This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: are regression hunts worthwhile for very old C++ regressions?
- From: Mark Mitchell <mark at codesourcery dot com>
- To: Janis Johnson <janis187 at us dot ibm dot com>, "gcc at gcc dot gnu dot org" <gcc at gcc dot gnu dot org>
- Date: Tue, 14 Jan 2003 19:56:17 -0800
- Subject: Re: are regression hunts worthwhile for very old C++ regressions?
--On Tuesday, January 14, 2003 11:23:00 AM -0800 Janis Johnson
<janis187@us.ibm.com> wrote:
There are several C++ bug reports for regressions that last worked in
2.95, some of which are fixed by the new parser. Given that there were
a lot of changes to G++ between 2.95 and 3.0; is it worthwhile to
identify the patches that caused these regressions?
Probably not.
First, things that are already fixed are going to be less important to
fix than things that aren't; if 3.3 is the last release with the bug
that's better than letting it go on forever.
Second, the changes are big, so a lot of times you're going to find that
some major rewrite caused the bug.
Third, the code has changed so much since then that the original patch may
well be gone.
It sure won't hurt, but I bet your efforts will be more fruitful elsewhere.
--
Mark Mitchell mark@codesourcery.com
CodeSourcery, LLC http://www.codesourcery.com