Re: checking changes for warning regressions

["H.J. Lu" <hjl dot tools at gmail dot com>]

On Tue, Oct 2, 2012 at 4:23 PM, Roland McGrath <roland@hack.frob.com> wrote:
> Your commit ae30640a of 2012-9-19 has two issues.
>
> First, you didn't update/collapse copyright years.

Fixed.

> Second, this change introduced several -Wpointer-to-int-cast warnings.
>
> Everybody, please consider it a responsibility for any change you do to
> try the compilation on some relevant configuration and verify that it
> does not introduce new warnings into the build with some fairly recent
> compiler (I'm using 4.6.3).
>
> In this case, the warning doesn't indicate any possible bug.  It's in an
> if branch that will always be eliminated dead code in any circumstance
> where the warning is issued.  But it's still bad to introduce such warnings.
> (I don't think I need to elaborate on why it's bad.  I can if necessary.)
>
> I haven't thought of a way to eliminate this warning without losing the
> functional aspect of your change.  But I haven't put much effort into
> finding one.  Please see if you can find one.
>

The warnings have always been there for x32 build.  I investigated it
while working on x32 and came to the same conclusion.   This is the
known problem:

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=4210

-- 
H.J.