From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757477AbZJPIdb (ORCPT ); Fri, 16 Oct 2009 04:33:31 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751591AbZJPIda (ORCPT ); Fri, 16 Oct 2009 04:33:30 -0400 Received: from mx2.mail.elte.hu ([157.181.151.9]:46627 "EHLO mx2.mail.elte.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751382AbZJPId3 (ORCPT ); Fri, 16 Oct 2009 04:33:29 -0400 Date: Fri, 16 Oct 2009 10:32:42 +0200 From: Ingo Molnar To: Christoph Hellwig Cc: fweisbec@gmail.com, linux-kernel@vger.kernel.org, Steven Rostedt , Peter Zijlstra , Paul Mackerras Subject: Re: perf trace errors Message-ID: <20091016083242.GA11174@elte.hu> References: <20091011025441.GA18152@lst.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20091011025441.GA18152@lst.de> User-Agent: Mutt/1.5.18 (2008-05-17) X-ELTE-SpamScore: -1.5 X-ELTE-SpamLevel: X-ELTE-SpamCheck: no X-ELTE-SpamVersion: ELTE 2.0 X-ELTE-SpamCheck-Details: score=-1.5 required=5.9 tests=BAYES_00 autolearn=no SpamAssassin version=3.2.5 -1.5 BAYES_00 BODY: Bayesian spam probability is 0 to 1% [score: 0.0000] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Christoph Hellwig wrote: > I get some rather interesting errors using perf trace with the block > tracepoints on current mainline, e.g.: > > perf record -f -e 'block:block_bio_queue:record' -R -F 1 -a -s > > ^C[ perf record: Woken up 1 times to write data ] > [ perf record: Captured and wrote 0.237 MB perf.data (~10362 samples) ] > > brick:/home/hch# perf trace > Fatal: unknown op '-' > xfssyncd-4232 [000] 87.798249255: block_bio_queue: 253, > > There are lots of occurances of that exactly same warning in > builtin-trace.c, and the code sourrouning them is sufficiently enough > convoluted that I'm not able to track it down. FYI, these bugs are fixed in latest -tip, your testcase works for me: aldebaran:~/linux/linux/tools/perf> perf record -f -e 'block:block_bio_queue:record' -R -F 1 -a -s ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.027 MB perf.data (~1168 samples) ] aldebaran:~/linux/linux/tools/perf> perf trace kjournald-1406 [000] 10900.853086: block_bio_queue: 8,0 W 68735 + 8 [kjournald] kjournald-1406 [000] 10900.853097: block_bio_queue: 8,0 W 68743 + 8 [kjournald] kjournald-1406 [000] 10900.853100: block_bio_queue: 8,0 W 68751 + 8 [kjournald] kjournald-1406 [000] 10900.853103: block_bio_queue: 8,0 W 68759 + 8 [kjournald] flush-8:0-20110 [007] 10900.853117: block_bio_queue: 8,0 W 64487535 + 8 [flush-8:0] flush-8:0-20110 [007] 10900.853129: block_bio_queue: 8,0 W 64487551 + 8 [flush-8:0] flush-8:0-20110 [007] 10900.853138: block_bio_queue: 8,0 W 64487559 + 8 [flush-8:0] flush-8:0-20110 [007] 10900.853149: block_bio_queue: 8,0 W 64700511 + 8 [flush-8:0] sync-16876 [008] 10900.851948: block_bio_queue: 8,0 R 3218919 + 40 [sync] sync-16876 [008] 10900.853062: block_bio_queue: 8,0 W 63 + 8 [sync] sync-16876 [008] 10900.853067: block_bio_queue: 8,0 W 71 + 8 [sync] sync-16876 [008] 10900.853071: block_bio_queue: 8,0 W 79 + 8 [sync] sync-16876 [008] 10900.853077: block_bio_queue: 8,0 W 15466559 + 8 [sync] sync-16876 [011] 10900.840473: block_bio_queue: 8,0 R 3218887 + 32 [bash] Mind checking whether it works for you? Thanks, Ingo