This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Bad stmt.c patch
- To: Richard Kenner <kenner at vlsi1 dot ultra dot nyu dot edu>
- Subject: Re: Bad stmt.c patch
- From: Roger Sayle <roger at eyesopen dot com>
- Date: Thu, 25 Oct 2001 09:32:12 -0600 (MDT)
- cc: <gcc at gcc dot gnu dot org>, Andreas Jaeger <aj at suse dot de>
Hi Richard,
> It looks like this patch:
>
> 2001-10-24 Roger Sayle <roger@eyesopen.com>
>
> * stmt.c (expand_end_case): Index jumptables from zero for
> suitably small values of minval.
>
> causes a bootstrap failure on, at least x86. The patch, at first glance,
> looks OK to me, so I don't fully understand what's going on.
I apologise for any inconvenience. Obviously I didn't stress strongly
enough in http://gcc.gnu.org/ml/gcc-patches/2001-10/msg01234.html that
it required http://gcc.gnu.org/ml/gcc-patches/2001-10/msg01233.html to
be applied. I split the two to make them easier to review, and the
first fix should probably be applied, even if the second optimization
patch was rejected for other reasons.
Alternatively, rth may have considered law's comments an implicit approval
in http://gcc.gnu.org/ml/gcc-patches/2001-10/msg01273.html.
As soon as I noticed (within minutes) that this patch had been applied
without the previous one that it depended upon, I sent e-mails to both
Richard Henderson and Jeff Law pointing out the mistake.
I also see that you've correctly diagnosed the problem yourself in
http://gcc.gnu.org/ml/gcc-patches/2001-10/msg01302.html Hopefully
someone will apply this patch soon, so that GCC can be restored to
bootstrap land.
Once again, my apologies for the confusion.
Roger
--