Hi, > On Wed, Feb 10, 2016 at 10:36:22AM -0600, Michael Welling wrote: > > On Wed, Feb 10, 2016 at 08:39:04PM +0530, Sricharan wrote: > > > > Hi Sricharan, > > > > > > > > Are you looking at pca9685_pwm_probe in drivers/pwm/pwm- > pca9685.c > > > > right? > > > > > > > Yes. > > > > > > > I'm asking this because this driver doesn't seem to support > > > > runtime pm and there is no check for regmap_write/regmap_write > > > > return code in the probe function. > > > Hmm to be clear, so it’s the pm_runtime_getsync from i2c-qup which > fails right ? > > > I was tracking that when there are i2c_xfers from pwm. I did not see > any failures there. > > > So wanted to know the correct sequence to reproduce. > > > > > > > The problem was discovered using the patch that this thread is on. The > > PWM driver does not have the problem. > > > > When the driver in this patch called pm_runtime_get_sync you got -EINVAL > back. > > I noticed some patches for the QUP I2C driver in linux-next so I built against > it. > > The ADC driver now appears to work as desired. > Ok, I suppose those were changes merged last week to add qup V2 tags support, but not sure how that is helping to solve the pm_runtime issue here.
Regards, Sricharan