Re: build with clang-3.4 warnings report

[Rich Felker <dalias@aerifal.cx>]

On Mon, Apr 07, 2014 at 11:17:55AM +0000, Oliver Schneider wrote:
> On 2014-04-05 01:54, Rich Felker wrote:
> >> src/env/__init_security.c:8:13: warning: function 'dummy' is not needed and will not be emitted [-Wunneeded-internal-declaration]
> >> static void dummy(void *ent)
> >>             ^
> >> 1 warning generated.
> > 
> > This is a bug in the compiler; the function dummy has an externally
> > visible alias, so it will not be omitted (and if it is actually
> > omitted, that means the compiler is generating broken output).
> There is yet another possibility ;)
> 
> The ccc-analyzer could simply be buggy in this case. Would be
> interesting to look at the optimized intermediate code whether this gets
> optimized out. I think it probably doesn't, but who knows.

At one point it was getting optimized out, which made it impossible to
build musl with clang, so I think this was fixed and it's just a
residual bug in the warning generation.

> The dangling else warnings could also be resolved by simply adding
> braces. It doesn't add to the code size, and the times when developers
> where paid by LoC are long over (I hope).

Our developers are paid by 1/LoC. ;-)

Rich