Received-SPF: pass (google.com: domain of linux-kernel+bounces-102571-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
Message-ID: <96325519f8f05d7289ec9b0af4e89da28c432e4f.camel@scylladb.com>
Subject: Re: [RFC PATCH 0/2] Introduce serialized smp_call_function APIs
From: Avi Kivity <avi@scylladb.com>
To: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>, Peter Oskolkov
	 <posk@google.com>
Cc: linux-kernel@vger.kernel.org, Peter Zijlstra <peterz@infradead.org>, 
 "Paul E . McKenney" <paulmck@linux.vnet.ibm.com>, Boqun Feng
 <boqun.feng@gmail.com>, Andrew Hunter <ahh@google.com>,  Maged Michael
 <maged.michael@gmail.com>, gromer@google.com, Benjamin Herrenschmidt
 <benh@kernel.crashing.org>,  Paul Mackerras <paulus@samba.org>, Michael
 Ellerman <mpe@ellerman.id.au>
Date: Wed, 13 Mar 2024 23:58:44 +0200
In-Reply-To: <20240313205622.2179659-1-mathieu.desnoyers@efficios.com>
References: <20240313205622.2179659-1-mathieu.desnoyers@efficios.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.48.4 (3.48.4-1.fc38) 
Precedence: bulk
MIME-Version: 1.0

[resend in plain text for the list]

On Wed, 2024-03-13 at 16:56 -0400, Mathieu Desnoyers wrote:
> commit 944d5fe50f3f ("sched/membarrier: reduce the ability to hammer
> on sys_membarrier")
> introduces a mutex over all membarrier operations to reduce its
> ability
> to slow down the rest of the system.
>=20
> This RFC series has two objectives:
>=20
> 1) Move this mutex to the smp_call_function APIs so other system
> calls
> =C2=A0=C2=A0 using smp_call_function IPIs are limited in the same way,
>=20
> 2) Restore scalability of MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ with
> =C2=A0=C2=A0 MEMBARRIER_CMD_FLAG_CPU, which targets specific CPUs with IP=
Is.
> =C2=A0=C2=A0 This may or may not be useful, and I would welcome benchmark=
s from
> =C2=A0=C2=A0 users of this feature to figure out if this is worth it.
>=20

I see this doesn't restore scaling of MEMBARRIER_CMD_PRIVATE_EXPEDITED,
which I use (and wasn't aware was broken).

I don't have comments on the patches, but do have ideas on how to work
around the problem in Seastar. So this was a useful heads-up for me.