From: Linus Torvalds <torvalds@linux-foundation.org>
To: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Ingo Molnar <mingo@elte.hu>,
linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org,
Steven Rostedt <rostedt@goodmis.org>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Thomas Gleixner <tglx@linutronix.de>,
David Miller <davem@davemloft.net>,
Ingo Molnar <mingo@redhat.com>, "H. Peter Anvin" <hpa@zytor.com>
Subject: Re: [RFC patch 15/15] LTTng timestamp x86
Date: Thu, 16 Oct 2008 17:12:31 -0700 (PDT) [thread overview]
Message-ID: <alpine.LFD.2.00.0810161709230.3288@nehalem.linux-foundation.org> (raw)
In-Reply-To: <alpine.LFD.2.00.0810161701470.3288@nehalem.linux-foundation.org>
On Thu, 16 Oct 2008, Linus Torvalds wrote:
>
> Perhaps more importantly - if the TSC really are out of whack, that just
> means that now all your timestamps are worthless, because the value you
> calculate ends up having NOTHING to do with the timestamp. So you cannot
> even use it to see how long something took, because it may be that you're
> running on the CPU that runs behind, and all you ever see is the value of
> LTT_MIN_PROBE_DURATION.
If it isn't clear: the alternative is to just always use local timestamps.
At least that way the timestamps mean _something_. You can get the
difference between two events when they happen on the same CPU, and it is
about as meaningful as it can be.
Don't even _try_ to make a global clock.
Yes, to be able to compare across CPU's you'd need to have extra
synchronization information (eg offset and frequency things), but quite
frankly, the "global TSC" thing is already worse than even a totally
non-synchronized TSC for the above reasons.
Linus
next prev parent reply other threads:[~2008-10-17 0:13 UTC|newest]
Thread overview: 66+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-10-16 23:27 [RFC patch 00/15] Tracer Timestamping Mathieu Desnoyers
2008-10-16 23:27 ` [RFC patch 01/15] get_cycles() : kconfig HAVE_GET_CYCLES Mathieu Desnoyers
2008-10-16 23:27 ` [RFC patch 02/15] get_cycles() : x86 HAVE_GET_CYCLES Mathieu Desnoyers
2008-10-16 23:27 ` [RFC patch 03/15] get_cycles() : sparc64 HAVE_GET_CYCLES Mathieu Desnoyers
2008-10-17 2:48 ` [RFC patch 03/15] get_cycles() : sparc64 HAVE_GET_CYCLES (update) Mathieu Desnoyers
2008-10-17 2:57 ` David Miller
2008-10-16 23:27 ` [RFC patch 04/15] get_cycles() : powerpc64 HAVE_GET_CYCLES Mathieu Desnoyers
2008-10-17 0:26 ` Paul Mackerras
2008-10-17 0:43 ` [RFC patch 04/15] get_cycles() : powerpc64 HAVE_GET_CYCLES (update) Mathieu Desnoyers
2008-10-17 0:54 ` Paul Mackerras
2008-10-17 1:42 ` David Miller
2008-10-17 2:08 ` Mathieu Desnoyers
2008-10-17 2:33 ` David Miller
2008-10-16 23:27 ` [RFC patch 05/15] get_cycles() : MIPS HAVE_GET_CYCLES_32 Mathieu Desnoyers
2008-10-26 10:18 ` Ralf Baechle
2008-10-26 20:39 ` Mathieu Desnoyers
2008-10-16 23:27 ` [RFC patch 06/15] LTTng build Mathieu Desnoyers
2008-10-17 8:10 ` KOSAKI Motohiro
2008-10-17 16:18 ` Mathieu Desnoyers
2008-10-16 23:27 ` [RFC patch 07/15] LTTng timestamp Mathieu Desnoyers
2008-10-17 8:15 ` KOSAKI Motohiro
2008-10-17 16:23 ` Mathieu Desnoyers
2008-10-16 23:27 ` [RFC patch 08/15] LTTng - Timestamping Mathieu Desnoyers
2008-10-16 23:27 ` [RFC patch 09/15] LTTng mips export hpt frequency Mathieu Desnoyers
2008-10-16 23:27 ` [RFC patch 10/15] LTTng timestamp mips Mathieu Desnoyers
2008-10-16 23:27 ` [RFC patch 11/15] LTTng timestamp powerpc Mathieu Desnoyers
2008-10-16 23:27 ` [RFC patch 12/15] LTTng timestamp sparc64 Mathieu Desnoyers
2008-10-16 23:27 ` [RFC patch 13/15] LTTng timestamp sh Mathieu Desnoyers
2008-10-16 23:27 ` [RFC patch 14/15] LTTng - TSC synchronicity test Mathieu Desnoyers
2008-10-16 23:27 ` [RFC patch 15/15] LTTng timestamp x86 Mathieu Desnoyers
2008-10-17 0:08 ` Linus Torvalds
2008-10-17 0:12 ` Linus Torvalds [this message]
2008-10-17 1:28 ` Mathieu Desnoyers
2008-10-17 2:19 ` Luck, Tony
2008-10-17 17:25 ` Steven Rostedt
2008-10-17 18:08 ` Luck, Tony
2008-10-17 18:42 ` Mathieu Desnoyers
2008-10-17 18:58 ` Luck, Tony
2008-10-17 20:23 ` Mathieu Desnoyers
2008-10-17 23:52 ` Luck, Tony
2008-10-18 17:01 ` Mathieu Desnoyers
2008-10-18 17:35 ` Linus Torvalds
2008-10-18 17:50 ` Ingo Molnar
2008-10-22 16:19 ` Mathieu Desnoyers
2008-10-22 15:53 ` Mathieu Desnoyers
2008-10-20 18:07 ` Luck, Tony
2008-10-22 16:51 ` Mathieu Desnoyers
2008-10-17 19:17 ` Steven Rostedt
2008-10-20 20:10 ` Linus Torvalds
2008-10-20 21:38 ` john stultz
2008-10-20 22:06 ` Linus Torvalds
2008-10-20 22:17 ` Ingo Molnar
2008-10-20 22:29 ` H. Peter Anvin
2008-10-21 18:10 ` Bjorn Helgaas
2008-10-23 15:47 ` Linus Torvalds
2008-10-23 16:39 ` H. Peter Anvin
2008-10-23 21:54 ` Paul Mackerras
2008-10-20 23:47 ` john stultz
2008-10-22 17:05 ` Mathieu Desnoyers
2008-10-17 19:36 ` Christoph Lameter
2008-10-17 7:59 ` [RFC patch 00/15] Tracer Timestamping Peter Zijlstra
2008-10-20 20:25 ` Mathieu Desnoyers
2008-10-21 0:20 ` Nicolas Pitre
2008-10-21 1:32 ` Mathieu Desnoyers
2008-10-21 2:32 ` Nicolas Pitre
2008-10-21 4:05 ` Mathieu Desnoyers
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=alpine.LFD.2.00.0810161709230.3288@nehalem.linux-foundation.org \
--to=torvalds@linux-foundation.org \
--cc=a.p.zijlstra@chello.nl \
--cc=akpm@linux-foundation.org \
--cc=davem@davemloft.net \
--cc=hpa@zytor.com \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@polymtl.ca \
--cc=mingo@elte.hu \
--cc=mingo@redhat.com \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®