Re: [PATCH] staging: nvec: use x instead of x != NULL

[Philipp Hortmann <philipp.g.hortmann@gmail.com>]

On 6/26/24 06:48, Julia Lawall wrote:
> 
> 
> On Wed, 26 Jun 2024, Philipp Hortmann wrote:
> 
>> On 6/25/24 22:56, Tom Mounet wrote:
>>> Comply with coding rules defined in checkpatch
>>>
>>> Signed-off-by: Tom Mounet <tommounet@gmail.com>
>>> ---
>>>    drivers/staging/nvec/nvec.c | 4 ++--
>>>    1 file changed, 2 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/staging/nvec/nvec.c b/drivers/staging/nvec/nvec.c
>>> index e5ca78e57..814eb121c 100644
>>> --- a/drivers/staging/nvec/nvec.c
>>> +++ b/drivers/staging/nvec/nvec.c
>>> @@ -300,7 +300,7 @@ int nvec_write_sync(struct nvec_chip *nvec,
>>>    {
>>>    	mutex_lock(&nvec->sync_write_mutex);
>>>    -	if (msg != NULL)
>>> +	if (msg)
>>>    		*msg = NULL;
>>>      	nvec->sync_write_pending = (data[1] << 8) + data[0];
>>> @@ -322,7 +322,7 @@ int nvec_write_sync(struct nvec_chip *nvec,
>>>      	dev_dbg(nvec->dev, "nvec_sync_write: pong!\n");
>>>    -	if (msg != NULL)
>>> +	if (msg)
>>>    		*msg = nvec->last_sync_msg;
>>>    	else
>>>    		nvec_msg_free(nvec, nvec->last_sync_msg);
>>
>>
>> Hi Tom,
>>
>> what you change in this patch is fine. But the Description is not so lucky.
>> Reason is that checkpatch is not defining the coding style. Not at all.
>> Sometimes checkpatch is even wrong. The description I like would be:
>>
>> Use x instead of x != NULL to shorten code.
>>
>> or
>>
>> Use x instead of x != NULL to improve readability.
>>
>> If you send in a second version of this patch please use a change history.
>> Description from Dan under:
>> https://staticthinking.wordpress.com/2022/07/27/how-to-send-a-v2-patch/
> 
> How about adding "Issue identified by checkpatch"?  Checkpatch helped find
> the problem, so it would be nice to acknowledge that.
> 
> julia
> 

Hi Julia,

The following lines sound very authoritative. It is only my opinion and 
can be wrong.

I think checkpatch is valued a lot because every patch send in is 
checked by checkpatch. checkpatch can be mentioned in the description. 
But the developer cannot hide at all behind a checkpatch warning/error 
message. The developer must take full responsibility for the patch. The 
developer needs to use common sense.

Please have a look at this email from Greg:
https://lore.kernel.org/linux-staging/2024062443-udder-spotted-cc0d@gregkh/T/#m280ebb2be94e434234f405e722fc35dc6d1db710

I think that Greg once wrote that he does not care about the tool that 
found the issue. He much more cares about if the change makes sense. The 
"Why" in the description is most important for him. And the why cannot 
be because checkpatch or any other tool told the developer so.

Thanks for your support.

Bye Philipp