Red Hat Bugzilla – Bug 80730
SMP-kernel Oops, UP works fine
Last modified: 2013-07-02 22:09:06 EDT
Description of problem:
The SMP-version of the kernel oopses every time i try to start it. The UP-kernel
works fine. Various 7.1 and 7.2 smp-kernels (including errata) have worked fine
on the same hardware.
This is the message on the screen.
EIP: 0060:[<00000068>] Not tainted
EIP is at Using_Versions 0x67 (2.4.20-2.2smp)
eax: 00000000 ebx: c2522000 ecx: 00000002 edx: c2522000 esi: c01070c0
edi: 00000000 ebp: 00000000 esp: c2523fb8
ds: 0068 es: 0068 ss: 0068
Process swapper (pid: 0, Stackpage = c2523000)
Stack: c0107164 00000002 00000000 00000cb9 00000cb9 c01226c5 00000cb9
00000282 ffffffff 00000000 00000001 00000282 00000001 c040de4a 00000246
0000002a 00000000 c0122932
Call Trace [<c0107164>] cpu_idle [kernel] 0x24 (0xc2523fb8)
[<c01226c5>] call_console_drivers [kernel] 0x65 (0xc2523fcc)
[<c0122932>] printk [kernel] 0x142 (0xc2523ffc)
Code: Bad EIP Value
<0> Kernel panic: Attempted to kill the idle task!
In idle task - not syncing
Version-Release number of selected component (if applicable): 2.4.20-2.2smp
How reproducible: Always
Steps to Reproduce:
1. Press Power on, wait.
Motherboard: Tyan 230
CPU: 2xPIII 933 MHz
Memory: 1.5 GB
Video card: Matrox G400 16MB
does this get fixed if you use "acpi=off" ?
Yes, the SMP-kernel now boots up fine. For a beta-release we're now OK, this
issue can be closed.
Please attach the output of 'dmidecode' from the kernel-utils package on your
Created attachment 89028 [details]
Output from dmidecode as requested
I ran successfully a test with Memtest86 3.0 on the hardware. I should also
mention that this test-machine has also successfully run Solaris 8 and FreeBSD
The reason that I mention this is that I get both SMP and UP kernel oops (with
acpi=off, obviously) when I have httpd started during boot-up, which was a big
surprise. If I start httpd from a Gnome-terminal, the kernel doesn't Oops, but
httpd goes into infinite 99.9% CPU consumption.
Unfortunately it is difficult to get good Oops messages for this. Sometimes the
screen just goes black, sometimes the messages are reported very fast in an
infinite loop. Should this be reported as a separate error?
*** This bug has been marked as a duplicate of 82123 ***
Changed to 'CLOSED' state since 'RESOLVED' has been deprecated.