Re: [Xen-devel] [PATCH RFC] xen/evtchn: Implement EVTCHNOP_send_imm as a companian to EVTCHNOP_send

[Wei Liu <wei.liu2@xxxxxxxxxx>]

On Thu, Jul 13, 2017 at 12:14:35PM +0200, Juergen Groß wrote:
> On 07/13/2017 09:50 AM, Andrew Cooper wrote:
> > The event channel ABI passes a pointer to a subop-specific structure.  Some 
> > of
> > these structures however are smaller than the pointers passed in the ABI;
> > EVTCHNOP_send specifically passes a 4 byte evtchn port, using a 4 or 8 byte
> > pointer.
> > 
> > For translated guests (x86 HVM and all ARM), passing the port number 
> > directly
> > avoids a guest pagetable walk, which can be very expensive for Xen to 
> > perform.
> > 
> > Signed-off-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
> 
> I'm absolutely in favor of the approach.
> 

+1

> Could we please discuss which hypercalls should offer an immediate
> variant before implementing one after the other? I believe those should
> all be introduced at the same time in order to make deciding which
> variant to use in domU easier: I don't want to have a boolean for each
> possible hypercall, but just one.

+1

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