Received-SPF: pass (google.com: domain of linux-kernel+bounces-13791-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
From: David Laight <David.Laight@ACULAB.COM>
To: "'linux-kernel@vger.kernel.org'" <linux-kernel@vger.kernel.org>,
	"'peterz@infradead.org'" <peterz@infradead.org>, "'longman@redhat.com'"
	<longman@redhat.com>
CC: "'mingo@redhat.com'" <mingo@redhat.com>, "'will@kernel.org'"
	<will@kernel.org>, "'boqun.feng@gmail.com'" <boqun.feng@gmail.com>, "'Linus
 Torvalds'" <torvalds@linux-foundation.org>,
	"'virtualization@lists.linux-foundation.org'"
	<virtualization@lists.linux-foundation.org>, 'Zeng Heng'
	<zengheng4@huawei.com>
Subject: [PATCH next v2 2/5] locking/osq_lock: Optimise the
 vcpu_is_preempted() check.
Thread-Topic: [PATCH next v2 2/5] locking/osq_lock: Optimise the
 vcpu_is_preempted() check.
Thread-Index: Ado8M69TGuBFtEJaSr+qwMS7CJLhkw==
Date: Sun, 31 Dec 2023 21:52:51 +0000
Message-ID: <3a9d1782cd50436c99ced8c10175bae6@AcuMS.aculab.com>
References: <2b4e8a5816a742d2bd23fdbaa8498e80@AcuMS.aculab.com>
In-Reply-To: <2b4e8a5816a742d2bd23fdbaa8498e80@AcuMS.aculab.com>
Accept-Language: en-GB, en-US
Precedence: bulk
MIME-Version: 1.0
Content-Language: en-US
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

The vcpu_is_preempted() test stops osq_lock() spinning if a virtual
cpu is no longer running.

Although patched out for bare-metal the code still needs the cpu number.
Reading this from 'prev->cpu' is a pretty much guaranteed have a cache miss
when osq_unlock() is waking up the next cpu.

Instead save 'prev->cpu' in 'node->prev_cpu' and use that value instead.
Update in the osq_lock() 'unqueue' path when 'node->prev' is changed.

This is simpler than checking for 'node->prev' changing and caching
'prev->cpu'.

Signed-off-by: David Laight <david.laight@aculab.com>
---
 kernel/locking/osq_lock.c | 16 +++++++---------
 1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/kernel/locking/osq_lock.c b/kernel/locking/osq_lock.c
index e0bc74d85a76..eb8a6dfdb79d 100644
--- a/kernel/locking/osq_lock.c
+++ b/kernel/locking/osq_lock.c
@@ -14,8 +14,9 @@
=20
 struct optimistic_spin_node {
 =09struct optimistic_spin_node *next, *prev;
-=09int locked; /* 1 if lock acquired */
-=09int cpu; /* encoded CPU # + 1 value */
+=09int locked;    /* 1 if lock acquired */
+=09int cpu;       /* encoded CPU # + 1 value */
+=09int prev_cpu;  /* encoded CPU # + 1 value */
 };
=20
 static DEFINE_PER_CPU_SHARED_ALIGNED(struct optimistic_spin_node, osq_node=
);
@@ -29,11 +30,6 @@ static inline int encode_cpu(int cpu_nr)
 =09return cpu_nr + 1;
 }
=20
-static inline int node_cpu(struct optimistic_spin_node *node)
-{
-=09return node->cpu - 1;
-}
-
 static inline struct optimistic_spin_node *decode_cpu(int encoded_cpu_val)
 {
 =09int cpu_nr =3D encoded_cpu_val - 1;
@@ -110,9 +106,10 @@ bool osq_lock(struct optimistic_spin_queue *lock)
 =09if (old =3D=3D OSQ_UNLOCKED_VAL)
 =09=09return true;
=20
-=09node->locked =3D 0;
+=09node->prev_cpu =3D old;
 =09prev =3D decode_cpu(old);
 =09node->prev =3D prev;
+=09node->locked =3D 0;
=20
 =09/*
 =09 * osq_lock()=09=09=09unqueue
@@ -144,7 +141,7 @@ bool osq_lock(struct optimistic_spin_queue *lock)
 =09 * polling, be careful.
 =09 */
 =09if (smp_cond_load_relaxed(&node->locked, VAL || need_resched() ||
-=09=09=09=09  vcpu_is_preempted(node_cpu(node->prev))))
+=09=09=09=09  vcpu_is_preempted(READ_ONCE(node->prev_cpu) - 1)))
 =09=09return true;
=20
 =09/* unqueue */
@@ -201,6 +198,7 @@ bool osq_lock(struct optimistic_spin_queue *lock)
 =09 * it will wait in Step-A.
 =09 */
=20
+=09WRITE_ONCE(next->prev_cpu, prev->cpu);
 =09WRITE_ONCE(next->prev, prev);
 =09WRITE_ONCE(prev->next, next);
=20
--=20
2.17.1

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1=
PT, UK
Registration No: 1397386 (Wales)