From: Chengming Zhou <chengming.zhou@linux.dev>
To: Nhat Pham <nphamcs@gmail.com>
Cc: hannes@cmpxchg.org, yosryahmed@google.com,
akpm@linux-foundation.org, linux-mm@kvack.org,
linux-kernel@vger.kernel.org,
Chengming Zhou <zhouchengming@bytedance.com>
Subject: Re: [PATCH 1/2] mm/zswap: don't return LRU_SKIP if we have dropped lru lock
Date: Sat, 27 Jan 2024 22:37:07 +0800 [thread overview]
Message-ID: <22e52669-2cf5-4375-9cb8-82d2f5faadc6@linux.dev> (raw)
In-Reply-To: <CAKEwX=NzDY57n6ogx6=VaaEx8j_Jtvz5BeuMDAW-KuDXmsyQFw@mail.gmail.com>
On 2024/1/27 02:01, Nhat Pham wrote:
> On Fri, Jan 26, 2024 at 12:31 AM <chengming.zhou@linux.dev> wrote:
>>
>> From: Chengming Zhou <zhouchengming@bytedance.com>
>>
>> LRU_SKIP can only be returned if we don't ever dropped lru lock, or
>> we need to return LRU_RETRY to restart from the head of lru list.
>
> Ooops. You're right! I just double checked and only LRU_REMOVED_RETRY
> and LRU_RETRY indicate we might have dropped the lock. My bad.
>
>>
>> Actually we may need to introduce another LRU_STOP to really terminate
>> the ongoing shrinking scan process, when we encounter a warm page
>
> Yup. This is what I was trying (and failing) to do. To be honest, this
> needs to be even stronger: short-circuit ALL concurrent/ongoing zswap
> shrinker scan processes that are touching this memcg (as they will
> also shrink into warmer regions going forward). But that's a bit more
> engineering to do. LRU_STOP, which stops this scan process, would be a
> good place to start.
Good suggestion, will look into that more later.
>
>> already in the swap cache. The current list_lru implementation
>> doesn't have this function to early break from __list_lru_walk_one.
>>
>> Fixes: b5ba474f3f51 ("zswap: shrink zswap pool based on memory pressure")
>> Signed-off-by: Chengming Zhou <zhouchengming@bytedance.com>
>
> Reviewed-by: Nhat Pham <nphamcs@gmail.com>
Thanks.
>
>> ---
>> mm/zswap.c | 4 +---
>> 1 file changed, 1 insertion(+), 3 deletions(-)
>>
>> diff --git a/mm/zswap.c b/mm/zswap.c
>> index 00e90b9b5417..81cb3790e0dd 100644
>> --- a/mm/zswap.c
>> +++ b/mm/zswap.c
>> @@ -901,10 +901,8 @@ static enum lru_status shrink_memcg_cb(struct list_head *item, struct list_lru_o
>> * into the warmer region. We should terminate shrinking (if we're in the dynamic
>> * shrinker context).
>> */
>> - if (writeback_result == -EEXIST && encountered_page_in_swapcache) {
>> - ret = LRU_SKIP;
>> + if (writeback_result == -EEXIST && encountered_page_in_swapcache)
>> *encountered_page_in_swapcache = true;
>> - }
>>
>> goto put_unlock;
>> }
>> --
>> 2.40.1
>>
prev parent reply other threads:[~2024-01-27 14:37 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-01-26 8:30 chengming.zhou
2024-01-26 8:30 ` [PATCH 2/2] mm/zswap: fix race between lru writeback and swapoff chengming.zhou
2024-01-26 15:31 ` Johannes Weiner
2024-01-26 19:31 ` Nhat Pham
2024-01-27 14:53 ` Chengming Zhou
2024-01-26 19:50 ` Nhat Pham
2024-01-27 15:12 ` Chengming Zhou
2024-01-26 14:28 ` [PATCH 1/2] mm/zswap: don't return LRU_SKIP if we have dropped lru lock Johannes Weiner
2024-01-27 14:33 ` Chengming Zhou
2024-01-26 18:01 ` Nhat Pham
2024-01-27 14:37 ` Chengming Zhou [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=22e52669-2cf5-4375-9cb8-82d2f5faadc6@linux.dev \
--to=chengming.zhou@linux.dev \
--cc=akpm@linux-foundation.org \
--cc=hannes@cmpxchg.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=nphamcs@gmail.com \
--cc=yosryahmed@google.com \
--cc=zhouchengming@bytedance.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox