Re: Xen spinlock questions

[Jeremy Fitzhardinge <jeremy@goop.org>]

Keir Fraser wrote:
> On 11/8/08 19:49, "Jeremy Fitzhardinge" <jeremy@goop.org> wrote:
>
>   
>> Keir Fraser wrote:
>>     
>>> That's right. The hypercall returns if any event is delivered to the vcpu,
>>> or if any port in the poll set is pending. The fact we don't clear the vcpu
>>> from the poll_mask if unblocked for event delivery isn't a correctness
>>> issue, but it will cause unnecessary extra work in future invocations of
>>> evtchn_set_pending(). Perhaps vcpu_unblock() should clear it.
>>>   
>>>       
>> That seems reasonable.  In this use-case, it's quite likely that if the
>> poll is interrupted by event delivery, on return it will find that the
>> spinlock is now free and never re-enter the poll.
>>     
>
> Attached is a new version of the patch which clears the vcpu from poll_mask
> when it is unblocked for any reason. Jan: please can you give this one a
> spin if you get time.
>   

Forgot to attach?

    J