Chuck Swiger <cswiger@mac.com> writes:

> You have a good point, but if the firmware/BIOS doesn't support ACPI
> at least to some extent, how is one going to get the system into a S4
> state?  :-)
>
> The older APM standard supports or supported something resembling S1,
> and of course the OS itself can execute HLT in the idle loop until an
> interrupt wakes it up, but in those cases the state of memory is kept
> active, and not flushed to disk.  On the other hand, if someone can
> figure out how to do hibernate-to-disk without ACPI or similar
> support from the firmware, that would be brilliant....

Well, it's more than just hardware support. the ACPI suspend states
model doesn't work well for complex battery operated devices which
might have dozens of different power-save states, depending on their
current activities; especially if the system has asymetric
multiprocessing in which it might make sense to halt one processor
while not halting the other.

The Linux guys have gone around this issue for years, and have never
come up with an ACPI based powermanagement system that's a good match
for embedded devices. It might be a good thing to look at the results
from their "power management summit" for some details.