| To: | Kanoj Sarcar <kanoj@xxxxxxxxxxxxxxxxxxx> |
|---|---|
| Subject: | Re: Kernel hang ... |
| From: | Ralf Baechle <ralf@xxxxxxxxxxx> |
| Date: | Wed, 10 May 2000 20:12:36 +0200 |
| Cc: | linux-origin@xxxxxxxxxxx |
| In-reply-to: | <200005101751.KAA05291@google.engr.sgi.com>; from kanoj@google.engr.sgi.com on Wed, May 10, 2000 at 10:51:41AM -0700 |
| References: | <20000510142407.A7547@uni-koblenz.de> <200005101751.KAA05291@google.engr.sgi.com> |
| Sender: | owner-linux-origin@xxxxxxxxxxx |
On Wed, May 10, 2000 at 10:51:41AM -0700, Kanoj Sarcar wrote:
> Depends on how you "disable" the other processors. Prom level disabling
> has not been tested at all, and I would suggest leaving that for a little
> later (maybe next week), since it also ties in to processor virtual/physical
> numbering.
>
> If you want to come up on a CONFIG_SMP kernel with just 1 processor,
> hack do_cpumask(), specifically, #if 0 out the code block
>
> if (cpuid > *highest)
> *highest = cpuid;
> /* Only let it join in if it's marked enabled */
> if (acpu->cpu_info.flags & KLINFO_ENABLE) {
> CPUMASK_SETB(*boot_cpumask, cpuid);
> cpus_found++;
> }
The machine actually freezes is the line
if (maxcpus > 1) while(atomic_read(&numstarted) == 0);
In my case there is only a single enabled CPU, so the numstarted will
stay 0 and the kernel freezes.
With both CPUs enabled I get somewhat further, the kernel then freezes
after probing the caches. Will take a look at this next.
Ralf
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: Kernel hang ..., Kanoj Sarcar |
|---|---|
| Next by Date: | Re: Kernel hang ..., Srinivasa Prasad Thirumalachar |
| Previous by Thread: | Re: Kernel hang ..., Kanoj Sarcar |
| Next by Thread: | Re: Kernel hang ..., Kanoj Sarcar |
| Indexes: | [Date] [Thread] [Top] [All Lists] |