Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.38 as permitted sender) client-ip=23.128.96.38;
Message-ID: <74bb08ba-25c8-4aca-af04-18a45208bdad@bytedance.com>
Date:   Wed, 13 Dec 2023 10:57:25 +0800
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH 0/7] mm/zswap: optimize the scalability of zswap rb-tree
Content-Language: en-US
To:     Nhat Pham <nphamcs@gmail.com>, Yosry Ahmed <yosryahmed@google.com>
Cc:     Chris Li <chriscli@google.com>,
        Vitaly Wool <vitaly.wool@konsulko.com>,
        Johannes Weiner <hannes@cmpxchg.org>,
        Michal Hocko <mhocko@kernel.org>,
        Seth Jennings <sjenning@redhat.com>,
        Dan Streetman <ddstreet@ieee.org>,
        Andrew Morton <akpm@linux-foundation.org>, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org, Matthew Wilcox <willy@infradead.org>
References: <20231206-zswap-lock-optimize-v1-0-e25b059f9c3a@bytedance.com>
 <CAKEwX=NXcY3_GuKbh0=Ceg9wj=7u4y0NgmfSVEG3-+deY0jiWQ@mail.gmail.com>
 <CAJD7tkbErWz7Rjn-JdY8LjSW=GzEyyrNeJ5P1ipChFSufmQmLQ@mail.gmail.com>
 <CAF8kJuMegmMD3+YZfvsW3h1y1z1-kH7SiyJHPVnrSc89OZASuA@mail.gmail.com>
 <77d628dc-ab8c-4d8c-bc63-7e4518ea92d7@bytedance.com>
 <CAJD7tkZCxwDpNm-jQv_ieDvYhrtvMTXXjRp9_dQW3_VqYgUUsw@mail.gmail.com>
 <CAKEwX=PfRNLL_t5q8vfEKA_rxgFcAer=9EFwLWDo9kDsmroAGw@mail.gmail.com>
From:   Chengming Zhou <zhouchengming@bytedance.com>
In-Reply-To: <CAKEwX=PfRNLL_t5q8vfEKA_rxgFcAer=9EFwLWDo9kDsmroAGw@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Precedence: bulk

On 2023/12/13 07:33, Nhat Pham wrote:
> On Tue, Dec 12, 2023 at 3:27 PM Yosry Ahmed <yosryahmed@google.com> wrote:
>>
>> Let's split the rbtree breakdown into a separate series. This series
>> has irrelevant (and very nice) cleanups and optimizations, let's get
>> them separately and defer the rbtree breakdown part until we get data

Ok, will split and just send the cleanups/optimizations with dstmem reuse.

>> about the xarray implementation. Perhaps the tree breakdown is not
>> needed as much with an xarray, or at the very least the implementation
>> would look different on top of an xarray.

Yeah, will retest on the xarray version of Chris, the implementation is
easy anyway.

> 
> Actually, kinda agree - I quite like the cleanup/optimization done
> w.r.t dstmem reuse :)

Thanks!