Marc Kleine-Budde <mkl@xxxxxxxxxxxxxx> writes: > On 09/13/2012 04:14 PM, Kevin Hilman wrote: >> AnilKumar Ch <anilkumar@xxxxxx> writes: >> >>> Move pm_runtime_enable/disable calls to c_can.c driver. Current >>> implementation is such that platform driver is doing pm_runtime >>> enable/disable and core driver is doing put_sync/get_sync. >>> >>> PM runtime calls should be invoked if there is a valid device >>> pointer from platform driver so moving enable/disable calls >>> to core driver. >>> >>> Signed-off-by: AnilKumar Ch <anilkumar@xxxxxx> >>> --- >>> Incorporated Kevin's comments on "can: c_can: Add runtime PM >>> support to Bosch C_CAN/D_CAN controller" patch. >> >> This looks better, but in addition, you can get rid of the >> runtime PM helper functions you added (the ones that check for >> priv->device) and call the pm_runtime_get/put APIs directly. > > But priv->device might be NULL. AFAICS pm_runtime_get() is not safe to > be called with a NULL pointer. Yes, you're right. Guess there's not a clean way to get rid of those helpers. Sorry for the noise, Kevin -- To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
