Re: Building 100 kernels; we suck at dependencies and drown in warnings

[Stephen Hemminger <shemminger@osdl.org>]

On Sun, 26 Feb 2006 23:12:51 +0100
"Jesper Juhl" <jesper.juhl@gmail.com> wrote:

> On 2/26/06, Lee Revell <rlrevell@joe-job.com> wrote:
> > On Sun, 2006-02-26 at 22:56 +0100, Jesper Juhl wrote:
> > > Yeah so gcc is not perfect, but that still doesn't change that the
> > > intention of the warning and the use of the word "might" is as I said
> > > above.
> >
> > Not a very compelling case for changing the kernel rather than getting
> > GCC fixed.
> >
> 
> I think we are misunderstanding eachother. Or rather, I seem to have
> misread what Nix wrote.
> 
> I saw  "(i.e., there's a reason that warning uses the word *might*.)"
> and mistakenly read it as a question - "is there a reason that warning
> uses the word *might*?".
> I then proceeded to answer that question.
> When I read your latest mail I then couldn't make sense of things any
> longer and went back and read the previous mails again and realized my
> mistake.
> 
> My bad, sorry.

I went hunting for this in the GCC bugzilla, and one bug basically said.
"Yeah, we know the initialization checking code doesn't work right, but
 fixing it is too hard"