Received-SPF: pass (google.com: domain of linux-kernel+bounces-143212-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
Message-ID: <58042847-7545-4621-abda-02b225cfa3ec@oracle.com>
Date: Fri, 12 Apr 2024 11:32:56 -0700
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH] slub: limit number of slabs to scan in count_partial()
To: "Christoph Lameter (Ampere)" <cl@linux.com>
Cc: Vlastimil Babka <vbabka@suse.cz>, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org, penberg@kernel.org, rientjes@google.com,
        iamjoonsoo.kim@lge.com, akpm@linux-foundation.org,
        junxiao.bi@oracle.com
References: <20240411164023.99368-1-jianfeng.w.wang@oracle.com>
 <e0222219-eb2d-5e1e-81e1-548eeb5f73e0@linux.com>
 <38ef26aa-169b-48ad-81ad-8378e7a38f25@suse.cz>
 <a8e208fb-7842-4bca-9d2d-3aae21da030c@oracle.com>
 <bb2cb21b-0a5e-0154-2a7d-7e630b50aa4a@linux.com>
Content-Language: en-US
From: Jianfeng Wang <jianfeng.w.wang@oracle.com>
In-Reply-To: <bb2cb21b-0a5e-0154-2a7d-7e630b50aa4a@linux.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Precedence: bulk
MIME-Version: 1.0
X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1
X-MS-Exchange-AntiSpam-MessageData-0: 
	=?utf-8?B?YXdabEluMWpTZnJEdzJRMkV0Tm9TTEFIU0dwdmVQdi96VkhZVkZqMlI3VkF5?=
 =?utf-8?B?c1N4WGEyRE82TENvVlBtbUxIUHY1ODFadlFqZnlaWmZvK1kvcURyZ2ZnU3dt?=
 =?utf-8?B?QkJiMVlDWmVOdFQ0VEtJYmEvZFcxY29EdVVCTnF0SzExYS9neWY1WHN3NDJG?=
 =?utf-8?B?cjdMZ0E1dFp5Vk1rQitOandFS2hTNmNYaTFvK3RtT00ycStuSmZuZGJuazRa?=
 =?utf-8?B?UHRRWC9DZWJMZEluMytuTWNlMjdLZXJkeXlWbGhrNWU2ZGEwWWdkTDZ1Z3Yz?=
 =?utf-8?B?cmExb2NSUUJweVJKM0x1M0FjMStWSFpMTjR2ZVFseWJoeXJqZ3BtcUJVR3cw?=
 =?utf-8?B?L28yTTZZYURHQmtzOUVqMnJ3bGNMRWxNMTlnVEFKaDkrNFdFOHpvYTNDbSti?=
 =?utf-8?B?by9aRmYrdzdlOG5xK1M5em03cDZHNXJLVk1xNHFEM3BWcFZyRkFXZkREYThQ?=
 =?utf-8?B?ZFEyVmwzbnZYM1k0cmpPUThqbU5rVGNYTkJ4YWgwdml1SFJQMllsa1lmV3ps?=
 =?utf-8?B?NWZmNXA4T3Mxd3YvMCtHKzRRa2lXWExML3pYMHZjSHQvZlJPZDR6WjVRMUk1?=
 =?utf-8?B?WHlyMDNKVEdNSmlkckJxeFlGK21aMElQVlhtdjRxWmZodWxhQVFNRzJWRjRO?=
 =?utf-8?B?ellsVTYwN0d4ZkRBRXd1R2NJS05rZWZlTkNmTkN2M3hyKzhsMU1Ia1o1UjdT?=
 =?utf-8?B?Q1lJanoxQmJNSnZtNHh0bnVNQnBUY3RnWlJjdWVOT01mS0hlQllXRjM1aFFN?=
 =?utf-8?B?ZnJXbXUyOS9SSWVKQzU0ZTd3cXRmT2RNb29LTXp0M3kzaDlDKzJEcUZmYUkr?=
 =?utf-8?B?Y2NBOTZGeStQMGovakVIc0V0eXM0M1Y0RG1McEg0OEEvSWRueWM0anlGT1Bh?=
 =?utf-8?B?Q3haOVFpNmtGcng0djdENjJvSnhva1BZNHpXbHZGMjY0NFBsV1dGbHF4RWZG?=
 =?utf-8?B?QldtNGIvMTNHVU5OTE1KSEE2dUtUaGUxN0tLbnp4U21seDQ4eEtvVnl4aW9m?=
 =?utf-8?B?emFGMmRTbEtIa1RNak1qcmNHNWhDeHJkY1hUQzB5UmllcXZIOEgydGluZFJl?=
 =?utf-8?B?eXhtWFBaRnBKS3dRMm1Rei91dC9XSjFvdnh0Z1l3L1R1UzhaMXFESnFRTzFM?=
 =?utf-8?B?cDBXa3k1cGJiNjZwSVZGdWh3L213aXdJSEc3bWw4Wnk3ZEdQQWpMRktUaWdF?=
 =?utf-8?B?VnZPb0ZEayswU3JvTVJsbTRmTUQ1b0R4U0p5dlVaQTROdkZGM2twYlhtM20x?=
 =?utf-8?B?N3JHRkRyeFJLeDF1MmZpSjVwa3dBY2JNWVVLL3dUTCsya1ZjY1FIVnhCaXV1?=
 =?utf-8?B?S3pFUzZwaUlDNCs1aXRIM1NzMDQ0cDc1bXNLK3NNeWhOVlZ1amdmM0pLQWt4?=
 =?utf-8?B?MG8yMitqQVZ6Z1NzbUgxMU00cnhLbGJ6Z0lPSHl1M0IyejJhTDFDbTFEZktB?=
 =?utf-8?B?TEVKYXNoaGdvajJETmlhWUk3VUxlYUY3R0lIMWZCZytXKy8wLzlVc2JHWkdo?=
 =?utf-8?B?OWpLNzhoQkVTWjFIdVFuSVE5R21ITVJmRTdFL0xPTFlSb255WEFzTGU5WmVZ?=
 =?utf-8?B?NUdNS3gwUUtGZDRZbG5xZUtwak8wUThxTGdnOGFDTVRQT0lzcUViWjFOYitD?=
 =?utf-8?B?SzBzeE11bnNLa1cxcUQvR3JMM2lVMlZ1SHVsU0E3ekszYjFvQU9YMlhhUjAw?=
 =?utf-8?B?dnFqcERkYWovQUY4TStvM3JscVlZUWtGSWJHYUlhdGxQQldiS0l3Z1MvRWVJ?=
 =?utf-8?B?UUpPWFhyTk5aRDBFcjBndkp2UExFMit4cFJOUlB4eHFBNVJoWnVNWk90MlRs?=
 =?utf-8?B?dXRDRTBxUXp2a1NaMmU0YmJqTkFlaFVHYTRhWE8yRVA5cTB2QnFmekVVbnQz?=
 =?utf-8?B?UjZFUUpOOENVM244RFROc0YyMFBhWVRvMkhhWWVEOXJIWWsveTRjV2srV0hB?=
 =?utf-8?B?ZFE4Q0h5ZlBpZW9PTUNKS29Zb09lbDdIL0w2S0VUNm5BUUJNc1dMU25xMDRv?=
 =?utf-8?B?b0FnbjlJTHcvZFhudWp1eGpOUURaNy9sSitDcy84M2l6RURldGV6S2ZTU0tw?=
 =?utf-8?B?b0k4ejlhaEpOS3JwSHJPQ3ZiU1NiRTRxa2V1SUJGMVR5bVREclVTRkw4bUwr?=
 =?utf-8?B?aTQ2TGtqZjNnZStqNndXcXZtZnNPbC9iNlQzV0Iwbk9KT24vdXZKTmo3cDBJ?=
 =?utf-8?B?RFE9PQ==?=
X-MS-Exchange-AntiSpam-ExternalHop-MessageData-ChunkCount: 1
X-MS-Exchange-AntiSpam-ExternalHop-MessageData-0: 
	V5XhYeYYSwIsyB0hEs5q8yHe9pVnkKqmOoozXNY0oT8YK9VUXfeMCNd/e6DMU9v4EkAiobrXSqU/yJjQ+io1a+RBBPxz5rnDLh+j4YNwXRHSrPLfkhwjMq8t5rJKRqzMVqN5AfPqTGAewSNnjSUUqOl5lHDdembZh6cTKK1RJlbgt3sfgkejv596Ee4+aQB06drMKlpQwyfQOrKn2++V81r/HSJdhh5U+80g9CfutI+9rF02IBkjuvxbII+k8+YTIwZczm7G00VDV+C7WnBZ/P96QSvbKCYEtzcj/gPXyADMpgJjGuh7pWM77QVXSgg1CqG8B2K5JggMGYFfGhyOx/8sGi8HkTiTPMX8F7Oz3R4SzuvL1RQ7p0tH2zDyhVXmdqWB6If9s5rYOWFUjc1a2IGhCkwXL7x7qua+Qvn+vhFU2xo2t5sGAojFVh+OOLdnbLzWRPS3fT9o6YGsp6Pmgmgdl94UlLOZPGFa0SCVN94l4Nqi2vH2t4FSFNZeVj5mRxryfsC8F9xipPWxGQ+d6ys9wdwzuALC6J/9mOJP1HsVC1lqsHEztckzpC1gu8W7ISnCVJOhVnFj7UcQE5NZEUwho2sZnlkxxkDGv5m2qqU=
X-OriginatorOrg: oracle.com
X-MS-Exchange-CrossTenant-Network-Message-Id: 270c9ae0-8257-4b9e-9689-08dc5b1f051d
X-MS-Exchange-CrossTenant-AuthSource: PH7PR10MB6379.namprd10.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Internal
