Skip to contentSkip to conversation

regulator/mfd: fix pointer-to-int cast

PATCH In review Arnd Bergmann 2 participants 2 messages Oct 2, 2018 – Oct 3, 2018

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 regulator/mfd: fix pointer-to-int cast

gcc points out that a pointer is longer than an int on 64-bit architectures, and that casting between the two may be dangerous:

drivers/mfd/rohm-bd718x7.c: In function 'bd718xx_i2c_probe':
drivers/mfd/rohm-bd718x7.c:101:23: error: cast from pointer to integer of different size [-Werror=pointer-to-int-cast]

In this driver it is correct, we just need the right kind of cast to avoid the warning.

Fixes 494edd266b94 ("regulator/mfd: Support ROHM BD71847 power management IC")
---
 drivers/mfd/rohm-bd718x7.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/mfd/rohm-bd718x7.c +1 -1
index 161c8aac6d86..e66d59190a82 100644
@@ -98,7 +98,7 @@ static int bd718xx_i2c_probe(struct i2c_client *i2c,
 		return -ENOMEM;
 
 	bd718xx->chip_irq = i2c->irq;
-	bd718xx->chip_type = (unsigned int)
+	bd718xx->chip_type = (uintptr_t)
 				of_device_get_match_data(&i2c->dev);
 	bd718xx->dev = &i2c->dev;
 	dev_set_drvdata(&i2c->dev, bd718xx);
Signature · 1 line
2.18.0
#1 lore raw

Hello Arnd,

Quoted 20 lines · Arnd Bergmann

gcc points out that a pointer is longer than an int on 64-bit architectures, and that casting between the two may be dangerous:

drivers/mfd/rohm-bd718x7.c: In function 'bd718xx_i2c_probe':
drivers/mfd/rohm-bd718x7.c:101:23: error: cast from pointer to integer of different size [-Werror=pointer-to-int-cast]

In this driver it is correct, we just need the right kind of cast to avoid the warning.

Fixes 494edd266b94 ("regulator/mfd: Support ROHM BD71847 power management IC")
---
 drivers/mfd/rohm-bd718x7.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/mfd/rohm-bd718x7.c +0 -0
index 161c8aac6d86..e66d59190a82 100644

drivers/mfd/rohm-bd718x7.c static int bd718xx_i2c_probe(struct i2c_client *i2c, · line 104

Quoted 8 lines
drivers/mfd/rohm-bd718x7.c +1 -1
@@ -98,7 +98,7 @@ static int bd718xx_i2c_probe(struct i2c_client *i2c,
 		return -ENOMEM;
 
 	bd718xx->chip_irq = i2c->irq;
-	bd718xx->chip_type = (unsigned int)
+	bd718xx->chip_type = (uintptr_t)
 				of_device_get_match_data(&i2c->dev);
 	bd718xx->dev = &i2c->dev;
 	dev_set_drvdata(&i2c->dev, bd718xx);

Big thanks for this activity =) I also got heads up mail from linux-next so I also did a patch fixing this. I think Mark alrady applied that patch to his tree.

(https://lore.kernel.org/lkml/[email protected]/)

Br,
Matti Vaittinen