Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   "Zhang, Qiang1" <qiang1.zhang@intel.com>
To:     "paulmck@kernel.org" <paulmck@kernel.org>
CC:     "frederic@kernel.org" <frederic@kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: RE: [PATCH] rcu: When rcuog kthreads is in polling mode, wakeup
 waitqueue is not requried
Thread-Topic: [PATCH] rcu: When rcuog kthreads is in polling mode, wakeup
 waitqueue is not requried
Thread-Index: AQHYE/UODHHjhGvlP0WE4iSSid3cl6x5dzEAgAAFxeA=
Date:   Sat, 29 Jan 2022 05:55:34 +0000
Message-ID: <PH0PR11MB5880073B7080265117EE2868DA239@PH0PR11MB5880.namprd11.prod.outlook.com>
References: <20220128031346.157961-1-qiang1.zhang@intel.com>
 <20220129051748.GP4285@paulmck-ThinkPad-P17-Gen-1>
In-Reply-To: <20220129051748.GP4285@paulmck-ThinkPad-P17-Gen-1>
Accept-Language: en-US
Content-Language: en-US
dlp-product: dlpe-windows
dlp-version: 11.6.200.16
dlp-reaction: no-action
x-ms-exchange-antispam-messagedata-chunkcount: 1
x-ms-exchange-antispam-messagedata-0: =?us-ascii?Q?NOJbvCurBrSS3kwv7sGmtuQiNI+aGEy+Aq/j+oPylCf2XQKkFJwUbbkfPzd6?=
 =?us-ascii?Q?eEabM28+9WduEUGmM9RUx4Zn0grTelm04SOgOTGxvO/LfFi7mF9gVr8h/KS0?=
 =?us-ascii?Q?Ci9O6cxBK9RmVigEYy6DX/gHbHA1KZCmmyAODeqITv6kgVr5W17vkRYXse/J?=
 =?us-ascii?Q?vKqXEYnzCLgXlgbIGHBt4EYVCxGwxotZXIx4nkBh0eERtwSlu+BOG4I10DLg?=
 =?us-ascii?Q?E/ENYOlMyLWg2L5Wdf6jhT4hLLK82SCxlNuuKTx2g6+D8h4CzQPiVNUwWBQm?=
 =?us-ascii?Q?CtNv8EiFHzOz5j/lzOVey/W7RxTTOyXi3jUUJBex6i3IXlwaKW2kv78gVhzI?=
 =?us-ascii?Q?NRgxSMABCfFCB1qgLxbi596VL6ymOPPb+sOfrMneGRr1u5MCOWV6t4UKtdLa?=
 =?us-ascii?Q?JvsU38bkRexBhVGs3m+C4TLvO3s2uEEAdEPt78nga02cESF0nbkkZUXWvlvD?=
 =?us-ascii?Q?gIsjVDZrMpub+XI6inHXMRGmIsUQDlHZiD7agzbSTDiKI3yaiihiwRUqyjL0?=
 =?us-ascii?Q?DaM98jzsFUlLLs6aqlsHPYfn0VZ0mX3t0WMybEprdRGWBCqfuYd+4IDDej2n?=
 =?us-ascii?Q?7VSfyxtS9voLYO1VDEf6BbgQEuPaITJdzEe3Jb+Fvp3pXqtO+wcsxCU3bYEL?=
 =?us-ascii?Q?Eu+d2BD7HgbVaRCxd5F2K0lreZ6/wF15Utg5q5RXv8uZqXkwxQUEVPyc56RM?=
 =?us-ascii?Q?ibdws7vD7rraSSO89CSU1JOPXKOMH3l0eqgJAGZUnCiB2nZEFAeJPax8yQM5?=
 =?us-ascii?Q?sm7awJ/YMEOHDjL3XbP0TQQG6R+hHilKLn4m2LB5/G6fVKgEP56di67RCM4K?=
 =?us-ascii?Q?NAITcV0fedrl03DktdDgOS0BVnVKpViyBi+B/3JjEPkJ4R7O1a59iIW3RVwV?=
 =?us-ascii?Q?9KGzr7QC69hAfgDV2rnucTsrqBlcJYG+Dle6k1PyFRCnL7hVwZ+9BmolraFI?=
 =?us-ascii?Q?U+CuFYHFp4/vgOMJ10cB/nnYQuB5+EU7QxRpdl6ZdWG8TwUw/ZoATXD3511q?=
 =?us-ascii?Q?cqS2J6z5J7quBwjk9ZDzyvdZleFsy9cws+p4RSZdrZXGT1FXAzwgrolOJHPw?=
 =?us-ascii?Q?G7ec7HVqFDtzfFypHSmpdxs9Nw/K6RMQr3hyQqXldBK29ZgEhIK65QpNICIl?=
 =?us-ascii?Q?umE9LtcO52TG5jXrxu2oxWkD3d5ZMm5hHrP2pEqwB6iebZOtTvLakrm7FwTg?=
 =?us-ascii?Q?mewYH7133zE6r97zjIA+reZDwG9bqTXsA8KMGC9srkpugY2SK/jD2LGaJ9Rd?=
 =?us-ascii?Q?bRRwDvUUbXDrr4TKigYPE7P+0xUG5BWDRsRzQgGVtFF4AvlaKXDUj2BxVuo3?=
 =?us-ascii?Q?lUw4mwpuG+NCFrX1ntgQDVxzhsn5vXW56EmUCKaDMxh9jgz8joCqIfDkJUKm?=
 =?us-ascii?Q?PktkNaDOQJt2SKPfD0qu35nRVgdDVp8vTojo+mPm5QEK7BM2idVRvYB6ugVk?=
 =?us-ascii?Q?2Mrh/szPFQi50JMX6ysVs4qld3Mk4opOnK0x2P0iR0FD4hMj7AbBGGsCMjqZ?=
 =?us-ascii?Q?5QJ4t/+JCMOft1iN5MZfNcIlomSq0Xe3Skll97XRjs4Cb+j6drQIXIRG8JOK?=
 =?us-ascii?Q?PZefGWWtuNVkqUM1BlVMMlVcAuqEFdMCmBlBxQT0QtjOfMoZ7C4iTKHmKI1t?=
 =?us-ascii?Q?jw=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: a6e707a0-47d6-4485-5f6a-08d9e2ebf742
X-MS-Exchange-CrossTenant-originalarrivaltime: 29 Jan 2022 05:55:34.2056
 (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: 88X/5J7VqmEpBR0d1LQXYHTLCZo/ovbsIGKaXz3i6giU/GxNhkhPbrTqHnlz8z9yy1sUvZOxwXZRiWr/roXt2g==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: PH0PR11MB5208
X-OriginatorOrg: intel.com
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org


