On Mon, 29 Mar 2021 18:25:46 +0300 Andy Shevchenko wrote:

On Mon, Mar 29, 2021 at 08:08:52AM -0700, Joe Perches wrote:

quoted

On Mon, 2021-03-29 at 14:59 +0300, Andy Shevchenko wrote:  

quoted

On Mon, Mar 29, 2021 at 2:43 PM Matti Vaittinen
[off-list ref] wrote:  

quoted

The checkpacth instructs to switch from ENOSUPP to EOPNOTSUPP.  

quoted

WARNING: ENOTSUPP is not a SUSV4 error code, prefer EOPNOTSUPP  

Make the gpiolib allow drivers to return both so driver developers
can avoid one of the checkpatch complaints.  

Internally we are fine to use the ENOTSUPP.
Checkpatch false positives there.

I doubt we need this change. Rather checkpatch should rephrase this to
point out that this is only applicable to _user-visible_ error path.
Cc'ed Joe.  

Adding CC for Jakub Kicinski who added that particular rule/test.

And the output message report of the rule is merely a suggestion indicating
a preference.  It's always up to an individual to accept/reject.

At best, perhaps wordsmithing the checkpatch message might be an OK option.  

Thanks, Joe!

Jakub, what do you think?

Agreed, weaving into the message that ENOTSUPP is okay internally
sounds good. Perhaps we should append "if error may be returned to 
user space"?