This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: function arguments offset computation with -fomit-frame-pointer

From: Jim Wilson <wilson at tuliptree dot org>
To: Bernardo Innocenti <bernie at develer dot com>
Cc: Hans-Peter Nilsson <hp at bitrange dot com>, gcc at gcc dot gnu dot org, Peter Barada <pbarada at mail dot wm dot sps dot mot dot com>
Date: 09 Jul 2003 12:28:34 -0700
Subject: Re: function arguments offset computation with -fomit-frame-pointer
References: <Pine.BSF.4.44.0307080400550.22413-100000@dair.pair.com><200307090947.20714.bernie@develer.com><1057738709.1057.180.camel@leaf.tuliptree.org> <200307092023.35366.bernie@develer.com>

On Wed, 2003-07-09 at 11:23, Bernardo Innocenti wrote:
>  You got me! How did you know? The check wasn't in the original gcc sources.

The m68k_save_reg function in the FSF tree handles the pic register. 
You posted your version of the INITIAL_FRAME_POINTER_OFFSET macro which
uses m68k_save_reg on all registers which is different from the FSF
code, but which still has the pic register check.  It wasn't hard for me
to notice that this caused the pic reg to be counted twice.

I don't see any obvious problem in your m68k_save_reg function.

Jim

References:
- Re: function arguments offset computation with -fomit-frame-pointer
  - From: Hans-Peter Nilsson
- Re: function arguments offset computation with -fomit-frame-pointer
  - From: Bernardo Innocenti
- Re: function arguments offset computation with -fomit-frame-pointer
  - From: Jim Wilson
- Re: function arguments offset computation with -fomit-frame-pointer
  - From: Bernardo Innocenti

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]