From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Ahern Subject: Re: [PATCH 2/3] perf events: Introduce realtime clock event Date: Fri, 18 Feb 2011 08:39:47 -0700 Message-ID: <4D5E92C3.70003@cisco.com> References: <1298008433-22911-1-git-send-email-daahern@cisco.com> <1298008433-22911-3-git-send-email-daahern@cisco.com> <1298027641.5226.666.camel@laptop> <4D5E84AC.4040104@cisco.com> <1298041106.5226.775.camel@laptop> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Return-path: Received: from sj-iport-4.cisco.com ([171.68.10.86]:38390 "EHLO sj-iport-4.cisco.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752003Ab1BRPjt (ORCPT ); Fri, 18 Feb 2011 10:39:49 -0500 In-Reply-To: <1298041106.5226.775.camel@laptop> Sender: linux-perf-users-owner@vger.kernel.org List-ID: To: Peter Zijlstra Cc: linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, mingo@elte.hu, acme@ghostprotocols.net, paulus@samba.org On 02/18/11 07:58, Peter Zijlstra wrote: >>> I'm really not sure why you want CLOCK_REALTIME and I think >>> CLOCK_MONOTONIC is more useful (I'd argue you want your system logs to >>> contain both, every admin who's ever had to untangle what happened >>> during DST switches will agree) >> >> I believe CLOCK_MONOTONIC is what perf_clock is tied to -- the >> timestamps for PERF_SAMPLE_TIME -- so we already have that. > > Its not (it mere _can_ be), it could be tied to the TSC which can > significantly drift wrt CLOCK_MONOTONIC. Ok, either way I would like correlation between perf_clock and the time sample data and gettimeofday. > >> Programs that generate time-of-day output are using gettimeofday which >> is tied to CLOCK_REALTIME. We want to be able to correlate a perf sample >> to an entry in an applications log file. > > Well, you can argue those programs are broken :-), Imagine the joys of > trying to figure out wth happens when DST jumps the clock back an hour > and you have an hour of duplicate data. > Luckily DST only happens twice a year. Of course reboots happen a little more often and those reset the monotonic clock. I can't change the known universe of programs that create pretty HH:MM:SS MM/DD/YY time strings. What I do want is to know why a program missed a heartbeat as noted by a log entry. Correlating with a perf event and seeing the backtrace is quite handy. David