Received-SPF: pass (google.com: domain of linux-kernel+bounces-34844-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: <572bfa89-83c3-45ed-abc6-d71b9519813c@bytedance.com>
Date: Tue, 23 Jan 2024 15:46:25 +0800
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH v2 2/2] mm/zswap: split zswap rb-tree
Content-Language: en-US
To: Yosry Ahmed <yosryahmed@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>, Nhat Pham <nphamcs@gmail.com>,
 Chris Li <chriscli@google.com>, linux-kernel@vger.kernel.org,
 linux-mm@kvack.org, Johannes Weiner <hannes@cmpxchg.org>
References: <20240117-b4-zswap-lock-optimize-v2-0-b5cc55479090@bytedance.com>
 <20240117-b4-zswap-lock-optimize-v2-2-b5cc55479090@bytedance.com>
 <CAJD7tkbYv_TfEZ3Dj1JE=NXA323MdxGR9ib34PUoCmbfFaSCRQ@mail.gmail.com>
From: Chengming Zhou <zhouchengming@bytedance.com>
In-Reply-To: <CAJD7tkbYv_TfEZ3Dj1JE=NXA323MdxGR9ib34PUoCmbfFaSCRQ@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On 2024/1/23 03:49, Yosry Ahmed wrote:
> On Fri, Jan 19, 2024 at 3:22 AM Chengming Zhou
> <zhouchengming@bytedance.com> wrote:
>>
>> Each swapfile has one rb-tree to search the mapping of swp_entry_t to
>> zswap_entry, that use a spinlock to protect, which can cause heavy lock
>> contention if multiple tasks zswap_store/load concurrently.
>>
>> Optimize the scalability problem by splitting the zswap rb-tree into
>> multiple rb-trees, each corresponds to SWAP_ADDRESS_SPACE_PAGES (64M),
>> just like we did in the swap cache address_space splitting.
>>
>> Although this method can't solve the spinlock contention completely, it
>> can mitigate much of that contention. Below is the results of kernel build
>> in tmpfs with zswap shrinker enabled:
>>
>>      linux-next  zswap-lock-optimize
>> real 1m9.181s    1m3.820s
>> user 17m44.036s  17m40.100s
>> sys  7m37.297s   4m54.622s
>>
>> So there are clearly improvements.
> 
> If/when you respin this, can you mention that testing was done with a
> single swapfile? I assume the improvements will be less with multiple
> swapfiles as lock contention should be better.
> 

Ok. Not sure how much improvement, may do some tests later.

>>
>> Acked-by: Johannes Weiner <hannes@cmpxchg.org>
>> Acked-by: Nhat Pham <nphamcs@gmail.com>
> 
> I think the diff in zswap_swapoff() should be much simpler with the
> tree(s) cleanup removed. Otherwise LGTM.
> 
> Acked-by: Yosry Ahmed <yosryahmed@google.com>

Right, thanks!