[Thinkpad] APM vs. ACPI on A31p under Linux

From: Richard Jelinek <rj_at_petamem.com>
Date: Thu Dec 02 2004 - 04:32:59 EST

Hi TPers,

finally I upgraded my Linux distribution to a 2.6.8 kernel-based one
and one of the many many problems I'm experiencing is the ACPI
support.

Has anyone experiences with ACPI on a Thinkpad under Linux? First the
good news:

* Suspend to disk seems possible without a separate partition, Linux
  just saves the contents to free space on the swap partition. Even
  better: Even if you have much more RAM than Swapspace, this method
  works, as long as you don't have too much "hard data" in your
  memory. I.e. the Linux ACPI implementation purges "cached" and
  "buffers" pages from memory, and stores the rest on the swap
  partition. So my 2GB RAM A31p had no problems with suspend-to-disk
  on a 500MB swap partition, if one has not too many/big applications
  running.

* I also managed to configure the ACPI daemon to perform certain
  operations, so the notebook would act as natural as it did with APM:
  closing the display should bring the machine to a "suspend-to-ram"
  status automatically. This also works - almost.

End of good news. Here's the bad ones:

* The "suspend-to-ram" (S3?) ACPI mode the machine goes in when
  closing the display, does stop all fans, HD and so on. Also the
  machine seems to go to hibernation, as the display is switched off
  and the "Moon" LED lights up. However - the backlight of the display
  GOES ON. Generating significant heat and of course draining the
  batteries, so after 2 hours in this mode, the main battery was only
  about 74% charged. Normally it could last for 8 hours in the
  "suspend" mode with APM and drain only 30% of the battery.

* Not sure, whether the "suspend-to-disk" works really as
  should. Status semms to get written to swap, but after the machine
  switches off, theres no way to "wake it up" but power on, which
  means the machine starts booting. During boot it sees a stored image
  in swap and loads that. Ok - so far so good. However the clock is
  set to the time when the "suspend-to-disk" was done

Any hints how to improve this situation (besides going back to APM of
course ;-))?

-- 
best regards,
     Dipl.-Inf. Richard Jelinek
     - The PetaMem Group - Prague/Nuremberg - www.petamem.com -
		       -= 3394928 Mind Units =-
_______________________________________________
Thinkpad mailing list
Thinkpad@stderr.org
http://stderr.org/cgi-bin/mailman/listinfo/thinkpad
Received on Thu Dec 2 13:29:38 2004

This archive was generated by hypermail 2.1.8 : Fri May 26 2006 - 16:03:58 EDT