On Tuesday, 5 December 2006 06:43, David Brownell wrote:
> [ off $SUBJECT ]
>
> On Monday 04 December 2006 11:44 am, Pavel Machek wrote:
> > > But I think I'll need to add TIF_FROZEN for all architectures, because suspend
> > > to RAM is supposed to work on all of them, isn't it?
> >
> > Well, yes, it should be added, but no, I do not think s2ram works on
> > that many machines.
>
> Userspace "s2ram" != PM_SUSPEND_MEM ("suspend-to-RAM") though.
>
> I'm not even sure there's a globally acceptable definition of
> what PM_SUSPEND_MEM indicates, beyond the fact that one expects
> it saves more power than PM_SUSPEND_STANDBY.
>
> Yes, the Documentation/power/states.txt regurgitates ACPI spec text.
> But lots of non-ACPI systems don't have any reason to do things
> like ACPI says. And in terms of latency, I've seen a lot of that
> be due to the PM framework taking so long to walk the device tree,
> with no substantial difference between "standby" and "str" costs.
Well, in this particular case the question is only which architectures will
actually use the freezer. :-)
Greetings,
Rafael
--
If you don't have the time to read,
you don't have the time or the tools to write.
- Stephen King
