Re: [PATCH] implement pm_ops.valid for everybody

[David Brownell <david-b@pacbell.net>]

On Friday 23 March 2007 6:43 am, Rafael J. Wysocki wrote:
> On Friday, 23 March 2007 00:55, David Brownell wrote:

> > You said that if the hardware doesn't support a "turn CPU off" mode, then
> > you'd define that as being incapable of implementing suspend-to-RAM.
> 
> That's _if_ the suspend-to-RAM is defined as the state in which the CPU
> is off, which I _think_ would be a reasonable definition.  

I disagree.


> I don't mean the 
> platforms incapable of doing this should be restricted from entering any
> system-wide low-power states, but perhaps we can call these states
> differently.

Well, we have ** ONLY TWO LABELS TO APPLY ** and you're saying that
one of them should be restricted to systems where the CPU can go into
an "off" state.  


> > That's a restriction ... a very arbitrary one.
> > 
> > 
> > ...
>
> My point is that _if_ we use lables like "standby", "STR", "STD", etc.,

That is, the strings in /sys/power/state.  That's a given for now...


> then they shouldn't mean different things on different platforms. 

Unreasonable.  The platforms are different.  And moreover the
specifics DO NOT MATTER to userspace.  Plus, they can differ
even on two x86 systems:  different D-states, different wakeup
events.  So nobody has any valid expectation that STR on one
box has exactly the same behavior on a different box.  And if
users are trained to expect anything, it's that platforms will
differ in those details.


> So, either 
> we don't use labels at all, or we should know what they mean regardless
> of what platform we're talking about.

That's a false choice, when you "mean" anything more than
fairly broad behavioral expectations:  STR saves more power
than "standby", and transitions to/from STR take more
time than to/from "standby".

- Dave