diff mbox series

[1/4] drm/bridge: ti-sn65dsi86: Don't compile GPIO bits if not CONFIG_OF_GPIO

Message ID 20200608104832.1.Ibe95d8f3daef01e5c57d4c8c398f04d6a839492c@changeid (mailing list archive)
State New, archived
Headers show
Series [1/4] drm/bridge: ti-sn65dsi86: Don't compile GPIO bits if not CONFIG_OF_GPIO | expand

Commit Message

Doug Anderson June 8, 2020, 5:48 p.m. UTC
The kernel test robot noted that if "OF" is defined (which is needed
to select DRM_TI_SN65DSI86 at all) but not OF_GPIO that we'd get
compile failures because some of the members that we access in "struct
gpio_chip" are only defined "#if defined(CONFIG_OF_GPIO)".

All the GPIO bits in the driver are all nicely separated out.  We'll
guard them with the same "#if defined" that the header has and add a
little stub function if OF_GPIO is not defined.

Fixes: 27ed2b3f22ed ("drm/bridge: ti-sn65dsi86: Export bridge GPIOs to Linux")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
---

 drivers/gpu/drm/bridge/ti-sn65dsi86.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

Comments

Stephen Boyd June 11, 2020, 9:56 a.m. UTC | #1
Quoting Douglas Anderson (2020-06-08 10:48:32)
> The kernel test robot noted that if "OF" is defined (which is needed
> to select DRM_TI_SN65DSI86 at all) but not OF_GPIO that we'd get
> compile failures because some of the members that we access in "struct
> gpio_chip" are only defined "#if defined(CONFIG_OF_GPIO)".
> 
> All the GPIO bits in the driver are all nicely separated out.  We'll
> guard them with the same "#if defined" that the header has and add a
> little stub function if OF_GPIO is not defined.
> 
> Fixes: 27ed2b3f22ed ("drm/bridge: ti-sn65dsi86: Export bridge GPIOs to Linux")
> Reported-by: kernel test robot <lkp@intel.com>
> Signed-off-by: Douglas Anderson <dianders@chromium.org>
> ---

Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Linus Walleij June 20, 2020, 8:21 p.m. UTC | #2
On Mon, Jun 8, 2020 at 7:48 PM Douglas Anderson <dianders@chromium.org> wrote:

> The kernel test robot noted that if "OF" is defined (which is needed
> to select DRM_TI_SN65DSI86 at all) but not OF_GPIO that we'd get
> compile failures because some of the members that we access in "struct
> gpio_chip" are only defined "#if defined(CONFIG_OF_GPIO)".
>
> All the GPIO bits in the driver are all nicely separated out.  We'll
> guard them with the same "#if defined" that the header has and add a
> little stub function if OF_GPIO is not defined.
>
> Fixes: 27ed2b3f22ed ("drm/bridge: ti-sn65dsi86: Export bridge GPIOs to Linux")
> Reported-by: kernel test robot <lkp@intel.com>
> Signed-off-by: Douglas Anderson <dianders@chromium.org>

Fair enough,
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>

Yours,
Linus Walleij
diff mbox series

Patch

diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi86.c b/drivers/gpu/drm/bridge/ti-sn65dsi86.c
index 2240e9973178..6fa7e10b31af 100644
--- a/drivers/gpu/drm/bridge/ti-sn65dsi86.c
+++ b/drivers/gpu/drm/bridge/ti-sn65dsi86.c
@@ -151,8 +151,10 @@  struct ti_sn_bridge {
 	u8				ln_assign;
 	u8				ln_polrs;
 
+#if defined(CONFIG_OF_GPIO)
 	struct gpio_chip		gchip;
 	DECLARE_BITMAP(gchip_output, SN_NUM_GPIOS);
+#endif
 };
 
 static const struct regmap_range ti_sn_bridge_volatile_ranges[] = {
@@ -925,6 +927,8 @@  static int ti_sn_bridge_parse_dsi_host(struct ti_sn_bridge *pdata)
 	return 0;
 }
 
+#if defined(CONFIG_OF_GPIO)
+
 static int tn_sn_bridge_of_xlate(struct gpio_chip *chip,
 				 const struct of_phandle_args *gpiospec,
 				 u32 *flags)
@@ -1092,6 +1096,15 @@  static int ti_sn_setup_gpio_controller(struct ti_sn_bridge *pdata)
 	return ret;
 }
 
+#else
+
+static inline int ti_sn_setup_gpio_controller(struct ti_sn_bridge *pdata)
+{
+	return 0;
+}
+
+#endif
+
 static void ti_sn_bridge_parse_lanes(struct ti_sn_bridge *pdata,
 				     struct device_node *np)
 {