This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: ix86 -fomit-frame-pointer bootstrap fix
- To: Jan Hubicka <hubicka at atrey dot karlin dot mff dot cuni dot cz>
- Subject: Re: ix86 -fomit-frame-pointer bootstrap fix
- From: Jeffrey A Law <law at cygnus dot com>
- Date: Tue, 28 Mar 2000 08:33:58 -0700
- cc: egcs-patches at egcs dot cygnus dot com
- Reply-To: law at cygnus dot com
In message <20000325185300.A13569@atrey.karlin.mff.cuni.cz>you write:
>
> Hi
> As jeff asked for, I've investigated this problem.
> The -O2 -fomit-frame-pointer/ix86 failures are caused by MEM sharing
> in combine_stack_adjustemnts. Sorry, I have no trivial testcase for this
> bug.
>
> Sat Mar 25 18:24:30 CET 2000 Jan Hubicka <jh@suse.cz>
>
> * regmove.c (struct csa_memlist): Make mem field rtx *.
> (record_one_stack_ref): Accept rtx * instead of rtx as parameter.
> (try_apply_stack_adjustment): Replace whole MEM rtx.
> (combine_stack_adjustments_for_block): Update calls
> to record_one_stack_ref.
Richard reviewed the patch a couple days ago; I just did an x86 bootstrap
(with -fomit-frame-pointer) and a run of the testsuite without any regressions.
So I went ahead and checked in the patch.
Thanks!
jeff
ps. It would help if you included some explanation of the problem -- it is
almost impossible to review patches when they do not include some explanation
of the bug and how the patch fixes the bug.