/343 https://lkml.org/lkml/2014/1/23/354 I've not had time to revisit/finish them, but you should definitely clean up the percpu_ida stuff and reduce existing contention before going overboard.