From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 24820 invoked by alias); 23 Sep 2008 02:27:20 -0000 Received: (qmail 24812 invoked by uid 22791); 23 Sep 2008 02:27:19 -0000 X-Spam-Status: No, hits=-1.0 required=5.0 tests=AWL,BAYES_00,SARE_MSGID_LONG40 X-Spam-Check-By: sourceware.org Received: from wa-out-1112.google.com (HELO wa-out-1112.google.com) (209.85.146.179) by sourceware.org (qpsmtpd/0.31) with ESMTP; Tue, 23 Sep 2008 02:26:38 +0000 Received: by wa-out-1112.google.com with SMTP id j5so1474747wah.18 for ; Mon, 22 Sep 2008 19:26:36 -0700 (PDT) Received: by 10.114.195.19 with SMTP id s19mr5644761waf.59.1222136796264; Mon, 22 Sep 2008 19:26:36 -0700 (PDT) Received: by 10.114.108.20 with HTTP; Mon, 22 Sep 2008 19:26:36 -0700 (PDT) Message-ID: Date: Tue, 23 Sep 2008 02:27:00 -0000 From: "Darren Hart" To: "Mathieu Desnoyers" Subject: Re: Unified tracing buffer Cc: "Roland Dreier" , "Linus Torvalds" , "Masami Hiramatsu" , "Martin Bligh" , "Linux Kernel Mailing List" , "Thomas Gleixner" , "Steven Rostedt" , "Frank Ch. Eigler" , systemtap-ml In-Reply-To: <20080923020216.GC24937@Krystal> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <33307c790809191433w246c0283l55a57c196664ce77@mail.gmail.com> <48D7F5E8.3000705@redhat.com> <33307c790809221313s3532d851g7239c212bc72fe71@mail.gmail.com> <48D81B5F.2030702@redhat.com> <33307c790809221616h5e7410f5gc37c262d83722111@mail.gmail.com> <48D832B6.3010409@redhat.com> <20080923020216.GC24937@Krystal> X-Google-Sender-Auth: 3d63f44a910f46f2 Mailing-List: contact systemtap-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Post: List-Help: , Sender: systemtap-owner@sourceware.org X-SW-Source: 2008-q3/txt/msg00743.txt.bz2 On Mon, Sep 22, 2008 at 7:02 PM, Mathieu Desnoyers > So only in the specific case of instrumentation of things like locking, > where it is possible to insure that instrumentation is synchronized with > the instrumented operation, does it make a difference to choose the TSC > (which implies a slight delta between the TSCs due to cache line delays > at synchronization and delay due to TSCs drifts caused by temperature) > over an atomic increment. > Hrm, i think that overlooks the other reason to use a time based counter over an atomic increment: you might care about time. Perhaps one might be less concerned with actual order tightly grouped events and more concerned with the actual time delta between more temporally distant events. In that case, using a clocksource would still be valuable. Although admitedtly the caller could embed that in their payload, but since we seem to agree we need some kind of counter, the time-based counter appears to be the most flexible. Thanks, -- Darren Hart