From: Mark Brown broonie@linaro.org
At least the sun6i DMA controller driver makes use of the reset controller API stubs when being built for coverage and otherwise fails to build.
Signed-off-by: Mark Brown broonie@linaro.org --- include/linux/reset.h | 7 +++++++ 1 file changed, 7 insertions(+)
diff --git a/include/linux/reset.h b/include/linux/reset.h index 349f150..28f119f 100644 --- a/include/linux/reset.h +++ b/include/linux/reset.h @@ -67,6 +67,13 @@ static inline int device_reset_optional(struct device *dev) return -ENOSYS; }
+static inline struct reset_control *devm_reset_control_get(struct device *dev, + const char *id) +{ + WARN_ON(1); + return ERR_PTR(-ENOSYS); +} + static inline struct reset_control *reset_control_get_optional( struct device *dev, const char *id) {
Hi Mark,
Am Freitag, den 01.08.2014, 18:11 +0100 schrieb Mark Brown:
From: Mark Brown broonie@linaro.org
At least the sun6i DMA controller driver makes use of the reset controller API stubs when being built for coverage and otherwise fails to build.
Signed-off-by: Mark Brown broonie@linaro.org
include/linux/reset.h | 7 +++++++ 1 file changed, 7 insertions(+)
diff --git a/include/linux/reset.h b/include/linux/reset.h index 349f150..28f119f 100644 --- a/include/linux/reset.h +++ b/include/linux/reset.h @@ -67,6 +67,13 @@ static inline int device_reset_optional(struct device *dev) return -ENOSYS; } +static inline struct reset_control *devm_reset_control_get(struct device *dev,
const char *id)+{
- WARN_ON(1);
- return ERR_PTR(-ENOSYS);
+}
I'd prefer to keep devm_reset_control_get without a stub. Drivers that can be built without CONFIG_RESET_CONTROLLER enabled should use devm_reset_control_get_optional, and this way the build tests already report this issue.
regards Philipp
linaro-kernel@lists.linaro.org
-
Mark Brown -
Philipp Zabel