From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755416AbYITA0R (ORCPT ); Fri, 19 Sep 2008 20:26:17 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752642AbYITA0F (ORCPT ); Fri, 19 Sep 2008 20:26:05 -0400 Received: from senator.holtmann.net ([87.106.208.187]:55383 "EHLO mail.holtmann.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752537AbYITA0F (ORCPT ); Fri, 19 Sep 2008 20:26:05 -0400 Subject: Re: Unified tracing buffer From: Marcel Holtmann To: Martin Bligh Cc: Linux Kernel Mailing List , Linus Torvalds , Thomas Gleixner , Mathieu Desnoyers , Steven Rostedt , od@novell.com, "Frank Ch. Eigler" In-Reply-To: <33307c790809191433w246c0283l55a57c196664ce77@mail.gmail.com> References: <33307c790809191433w246c0283l55a57c196664ce77@mail.gmail.com> Content-Type: text/plain Date: Sat, 20 Sep 2008 02:26:52 +0200 Message-Id: <1221870412.6782.69.camel@californication> Mime-Version: 1.0 X-Mailer: Evolution 2.22.3.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Martin, > During kernel summit and Plumbers conference, Linus and others > expressed a desire for a unified > tracing buffer system for multiple tracing applications (eg ftrace, > lttng, systemtap, blktrace, etc) to use. > This provides several advantages, including the ability to interleave > data from multiple sources, > not having to learn 200 different tools, duplicated code/effort, etc. I talked with Thomas and Steven about it during lunch and we might wanna also use it for sniffing/monitoring/capturing efforts. Inaky and I talked about unifying things like usbmon, hcidump, tcpdump etc. with a common interface into the kernel. Currently every subsystem does it differently. Especially when it comes to add monitor support to SDIO we ran into the problem that we don't just wanna invent another interface. Right now I haven't spent much time on this and regret that I didn't sit together with you guys yesterday, but it was not that high priority on my list of nice things to have. Anyway, since I am mostly interested in subsystems that copy a lot of packets around, it would be nice to only reference to the data. For example for structures like SKBs and URBs and only have to copy the data if we have a consumer. My thoughts :) Regards Marcel