All of lore.kernel.org
 help / color / mirror / Atom feed
From: Cen Zhang <zzzccc427@gmail.com>
To: mark@fasheh.com, jlbec@evilplan.org, joseph.qi@linux.alibaba.com,
	heming.zhao@suse.com, akpm@linux-foundation.org,
	zzzccc427@gmail.com, kmehltretter@gmail.com,
	ericterminal@gmail.com, kees@kernel.org,
	sunil.mushran@oracle.com, christophe.jaillet@wanadoo.fr
Cc: ocfs2-devel@lists.linux.dev, linux-kernel@vger.kernel.org,
	baijiaju1990@gmail.com, jjzuming@gmail.com
Subject: [PATCH] ocfs2: Pin the node manager module for open debugfs files
Date: Mon, 12 Oct 2026 09:31:38 +0800	[thread overview]
Message-ID: <pm-ocfs2-objects-candidate-0089-v2-4c67f48d4e2d0fc85990@gmail.com> (raw)

The node manager debugfs file operation tables do not set owner.
The debugfs full proxy obtains real_fops with fops_get() at open and
keeps it until final release. Without an owner, this does not acquire
a module reference. Normal unload can free the ocfs2_nodemanager
module image while an open file still holds its operation table.

Closing connected_nodes then makes full_proxy_release() read
real_fops->release from freed module memory. Removing the debugfs
entries during module exit does not drain retained open files.

Set owner to THIS_MODULE in the o2net and heartbeat debugfs operation
tables. The existing debugfs get/put pair then keeps the module and
its callbacks alive until the last open file is released.

A controlled test used an unloadable OCFS2 module and case-local probe
filters without changing the candidate open, unload or release logic.
Closing a retained connected_nodes file after normal unload produced:

  BUG: unable to handle page fault for address: fffffbfff804cfef
  Oops: Oops: 0000 [#1] SMP KASAN NOPTI
  RIP: 0010:full_proxy_release+0xbf/0x190
  Call Trace:
   <TASK>
   __fput+0x39a/0xa60
   fput_close_sync+0xff/0x200
   __x64_sys_close+0x8c/0xf0
   do_syscall_64+0x114/0x620
   entry_SYSCALL_64_after_hwframe+0x77/0x7f
   </TASK>
  Modules linked in: [last unloaded: ocfs2_nodemanager]

Fixes: 2309e9e040fe ("ocfs2/net: Add debug interface to o2net")
Fixes: 87d3d3f3931f ("ocfs2/hb: Expose the list of heartbeating nodes via debugfs")
Fixes: 3ba169ccec1c ("ocfs2/cluster: Add new function o2net_fill_node_map()")
Assisted-by: LLM
Signed-off-by: Cen Zhang <zzzccc427@gmail.com>
---
diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c
index 8e6ead433228..1d4f71190d27 100644
--- a/fs/ocfs2/cluster/heartbeat.c
+++ b/fs/ocfs2/cluster/heartbeat.c
@@ -1447,6 +1447,7 @@ static ssize_t o2hb_debug_read(struct file *file, char __user *buf,
 #endif  /* CONFIG_DEBUG_FS */
 
 static const struct file_operations o2hb_debug_fops = {
+	.owner = THIS_MODULE,
 	.open =		o2hb_debug_open,
 	.release =	o2hb_debug_release,
 	.read =		o2hb_debug_read,
diff --git a/fs/ocfs2/cluster/netdebug.c b/fs/ocfs2/cluster/netdebug.c
index f7ab66e70799..5d373002297d 100644
--- a/fs/ocfs2/cluster/netdebug.c
+++ b/fs/ocfs2/cluster/netdebug.c
@@ -183,6 +183,7 @@ static int nst_fop_release(struct inode *inode, struct file *file)
 }
 
 static const struct file_operations nst_seq_fops = {
+	.owner = THIS_MODULE,
 	.open = nst_fop_open,
 	.read = seq_read,
 	.llseek = seq_lseek,
@@ -415,6 +416,7 @@ static int stats_fop_open(struct inode *inode, struct file *file)
 }
 
 static const struct file_operations stats_seq_fops = {
+	.owner = THIS_MODULE,
 	.open = stats_fop_open,
 	.read = seq_read,
 	.llseek = seq_lseek,
@@ -427,6 +429,7 @@ static int sc_fop_open(struct inode *inode, struct file *file)
 }
 
 static const struct file_operations sc_seq_fops = {
+	.owner = THIS_MODULE,
 	.open = sc_fop_open,
 	.read = seq_read,
 	.llseek = seq_lseek,
@@ -476,6 +479,7 @@ static ssize_t o2net_debug_read(struct file *file, char __user *buf,
 }
 
 static const struct file_operations nodes_fops = {
+	.owner		= THIS_MODULE,
 	.open		= nodes_fop_open,
 	.release	= o2net_debug_release,
 	.read		= o2net_debug_read,

             reply	other threads:[~2026-10-12  1:31 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-12  1:31 Cen Zhang [this message]
2026-10-12  6:02 ` [PATCH] ocfs2: Pin the node manager module for open debugfs files Heming Zhao
2026-10-12 12:01   ` Cen Zhang

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=pm-ocfs2-objects-candidate-0089-v2-4c67f48d4e2d0fc85990@gmail.com \
    --to=zzzccc427@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=baijiaju1990@gmail.com \
    --cc=christophe.jaillet@wanadoo.fr \
    --cc=ericterminal@gmail.com \
    --cc=heming.zhao@suse.com \
    --cc=jjzuming@gmail.com \
    --cc=jlbec@evilplan.org \
    --cc=joseph.qi@linux.alibaba.com \
    --cc=kees@kernel.org \
    --cc=kmehltretter@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mark@fasheh.com \
    --cc=ocfs2-devel@lists.linux.dev \
    --cc=sunil.mushran@oracle.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.