linux-kernel - Re: [PATCH] blk-throtl: optimize IOPS throttle for large IO scenarios

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] [day] [month] [year] [list]

Message-ID: <YQELtmeLjlDHCk2i@mtj.duckdns.org>
Date:   Tue, 27 Jul 2021 21:48:06 -1000
From:   Tejun Heo <tj@...nel.org>
To:     brookxu <brookxu.cn@...il.com>
Cc:     axboe@...nel.dk, cgroups@...r.kernel.org,
        linux-block@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] blk-throtl: optimize IOPS throttle for large IO scenarios

Hello,

On Wed, Jul 28, 2021 at 10:33:20AM +0800, brookxu wrote:
> If we do this, I think we will encounter some problems, as follows:
> 1. blk-throttle is now at the top of the IO stack. Changing the position of the throttle hook
>    will make this mechanism invalid for some devices.
> 2. We may also need to add a new hook to handle back-merge, otherwise I think the isolation
>    effect will be worse in the sequential IO scene.
> 3. Since bio has entered the IO stack, if it is suspended and resubmitted by blk-throttle,
>    then there is an IO stack reentry problem, which I think may cause many small problems.

Yeah, you're right. I don't think it's worth rocking the boat with
blk-throtl at this point. An extra callback it is then.

Thanks.

-- 
tejun