From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751822AbZIBTRu (ORCPT ); Wed, 2 Sep 2009 15:17:50 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751425AbZIBTRt (ORCPT ); Wed, 2 Sep 2009 15:17:49 -0400 Received: from casper.infradead.org ([85.118.1.10]:34173 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751188AbZIBTRt (ORCPT ); Wed, 2 Sep 2009 15:17:49 -0400 Subject: Re: tree rcu: call_rcu scalability problem? From: Peter Zijlstra To: Nick Piggin Cc: Paul McKenney , Linux Kernel Mailing List In-Reply-To: <20090902122756.GC12251@wotan.suse.de> References: <20090902094835.GB12251@wotan.suse.de> <20090902122756.GC12251@wotan.suse.de> Content-Type: text/plain Date: Wed, 02 Sep 2009 21:17:44 +0200 Message-Id: <1251919064.10394.25.camel@laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.26.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 2009-09-02 at 14:27 +0200, Nick Piggin wrote: > It seems like nearly 2/3 of the cost is here: > /* Add the callback to our list. */ > *rdp->nxttail[RCU_NEXT_TAIL] = head; <<< > rdp->nxttail[RCU_NEXT_TAIL] = &head->next; > > In loading the pointer to the next tail pointer. If I'm reading the profile > correctly. Can't see why that should be a probem though... > > ffffffff8107dee0 <__call_rcu>: /* __call_rcu total: 320971 100.000 */ > 697 0.2172 :ffffffff8107dee0: push %r12 > 921 0.2869 :ffffffff8107df57: push %rdx > 151 0.0470 :ffffffff8107df58: popfq > 183507 57.1725 :ffffffff8107df59: mov 0x50(%rbx),%rax > 995 0.3100 :ffffffff8107df5d: mov %rdi,(%rax) I'd guess at popfq to be the expensive op here.. skid usually causes the attribution to be a few ops down the line.