Received-SPF: pass (google.com: domain of linux-kernel+bounces-29586-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Message-ID: <adaefdf7-4a9f-4055-be0b-ad30bf1683b8@intel.com>
Date: Wed, 17 Jan 2024 16:26:01 -0800
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH] x86/resctrl: Implement new MBA_mbps throttling heuristic
Content-Language: en-US
To: Tony Luck <tony.luck@intel.com>, Fenghua Yu <fenghua.yu@intel.com>, "Peter
 Newman" <peternewman@google.com>, <x86@kernel.org>
CC: James Morse <james.morse@arm.com>, Jamie Iles <quic_jiles@quicinc.com>,
	Babu Moger <babu.moger@amd.com>, Xiaochen Shen <xiaochen.shen@intel.com>,
	<linux-kernel@vger.kernel.org>, <patches@lists.linux.dev>
References: <20231201214737.104444-1-tony.luck@intel.com>
 <20240109220003.29225-1-tony.luck@intel.com>
From: Reinette Chatre <reinette.chatre@intel.com>
In-Reply-To: <20240109220003.29225-1-tony.luck@intel.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit
Precedence: bulk
MIME-Version: 1.0
X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1
X-MS-Exchange-AntiSpam-MessageData-0: =?utf-8?B?V1JqdTRIRzBMQjZRRVc0ZkJaK3prbWJRSmlzNHRxL25aRkhTTGxuMWtOdVhp?=
 =?utf-8?B?S2pqdlpmbzlTVktCczErOXJrRXJCaHV6ZjZySnBVbno1WlMzR1FYZzEvKzZK?=
 =?utf-8?B?eitkNlhRWmh1amkwUktNM2tpem95RXBqK1VxMm8rTkZvSC9ORGp5Q3p1M0Va?=
 =?utf-8?B?NGxCeVVTc0pUUlFxOFNVdzU3Q1ZsSWJuNTVsVnZGb0xRalZNekVDaU9CNDUz?=
 =?utf-8?B?eTcrUCtJcTd3QjBXQUhJVE96Zng5bmxiQUhEVytWdy81QjJnaGRtR2pVeDdQ?=
 =?utf-8?B?VE5DaE5vQ0tmeHIydXBGRlUrNnlKWko3TWlTd1N2TFJMeU5heHVQQ0JWRHpM?=
 =?utf-8?B?Q2ZuSjQzYzlCQVJJQnI0VEZLUEZXNFRmQURnUEgxWkNnK1A5aTlwK2dkTloy?=
 =?utf-8?B?c0hyMGt5aEh6Q1AvMDc3Y05pdFd2VXZXS3VpaTlaVkFvTXBOeHUrMnFtVmNx?=
 =?utf-8?B?ZkVtWVl6V3Irc25pV0lzTmtzRnRxZE5BdXlEdEJ4dHlUOTlHbmZDTW9rOXdM?=
 =?utf-8?B?dVJuZ2xTb1RoUFBWeW1nZnFDT1IyUm5oVlRicGJ0dWI3Wk1mMkwveldKdFRS?=
 =?utf-8?B?QmtmQzRwcUtyZEhlaXdPUlJoREttdTNMenU4Vm1XWEFEci9HVHNzOG9CQk1q?=
 =?utf-8?B?QlRPL2tHczlPeXBIMnQ4cDdpWUlpRzNwSkloUjM2SGs0M1NtT3lacWdLcVVG?=
 =?utf-8?B?dkNpdndSb1grMWhsSnZVUFFwbjFmTDg4WXViTGZhWWdJemdoZTU2Nld0RS8x?=
 =?utf-8?B?Z0E2cnJiSVMzYTFXcUZOeVkxZytBR2JEUG03aG5pSGlZMDJKVzUxUUgvUGJv?=
 =?utf-8?B?UUh1WjlZMEFXQmxnN0RvM2VYalpEb2Q2YWwyeVBzN3hJdE9rQm5JbFIrTjJa?=
 =?utf-8?B?aGxGNHBHMDQzajJ4SzhIVnJPYTBYOVU2ZzFoQzE1dHRza0VaVWFvRC9pVFZL?=
 =?utf-8?B?OXBmSXEweTVPbEZPNGxZK0xaNkNkNkhmNzlmM2g5eW9TQjJsN3VjNzVZeUR4?=
 =?utf-8?B?TjZ6cTEyTnJSbnRFbjk0S3pDdmhnbkQrYi83ZFh5dFFBMG9mbzFLcTRsczUr?=
 =?utf-8?B?TnB2bXZwTjJ0Q28reEJub2Npcm93b20zc2hQdEpBbE42b01IRFJjb3FwWjdl?=
 =?utf-8?B?TWN1U0VUQzM0MGFsSzFVdEpuWkI0aEt5QzR3UzkrZllLUDNtc05OdjVmNGpU?=
 =?utf-8?B?VWRFakFtQkVnbWZuUitEMHlRTjdGNDVhT1NnMWlFRGdldFc3S1ZzL0o0cGFm?=
 =?utf-8?B?MjMzRnBWbFlhQTYwWG8rbTY0a1RRbmxDa2VoMTJWSmtRRmhUcFc5MUN2ak54?=
 =?utf-8?B?WldsUzI4WUtNQUNBYWRWQVZTNUFIQ0tSdksvLzFZWDJ2OGZ5L2VkU29xVjdu?=
 =?utf-8?B?NkM4ZnhZWm5zUmVucXBFMkJvcjdxWTI1cHU3TldyZmRpR3AxZHB6amZ5amVR?=
 =?utf-8?B?dm5WclloeTVVMGZpVXl0SGJRSEduSXlyWlZyTitScndsVUFvQnFkZElhb3BE?=
 =?utf-8?B?UUFTMng2b3dVaEkwdUVqZmQvR2dkUExrc0E3ZGVTcmx0eGhhMElTK3FrMm9i?=
 =?utf-8?B?bG92ZWQ3b1F1bU9XTVArcm5MTmhIeTBXVUZwbmIwMFVESjQ1Z2tIbmhqelF5?=
 =?utf-8?B?UWh5YVBSNWpGRzh3OHlxVGVacEt6LzBnaXlTMVJHNGx3S2U3TGZkc01WUVdL?=
 =?utf-8?B?cGMzZVh4eXhPdnNUVld0QVF1c3Y0T3dSN2hITjRyNHQwSGVtRTlVQ01sa2pR?=
 =?utf-8?B?R3hxQVlack5laklWb3EyU2diaTRVYjBKWUdocTk0TkQ3ZmhENFhTeXdKa1Jw?=
 =?utf-8?B?aWxqMXQ5UmhjSjd6NFFkM0RSQTd3MlZsQ1RVdWxJTURrbk1YL2Z4T1VMMFVF?=
 =?utf-8?B?VVNHTzlSa3B4RWRZTWtqVUVkNlEwaDJ0dFRwb1NtS0wyR2R4clJTYkkyRjh3?=
 =?utf-8?B?WFY3dmovSXpMZ2pOdTl6NFZNSW81emhIUmZLYjlmTExOeFVOWkp0b1E4alVL?=
 =?utf-8?B?dlg2em9XS2MzMi9yQkU0VkRLVENOSTBrL0RmaEwwMjg1djBDaU43b1BhSVUy?=
 =?utf-8?B?cWVIaXhzUnQyQUpady91eGtyamxwQWdXK05xMVFZeExxaDVQcGk1bVdLRDJN?=
 =?utf-8?B?cGNmQmVFSVBQUFhNWlBvREFFVXFYNHF4RGVBME1GVzFvZmt2UER4ZnF0b0N3?=
 =?utf-8?B?dXc9PQ==?=
X-MS-Exchange-CrossTenant-Network-Message-Id: 8bf07aca-7886-4c3a-fa58-08dc17bc0e5a
X-MS-Exchange-CrossTenant-AuthSource: SJ2PR11MB7573.namprd11.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Internal
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 18 Jan 2024 00:26:04.3290
 (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: i8EuIoKN0pkZ+KsT8mZyiJeVGkT+yId7+lgj/fHPzLZYJilqcc45yCTB3985TT4axMIkLM+y9cLNSu7uF5WLF7Un8IDh6GCWcVdqanK1TDo=
X-MS-Exchange-Transport-CrossTenantHeadersStamped: CO1PR11MB4852
X-OriginatorOrg: intel.com

