From:   "Zhuo, Qiuxu" <qiuxu.zhuo@intel.com>
To:     "Uladzislau Rezki (Sony)" <urezki@gmail.com>,
        "Paul E . McKenney" <paulmck@kernel.org>
CC:     RCU <rcu@vger.kernel.org>,
        "quic_neeraju@quicinc.com" <quic_neeraju@quicinc.com>,
        Boqun Feng <boqun.feng@gmail.com>,
        Joel Fernandes <joel@joelfernandes.org>,
        LKML <linux-kernel@vger.kernel.org>,
        "Oleksiy Avramchenko" <oleksiy.avramchenko@sony.com>,
        Steven Rostedt <rostedt@goodmis.org>,
        Frederic Weisbecker <frederic@kernel.org>
Subject: RE: [PATCH 1/1] Reduce synchronize_rcu() waiting time
Thread-Topic: [PATCH 1/1] Reduce synchronize_rcu() waiting time
Thread-Index: AQHZW9/cgEn1/2PZAki4vZrvu1Ahi68FQZQw
Date:   Tue, 21 Mar 2023 14:03:19 +0000
Message-ID: <IA1PR11MB617169A10E6DDE3C1168605D89819@IA1PR11MB6171.namprd11.prod.outlook.com>
References: <20230321102748.127923-1-urezki@gmail.com>
In-Reply-To: <20230321102748.127923-1-urezki@gmail.com>
Accept-Language: en-US
Content-Language: en-US
x-ms-exchange-antispam-messagedata-chunkcount: 1
x-ms-exchange-antispam-messagedata-0: =?us-ascii?Q?MNCy1N/BAKjziHCSo5TxX3t/oh0p0y29J2sWMymE+5y59cLXs3QRvMmeaoIt?=
 =?us-ascii?Q?hXvZSk3fJcFbrA/RgRT9VT/hqw7OMos3S1S6QXgOedr9jGtAjhmW9RmFzOGe?=
 =?us-ascii?Q?Ftbdkla0KWgkV1UgDRA3YDo4OGL8AJRAEvizzkuVlm4yiNIwB6UoIsongybT?=
 =?us-ascii?Q?qPCDdYI41Ajv4N1maTMPrkeYQgGp4NYycglsNRt2jyC1rJdw6SoMAK2MaHxZ?=
 =?us-ascii?Q?WHaIX7S/WdrXKK8ThdhBzg8kPPI0RPV15U7zyca/BC7yJTaJDHaRI5NHh4Go?=
 =?us-ascii?Q?OjP4ITVkGkX3lpojQ3SGmCGWP24t7MEg/g4nPiNdikm4oXVKUY5CnWXpDM9t?=
 =?us-ascii?Q?/W4yQsSr5eVl9kizREpbh1/eS9W3ZdS0qcbuUaZpjTn6+92tZB7lkYIOOhwd?=
 =?us-ascii?Q?TFBYC8bY2uELWlELN7zgH8Rhy/t393pUUpC4kVtn/+VOjdN6VTltm9uGx6yM?=
 =?us-ascii?Q?DIXEL1xj3RzemQG4SfgzQK+hj14B1RNv+oG7uFde2bFTZkNcr0PabYaXMJdg?=
 =?us-ascii?Q?0aE6UWUtQ4fPmnEyByZOUUrsxxJzVONWcDldLYsNnR9VofGirmysuLZdNUpU?=
 =?us-ascii?Q?zR3mHKjowkTGyXU/ihfe10B/IM6pvWDkaxiuI0gAw8RMZtHvB9zFyPPKzrYJ?=
 =?us-ascii?Q?ePLclVGETcULxlaZ0SSkUG7KqII4udpYOpK0g+kMkyqNljzLQb8OfMcK7oIY?=
 =?us-ascii?Q?WjDT+W8XyEUfk3WWQqP8QDnP8cq9cAfTRhNvABqXOo6wr3PwpVJeBMOjni0J?=
 =?us-ascii?Q?aBc42Fcoa2N+t044bMfnT6tfd/vnstNkbET2nc4a3S0ilFHpJRNTeeF8Qviu?=
 =?us-ascii?Q?JqV8d9TxhBXcl6UvrOg8PIcjaNt7RNFLg4LqBGbw4U7bTfhllS5rDqWcn3/g?=
 =?us-ascii?Q?thFLBdXtcfVZ5wS/xo3kq3rl/cTACwFM4O3ilRNcUmxC0Y288ZtCKkd/aOMY?=
 =?us-ascii?Q?vf15WYOAHvASXALpnHuGRuTZsXVGamNiNYX8ESZcFTn8aFwmfhELycS5mBsQ?=
 =?us-ascii?Q?7LjmjMLrCHk6hsizMckRR8NZp5EjqxWfyjBfXwae6vBKrPXFaNDuF4wgDBUT?=
 =?us-ascii?Q?/08FVcWes9pVlwEc4qp101GAQVy1ol50q89+vXsD4S6ue6l9o+wQIRIiunrj?=
 =?us-ascii?Q?yBpQWr37DgVz/KIW1dw2+wx2Zm4GIAiZQzAbL/Nv/0l+NA18q6vEfZHQCXez?=
 =?us-ascii?Q?wVpf2QaFEjOO2Estp58GmPJVQmX/F7UP4Q3461oFKnvLaFoSWljMMuV5+QA1?=
 =?us-ascii?Q?R0OJHuYiK2TnZx5kEFQ5nfztQJyonRu05VA3P1J7ku1STZ0LUU/TQVokmTmh?=
 =?us-ascii?Q?YuVbKJkmcMuv6yHQvVU4NoTw9nI5JR76Y1ygaoDlD9dBfN/y6s0zhMWitlME?=
 =?us-ascii?Q?4gD8okmuBXphV6zeUl4aiFFqBLX9YXhT2vtXeiPRr1vbtDAGU9vNOafD8GRN?=
 =?us-ascii?Q?4+9NYCiXvHTJppoG8Sezhg3CSWDGotJEutiVO/OtgEhX3T2WQb6dgpIBLoxw?=
 =?us-ascii?Q?Dlyrl+aZr0wTqFadl3lY6IWfW+xOi4P5XO5b6nC0FlmnNyi6VvHwUCFHa9EF?=
 =?us-ascii?Q?fjol0JzpM4qAMJlqz9AyNnQsUIe4+qPEsaM+CwJJ?=
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-AuthAs: Internal
X-MS-Exchange-CrossTenant-AuthSource: IA1PR11MB6171.namprd11.prod.outlook.com
X-MS-Exchange-CrossTenant-Network-Message-Id: 67d7f77f-22b2-4a0a-9f44-08db2a150683
X-MS-Exchange-CrossTenant-originalarrivaltime: 21 Mar 2023 14:03:19.4870
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 46c98d88-e344-4ed4-8496-4ed7712e255d
X-MS-Exchange-CrossTenant-mailboxtype: HOSTED
X-MS-Exchange-CrossTenant-userprincipalname: EoIH8WadbuX8tuJEab1rcW5mBACnCKQ+8sGMxW7dH5mkk6jCX6fb8pNCeBs9+tIyPRO3Q1xoqV9jBTi9jN3H9Q==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: PH8PR11MB8040
X-OriginatorOrg: intel.com
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

> From: Uladzislau Rezki (Sony) <urezki@gmail.com>
> Sent: Tuesday, March 21, 2023 6:28 PM
> [...]
> Subject: [PATCH 1/1] Reduce synchronize_rcu() waiting time
>=20
> A call to a synchronize_rcu() can be expensive from time point of view.
> Different workloads can be affected by this especially the ones which use=
 this
> API in its time critical sections.
>=20

This is interesting and meaningful research. ;-)

> For example in case of NOCB scenario the wakeme_after_rcu() callback
> invocation depends on where in a nocb-list it is located. Below is an exa=
mple
> when it was the last out of ~3600 callbacks:
>=20
> <snip>
>   <...>-29      [001] d..1. 21950.145313: rcu_batch_start: rcu_preempt
> CBs=3D3613 bl=3D28
> ...
>   <...>-29      [001] ..... 21950.152578: rcu_invoke_callback: rcu_preemp=
t
> rhp=3D00000000b2d6dee8 func=3D__free_vm_area_struct.cfi_jt
>   <...>-29      [001] ..... 21950.152579: rcu_invoke_callback: rcu_preemp=
t
> rhp=3D00000000a446f607 func=3D__free_vm_area_struct.cfi_jt
>   <...>-29      [001] ..... 21950.152580: rcu_invoke_callback: rcu_preemp=
t
> rhp=3D00000000a5cab03b func=3D__free_vm_area_struct.cfi_jt
>   <...>-29      [001] ..... 21950.152581: rcu_invoke_callback: rcu_preemp=
t
> rhp=3D0000000013b7e5ee func=3D__free_vm_area_struct.cfi_jt
>   <...>-29      [001] ..... 21950.152582: rcu_invoke_callback: rcu_preemp=
t
> rhp=3D000000000a8ca6f9 func=3D__free_vm_area_struct.cfi_jt
>   <...>-29      [001] ..... 21950.152583: rcu_invoke_callback: rcu_preemp=
t
> rhp=3D000000008f162ca8 func=3Dwakeme_after_rcu.cfi_jt
>   <...>-29      [001] d..1. 21950.152625: rcu_batch_end: rcu_preempt CBs-
> invoked=3D3612 idle=3D....
> <snip>
>

Did the results above tell us that CBs-invoked=3D3612 during the time 21950=
.145313 ~ 21950.152625?

If possible, may I know the steps, commands, and related parameters to prod=
uce the results above?
Thank you!

- Qiuxu
=20
> As performance results, i will provide it once we are ready with a patch.
>=20
> Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
> ---
>  kernel/rcu/tree.c | 104
> +++++++++++++++++++++++++++++++++++++++++++++-
[...]