Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:18 as permitted sender) client-ip=2620:137:e000::1:18;
From:   "Zhang, Qiang1" <qiang1.zhang@intel.com>
To:     Uladzislau Rezki <urezki@gmail.com>
CC:     "paulmck@kernel.org" <paulmck@kernel.org>,
        Neeraj Upadhyay <quic_neeraju@quicinc.com>,
        "frederic@kernel.org" <frederic@kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "rcu@vger.kernel.org" <rcu@vger.kernel.org>,
        "bigeasy@linutronix.de" <bigeasy@linutronix.de>
Subject: RE: [PATCH] rcu: Only boost rcu reader tasks with lower priority than
 boost kthreads
Thread-Topic: [PATCH] rcu: Only boost rcu reader tasks with lower priority
 than boost kthreads
Thread-Index: AQHYL6nedKu+SJDFEUmZSHHikE8cTayvGo8AgAQNpZCAASfOAIABfomAgACVE5A=
Date:   Wed, 9 Mar 2022 03:04:19 +0000
Message-ID: <PH0PR11MB5880569E024586505CD18630DA0A9@PH0PR11MB5880.namprd11.prod.outlook.com>
References: <20220304092631.4123188-1-qiang1.zhang@intel.com>
 <81f69dd4-6ca9-760c-bec5-5cb27afbe788@quicinc.com>
 <PH0PR11MB5880026EBBE3E195549E2245DA089@PH0PR11MB5880.namprd11.prod.outlook.com>
 <20220307191512.GN4285@paulmck-ThinkPad-P17-Gen-1>
 <YieapfzKvy2chaHT@pc638.lan>
In-Reply-To: <YieapfzKvy2chaHT@pc638.lan>
Accept-Language: en-US
Content-Language: en-US
dlp-product: dlpe-windows
dlp-version: 11.6.401.20
dlp-reaction: no-action
x-ms-exchange-antispam-messagedata-chunkcount: 1
x-ms-exchange-antispam-messagedata-0: =?us-ascii?Q?t+8RWRe/InP4H93ety2NLktwJ4N5Rz9NNC4B+ZwW57CVgjhyajidPxk/3M/2?=
 =?us-ascii?Q?p6zfwewmU3G2uZX8Uc50if2Kou9YCdgsWHYkxrayXMXKQ9aZ0OrJ3YxsSjNv?=
 =?us-ascii?Q?GAPOFedmgEQHDooULcKLw3pZt+Knt6hh5674zhzLK80eMc9iRs4kSOBECjJl?=
 =?us-ascii?Q?RCYdJEomUAEa+5QnggOeu8MqW+eLeTjdYlHKxefLtxebxJ3NWb6GtNuZc3K6?=
 =?us-ascii?Q?PNAVZPOHhb5Kb5Ub0vja9CIyh2k4/YnTHhyHti+/9KtUwKUkYrkK7O3fcU9e?=
 =?us-ascii?Q?HW972TyXtFGTcQc0lTuDLBAPI+X+9aRZcL/B67/BvlDHziOZbGh7NEk4q8Sq?=
 =?us-ascii?Q?+EbFlFCfVqh6bSoMDqVBkRN/2lx8beLaytS5Srl9hYq3BSgi2JnugWIEKe+j?=
 =?us-ascii?Q?m1hPyDUzQrS7qCACbPAWbno+D9Do4h9NPwwMXB+Gig2T6MnIrRVOZ3AliQqc?=
 =?us-ascii?Q?a78GZ/wfsTevpQARJkj5jzliX/2Ux4WMbkf1EWWanJo1RxGnz3U8iLHmcTN1?=
 =?us-ascii?Q?TpcL3MKIN1nDK8CY4XYj5lPPs1QPU4oeupmRCuzjk3uKJCX68VTRVotJCTBr?=
 =?us-ascii?Q?MyuQRf09vSoyXMpTVstcozpAC5XV7pd26zKp5RdtXak9zpXPkLa+/69quAeO?=
 =?us-ascii?Q?GpfPx9jGQClGOXoKnl6ubf7P/FUifp62cA/0kLHWj39LoDOrS1HM9uFHSYy0?=
 =?us-ascii?Q?8PhGgQf1Bw5MRTwme+/9aNRz5RreZAvsCCOvgmFMAu+f7G4TGWKipLnwnhL/?=
 =?us-ascii?Q?gliQy7nl7Myxbm/81Grr5Y8iVA2chnHCLWPADljVtUnIeIq1fQD9UPF0COnJ?=
 =?us-ascii?Q?Lq7G9ryYqkxvmgVL6qDmS/bUVnNQ99IwZLiwWA/Hum4Q9j71ZwujOdJhZhGO?=
 =?us-ascii?Q?Y16DhZe3SsCjEIbgdT97EL/9fpi0RqPPAAmvipp9UOyhQW3i3Kfzzazx4P4z?=
 =?us-ascii?Q?vQSN9EV1pcTXUEaP7BkFWrMuv2svp30dNxJCn+GpOKpnOeUzBOAo6Dc9jBLv?=
 =?us-ascii?Q?BfWarqF/WFdRCPKFI+4wKBIyKLa4BQl5orjxts72CW3roh98DU4a/2xPwBc4?=
 =?us-ascii?Q?EpbC5eUhNW2iOhBOm2CrS7e++HjM8e3z4nHBeJoeoWXKmGFNk3rCHLWF51FD?=
 =?us-ascii?Q?BJuUkNdh+AtNgRTTKnPMocfHYEiR5SPlPDLJqAkYf2zFqJ8qC97t03s19IWV?=
 =?us-ascii?Q?outgO24tPHZ85mO0TLGumpIrVEpboy3J7boqbm5ysbV9RKH0XJjXw3eDvEHP?=
 =?us-ascii?Q?axuNisa8s8GU5J+eDir6jEqNoCAFVxmhNEtiilSNA27Kub5h3tX+o0STZsjm?=
 =?us-ascii?Q?gkFDiHSBI7BjkERujx30scXIOqLZ3sb0o59rKPaLyBmT0xYmgmahh3S5JV+e?=
 =?us-ascii?Q?KrvaPjWTCyHbC0K6QaH6NUBBtfzE70hXpLdza2yZa6yhkYt9R6ytdBU3osay?=
 =?us-ascii?Q?2+JYNTbOPUl3iPmOn4FAm+iDXf2gCQHcvO7k1In4X7t6z1yn28l8viX2Kb6V?=
 =?us-ascii?Q?nuQWsNaGyqGfdZOUMcL3iXmyjA8Wua3gxYUKpjJzKGZD2bPUNQPjFdbdlBnb?=
 =?us-ascii?Q?bADVwHaa+mLyY2DyjUHhaNX88cOLc0K9K0WmsuHH0TF8VhsbFRTK7KOy8EXj?=
 =?us-ascii?Q?Ow=3D=3D?=
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: PH0PR11MB5880.namprd11.prod.outlook.com
X-MS-Exchange-CrossTenant-Network-Message-Id: a72e7d2f-9ce0-4535-fe7d-08da01798117
X-MS-Exchange-CrossTenant-originalarrivaltime: 09 Mar 2022 03:04:19.4056
 (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: sMb9Ko4FoRHeH7KrfQ5bWWJtWitnGgyBNP+1GZdz3vs1MKkMnSPK+YhU8mDSiLZsrIOH1MdXPvzhPY/HfYGF0A==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: CH2PR11MB4341
X-OriginatorOrg: intel.com
X-Spam-Status: No, score=-2.4 required=5.0 tests=BAYES_00,DKIMWL_WL_HIGH,
	DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,
	MAILING_LIST_MULTI,RDNS_NONE,SPF_HELO_NONE,T_SCC_BODY_TEXT_LINE
	autolearn=unavailable autolearn_force=no version=3.4.6
X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on
	lindbergh.monkeyblade.net
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org


> On Mon, Mar 07, 2022 at 02:03:17AM +0000, Zhang, Qiang1 wrote:
> > On 3/4/2022 2:56 PM, Zqiang wrote:
> > > When RCU_BOOST is enabled, the boost kthreads will boosting=20
> > > readers who are blocking a given grace period, if the current=20
> > > reader tasks have a higher priority than boost kthreads(the boost=20
> > > kthreads priority not always 1, if the kthread_prio is set),=20
> > > boosting is useless, skip current task and select next task to=20
> > > boosting, reduce the time for a given grace period.
> > >=20
> > > Signed-off-by: Zqiang <qiang1.zhang@intel.com>
>=20
> Adding to CC to get more eyes on this.  I am not necessarily opposed=20
> to it, but I don't do that much RT work myself these days.
>=20
> 							Thanx, Paul
>=20
> > > ---
> > >   kernel/rcu/tree_plugin.h | 10 +++++++++-
> > >   1 file changed, 9 insertions(+), 1 deletion(-)
> > >=20
> > > diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h=20
> > > index c3d212bc5338..d35b6da66bbd 100644
> > > --- a/kernel/rcu/tree_plugin.h
> > > +++ b/kernel/rcu/tree_plugin.h
> > > @@ -12,6 +12,7 @@
> > >    */
> > >  =20
> > >   #include "../locking/rtmutex_common.h"
> > > +#include <linux/sched/deadline.h>
> > >  =20
> > >   static bool rcu_rdp_is_offloaded(struct rcu_data *rdp)
> > >   {
> > > @@ -1065,13 +1066,20 @@ static int rcu_boost(struct rcu_node *rnp)
> > >   	 * section.
> > >   	 */
> > >   	t =3D container_of(tb, struct task_struct, rcu_node_entry);
> > > +	if (!rnp->exp_tasks && (dl_task(t) || t->prio <=3D current->prio)) =
{
> > > +		tb =3D rcu_next_node_entry(t, rnp);
> > > +		WRITE_ONCE(rnp->boost_tasks, tb);
> > > +		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
> > > +		goto end;
> > > +	}
> > > +
>>>>Why do you bypass the expedited grace period and boost any tasks anyway=
?
>>>>Same way the expedited gp can be blocked by higher prior tasks SCHED_DE=
ADLINE or SCHED_FIFO.

Thanks advice, it really ignores expedited grace period, may be
should generate 'boost_exp_tasks' pointer to point 'rnp->exp_tasks', do the=
 same thing
as normal grace period.

Thanks
Zqiang

>>>>
>>>>Thanks!
>>>>
>>>>--
>>>>Vlad Rezki