This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Patch fix rtx_unstable_p and improve rtx_varies_p MEM handling
- To: rth at cygnus dot com
- Subject: Re: Patch fix rtx_unstable_p and improve rtx_varies_p MEM handling
- From: John Wehle <john at feith dot com>
- Date: Tue, 22 Aug 2000 17:51:09 -0400 (EDT)
- Cc: gcc-patches at gcc dot gnu dot org
> > * rtlanal.c (rtx_unstable_p): An unchanging MEM is
> > only stable if its address is stable.
> > (rtx_varies_p): An unchanging MEM can't vary if
> > its address doesn't vary.
>
> Better "An unchanging MEM can vary if its address varies",
> since that's the change you're making.
Umm ... actually that's not the change I'm making to rtx_varies_p.
The existing rtx_varies_p assumes that all MEMs vary ... my change
allows it to realize that a MEM which is unchanging and has an
unvarying address can't vary.
The logic itself is essentially the same for the rtx_unstable_p
and rtx_varies_p changes, however rtx_unstable_p is being made
more restrictive since it currently believes that all unchanging
MEMs are stable and rtx_varies_p is being made more liberal.
In any event I'm happy to use your wording if you feel it's clearer.
-- John
-------------------------------------------------------------------------
| Feith Systems | Voice: 1-215-646-8000 | Email: john@feith.com |
| John Wehle | Fax: 1-215-540-5495 | |
-------------------------------------------------------------------------