From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 25192 invoked by alias); 22 Mar 2006 23:46:45 -0000 Received: (qmail 25184 invoked by uid 22791); 22 Mar 2006 23:46:44 -0000 X-Spam-Status: No, hits=-2.6 required=5.0 tests=BAYES_00 X-Spam-Check-By: sourceware.org Received: from mga03.intel.com (HELO azsmga101-1.ch.intel.com) (143.182.124.21) by sourceware.org (qpsmtpd/0.31) with ESMTP; Wed, 22 Mar 2006 23:46:43 +0000 Received: from azsmga001.ch.intel.com ([10.2.17.19]) by azsmga101-1.ch.intel.com with ESMTP; 22 Mar 2006 15:46:42 -0800 Received: from scsmsx332.sc.intel.com (HELO scsmsx332.amr.corp.intel.com) ([10.3.90.6]) by azsmga001.ch.intel.com with ESMTP; 22 Mar 2006 15:46:42 -0800 X-IronPort-AV: i="4.03,120,1141632000"; d="scan'208"; a="14410582:sNHT99449651" Received: from scsmsx403.amr.corp.intel.com ([10.3.90.18]) by scsmsx332.amr.corp.intel.com with Microsoft SMTPSVC(6.0.3790.211); Wed, 22 Mar 2006 15:46:41 -0800 X-MimeOLE: Produced By Microsoft Exchange V6.5 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Subject: RE: Proposed systemtap access to perfmon hardware Date: Wed, 22 Mar 2006 23:46:00 -0000 Message-ID: X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: Proposed systemtap access to perfmon hardware Thread-Index: AcZN/mQX71oboISnTw678HD5aQiVlQACT5/Q From: "Stone, Joshua I" To: , "William Cohen" Cc: "SystemTAP" X-OriginalArrivalTime: 22 Mar 2006 23:46:41.0999 (UTC) FILETIME=[DEB9B5F0:01C64E0A] X-IsSubscribed: yes Mailing-List: contact systemtap-help@sourceware.org; run by ezmlm Precedence: bulk List-Subscribe: List-Post: List-Help: , Sender: systemtap-owner@sourceware.org X-SW-Source: 2006-q1/txt/msg00858.txt.bz2 Maynard Johnson wrote: > William Cohen wrote: >> The individual start and stop operations would be allowed. > This is not so good. Besides the fact that it may be difficult (or > impossible) to do, I don't see it being all that useful. But then, > I'm a tool developer, not a performance analyst, so I could be > missing the point. Enabling start & stop lets you narrow the context that you want to measure. Perfmon can only give you thread level virtualization of the counters. With start & stop I can, for example, start the counters when I enter sys_open and stop when I return. Now if I want I can get a microbenchmark of IPC for the sys_open call (and its callees). But this also opens up possibilities for more obscure "contexts" - perhaps I want to start counting when a network packet is received and stop when it is delivered to the thread. Any probepoint you can do today can become a start/stop point for the counters. Josh