Re: [PATCH 08/11] OMAP3: PM: Ensure MUSB block can idle when driver not loaded

[Tony Lindgren <tony@atomide.com>]

* Kevin Hilman <khilman@deeprootsystems.com> [090518 07:50]:
> Russell King - ARM Linux <linux@arm.linux.org.uk> writes:
> 
> > On Fri, May 15, 2009 at 11:40:48AM -0700, Kevin Hilman wrote:
> >> Otherwise, bootloaders may leave MUSB in a state which prevents
> >> retention.
> >
> > Hmm, so what happens if a boot loader has touched this MUSB thing, but
> > the kernel which is being run doesn't have CONFIG_USB_MUSB_SOC enabled?
> 
> Then the OMAP will not be able to hit retention.
> 
> That Makefile change happened after this series was initially made.
> I think we should change the Makefile back to always compile usb-musb.c
> and move the #ifdef CONFIG_USB_MUSB_SOC into usb-musb.c
> 
> Tony, any objections?

Sounds good to me, I like patches that make things behave in a predictable
way.

We already have enough of these mysterious problems where "the same kernel
behaves in a different way on my board" when some registers are only
tweaked in the bootloader.

Regards,

Tony