This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: PATCH for fold-const.c: take 2
- To: mark at markmitchell dot com
- Subject: Re: PATCH for fold-const.c: take 2
- From: Franz Sirl <Franz dot Sirl-kernel at lauterbach dot com>
- Date: Fri, 26 Jun 1998 13:55:37 +0200
- Cc: law at cygnus dot com, egcs-patches at cygnus dot com
- References: <465.898820062@hurl.cygnus.com><465.898820062@hurl.cygnus.com>
At 02:29 26.06.98 , Mark Mitchell wrote:
>>>>>> "Jeffrey" == Jeffrey A Law <law@hurl.cygnus.com> writes:
>
> Jeffrey> In message
> Jeffrey> <199806252036.NAA11445@smtp.earthlink.net>you write:
> >> The patch I just sent was written while I was under an evil
> >> influence, and is total garbage.
> >>
> >> Here's a better version.
> >>
> >>
> >> Thu Jun 25 12:47:41 1998 Mark Mitchell <mark@markmitchell.com>
> >>
> >> * fold-const.c (make_range): Don't go looking at TREE_OPERANDs
> >> of nodes that are not expressions.
>
> Jeffrey> Just curious -- what kinds of nodes were showing up which
> Jeffrey> caused you to look at this?
>
>PARM_DECLs. If you look at the code, you'll see that it was goes
>through the loop until it finds something it doesn't understand. On
>that last iteration, though, it was still grabbing TREE_OPERANDs. I
>found this by configuring with --enable-checking as I was doing one
>more test of the type-based alias stuff before checking it in.
Aah, this hopefully means that I can now bootstrap with --enable-checking
on powerpc-unknown-linux-gnu. Before it was bailing out in fold-const.c
while compiling libgcc as reported in
http://www.cygnus.com/ml/egcs-bugs/1998-Jun/0017.html
(In the meantime I have stripped down the original bug to
http://www.cygnus.com/ml/egcs-bugs/1998-Jun/0275.html)
I will try the patch, Thanks!
Franz.