This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: question regarding regrename and failure of 950704-1.c on main
- From: law at redhat dot com
- To: "John David Anglin" <dave at hiauly1 dot hia dot nrc dot ca>
- Cc: rth at redhat dot com, gcc at gcc dot gnu dot org
- Date: Fri, 10 May 2002 07:28:39 -0600
- Subject: Re: question regarding regrename and failure of 950704-1.c on main
- Reply-to: law at redhat dot com
In message <200205100513.g4A5DYop015013@hiauly1.hia.nrc.ca>, "John David Anglin
" writes:
> > Let's hold on here. Clearly something is wrong as HARD_REGNO_MODE_OK
> > is supposed to reject one of those two register pairs. I think we need
> > to figure that out before we go mucking up the DImode patterns.
>
> The patch which caused the regression is
>
> Tue Apr 30 09:31:59 2002 Jeffrey A Law (law@cygnus.com)
>
> * pa.c (override_options): Default to PA8000 scheduling.
> * doc/invoke.texi (HP-PA options): Mention newly added 7300
> scheduling parameter.
>
> At first glance, it seems totally unrelated to the problem but the test
> fails with -mschedule=8000 (default) and passes with -mschedule=7100.
It's got to be merely exposing a latent problem. Dig deeper. Again,
why do we have reg:di 22 and reg:di 23. That's simply wrong.
jeff