From nobody Sat Sep 26 02:32:49 2026 Received: from dggsgout12.his.huawei.com (dggsgout12.his.huawei.com [45.249.212.56]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id D62092BDC2A; Sat, 5 Sep 2026 02:06:34 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=45.249.212.56 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1788573998; cv=none; b=Cl9VSfljxQnA+uonKEZeXzw9Xd7M6ZOiURlRoeanWd6yHXYrY477Mpok9nkWALiKFgixrvHNpA9n5iaWdRYtTN5yPLjXcZKBfIDPn2xTdm2dUZ8uBvDYIuSUj9nMPW5Z5L2QKSree4sqymef5GTbP33P2IM/Z2DfuiG0jlvzI8k= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1788573998; c=relaxed/simple; bh=TP6FLXOhz2BdUy+YQ34NHJkGdm2IsYYyKNqxHrB0YWo=; h=From:To:Cc:Subject:Date:Message-Id:MIME-Version; b=ubADVG/lFGVJzNsBrKRasAkUlq/Fy73og838azkgNcdJzgd8MxXGyO3sWhBtQw2B9LiYlGdzn0KkQq+Rh/igDA6R36anTSR/9yLSWYjBVU6gpjF2121JbGo3JiqWgmz5RnRjGsAtl3OIa/UCewpHmxmZfxpcMalylBFRCXyDJG0= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=huaweicloud.com; spf=pass smtp.mailfrom=huaweicloud.com; arc=none smtp.client-ip=45.249.212.56 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=huaweicloud.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=huaweicloud.com Received: from mail.maildlp.com (unknown [172.19.163.170]) by dggsgout12.his.huawei.com (SkyGuard) with ESMTPS id 4hcGrc25YBzKHMLs; Sat, 5 Sep 2026 10:05:36 +0800 (CST) Received: from mail02.huawei.com (unknown [10.116.40.112]) by mail.maildlp.com (Postfix) with ESMTP id 7EE9C4056D; Sat, 5 Sep 2026 10:06:32 +0800 (CST) Received: from ultra.huawei.com (unknown [10.90.53.71]) by APP1 (Coremail) with UTF8SMTPA id cCh0CgB3XwoneZtqm8crAw--.7745S2; Sat, 05 Sep 2026 10:06:32 +0800 (CST) From: Pu Lehui To: bpf@vger.kernel.org, linux-kernel@vger.kernel.org, Alexei Starovoitov , Hou Tao , Leon Hwang Cc: Daniel Borkmann , Andrii Nakryiko , Eduard Zingerman , Kumar Kartikeya Dwivedi , Martin KaFai Lau , Yonghong Song , Song Liu , Jiri Olsa , Emil Tsalapatis , Ihor Solodrai , Pu Lehui , Pu Lehui Subject: [PATCH bpf v4] bpf: Fix UAF due to concurrent consumption of ttrace lists in alloc_bulk Date: Sat, 5 Sep 2026 02:11:39 +0000 Message-Id: <20260905021139.4116529-1-pulehui@huaweicloud.com> X-Mailer: git-send-email 2.34.1 Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-CM-TRANSID: cCh0CgB3XwoneZtqm8crAw--.7745S2 X-Coremail-Antispam: 1UD129KBjvJXoW3GF4fur1rXF1Uur4rury5twb_yoWxZry8pF W3Jr95XF4UZF4av3WSqr48Ca9Iqw4vq343J3y7uryS9r1Yvwn0gFyfCry3uF90yrZ29a4a yr1qgry8AF4UZw7anT9S1TB71UUUUU7qnTZGkaVYY2UrUUUUjbIjqfuFe4nvWSU5nxnvy2 9KBjDU0xBIdaVrnRJUUU9Y14x267AKxVW5JVWrJwAFc2x0x2IEx4CE42xK8VAvwI8IcIk0 rVWrJVCq3wAFIxvE14AKwVWUJVWUGwA2ocxC64kIII0Yj41l84x0c7CEw4AK67xGY2AK02 