perry@piermont.com <Perry E.Metzger> writes:
>>>> It isn't actually that bad. There are various different levels of ACPI
>>>> suspend, and the lightest sleeps are very much like APM sleeps -- we
>>>> could implement it very trivially, or, even better, we could just take
>>>> the code from FreeBSD that handles all this.
Lennart Augustsson <lennart@augustsson.net> writes:
>>> Excellent!  When will you be done? ;-)
Alicia da Conceicao <alicia@engine.ca> writes:
>> Hi Perry:
>> Any progress on implimenting ACPI sleep/suspend using the FreeBSD code?
perry@piermont.com <Perry E.Metzger> writes:
> Lennart was being sarcastic -- I'm not working on this. Perhaps
> someday I'll get frustrated enough to do it, but for the moment I have
> a laptop that works fine with APM still...

Hi Perry:

Well, I took a look at the FreeBSD apciconf source code, and all
it is, is just a single "ioctl" call to the "/dev/acpi" device node.
To port it to NetBSD will require modifing the existing ACPI driver
and integrating additional code for low power (sleep mode 1) suspend
that stops the CPU clock, not to mention we would have to add an
additional IOCTL request to the existing "/dev/power" device node,
or prehaps add support for a new "/dev/acpi" device node.

Adding IOCTL requests and nodes into a Unix driver is trivial, but
unless one is familar with the current NetBSD ACPI driver, integrating
the additional code for a low power (sleep mode 1) suspend that stops
the CPU clock is much more difficult for us mere-mortals.

So it looks like I will have to wait patiently for someone else to
do this, since my only experience with writing hardware drivers from
scratch has been on the Linux/uClinux platforms.  :-(

Alicia,