[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH] docs: add vpmu description to xen-command-line.markdown


  • To: xen-devel@xxxxxxxxxxxxx
  • From: Dietmar Hahn <dietmar.hahn@xxxxxxxxxxxxxx>
  • Date: Fri, 27 Apr 2012 12:00:07 +0200
  • Cc: Ian Campbell <Ian.Campbell@xxxxxxxxxx>
  • Delivery-date: Fri, 27 Apr 2012 10:00:47 +0000
  • Domainkey-signature: s=s1536a; d=ts.fujitsu.com; c=nofws; q=dns; h=X-SBRSScore:X-IronPort-AV:Received:X-IronPort-AV: Received:Received:From:To:Cc:Subject:Date:Message-ID: User-Agent:In-Reply-To:References:MIME-Version: Content-Transfer-Encoding:Content-Type; b=fXkTIaMd3utQ2YSGCVAWorck5dvHsxMJ8fmyxP/I2/WTVQHDHZpn0QV9 O7S+q2+i69v+dhEalUaGXJvHPi9f11tKj3i7DiwL6frhTIiBdQwu66ERM JRcZai8lFvuY0stTJZzZOeiIcDu+jw99C59rU/45RlRWAHxbMPLA4JanC HZ7yPYAq2R5LF8yYZZRljzFzQ5lKjmpKG3wVR0W0XHposFBZYRRgd6E8/ pGZFBgTTG+YA7d6TqvyOeP+aPAg6j;
  • List-id: Xen developer discussion <xen-devel.lists.xen.org>

Am Freitag 27 April 2012, 10:29:55 schrieb Ian Campbell:
> On Fri, 2012-04-27 at 10:20 +0100, Dietmar Hahn wrote:
> > # HG changeset patch
> > # User Dietmar Hahn <dietmar.hahn@xxxxxxxxxxxxxx>
> > # Date 1335517766 -7200
> > # Node ID 9ba4e4a5f3b1b7dad77e2801afc0253d6d30d565
> > # Parent  107285938c50f82667bd4d014820b439a077c22c
> > 
> > Add a short description to the vpmu boot option in the
> > xen-command-line.markdown
> > 
> > 
> > Signed-off-by:  Dietmar Hahn <dietmar.hahn@xxxxxxxxxxxxxx>
> 
> Acked-by: Ian Campbell <ian.campbell@xxxxxxxxxx>
> 
> Can't we just blacklist systems with the nehalem quirk and otherwise
> enable it?

I'm not sure how to create a blacklist. The problem is, the processor models
named in the check_pmc_quirk() function were found be doing some performance
counter tests with machines accessible be me and looking what happens.
As some other models may have the same PMU the hang may occur on other machines
too.
Maybe we can change the policy to switch always on the vpmu stuff for the
AMD cpus and enforce a flag for Intel?

Dietmar.

> 
> Ian.
> 
> > 
> > 
> > diff -r 107285938c50 docs/misc/xen-command-line.markdown
> > --- a/docs/misc/xen-command-line.markdown   Thu Apr 26 10:03:08 2012 +0100
> > +++ b/docs/misc/xen-command-line.markdown   Fri Apr 27 11:09:26 2012 +0200
> > @@ -543,7 +543,28 @@ console even after dom0 has been started
> >  relinquish control to dom0.
> >  
> >  ### vpid
> > +
> >  ### vpmu
> > +> `= ( bts )`
> > +
> > +> Default: `off`
> > +
> > +Switch on the virtualized performance monitoring unit for HVM guests.
> > +
> > +If the current cpu isn't supported a message like  
> > +'VPMU: Initialization failed. ...'  
> > +is printed on the hypervisor serial log.
> > +
> > +For some Intel Nehalem processors a quirk handling exist for an unknown
> > +wrong behaviour (see handle_pmc_quirk()).
> > +
> > +If 'vpmu=bts' is specified the virtualisation of the Branch Trace Store 
> > (BTS)
> > +feature is switched on on Intel processors supporting this feature.
> > +
> > +*Warning:*
> > +As the BTS virtualisation is not 100% safe and because of the nehalem quirk
> > +don't use the vpmu flag on production systems with Intel cpus!
> > +
> >  ### vti\_vhpt\_size
> >  ### vti\_vtlb\_size
> >  ### watchdog

-- 
Company details: http://ts.fujitsu.com/imprint.html

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.