Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752313AbZDMUMa (ORCPT ); Mon, 13 Apr 2009 16:12:30 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752192AbZDMUMS (ORCPT ); Mon, 13 Apr 2009 16:12:18 -0400 Received: from mx2.mail.elte.hu ([157.181.151.9]:55128 "EHLO mx2.mail.elte.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751327AbZDMUMR (ORCPT ); Mon, 13 Apr 2009 16:12:17 -0400 Date: Mon, 13 Apr 2009 22:11:55 +0200 From: Ingo Molnar To: Jens Axboe , Steven Rostedt , Tom Zanussi Cc: Li Zefan , "Theodore Ts'o" , Arnaldo Carvalho de Melo , LKML Subject: Re: [PATCH] blktrace: move trace/ dir to /sys/block/sda/ Message-ID: <20090413201155.GA17450@elte.hu> References: <49E31931.5000506@cn.fujitsu.com> <20090413180054.GZ5178@kernel.dk> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20090413180054.GZ5178@kernel.dk> User-Agent: Mutt/1.5.18 (2008-05-17) X-ELTE-VirusStatus: clean 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=3.2.3 -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 Content-Length: 1693 Lines: 40 * Jens Axboe wrote: > On Mon, Apr 13 2009, Li Zefan wrote: > > Impact: allow ftrace-plugin blktrace to trace device-mapper devices > > > > blktrace can't trace a single partition, so it makes no sense to > > have one trace/ dir in each /sys/block/sda/sdaX. Move it to > > /sys/block/sda/. > > > > Thus we fix an issue reported by Ted, that ftrace-plugin > > blktrace can't be used to trace device-mapper devices. > > Perhaps I never committed that patch, but it would be trivial to > do partition based blktrace tracing. It's also quite useful. So > please don't go changing things to make that harder to support, it > would be nicer to just add the (small) bits to support > per-partition tracing. It's basically just a start/stop sector > range, while some events are per-device and should just be > included always. btw., per tracepoint filters could be used for that. That would allow multiple partitions to be traced at once. blktrace user-space could make use of sector range filters straight away [ hm, Tom - do we have the <= comparison operator already, or is that still WIP? ] - but i think it's better to do this in a more integrated way: via the sysfs API, via /sys/block/sda/sda2/trace/. So when a partition's trace entry is activated, it would auto-install a specific filter expression for sda, with the sector range of that partition. Or something like that. How does this sound? Ingo -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/