linux-ext4 - [BUG] INFO: rcu_tasks_trace detected stalls on tasks:

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>] [day] [month] [year] [list]

Message-ID: <CAJNGr6uGfUpvHkPdE-OrWL4_cqd+_AG9Z4ZS9DB1KfYm3CzQAA@mail.gmail.com>
Date: Thu, 15 May 2025 14:54:43 +0800
From: Guoyu Yin <y04609127@...il.com>
To: tytso@....edu, adilger.kernel@...ger.ca, linux-ext4@...r.kernel.org, 
	linux-kernel@...r.kernel.org
Subject: [BUG] INFO: rcu_tasks_trace detected stalls on tasks:

Hi,

This crash can be triggered by repeatedly executing a syzkaller test
case that stresses the signal handling path and EXT4 filesystem
operations. The reproducer causes the kernel to process signals (e.g.,
via do_group_exit) while performing delayed block allocation on a
corrupted EXT4 filesystem.

The call trace shows the task is stuck in get_signal
(kernel/signal.c:3036) and exit_to_user_mode_prepare
(include/linux/entry-common.h:329) while handling signals.I think the
key failure occurs in ext4_validate_block_bitmap due to inconsistent
block bitmap padding, combined with an ENOSPC error during delayed
allocation. This leads to RCU stalls when the task (syz-executor.4)
fails to exit the RCU read-side critical section during resource
cleanup.

Maybe we can add sanity checks for block bitmap padding in
ext4_validate_block_bitmap?

This can be reproduced on:
HEAD commit: 38fec10eb60d687e30c8c6b5420d86e8149f7557

report: https://pastebin.com/raw/A7yVpRQJ

console output : https://pastebin.com/raw/Me02WbSY

kernel config : https://pastebin.com/raw/u0Efyj5P

C reproducer : https://pastebin.com/raw/TLx7rz0Q