linux-kernel - Re: [PATCH v4] zram: Implement multi-page write-back

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <tencent_FE5751D11BB7EC389B2BC0B957DEE0D75E06@qq.com>
Date: Thu, 13 Nov 2025 15:55:20 +0800
From: Yuwen Chen <ywen.chen@...mail.com>
To: senozhatsky@...omium.org
Cc: akpm@...ux-foundation.org,
	axboe@...nel.dk,
	bgeffon@...gle.com,
	licayy@...look.com,
	linux-block@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	linux-mm@...ck.org,
	liumartin@...gle.com,
	minchan@...nel.org,
	richardycc@...gle.com,
	ywen.chen@...mail.com
Subject: Re: [PATCH v4] zram: Implement multi-page write-back

On Thu, 13 Nov 2025 11:04:04 +0900, Sergey Senozhatsky wrote:
> How was this number chosen?  Did you try lower/higher values?
> I think we might want this to be runtime tunable via sysfs, e.g.
> writeback_batch_size attr, with min value of 1.

I haven't conducted any tests on this value. I just set an empirical
value of 32 based on the submission queue length of the storage device.

As you said, providing a sys node for configuration might offer
performance advantages for mechanical hard drives.

On Thu, 13 Nov 2025 11:20:15 +0900, Sergey Senozhatsky wrote:
> So I wonder if things will look simpler (is this the word I'm looking
> for?) if you just have two lists for requests: one list for completed/idle
> requests and one list for in-flight requests (and you move requests
> around accordingly).  Then you don't need to iterate the pool and check
> flags, you just can check list_empty(&idle_requests) and take the first
> (front) element.

Yes, using two linked lists can reduce the complexity. It's just that
before I saw your submission, I couldn't find a better way to avoid
introducing locks. Thank you very much!