regulator: core: Sleep (not delay) in set_voltage()

These delays can be relatively large (e.g., hundreds of microseconds to
several milliseconds on RK3399 Gru systems). Per
Documentation/timers/timers-howto.rst, that should usually use a
sleeping delay. Let's use the existing regulator delay helper to handle
both large and small delays appropriately. This avoids burning a bunch
of CPU time and hurting scheduling latencies when hitting regulators a
lot (e.g., during cpufreq).

The sleep vs. delay issue choice has been made differently over time --
early versions of RK3399 Gru PWM-regulator support used usleep_range()
in pwm-regulator.c. More of this got moved into the regulator core,
in commits like:

73e705bf81 regulator: core: Add set_voltage_time op

At the same time, the sleep turned into a delay.

It's OK to sleep in _regulator_do_set_voltage(), as we aren't in an
atomic context. (All our callers grab various mutexes already.)

I avoid using fsleep() because it uses a usleep_range() of [N to N*2],
and usleep_range() very commonly biases to the high end of the range. We
don't want to double the expected delay, especially for long delays.

Signed-off-by: Brian Norris <briannorris@chromium.org>
Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
Link: https://lore.kernel.org/r/20220420141511.v2.2.If0fc61a894f537b052ca41572aff098cf8e7e673@changeid
Signed-off-by: Mark Brown <broonie@kernel.org>

drivers/regulator/core.c

@@ -3566,12 +3566,7 @@ static int _regulator_do_set_voltage(struct regulator_dev *rdev,
 	}
 
 	/* Insert any necessary delays */
-	if (delay >= 1000) {
-		mdelay(delay / 1000);
-		udelay(delay % 1000);
-	} else if (delay) {
-		udelay(delay);
-	}
+	_regulator_delay_helper(delay);
 
 	if (best_val >= 0) {
 		unsigned long data = best_val;