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 239AA3858D33; Tue,  7 Feb 2023 18:30:25 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 239AA3858D33
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1675794625;
	bh=zYo2l0RRO+nX2o3Qt0zzC91jRY4nb7zqSl/aap1FZj8=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=LGlUS+FSaCYsCyj6EH6TN9NLaTaB7NGYzbuvPUUe9tk18uD2IW43fLNOi/iYQeiqC
	 D2v4MKlAP4o3laBcJ0eo/Gx/LklnFyPVghpbTCGCMk8U4ghNmG1PzmP0AwrCZWysdS
	 Bvw/7CwNq6gnX8a4dRPg2xx0aUkYAYm8h1bqm0RI=
From: "amacleod at redhat dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/108684] [13 Regression] ICE: verify_ssa
 failed
Date: Tue, 07 Feb 2023 18:30:24 +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: 13.0
X-Bugzilla-Keywords: ice-on-valid-code, inline-asm
X-Bugzilla-Severity: normal
X-Bugzilla-Who: amacleod at redhat dot com
X-Bugzilla-Status: NEW
X-Bugzilla-Resolution: 
X-Bugzilla-Priority: P1
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 13.0
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-108684-4-t8TIiUmw5B@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-108684-4@http.gcc.gnu.org/bugzilla/>
References: <bug-108684-4@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
List-Id: <gcc-bugs.sourceware.org>

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D108684

--- Comment #3 from Andrew Macleod <amacleod at redhat dot com> ---
OK, its been a while.   Why is there a VUSE on the return?  this is the IL
right from into-ssa:

int f (int a)
{
  int _4;

  <bb 2> :
  # .MEM_2 =3D VDEF <.MEM_1(D)>
  __asm__(" " : "=3DX" a_3 :  : "memory");
  if (a_3 !=3D 0)
    goto <bb 3>; [INV]
  else
    goto <bb 4>; [INV]

  <bb 3> :
  _4 =3D 0;
  // predicted unlikely by early return (on trees) predictor.
  # VUSE <.MEM_2>
  return _4;

  <bb 4> :
  __builtin_unreachable ();

}

The problem is that in VRP, we change a_3's global range to be ~[0,0], rewr=
ite
the condition to remove the block with the builtin_unreachable... which the=
re
leaves the IL something like:

 <bb 2> [local count: 1073741824]:
  # .MEM_2 =3D VDEF <.MEM_1(D)>
  __asm__(" " : "=3DX" a_3 :  : "memory");

  <bb 3> [local count: 1073741824]:
  # VUSE <.MEM_2>
  return 0;

and since there are no more uses of a_3, the asm is considered dead and rem=
oved
by simple_dce_from_worklist.

BUt it does not adjust the vdefs at all.. so we when its deleted, we are st=
ill
eft with:
  <bb 3> [local count: 1073741824]:
  # VUSE <.MEM_2>
  return 0;

but now there is no longer a definition of MEM_2  and verify_ssa fails.

I can fix it by not trying to dce any defs that have VDEFS on them... but t=
hat
doesnt seem  like the correct long term solution... Why is there a VUSE on =
that
return anyway?  to keep it from being hoisted above the asm?

Whats the correct thing to do here?   Im surprised that somewhere in the st=
mt
deletion process, it doesnt rewrite all uses of MEM_2 to MEM_1 before delit=
ing
it.   Or is that a manual step?=