Re: [Xen-devel] RFC: PVH set vcpu info context in vmcs....

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 15.08.13 at 03:58, Mukesh Rathor <mukesh.rathor@xxxxxxxxxx> wrote:
>>     __vmwrite(GUEST_FS_BASE, ctxtp->fs_base);
>>     __vmwrite(GUEST_GS_BASE, ctxtp->gs_base_kernel);
>> 
>>     if ( (rc = vmx_add_guest_msr(MSR_SHADOW_GS_BASE)) )
>>         goto out;
>>     vmx_write_guest_msr(MSR_SHADOW_GS_BASE, ctxtp->gs_base_user);
>> 
> 
> Thinking about this more, I realized we are unnecessarily creating
> a vmcs intercept for gs_base_user. How about we not allow that either?

Why would requiring this base to be zero avoid the need to create
an intercept?

Also, remember that the user/kernel distinction here is sort of a
mis-naming by the Xen public interface - the CPU architecture
calls them GS base and shadow GS base, so other usage models
by kernels are possible (albeit admittedly unlikely).

Jan


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel