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 0F28F3858D33; Sun, 10 Dec 2023 09:54:15 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 0F28F3858D33
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1702202055;
	bh=ao8TTnrQYcADF/w//9UGyU+6e7BY/I5o9tCMsT8FjKo=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=yFaKLTJkH3s3zMFdDIfXaXsWVxT2zwtNs5LWdKF49c5ElQrUd23xhZXOcgYKy7Rnq
	 NT74aNLJiRb3T+qpS/b67iKT7fXXTFdebAhY6rw8cyuv9Y2fu+ZRcoQPzhu8+2kYaT
	 R0zVomrrFSJF+43ZEDrePWCGI/9crK2uqsPu0eM4=
From: "lh_mouse at 126 dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/80878] -mcx16 (enable 128 bit CAS) on x86_64 seems not
 to work on 7.1.0
Date: Sun, 10 Dec 2023 09:54:10 +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: 7.1.0
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: lh_mouse at 126 dot com
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-80878-4-9iYTZgLomP@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-80878-4@http.gcc.gnu.org/bugzilla/>
References: <bug-80878-4@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=3D80878

--- Comment #42 from LIU Hao <lh_mouse at 126 dot com> ---
(In reply to Yongwei Wu from comment #27)
> Anyone can show a valid use case for a non-lock-free version of 128-bit
> atomic_compare_exchange?
>=20
> I am trying to use it in a data structure intended to be lock-free. I am
> surprised to find that the C++ std::atomic::compare_exchange_weak does not
> result in lock-free code for a 128-bit struct intended for ABA-free CAS. =
As
> a result, the GCC-generated code is MUCH slower than the mutex-based vers=
ion
> in my 8-thread contention test, defeating all its valid purposes. I am
> talking about a 10x difference. And the Clang-generated code is more than
> 200x faster in the same test.

[I think this is off topic though.]

I tested CMPXCHG16B with inline assembly on an i7-1165G7 (Dell XPS 13 9305)=
 and
it turned out to be much slower than CMPXCHG, even slower than a pair of ca=
lls
to `pthread_mutex_lock()` and unlock. Similar results were observed on a
desktop i7 11700 and a server Xeon Cascadelake. The performance degeneration
might be caused by more =CE=BCops, more locking work for the extra width of
operands, and more cache synchronization, which makes some sense if we assu=
me
the CPU should be optimized mostly for 8-byte access.

The conclusion is probably that 16-byte compare-and-swap isn't recommended.=