[<prev] [next>] [day] [month] [year] [list]
Message-ID: <CAJNGr6uG6h=dBLjfB5ajZjv4qdf8-LjcW-i8L3_oqhbvBosSuA@mail.gmail.com>
Date: Thu, 15 May 2025 15:36:04 +0800
From: Guoyu Yin <y04609127@...il.com>
To: ericvh@...nel.org
Cc: lucho@...kov.net, asmadeus@...ewreck.org, linux_oss@...debyte.com,
v9fs@...ts.linux.dev, linux-kernel@...r.kernel.org
Subject: [BUG] INFO: rcu_tasks_trace detected stalls? (Late IPI?)
Hi,
I discovered a kernel crash with the description "INFO:
rcu_tasks_trace detected stalls? (Late IPI?)". Analysis of the logs
suggests that the issue may be related to numerous 9P network file
system connection failures and high system load.
Specifically, the p9_fd_create_tcp function repeatedly fails when
attempting to connect to 127.0.0.1, with the error "problem connecting
socket to 127.0.0.1". This could lead to resource exhaustion,
affecting CPU scheduling and interrupt handling, ultimately triggering
the RCU stall. The logs also show EXT4 file system errors (e.g., block
bitmap inconsistencies and delayed block allocation failures), likely
a side effect of Syzkaller testing.
I recommend investigating the 9P network file system’s connection
logic (particularly the socket connection part in p9_fd_create_tcp)
and system resource usage to pinpoint the root cause of the connection
failures and high load.
This can be reproduced on:
HEAD commit:
38fec10eb60d687e30c8c6b5420d86e8149f7557
report: https://pastebin.com/raw/0LB0VhWw
console output : https://pastebin.com/raw/EtnaHBBq
kernel config : https://pastebin.com/raw/u0Efyj5P
C reproducer : https://pastebin.com/raw/WzZ023uX
Powered by blists - more mailing lists