Hi Tony,

Please note in the subject and changelog that the mount option
is mba_MBps. The subject and first sentence of changelog seems
to fuse the mount option with the software controller enabled by it,
but I do not think doing so dilutes the message.

On 1/9/2024 2:00 PM, Tony Luck wrote:
> The MBA_mbps feedback loop increases throttling when a group is using
> more bandwidth than the target set by the user in the schemata file, and
> decreases throttling when below target.
> 
> To avoid possibly stepping throttling up and down on every poll a
> flag "delta_comp" is set whenever throttling is changed to indicate
> that the actual change in bandwidth should be recorded on the next
> poll in "delta_bw". Throttling is only reduced if the current bandwidth
> plus delta_bw is below the user target.
> 
> This algorithm works well if the workload has steady bandwidth needs.
> But it can go badly wrong if the workload moves to a different phase
> just as the throttling level changed. E.g. if the workload becomes
> essentially idle right as throttling level is increased, the value
> calculated for delta_bw will be more or less the old bandwidth level.
> If the workload then resumes, Linux may never reduce throttling because
> current bandwidth plu delta_bw is above the target set by the user.

"plu" -> "plus"

> 
> Implement a simpler heuristic by assuming that in the worst case the
> currently measured bandwidth is being controlled by the current level of
> throttling. Compute how much it may increase if throttling is relaxed to
> the next higher level. If that is still below the user target, then it
> is ok to reduce the amount of throttling.
> 
> Fixes: ba0f26d8529c ("x86/intel_rdt/mba_sc: Prepare for feedback loop")
> Reported-by: Xiaochen Shen <xiaochen.shen@intel.com>
> Signed-off-by: Tony Luck <tony.luck@intel.com>
> ---

Thank you very much for catching and fixing this.

With the issues ("MBA_mbps" -> "mba_MBps", "plu" -> "plus") addressed:

Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>

Reinette