Suspend2 isn't invasive.

Nigel Cunningham nigel at
Thu Nov 30 08:47:34 GMT 2006

(Sorry, Scott - resending because I replied with wrong From:, so the
email got moderated).


On Wed, 2006-11-29 at 22:33 -0800, Scott Robinson wrote:
> Might I interject here and point at a document I have started preparing:
> If we want to improve the suspend process, there are far better places
> to start than adding suspend2 patches. Nigel, am I mistaken in asserting
> that the kernel patches do not fundamentally change the sleep/hibernate
> itself.. but rather add a richer capabilities like encryption and
> compression?

That's right. They provide more options for how and where to store the
image, but use the same basic functions to (eg) suspend drivers and do
the atomic copy.

> Adding some sanity to our current infrastructure would be a start.
> Harmonizing policy and reducing differences from upstream.
> Comprehensive analysis of the acpi-support vs. suspend's hibernate
> script actions would also be a great effort. It could potentially
> improve both scripts. (I have started on this and would be happy to
> share some of my work.)

I agree. I'd really like to see a better mechanism in place than either
the hibernate script or the existing hal/dbus solution, one that lets
you configure any of the solutions simply and easily, ideally even from
some sort of X/console based 'wizard'.

> (Me, I dream of a sleep -> hibernate automatic timeout.)

That's a kernel thing. I would implement it if the driver support was
there, but at the moment I have no way of saying 'wake up in 20 minutes'
or of telling when the wake up occurs whether it was because the timer
ran out or because the user manually woke us at (say) 19 minutes and 55

> What I'm saying is, there is a lot of work that could be done on the
> current infrastructure. And if interested people can make some progress
> there, then I'm sure arguments for including suspend2 patches would
> likely be considered a bit more...

I think they're really separate issues, but since I don't have the time
at the moment to make progress in either area, I'll just listen and try
to contribute what I can to help the discussion progress.



More information about the ubuntu-devel mailing list