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 12 Apr 2024 18:33:16.9905
 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-CrossTenant-Id: 4e2c6054-71cb-48f1-bd6c-3a9705aca71b
X-MS-Exchange-CrossTenant-MailboxType: HOSTED
X-MS-Exchange-CrossTenant-UserPrincipalName: Sf21k06/Af9vWQltQVhASchaM/fLESe1nhkMPNfK+Agf5cKxOnuX3ZBLJF3wfj3cpmbfnRJampk0FEQaEt39Me3aiYo1djQsZ1b+wEiov6U=
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BLAPR10MB5202
X-Proofpoint-Virus-Version: vendor=baseguard
 engine=ICAP:2.0.272,Aquarius:18.0.1011,Hydra:6.0.619,FMLib:17.11.176.26
 definitions=2024-04-12_14,2024-04-09_01,2023-05-22_02
X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 mlxlogscore=904 malwarescore=0
 suspectscore=0 spamscore=0 mlxscore=0 bulkscore=0 adultscore=0
 phishscore=0 classifier=spam adjust=0 reason=mlx scancount=1
 engine=8.12.0-2404010000 definitions=main-2404120137
X-Proofpoint-ORIG-GUID: wF_X5pQgMPPHXSqxEatfWs9sJNOFJ83f
X-Proofpoint-GUID: wF_X5pQgMPPHXSqxEatfWs9sJNOFJ83f


On 4/12/24 11:16 AM, Christoph Lameter (Ampere) wrote:
> On Fri, 12 Apr 2024, Jianfeng Wang wrote:
> 
>>>> Can you run some tests showing the difference between the estimation and
>>>> the real count?
>>
>> Yes.
>> On a server with one NUMA node, I create a case that uses many dentry objects.
>> For "dentry", the length of partial slabs is slightly above 250000. Then, I
>> compare my approach of scanning N slabs from the list's head v.s. the original
>> approach of scanning the full list. I do it by getting both results using
>> the new and the original count_partial() and printing them in /proc/slabinfo.
>>
>> N = 10000
>> my_result = 4741651
>> org_result = 4744966
>> diff = (org_result - my_result) / org_result = 0.00069 = 0.069 %
>>
>> Increasing N further to 25000 will only slight improve the accuracy:
>> N = 15000 -> diff =  0.02 %
>> N = 20000 -> diff =  0.01 %
>> N = 25000 -> diff = -0.017 %
>>
>> Based on the measurement, I think the difference between the estimation and
>> the real count is very limited (i.e. less than 0.1% for N = 10000). The
>> benefit is significant: shorter execution time for get_slabinfo(); no more
>> soft lockups or crashes caused by count_partial().
> 
> Wow. That is good. Maybe decrease N to 1000 instead?
> 

Yes, the diff is still limited. Here are some numbers:
N = 5000 -> diff = 0.0019 = 0.19 %
N = 3000 -> diff = 0.0023 = 0.23 %
N = 1000 -> diff = 0.0040 = 0.40 %

So, the estimation is quite accurate.