From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-bugs-return-384139-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 849 invoked by alias); 21 Feb 2012 13:28:32 -0000
Received: (qmail 836 invoked by uid 22791); 21 Feb 2012 13:28:31 -0000
X-SWARE-Spam-Status: No, hits=-2.9 required=5.0	tests=ALL_TRUSTED,AWL,BAYES_00
X-Spam-Check-By: sourceware.org
Received: from localhost (HELO gcc.gnu.org) (127.0.0.1)    by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Tue, 21 Feb 2012 13:28:18 +0000
From: "rguenth at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug rtl-optimization/52326] [4.6 Regression] float result incorrect with -O1 and calling external function.
Date: Tue, 21 Feb 2012 13:34:00 -0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: rtl-optimization
X-Bugzilla-Keywords: wrong-code
X-Bugzilla-Severity: normal
X-Bugzilla-Who: rguenth at gcc dot gnu.org
X-Bugzilla-Status: NEW
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 4.6.3
X-Bugzilla-Changed-Fields:
Message-ID: <bug-52326-4-G5eoMaVKax@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-52326-4@http.gcc.gnu.org/bugzilla/>
References: <bug-52326-4@http.gcc.gnu.org/bugzilla/>
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
Content-Type: text/plain; charset="UTF-8"
MIME-Version: 1.0
Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-bugs.gcc.gnu.org>
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: 2012-02/txt/msg02073.txt.bz2

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52326

--- Comment #2 from Richard Guenther <rguenth at gcc dot gnu.org> 2012-02-21 13:27:23 UTC ---
Reduced single-file testcase, fails at -O1:

float fabsf(float x);
void abort (void);

static float minf(float a, float b)
{
  return (a < b) ? a: b;
}

static float maxf(float a, float b)
{
  return (a > b) ? a: b;
}

struct EC {
    float fftout;
    int extragainOn;
};

static float __attribute__((noinline))
foo (float a)
{
  asm("");
  return a;
}

static struct EC * __attribute__((noinline))
ec_create(void)
{
  static struct EC p;

  p.fftout = 1.234567f;
  p.extragainOn = 1;
  return &p;
}

static float ec_process(struct EC *p, float beflev, float estlev)
{
  float fullgCorr;
  float extrag;

  fullgCorr = maxf(0.0f, beflev - estlev) / beflev;
  extrag = 1.0f;

  if (p->extragainOn)
    extrag = minf(1.0f, fullgCorr);

  foo(0.0f);
  return extrag * p->fftout;
}

int main(void)
{
  unsigned int i = 0;
  float beflev[] = {11.6f, 1.0f};
  float estlev[] = {11.5f, 1.0f};
  float output;
  struct EC *p = ec_create();

  /* NOTE: If the for-loop is removed, the error disappears */
  for (i = 0; i < 1; i++)
    output = ec_process(p, beflev[i], estlev[i]);

  if (fabsf(output - 0.010643f) < 1e-5f)
    return 0;
  else
    abort ();
}