On Fri, Jan 28, 2022 at 11:13:46AM +0800, Zqiang wrote:
> When grace period cleanup, the rcuog kthreads that waiting in sq=20
> waitqueue will be awakened, however if the 'rcu_nocb_poll' is set, the=20
> sq waitqueue always empty, so if 'rcu_nocb_poll' is set, return=20
> directly.

>This does decrease grace-period-cleanup overhead in kernels built with CON=
FIG_RCU_NOCB_CPU=3Dy and booted with the rcu_nocb_poll kernel boot paramete=
r set.  On the other hand, it increases grace-period-cleanup overhead in ke=
rnels built with CONFIG_RCU_NOCB_CPU=3Dy but booted without the rcu_nocb_po=
ll kernel boot parameter set.
>
>Last I checked, more kernels were booted without the rcu_nocb_poll kernel =
boot parameter set.  If this is still the case, this patch would slow thing=
s down for most systems.
>
>Or are there now lots of systems booted with rcu_nocb_poll?

Hi Paul
I found that some of our customers configured rcu_nocb_poll startup paramet=
ers under Preempt-RT kernel.
at each grace period cleanup, we'll all wakeup sq waitqueue, however when r=
cuog kthreads is in polling mode,=20
the wakeup operation doesn't required, because the sq waitqueue always empt=
y.

Thanks,
Zqiang

>
>							Thanx, Paul

> Signed-off-by: Zqiang <qiang1.zhang@intel.com>
> ---
>  kernel/rcu/tree_nocb.h | 2 ++
>  1 file changed, 2 insertions(+)
>=20
> diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h index=20
> 636d0546a4e9..9e106c590e56 100644
> --- a/kernel/rcu/tree_nocb.h
> +++ b/kernel/rcu/tree_nocb.h
> @@ -201,6 +201,8 @@ static void rcu_lockdep_assert_cblist_protected(struc=
t rcu_data *rdp)
>   */
>  static void rcu_nocb_gp_cleanup(struct swait_queue_head *sq)  {
> +	if (rcu_nocb_poll)
> +		return;
>  	swake_up_all(sq);
>  }
> =20
> --
> 2.25.1
>=20