Re: [PATCH -next] I2C: Use helper function IS_ERR_OR_NULL()

[Ruan Jinjie <ruanjinjie@huawei.com>]

On 2023/8/16 18:08, Russell King (Oracle) wrote:
> On Wed, Aug 16, 2023 at 05:48:34PM +0800, Ruan Jinjie wrote:
>>
>>
>> On 2023/8/16 17:46, Ruan Jinjie wrote:
>>> Use IS_ERR_OR_NULL() instead of open-coding it
>>> to simplify the code.
>>>
>>> Signed-off-by: Ruan Jinjie <ruanjinjie@huawei.com>
>>> ---
>>>  drivers/i2c/busses/i2c-at91-master.c | 2 +-
>>>  drivers/i2c/busses/i2c-imx.c         | 2 +-
>>>  2 files changed, 2 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/i2c/busses/i2c-at91-master.c b/drivers/i2c/busses/i2c-at91-master.c
>>> index 94cff1cd527e..0e454c04a145 100644
>>> --- a/drivers/i2c/busses/i2c-at91-master.c
>>> +++ b/drivers/i2c/busses/i2c-at91-master.c
>>> @@ -831,7 +831,7 @@ static int at91_init_twi_recovery_gpio(struct platform_device *pdev,
>>>  	struct i2c_bus_recovery_info *rinfo = &dev->rinfo;
>>>  
>>>  	rinfo->pinctrl = devm_pinctrl_get(&pdev->dev);
>>> -	if (!rinfo->pinctrl || IS_ERR(rinfo->pinctrl)) {
>>> +	if (IS_ERR_OR_NULL(rinfo->pinctrl)) {
>>>  		dev_info(dev->dev, "can't get pinctrl, bus recovery not supported\n");
>>>  		return PTR_ERR(rinfo->pinctrl);
>>>  	}
>>> diff --git a/drivers/i2c/busses/i2c-imx.c b/drivers/i2c/busses/i2c-imx.c
>>> index 10e89586ca72..8807c90df749 100644
>>> --- a/drivers/i2c/busses/i2c-imx.c
>>> +++ b/drivers/i2c/busses/i2c-imx.c
>>> @@ -1388,7 +1388,7 @@ static int i2c_imx_init_recovery_info(struct imx_i2c_struct *i2c_imx,
>>>  	struct i2c_bus_recovery_info *rinfo = &i2c_imx->rinfo;
>>>  
>>>  	i2c_imx->pinctrl = devm_pinctrl_get(&pdev->dev);
>>> -	if (!i2c_imx->pinctrl || IS_ERR(i2c_imx->pinctrl)) {
>>> +	if (!IS_ERR_OR_NULL(i2c_imx->pinctrl)) {
>>
>> Sorry, there is a problem, please ignore it.
> 
> I don't think these conversions are appropriate for devm_pinctrl_get().
> If one reads the code for devm_pinctrl_get() and pinctrl_get(), then
> one finds that it will never return NULL.

Right! devm_pinctrl_get() never return NULL.

> 
> So the tests for NULL do not serve any practical purpose than giving
> coders a warm fuzzy feeling that they've tested for a NULL pointer.
> Practically, they are of no use, so should be removed... thus making
> the conversion to IS_ERR_OR_NULL() irrelevant.
>