Re: problems getting SMP to work with vanilla 2.4.26

[Len Brown <len.brown@intel.com>]

On Tue, 2004-07-06 at 12:48, Zack Brown wrote:
> Hi Len,
> 
> On Mon, Jul 05, 2004 at 03:07:44PM -0400, Len Brown wrote:
> > On Sat, 2004-07-03 at 22:05, Zack Brown wrote:
> > > Hi folks,
> > > 
> > > When booting vanilla 2.4.26 with SMP enabled, I get a lockup
> before
> > > the
> > > boot sequence is completed. The same kernel with SMP disabled
> boots
> > > and runs
> > > just fine. Both CPUs are detected by the system at bootup, before
> lilo
> > > takes
> > > over. Here's the error as I wrote it down from the screen,
> followed by
> > > the
> > > .config file:
> > e 
> > > ------------------------------ cut here
> ------------------------------
> > > Using local APIC timer interrupts.
> > > Calibrating APIC timer...
> > > ..... CPU clock speed is 1004.4785 MHZ
> > > ..... hostbus clock speed is 133.9304 MHz
> > > cpu: 0, clocks: 1339304, slice: 446434
> > > CPU0<T0:1339296,T1:892848,D:14,S:446434,C:1339304>
> > > cpu: 1, clocks: 1339304, slice: 446434
> > > CPU1<T0:1339296,T1:446416,D:12,S:446434,C:1339304>
> > > ------------------------------ cut here
> ------------------------------
> > 
> > complete dmesg from success case would help,
> > but try booting with "acpi=off", as that will
> > disable ACPI for CPU enumeration, and if there
> > is a bug in your ACPI tables, it would avoid it.
> 
> I tried acpi=off, and got a hang at boot right after the line:
> 
> CPU0<T0:1339376,T1:892912,D:4,S:446460,C:1339380

sounds like there is no SMP success case on this box with any OS?
Maybe a hardware problem.  See if you
can simplify the hardware by ripping stuff out
until it works.  Alternatively, if you find a
different OS or version of linux that boots SMP,
then that suggests a Linux specific issue.

cheers,
-Len