Re: RFC: Possiblities for APM implementations.

Peter Horton (pdh@berserk.demon.co.uk)
Thu, 8 Oct 1998 23:17:37 +0100 (BST)


I don't think Linux can do APM. APM relies on SMIs (system management
interrupts), and the code that is executed on SMI is BIOS code, and on
modern MBs this code is protected and locked by hardware.

On some motherboards we might be able to shut the power supply down, and
we can obviously spin hard drives down etc, but that's about it ...

>From the Linux point of view, ACPI will provide a better solution, but
it's still early days for ACPI ...

P.

---
pdh@colonel-panic.com
pdh@berserk.demon.co.uk
http://www.berserk.demon.co.uk

On Thu, 8 Oct 1998, Alex Buell wrote:

> On Thu, 8 Oct 1998, Michael L. Galbraith wrote: > > > The kernel uses bios routines exclusively. An implimentation without > > BIOS hooks is possible ('course) but doesn't exist as we speak. > > So it can be done. I understand that each type of motherboard would have > a different implementation of APM. > > Here's a short proposal for true Linux APM (and SMP safe) handling: > > It perhaps would be nice to build an APM module for Linux that supports > each type of motherboard that is APM-capable with a 'driver' a la the > KGICON part of the GGI project. This module would come in two parts; the > top high level part of the APM module would interface seamlnessly with the > Linux kernel proper by either being loaded or by being compiled in. The > lower part of the APM module would then contain an APM driver specific to > that particular motherboard it is running upon. There could be any number > of APM drivers (ie. say. Tyan S1662S/D APM driver, Supermicro P6DNE/F APM > driver and so on, even Alpha, Sparc and so forth) that could be compiled > into the APM module and linked as a single module or linked into the > kernel itself. It remains to be seen how easily we could do this. > > This then removes the need to depend on the BIOS's APM implementation > which might leave a lot to be desired. Also using our own APM > implementations can ensure that APM management is then SMP safe. We could > even specify which processor to put to sleep... 8) > > Any comments, flames, and knuckledusters welcome. > > Cheers, > Alex. > > --- > /\_/\ Legalise cannabis now! > ( o.o ) Grow some cannabis today! > > ^ < Peace, Love, Unity and Respect to all. > > It BOOTED! Wow. Amazing. > > Check out http://www.tahallah.demon.co.uk > Linux lo-pc3035a 2.1.124 #3 Mon Oct 5 13:30:26 EDT 1998 > One Intel Pentium 75+ processor, 66.36 total bogomips, 16M RAM > libc 5.4.44 > > > - > 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/ >

- 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/