Re: Solving suspend-level confusion

[Benjamin Herrenschmidt]

> 
> Passing PM_SUSPEND_* values (0-3) instead of PCI power states (0-4),
> as PCI drivers have previously expected, looks like it ought to be a
> simplifying assumption.  Simpler is often good ... though this makes
> me suspect "too simple".  Especially since it pushes policy choices
> into device drivers (normally not good, except for quirks like "this
> driver+hardware can't do D3, so let's use D2")..

Actually, I took a shortcut with my PPC implementation of swsusp,
which was to tweak the numbering of PM_SUSPEND_* so that

 PM_SUSPEND_STANDBY	= 1
 PM_SUSPEND_MEM		= 3
 PM_SUSPEND_DISK	= 4

Which has the "side effect" of matching S states and mostly D states
with the exception of disk, for the few drivers that care...

But in the long run, this may add confusion instead of clearing things,
I agree we should rather move to completely different types, though I
don't feel like re-typing every callbacks in the tree right now...

Ben.


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/