This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: More useful .greg dumps
- To: Michael Meissner <meissner at cygnus dot com>
- Subject: Re: More useful .greg dumps
- From: Bernd Schmidt <bernds at redhat dot co dot uk>
- Date: Tue, 31 Oct 2000 09:24:07 +0000 (GMT)
- cc: gcc-patches at gcc dot gnu dot org
On Mon, 30 Oct 2000, Michael Meissner wrote:
> On Mon, Oct 30, 2000 at 03:08:01PM +0000, Bernd Schmidt wrote:
> > When debugging reload, it would be nice to get a dump of the insns left after
> > it completes. However, the .greg dump is done after a few other optimization
> > passes have been run, e.g. reload_cse_regs and most importantly
> > split_all_insns, making the dump less useful than it could be.
> >
> > This patch splits up .19.greg into .19.greg (done after reload) and
> > .20.postreload (done immediately before flow2).
>
> I think it would be useful to do a dump after global has been run, but before
> reload gets its grubby little paws on the RTL. Unfortunately, global calls
> reload directly, rather than going back to toplev.
You can use the .lreg dump. global shouldn't modify rtl.
What might be useful is to dump out the initial register assignments before reload
has spilled anything.
Bernd