Hi, I have a Thinkpad laptop (T43) and I'm about to install OpenBSD on it. I have a few questions regarding hibernation though. I've read various documents online so I'm fairly confident with regard to the "how" but out of curiosity I have some questions below regarding the "why", plus a few comments.
- http://www.openbsd.org/i386-laptop.html indicates that there's no ACPI support in OpenBSD, but at the same time I've found a man page for acpid(8) on the OpenBSD website. So what is the current situation? - http://www.openbsd.org/i386-laptop.html doesn't say anything about the size of the hibernation partition. I looked around for that information and found on http://jcs.org/laptops/#x40 that the size of the hibernation partition should be the size of memory + video memory + a few MB, though that web page doesn't give any hint as to were that information came from or if it was simply guessed by trial-and-error. After searching around some more I finally found http://samba.org/ftp/unpacked/junkcode/tphdisk.c which seems to be the source of the information and indicates that the size of the partition should be main memory + video memory + 2 MB. I think it would be nice if that information could be placed somewhere on http://www.openbsd.org/i386-laptop.html because, unless one searches hard for that information in advance, by the time one might see tphdisk's usage message the hibernation partition will presumably already have been created (and therefore will potentially have been created with the wrong size...) - I'm not sure I understand why the video memory has to be saved when hibernating. Once the computer has been restarted I would expect programs to simply redraw themselves, no? Is it because in OpenBSD hibernation is done via the BIOS and the BIOS simply does it that way no matter what? - Which brings me to my last point: has anybody worked / is working / plans on working on handling hibernation and suspend-to-RAM entirely in software rather than through the BIOS? The way swsusp or suspend2 do it for linux? I'd guess that a good chunk of the machinery necessary for something like that is already in the kernel, save maybe for things like copying to disk the non-pageable kernel memory and bringing it back into memory after restart. So are there any plans to go into that direction, by any chance? Thanks, Philippe
