lkml.org 
[lkml]   [2026]   [Mar]   [11]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Subject[PATCH 6.1.y] binfmt_misc: restore write access before closing files opened by open_exec()
Date
From: Zilin Guan <zilin@seu.edu.cn>

[ Upstream commit 90f601b497d76f40fa66795c3ecf625b6aced9fd ]

bm_register_write() opens an executable file using open_exec(), which
internally calls do_open_execat() and denies write access on the file to
avoid modification while it is being executed.

However, when an error occurs, bm_register_write() closes the file using
filp_close() directly. This does not restore the write permission, which
may cause subsequent write operations on the same file to fail.

Fix this by calling exe_file_allow_write_access() before filp_close() to
restore the write permission properly.

Fixes: e7850f4d844e ("binfmt_misc: fix possible deadlock in bm_register_write")
Signed-off-by: Zilin Guan <zilin@seu.edu.cn>
Link: https://patch.msgid.link/20251105022923.1813587-1-zilin@seu.edu.cn
Signed-off-by: Christian Brauner <brauner@kernel.org>
[ Use allow_write_access() instead of exe_file_allow_write_access()
according to commit 0357ef03c94ef
("fs: don't block write during exec on pre-content watched files"). ]
Signed-off-by: Robert Garcia <rob_garcia@163.com>
---
fs/binfmt_misc.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c
index 740dac1012ae..05c235309421 100644
--- a/fs/binfmt_misc.c
+++ b/fs/binfmt_misc.c
@@ -816,8 +816,10 @@ static ssize_t bm_register_write(struct file *file, const char __user *buffer,
inode_unlock(d_inode(root));

if (err) {
- if (f)
+ if (f) {
+ allow_write_access(f);
filp_close(f, NULL);
+ }
kfree(e);
return err;
}
--
2.34.1

\
 
 \ /
  Last update: 2026-03-11 03:10    [W:0.088 / U:10.724 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and my Meterkast|Read the blog