Re: Pending patches mostly pushed, please check

[Todd Poynor <tpoynor@mvista.com>]

Tony Lindgren wrote:

>> 9) LDM wakeup flags for OMAP keypad
>> http://linux.omap.com/pipermail/linux-omap-open-source/2005-December/005897.html
> 
> Let's think about this one a bit more. Maybe we should have u32 device_wakeup in
> pm.c, and then various drivers would mask it with OMAP_WAKEUP_KEYPAD etc?
> 
> Or maybe device_init_wakeup() should register a callback function in the driver,
> and them pm.c just calls all registered wakeup callback functions? That would
> move the enable/disable code to drivers.

Agreed, I should have labeled this patch as at an RFC stage.  It's the 
first attempt I'm aware of to apply the wakeup flags stuff begun for 
PCI/USB to an embedded SoC, and was intended to start a discussion on 
the above sorts of issues, which should also occur on 
linux-pm@lists.osdl.org (and other embedded-oriented wakeup patches are 
being floated there as well).  It would be nice to move the wakeup 
enable code to drivers, but with the per-core keypad IRQ assignments, 
the need to enable level-2 IRQs, etc., not sure if a platform core file 
ends up being more suitable, don't have a strong opinion myself so far.

On a related note, if anybody knows how to tell what source woke up an 
OMAP from deep/big sleep I'd appreciate hearing about it.  I haven't 
found a register devoted to it, and keypad wakeup on my H3 seemed to 
show a pending Level-2 interrupt in the level-1 ITR, but the level-2 
ITRs all read zero (reading prior to enabling interrupts upon deep sleep 
resume).  Thanks,


-- 
Todd