Re: make-bogus-warnings-go-away tree [was: 2.6.18-mm3]

[Kyle Moffett <mrmacman_g4@mac.com>]

On Oct 05, 2006, at 06:05:16, Jeff Garzik wrote:
> Ingo Molnar wrote:
> > A small suggestion: to give GCC folks a chance to actually fix  
> > this, could we actively annotate these places instead of working  
> > them around?
>
> There was a patch posted in the past, mentioned in the thread  
> discussed my #gccbug branch, that permitted annotations with zero  
> code size changes.  I think that sort of annotation approach would  
> be preferred. It was something like
>
> #define noinit_warning(x) \
> 	do { (void) (x) = (x); } while (0)
>
> but given my memory, that's probably all wrong.

The simplest way given the current GCC feature-set is:

  #ifdef HIDE_GCC_FALSE_POSITIVES
  # define correct_init(x) x = x
  #else
  # define correct_init(x) x
  #endif

Then:

  int correct_init(arg);
  struct some_struct correct_init(foo);

Alternatively if only some struct member has problems and the rest  
are OK:

  struct some_struct foo;
  correct_init(foo.bar);

Cheers,
Kyle Moffett

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/