Re: [PATCH 3/4] pm: Handle HID0_SLEEP in the TLF_NAPPING hack.

[Scott Wood <scottwood@freescale.com>]

Paul Mackerras wrote:
> Scott Wood writes:
> 
> 
>>The e300 core (and probably most other 6xx chips) can only come out of
>>sleep mode with an interrupt.  However, interrupts are logically disabled
>>by the power management layer.
> 
> 
> On powerbooks it's typically a hard reset rather than an interrupt.
> Is it possible to use a hard reset on e300-based systems?

The 8313 can do this (see the deep sleep portion of the 83xx PM patch 
that I posted), but others can't.  Even on the 8313, it's not always 
desireable to use deep sleep, due to limited wakeup sourcers, higher 
latency, etc.

> Also, if you use an interrupt, presumably the cpu has to do something
> to clear the interrupt condition.  What would that be?

The caller's MSR[EE] is cleared in power_save_6xx_restore, so the 
interrupt won't happen again until the PM code re-enables MSR[EE].

-Scott