Subject: Re: [acpi-jp 661] Re: Experimental port of FreeBSD ACPI stuffs
To: None <,>
From: Mitsuru IWASAKI <>
List: tech-kern
Date: 09/26/2000 10:58:16

From: Warner Losh <>
Subject: [acpi-jp 661] Re: Experimental port of FreeBSD ACPI stuffs 
Date: Mon, 25 Sep 2000 11:43:03 -0600
Message-ID: <>

> In message <> "T.SHIOZAKI" writes:
> : I hacked FreeBSD ACPI stuffs to work under NetBSD.
> ...
> : Comments?
> Given the large scope of ACPI, I think it would be desirable to share
> this code between the two BSDs as much as is possible.

Yes, it's one of our (my at least) goals.  I believe that most of the
code can be shared among BSDs except for the part of device-tree and
ACPI namespace integration.  Of course there are some differences on
kernel API between them, but it's no so serious matter, we can have
compat. layer just like USB stuff is doing.  I'm glad to help everything
I can do if our code is marged into BSDs :-)

From: Bill Sommerfeld <>
Subject: [acpi-jp 662] Re: Experimental port of FreeBSD ACPI stuffs 
Date: Mon, 25 Sep 2000 14:15:13 -0400
Message-ID: <>

> As you may be aware I'm working on multiprocessor support for
> NetBSD/i386.


> Given that APM doesn't deal with multiprocessors, and the intel MP
> spec tables are often highly inaccurate, I'm very interested in using
> ACPI for interrupt wiring and power management.

Yes, replacing APM by ACPI was one of the my motivations.  APM BIOS on
modern PC seems bogus even on commonplace laptops which supported by
WinXX already.

> I'm very glad that someone else wrote the AML interpreter. :-)

Thanks, we're very happy to hear that :-) I hope that a lot of people
improve the code, we know it's still experimental quality.

From: Frank van der Linden <>
Subject: [acpi-jp 663] Re: Experimental port of FreeBSD ACPI stuffs
Date: Mon, 25 Sep 2000 20:36:26 +0200
Message-ID: <>

> That's cool, thanks for doing this.
> When I looked at the FreeBSD code recently, it seemed that the power
> management code was mostly there, but that other ACPI features
> weren't actively used yet (although hooks are present). For example,
> io/memory accounting. Is that correct?

As Warner-san mentioned, many compoments are not yet implemented, many
developers are needed in this area.  I think BSDs co-devepment ACPI
stuff and avoiding duplicate effort would be good idea.  If you are
interested in the development, please visit and see `Who is woking on what'
section, then contact (in English OK).

We're looking forward to working together with you :-)