Re: Xen 4.3 development update / winxp AMD performance regression

[George Dunlap <george.dunlap@eu.citrix.com>]

On 04/25/2013 02:51 PM, Pasi Kärkkäinen wrote:
> On Wed, Apr 03, 2013 at 11:34:13AM -0400, Andres Lagar-Cavilla wrote:
>> On Apr 3, 2013, at 6:53 AM, George Dunlap <george.dunlap@eu.citrix.com> wrote:
>>
>>> On 03/04/13 08:27, Jan Beulich wrote:
>>>>>>> On 02.04.13 at 18:34, Tim Deegan <tim@xen.org> wrote:
>>>>> At 16:42 +0100 on 02 Apr (1364920927), Jan Beulich wrote:
>>>>>>>>> On 02.04.13 at 16:07, George Dunlap <George.Dunlap@eu.citrix.com> wrote:
>>>>>>> * AMD NPT performance regression after c/s 24770:7f79475d3de7
>>>>>>>    owner: ?
>>>>>>>    Reference: http://marc.info/?l=xen-devel&m=135075376805215
>>>>>> This is supposedly fixed with the RTC changes Tim committed the
>>>>>> other day. Suravee, is that correct?
>>>>> This is a separate problem.  IIRC the AMD XP perf issue is caused by the
>>>>> emulation of LAPIC TPR accesses slowing down with Andres's p2m locking
>>>>> patches.  XP doesn't have 'lazy IRQL' or support for CR8, so it takes a
>>>>> _lot_ of vmexits for IRQL reads and writes.
>>>> Ah, okay, sorry for mixing this up. But how is this a regression
>>>> then?
>>>
>>> My sense, when I looked at this back whenever that there was much more to this.  The XP IRQL updating is a problem, but it's made terribly worse by the changset in question.  It seemed to me like the kind of thing that would be caused by TLB or caches suddenly becoming much less effective.
>>
>> The commit in question does not add p2m mutations, so it doesn't nuke the NPT/EPT TLBs. It introduces a spin lock in the hot path and that is the problem. Later in the 4.2 cycle we changed the common case to use an rwlock. Does the same perf degradation occur with tip of 4.2?
>>
>
> Adding Peter to CC who reported the original winxp performance problem/regression on AMD.
>
> Peter: Can you try Xen 4.2.2 please and report if it has the performance problem or not?

Do you want to compare 4.2.2 to 4.2.1, or 4.3?

The changeset in question was included in the initial release of 4.2, so 
unless you think it's been fixed since, I would expect 4.2 to have this 
regression.

  -George