Grr!

[Matt Graham]

"Jeremy Paul Bowers" <bowersj2@pilot.msu.edu> wrote:
> Ben Pfaff, you said. . .
> >As I understand it, this is true for BIOS-assisted
> >suspend-to-disk.  But I don't think that BIOS-assisted
> >suspend-to-disk is supported by Linux.  Fortunately, it doesn't
> >need to be: there is a kernel "hibernation" patch that does this
> >without any assistance from hardware or BIOS, and this patch
> >works on any hardware.

Correct about the patch.  However, some laptops (lots of ThinkPads) have a
hardware "suspend-to-disk" that you can trigger with a key sequence.

> I run Linux on a P133 laptop, and I really wanted the suspend-to-disk. I
> eventually set it up in the BIOS so that after 20 minutes of being
suspended,
> it saves everything to the disk (to a DOS parition at the start of the
drive)
> and shuts off. When next I turn on the computer, Linux comes back off the
disk
> and life is mostly good.

As I understand it, the suspend-to-disk on many laptops requires that you make
a "hibernation" file that has a special name, is on a FAT partition, and is
composed of contiguous sectors.  The starting and ending sectors are written
to somewhere in the NVRAM, generally by the DOS utility that creates the
hibernation file.  Then, either a key sequence or a super-secret series of
BIOS calls does the actual suspend work.

FWIW, a Thinkpad 600X with a full battery loses roughly 30% of its battery
charge for each 24 hours of suspend-to-RAM operation.  Yeah, it's not that
great, but perfectly fine for my usage patterns....

-- 
Matt G / Dances With Crows
There is no Darkness in Eternity/But only Light too dim for us to see
"I backed up my brain to tape, but tar says the tape contains no data...."

____________________________________________________________________
Get free email and a permanent address at http://www.netaddress.com/?N=1