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 E40163858D32; Wed, 28 Jun 2023 05:07:30 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org E40163858D32
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1687928850;
	bh=yf2PY0QKink3zgtZzBVRIEXimrVYmxOnGBPX5FdT6ZQ=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=CMr2P9ZqYhee6f3ShML0fV/QbXwPSNVV5F5PE9ynL+Uah9dvC34b6LhpkDGoFyv6K
	 Prmry2wLmHlY8C1p1FRgDJk17R7qTkbz7HFZZc0O4Ri8eCXEaoSIhXRBC3SDzSR0nC
	 q9+V8u4MnOcax4OzaWLJ1jhcYtXMfjfg66MCwrPg=
From: "amonakov at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug rtl-optimization/110202] _mm512_ternarylogic_epi64 generates
 unnecessary operations
Date: Wed, 28 Jun 2023 05:07:29 +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-Version: 13.1.0
X-Bugzilla-Keywords: missed-optimization
X-Bugzilla-Severity: enhancement
X-Bugzilla-Who: amonakov 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-110202-4-gv8BKUx4P5@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-110202-4@http.gcc.gnu.org/bugzilla/>
References: <bug-110202-4@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=3D110202

--- Comment #9 from Alexander Monakov <amonakov at gcc dot gnu.org> ---
(In reply to Hongtao.liu from comment #8)
>=20
> For this one, we can load *a into %zmm0 to avoid false_dependence.
>=20
> vmovdqau ZMMWORD PTR [rdi], zmm0
> vpternlogq      zmm0, zmm0, zmm0, 85

Yes, since ternlog with memory operand needs two fused-domain uops on Intel
CPUs, breaking out the load would be more efficient for both negate1 and
negate2.=