On 20.11.2018 3:26, Douglas Anderson wrote: > In the commit f8702f9e4aa7 ("regulator: core: Use ww_mutex for > regulators locking") disabling of the supply was moved into > _regulator_disable(). That means regulator_disable_work() shouldn't > be disabling since that double-disables the supply. > > Fixes: f8702f9e4aa7 ("regulator: core: Use ww_mutex for regulators locking") > Signed-off-by: Douglas Anderson <dianders@xxxxxxxxxxxx> > --- > > drivers/regulator/core.c | 10 ---------- > 1 file changed, 10 deletions(-) > > diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c > index 0052bbc8c531..63a8af1e2256 100644 > --- a/drivers/regulator/core.c > +++ b/drivers/regulator/core.c > @@ -2703,16 +2703,6 @@ static void regulator_disable_work(struct work_struct *work) > regulator_balance_voltage(rdev, PM_SUSPEND_ON); > > regulator_unlock_dependent(rdev, &ww_ctx); > - > - if (rdev->supply) { > - for (i = 0; i < count; i++) { > - ret = regulator_disable(rdev->supply); > - if (ret != 0) { > - rdev_err(rdev, > - "Supply disable failed: %d\n", ret); > - } > - } > - } > } > > /** > Good catch! Reviewed-by: Dmitry Osipenko <digetx@xxxxxxxxx>