Re: [Xen-devel] collecting info on platforms with VT-d BIOS problems

[Andrew Oakley <andrew@xxxxxxxxxxxxxxxxx>]

Sander Eikelenboom wrote:
> Hello Andrew,
> 
> Seems this problem is on a lot of Asus motherboards, i have a similar
> (not the same) motherboard (P5Q-EM DO), and have had the same
> problem. But Weidongs recent workarounds in xen-unstable don't
> disable vt-d anymore but ignore the RMRR, so if you wish to give Xen
> and vt-d a try again, I think it will work for you as well.

This does seem to "work" now on my board - at least Xen starts without
disabling VT-d (I haven't managed to actually use a passed through
device).

> 
> -- Sander
> 
> 
> Monday, January 25, 2010, 4:25:31 PM, you wrote:
> 
>> Kay, Allen M wrote:
>>> We would like to collect information on platforms that have
>>> problem with VT-d PCI device passthrough using non-IGD PCI
>>> devices.  Note that IGD (integrated display device) device
>>> passthrough code is still being worked on.
>>> 
>>> If you have encountered  VT-d problems on a particular platform
>>> using latest xen-unstable, please provide the following
>>> information so that we can work with OEM's to get their BIOS's
>>> fixed:
> 
>> I've had issues with the ASUS P5E-VM DO on the latest (0902) BIOS.
>> I had given up trying to get this fixed, so I don't have a Xen
>> install on there at the moment.
> 
>> I investigated the problem in May last year and discovered that one
>> of the RMRR entries had a base address that was greater than its
>> limit address.  This didn't seem to happen when using integrated
>> graphics.
> 
>> Let me know if this is sufficient information, or should I be
>> getting Xen booting again so you can see the log, or perhaps a dump
>> of the DMAR table is sufficient?
> 
>> Thanks
> 
> 
> 
> 


-- 
Andrew Oakley

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel