RE: [PATCH][2.4] generic support for systems with more than 8 CPUs (1/2)

Pallipadi, Venkatesh (venkatesh.pallipadi@intel.com)
Sun, 22 Dec 2002 17:24:15 -0800


> From: Martin J. Bligh [mailto:mbligh@aracnet.com]
> > 1/2 : checking for xAPIC support in the system
>
> OK, that looks pretty sane - one question:
>
> > - if ((clustered_apic_mode != CLUSTERED_APIC_XAPIC) &&
> > + if (!xapic_support &&
> > + (clustered_apic_mode != CLUSTERED_APIC_XAPIC) &&
>
> When does xapic_support differ from
> (clustered_apic_mode == CLUSTERED_APIC_XAPIC) ?
>

They are quite different.
Infact CLUSTERED_APIC_XAPIC just means using physical APIC mode and is kind of a
misnomer as xAPIC doesn't necessariy mean physical APIC mode.
xapic_support says whether xAPIC support is there or not. Then APICs
can be configured either in physical or logical modes. I mainly need this as
with xAPIC support, we have:
- LAPIC and IOAPIC have there own name space,
- max or 255 CPUS with 0xff as broadcast, as opposed to 0xf broadcast in case of no xAPIC

> Do you want to use a physical flat xapic mode for your stuff, or the
> same clustered physical mode as the Summit stuff? If the latter, then
> the new switch seems unnecessary ....

Now I am getting a bit confused here. I am using physical mode with no clustering
whatsoever. Thats what I felt even Summit was doing in 2.4.

Thanks,
-Venkatesh
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/