From mboxrd@z Thu Jan 1 00:00:00 1970 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757610Ab0AOOBE (ORCPT ); Fri, 15 Jan 2010 09:01:04 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1756691Ab0AOOBD (ORCPT ); Fri, 15 Jan 2010 09:01:03 -0500 Received: from mx1.redhat.com ([209.132.183.28]:10623 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756380Ab0AOOBC (ORCPT ); Fri, 15 Jan 2010 09:01:02 -0500 Date: Fri, 15 Jan 2010 09:00:42 -0500 From: "Frank Ch. Eigler" To: Peter Zijlstra Cc: Jim Keniston , Arnaldo Carvalho de Melo , Frederic Weisbecker , LKML , Mark Wielaard , utrace-devel Subject: Re: [RFC] [PATCH 4/7] Uprobes Implementation Message-ID: <20100115140042.GR4822@redhat.com> References: <20100111122553.22050.46895.sendpatchset@srikar.in.ibm.com> <1263467394.4244.291.camel@laptop> <1263509380.4875.35.camel@localhost.localdomain> <1263546632.4244.352.camel@laptop> <1263548124.4244.358.camel@laptop> <20100115131037.GP4822@redhat.com> <1263561930.4244.417.camel@laptop> <20100115133825.GQ4822@redhat.com> <1263563276.4244.426.camel@laptop> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1263563276.4244.426.camel@laptop> User-Agent: Mutt/1.4.2.2i Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi - On Fri, Jan 15, 2010 at 02:47:56PM +0100, Peter Zijlstra wrote: > [...] > > I'm not sure, but it sounds like the part you're complaining about is > > how utrace ultimately reports the trap to uprobes: i.e., > > utrace_get_signal()? Is that the "insane amount of code"? > > Well when tracing/profiling every instruction is too much. Having to > needlessly raise a signal only to catch it again a short bit later > sounds like obvious waste to me. Well, I'm not in a position to argue line by line about the necessity or the cost of utrace low level guts, but this may represent the most practical engineering balance between functionality / modularity / undesirably intrusive modifications. Perhaps there exists a tool with which one can confirm your worry about excess cost of this particular piece. > > > Furthermore it requires stopping and resuming tasks and nonsense like > > > that, that's unwanted in many cases, just run stuff from the trap site > > > and you're done. > > > > I don't know what you mean exactly. A trap already stopped task. > > utrace merely allows various clients to inspect/manipulate the state > > of the task at that moment. It does not add any context switches or > > spurious stop/resumue operations. > > Srikar seemed to suggest it needed stop/resume. You may be confusing breakpoint insertion/removal operations versus breakpoint hits. - FChE