[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87a5mkm5wg.fsf@yhuang6-desk2.ccr.corp.intel.com>
Date: Wed, 27 Mar 2024 16:32:31 +0800
From: "Huang, Ying" <ying.huang@...el.com>
To: Ryan Roberts <ryan.roberts@....com>
Cc: Kairui Song <ryncsn@...il.com>, <linux-mm@...ck.org>, Chris Li
<chrisl@...nel.org>, Minchan Kim <minchan@...nel.org>, Barry Song
<v-songbaohua@...o.com>, Yu Zhao <yuzhao@...gle.com>, SeongJae Park
<sj@...nel.org>, David Hildenbrand <david@...hat.com>, Yosry Ahmed
<yosryahmed@...gle.com>, Johannes Weiner <hannes@...xchg.org>, Matthew
Wilcox <willy@...radead.org>, Nhat Pham <nphamcs@...il.com>, Chengming
Zhou <zhouchengming@...edance.com>, "Andrew Morton"
<akpm@...ux-foundation.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [RFC PATCH 00/10] mm/swap: always use swap cache for
synchronization
Ryan Roberts <ryan.roberts@....com> writes:
> [...]
>
>>>> Test 1, sequential swapin/out of 30G zero page on ZRAM:
>>>>
>>>> Before (us) After (us)
>>>> Swapout: 33619409 33886008
>>>> Swapin: 32393771 32465441 (- 0.2%)
>>>> Swapout (THP): 7817909 6899938 (+11.8%)
>>>> Swapin (THP) : 32452387 33193479 (- 2.2%)
>>>
>>> If my understanding were correct, we don't have swapin (THP) support,
>>> yet. Right?
>>
>> Yes, this series doesn't change how swapin/swapout works with THP in
>> general, but now THP swapout will leave shadows with large order, so
>> it needs to be splitted upon swapin, that will slow down later swapin
>> by a little bit but I think that's worth it.
>>
>> If we can do THP swapin in the future, this split on swapin can be
>> saved to make the performance even better.
>
> I'm confused by this (clearly my understanding of how this works is incorrect).
> Perhaps you can help me understand:
>
> When you talk about "shadows" I assume you are referring to the swap cache? It
> was my understanding that swapping out a THP would always leave the large folio
> in the swap cache, so this is nothing new?
>
> And on swap-in, if the target page is in the swap cache, even if part of a large
> folio, why does it need to be split? I assumed the single page would just be
> mapped? (and if all the other pages subsequently fault, then you end up with a
> fully mapped large folio back in the process)?
>
> Perhaps I'm misunderstanding what "shadows" are?
Perhaps, shadow is used to support workingset protection/detection on
the anonymous LRU list as in the following patchset (merged).
https://lore.kernel.org/all/1595490560-15117-5-git-send-email-iamjoonsoo.kim@lge.com/T/#m962395eb5968c74b0c4c8e41d4b0dcdd3f28b2e6
--
Best Regards,
Huang, Ying
Powered by blists - more mailing lists