[Pm-utils] [RFC] check for swap partition in pm-is-supported for hibernate / suspend-hybrid

[Dan Nicholson]

On Feb 8, 2008 1:40 PM, Till Maas <opensource at till.name> wrote:
> On Fri February 8 2008, Victor Lowther wrote:
>
> > I would not add it as an intergal part of pm-utils.  tuxonice,
> > for example, can be configured to hibernate/resume without touching a
> > swap partition or a swap file.
>
> Iirc did you create some patches that separte the checks for the different
> sleep methods for the different backends.

Yeah, once Victor's patch to separate the backends, this check should
be part of the kernel check_hibernate sequence, IMO. Then again, I'm
not intimately familiar with the kernel process to say that swap is
definitely mandatory for kernel hibernating.

> > Well, I would not have pm-is-supported check for the presence of an
> > available swap partition -- a usage scenario that would break involves
> > someone who normally runs without swap, and mounts the swap partition just
> > for hibernate
>
> How does this break? As soon as someone activates a swap partition, the output
> of pm-is-supported would change. And in case it is not activated, hibernation
> would fail anyways. Unless he uses a different sleep-backend than the kernel
> of course.

I agree. If you're in the situation where you only mount swap to do
the hibernate, then it would be nice to have pm-utils tell you when
you forgot to do that.

--
Dan