linux-ext4 - Re: ext4/004 hangs with -o inlinecrypt,test_dummy

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: <80f77860-2d5d-4ff9-9bb8-1e5bc46a4692@bytedance.com>
Date: Tue, 9 Dec 2025 16:40:46 +0800
From: Julian Sun <sunjunchao@...edance.com>
To: Christoph Hellwig <hch@...radead.org>, linux-ext4@...r.kernel.org
Subject: Re: ext4/004 hangs with -o inlinecrypt,test_dummy_encryption

On 12/8/25 2:59 PM, Christoph Hellwig wrote:
> Hi all,
> 
> I've just been wanting to test my changes to the inline encruption
> fallback code, and it seems like ext/004 (the only ext4 dump test)
> hangs when using the following options:
> 
> export MKFS_OPTIONS='-O encrypt'
> export MOUNT_OPTIONS="-o inlinecrypt,test_dummy_encryption"
> 
> I thought I did not see this before, but it reproduces back to at least
> Linux 6.17.  This is in an uptodate Debian trixie VM.  The dump/restore
> process look like in weird states:
> 
>     4727 ttyS0    T      0:00 /usr/sbin/dump -0 -f - /mnt/scratch/dump_restore_dir
>     4728 ttyS0    T      0:00 /usr/sbin/restore -urvf -
>     4729 ttyS0    T      0:00 /usr/sbin/dump -0 -f - /mnt/scratch/dump_restore_dir
>     4730 ttyS0    T      0:00 /usr/sbin/dump -0 -f - /mnt/scratch/dump_restore_dir
>     4731 ttyS0    T      0:00 /usr/sbin/dump -0 -f - /mnt/scratch/dump_restore_dir
>     4732 ttyS0    T      0:00 /usr/sbin/dump -0 -f - /mnt/scratch/dump_restore_dir
> 

I can reproduce this issue locally with both v6.18 and v6.0. The problem 
disappears after removing test_dummy_encryption, and it still reproduces 
when test_dummy_encryption is set alone in MOUNT_OPTIONS. Therefore, I 
believe the issue lies in test_dummy_encryption — it is an 
implementation of fscrypt.

CC: linux-fscrypt

Thanks,
-- 
Julian Sun <sunjunchao@...edance.com>