From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-return-54125-listarch-gcc=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 21078 invoked by alias); 31 May 2002 23:13:00 -0000
Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Archive: <http://gcc.gnu.org/ml/gcc/>
List-Post: <mailto:gcc@gcc.gnu.org>
List-Help: <http://gcc.gnu.org/ml/>
Sender: gcc-owner@gcc.gnu.org
Received: (qmail 21070 invoked from network); 31 May 2002 23:13:00 -0000
Received: from unknown (HELO cygnus.com) (205.180.83.203)
  by sources.redhat.com with SMTP; 31 May 2002 23:13:00 -0000
Received: from greed.delorie.com (cse.sfbay.redhat.com [205.180.230.236])
	by runyon.cygnus.com (8.8.7-cygnus/8.8.7) with ESMTP id QAA06876
	for <gcc@gcc.gnu.org>; Fri, 31 May 2002 16:12:58 -0700 (PDT)
Received: (from dj@localhost)
	by greed.delorie.com (8.11.6/8.11.6) id g4VN7O126234;
	Fri, 31 May 2002 19:07:24 -0400
Date: Fri, 31 May 2002 22:44:00 -0000
Message-Id: <200205312307.g4VN7O126234@greed.delorie.com>
From: DJ Delorie <dj@redhat.com>
To: gcc@gcc.gnu.org
Subject: output_func_start_profiler vs __bb_init_func
X-SW-Source: 2002-05/txt/msg03053.txt.bz2


It looks like there is an inconsistency in how the call to
__bb_init_func is created in profile.c.  Note the return mode as if
the function returned a GCOV_TYPE_SIZE int:

output_func_start_profiler ()
{
  enum machine_mode mode = mode_for_size (GCOV_TYPE_SIZE, MODE_INT, 0);
  ...
  emit_library_call (gen_rtx_SYMBOL_REF (Pmode, "__bb_init_func"), LCT_NORMAL,
		     mode, 1, table_address, Pmode);
}

But in libgcc2.c:

void
__bb_init_func (struct bb *blocks)

On targets where 64-bit ints are returned by reference (like big
structs), the call to __bb_init_func will be corrupted by the
unexpected extra (zeroth) parameter.

A simple replacement of mode -> VOIDmode seems to work.  Does this
sound right?