eventfs: Have eventfs_iterate() stop immediately if ei->is_freed is set
authorSteven Rostedt (Google) <rostedt@goodmis.org>
Thu, 4 Jan 2024 21:57:12 +0000 (16:57 -0500)
committerSteven Rostedt (Google) <rostedt@goodmis.org>
Thu, 4 Jan 2024 22:11:57 +0000 (17:11 -0500)
If ei->is_freed is set in eventfs_iterate(), it means that the directory
that is being iterated on is in the process of being freed. Just exit the
loop immediately when that is ever detected, and separate out the return
of the entry->callback() from ei->is_freed.

Link: https://lore.kernel.org/linux-trace-kernel/20240104220048.016261289@goodmis.org
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
fs/tracefs/event_inode.c

index 72912b5f9a90bc56146bf193998bed76689965e6..0aca6910efb3f5e1ca2664c447074e177a1ca7cb 100644 (file)
@@ -788,11 +788,12 @@ static int eventfs_iterate(struct file *file, struct dir_context *ctx)
                name = entry->name;
 
                mutex_lock(&eventfs_mutex);
-               /* If ei->is_freed, then the event itself may be too */
-               if (!ei->is_freed)
-                       r = entry->callback(name, &mode, &cdata, &fops);
-               else
-                       r = -1;
+               /* If ei->is_freed then just bail here, nothing more to do */
+               if (ei->is_freed) {
+                       mutex_unlock(&eventfs_mutex);
+                       goto out;
+               }
+               r = entry->callback(name, &mode, &cdata, &fops);
                mutex_unlock(&eventfs_mutex);
                if (r <= 0)
                        continue;