Re: [Xen-devel] [PATCH RFC for-4.13 03/10] xen/arm: traps: Rework entry/exit from the guest path

[Julien Grall <julien.grall@xxxxxxx>]

Hi Stefano,

On 10/2/19 1:16 AM, Stefano Stabellini wrote:
> On Tue, 1 Oct 2019, Julien Grall wrote:
> > On 01/10/2019 21:12, Stefano Stabellini wrote:
> > > On Thu, 26 Sep 2019, Julien Grall wrote:
> > > I am OK with the general approach but one thing to note is that the fiq
> > > handler doesn't use the guest_vector macro at the moment.
> > do_trap_fiq() is not called from arm64 Instead, we call do_bad_mode().
> > So I don't see an issue here.
> >
> > As do_bad_mode() does not call enter_hypervisor_head(), the fiq handler
> > does not use guest_vector.
> >
> > That said, it is interesting to see that we don't deal the same way the
> > FIQ on Arm32 and Arm64. On the former, we will call do_IRQ while the
> > latter will crash the guest.
> >
> > It would be good if we can have the same behavior accross the two arch
> > if possible. I vaguely recall someone (Andre?) mentioning some changes
> > around FIQ in KVM recently. Andre, are FIQ meant to work with Guest?
> >
> > Also, a side effect of not calling enter_hypervisor_head() is workaround
> > are not re-enabled. We are going to panic soon after, so it may not be
> > that much an issue.
> Right, that is what I was thinking too, but I wanted to highlight it. At
> least it would be worth adding a sentence to the commit message about
> it.
As I pointed out above, this patch does not change anything in this 
particular Arm64 vector so I don't see why I should mention it in the 
commit message.
Cheers,

--
Julien Grall

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel