> you can see, I played with some kernel options. Among those were:
> acpi=3Don/off/force/ht, apm=3Don/off, reboot=3D[cbhw],
> pci=3Dbios/nobios/conf1/conf2/nosort/biosirq/noacpi. But whatever
> combination I tried, I never had success with rebooting the=20
> laptop. The
> only way to get a working reboot is with a kernel 2.2.x. Other 2.4.x
> kernels (x=3D18,20,22,24,25) always failed to reboot the machine.
> I hope, you can help me. If you need any further information,=20
> please ask
> me. TIA.
I think you need to set up serial console to monitor the kernel message
rebooting process to sort out valuable clues. I presume that if =
doesn't work, then acpi is exempt from any accusation. :)
> I think you need to set up serial console to monitor the kernel message
> of rebooting process to sort out valuable clues.
Here are the last lines from my kernel log:
kernel: unloading PCMCIA Card Services
kernel: Kernel logging (proc) stopped.
kernel: Kernel log daemon terminating.
The last message on the console is 'rebooting system ...'. That's all
and then the lights went off. As far as I can see the shutdown process
comes to a complete end. On booting the next time there's no fsck'ing
because of unproperly unmounting.
> I presume that if acpi=off doesn't work, then acpi is exempt from
> any accusation. :)
Sorry, but I do not agree. Suspending and power management is one side
of ACPI, but ACPI is also used to setup und configure the system. As far
as I noted, the kernel still uses some ACPI information to setup up the
system, even if you have 'acpi=off' on your command line. In the case of
'acpi=off' you definitely have no suspend possibilities.
As Bernd Schubert suggested, I'll give kernel 2.6.x a try. In regards to
PCI kernel 2.2.x had a very 'simple' approach (compared with 2.4.x).
2.4.x makes a better job. But with my Dell I7500 it broke something
(okay, I admit the I7500 has a broken BIOS, too). My only hope is that
2.6.x corrects what 2.4.x didn't achieve.
Every mail not from this mailing list will be deleted.