This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Beyond GCC 3.0: Summing Up
- To: Alexandre Oliva <aoliva at redhat dot com>, "dewar at gnat dot com" <dewar at gnat dot com>
- Subject: Re: Beyond GCC 3.0: Summing Up
- From: Mark Mitchell <mark at codesourcery dot com>
- Date: Tue, 10 Jul 2001 00:01:48 -0700
- cc: "gcc at gcc dot gnu dot org" <gcc at gcc dot gnu dot org>, "kenner at vlsi1 dot ultra dot nyu dot edu" <kenner at vlsi1 dot ultra dot nyu dot edu>
--On Tuesday, July 10, 2001 04:10:50 AM -0300 Alexandre Oliva
<aoliva@redhat.com> wrote:
> On Jul 10, 2001, dewar@gnat.com wrote:
>
>> Well I do think so! It is great to "expose" latent bugs *if* they
>> get fixed quickly. If not, then there is no choice but to back out
>> a change that causes regressions.
>
> But backing out the change would *also* cause a regression. The bug
> fixed by the patch would then start failing again. We'd be trading
> one failure for another.
I see what you mean, but this is not what I mean by `regression'.
A failure that did not exist in a previous release is different
from a failure that did. A failure that existed in a previous
release is one that people are already dealing with; a new failure
adds to their burden.
If you check in a patch that fixes a bug, but breaks something else,
and then back it out, you are back where you started. That is not
a regression; you just still have the same bug.
--
Mark Mitchell mark@codesourcery.com
CodeSourcery, LLC http://www.codesourcery.com