Criteria for a warning to be in -Wall? (was: Re: a warning to implement)

[Olivier Galibert]

On Thu, Feb 07, 2002 at 01:12:34AM -0200, Alexandre Oliva wrote:
> As Joe Buck pointed out, we *do* warn about the use of the
> uninitialized variable when writing `int a = a;'.  I'm pretty sure we
> didn't before, but now we do.

Probably only with -O, like every warning requiring flow analysis,
which may explain why you could have missed it in some cases.

I'd love a "perfect debugging as with -g only but with flow analysis
for uninitialized warnings" option, but it's probably not easy to add.

  OG.