Received-SPF: pass (google.com: domain of linux-kernel+bounces-106923-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Message-ID: <a2272aaf-60bf-4493-bee6-4f4efb53bbb2@intel.com>
Date: Tue, 19 Mar 2024 10:31:01 +0800
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH v4 5/6] mm: vmscan: Avoid split during shrink_folio_list()
Content-Language: en-US
To: David Hildenbrand <david@redhat.com>, "Huang, Ying"
	<ying.huang@intel.com>, Ryan Roberts <ryan.roberts@arm.com>
CC: "linux-mm@kvack.org" <linux-mm@kvack.org>, "linux-kernel@vger.kernel.org"
	<linux-kernel@vger.kernel.org>, Andrew Morton <akpm@linux-foundation.org>,
	Matthew Wilcox <willy@infradead.org>, Gao Xiang <xiang@kernel.org>, Yu Zhao
	<yuzhao@google.com>, Yang Shi <shy828301@gmail.com>, Michal Hocko
	<mhocko@suse.com>, Kefeng Wang <wangkefeng.wang@huawei.com>, Barry Song
	<21cnbao@gmail.com>, Chris Li <chrisl@kernel.org>
References: <20240311150058.1122862-1-ryan.roberts@arm.com>
 <20240311150058.1122862-6-ryan.roberts@arm.com>
 <d8086108-05e7-4765-a6f5-dfb85fb44b90@redhat.com>
 <1db242d3-5ff1-4ef5-b20a-578a317fa859@arm.com>
 <caafeabd-fb94-45de-9dcc-2451ddc0ef50@redhat.com>
 <5cc147fa-492e-46c3-963d-2f6e6680dce9@arm.com>
 <878r2gwahg.fsf@yhuang6-desk2.ccr.corp.intel.com>
 <a75ea34a-7512-4169-b987-95f11a7f3dd0@intel.com>
 <b084270d-6c7a-400c-8f4b-0797d2ae5fa6@redhat.com>
From: Yin Fengwei <fengwei.yin@intel.com>
In-Reply-To: <b084270d-6c7a-400c-8f4b-0797d2ae5fa6@redhat.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?OFBlbkQrd3Q5V1UxWU1sUGJlRlBEOG5HenA2QlNHVC9SRXp0TjlKOEdrSWd4?=
 =?utf-8?B?dkwzUkNiVHlBL01qenFaNUdsZVZwaVMyRC9tYjFvQXZjQ2hzMVBsOExQOHM1?=
 =?utf-8?B?YnNZd0VCYWp6ZWtHVVp5ZXdsL0NnNUlxL3VOc0ozSkU1U1QrWm1oM29vQTNy?=
 =?utf-8?B?MmhPeTFyL0lIakNLMDkwOVhTRjRhLzNwMXVXRTRDcmo3akhnWGpWL3ZWQ1cz?=
 =?utf-8?B?L1lPMENOVm5rUUVyZUZNanZsSVIrd2JqeHJidnFEdDYxY2RJMHpMTlBPNWpt?=
 =?utf-8?B?UUllc3RKeTlIR3lKMG5FU0V5MDg0V3QraTJtWGhSdkxLTW9rLzNoTmFGcTdo?=
 =?utf-8?B?cUZ6ei95L1pMbW1aVGxxaTM0aVJpL2JaT2ZTb2k4QnFpTmc1Z3Y3UjNMdnV3?=
 =?utf-8?B?OVJWLzEwdFU2R2F5NGtxSksvQmxRcGJFTEE4ZlNPOWpVUXVmem1FYUlmQzZS?=
 =?utf-8?B?Z1VvTWE2U1pQUGhzejVBNmtJUmg3M3htN09wWG5KUkQzTE1oWnNtdlhYdkZ1?=
 =?utf-8?B?eXAyZ1NCclRFNFZOWG9lYzVWNjVqT0tCQUlZK29YU2habGlZUi83YkhLem9s?=
 =?utf-8?B?Mm41OVNyc0cvQUlpSmFjekpaanlndHFkd2VpcWpJQXM4VytZU2tKM3Rpbm10?=
 =?utf-8?B?UXo5VTBLT1BWWGZGeEVMWjUyVmJEMU11ZFFkVlJ6Mkp0dU1EcUZYNklBTU04?=
 =?utf-8?B?VkhYRHlGYjhZOG9oNldqaWFRM1A5QnFZUWd1cWszbUF5cStFdEx6TTlaNUds?=
 =?utf-8?B?R0xVbnl2TVBSbVFNR0tIMUJaVSsvanBEeHNKYklFVDVHTVIySHlIcjNVSVFH?=
 =?utf-8?B?LzdMMHFvWllLL1B0akhlL2ROV1o5QVNVVlgrWjNodktROFlIQ1A2MmdNRTI3?=
 =?utf-8?B?S3F1M0laZDhXYy8yR0xKV0tPK3h3TjhVOGZua21UMGcwNnA3ZE1iZGoxOFpT?=
