NOTE: This page is outdated. Please see Joerg Sommrey's page for 2.6.12 and
later versions here.
Amd76x_pm enables the power savings mode for Dual Athlon processors running
Linux. It is currently tested on AMD-760MP and AMD-760MPX boards, for example
Tyan S2460 and S2462 boards.
The module saves about 70 - 90 Watts of energy in the idle mode compared to
the default idle mode.
Currently amd76x_pm does not calculate the CPU load, instead the system
will wake up from the idle mode if the idle function runs twice in a row on
the same processor. This means that amd76x_pm currently only supports SMP
systems. Adding support for single processor systems and sother chipsets
should be relatively easy.
Amd76x_pm is now included in the Linux-2.4 kernel starting from 2.4.19rc3-ac2.
The driver can can be enabled under Character devices -- AMD 76x native
Hopefully this module will eventually be integrated to the ACPI to provide
Pasi Savolainen did a port of the driver to Linux-2.6, and I recently did
some cleanup on Pasi's version of the driver. You can download the current
version of the amd76x_pm for Linux-2.6.0-test9 here. Pasi's older version
of the driver is available here.
UPDATED: The current versions available are:
If you are using 2.4.26, Kay Diederichs has ported Arjan van de Ven's
original 2.4.20 pmtimer patch to 2.4.26. Using the pmtimer should take
care of the clock skew problems with the amd76x_pm. You can download
the patch here.
There are at least two known problems remaining with this driver:
- System goes into sleep mode on S2460 when the driver is compiled in and
loaded before other ACPI initializations. The system wakes up from an
ACPI interrupt at this point, such as pressing the power button once.
Current workaround is to load amd76x_pm as module.
You can also wake up the system automatically when loading the driver
after a reboot. To do this, put the system to sleep mode for few
seconds, and have it wake up to an RTC interrupt. Here's a script
to do it from /etc/init.d on startup: amd-idle.
- System clock skew on certain systems. This seems to be fixed on
2.6.0-test9 for most part, but still appears on some systems. This may
be caused by out of sync TSCs. This problem may also be related to
spurious interrupts on line 9 (ACPI). Dropping HZ to 100 is the current
workaround for this problem.
If you are experiencing the clock skew problem, and want to help debug
the problem, please run this script as root and send the results
to tony at atomide.com.
In case you are looking for the original amd-smp-idle, you can get it here.