From: Hui Peng <benquike@gmail.com>
To: jlayton@kernel.org, cel@kernel.org, viro@zeniv.linux.org.uk,
brauner@kernel.org
Cc: alex.aring@gmail.com, jack@suse.cz,
linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH] filelock: prevent FL_LEASE and FL_DELEG flavor confusion in generic_add_lease()
Date: Sat, 19 Sep 2026 21:06:17 +0000 [thread overview]
Message-ID: <20260919210617.3028917-1-benquike@gmail.com> (raw)
generic_add_lease() searches ctx->flc_lease for an existing entry
matching (flc_file == filp && flc_owner == lease->c.flc_owner) in order
to modify an existing lease in place via lease_modify(). However, it
does not verify that the existing entry's flavor (FL_LEASE vs FL_DELEG)
matches the newly requested lease's flavor in flc_flags.
Since both F_SETLEASE and F_SETDELEG pass filp as flc_owner from
userland (fcntl_setlease() and fcntl_setdeleg()), calling F_SETDELEG on
a file descriptor that already holds an FL_LEASE modifies flc_type in
place and returns 0 without setting FL_DELEG in flc_flags. Conversely,
calling F_SETLEASE on a file descriptor that holds an active FL_DELEG
modifies the delegation in place without holding inode_lock() (which
vfs_setlease() only acquires when is_deleg is true).
Reject cross-flavor modifications in generic_add_lease() when the
existing entry on ctx->flc_lease has a different (FL_LEASE | FL_DELEG)
flag mask from the requested lease.
Fixes: 1602bad16d7d ("vfs: expose delegation support to userland")
Assisted-by: LLM
Signed-off-by: Hui Peng <benquike@gmail.com>
---
fs/locks.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/fs/locks.c b/fs/locks.c
index 6e4ff7fcec05..15d73f8f48cd 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -1938,6 +1938,9 @@ generic_add_lease(struct file *filp, int arg, struct file_lease **flp, void **pr
list_for_each_entry(fl, &ctx->flc_lease, c.flc_list) {
if (fl->c.flc_file == filp &&
fl->c.flc_owner == lease->c.flc_owner) {
+ if ((fl->c.flc_flags & (FL_LEASE | FL_DELEG)) !=
+ (lease->c.flc_flags & (FL_LEASE | FL_DELEG)))
+ goto out;
my_fl = fl;
continue;
}
--
2.55.0.1082.g2b9226bbc0-goog
reply other threads:[~2026-09-19 21:06 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=20260919210617.3028917-1-benquike@gmail.com \
--to=benquike@gmail.com \
--cc=alex.aring@gmail.com \
--cc=brauner@kernel.org \
--cc=cel@kernel.org \
--cc=jack@suse.cz \
--cc=jlayton@kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=viro@zeniv.linux.org.uk \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®