Received-SPF: pass (google.com: domain of linux-kernel+bounces-28264-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
Date: Tue, 16 Jan 2024 21:44:01 +0100
From: Pavel Machek <pavel@denx.de>
To: Sasha Levin <sashal@kernel.org>
Cc: linux-kernel@vger.kernel.org, stable@vger.kernel.org,
	"Borislav Petkov (AMD)" <bp@alien8.de>, tglx@linutronix.de,
	mingo@redhat.com, dave.hansen@linux.intel.com, x86@kernel.org,
	puwen@hygon.cn, seanjc@google.com, kim.phillips@amd.com,
	reinette.chatre@intel.com, babu.moger@amd.com, jmattson@google.com,
	peterz@infradead.org, ashok.raj@intel.com,
	rick.p.edgecombe@intel.com, brgerst@gmail.com, mjguzik@gmail.com,
	jpoimboe@kernel.org, nik.borisov@suse.com, aik@amd.com,
	vegard.nossum@oracle.com, daniel.sneddon@linux.intel.com,
	acdunlap@google.com
Subject: Re: [PATCH AUTOSEL 5.10 09/10] x86/barrier: Do not serialize MSR
 accesses on AMD
Message-ID: <ZabqkZ2vXaicy3pZ@duo.ucw.cz>
References: <20240115232818.210010-1-sashal@kernel.org>
 <20240115232818.210010-9-sashal@kernel.org>
Precedence: bulk
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature"; boundary="4Ia2zJT+I07Stjth"
Content-Disposition: inline
In-Reply-To: <20240115232818.210010-9-sashal@kernel.org>


--4Ia2zJT+I07Stjth
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

Hi!

> From: "Borislav Petkov (AMD)" <bp@alien8.de>
>=20
> [ Upstream commit 04c3024560d3a14acd18d0a51a1d0a89d29b7eb5 ]
>=20
> AMD does not have the requirement for a synchronization barrier when
> acccessing a certain group of MSRs. Do not incur that unnecessary
> penalty there.
=2E..
> Performance captured using an unmodified ipi-bench using the 'mesh-ipi' o=
ption
> with and without weak_wrmsr_fence() on a Zen4 system also showed signific=
ant
> performance improvement without weak_wrmsr_fence(). The 'mesh-ipi' option=
 ignores
> CCX or CCD and just picks random vCPU.
>=20
>   Average throughput (10 iterations) with weak_wrmsr_fence(),
>         Cumulative throughput: 4933374 IPI/s
>=20
>   Average throughput (10 iterations) without weak_wrmsr_fence(),
>         Cumulative throughput: 6355156 IPI/s
>=20
> [1] https://github.com/bytedance/kvm-utils/tree/master/microbenchmark/ipi=
-bench

Speed improvement, not a bugfix. Please drop.

BR,
									Pavel
									--=20
DENX Software Engineering GmbH,        Managing Director: Erika Unter
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany

--4Ia2zJT+I07Stjth
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----

iF0EABECAB0WIQRPfPO7r0eAhk010v0w5/Bqldv68gUCZabqkQAKCRAw5/Bqldv6
8uONAJ0WBAl/1FnlM9UMYG/+rhKbGP0K0gCgm6w4SGSIymFTPMqja4N0JqK9w7k=
=ojzs
-----END PGP SIGNATURE-----

--4Ia2zJT+I07Stjth--