All of lore.kernel.org
 help / color / mirror / Atom feed
From: Glauber Costa <glommer@redhat.com>
To: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
	avi@redhat.com, Marcelo Tosatti <mtosatti@redhat.com>,
	Zachary Amsden <zamsden@redhat.com>
Subject: Re: [PATCH 1/5] Add a global synchronization point for pvclock
Date: Mon, 19 Apr 2010 15:25:43 -0300	[thread overview]
Message-ID: <20100419182542.GI14158@mothafucka.localdomain> (raw)
In-Reply-To: <4BCC829A.6000803@goop.org>

On Mon, Apr 19, 2010 at 09:19:38AM -0700, Jeremy Fitzhardinge wrote:
> On 04/19/2010 07:26 AM, Glauber Costa wrote:
> >> Is the problem that the tscs are starting out of sync, or that they're
> >> drifting relative to each other over time?  Do the problems become worse
> >> the longer the uptime?  How large are the offsets we're talking about here?
> >>     
> > The offsets usually seem pretty small, under a microsecond. So I don't think
> > it has anything to do with tscs starting out of sync. Specially because the
> > delta-based calculation has the exact purpose of handling that case.
> >   
> 
> So you think they're drifting out of sync from an initially synced
> state?  If so, what would bound the drift?
I think delta calculation introduces errors.

Marcelo can probably confirm it, but he has a nehalem with an appearently
very good tsc source. Even this machine warps.

It stops warping if we only write pvclock data structure once and forget it,
(which only updated tsc_timestamp once), according to him.

Obviously, we can't do that everywhere.

  reply	other threads:[~2010-04-19 18:25 UTC|newest]

Thread overview: 76+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-04-15 18:37 [PATCH 0/5] pv clock misc fixes Glauber Costa
2010-04-15 18:37 ` [PATCH 1/5] Add a global synchronization point for pvclock Glauber Costa
2010-04-15 18:37   ` [PATCH 2/5] change msr numbers for kvmclock Glauber Costa
2010-04-15 18:37     ` [PATCH 3/5] Try using new kvm clock msrs Glauber Costa
2010-04-15 18:37       ` [PATCH 4/5] export new cpuid KVM_CAP Glauber Costa
2010-04-15 18:37         ` [PATCH 5/5] add documentation about kvmclock Glauber Costa
2010-04-15 19:28           ` Randy Dunlap
2010-04-15 20:10             ` Glauber Costa
2010-04-17 18:58         ` [PATCH 4/5] export new cpuid KVM_CAP Avi Kivity
2010-04-19 14:50           ` Glauber Costa
2010-04-20  9:29             ` Avi Kivity
2010-04-17 18:55       ` [PATCH 3/5] Try using new kvm clock msrs Avi Kivity
2010-04-17 18:51     ` [PATCH 2/5] change msr numbers for kvmclock Avi Kivity
2010-04-16 20:23   ` [PATCH 1/5] Add a global synchronization point for pvclock Marcelo Tosatti
2010-04-16 20:36   ` Jeremy Fitzhardinge
2010-04-16 21:05     ` Zachary Amsden
2010-04-19 10:39     ` Peter Zijlstra
2010-04-19 10:50       ` Avi Kivity
2010-04-19 11:05         ` Peter Zijlstra
2010-04-19 11:10           ` Avi Kivity
2010-04-19 14:21             ` Glauber Costa
2010-04-19 14:33               ` Avi Kivity
2010-04-19 14:46                 ` Peter Zijlstra
2010-04-19 16:18                   ` Jeremy Fitzhardinge
2010-04-20  9:31                     ` Avi Kivity
2010-04-20 18:23                       ` Jeremy Fitzhardinge
2010-04-20 18:54                         ` Avi Kivity
2010-04-20 19:42                           ` Jeremy Fitzhardinge
2010-04-21  0:07                             ` Zachary Amsden
2010-04-22 13:11                             ` Glauber Costa
2010-04-23  1:44                               ` Zachary Amsden
2010-04-23  9:34                                 ` Avi Kivity
2010-04-23 19:22                                   ` Jeremy Fitzhardinge
2010-04-23 19:25                                     ` Avi Kivity
2010-04-23 21:31                                   ` Zachary Amsden
2010-04-23 21:35                                     ` Jeremy Fitzhardinge
2010-04-23 21:41                                       ` Zachary Amsden
2010-04-24  9:30                                         ` Avi Kivity
2010-04-24  9:29                                     ` Avi Kivity
2010-04-19 16:11                 ` Jeremy Fitzhardinge
2010-04-19 14:26     ` Glauber Costa
2010-04-19 16:19       ` Jeremy Fitzhardinge
2010-04-19 18:25         ` Glauber Costa [this message]
2010-04-20  1:57           ` Marcelo Tosatti
2010-04-20  9:35             ` Avi Kivity
2010-04-20 12:59               ` Glauber Costa
2010-04-20 15:16                 ` Avi Kivity
2010-04-21  0:01               ` Zachary Amsden
2010-04-21  8:06                 ` Avi Kivity
2010-04-17 18:48   ` Avi Kivity
2010-04-17 18:49     ` Avi Kivity
2010-04-19 10:43       ` Peter Zijlstra
2010-04-19 10:47         ` Avi Kivity
2010-04-19 10:56           ` Peter Zijlstra
2010-04-19 11:13             ` Avi Kivity
2010-04-19 11:19               ` Peter Zijlstra
2010-04-19 11:40                 ` Avi Kivity
2010-04-19 14:32                 ` Glauber Costa
2010-04-19 14:37                   ` Avi Kivity
2010-04-19 10:46     ` Peter Zijlstra
2010-04-19 10:49       ` Avi Kivity
2010-04-19 10:51         ` Peter Zijlstra
2010-04-19 10:54           ` Avi Kivity
2010-04-19 18:35             ` Zachary Amsden
2010-04-20  9:39               ` Avi Kivity
2010-04-21  0:05                 ` Zachary Amsden
2010-04-21  8:08                   ` Avi Kivity
2010-04-19 10:49       ` Peter Zijlstra
2010-04-19 10:53         ` Avi Kivity
2010-04-19 10:59           ` Peter Zijlstra
2010-04-19 11:35             ` Avi Kivity
2010-10-25 23:30   ` Jeremy Fitzhardinge
2010-10-25 23:30     ` Jeremy Fitzhardinge
2010-10-26  8:14     ` Avi Kivity
2010-10-26 10:49       ` Glauber Costa
2010-10-26 17:04       ` Jeremy Fitzhardinge

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=20100419182542.GI14158@mothafucka.localdomain \
    --to=glommer@redhat.com \
    --cc=avi@redhat.com \
    --cc=jeremy@goop.org \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mtosatti@redhat.com \
    --cc=zamsden@redhat.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.