sched isolcpus=1 related OOPS in 2.6.9

[devik]

Hello,

in Soyo dual CPU PII/350 system I experience early
OOPS (even ksymdump can't save it) during CPU#1
initialization when I use cmdline isolcpus=1 to force
only CPU#0 use (I want to use affinity to select CPU#1).
The OOPS triggers every time when I use isolcpus.

I traced the problem down into sched.c:1928 (find_busiest_group)
where group->cpu_power was zero (thus division by zero occured).
In call trace it goes swapper->schedule()->........->find_busiest_group.
Important registers there: eax=ecx=edx=0, ebx!=0.

Config and vmlinux:
http://luxik.cdi.cz/~devik/files/isolcpus-oops/

Sorry no oops yet (can't get it via ksymoops nor serial),
I can provide further info (screen photo).
Can anyone at least direct me where to look further ?
(I found no general description of group scheduling code
so that I'm lost in it).

thanks a much,

-------------------------------
    Martin Devera aka devik
Linux kernel QoS/HTB maintainer
  http://luxik.cdi.cz/~devik/


-
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/