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]

Date: Wed, 07 Feb 2024 11:44:41 +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

Matteo Italia (matteo@...alia.net) changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |matteo@...alia.net

--- Comment #66 from Matteo Italia (matteo@...alia.net) ---
I think I'm experiencing the same problem; relatively big (1.5T), relatively
empty (716 G used) ext4 partition on a slow-ish SATA magnetic HDD
(ST3000DM007-1WY10G), stripe=32747, 6.5.0-14-generic kernel from Ubuntu 22.04
HWE stack; during the installation of COMSOL Multiphysics 6.2 (which
essentially untars ~16 GB of stuff) I got kworker flush process stuck at 100%
of CPU for multiple hours, and IO cache filling my 32 GB of RAM. I tried the
various parameters tweaking suggested here with no particular change of pace
(although maybe the pending IO still worked according to the old values?). I
run the bfq scheduler on this disk, but I tried to switch to pretty much all
other schedulers mid-process with no perceptible change.

After ~3 hours of agonizingly slow progress (with disk write performance for
the installation often in the tens of KB/s) I decided to abort the
installation. I rebooted in the Ubuntu GA kernel (5.15.0-92-generic) without
touching any specific parameter and restarted the installation; this time it
completed in ~10 minutes.

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

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