[ltp] radeonfb sleep patch for kernel 2.6.15

[Henrique de Moraes Holschuh]

On Wed, 21 Jun 2006, Guarded Identity wrote:
> If it is possible to drain a battery too far, would a ACPI S3 sleep quit in 
> time to not cause any damage?  Maybe this is a candidate question for the 
> acpi4linux mailing list.  

Depends on the hardware and ACPI BIOS, and yes, many laptops wake from sleep
as soon as battery hits the critical threshold.

And AFAIK, some thinkpads CAN be configured to do so in the BIOS (wake from
S3 if batteries are critical, or overheating).

The correct thing to do for maximum hardware safety (remember the machine
was sleeping, it might be under heavy phisical or thermal stress when it
wakes up) is to teach the kernel power management to know if it is doing S3
sleep, or S3-backed-by-disk sleep.  Then, upon wakeup, immediately switch to
S5 (power off) *before* powering up buses and devices, if it notices there
is a critical condition.

Examples of critical conditions that warrant S3 partial wakeup -> immediate
S5:

1. All of the ACPI batteries are critical

2. Any of the ACPI thermal zones are critical.

You will have to ask if this is possible on the power-management kernel
list.  Maybe it just isn't doable.

-- 
  "One disk to rule them all, One disk to find them. One disk to bring
  them all and in the darkness grind them. In the Land of Redmond
  where the shadows lie." -- The Silicon Valley Tarot
  Henrique Holschuh