Re: [patch 01/10] compiler: define __attribute_unused__

[Adrian Bunk <bunk@stusta.de>]

On Wed, May 02, 2007 at 12:22:24AM -0700, David Rientjes wrote:
> On Wed, 2 May 2007, Rusty Russell wrote:
> 
> > 	That sounds exactly right to me!  If the author says it's optional, it
> > might be discarded.  If they say it's needed, it won't be.  At least,
> > when I'm coding and gcc warns me something is unused, this is the
> > decision I have to make ("is this really needed or not?").
> > 
> 
> Hi Rusty,
> 
> There are many instances in the tree of functions that have no callers 
> whatsoever because they've been commented out temporarily, disabled 
> through configuration, etc.  These are marked __attribute__ ((unused)) 
> right now so that the compiler doesn't emit a warning (and with gcc >=3.4 
> it doesn't even emit code for them).  What's __optional about these 
> functions if they have no callers?  They're unused.  So we cover all our 
> bases with __maybe_unused.

"many ... are marked __attribute__ ((unused))" is not true:
$ grep -r __attribute_used__ * | wc -l
60
$ 

static inline functions don't result in warnings.

And for global functions, it is technically impossible for gcc to figure
out whether a function has any users.

Unused static non-inline functions are the only functions resulting in 
warnings when being unused.
If we don't want gcc to emit warnings for such, we could disable them 
globally.

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed