Andy Henroid wrote:
> > The only visible effect is that the laptop won't suspend any more.
> > Something obviously got set because when acpid is killed, the laptop
> > still doesn't suspend.
>
> Right, APM is not being engaged when it is
> overriden by ACPI so I wouldn't expect APM
> suspend to work at all.
Ah, but it does. With older kernels it's immediate because of course
APM is enabled. With newer kernels, when I close the lid the laptop
beeps as it does for a suspend, but carries on beeping for about 5
seconds. I presume it's waiting for the all-clear from the kernel which
it doesn't get. Then it suspends anyway.
The clock is screwed of course because apmd is not there to restore it
later.
What's interesting is that after running acpid, the laptop doesn't do
that any more. Closing the lid makes it beep once and not try to shut
down. That corresponds to the "don't shut down when I close the lid"
mode that is settable via the keyboard. Trying to change back to the
"do shut down" mode using the keyboard doesn't work.
> Further, once acpid
> is started, ACPI is enabled in the BIOS. After
> that any attempt to use APM features is going
> to fail (yes, even after acpid is killed).
Ah, it calls the BIOS. No wonder the default APM facilities are
deactivated.
> I don't know what the actual power savings of
> C-states vs. APM CPU slowing are on your system.
> It might be interesting to test it out and see
> which causes your battery to last longer.
That's not really informative because power consumption varies so much
with what the computer's doing. The estimates from apmd vary by a
factor of two depending on how bright the screen is, whether I dragged a
window recently, whether the disk is spun down, that sort of thing.
(I'm not quite sure why the disk ever spins down, what with atimes and
cron, but mysteriously it does ever so occasionally).
-- Jamie
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Jan 23 2000 - 21:00:23 EST