This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: cselib fix
- To: Bernd Schmidt <bernds at balti dot cygnus dot co dot uk>
- Subject: Re: cselib fix
- From: Jeffrey A Law <law at cygnus dot com>
- Date: Wed, 15 Mar 2000 13:22:07 -0700
- cc: gcc-patches at gcc dot gnu dot org
- Reply-To: law at cygnus dot com
In message <Pine.LNX.4.10.10003151851050.575-100000@balti.cygnus.co.uk>you wr
ite:
> This fixes the problem found by Jeff.
Thanks.
> I'm currently bootstrapping with this change; so far everything looks OK.
My bootstrap completed successfully.
> The problem was a rather subtle one arising from using the new hash
> tables. While I was there, I discovered that the hash function was
> returning rather stupid values and fixed that, too.
>
>
> Bernd
>
> * simplify-rtx.c (hash_rtx, case MEM/REG): Take into account that
> HASH may already be nonzero. Add code/mode into hash value
> immediately after repeat label.
> (cselib_lookup): Don't leave the hash table in an inconsistent
> state before a hash lookup operation.
This is fine. Please install it.
jeff