Difference between revisions of "Problem with LCD backlight remaining on during ACPI sleep"

From ThinkWiki
Jump to: navigation, search
Line 1: Line 1:
 
==Problem description==
 
==Problem description==
On some models the LCD backlight remains on during ACPI sleep. This apparently affects various models (the T30, X31, A31p are mentioned on the relevant bug on the Kernel bug tracker ([http://bugzilla.kernel.org/show_bug.cgi?id=2576 bug 2576]).
+
On some models the LCD backlight remains on during ACPI sleep.
 +
 
 +
==Affected Models==
 +
*ThinkPad {{A31p}}
 +
*ThinkPad {{T30}}
 +
*ThinkPad {{X31}}
 +
(according to the relevant bug on the Kernel bug tracker ([http://bugzilla.kernel.org/show_bug.cgi?id=2576 bug 2576]))
  
 
==Partial solution==
 
==Partial solution==
Line 22: Line 28:
  
 
Improvements are clearly possible. :-)
 
Improvements are clearly possible. :-)
 
 
[[Category:A31p]] [[Category:T30]] [[Category:X31]]
 

Revision as of 11:38, 10 May 2005

Problem description

On some models the LCD backlight remains on during ACPI sleep.

Affected Models

(according to the relevant bug on the Kernel bug tracker (bug 2576))

Partial solution

If the ThinkPad has a radeon video chip, then switching off the light with the command

 radeontool light off

will work, but only in a text-mode virtual terminal. If X is running, the light apparently comes back on when the kernel switches to text mode prior to sleeping. The solution in that case it to switch to a text mode console first.

I use the following script to suspend my machine. The hwclock lines compensate for the clock running at double speed during suspend (this is fixed in the 2.6.11 release candidate kernels, various distributions may backport the fix).

logger "Software suspend to ram."
sync
FGCONSOLE=`fgconsole`
/sbin/hwclock --systohc
chvt 1
radeontool light off
echo -n 3 >/proc/acpi/sleep
/sbin/hwclock --adjust
/sbin/hwclock --hctosys
logger "Woke up from suspend."
radeontool light on
chvt $FGCONSOLE

Improvements are clearly possible. :-)