On 30.09.2020 12:40, Roger Pau Monne wrote:
> --- a/xen/arch/x86/hvm/vlapic.c
> +++ b/xen/arch/x86/hvm/vlapic.c
> @@ -459,13 +459,10 @@ void vlapic_EOI_set(struct vlapic *vlapic)
>
> void vlapic_handle_EOI(struct vlapic *vlapic, u8 vector)
> {
> - struct vcpu *v = vlapic_vcpu(vlapic);
> - struct domain *d = v->domain;
> -
> if ( vlapic_test_vector(vector, &vlapic->regs->data[APIC_TMR]) )
> - vioapic_update_EOI(d, vector);
> + vioapic_update_EOI(vector);
>
> - hvm_dpci_msi_eoi(d, vector);
> + hvm_dpci_msi_eoi(vector);
> }
What about viridian_synic_wrmsr() -> vlapic_EOI_set() ->
vlapic_handle_EOI()? You'd probably have noticed this if you
had tried to (consistently) drop the respective parameters from
the intermediate functions as well.
Question of course is in how far viridian_synic_wrmsr() for
HV_X64_MSR_EOI makes much sense when v != current. Paul, Wei?
A secondary question of course is whether passing around the
pointers isn't really cheaper than the obtaining of 'current'.
Jan
