linux-ext4 - [Bug 217965] ext4(?) regression since 6.5.0 on sata hdd

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: <bug-217965-13602-olhSE5cbcu@https.bugzilla.kernel.org/>
Date: Tue, 12 Mar 2024 20:27:43 +0000
From: bugzilla-daemon@...nel.org
To: linux-ext4@...r.kernel.org
Subject: [Bug 217965] ext4(?) regression since 6.5.0 on sata hdd

https://bugzilla.kernel.org/show_bug.cgi?id=217965

denk@...t.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |denk@...t.com

--- Comment #67 from denk@...t.com ---
Hi,

I just stumbled upon this issue, which it seems is the same I am suffering from
on a Raspberry Pi 5. Th e OS is Arch Linux ARM and linux-rpi 6.6.20 running on
a NVME SSD. After reading this issue, I checked the ext4 mount options and
found stripe=8191. As a first test I downgraded linux-rpi to 6.1.69 and my
issue of a CPU hogging kworker process, when pulling and creating docker
containers did not happen anymore. With kernel 6.6.20 and all earlier 6.6.x
versions, docker operations were really slow, in some cases pulling and
recreating a container took over 1000 seconds. With kernel 6.1.69 these
operations are back to normal speed.
As a next step I would like to try to set stripe=0. What is the right approach
here, as I am a bit unsure. Is it ok to remount the running system partition
with mount -o remount,stripe=0 <dev> without any risks for a first test? And to
change permanently to stripe=0,id it ok and sufficient to run tune2fs -f -E
stride=0 <dev> on the mounted system partition without any risks? I would be
thankful for a helping hand here to not do anything wrong or dangerous. 

I hope, that this works around my issue for now, until a fix has been found for
this. Thank you very much for looking into this!

Kind regards
denk

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.