AW: Re: [Xen-devel] Xen BUG in mm / Xen 4.0.1 with 2.6.32.18/21 pvops Kernel?

["Jan Beulich" <JBeulich@xxxxxxxxxx>]

>>> On 09.09.10 at 20:27, "Carsten Schiers" <carsten@xxxxxxxxxx> wrote:
> and provided the output, just in case it is what you needed. I will for 
> sure need some time to understand that.
> Is the hypothesis that a) the BIOS is wrong, and b) Xen is checking more 
> extreme than bare metal, so we need to
> make the Xen Code a bit more relaxed (error message instead of crash)?

I think the underlying problem was pointed out before: The page table
accessor functions can't currently propagate errors, and hence
ioremap() can't be forced to fail if Xen refused a mapping. And then, as
I wrote in an earlier reply, the check for not trying to map RAM from
ioremap() is pretty pointless (and hence no help) currently for the Xen
case.

Jan


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