Skip to contentSkip to conversation

drm/bridge: ti-sn65dsi86: rename GPIO register bits

PATCH In review Shawn Guo 3 participants 4 messages Dec 10, 2020 – Jan 12, 2021

Subsystems (4)
lore
Share & tools
lore

Shawn Guo

author
#0 lore raw

PATCH 1/1 drm/bridge: ti-sn65dsi86: rename GPIO register bits

From: Shawn Guo <[email protected]>

It renames GPIO register bits to drop 'SN_' prefix, so that they are consistent to other definitions - prefixing register name with 'SN_' but not for bit fields.

---
 drivers/gpu/drm/bridge/ti-sn65dsi86.c | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)
drivers/gpu/drm/bridge/ti-sn65dsi86.c +13 -13
index f27306c51e4d..8eac9d77eba1 100644
@@ -63,13 +63,13 @@
 #define SN_HPD_DISABLE_REG			0x5C
 #define  HPD_DISABLE				BIT(0)
 #define SN_GPIO_IO_REG				0x5E
-#define  SN_GPIO_INPUT_SHIFT			4
-#define  SN_GPIO_OUTPUT_SHIFT			0
+#define  GPIO_INPUT_SHIFT			4
+#define  GPIO_OUTPUT_SHIFT			0
 #define SN_GPIO_CTRL_REG			0x5F
-#define  SN_GPIO_MUX_INPUT			0
-#define  SN_GPIO_MUX_OUTPUT			1
-#define  SN_GPIO_MUX_SPECIAL			2
-#define  SN_GPIO_MUX_MASK			0x3
+#define  GPIO_MUX_INPUT				0
+#define  GPIO_MUX_OUTPUT			1
+#define  GPIO_MUX_SPECIAL			2
+#define  GPIO_MUX_MASK				0x3
 #define SN_AUX_WDATA_REG(x)			(0x64 + (x))
 #define SN_AUX_ADDR_19_16_REG			0x74
 #define SN_AUX_ADDR_15_8_REG			0x75
@@ -1035,7 +1035,7 @@ static int ti_sn_bridge_gpio_get(struct gpio_chip *chip, unsigned int offset)
 	if (ret)
 		return ret;
 
-	return !!(val & BIT(SN_GPIO_INPUT_SHIFT + offset));
+	return !!(val & BIT(GPIO_INPUT_SHIFT + offset));
 }
 
 static void ti_sn_bridge_gpio_set(struct gpio_chip *chip, unsigned int offset,
@@ -1051,8 +1051,8 @@ static void ti_sn_bridge_gpio_set(struct gpio_chip *chip, unsigned int offset,
 
 	val &= 1;
 	ret = regmap_update_bits(pdata->regmap, SN_GPIO_IO_REG,
-				 BIT(SN_GPIO_OUTPUT_SHIFT + offset),
-				 val << (SN_GPIO_OUTPUT_SHIFT + offset));
+				 BIT(GPIO_OUTPUT_SHIFT + offset),
+				 val << (GPIO_OUTPUT_SHIFT + offset));
 	if (ret)
 		dev_warn(pdata->dev,
 			 "Failed to set bridge GPIO %u: %d\n", offset, ret);
@@ -1069,8 +1069,8 @@ static int ti_sn_bridge_gpio_direction_input(struct gpio_chip *chip,
 		return 0;
 
 	ret = regmap_update_bits(pdata->regmap, SN_GPIO_CTRL_REG,
-				 SN_GPIO_MUX_MASK << shift,
-				 SN_GPIO_MUX_INPUT << shift);
+				 GPIO_MUX_MASK << shift,
+				 GPIO_MUX_INPUT << shift);
 	if (ret) {
 		set_bit(offset, pdata->gchip_output);
 		return ret;
@@ -1103,8 +1103,8 @@ static int ti_sn_bridge_gpio_direction_output(struct gpio_chip *chip,
 
 	/* Set direction */
 	ret = regmap_update_bits(pdata->regmap, SN_GPIO_CTRL_REG,
-				 SN_GPIO_MUX_MASK << shift,
-				 SN_GPIO_MUX_OUTPUT << shift);
+				 GPIO_MUX_MASK << shift,
+				 GPIO_MUX_OUTPUT << shift);
 	if (ret) {
 		clear_bit(offset, pdata->gchip_output);
 		pm_runtime_put(pdata->dev);
Signature · 6 lines
2.17.1

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
reviewer Reviewed-by on v1 1/1
#1 lore raw

Hi,

Quoted 11 lines · Shawn Guo

From: Shawn Guo <[email protected]>

It renames GPIO register bits to drop 'SN_' prefix, so that they are consistent to other definitions - prefixing register name with 'SN_' but not for bit fields.

---
 drivers/gpu/drm/bridge/ti-sn65dsi86.c | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)

Sorry for taking so long to get back to this. I think it came into my inbox at the same time as a pile of other things and then got lost. Your change looks good to me. Sorry for being inconsistent when adding those defines and thanks for fixing them!

-Doug
_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel

reviewer
#2 lore raw

Hi,

Quoted 16 lines · Doug Anderson

Hi,

Quoted 11 lines · Shawn Guo

From: Shawn Guo <[email protected]>

It renames GPIO register bits to drop 'SN_' prefix, so that they are consistent to other definitions - prefixing register name with 'SN_' but not for bit fields.

---
 drivers/gpu/drm/bridge/ti-sn65dsi86.c | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)
Quoted 6 lines

Sorry for taking so long to get back to this. I think it came into my inbox at the same time as a pile of other things and then got lost. Your change looks good to me. Sorry for being inconsistent when adding those defines and thanks for fixing them!

Hrm, I just caught up on more email and found that in:

https://lore.kernel.org/r/[email protected]

...that Uwe would prefer to keep these bits what I have and change all the others. ;-) I don't have a strong opinion either way, but I definitely agree that it'd be better for all the defines to be consistent. If I had to arbitrarily make the decision one way or the other I'd probably land Shawn's patch but I certainly wouldn't object if we went Uwe's way either. :-P

-Doug
_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel

#3 lore raw
Quoted 28 lines · Doug Anderson

Hi,

Quoted 16 lines · Doug Anderson

Hi,

Quoted 11 lines · Shawn Guo

From: Shawn Guo <[email protected]>

It renames GPIO register bits to drop 'SN_' prefix, so that they are consistent to other definitions - prefixing register name with 'SN_' but not for bit fields.

---
 drivers/gpu/drm/bridge/ti-sn65dsi86.c | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)
Quoted 6 lines

Sorry for taking so long to get back to this. I think it came into my inbox at the same time as a pile of other things and then got lost. Your change looks good to me. Sorry for being inconsistent when adding those defines and thanks for fixing them!

Hrm, I just caught up on more email and found that in:

Quoted 8 lines

https://lore.kernel.org/r/[email protected]

...that Uwe would prefer to keep these bits what I have and change all the others. ;-) I don't have a strong opinion either way, but I definitely agree that it'd be better for all the defines to be consistent. If I had to arbitrarily make the decision one way or the other I'd probably land Shawn's patch but I certainly wouldn't object if we went Uwe's way either. :-P

For the the relevant argument for prefixes is that tools like ctags and cscope work more reliable. Take for example the name TX_TIMEOUT. There are around 60 symbols with that name in the kernel tree. This is quite annoying if you want to jump from a certain use to its definition.

Best regards
Uwe

Signature · 2 lines
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | https://www.pengutronix.de/ |

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel