Re: How can I test the SMP ?

li qi cheng (tank@public.bjnet.edu.cn)
Sat, 21 Nov 1998 23:17:11 +0800 (CST)


Thanks for the people who answer my question so quickly. Just two hours
after I send that mail, I realize I ask a stupid question. I have recompile
the kernel 2.0.36. and add one line "SMP=yes" in the .config
and the two cpu works fine.

in /proc/cpuinfo I find two cpu working. Both bogoMIPS was 333.4.
and I compile file " main(){while(1){}} " to run. It was almost the same
speed with one cpu when I compile the kernel again. It was really grate.

But I find one problem in the kernel 2.0.36. The scsi did not work fine.
The first time I turn all scsi option on. When boot, it lock on searching
the second scsi. I get something like reset after 5 second. and no response
When I turn off some option, next boot can pass on.
It was my firend's computer and it was very late I have no time to check it
very clearly exactly which option cause it can't boot.

Li Qicheng

On Tue, 17 Nov 1998, Zow Terry Brugger wrote:

> > Any body having played with SMP ?
>
> I rely on it.
>
> > We have a two P2-330 cpu SMP with 128M RAM, runing kernel 2.0.25. How can
> > I know that the linux was using the two CPU not one? And how can I test
> > how effection it is?
> >
> The best way I've found is to recompile the kernel (as compiling is CPU and
> I/O intensive), using the time command or some such to see how long it takes.
> First time through, use the default Makefile. Second time through, add -J3 to
> the MAKE= declaration in the makefile. I think this is all documented in
> docs/SMP.txt, or by one of the documents it references.
>

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