The assumption that the fid cannot be used by any other operation is
wrong. At least, nothing prevents a misbehaving client to create a
file with a given fid, and to pass this fid to some other operation
at the same time (ie, without waiting for the response to the creation
request). The call to v9fs_path_copy() performed by the worker thread
after the file was created can race with any access to the fid path
performed by some other thread. This causes use-after-free issues that
can be detected by ASAN with a custom 9p client.
Unlike other operations that only read the fid path, v9fs_co_open2()
does modify it. It should hence take the write lock.
Cc: P J P <ppan...@redhat.com>
Reported-by: zhibin hu <noirf...@gmail.com>
Signed-off-by: Greg Kurz <gr...@kaod.org>
---
hw/9pfs/cofile.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/hw/9pfs/cofile.c b/hw/9pfs/cofile.c
index 88791bc327ac..9c22837cda32 100644
--- a/hw/9pfs/cofile.c
+++ b/hw/9pfs/cofile.c
@@ -140,10 +140,10 @@ int coroutine_fn v9fs_co_open2(V9fsPDU *pdu, V9fsFidState
*fidp,
cred.fc_gid = gid;
/*
* Hold the directory fid lock so that directory path name
- * don't change. Read lock is fine because this fid cannot
- * be used by any other operation.
+ * don't change. Take the write lock to be sure this fid
+ * cannot be used by another operation.
*/
- v9fs_path_read_lock(s);
+ v9fs_path_write_lock(s);
v9fs_co_run_in_worker(
{
err = s->ops->open2(&s->ctx, &fidp->path,
--
2.17.2
