Linux kernel -stable discussions
 help / color / mirror / Atom feed
* FAILED: patch "[PATCH] apparmor: fix profile leak when a policy load fails" failed to apply to 6.18-stable tree
@ 2026-11-07 10:15 Sasha Levin
  2026-11-12  4:05 ` [PATCH 6.18.y 1/4] apparmor: refactor xattr attachment, to take the file path Sasha Levin
  0 siblings, 1 reply; 5+ messages in thread
From: Sasha Levin @ 2026-11-07 10:15 UTC (permalink / raw)
  To: georgia.garcia, john.johansen; +Cc: stable

The patch below does not apply to the 6.18-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable@vger.kernel.org>.

To reproduce the conflict and resubmit, you may use the following commands:

git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.18.y
git checkout FETCH_HEAD
git cherry-pick -x ce94927fa4173fcc1eca27bdc65a140064c8f1e6
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<stable@vger.kernel.org>' --in-reply-to '20261107051557-failed-6.18-ce94927fa417@kernel.org' --subject-prefix 'PATCH 6.18.y' 'HEAD^..'

Thanks,
Sasha

------------------ original commit in Linus's tree ------------------

From ce94927fa4173fcc1eca27bdc65a140064c8f1e6 Mon Sep 17 00:00:00 2001
From: Georgia Garcia <georgia.garcia@canonical.com>
Date: Wed, 14 Oct 2026 19:08:21 -0300
Subject: [PATCH] apparmor: fix profile leak when a policy load fails

The error paths of aa_replace_profiles() and aa_unpack() do not break
the refcount cycle between label and proxy. They just aa_put_profile()
the new profiles, which drops the label refcount from 2 to 1, and the
proxy keeps the profile alive. Every failed atomic policy load
(e.g. "profile can not be replaced", or a later profile in the blob
failing to unpack) leaks every new profile in the set, along with its
proxy, rules and its loaddata reference.

The leak is usually not visible from userspace, because the raw_data
directory for a failed load is not created, but it becomes visible
when the blob is deduplicated onto an existing loaddata, which then
keeps its policy/raw_data/<rev> directory forever.

Add aa_put_unpublished_profile(), which drops the profile's own proxy
before putting the profile, and use it on those error paths. The
proxy of an unpacked profile is always freshly allocated and not
referenced by anything else before the profile is published, so
putting it is safe.

Fixes: 637f688dc3dc ("apparmor: switch from profiles to using labels on contexts")
Cc: stable@vger.kernel.org
Signed-off-by: Georgia Garcia <georgia.garcia@canonical.com>
Signed-off-by: John Johansen <john.johansen@canonical.com>
---
 security/apparmor/include/policy.h | 20 ++++++++++++++++++++
 security/apparmor/policy.c         |  3 +++
 security/apparmor/policy_unpack.c  |  4 +++-
 3 files changed, 26 insertions(+), 1 deletion(-)

diff --git a/security/apparmor/include/policy.h b/security/apparmor/include/policy.h
index 0dcbf8cf1029a..bb851654cb80f 100644
--- a/security/apparmor/include/policy.h
+++ b/security/apparmor/include/policy.h
@@ -436,6 +436,26 @@ static inline void aa_put_profile(struct aa_profile *p)
 		kref_put(&p->label.count.count, aa_label_kref);
 }
 
+/**
+ * aa_put_unpublished_profile - put a profile that was never made visible
+ * @p: profile  (MAYBE NULL)
+ *
+ * A newly allocated profile and its proxy reference each other
+ * (proxy->label holds a label refcount). That cycle is normally broken
+ * when the profile is replaced or removed, which redirects the proxy.
+ * A profile that is discarded before being added to the policy tree
+ * never goes through either path, so break the cycle here, otherwise
+ * the profile, its proxy and its rawdata are leaked.
+ */
+static inline void aa_put_unpublished_profile(struct aa_profile *p)
+{
+	if (p && p->label.proxy) {
+		aa_put_proxy(p->label.proxy);
+		p->label.proxy = NULL;
+	}
+	aa_put_profile(p);
+}
+
 static inline int AUDIT_MODE(const struct aa_profile *profile)
 {
 	if (aa_g_audit != AUDIT_NORMAL)
diff --git a/security/apparmor/policy.c b/security/apparmor/policy.c
index bfe969b869577..63f7821b4dfff 100644
--- a/security/apparmor/policy.c
+++ b/security/apparmor/policy.c
@@ -1462,6 +1462,9 @@ ssize_t aa_replace_profiles(struct aa_ns *policy_ns, struct aa_label *label,
 	}
 	list_for_each_entry_safe(ent, tmp, &lh, list) {
 		list_del_init(&ent->list);
+		/* new was never added to the policy tree */
+		aa_put_unpublished_profile(ent->new);
+		ent->new = NULL;
 		aa_load_ent_free(ent);
 	}
 
diff --git a/security/apparmor/policy_unpack.c b/security/apparmor/policy_unpack.c
index 7ebffb4ff74fa..35dfa3b451f39 100644
--- a/security/apparmor/policy_unpack.c
+++ b/security/apparmor/policy_unpack.c
@@ -1822,13 +1822,15 @@ int aa_unpack(struct aa_loaddata *udata, struct list_head *lh,
 
 fail_profile:
 	kfree(ns_name);
-	aa_put_profile(profile);
+	aa_put_unpublished_profile(profile);
 
 fail:
 	if (compressed_data)
 		kvfree(compressed_data);
 	list_for_each_entry_safe(ent, tmp, lh, list) {
 		list_del_init(&ent->list);
+		aa_put_unpublished_profile(ent->new);
+		ent->new = NULL;
 		aa_load_ent_free(ent);
 	}
 
-- 
2.53.0





^ permalink raw reply related	[flat|nested] 5+ messages in thread
end of thread, other threads:[~2026-11-12  4:05 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-07 10:15 FAILED: patch "[PATCH] apparmor: fix profile leak when a policy load fails" failed to apply to 6.18-stable tree Sasha Levin
2026-11-12  4:05 ` [PATCH 6.18.y 1/4] apparmor: refactor xattr attachment, to take the file path Sasha Levin
2026-11-12  4:05   ` [PATCH 6.18.y 2/4] apparmor: constify aa_perms parameters that are read-only Sasha Levin
2026-11-12  4:05   ` [PATCH 6.18.y 3/4] apparmor: constify aa_profile parameters on read-only compute paths Sasha Levin
2026-11-12  4:05   ` [PATCH 6.18.y 4/4] apparmor: fix profile leak when a policy load fails Sasha Levin

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox