From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-bugs-return-195750-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 5896 invoked by alias); 2 Sep 2006 21:14:24 -0000
Received: (qmail 5875 invoked by uid 48); 2 Sep 2006 21:14:17 -0000
Date: Sat, 02 Sep 2006 21:14:00 -0000
Subject: [Bug tree-optimization/28940]  New: Overzealous CSE with static array access
X-Bugzilla-Reason: CC
Message-ID: <bug-28940-10833@http.gcc.gnu.org/bugzilla/>
Reply-To: gcc-bugzilla@gcc.gnu.org
To: gcc-bugs@gcc.gnu.org
From: "lmakhlis at bmc dot com" <gcc-bugzilla@gcc.gnu.org>
Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Archive: <http://gcc.gnu.org/ml/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-help@gcc.gnu.org>
Sender: gcc-bugs-owner@gcc.gnu.org
X-SW-Source: 2006-09/txt/msg00163.txt.bz2
List-Id: <gcc-bugs.sourceware.org>

char a[10], b[10];

int f1(int i)
{
        return a[i+1] + b[i+1];
}

With -O1 and higher, gcc performs CSE on i+1:

        addl    $1, %edi
        movslq  %edi,%rdi
        movsbl  a(%rdi),%eax
        movsbl  b(%rdi),%edx
        addl    %edx, %eax
        ret

This doesn't happen with the equivalent
        return (&a[0])[i+1] + (&b[0])[i+1];
or
        return *(a + i + 1) + *(b + i + 1);
which both compile to
        movslq  %edi,%rdi
        movsbl  a+1(%rdi),%eax
        movsbl  b+1(%rdi),%edx
        addl    %edx, %eax
        ret


-- 
           Summary: Overzealous CSE with static array access
           Product: gcc
           Version: 4.1.1
            Status: UNCONFIRMED
          Severity: minor
          Priority: P3
         Component: tree-optimization
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: lmakhlis at bmc dot com
 GCC build triplet: x86_64-redhat-linux
  GCC host triplet: x86_64-redhat-linux
GCC target triplet: x86_64-redhat-linux


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28940