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: Richard Kenner <kenner at vlsi1 dot ultra dot nyu dot edu>
- Subject: Re: Beyond GCC 3.0: Summing Up
- From: "H . J . Lu" <hjl at lucon dot org>
- Date: Tue, 10 Jul 2001 09:24:50 -0700
- Cc: gcc at gcc dot gnu dot org
- References: <10107101616.AA12204@vlsi1.ultra.nyu.edu>
On Tue, Jul 10, 2001 at 12:16:12PM -0400, Richard Kenner wrote:
> Presumably, changes are relatively small. Shouldn't put our energy
> into getting more releases out rather than making full qualifications?
>
> No, because every change has the potential of causing regressions and/or
> more bugs, even a "bug fix". People expect bug fix releases to be *higher*
> quality than the original release, so, if anything, we should do *more*
> qualifications of those releases than the original release.
That is a fallacy. Who are making "*more* qualifications", gcc
developers? Unless you are planning to rebuild a whole Linux/xxxBSD
distribution with the new gcc, there is very little you can do beyond
bootstrap and testsuite, which should be done for any releases. The
benefits of making frequent bug fix releaes are
1. Users don't have to wait 2 months to bootstrap on his/her platform.
2. We get less duplicate bug reports.
3. Reduce the number of regressions in the current major release over
the previous major release.
4. Users help us discover regressions.
5. Since changes are small, it is easier to fix the new regressions.
H.J.