Re: Status of SR-IOV for Xen?

[Keir Fraser <keir.fraser@eu.citrix.com>]

Hopefully I will have MSI support fixed and check in by tomorrow. The issue
was not with Intel's locking changes to pci/msi, but with Jan Beulich's
changes to ACKTYPE_NONE/ACKTYPE_EOI handling of MSIs. I'm just doing some
testing of my fix now.

 -- Keir

On 20/01/2009 15:20, "Zhao, Yu" <yu.zhao@intel.com> wrote:

> I'll respin those patches and resubmit them later. One thing is that
> SR-IOV depends on MSI/MSI-x but they are disabled as Keir said in thread
> 'x86: Disable MSI as it seems to be triggering ASSERT at irq.c:269'.
> 
> Wei Huang wrote:
>> It is better not to duplicate the efforts if Yu Zhao is working on it.
>> Otherwise, I would be interested on it too.
>> 
>> -Wei
>> 
>> Simon Horman wrote:
>>> On Mon, Jan 19, 2009 at 11:13:37AM -0600, Wei Huang wrote:
>>>> Hi,
>>>> 
>>>> Any further update on SR-IOV support for Xen? Are we going to include
>>>> this feature soon?
>>> 
>>> I am also interested in this and would be willing to do some work
>>> on breaking out the patches to address Jan Beulich's concerns[1]
>>> about accreditation. And to bring the patches up to date with the most
>>> recent Linux patches. The main problem that I see with the latter
>>> that 2.6.18.8's PCI stack is now quite old, so using that as
>>> a target would be more work than for instance using the paravirt ops
>>> work that Jeremy Fitzhardinge has been working on.
>>> 
>>> [1] http://lists.xensource.com/archives/html/xen-devel/2008-09/msg00965.html
>>> 
>>> --
>>> Simon Horman
>>>   VA Linux Systems Japan K.K., Sydney, Australia Satellite Office
>>>   H: www.vergenet.net/~horms/             W: www.valinux.co.jp/en
>>> 
>>> 
>> 
>