From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S267194AbUHWAaV (ORCPT ); Sun, 22 Aug 2004 20:30:21 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S267264AbUHWAaV (ORCPT ); Sun, 22 Aug 2004 20:30:21 -0400 Received: from mail.ocs.com.au ([202.147.117.210]:40644 "EHLO mail.ocs.com.au") by vger.kernel.org with ESMTP id S267194AbUHWAaR (ORCPT ); Sun, 22 Aug 2004 20:30:17 -0400 X-Mailer: exmh version 2.6.3_20040314 03/14/2004 with nmh-1.0.4 From: Keith Owens To: Zwane Mwaikambo Cc: Linus Torvalds , Linux Kernel , Andrew Morton , Andi Kleen , William Lee Irwin III Subject: Re: [PATCH][1/4] Completely out of line spinlocks / i386 In-reply-to: Your message of "Sun, 22 Aug 2004 17:53:29 -0400." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 23 Aug 2004 10:29:48 +1000 Message-ID: <22241.1093220988@ocs3.ocs.com.au> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sun, 22 Aug 2004 17:53:29 -0400 (EDT), Zwane Mwaikambo wrote: >With the readprofile and oprofile changes it's still not that easy to >determine which locks are being contended as the samples generally are >being charged to the function the lock is being contended in. So some >investigation has to be done when looking at profiles. This could be >remedied by making the valid PC range include data or, preferably, moving >spinlock variables into a special section. That way we can simply >report back the lock word during sampling. kdb attempts to decode the lock address on ia64. A lot of the time, the lock is dynamically allocated (think inodes) so symbol lookup is no good. I find that the decoding the lock is useful but not required, the function that contended on the lock is more interesting.