From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-bugzilla@gcc.gnu.org>
Received: by sourceware.org (Postfix, from userid 48)
 id 22F1C3857C7E; Fri, 19 Nov 2021 16:11:19 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 22F1C3857C7E
From: "aldyh at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/103088] [12 regression] 500.perlbench from
 spec 2017 fails since r12-4698
Date: Fri, 19 Nov 2021 16:11:18 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: tree-optimization
X-Bugzilla-Version: 12.0
X-Bugzilla-Keywords: wrong-code
X-Bugzilla-Severity: normal
X-Bugzilla-Who: aldyh at gcc dot gnu.org
X-Bugzilla-Status: RESOLVED
X-Bugzilla-Resolution: INVALID
X-Bugzilla-Priority: P1
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 12.0
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-103088-4-AgscOA7JhB@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-103088-4@http.gcc.gnu.org/bugzilla/>
References: <bug-103088-4@http.gcc.gnu.org/bugzilla/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
MIME-Version: 1.0
X-BeenThere: gcc-bugs@gcc.gnu.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: Gcc-bugs mailing list <gcc-bugs.gcc.gnu.org>
List-Unsubscribe: <https://gcc.gnu.org/mailman/options/gcc-bugs>,
 <mailto:gcc-bugs-request@gcc.gnu.org?subject=unsubscribe>
List-Archive: <https://gcc.gnu.org/pipermail/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-request@gcc.gnu.org?subject=help>
List-Subscribe: <https://gcc.gnu.org/mailman/listinfo/gcc-bugs>,
 <mailto:gcc-bugs-request@gcc.gnu.org?subject=subscribe>
X-List-Received-Date: Fri, 19 Nov 2021 16:11:19 -0000

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D103088

--- Comment #19 from Aldy Hernandez <aldyh at gcc dot gnu.org> ---
Ughh, I was nerd sniped.  Couldn't let it go ;-).

The problem is this construct in Perl_do_ncmp:

      if (lnv < rnv)
        return -1;
      if (lnv > rnv)
        return 1;
      if (lnv =3D=3D rnv)
        return 0;
      return 2;

These are all doubles.  The code is depending on a pair of values that are
neither <, >, nor =3D=3D, being a NAN.

For -ffast-math with unsafe optimizations we end up inling Perl_do_ncmp into
pp_ge():

SETs(boolSV(
        (SvIOK_notUV(left) && SvIOK_notUV(right))
        ? (SvIVX(left) >=3D SvIVX(right))
        : ( (do_ncmp(left, right) & 2) =3D=3D 0)
    ));

but in doing so we destroy the 3 separate conditionals.  At one point we're
making decisions based on <=3D on the 35->38 edge:

struct OP * Perl_pp_ge ():
...
  <bb 35> [local count: 590686791]:
  # iftmp.547_110 =3D PHI <iftmp.547_107(33), iftmp.547_108(34)>
  if (iftmp.546_109 > iftmp.547_110)
    goto <bb 23>; [1.04%]
  else
    goto <bb 38>; [98.96%]

<snip>
<snip>

  <bb 38> [local count: 644407314]:

  <bb 39> [local count: 953267993]:
  # iftmp.532_21 =3D PHI <&PL_sv_yes(38), &PL_sv_no(37)>
  MEM[(struct SV * *)sp_27 + -8B] =3D iftmp.532_21;
  PL_stack_sp =3D sp_30;
  PL_op.534_19 =3D PL_op;
  _38 =3D PL_op.534_19->op_next;

  <bb 40> [local count: 1073741826]:
  # _20 =3D PHI <_29(5), _38(39)>
  return _20;

For -fno-unsafe-math-optimizations we avoid inlining Perl_no_ncmp, which ke=
eps
the conditionals.

So it looks like we inline the NAN checking code in unsafe mode and the
threader ends can make decisions on the return value for pp_ge().  As I sai=
d,
the threads are correct.

If anyone is curious, you can see what's going on by tagging Perl_do_ncmp()
with __attribute__((optimize("O3", "fast-math",
"no-unsafe-math-optimizations")))

and seeing the final pp_ge() output with the munged conditionals versus the
pristine code in the output for Perl_do_ncmp.

I think we can keep this PR closed.  Don't use -ffast-math unless followed =
by
-fno-unsafe-math-optimizations.=