Messages in this thread Patch in this message | | | Date | Mon, 8 Dec 2003 23:03:08 -0600 | From | Ryan Underwood <> | Subject | Re: APIC support on Slot-A Athlon, K6 [PATCH] |
| |
On Mon, Dec 08, 2003 at 06:07:11PM +0100, Mikael Pettersson wrote: > On Mon, 8 Dec 2003 05:58:59 -0600, Ryan Underwood wrote: > >> Furthermore, I/O-APIC usage requires (in hardware) that the > >> processor has a local APIC. > > > >What can the APIC support alone accomplish, without an I/O-APIC? > >Just NMI watchdog and related things? (looking at CONFIG_APIC help) > >Looks like I/O-APIC is the real desired feature, but a functioning local > >APIC, though not very useful by itself, is a prerequisite for it. > > Local APIC gives you: > - using the local APIC timer instead of the mobo's legacy timer > - thermal management interrupts on P4s (dunno about K7s/K8s) > - performance counter interrupts, which can be used as an NMI > source (watchdog, profiling) or for performance analysis > > I/O-APIC gives you: > - more interrupt vectors ==> less IRQ sharing ==> improved > stability and performance > - lower-overhead interrupt management > - I/O-APIC NMI watchdog (another timer with NMI delivery mode)
I attached a Kconfig patch more explicitly stating the prerequisites for using the I/O-APIC.
-- Ryan Underwood, <nemesis@icequake.net> --- arch/i386/Kconfig.bak 2003-08-08 23:32:43.000000000 -0500 +++ arch/i386/Kconfig 2003-12-08 23:00:36.000000000 -0600 @@ -491,6 +491,10 @@ If you have a system with several CPUs, you do not need to say Y here: the IO-APIC will be used automatically. + A functioning local APIC is a prerequisite for using the I/O-APIC + on uniprocessor machines. Sometimes working ACPI support is also + required (depending on the BIOS implementation). + config X86_LOCAL_APIC bool depends on !SMP && X86_UP_APIC[unhandled content-type:application/pgp-signature] | |