Lonni J Friedman wrote:
On 6/27/06, Gene Heskett <gene.heskett@xxxxxxxxxxx> wrote:
I believe one item is fixed, I was getting nuisance messages in the
messages file from the builtin lightscribe drive, and a tail -f on it
doesn't show them so far since it did an auto powerdown after the bios
flash. Duh, too soon, just got one:
diablo kernel: hdc: cdrom_pc_intr: The drive appears confused (ireason =
0x01)
and I see this stanza in the messages file is still there. Someone said
it looked like a bios problem...
Jun 27 21:18:08 diablo kernel: Kernel command line: ro root=LABEL=/
irqpoll noapic noapci pci=assign-busses lapic
Jun 27 21:18:08 diablo smartd[1861]: smartd has fork()ed into background
mode. New PID=1861.
Jun 27 21:18:08 diablo kernel: Misrouted IRQ fixup and polling support
enabled
Jun 27 21:18:08 diablo kernel: This may significantly impact system
performance
But I'm wondering if it might be related to that long string of kernel
arguments it took to get it to boot FC5 originally? Does this tell us
one or more of those "irqpoll noapic noapci pci=assign-busses lapic"
should be removed?
I dont' think noacpi is a valid kernel parameter, so that one did
nothing. noapic can certainly cause all kinds of problems on some
systems, and almost always results in a performance hit. The other
options may or may not have caused problems. Why did you use all of
those options initially?
It seemed to be the only way to make it boot when I first installed FC5.
I'll take those 2 out of the previous kernel for effect.
Half an hour later:
Ok, played around for a while and took those 2 plus the irqpoll out of
the argument list. It still boots, and those messages are now gone.
I'll leave a tail running on messages and see if that stopped the
mewling from the lightscribe. It works ok otherwise.
This touchpad may not be the whole problem, the system seems to want to
take off and do strange things after typing a 't' in particular. I
wonder what the two key rollover delays are in this POC?
--
Cheers, Gene
