Re: [RFC] clang: 'unused-function' warning on static inline functions

[Matthias Kaehlcke]

El Tue, Jun 06, 2017 at 09:32:35AM -0700 Linus Torvalds ha dit:

> On Tue, Jun 6, 2017 at 4:16 AM, Arnd Bergmann <arnd@xxxxxxxx> wrote:
> >
> > Those should all be fairly easy to address, I'd vote for leaving the
> > warning enabled
> > in clang, and possibly asking the gcc maintainers to add a similar feature for
> > warning about it.
> 
> Hell no. That warning is pointless shit.

I tend to disagree, the warning is useful to detect truly unused
static inline functions, which should be removed, rather than be
carried around/maintained for often long periods of time.

Also lets not forget that the kernel is somewhat special with its
heavy use of #ifdefs , which leads to the larger number of false
positives. At least for many other projects the warning makes perfect
sense IMO. A switch to enable/disable it could be interesting.

> The function is inlined. If it's not used, why the f*ck should you get
> a warning?

To detect dead code?

> Just to add more pointless #ifdef'fery around a function that is
> perhaps only used under certain circumstances?

The vast majority of instances could be fixed by simply marking the
function as __maybe_unused, which is already widely used in the
kernel. I don't fully understand why adding the attribute to static
inline functions (in .c files) is considered an offense, while it is a
standard practice for non-inline function. Granted, unused static
inline functions don't waste (binary) space, but doesn't the extra
capability to detect dead code also have a value?