Hello David,
On Fri, 19 Feb 1999, David Mentr\\'e wrote:
> Hi Sanna,
>
> Sanna Lee <[EMAIL PROTECTED]> writes:
>
> I may be wrong, but as far as I know :
>
> > 1) Do we absolutely /need/ the IO-APIC for SMP functionality?
>
> I though no as far as BIOS detect second processor. But it appears I'm
> wrong. Ingo Molnar should have a clear advice on this (he has developed
> most of [all?] IO-APIC support).
Ingo is guiding us to find a solution. :-)
>
> BTW, is BIOS MP1.x option enabled ?
Yes.
>
> > 2) Would there be a performance penalty in running these motherboards
without
> > an IO-APIC, if at all possible?
>
> yes. IO-APIC allows to dispatch interrupts on both CPU. Without IO-APIC,
> only cpu 0 takes interrupts ==> I/O slow down.
Hmmm, that's very interesting. Do you think this slowdown justifies adding the
IO-APIC? It increases the price of the motherboard by 30%, unfortunately. :-(
>
> > 3) What would we need to patch/program to get these motherboards to work
> > without an IO-APIC
>
> Try an old SMP kernel without IO-APIC (I would say a 2.0.x, maybe
> 2.0.36).
Oh, we wanted so much to use kernel 2.2.1. :-(
BTW we tried the noapic option but it gives exactly the same message:
the kernel thinks we are using a uniprocessor motherboard and gives up setting
the SMP part... :-(
Is there any way to make the kernel setup the MPTABLE for us if the BIOS
doesn't do it?
And boot the second processor?
Last minute news: THX to Ingo Molnar, we now have the mainboard booting in SMP
mode!!!! (we applied a patch sent to us by Ingo, and booted with the "noapic"
option).
We just have to sort a small problem with the IDE DMA driver now...
Merci a tous!
--
Sanna Lee
A Nice Hong Kong Girl
____________________________________________________________________
Get free e-mail and a permanent address at http://www.netaddress.com/?N=1
-
Linux SMP list: FIRST see FAQ at http://www.irisa.fr/prive/mentre/smp-faq/
To Unsubscribe: send "unsubscribe linux-smp" to [EMAIL PROTECTED]
