Re: nolibc and __attribute__((__unused__))

[Willy Tarreau]

Hi Paul!

On Wed, Jul 21, 2021 at 01:33:49PM -0700, Paul E. McKenney wrote:
> Hello!
> 
> My guess is that I should ignore the following checkpatch complaint on
> the assumption that checkpatch doesn't realize that this is not built
> as part of the Linux kernel.  But if my guess is incorrect, please let
> me know, as it is a trivial change to make.
(...)
> WARNING: __always_unused or __maybe_unused is preferred over __attribute__((__unused__))
> #24: FILE: tools/include/nolibc/nolibc.h:2246:
> +static __attribute__((unused))

Yes you're totally right, we try to keep both trees in sync by minimizing
the differences between the two, so as long as that doesn't become a
problem I prefer to keep the warning than having to manually apply
future patches due to context differences.

Thanks!
Willy