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

Bug#524521: marked as done (Fix time warps with VMI clocksource)

Your message dated Sat, 20 Feb 2010 10:34:16 +0100
with message-id <20100220093416.GA17806@galadriel.inutil.org>
and subject line Re: Fix time warps with VMI clocksource
has caused the Debian Bug report #524521,
regarding Fix time warps with VMI clocksource
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
524521: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=524521
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message --- 
Package: linux-2.6
Version: 2.6.26-13

This is a bug in the VMI_timer clocksource implementation and affects
only when running with VMI enabled on VMware platform.
There can be situations when the VMI_CYCLES_COUNTER can go back between
subsequent reads of that counter. This could result in kernels hangs due
to the unsigned calculation that are done in the clocksource code. 
This situation is only seen when the guest is running boot-halt tests
(as the VMI timer could mostly be skewed only during the boot process)
in a highly over-committed situation. 
So though the bug is not very severe it has negligible impact as it
affects code which will be run only when running under VMware and that
too in 32bit configuration.

Please include it for the next update of the kernel for lenny.

Commit details are 

commit 48ffc70b675aa7798a52a2e92e20f6cce9140b3d
Author: Alok N Kataria <akataria@vmware.com>
Date:   Wed Feb 18 12:33:55 2009 -0800

    x86, vmi: TSC going backwards check in vmi clocksource

Thanks,
Alok

--- End Message ---
--- Begin Message --- 
Version: 2.6.29-1

On Fri, Apr 17, 2009 at 11:17:14AM -0700, Alok Kataria wrote:
> Package: linux-2.6
> Version: 2.6.26-13
> 
> This is a bug in the VMI_timer clocksource implementation and affects
> only when running with VMI enabled on VMware platform.
> There can be situations when the VMI_CYCLES_COUNTER can go back between
> subsequent reads of that counter. This could result in kernels hangs due
> to the unsigned calculation that are done in the clocksource code. 
> This situation is only seen when the guest is running boot-halt tests
> (as the VMI timer could mostly be skewed only during the boot process)
> in a highly over-committed situation. 
> So though the bug is not very severe it has negligible impact as it
> affects code which will be run only when running under VMware and that
> too in 32bit configuration.
> 
> Please include it for the next update of the kernel for lenny.
> 
> Commit details are 
> 
> commit 48ffc70b675aa7798a52a2e92e20f6cce9140b3d
> Author: Alok N Kataria <akataria@vmware.com>
> Date:   Wed Feb 18 12:33:55 2009 -0800
> 
>     x86, vmi: TSC going backwards check in vmi clocksource

Sorry for the late followup, this fell through the cracks.

Thanks, it will be included in the next Lenny updated.

Cheers,
        Moritz

--- End Message ---
Reply to: