On Thu, Apr 18, 2024 at 06:56:09PM +0300, Andy Shevchenko wrote: > On Thu, Apr 18, 2024 at 6:06 PM Aren <aren@xxxxxxxxxxxxxxxxx> wrote: > > On Mon, Apr 15, 2024 at 05:04:53PM +0300, Andy Shevchenko wrote: > > > On Sun, Apr 14, 2024 at 8:57 PM Aren Moynihan <aren@xxxxxxxxxxxxxxxxx> wrote: > > ... > > > > > stk3310_set_state(iio_priv(indio_dev), STK3310_STATE_STANDBY); > > > > + if (data->vdd_reg) > > > > + regulator_disable(data->vdd_reg); > > > > > > I forgot to check the order of freeing resources, be sure you have no > > > devm_*() releases happening before this call. > > > > If I understand what you're saying, this should be fine. The driver just > > uses devm to clean up acquired resources after remove is called. Or am I > > missing something and resources could be freed before calling > > stk3310_remove? > > I'm not objecting to that. The point here is that the resources should > be freed in the reversed order. devm-allocated resources are deferred > to be freed after the explicit driver ->remove() callback. At the end > it should not interleave with each other, i.o.w. it should be > probe: devm followed by non-devm > remove: non-devm only. I think what you're describing is already the case, with the exception of parts of the probe function not changed in this patch mixing acquiring resources through devm with configuring the device. I hope I'm not being dense, thanks for the clarification - Aren
