Re: SMP
A long time ago, in a galaxy far, far way, someone said...
> On Wed, 16 Feb 2000, Phil Brutsche wrote:
> > It's conceivable that the BIOS would enable it's SMP support when the
> > second CPU card in added. It's also possible that there was BIOS on the
> > CPU card that 'patched' (somehow) the motherboard's BIOS.
>
> I see, so it's worth trying.
> I think the best way is to insert the second CPU board into the slot and
> boot the machine, right...?
Trying it is the best way of finding out. Just keep in mind that pretty
much any 486 is slow, and adding a second processor will mean that it's
still slower than a Pentium.
> BTW, more /proc/interrupts says:
> bdg:/proc# more inter*
> CPU0
> 0: 9016771 XT-PIC timer
> 1: 2548 XT-PIC keyboard
> 2: 0 XT-PIC cascade
> 3: 100039 XT-PIC NE2000
> 11: 19708 XT-PIC aic7xxx
> 13: 0 XT-PIC fpu
> NMI: 0
> ERR: 0
>
> Shouldn't those XT-PICs be IO-APICs..?
Depends on the kernel revision - kernels before 2.2 don't support APICs.
This is what my dual Pentium says (running 2.3.45):
CPU0 CPU1
0: 15208328 15105019 IO-APIC-edge timer
1: 445 396 IO-APIC-edge keyboard
2: 0 0 XT-PIC cascade
3: 3 8 IO-APIC-edge serial
4: 3 8 IO-APIC-edge serial
8: 2 4 IO-APIC-edge rtc
13: 1 0 XT-PIC fpu
14: 100268 95607 IO-APIC-edge ide0
15: 0 2 IO-APIC-edge ide1
17: 5969889 5966778 IO-APIC-level eth0
19: 516 464 IO-APIC-level aic7xxx
NMI: 30313283 30313283
LOC: 30305847 30305846
ERR: 0
(Oops. I guess Pentiums support IO-APIC as well)
IO-APICs would only show in /proc/interrupts if the hardware supported
them. 486s predate IO-APICs, and thus they won't be used.
--
----------------------------------------------------------------------
Phil Brutsche pbrutsch@creighton.edu
"There are two things that are infinite; Human stupidity and the
universe. And I'm not sure about the universe." - Albert Einstein
Reply to:
- Follow-Ups:
- Re: SMP
- From: Oki DZ <litbang@haltek.telkom.net.id>
- References:
- Re: SMP
- From: Oki DZ <litbang@haltek.telkom.net.id>