PM / runtime: Use pm_runtime_put_sync() in __device_release_driver()
authorRafael J. Wysocki <rafael.j.wysocki@intel.com>
Thu, 7 Nov 2013 00:51:15 +0000 (01:51 +0100)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 4 Dec 2013 18:56:59 +0000 (10:56 -0800)
commit2aed351e3f2014138a9551e0cb481128d3f3bd91
tree0cad7c4666b3960b4ecf9767f609341a8f241ea8
parent67b51f87ef941607e1e891bf477bd18df53ce6f1
PM / runtime: Use pm_runtime_put_sync() in __device_release_driver()

commit baab52ded242c35a2290e1fa82e0cc147d0d8c1a upstream.

Commit fa180eb448fa (PM / Runtime: Idle devices asynchronously after
probe|release) modified __device_release_driver() to call
pm_runtime_put(dev) instead of pm_runtime_put_sync(dev) before
detaching the driver from the device.  However, that was a mistake,
because pm_runtime_put(dev) causes rpm_idle() to be queued up and
the driver may be gone already when that function is executed.
That breaks the assumptions the drivers have the right to make
about the core's behavior on the basis of the existing documentation
and actually causes problems to happen, so revert that part of
commit fa180eb448fa and restore the previous behavior of
__device_release_driver().

Reported-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Fixes: fa180eb448fa (PM / Runtime: Idle devices asynchronously after probe|release)
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: Kevin Hilman <khilman@linaro.org>
Acked-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/base/dd.c