Fwd: Tegra 30 System hangs because of commit "pinctrl: move subsystem mutex to pinctrl_dev struct"

[Linus Walleij]

On Mon, Apr 29, 2013 at 9:59 PM, Stephen Warren <swarren at wwwdotorg.org> wrote:

> I would suggest simply dropping or reverting this patch; I had pointed out a
> long time ago that it was unlikely to be correct.
>
> My guess/suspicion is a recursive lock attempt during processing hogs during
> pinctrl device registration, or perhaps during dt->map conversion. Bryan,
> are you able to confirm this?

This was not the case at all, it was an mutex_unlock misspelled as
mutex_lock()... totally unrecursive dual locking :-)

I did test the patch quite extensively, albeit not on a system using
DT to populate the pinctrl tables so I missed this, mea culpa.

I really want to get rid of the big pinctrl lock and make this more
fine-granular, and that is why the patch is being pushed, bear
with me.

Yours,
Linus Walleij