From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1030194AbWJRK7g (ORCPT ); Wed, 18 Oct 2006 06:59:36 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1030191AbWJRK7g (ORCPT ); Wed, 18 Oct 2006 06:59:36 -0400 Received: from cantor.suse.de ([195.135.220.2]:17073 "EHLO mx1.suse.de") by vger.kernel.org with ESMTP id S1751056AbWJRK7T (ORCPT ); Wed, 18 Oct 2006 06:59:19 -0400 From: Andi Kleen To: "Jan Beulich" Subject: Re: [PATCH] Re: UNWIND_INFO slowdown in -mm1 Date: Wed, 18 Oct 2006 12:50:41 +0200 User-Agent: KMail/1.9.3 Cc: "Ingo Molnar" , "Andrew Morton" , "Thomas Gleixner" , linux-kernel@vger.kernel.org References: <20060928192048.GA17436@elte.hu> <45351782.76E4.0078.0@novell.com> In-Reply-To: <45351782.76E4.0078.0@novell.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200610181250.41423.ak@suse.de> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Tuesday 17 October 2006 17:48, Jan Beulich wrote: > >in 2.6.18-mm1 we are seeing _really_ long delays in the unwind code. > >(full trace attached) On an Athlon64 3800+ CPU: > > Below patch should help, can you try it out? Short of the linker > supporting building a binary lookup table at build time, it creates > one as soon as the bootmem allocator is usable (so you'll continue > using the linear lookup for the first [hopefully] few calls). > The code should be ready to utilize a build-time created table once > a fixed linker becomes available. I added the patch now, thanks. Not sure this is still .19 material though or better delayed for .20. What do you think? -Andi