 =?utf-8?B?MUJSMEFNMFgxaDl0akkzQXNBeng4WW16a3ZsQW5vVXVBVDVVaFc2U3dvUGZO?=
 =?utf-8?B?MzZjQXMwR3g2cE1kaE4yNWp3Y0MwVXR4YmZ4dWp2bG5ROWNOS1BreWNHdFB4?=
 =?utf-8?B?MUNkTUVnMThCSE52TDF4R1RHQXNMMzdDT2hqOXJ4bDNPcDl5RVVZbUY3VUJW?=
 =?utf-8?B?NENYQyt3SXBVdEZjOG1FWjc5UzlHckUwalFId3RvKzNkWWtTY1lDT2Y4ZGNs?=
 =?utf-8?B?TUNWTkFqK0FsS2Nvb0Q3RUV6VW5mNDZ2K0o3dXViVDQ1SXlIQ0t1ZTFVYlRC?=
 =?utf-8?B?OGpDRzl6VjQyclc1WVAzbEVCZnlYdUV1b05OOUFlME5sdzlPQy9xenRnOTY5?=
 =?utf-8?B?aU1DYzhEYlVuRFF4eVJIUHZDYTBSc1FQZ0x1RFdHbVBuZ1lsaFBJVzk0a0kx?=
 =?utf-8?B?eFdXZlN3MTJkR1BKdytoVFBlRHR0TTkxSnJOWXJNTjhlUEhkc2lJZW9GbnJR?=
 =?utf-8?B?TTRTeDdKTTA5dmNKR1RKNzF2ZGhpaHdXUHUyQk9LNFJKclE4ckl0NTlGWUJh?=
 =?utf-8?B?YlBPVTI3RUNYK1l3eWt6a1RKTTUwWnAzcXRuWW9WeSsxUVF3RkxtU0xWZm5v?=
 =?utf-8?B?K01UNUJNVjNwZWV5V0VWTkZoMXZhWnU0UUxkTFlzbDBIclVKeG9IMUR2Z2Iz?=
 =?utf-8?B?SElpWnl3QVIxbWl4bGdNcUtnb0VYV1NNd09BRzhpYUN6NlNEQ3piQ25tb01J?=
 =?utf-8?B?cloySTNYaEROL3RsVGhtYUVUMjlKOHJ0akQwOWgzTjVUOVZNeHk2ZnpYMVEx?=
 =?utf-8?B?bElNVnlVSVVOZVg4alkrU1FkM2Z4V05lbmRaZmUzaXBUTUROemFselNEQU9Y?=
 =?utf-8?B?Qm9EbFBkcHM0d3Z1QWJhM25JRHpMclBWRGJMV05ER2tFN3lITG1PMHh1VDhQ?=
 =?utf-8?B?Y3dtaHRsZ3V6MUpRYmt0ODRpcGJvUmZwZ2hVaTZrdEVQeDhPaUs3QXdTRnh0?=
 =?utf-8?B?MmZnQ3hzd3hOSWJhdEtnNXloaFJxRXN0OXMzbm8yVlRCNzVMeG4vTkpMSmhJ?=
 =?utf-8?B?R3Q3Vm0zaWlnMWJyOW1RUFUvbmsyZVFwdGs5MjhEVGk0WGM4K1owNmMwayto?=
 =?utf-8?B?a0hFbUZyWFlONE84T3F3WDhpQityVklBZmVPemYrbmF4Q3VodXcxUkRaa2ox?=
 =?utf-8?Q?mbdNkan4+EDW0q0EzD4I5eR2S?=
X-MS-Exchange-CrossTenant-Network-Message-Id: 2385e667-c8b7-4281-fbbf-08dc47bc9a1c
X-MS-Exchange-CrossTenant-AuthSource: CO1PR11MB4820.namprd11.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Internal
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 19 Mar 2024 02:30:54.7272
 (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: UbPhszZRgH92VYFfe4uc1n1Tq4YOlF9zkMQ44yaXsTQky6GmlSf323GvVvOT2C2LihWd6XMMM4S9f4P7Zs/VPQ==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: SJ2PR11MB8472
X-OriginatorOrg: intel.com


On 3/18/24 18:05, David Hildenbrand wrote:
> On 18.03.24 11:00, Yin, Fengwei wrote:
>>
>>
>> On 3/18/2024 10:16 AM, Huang, Ying wrote:
>>> Ryan Roberts <ryan.roberts@arm.com> writes:
>>>
>>>> Hi Yin Fengwei,
>>>>
>>>> On 15/03/2024 11:12, David Hildenbrand wrote:
>>>>> On 15.03.24 11:49, Ryan Roberts wrote:
>>>>>> On 15/03/2024 10:43, David Hildenbrand wrote:
>>>>>>> On 11.03.24 16:00, Ryan Roberts wrote:
>>>>>>>> Now that swap supports storing all mTHP sizes, avoid splitting large
>>>>>>>> folios before swap-out. This benefits performance of the swap-out path
>>>>>>>> by eliding split_folio_to_list(), which is expensive, and also sets us
>>>>>>>> up for swapping in large folios in a future series.
