Re: [Xen-ia64-devel][PATCH]fix vti inital broken after merge

[Alex Williamson <alex.williamson@xxxxxx>]

On Fri, 2006-12-01 at 11:50 +0800, Zhang, Xing Z wrote:
> Hi Alex:
> 
>          This patch base on your patch which made xen-ia64-unstable
> pass âmakeâ. It can initial Vti and show qemu window. 

Hi Wing,

   Thanks for working on this.  Did you try booting an SMP VTi guest?
I'm wondering if the HVM_INFO code that is #ifdef'd out for ia64 in your
patch will be necessary to enable more vcpus.

>  
>      xc_get_hvm_param(self->xc_handle, dom, HVM_PARAM_STORE_PFN,
> &store_mfn);
> +#if !defined(__ia64__)
>      xc_set_hvm_param(self->xc_handle, dom, HVM_PARAM_PAE_ENABLED,
> pae);
> +#endif
>      xc_set_hvm_param(self->xc_handle, dom, HVM_PARAM_STORE_EVTCHN,
>                       store_evtchn);

   I think it would be cleaner just to make the ia64 specific
xc_set_hvm_param() ignore HVM_PARAM_PAE_ENABLED (or have the xen side of
the hypercall ignore it).

 
> diff -r edaf70bb5bcc xen/common/memory.c
> --- a/xen/common/memory.c       Thu Nov 30 02:58:43 2006 +0800
> +++ b/xen/common/memory.c       Fri Dec 01 02:30:19 2006 +0800
> @@ -137,6 +137,10 @@ static void populate_physmap(struct memo
>          {
>              for ( j = 0; j < (1 << a->extent_order); j++ )
>                  guest_physmap_add_page(d, gpfn + j, mfn + j);
> +
> +            /* Inform the domain of the new page's machine address.
> */ 
> +            if ( unlikely(__copy_to_guest_offset(a->extent_list, i,
> &mfn, 1)) )
> +                goto out;
>          }
>          else
>          { 

   This chunk will need to be submitted to xen-devel.  Thanks,

        Alex

-- 
Alex Williamson                             HP Open Source & Linux Org.


_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel