Re: [XEN PATCH] xen: add acmacros.h to exclude-list.json

[Nicola Vetrini <nicola.vetrini@bugseng.com>]

On 2023-12-20 09:50, Luca Fancellu wrote:
>> On 20 Dec 2023, at 08:41, Jan Beulich <jbeulich@suse.com> wrote:
>> 
>> On 20.12.2023 01:17, Stefano Stabellini wrote:
>>> On Tue, 19 Dec 2023, Luca Fancellu wrote:
>>>>> On 19 Dec 2023, at 11:05, Nicola Vetrini 
>>>>> <nicola.vetrini@bugseng.com> wrote:
>>>>> On 2023-12-19 11:51, Nicola Vetrini wrote:
>>>>>> On 2023-12-19 11:37, Jan Beulich wrote:
>>>>>>> On 19.12.2023 10:02, Nicola Vetrini wrote:
>>>>>>>> --- a/docs/misra/exclude-list.json
>>>>>>>> +++ b/docs/misra/exclude-list.json
>>>>>>>> @@ -209,6 +209,10 @@
>>>>>>>>          "rel_path": "include/acpi/acglobal.h",
>>>>>>>>          "comment": "Imported from Linux, ignore for now"
>>>>>>>>        },
>>>>>>>> +        {
>>>>>>>> +          "rel_path": "include/acpi/acmacros.h",
>>>>>>>> +          "comment": "Imported from Linux, ignore for now"
>>>>>>>> +        },
>>>>>>> Together with what's already there (in context), wouldn't it 
>>>>>>> better be
>>>>>>> the entire directory then which is excluded, or at least all
>>>>>>> include/acpi/ac*.h collectively (and perhaps also
>>>>>>> include/acpi/platform/ac*.h)?
>>>>>>> Jan
>>>>>> +Cc Luca Fancellu
>>>>>> Sure. I wasn't certain which files are imported from ACPI CA and 
>>>>>> which aren't.
>>>>>> I'm also not sure whether "include/acpi/ac*.h" would be properly 
>>>>>> recognized by other tooling that uses exclude-list.json (only 
>>>>>> cppcheck I think). I Cc-ed Luca Fancellu on this.
>>>>> 
>>>>> It occurred to me that it's surely ok to use "include/acpi/ac*" and 
>>>>> "include/acpi/platform/ac*".
>>>> 
>>>> Yes I think it’s fine, it just come to my mind now that this could 
>>>> have the risk that if
>>>> another file is added with ‘ac' prefix, even if it could be subject 
>>>> to MISRA compliance,
>>>> it will be excluded.
>>>> 
>>>> If that risk is negligible for the maintainer of that part, then 
>>>> it’s fine.
>>> 
>>> I think it is OK either way, I'll let Jan pick his preference.
>> 
>> It hasn't become clear to me what the benefit would be of omitting the
>> trailing .h.
> 
> Yes, with the extension is better, the same as we already do here:
> 
> [...]
> {
>     "rel_path": "common/un*.c”,
>     "comment": "unlz4.c implementation by Yann Collet, the others un* 
> are from Linux, ignore for now"
> },
> [...]

Ok, I'll send a v2

-- 
Nicola Vetrini, BSc
Software Engineer, BUGSENG srl (https://bugseng.com)