Re: [Xen-devel] [PATCH v5 19/22] hvm/params: Add a new delivery type for event-channel in HVM_PARAM_CALLBACK_IRQ

[Stefano Stabellini <stefano.stabellini@xxxxxxxxxxxxx>]

On Wed, 16 Mar 2016, Konrad Rzeszutek Wilk wrote:
> On Wed, Mar 16, 2016 at 04:34:19PM +0000, Julien Grall wrote:
> > Hi Konrad,
> > 
> > On 04/03/2016 21:19, Konrad Rzeszutek Wilk wrote:
> > >Anyhow what I am wondering if there are some semantincs when it comes to 
> > >PPI
> > >and it being able to 'mask' an vector until it exits or such? If so
> > >you should document that.
> > 
> > I'm not sure to understand what you are asking.
> > 
> > Xen takes advantage of the virtual interrupt controller to notify the guest
> > of new pending event channels.
> > 
> > The life cycle (mask/eoi) of the interrupt associated to the notification is
> > handled by the interrupt controller. The interrupt can't be re-entrant but
> > we may receive spurious notification.
> 
> Excellent. That is what I wanted to know. It would be good to make
> sure that is mentioned in the header file.
> 
> > 
> > AFAIK, there is no specific semantics. Stefano, can you confirm it?

Yes, that's right.


> Because on x86 the vector callback bypasses the interrupt controller and
> injects the vector using the mechanism that is usually used for exceptions
> (or emulating software interrupts).

Right, that's the big difference.

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