[Testday]Xen 4.5 RC4

[Testday]Xen 4.5 RC4

[Hu, Robert]

Hi

This is test report on Xen 4.5 RC4, from Intel OTC VMM Team.

Platform: Grantley-EP, Ivytown-EP

We found these issue blow. Hoping corresponding patches can be committed in Xen 4.5 release.

Issue 1 -- detach a vt-d assigned device from guest, then reattach it to guest, will fail.
http://lists.xen.org/archives/html/xen-devel/2014-11/msg01342.html

Issue 2 -- attach/detach vt-d devices (both physical and virtual fucntions) to domain, more than 1 time, will fail with errors
http://lists.xenproject.org/archives/html/xen-devel/2014-12/msg02016.html
Note: this depends on issue 1. need to apply issue 1's patch to get it.

Issue 3 -- 1GB hugepage support on X86_64 with HVM guest requires hacks in the guest config
http://lists.xenproject.org/archives/html/xen-devel/2014-11/msg02624.html


Best Regards,
Robert Ho

Re: [Testday]Xen 4.5 RC4

[Konrad Rzeszutek Wilk]

On Thu, Dec 25, 2014 at 02:58:06AM +0000, Hu, Robert wrote:
> Hi
> 
> This is test report on Xen 4.5 RC4, from Intel OTC VMM Team.

Thank you!
> 
> Platform: Grantley-EP, Ivytown-EP
> 
> We found these issue blow. Hoping corresponding patches can be committed in Xen 4.5 release.

Are they regressions?
> 
> Issue 1 -- detach a vt-d assigned device from guest, then reattach it to guest, will fail.
> http://lists.xen.org/archives/html/xen-devel/2014-11/msg01342.html
> 
> Issue 2 -- attach/detach vt-d devices (both physical and virtual fucntions) to domain, more than 1 time, will fail with errors
> http://lists.xenproject.org/archives/html/xen-devel/2014-12/msg02016.html
> Note: this depends on issue 1. need to apply issue 1's patch to get it.
> 
> Issue 3 -- 1GB hugepage support on X86_64 with HVM guest requires hacks in the guest config
> http://lists.xenproject.org/archives/html/xen-devel/2014-11/msg02624.html

This one has been in existence for quite a while so it can go in Xen 4.6
> 
> 
> Best Regards,
> Robert Ho
> 

Re: [Testday]Xen 4.5 RC4

[Boris Ostrovsky]

On 12/26/2014 02:02 PM, Konrad Rzeszutek Wilk wrote:
> On Thu, Dec 25, 2014 at 02:58:06AM +0000, Hu, Robert wrote:
>> Hi
>>
>> This is test report on Xen 4.5 RC4, from Intel OTC VMM Team.
> Thank you!
>> Platform: Grantley-EP, Ivytown-EP
>>
>> We found these issue blow. Hoping corresponding patches can be committed in Xen 4.5 release.
> Are they regressions?
>> Issue 1 -- detach a vt-d assigned device from guest, then reattach it to guest, will fail.
>> http://lists.xen.org/archives/html/xen-devel/2014-11/msg01342.html


This is regression introduced by abfb006f.


-boris


>>
>> Issue 2 -- attach/detach vt-d devices (both physical and virtual fucntions) to domain, more than 1 time, will fail with errors
>> http://lists.xenproject.org/archives/html/xen-devel/2014-12/msg02016.html
>> Note: this depends on issue 1. need to apply issue 1's patch to get it.
>>
>> Issue 3 -- 1GB hugepage support on X86_64 with HVM guest requires hacks in the guest config
>> http://lists.xenproject.org/archives/html/xen-devel/2014-11/msg02624.html
> This one has been in existence for quite a while so it can go in Xen 4.6
>>
>> Best Regards,
>> Robert Ho
>>

Re: [Testday]Xen 4.5 RC4

[Jan Beulich]

>>> Boris Ostrovsky <boris.ostrovsky@oracle.com> 12/26/14 9:12 PM >>>
>> On Thu, Dec 25, 2014 at 02:58:06AM +0000, Hu, Robert wrote:
>> Issue 1 -- detach a vt-d assigned device from guest, then reattach it to guest, will fail.
>> http://lists.xen.org/archives/html/xen-devel/2014-11/msg01342.html
>
>This is regression introduced by abfb006f.

And the referenced mail even contains a proposed fix - Robert, did you give this a try?

Jan

Re: [Testday]Xen 4.5 RC4

[Robert Hu]

On Fri, 2015-01-02 at 16:47 +0000, Jan Beulich wrote:
> >>> Boris Ostrovsky <boris.ostrovsky@oracle.com> 12/26/14 9:12 PM >>>
> >> On Thu, Dec 25, 2014 at 02:58:06AM +0000, Hu, Robert wrote:
> >> Issue 1 -- detach a vt-d assigned device from guest, then reattach it to guest, will fail.
> >> http://lists.xen.org/archives/html/xen-devel/2014-11/msg01342.html
> >
> >This is regression introduced by abfb006f.
> 
> And the referenced mail even contains a proposed fix - Robert, did you give this a try?
Yes, we have verified it. It did fixed the issue.
> 
> Jan
>