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 E12FD3857837; Fri, 28 Oct 2022 06:55:09 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org E12FD3857837
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1666940109;
	bh=FH/aXufIxJerdioCLQC0gMN9gO9DpBAz7WTovheYhHk=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=LA46ClJv+LJCSC2+wAwz9FwienWC/9FOe9g9yQ3JgZPoDMbApMH2EVV6tt0PQ0lkD
	 vGDodYPwXNJ0ih1lM4F7v/C/AAtIabA6f5HVtgW/1w66esSQTaHwIv8QB1NcNZiOTd
	 oQ3hfAUnAZyjEumfklhityZzgDMgaSyUntq4FdkY=
From: "crazylht at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/107432] __builtin_convertvector generates inefficient
 code
Date: Fri, 28 Oct 2022 06:55:09 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: target
X-Bugzilla-Version: unknown
X-Bugzilla-Keywords: missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: crazylht at gmail dot com
X-Bugzilla-Status: UNCONFIRMED
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-107432-4-xkOimulSEI@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-107432-4@http.gcc.gnu.org/bugzilla/>
References: <bug-107432-4@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=3D107432

--- Comment #7 from Hongtao.liu <crazylht at gmail dot com> ---
(In reply to Hongtao.liu from comment #6)
> > Guess expand_vector_conversion can be optimized.
>=20
>   if (INTEGRAL_TYPE_P (TREE_TYPE (ret_type))
>       && SCALAR_FLOAT_TYPE_P (TREE_TYPE (arg_type)))
>     code =3D FIX_TRUNC_EXPR;
>   else if (INTEGRAL_TYPE_P (TREE_TYPE (arg_type))
> 	   && SCALAR_FLOAT_TYPE_P (TREE_TYPE (ret_type)))
>     code =3D FLOAT_EXPR;
>=20
> It only supports floatmn2/fix_truncmn2 for float <-> integer.
>=20
> But we can also supports extendmn2/zero_extendmn2/truncmn2 for float <->
> float, integer <-> integer.
>=20
> Or are there any concerns and VEC_PACK_TRUNC_EXPR,
> VEC_PACK_FIX_TRUNC_EXPR,VEC_PACK_FLOAT_EXPR are used on purpose?

May be we can add some gimple simplication in match.pd to hanlde=20
  _4 =3D VEC_PACK_TRUNC_EXPR <a_1(D), { 0, 0, 0, 0 }>;
  _5 =3D BIT_FIELD_REF <_4, 128, 0>;

and

  _4 =3D [vec_unpack_lo_expr] a_1(D);
  _5 =3D [vec_unpack_hi_expr] a_1(D);
  _2 =3D {_4, _5};

Since loop vectorizer may also create vec_unpack_lo_expr/vec_unpack_hi_expr=
.=