RE: [Xen-ia64-devel] RE: vcpu_translate issue

["Tian, Kevin" <kevin.tian@xxxxxxxxx>]

>From: Matt Chapman [mailto:matthewc@xxxxxxxxxxxxxxx]
>Sent: 2005年11月10日 16:53
>
>On Thu, Nov 10, 2005 at 04:44:19PM +0800, Tian, Kevin wrote:
>>
>> A correction here. I'm not against propagate NaTpage attribute to
>> machine TLB, and only claim that Xen can just respond per attribute what
>> guest specified. No need to care whether page 0 is mapped specially. ;-)
>
>Yes, that's what I meant to say - we can propagate the NaTpage
>attribute to the machine TLB, then we can simply reflect the NaT
>fault.

Yes, when in machine TLB, Nat fault is injected automatically. Or else we still 
need to walk vTLB to check which fault should be injected. If no entry in vTLB, 
normal TLB miss needs to be reflected back to guest. ;-)

>
>I think delivering a TLB miss is not sufficient.  Because the
>mapping for the zero page is indeed in Linux's pagetable, the TLB
>miss handler succeeds in finding and inserting it, and the page
>fault handler is not triggered.
>
>Matt

Actually that depends. Only when short format vhpt entry exists in TLB which 
contains that mapping for zero page, low level TLB miss handler can hit it. Or 
else the flow still goes to C code page fault handler, right? 

Thanks,
kevin

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