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 3D5563857C5A; Thu,  5 Oct 2023 07:48:11 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 3D5563857C5A
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1696492091;
	bh=2/zojFQZXMWGnqgs6V1GVv+q71fK294DZYm7/C6Dy6E=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=cxErGWal+HTh3eJLXScD4EyR18uo152PytRKyY6n6f58rzdZATgGHiytehhV/LWMY
	 HR8Lo95hSMYIkwPmzJ7TwnAmtaz5i5RfBKx9skYVjfVWotlMbl95DeqZ93LkIn7NNe
	 NwvwH1FnP7h37alQjcVIcsL3CXFsZwaI9avW+B2k=
From: "rguenth at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/111697] Sub optimal code gen for initialising
 vector using loop
Date: Thu, 05 Oct 2023 07:48:10 +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: 14.0
X-Bugzilla-Keywords: missed-optimization
X-Bugzilla-Severity: enhancement
X-Bugzilla-Who: rguenth at gcc dot gnu.org
X-Bugzilla-Status: NEW
X-Bugzilla-Resolution: 
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-111697-4-AmR7h8B1Vf@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-111697-4@http.gcc.gnu.org/bugzilla/>
References: <bug-111697-4@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=3D111697

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
(In reply to Richard Biener from comment #2)
> We have quite some code doing vector CTOR stuff in tree-ssa-forwprop.cc a=
nd
> this should be optimized to
>=20
>  v_2 =3D { x_6(D), x_6(D), x_6(D), x_6(D) };
>=20
> note SLP vectorization can do this but it fails because it doesn't handle
> a default def insert - it handles a group of BIT_INSERT_EXPRs as
> vector CTOR and SLP discovery doesn't know how to start from external defs
> (it needs actual definition stmts).
>=20
> A more general approach would be to try to track vector construction thro=
ugh
> symbolic execution like we form bswap in the bswap pass.

You could "steal" the code in vect_slp_check_for_roots,

      else if (code =3D=3D BIT_INSERT_EXPR
               && VECTOR_TYPE_P (TREE_TYPE (rhs))
               && TYPE_VECTOR_SUBPARTS (TREE_TYPE (rhs)).is_constant ()
               && TYPE_VECTOR_SUBPARTS (TREE_TYPE (rhs)).to_constant () > 1
               && integer_zerop (gimple_assign_rhs3 (assign))
               && useless_type_conversion_p
                    (TREE_TYPE (TREE_TYPE (rhs)),
                     TREE_TYPE (gimple_assign_rhs2 (assign)))
               && bb_vinfo->lookup_def (gimple_assign_rhs2 (assign)))
        {
          /* We start to match on insert to lane zero but since the
             inserts need not be ordered we'd have to search both
             the def and the use chains.  */
...

and put it into tree-ssa-forwprop.cc, explicitly creating the vector CTOR.=