Re: [ACPI] swsusp with 2.5.33

[Pavel M. <pa...@su...>]

Hi!

> lately I observed problems with swsusp. Only if I am suspending from
> console right after the boot there are enough pages free....I did some
> research on the net and found that the rmap things does not free enough
> pages. 

Then either rmap is seriously broken or we are not calling it the
right way.

> AFAIU when suspending to disk the kernel (respective the vm) is asked to
> swap as many pages out as possible and creates an image from the remaining
> pages and writes that to disk. While that seems to be at first a great
> idea it depends - by design - upon the ability to free more than half of
> the ram. Why can one not stop all processes, lock the kernel, and write
> all pages to an image in swap space? 

It is far easier to fix VM than do that.
								Pavel 
-- 
Worst form of spam? Adding advertisment signatures ala sourceforge.net.
What goes next? Inserting advertisment *into* email?