Suspend to RAM

[Ed Hill]

On Wed, 2004-12-15 at 19:40 -0500, Charles E Taylor IV wrote:
> On Wed, 15 Dec 2004 15:35:49 -0600 (CST)
> Satish Balay <balay at fastmail.fm> wrote:
> 
> > On Wed, 15 Dec 2004, Charles E Taylor IV wrote:
> 
> > > - APM suspend (to RAM) won't work unless you stop the PCMCIA service
> > > before suspending,
> 
> > If I have pcmcia cards inserted - I have to do an 'cardctl eject' -
> > before suspending. This has been true for ever (since rhl7 days)
> 
> This is without a card in the slot.  I've not even tried to suspend with a
> card in the slot, since I've *never* had that work on a Thinkpad. :)
> 
> Funny thing is - it's not necessary to stop the PCMCIA drivers for APM
> hibernation.


With recent (eg. 2.6.9-1.681_FC3) kernels, my ThinkPad A22p is able to
suspend-to-RAM with both ACPI and APM.  Using the *stock* FC2 and FC3
kernels, the differences I've noticed are:

 ACPI:
  - much faster suspending (just 1--2 seconds)
  - uses battery at 2--4X the rate of APM suspend-to-RAM
  - gives odd kernel error on wakeup (something about
    interruptable sleep) but seems to work none the less
  - fails to power down the system completely at the end 
    of a shutdown (need to pull power cord and remove
    the battery--annoying!)

 APM:
  - slower to suspend-to-RAM (by a few seconds)
  - uses much (!) less battery while suspended to RAM
  - crashes about once every 20--40 suspend-resume cycles
  - powers down the system cleanly at the end of a shutdown

And I've had some kernel versions where you *had* to rmmod the pcmcia
bits to get suspend-to-RAM to work for either ACPI or APM.  Recent
kernels now seem to be OK with pcmcia and suspend-to-RAM (no need to
rmmod) when using either ACPI or APM.

Ed

-- 
Edward H. Hill III, PhD
office:  MIT Dept. of EAPS;  Rm 54-1424;  77 Massachusetts Ave.
             Cambridge, MA 02139-4307
emails:  eh3 at mit.edu                ed at eh3.com
URLs:    http://web.mit.edu/eh3/    http://eh3.com/
phone:   617-253-0098
fax:     617-253-4464