All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jeremy Fitzhardinge <jeremy@goop.org>
To: "Xu, Dongxiao" <dongxiao.xu@intel.com>
Cc: Dan Magenheimer <dan.magenheimer@oracle.com>,
	"xen-devel@lists.xensource.com" <xen-devel@lists.xensource.com>,
	"kurt.hackel@oracle.com" <kurt.hackel@oracle.com>,
	"Dugger, Donald D" <donald.d.dugger@intel.com>,
	Keir Fraser <keir.fraser@eu.citrix.com>,
	"Nakajima, Jun" <jun.nakajima@intel.com>,
	"Zhang, Xiantao" <xiantao.zhang@intel.com>
Subject: Re: Live migration fails due to c/s 20627
Date: Tue, 15 Dec 2009 09:12:00 -0800	[thread overview]
Message-ID: <4B27C360.7080906@goop.org> (raw)
In-Reply-To: <6CADD16F56BC954D8E28F3836FA7ED7105AC868872@shzsmsx501.ccr.corp.intel.com>

On 12/15/09 08:10, Xu, Dongxiao wrote:
>> Why not just add the code to do rdtscp emulation,
>> which would NOT break live migration?
>>      
> Add rdtscp emulation has such problem that, in Intel VMX, the
> vmexit control for rdtsc and rdtscp is the same, so if we trap
> rdtscp for emulation, OS will suffer from looooots of rdtsc vmexit,
> which will bring performance downgrade.
>    

I don't see why that's relevant.  In the case where you've migrated the 
domain, if the CPU has rdtsc but not rdtscp, won't the rdtscp vmexit 
with an illegal instruction trap?  In that case you can emulate rdtscp 
while still having direct execution of rdtsc.

Of course, having a wide difference between rdtscp and rdtsc performance 
may cause its own set of problems.

     J

  parent reply	other threads:[~2009-12-15 17:12 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-12-14 18:02 Live migration fails due to c/s 20627 Dan Magenheimer
2009-12-14 23:19 ` Keir Fraser
2009-12-15  4:40   ` Xu, Dongxiao
2009-12-15 15:56     ` Dan Magenheimer
2009-12-15 16:10       ` Xu, Dongxiao
2009-12-15 16:31         ` Dan Magenheimer
2009-12-15 17:08         ` Dan Magenheimer
2009-12-15 18:04           ` Xu, Dongxiao
2009-12-16  2:07           ` Nakajima, Jun
2009-12-16  2:43           ` Zhang, Xiantao
2009-12-16  4:14             ` Dan Magenheimer
2009-12-16  5:07               ` Zhang, Xiantao
2009-12-16 16:23                 ` Dan Magenheimer
2009-12-16 17:38                   ` Jeremy Fitzhardinge
2009-12-16  6:19               ` Jeremy Fitzhardinge
2009-12-16 15:20                 ` Dan Magenheimer
2009-12-16 17:31                   ` Jeremy Fitzhardinge
2009-12-15 17:12         ` Jeremy Fitzhardinge [this message]
2009-12-15 17:24           ` Xu, Dongxiao
2009-12-15 18:25             ` Jeremy Fitzhardinge
2009-12-15 19:20               ` Dan Magenheimer
2009-12-15 19:31                 ` Keir Fraser
2009-12-15 19:52                   ` Keir Fraser
2009-12-16 15:41                 ` Andre Przywara
2009-12-16 16:54                   ` Dan Magenheimer
2009-12-15 19:25               ` Keir Fraser
2009-12-15 18:05     ` Xu, Dongxiao

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=4B27C360.7080906@goop.org \
    --to=jeremy@goop.org \
    --cc=dan.magenheimer@oracle.com \
    --cc=donald.d.dugger@intel.com \
    --cc=dongxiao.xu@intel.com \
    --cc=jun.nakajima@intel.com \
    --cc=keir.fraser@eu.citrix.com \
    --cc=kurt.hackel@oracle.com \
    --cc=xen-devel@lists.xensource.com \
    --cc=xiantao.zhang@intel.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.