On Tuesday 10 September 2013, Linus Walleij wrote:

The platform using this beeper has support for gpiolib, so there
is no point to use the custom gpio_line* API. A strange ambiguity
where a line was first set as input and then driven high was
solved by first driving the line high as output and then switch
it to input.

Cc: Imre Kaloz <kaloz@openwrt.org>
Cc: Krzysztof Halasa <khc@pm.waw.pl>
Cc: Alexandre Courbot <acourbot@nvidia.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Linus Walleij <redacted>
---
Arnd/Greg: seeking your ACK to take this through the GPIO tree
as part of the attempt at cleaning out custom GPIO implementations.

Acked-by: Arnd Bergmann <arnd@arndb.de>

 
-	 if (count) {
-		gpio_line_config(pin, IXP4XX_GPIO_OUT);
-		gpio_line_set(pin, IXP4XX_GPIO_LOW);
-
+	if (count) {
+		gpio_direction_output(pin, 0);
 		*IXP4XX_OSRT2 = (count & ~IXP4XX_OST_RELOAD_MASK) | IXP4XX_OST_ENABLE;
 	} else {

Too bad we can't just clean up all the open-coded mmio accesses as well. It shouldn't
be hard, but some platforms are full of them, and there is definitely some regression
potential.

	Arnd