On Sun, Apr 18, 2021 at 1:54 PM Jonathan Cameron <ji...@kernel.org> wrote: > On Wed, 14 Apr 2021 22:54:51 +0300 > Andy Shevchenko <andriy.shevche...@linux.intel.com> wrote: > > > In case we would initialize two IIO devices from one physical device, > > we shouldn't have a clash on regulators. That's why move > > st_sensors_power_enable() call from core to bus drivers. > > Why is this a problem?
You can't have two regulators of the same name on the same device. IIRC the regulator framework produces a good splat for this. > The two instances would double up and both get + > enable + disable the regulators. However, that shouldn't matter as > they are reference counted anyway. > > Perhaps an example? Even in patch 6 I can only see that it is wasteful > to do it twice, rather than wrong as such. Believe me, I would like to avoid that, but it seems a limitation of the regulator framework. It simply produces a splat. I'll try to reproduce it again (because this series started like a couple of years ago, just eventually I found a time to clean up and submit) and will tell you how it is going. -- With Best Regards, Andy Shevchenko
