Re: [Xen-ia64-devel][PATCH] handle ld.s on guest tr mapped page.

[Alex Williamson <alex.williamson@xxxxxx>]

On Tue, 2006-08-15 at 14:09 +0800, Xu, Anthony wrote:
> diff -r 6f02c4b9a9d0 xen/arch/ia64/vmx/vmx_phy_mode.c
> --- a/xen/arch/ia64/vmx/vmx_phy_mode.c  Wed Aug 09 12:05:13 2006 +0800
> +++ b/xen/arch/ia64/vmx/vmx_phy_mode.c  Tue Aug 15 10:50:54 2006 +0800
> @@ -195,7 +195,8 @@ vmx_load_all_rr(VCPU *vcpu)
>                         (void *)vcpu->domain->shared_info,
>                         (void *)vcpu->arch.privregs,
>                         (void *)vcpu->arch.vhpt.hash, pal_vaddr );
> -       ia64_set_pta(vcpu->arch.arch_vmx.mpta);
> +       ia64_set_pta(VMX(vcpu,mpta));
> +       ia64_set_dcr(VMX(vcpu,mdcr));

Hi Anthony,

   I don't see where non-VTI domains get switched back to a default dcr
value.  Did I miss it?  Seems like this only covers the VTI domains and
we'll end up "leaking" cr.dcr state out to everyone else.  How much will
adding cr.dcr to the domain state affect performance?  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