RE: [Xen-ia64-devel] Some things to be considered for ptc.ga emulation

["Xu, Anthony" <anthony.xu@xxxxxxxxx>]

>From: Tian, Kevin
>4. How long does ptc.ga emulation need to wait?
>       By above IPI style, the best case is that all target vcpus are
>not in
>running state and thus all emulation work can be done within IPI handler
>
>       The worst case is that target vcpu is in running state. In that
>case,
>extra time is added between IPI handler and the point resuming back to
>guest. But that extra time is normally short and tolerable.
>
>       Then the whole emulation can always ensure all other vcpus
>observe
>the effect before resuming back to guest.


Considering VTIdomain, I have below concerns,
If a vcpu receives a IPI, just before do_block( waiting device model
on dom0 to handle IO request), do_block will schedule out this vcpu,
and the schedule in vcpu may use the old tlb mapping.
And the wait maybe long.


Moreover, if this IO request is sent to the vcpu of domain0, which runs
On the same vcpu that send out IPI, because this vcpu is waiting other
Vcpus' response of IPI handling, it can't be scheduled to vcpu of domain0,
Then dead loop happens.

The vcpu that sends IPI should not wait by tight polling, this vcpu should yield
LP and poll.

Thank,
Anthony


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

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