linux-kernel - [REGRESSION] ioprio performance hangs, bisected

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-next>] [day] [month] [year] [list]

Message-ID: <CAP-bSRbCo7=wfUBZ8H7c3Q-7XSG+SB=R4MHHNNGPvBoinsVSZg@mail.gmail.com>
Date: Mon, 25 Nov 2024 15:44:49 +0000
From: Chris Bainbridge <chris.bainbridge@...il.com>
To: hch@....de
Cc: LKML <linux-kernel@...r.kernel.org>, axboe@...nel.dk, bvanassche@....org, 
	Linux regressions mailing list <regressions@...ts.linux.dev>, linux-block@...r.kernel.org
Subject: [REGRESSION] ioprio performance hangs, bisected

The commit 6975c1a486a4 ("block: remove the ioprio field from struct
request") appears to have introduced a performance regression. Test
case is the script /etc/cron.daily/locate from package `locate` (which
is from findutils locate on Debian). The script runs:

  ionice -c 3 -p 3965
  nice -n 10 updatedb.findutils

This locate script will usually complete in about 1m20s on my machine.
But, following commit 6975c1a486a4, the script doesn't seem to
complete in a reasonable time (I waited 20 minutes). `top` shows that
no or little CPU time seems to be dedicated to the find process. Also
it seems that this affects other processes too - once find is in a
"hung" state, doing `ls -R` on the affected drive will also hang, even
though the priority of the terminal/bash/ls process hasn't changed,
and is supposedly independent of the locate/find script. I'm running
btrfs, on an external USB SSD drive.

#regzbot introduced: 6975c1a486a40446b5bc77a89d9c520f8296fd08