Re: Re: [PATCH 0 of 12] PV on HVM Xen

[Jeremy Fitzhardinge <jeremy@goop.org>]

On 05/18/2010 03:55 AM, Christian Tramnitz wrote:
> Hi Stefano,
>
> what are the particular advantages of running PVonHVM vs. traditional
> PV (vs pure HVM)?
> I'd like to update the wiki with some info about it...

That's an interesting question.  It probably has two dimensions: ease of
deployment and performance.  Some users may just find it easier to
deploy a guest image in pure HVM mode since its most similar to the
native environment (though with pvgrub these differences are fairly
small).  Performance is more complex, as it depends on both your
workload and the capabilities of the processor you're running on.

So, try it and see ;)

    J

>
>
>
> Thanks,
>    Christian
>
> Am 18.05.2010 12:22, schrieb Stefano Stabellini:
>> Hi all,
>> this is the fixed, updated and rebased version of the PV on HVM series:
>> the series is based on 2.6.34 now and supports Xen PV frontends running
>> in a HVM domain, including netfront, blkfront and the VIRQ_TIMER.
>>
>> The list of bugs fixed in this update includes: xenbus drivers crashes
>> when xenbus is not properly initialized, a memory corruption bug in
>> suspend/resume and testing for the xen platform pci version and protocol
>> has been moved to enlighten.c (before unplugging emulated devices).
>>
>> In order to be able to use VIRQ_TIMER and to improve performances you
>> need a patch to Xen to implement the vector callback mechanism
>> for event channel delivery.
>>
>> A git tree is also available here:
>>
>> git://xenbits.xen.org/people/sstabellini/linux-pvhvm.git
>>
>> branch name 2.6.34-pvhvm.
>>
>> Cheers,
>>
>> Stefano
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel
>