linux-kernel - Re: [LKP] [ext4] 05c2c00f37: aim7.jobs-per-min -11.8% regression

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: <YTIV1SozOGwT/DQw@mit.edu>
Date:   Fri, 3 Sep 2021 08:32:21 -0400
From:   "Theodore Ts'o" <tytso@....edu>
To:     Xing Zhengjun <zhengjun.xing@...ux.intel.com>
Cc:     Jan Kara <jack@...e.cz>, kernel test robot <oliver.sang@...el.com>,
        LKML <linux-kernel@...r.kernel.org>, lkp@...ts.01.org,
        lkp@...el.com
Subject: Re: [LKP] [ext4] 05c2c00f37: aim7.jobs-per-min -11.8% regression

On Fri, Sep 03, 2021 at 01:28:25PM +0800, Xing Zhengjun wrote:
> Hi Jan,
> 
>      Do you have time to look at this? I re-test it in v5.13 and v5.14, the
> regression still existed. Thanks.

> > So as we chatted on today's call probably the best option is to leave the
> > code as is for now and instead work on moving away from orphan list
> > altogether. I'll revive my patches to do that.

The orphan_file feature has landed in for the 5.15 merge window.  When
enabled, it provides multiple blocks to store the set of orphaned
inodes, which provides parallelism for parallel truncate and unlink
operations.

It requires a development version of e2fsprogs to use, and so the
zero-day bot won't be able to test it for a little while.

Cheers,

						- Ted