This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Help with define_insn and constraints
- From: Ian Lance Taylor <ian at airs dot com>
- To: "Dave Korn" <dave dot korn at artimi dot com>
- Cc: "'Robert Baruch'" <autophile at gmail dot com>, <gcc at gcc dot gnu dot org>
- Date: 23 Dec 2004 16:37:40 -0500
- Subject: Re: Help with define_insn and constraints
- References: <NUTMEGR680We0j5y6vQ00000a60@NUTMEG.CAM.ARTIMI.COM>
"Dave Korn" <dave.korn@artimi.com> writes:
> > -----Original Message-----
> > From: gcc-owner On Behalf Of Robert Baruch
> > Sent: 23 December 2004 21:21
>
> > Well, after stepping through reload() a bunch of times, I determined
> > that first, the original instruction looks like this:
> >
> > [ P56 + 1 ] <- [ R20 + 17 ]
> >
> > where P56 is a pseudoreg, and R20 is my frame pointer.
>
> ?? In your first post, you said you had 19 hard regs. Does this not imply
> you're trying to keep your frame pointer in a pseudo ??
In gcc, it's normal for the frame pointer to be a pseudo which is
eliminated to the stack pointer, or, in the case of a function which
calls alloca, to be eliminated to some hard register which serves as
the so-called hard frame pointer. See FRAME_POINTER_REGNUM and
HARD_FRAME_POINTER_REGNUM.
Ian