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 7ABDF3858C52; Fri,  3 Feb 2023 22:45:33 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 7ABDF3858C52
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1675464333;
	bh=6jLZlpXY1gr0FfAF04N88//fWHuvZzdn2GzXEjO5PlQ=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=XKSNJaQ9JXKT8DA6kUkc2Ry5LzvvKP5E+h+t8FCwM7jvzOggFWAkOxdAUUL1EPVjf
	 J7DuPQcrMfjQBN7zD8TjmKIo/Gkzyxtc8k8XCg0l6O52M9++fH4LIe3mj+AWCbJKXY
	 0NF5HqlClQHdUzRwELB+OcUggfnjGyZ6DuqyBCLc=
From: "wilco at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/108659] Suboptimal 128 bit atomics codegen on AArch64
 and x64
Date: Fri, 03 Feb 2023 22:45:33 +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: 12.2.1
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: wilco at gcc dot gnu.org
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-108659-4-TUKq9sXg4G@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-108659-4@http.gcc.gnu.org/bugzilla/>
References: <bug-108659-4@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=3D108659

--- Comment #11 from Wilco <wilco at gcc dot gnu.org> ---
(In reply to Niall Douglas from comment #10)
> (In reply to Jakub Jelinek from comment #9)
> > (In reply to Wilco from comment #8)
> > > Yes that sounds like a reasonable approach.
> >=20
> > I don't think so.  Not all variables on which __atomic_* intrinsics are=
 used
> > are actually _Atomic, the vars can be embedded in const aggregates etc.
>=20
> I'd have the attribute propagate to enclosing types, like over-alignment.

Yes, a structure with a 128-bit Atomic type in a subfield/union would be fo=
rced
to rwdata.

And arbitrary casts (eg. from char* to an atomic type) wouldn't work due to
Atomics requiring strict alignment. A 128-bit atomic type might have a high=
er
alignment than a 128-bit integer so even casting that seems questionable.=