>>>>>>>>
>>>>>>>> If the folio is partially mapped, we continue to split it since we want
>>>>>>>> to avoid the extra IO overhead and storage of writing out pages
>>>>>>>> uneccessarily.
>>>>>>>>
>>>>>>>> Signed-off-by: Ryan Roberts <ryan.roberts@arm.com>
>>>>>>>> ---
>>>>>>>>      mm/vmscan.c | 9 +++++----
>>>>>>>>      1 file changed, 5 insertions(+), 4 deletions(-)
>>>>>>>>
>>>>>>>> diff --git a/mm/vmscan.c b/mm/vmscan.c
>>>>>>>> index cf7d4cf47f1a..0ebec99e04c6 100644
>>>>>>>> --- a/mm/vmscan.c
>>>>>>>> +++ b/mm/vmscan.c
>>>>>>>> @@ -1222,11 +1222,12 @@ static unsigned int shrink_folio_list(struct list_head
>>>>>>>> *folio_list,
>>>>>>>>                          if (!can_split_folio(folio, NULL))
>>>>>>>>                              goto activate_locked;
>>>>>>>>                          /*
>>>>>>>> -                     * Split folios without a PMD map right
>>>>>>>> -                     * away. Chances are some or all of the
>>>>>>>> -                     * tail pages can be freed without IO.
>>>>>>>> +                     * Split partially mapped folios map
>>>>>>>> +                     * right away. Chances are some or all
>>>>>>>> +                     * of the tail pages can be freed
>>>>>>>> +                     * without IO.
>>>>>>>>                           */
>>>>>>>> -                    if (!folio_entire_mapcount(folio) &&
>>>>>>>> +                    if (!list_empty(&folio->_deferred_list) &&
>>>>>>>>                              split_folio_to_list(folio,
>>>>>>>>                                      folio_list))
>>>>>>>>                              goto activate_locked;
>>>>>>>
>>>>>>> Not sure if we might have to annotate that with data_race().
>>>>>>
>>>>>> I asked that exact question to Matthew in another context bt didn't get a
>>>>>> response. There are examples of checking if the deferred list is empty with and
>>>>>> without data_race() in the code base. But list_empty() is implemented like this:
>>>>>>
>>>>>> static inline int list_empty(const struct list_head *head)
>>>>>> {
>>>>>>       return READ_ONCE(head->next) == head;
>>>>>> }
>>>>>>
>>>>>> So I assumed the READ_ONCE() makes everything safe without a lock? Perhaps not
>>>>>> sufficient for KCSAN?
>> I don't think READ_ONCE() can replace the lock.
>>
>>>>>
>>>>> Yeah, there is only one use of data_race with that list.
>>>>>
>>>>> It was added in f3ebdf042df4 ("THP: avoid lock when check whether THP is in
>>>>> deferred list").
>>>>>
>>>>> Looks like that was added right in v1 of that change [1], so my best guess is
>>>>> that it is not actually required.
>>>>>
>>>>> If not required, likely we should just cleanup the single user.
>>>>>
>>>>> [1]
>>>>> https://lore.kernel.org/linux-mm/20230417075643.3287513-2-fengwei.yin@intel.com/
>>>>
>>>> Do you have any recollection of why you added the data_race() markup?
>>>
>>> Per my understanding, this is used to mark that the code accesses
>>> folio->_deferred_list without lock intentionally, while
>>> folio->_deferred_list may be changed in parallel.  IIUC, this is what
>>> data_race() is used for.  Or, my understanding is wrong?
>> Yes. This is my understanding also.
> 
> Why don't we have a data_race() in deferred_split_folio() then, before taking the lock?
No idea why there is no data_race() added. But I think we should add data_race().

Regards
Yin, Fengwei

> 
> It's used a bit inconsistently here.
>