回复: Re: [PATCH] migrate_pages: modify max number of pages to migrate in batch

回复: Re: [PATCH] migrate_pages: modify max number of pages to migrate in batch

[李真能]

[-- Attachment #1: Type: text/html, Size: 3302 bytes --]

[-- Attachment #2: migrate_pages-modify-max-number-of-pages-to-migrate-in-batch.patch --]
[-- Type: application/octet-stream, Size: 1364 bytes --]

From 25c3d2df1d38b49f389234e074978d05946a0a63 Mon Sep 17 00:00:00 2001
From: Zhenneng Li <lizhenneng@kylinos.cn>
Date: Mon, 24 Jun 2024 12:19:08 +0800
Subject: [PATCH] migrate_pages: modify max number of pages to migrate in batch

We restrict the number of pages to be migrated to no more than
HPAGE_PMD_NR or NR_MAX_BATCHED_MIGRATION, but in fact, the
number of pages to be migrated may reach 2*HPAGE_PMD_NR-1 or 2
*NR_MAX_BATCHED_MIGRATION-1, it's not in inconsistent with the context.

Please refer to the patch: 42012e0436d4(migrate_pages: restrict number
of pages to migrate in batch)

Signed-off-by: Zhenneng Li <lizhenneng@kylinos.cn>
---
 mm/migrate.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/mm/migrate.c b/mm/migrate.c
index 781979567f64..cce8e2b85e89 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -1957,11 +1957,12 @@ int migrate_pages(struct list_head *from, new_folio_t get_new_folio,
 		}
 
 		nr_pages += folio_nr_pages(folio);
-		if (nr_pages >= NR_MAX_BATCHED_MIGRATION)
+		if ((nr_pages >= NR_MAX_BATCHED_MIGRATION) &&
+		    (!list_first_entry(from, struct folio, lru)))
 			break;
 	}
 	if (nr_pages >= NR_MAX_BATCHED_MIGRATION)
-		list_cut_before(&folios, from, &folio2->lru);
+		list_cut_before(&folios, from, &folio->lru);
 	else
 		list_splice_init(from, &folios);
 	if (mode == MIGRATE_ASYNC)
-- 
2.25.1

Re: 回复: Re: [PATCH] migrate_pages: modify max number of pages to migrate in batch

[Huang, Ying]

李真能 <lizhenneng@kylinos.cn> writes:

> ----
>
>  
>
>  
>
> 主　题：Re: [PATCH] migrate_pages: modify max number of pages to migrate in batch 
> 日　期：2024-06-25 09:17 
> 发件人：ying.huang 
> 收件人：李真能;
>
> Hi, Zhenneng,
>
> Zhenneng Li writes:
>
>> We restrict the number of pages to be migrated to no more than
>> HPAGE_PMD_NR or NR_MAX_BATCHED_MIGRATION, but in fact, the
>> number of pages to be migrated may reach 2*HPAGE_PMD_NR-1 or 2
>> *NR_MAX_BATCHED_MIGRATION-1, it's not in inconsistent with the context.
>
> Yes. It's not HPAGE_PMD_NR exactly.
>
>> Please refer to the patch: 42012e0436d4(migrate_pages: restrict number
>> of pages to migrate in batch)
>>
>> Signed-off-by: Zhenneng Li 
>> ---
>> mm/migrate.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/mm/migrate.c b/mm/migrate.c
>> index 781979567f64..7a4b37aac9e8 100644
>> --- a/mm/migrate.c
>> +++ b/mm/migrate.c
>> @@ -1961,7 +1961,7 @@ int migrate_pages(struct list_head *from, new_folio_t get_new_folio,
>> break;
>> }
>> if (nr_pages >= NR_MAX_BATCHED_MIGRATION)
>> - list_cut_before(&folios, from, &folio2->lru);
>> + list_cut_before(&folios, from, &folio->lru);
>
> If the first entry of the list "from" is a THP with size HPAGE_PMD_NR,
> "folio" will be the first entry of from, so that "folios" will be empty.
> Right?
>  
> Yes, It's right, so we can check whether it is the first entry of the list "from", new patch are as follows(attachment is
> patch file):
> diff --git a/mm/migrate.c b/mm/migrate.c
> index 781979567f64..cce8e2b85e89 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -1957,11 +1957,12 @@ int migrate_pages(struct list_head *from, new_folio_t get_new_folio,
> }
>
> nr_pages += folio_nr_pages(folio);
> - if (nr_pages >= NR_MAX_BATCHED_MIGRATION)
> + if ((nr_pages >= NR_MAX_BATCHED_MIGRATION) &&
> + (!list_first_entry(from, struct folio, lru)))
> break;
> }
> if (nr_pages >= NR_MAX_BATCHED_MIGRATION)
> - list_cut_before(&folios, from, &folio2->lru);
> + list_cut_before(&folios, from, &folio->lru);
> else
> list_splice_init(from, &folios);
> if (mode == MIGRATE_ASYNC)
>

I don't think that it's necessary to change this.  Yes, the max batch
number of the page migration may be 2*HPAGE_PMD_NR-1.  Does it cause any
latency issue for you?

>
>> else
>> list_splice_init(from, &folios);
>> if (mode == MIGRATE_ASYNC)

--
Best Regards,
Huang, Ying