From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-return-101807-listarch-gcc=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 28464 invoked by alias); 31 Aug 2004 23:15:56 -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 28456 invoked from network); 31 Aug 2004 23:15:55 -0000
Received: from unknown (HELO vlsi1.ultra.nyu.edu) (128.122.140.213)
  by sourceware.org with SMTP; 31 Aug 2004 23:15:55 -0000
Received: by vlsi1.ultra.nyu.edu (4.1/1.34)
	id AA10724; Tue, 31 Aug 04 19:18:48 EDT
Date: Tue, 31 Aug 2004 23:52:00 -0000
From: kenner@vlsi1.ultra.nyu.edu (Richard Kenner)
Message-Id: <10408312318.AA10724@vlsi1.ultra.nyu.edu>
To: gcc@gcc.gnu.org
Subject: TREE_READONLY on SAVE_EXPR
X-SW-Source: 2004-08/txt/msg01697.txt.bz2

When we gimplify a SAVE_EXPR, it looks like TREE_READONLY isn't set on
the variable we make.   Therefore it seems wrong to set TREE_READONLY
on the SAVE_EXPR when it is created.  Should that be removed?