Re: code generation bug

["Martin v. Loewis" <martin at loewis dot home dot cs dot tu-berlin dot de>]

> > here I don't investigate it because I trust
> > that the submitters will re-report it if it does not get fixed.
> 
> You disagree with Alexandre Oliva here, who writes
> (http://gcc.gnu.org/ml/gcc-bugs/2000-02/msg00617.html)

> I will gladly resubmit a bug report if it has not been resolved
> after a month or so;

I don't really disagree; re-reporting after only a month is not
necessary. I was rather thinking of re-reporting when it is not fixed
at the time the release branch is created.

> I will do this, thanks! It would be good to know what the different
> bug categories exactly mean, though. If you could give me a pointer
> to more information about GNATS and special GCC GNATS issues (like
> bug types etc.)  I could provide a patch for the HTML docs.

GNATS documentation is available at

http://sourceware.cygnus.com/gnats/gnats_toc.html

In the multiple-choice fields, it is probably easiest if you fill in
what you understand and what you think is most appropriate; if it is
incorrectly filed, somebody will correct the PR. Some of the choices
are really GCC specific; I'm not convinced that we need all the
categories, so I don't want to document it too early. The mainline is
primarily intended for compiler developers and experienced users; I
hope they can deal with the complexity.

Regards,
Martin