Re: [Xen-devel] [PATCH v3 2/4] x86/apic: force phys mode if interrupt remapping is disabled

[Roger Pau Monné <roger.pau@xxxxxxxxxx>]

On Mon, Dec 09, 2019 at 03:30:27PM +0100, Jan Beulich wrote:
> On 09.12.2019 11:25, Roger Pau Monné wrote:
> > On Thu, Dec 05, 2019 at 10:32:34AM +0100, Jan Beulich wrote:
> >> On 04.12.2019 17:20, Roger Pau Monne wrote:
> >>> Cluster mode can only be used with interrupt remapping support, since
> >>> the top 16bits of the APIC ID are filled with the cluster ID, and
> >>> hence on systems where the physical ID is still smaller than 255 the
> >>> cluster ID is not. Force x2APIC to use physical mode if there's no
> >>> interrupt remapping support.
> >>>
> >>> Note that this requires a further patch in order to enable x2APIC
> >>> without interrupt remapping support.
> >>>
> >>> Signed-off-by: Roger Pau Monné <roger.pau@xxxxxxxxxx>
> >>
> >> Reviewed-by: Jan Beulich <jbeulich@xxxxxxxx>
> >> albeit ...
> >>
> >>> --- a/xen/arch/x86/genapic/x2apic.c
> >>> +++ b/xen/arch/x86/genapic/x2apic.c
> >>> @@ -226,7 +226,23 @@ boolean_param("x2apic_phys", x2apic_phys);
> >>>  const struct genapic *__init apic_x2apic_probe(void)
> >>>  {
> >>>      if ( x2apic_phys < 0 )
> >>> -        x2apic_phys = !!(acpi_gbl_FADT.flags & ACPI_FADT_APIC_PHYSICAL);
> >>> +    {
> >>> +        if ( !iommu_intremap )
> >>> +            /*
> >>> +             * Force physical mode if there's no interrupt remapping 
> >>> support:
> >>> +             * the ID in clustered mode requires a 32 bit destination 
> >>> field due
> >>> +             * to the usage of the high 16 bits to store the cluster ID.
> >>> +             */
> >>> +            x2apic_phys = true;
> >>> +        else
> >>> +            x2apic_phys = !!(acpi_gbl_FADT.flags & 
> >>> ACPI_FADT_APIC_PHYSICAL);
> >>
> >> ... I wonder why you didn't make this
> >>
> >>         x2apic_phys = !iommu_intremap || (acpi_gbl_FADT.flags & 
> >> ACPI_FADT_APIC_PHYSICAL);
> >>
> >> (not the least because of allowing to drop the somewhat ugly !!).
> > 
> > Feel free to do it at commit (and reindent the comment), or else I can
> > resend a new version if that's too intrusive.
> 
> Doing these adjustments at commit time ought to be fine. It's
> just that I'd prefer to wait with committing this series until
> 4.13 is fully finished.

That's fine, I don't have any hurry. All patches are Acked or RB now,
so I will hold off sending a new version. Let me know if the patches
don't apply cleanly when committing and I can send an updated version
with the minor comments from this last round.

Thanks, Roger.

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