RE: PROBLEM: Second processor not responding in 2.4.21 and later

[Marc Rieffel]

> > 2.4.18-27.7.xsmp                0       3314    0.0000
> > 2.4.18                  5       4206    0.0012
> > 2.4.19                  12      25786   0.0005
> > 2.4.20-pre4                     2       586     0.0034
> > 2.4.20-pre5                     2       49      0.0392
> > 2.4.20-pre6                     12      745     0.0159
> > 2.4.20                  55      3128    0.0173
> > 2.4.20-20.7smp          483     15427   0.0304
> > 2.4.21-4.0.1.ELsmp      155     7278    0.0209
> 
> While a difference between 2.4.20-pre4 and -pre5 may be a clue,
> it isn't the root cause, because they're both broken.
> Looks like only 2.4.18-27.7.xsmp (whatever that is) got 0 failures.

True, I haven't seen any failures with that kernel (which comes with Rocks 2.3.2, which is based on RH 7.3).  On the other hand, I haven't rebooted it nearly as often as the others, so I can't be confident the error doesn't occur there.

> Any chance you can run a test with, say, 2.6.5?
> It might also be interesting to know what compiler built each kernel..

I did try 2.4.26 and it has the same problem.  

I'm afraid I've already spent as much time (and as much hardware) as I can afford on this problem.  I wrote a script to reboot if processors are missing.  That should suffice for my needs for now.

If anyone else encounters this problem, or someone who knows the smp boot code or the Intel motherboards is interested in helping, please let me know and I'll see what else I can do.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/