From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755351AbYIAItc (ORCPT ); Mon, 1 Sep 2008 04:49:32 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751990AbYIAItY (ORCPT ); Mon, 1 Sep 2008 04:49:24 -0400 Received: from one.firstfloor.org ([213.235.205.2]:44736 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750840AbYIAItX (ORCPT ); Mon, 1 Sep 2008 04:49:23 -0400 Date: Mon, 1 Sep 2008 10:52:17 +0200 From: Andi Kleen To: Rusty Russell Cc: Andi Kleen , paulmck@us.ibm.com, linux-kernel@vger.kernel.org, akpm@osdl.org Subject: Re: [PATCH] Remove stop_machine during module load Message-ID: <20080901085217.GA18288@one.firstfloor.org> References: <20080829191734.GA28329@basil.nowhere.org> <200809011702.00956.rusty@rustcorp.com.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200809011702.00956.rusty@rustcorp.com.au> User-Agent: Mutt/1.4.2.1i Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Sep 01, 2008 at 05:02:00PM +1000, Rusty Russell wrote: > On Saturday 30 August 2008 05:17:34 Andi Kleen wrote: > > Remove stop_machine during module load > > > > module loading currently does a stop_machine on each module load to insert > > the module into the global module lists. Especially on larger systems this > > can be quite expensive. > > Thanks Andi, but doesn't this mean that we should be doing list_for_each_rcu > in all the kallsyms readers? Yes I did that in v2. But only for the ones that don't hold modules_mutex anyways. Also I found that print_modules() does not always disable preemption. -Andi -- ak@linux.intel.com