Add check for the return value of clk_enable() to catch the potential
error.

Fixes: e70a540b4e02 ("pwm: Add STM32 LPTimer PWM driver")
Signed-off-by: Mingwei Zheng <zmw12306@gmail.com>
Signed-off-by: Jiasheng Jiang <jiashengjiangcool@gmail.com>
---
 drivers/pwm/pwm-stm32-lp.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/pwm/pwm-stm32-lp.c b/drivers/pwm/pwm-stm32-lp.c
index 989731256f50..4abef304417d 100644
--- a/drivers/pwm/pwm-stm32-lp.c
+++ b/drivers/pwm/pwm-stm32-lp.c
@@ -163,12 +163,16 @@ static int stm32_pwm_lp_get_state(struct pwm_chip *chip,
 	unsigned long rate = clk_get_rate(priv->clk);
 	u32 val, presc, prd;
 	u64 tmp;
+	int ret;
 
 	regmap_read(priv->regmap, STM32_LPTIM_CR, &val);
 	state->enabled = !!FIELD_GET(STM32_LPTIM_ENABLE, val);
 	/* Keep PWM counter clock refcount in sync with PWM initial state */
-	if (state->enabled)
-		clk_enable(priv->clk);
+	if (state->enabled) {
+		ret = clk_enable(priv->clk);
+		if (ret)
+			return ret;
+	}
 
 	regmap_read(priv->regmap, STM32_LPTIM_CFGR, &val);
 	presc = FIELD_GET(STM32_LPTIM_PRESC, val);
-- 
2.34.1

On Thu, Dec 05, 2024 at 12:17:46AM -0500, Mingwei Zheng wrote:
> Add check for the return value of clk_enable() to catch the potential
> error.

Is this something that you actually hit, or just a janitoral fix you
noticed while browsing the code (or reading some checker output)?

> Fixes: e70a540b4e02 ("pwm: Add STM32 LPTimer PWM driver")
> Signed-off-by: Mingwei Zheng <zmw12306@gmail.com>
> Signed-off-by: Jiasheng Jiang <jiashengjiangcool@gmail.com>
> ---
>  drivers/pwm/pwm-stm32-lp.c | 8 ++++++--
>  1 file changed, 6 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/pwm/pwm-stm32-lp.c b/drivers/pwm/pwm-stm32-lp.c
> index 989731256f50..4abef304417d 100644
> --- a/drivers/pwm/pwm-stm32-lp.c
> +++ b/drivers/pwm/pwm-stm32-lp.c
> @@ -163,12 +163,16 @@ static int stm32_pwm_lp_get_state(struct pwm_chip *chip,
>  	unsigned long rate = clk_get_rate(priv->clk);
>  	u32 val, presc, prd;
>  	u64 tmp;
> +	int ret;

Please move this variable to the block where it's used. No need for such
a big scope.

Otherwise looks fine.

Best regards
Uwe

> On Dec 5, 2024, at 3:34 AM, Uwe Kleine-König <ukleinek@kernel.org> wrote:
> 
> On Thu, Dec 05, 2024 at 12:17:46AM -0500, Mingwei Zheng wrote:
>> Add check for the return value of clk_enable() to catch the potential
>> error.
> 
> Is this something that you actually hit, or just a janitoral fix you
> noticed while browsing the code (or reading some checker output)?

We detected this through static analysis, instead of actually hit.

> 
>> Fixes: e70a540b4e02 ("pwm: Add STM32 LPTimer PWM driver")
>> Signed-off-by: Mingwei Zheng <zmw12306@gmail.com>
>> Signed-off-by: Jiasheng Jiang <jiashengjiangcool@gmail.com>
>> ---
>> drivers/pwm/pwm-stm32-lp.c | 8 ++++++--
>> 1 file changed, 6 insertions(+), 2 deletions(-)
>> 
>> diff --git a/drivers/pwm/pwm-stm32-lp.c b/drivers/pwm/pwm-stm32-lp.c
>> index 989731256f50..4abef304417d 100644
>> --- a/drivers/pwm/pwm-stm32-lp.c
>> +++ b/drivers/pwm/pwm-stm32-lp.c
>> @@ -163,12 +163,16 @@ static int stm32_pwm_lp_get_state(struct pwm_chip *chip,
>> unsigned long rate = clk_get_rate(priv->clk);
>> u32 val, presc, prd;
>> u64 tmp;
>> + int ret;
> 
> Please move this variable to the block where it's used. No need for such
> a big scope.
> 
> Otherwise looks fine.
> 
> Best regards
> Uwe

Thank you! We will submit a v2 patch for you to review!

Best,
Mingwei