Re: [Cocci] string.h: Mark 34 functions with __must_check

Re: [Cocci] string.h: Mark 34 functions with __must_check

[Markus Elfring]

> I reviewed the functions here and believe the ones you added checks
> for all look good.

Thanks for your positive feedback.


> Though Joe's comment on the relative order of where the
> annotation appears in the function declarations should be addressed in
> a V2 IMO.

Would you be looking for a subsequent change also by the means of
the semantic patch language with which the (function) attributes
can be adjusted to the ordering that you would prefer finally?

Regards,
Markus
_______________________________________________
Cocci mailing list
Cocci@systeme.lip6.fr
https://systeme.lip6.fr/mailman/listinfo/cocci

Re: [Cocci] Searching for missing variable checks

[Markus Elfring]

> The problem is the __must_check does not mean that the
> return value must be followed by a comparison to NULL and bailing out
> (that can't really be checked), it simply ensures the return value is
> assigned somewhere or used in an if(). So foo->bar = kstrdup() not
> followed by a check of foo->bar won't warn.

Higher level source code analysis tools like the semantic patch language
(Coccinelle software) can help to find such missing checks.
Would you like to point any additional development tools out
for this purpose?

Regards,
Markus
_______________________________________________
Cocci mailing list
Cocci@systeme.lip6.fr
https://systeme.lip6.fr/mailman/listinfo/cocci