Re: [Xen-devel] [PATCH v2] x86/hvm: Add per-vcpu evtchn upcalls

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 06.11.14 at 16:33, <paul.durrant@xxxxxxxxxx> wrote:
> HVM guests have always been confined to using the domain callback
> via (see HVM_PARAM_CALLBACK_IRQ) to receive event notifications
> which is an IOAPIC vector and is only used if the event channel is
> bound to vcpu 0.

Iirc the callback-via-vector method was specifically added to have a
way to spread the IRQ handling load. And even if this didn't work out
as intended, wouldn't simply setting a flag to avoid the restriction in


> This patch adds a new HVM op allowing a guest to specify a local
> APIC vector to use as an upcall notification for a specific vcpu.
> This therefore allows a guest which sets a vector for a vcpu
> other than 0 to then bind event channels to that vcpu.

So is there really a need for a per-vCPU vector value (rather than
a single domain wide one)?

> @@ -220,6 +227,8 @@ void hvm_assert_evtchn_irq(struct vcpu *v)
>  
>      if ( is_hvm_pv_evtchn_vcpu(v) )
>          vcpu_kick(v);
> +    else if ( v->arch.hvm_vcpu.evtchn_upcall_vector != 0 )
> +        hvm_set_upcall_irq(v);

The context code above your insertion is clearly not enforcing
vCPU 0 only; the code below this change is.

Jan


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