This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] ifcvt: Handle INSV patterns properly when a temp reg is needed
- From: James E Wilson <wilson at specifix dot com>
- To: Roger Sayle <roger at eyesopen dot com>
- Cc: Andreas Krebbel <krebbel1 at de dot ibm dot com>, gcc-patches at gcc dot gnu dot org
- Date: Fri, 07 Apr 2006 13:25:50 -0700
- Subject: Re: [PATCH] ifcvt: Handle INSV patterns properly when a temp reg is needed
- References: <Pine.LNX.4.44.0604050759330.23200-100000@www.eyesopen.com>
On Wed, 2006-04-05 at 07:10, Roger Sayle wrote:
> On Mon, 3 Apr 2006, Andreas Krebbel wrote:
> I've Cc'd Jim on this e-mail. If he's also happy with this
> solution, it's OK for mainline. We can investigate the paradoxical
> subreg issue and what to do about 4.1 as follow-up patches once
> this is in the tree.
Well, I wouldn't say that I'm happy with it, but I can live with it.
Long term I think we should rip this code out and replace it with
something more sensible. Short term, this looks like a reasonable
workaround for the current bug.
There is one issue I see. There is a line
! unsigned HOST_WIDE_INT size = INTVAL (XEXP (x, 1));
but nowhere do we verify that this is a CONST_INT before we apply
INTVAL. We do verify that XEXP (x, 2) is a CONST_INT before using
INTVAL. So I think that code should be extended to also check XEXP (x,
1).
--
Jim Wilson, GNU Tools Support, http://www.specifix.com