RE: [RFC][PATCH] i386 x86-64 Eliminate Local APIC timer interrupt

[Maciej W. Rozycki]

On Thu, 5 May 2005, Len Brown wrote:

> Re: SMP using i8259
> While Linux in ACPI mode allows "noapic" on SMP, it isn't recommended.
> It is there for comparisons, debugging, and to work-around the odd
> broken system.  It is an exception configuration, and supporting it
> should in no way impact the design for other 99.99% normal systems.
> 
> Indeed, note that SMP systems using i8259 instead of IOAPIC
> is explicity forbidden by MPS, and thus would probably fail
> the compatibility test for your favorite high volume binary OS.

 I'm not quite sure if that's forbidden by MPS -- the mixed mode certainly 
is not as not all interrupt sources may necessarily be routed to one of 
I/O APICs, and "noapic" can probably be treated as a special case of the 
mixed mode.  Regardless, there used to be systems in existence that 
wouldn't route IRQ 0 to an I/O APIC, so using that interrupt requires 
either the mixed mode or using the "through-i8259A" trick (which 
unfortunately does not work for a subset of affected systems as a result 
of manufacturers implementing the Intel-recommended glue logic at the 
output of the master i8259).

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