Mark Brown wrote at Thursday, August 11, 2011 9:15 PM:

On Thu, Aug 11, 2011 at 07:59:27PM -0700, Colin Cross wrote:

quoted

On Thu, Aug 11, 2011 at 5:45 PM, Mark Brown

quoted

quoted

For example with ASoC we'd sort all the components before the ASoC card
without regard for their bus dependencies or any other dependencies they
have (eg, their regulators). Since the ASoC card is a platform device
it's likely to have registered early with no regard for where the buses
the card needs are registered. I'd expect there's a reasonable chance
it'll actually make things worse in the short term.

quoted

You can't just move everything after the card, you have to move
everything after the last device that was probed, and it only works if
nothing depends on any of the devices that are moved.

Sorry, I said that the wrong way round due to trying to reply quickly -
the card would be the thing that moves since that's the thing that
actually does the suspend but we've *no* idea which device we need to
move it after.  Since all the function does is a direct move after or
before a single device all we can do is pick one and pray that it's the
right device.

Colin, 

This thread seems to have died down; how should we make progress?

It sounds like the suspend_irq solution is the current de-facto standard;
not optimal, but all we really have right now and already in use. I could
certainly see avoiding this solution if it was the first time it was
employed, but re-using it seems reasonable to me?

Alternatively, are you attending either Linux Plumbers Conference or the
Kernel Summit? Mark implied this topic might well come up for discussion
there. Unfortunately, I won't be able to make LPC due to a conflict.

(and you'd mentioned having the subsystem maintainers weigh in on this;
which sub-system; IRQ, power, I2C, ...?)

Thanks!

-- 
nvpublic