Re: [XEN PATCH 3/5] x86/irq: address violation of MISRA C Rule 5.5

[Jan Beulich <jbeulich@suse.com>]

On 07.07.2025 23:25, Stefano Stabellini wrote:
> On Mon, 7 Jul 2025, Jan Beulich wrote:
>> On 04.07.2025 22:39, Dmytro Prokopchuk1 wrote:
>>> Address a violation of MISRA C:2012 Rule 5.5:
>>> "Identifiers shall be distinct from macro names".
>>>
>>> Reports for service MC3A2.R5.5:
>>> xen/include/xen/irq.h: non-compliant function `pirq_cleanup_check(struct pirq*, struct domain*)'
>>> xen/include/xen/irq.h: non-compliant macro `pirq_cleanup_check'
>>>
>>> The primary issue stems from the macro and function
>>> having identical names, which is confusing and
>>> non-compliant with common coding standards.
>>>
>>> Change the function name by adding two underscores at the end.
>>>
>>> Signed-off-by: Dmytro Prokopchuk <dmytro_prokopchuk1@epam.com>
>>
>> I'm not going to NAK this, but I dislike the transformation done. The aliasing
>> in this case was intentional, to avoid any caller appearing that would bypass
>> the macro. Yes, the double underscores will also stand out (as much as the
>> parenthesization that would have been needed to override the protection), but
>> still ...
> 
> Maybe you can suggest a different name?

As per my earlier reply, using the same name was intentional here. Hence
it's not a matter of what (different) name to pick, but the mere fact that
a different name is being suggested to be used. Yet as said - I'm not
going to NAK this, but I also don't like the change.

Jan

> Looking at the diff, this patch also seems OKish.
> 
> It is possible but difficult to deviate specific instances like this: if
> a SAF in-code comment works, then great, otherwise we have to resort to
> a regex which makes thing harder to maintain.
> 
> Unless a SAF in-code comment works, I think this patch is the best way
> to go.