This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Adding files for Ada
- To: jsm28 at cam dot ac dot uk, kenner at vlsi1 dot ultra dot nyu dot edu
- Subject: Re: Adding files for Ada
- From: dewar at gnat dot com
- Date: Fri, 25 May 2001 11:26:58 -0400 (EDT)
- Cc: Florian dot Weimer at rus dot uni-stuttgart dot de, gcc at gcc dot gnu dot org
<<Whereas I think we should keep the bug reporting procedures simple, and
have exactly one place - the public GCC GNATS system, with gccbug used to
submit reports. Presumably you and the other Ada maintainers who have or
will have GCC CVS write access also have or will also have GNATS access.
GNATS can be configured for all bugs in category "ada" to be assigned to
you or another GNU Ada maintainer automatically.
I don't think we should direct bug reports for any part of public GCC
releases or CVS to any non-public bug tracking system. Are bug reports
the public have submitted accessible to the public in the ACT system to
which you refer?
>>
The answer to the last question is no, reports on the public version of
GNAT go into a black hole from the point of the submitter when sent
to report@gnat.com (i.e. we receive bug reports there, but we don't
track them or correspond with people on them, though we do examine
all reports, and fix things where appropriate).
Indeed I agree with Joseph's position on this point, and my reply above
indeed stregthens the point here. I would like GNAT to look as much a
part of stanbdard gcc as possible in all respects, and bug reporting is
one of these respects.
Many bug reports are, as everyone knows, not bug reports at all, but
rather confusion or misunderstandings. Right now, if such a report
goes to report@gnat.com without a customer support number, it is
basically ignored (we don't provide the service there of answering
such questions, and indeed the auto-response directs them elsewhere,
currently to comp.lang.ada for language questions, and to ??? for
compiler questions. The ??? essentially is undefined at the moment,
pending sorting out the answer to the question being discussed here.
I think it is much better if all reports on the public GCC releases
go to a public bug tracking system, where the community can respond.
The community of course includes individuals working at ACT, and
indeed it is somewhat easier for us to participate as individuals
in this context than operate from gnat.com giving support to
unsupported users.
I don't think the volume of reports will be that high in any case,
and most certainly we can take "real" bug reports into our internal
system as we need to, so they won't get ignored.
Robert Dewar