1l84ACjcxK6xIIjxv20xvE14v26r1I6r4UM28EF7xvwVC0I7IYx2IY6xkF7I0E14v26r4j 6F4UM28EF7xvwVC2z280aVAFwI0_Gr1j6F4UJwA2z4x0Y4vEx4A2jsIEc7CjxVAFwI0_Gc CE3s1le2I262IYc4CY6c8Ij28IcVAaY2xG8wAqx4xG64xvF2IEw4CE5I8CrVC2j2WlYx0E 2Ix0cI8IcVAFwI0_Jr0_Jr4lYx0Ex4A2jsIE14v26r1j6r4UMcvjeVCFs4IE7xkEbVWUJV W8JwACjcxG0xvY0x0EwIxGrwACjI8F5VA0II8E6IAqYI8I648v4I1lFIxGxcIEc7CjxVA2 Y2ka0xkIwI1lc7CjxVAaw2AFwI0_GFv_Wryl42xK82IYc2Ij64vIr41l4I8I3I0E4IkC6x 0Yz7v_Jr0_Gr1lx2IqxVAqx4xG67AKxVWUJVWUGwC20s026x8GjcxK67AKxVWUGVWUWwC2 zVAF1VAY17CE14v26r4a6rW5MIIYrxkI7VAKI48JMIIF0xvE2Ix0cI8IcVAFwI0_Jr0_JF 4lIxAIcVC0I7IYx2IY6xkF7I0E14v26r4j6F4UMIIF0xvE42xK8VAvwI8IcIk0rVWUJVWU CwCI42IY6I8E87Iv67AKxVWUJVW8JwCI42IY6I8E87Iv6xkF7I0E14v26r4j6r4UJbIYCT nIWIevJa73UjIFyTuYvjTRNJ5oDUUUU X-CM-SenderInfo: psxovxtxl6x35dzhxuhorxvhhfrp/ Content-Type: text/plain; charset="utf-8" From: Pu Lehui Syzkaller repeatedly triggered UAF splats related to nodes in waiting_for_gp_ttrace within the bpf memalloc: BUG: KASAN: slab-use-after-free in llist_del_first+0x85/0x110 lib/llist.c:61 Read of size 8 at addr ffff8881572cd080 by task syz.4.470/5112 ... llist_del_first+0x85/0x110 lib/llist.c:61 alloc_bulk+0x193/0x460 kernel/bpf/memalloc.c:229 bpf_mem_refill+0x386/0x560 kernel/bpf/memalloc.c:436 Freed by task 14: ... __free_rcu kernel/bpf/memalloc.c:281 [inline] __free_rcu_tasks_trace+0x48/0xd0 kernel/bpf/memalloc.c:291 rcu_tasks_invoke_cbs+0x1ec/0x3e0 kernel/rcu/tasks.h:571 rcu_tasks_one_gp+0x13d/0x220 kernel/rcu/tasks.h:621 rcu_tasks_kthread+0xf3/0x120 kernel/rcu/tasks.h:651 The reason is that the UAF occurs after the RCU Tasks Trace GP expires: when the __free_rcu() callback runs, there is no synchronization protecting llist_del_all() against concurrent alloc_bulk() operating on waiting_for_gp_ttrace, leading to the race condition below: CPU0 CPU1 __free_rcu (RCU Tasks Trace = callback) alloc_bulk llist_del_first(&c->waiting_for_gp_ttrace) entry =3D smp_load_acquire(&head->first); do { if (entry =3D=3D NULL) return NULL; free_all(llist_del_all(&c->w= aiting_for_gp_ttrace)) llist_for_each_safe(pos, t= , llnode) free_one(pos); next =3D READ_ONCE(entry->next); <-- trigger UAF } while (!try_cmpxchg(&head->first, &entry, next)); In addition, there is also a theoretical race condition on the free_by_rcu_ttrace list. This race requires two preconditions: an in-flight Tasks Trace GP keeping c->call_rcu_ttrace_in_progress =3D=3D 1, and concurrent cross-CPU frees repopulating c->free_by_rcu_ttrace with new nodes. Under these conditions, the following scenario triggers UAF: // CPU0 // irq work is still busy (on PREEMPT_RT) alloc_bulk() llist_del_first(&c->free_by_rcu_ttrace) entry =3D smp_load_acquire(&head->first); do { if (entry =3D=3D NULL) return NULL; // CPU1 bpf_mem_alloc_destroy() WRITE_ONCE(c->draining, true) // wait for CPU0 irq_work_sync() // CPU2 do_call_rcu_ttrace(tgt(CPU0)) if (c->draining) { llist_del_all(&c->free_by_rcu_ttrace) free_all() } // CPU0 continue next =3D READ_ONCE(entry->next); <-- trigger UAF while (!try_cmpxchg(&head->first, &entry, next)); Fix this by introducing a raw spinlock to synchronize the concurrent consumption on waiting_for_gp_ttrace and free_by_rcu_ttrace. Fixes: 04fabf00b4d3 ("bpf: Allow reuse from waiting_for_gp_ttrace list.") Suggested-by: Alexei Starovoitov Suggested-by: Hou Tao Signed-off-by: Pu Lehui Acked-by: Hou Tao --- v4: - Update comments for ttrace list in alloc_bulk. (BPF CI) - Drop initial troubleshooting explaining in commit message. v3: https://lore.kernel.org/bpf/20260901150307.3474349-1-pulehui@huaweiclou= d.com - Fix concurrent issue also for free_by_rcu_ttrace. (bpfci and Hou Tao) - Use scoped_guard. (Leon) v2: https://lore.kernel.org/bpf/20260827084013.1062816-1-pulehui@huaweiclou= d.com - Use raw spinlock to fix concurrent alloc_bulk and __free_rcu on waiting_for_gp_ttrace after GP. (Hou Tao) v1: https://lore.kernel.org/bpf/20260826103615.932094-1-pulehui@huaweicloud= .com kernel/bpf/memalloc.c | 50 ++++++++++++++++++++++++------------------- 1 file changed, 28 insertions(+), 22 deletions(-) diff --git a/kernel/bpf/memalloc.c b/kernel/bpf/memalloc.c index e9662db7198f..8a8f088e83e6 100644 --- a/kernel/bpf/memalloc.c +++ b/kernel/bpf/memalloc.c @@ -119,6 +119,7 @@ struct bpf_mem_cache { struct llist_head waiting_for_gp_ttrace; struct rcu_head rcu_ttrace; atomic_t call_rcu_ttrace_in_progress; + raw_spinlock_t lock; }; =20 struct bpf_mem_caches { @@ -214,25 +215,24 @@ static void alloc_bulk(struct bpf_mem_cache *c, int c= nt, int node, bool atomic) gfp =3D __GFP_NOWARN | __GFP_ACCOUNT; gfp |=3D atomic ? GFP_NOWAIT : GFP_KERNEL; =20 - for (i =3D 0; i < cnt; i++) { - /* - * For every 'c' llist_del_first(&c->free_by_rcu_ttrace); is - * done only by one CPU =3D=3D current CPU. Other CPUs might - * llist_add() and llist_del_all() in parallel. - */ - obj =3D llist_del_first(&c->free_by_rcu_ttrace); - if (!obj) - break; - add_obj_to_free_list(c, obj); - } - if (i >=3D cnt) - return; + /* + * c->lock serializes concurrent llist_del_first() against + * llist_del_all() in __free_rcu() and do_call_rcu_ttrace(). + */ + scoped_guard(raw_spinlock_irqsave, &c->lock) { + for (i =3D 0; i < cnt; i++) { + obj =3D llist_del_first(&c->free_by_rcu_ttrace); + if (!obj) + break; + add_obj_to_free_list(c, obj); + } =20 - for (; i < cnt; i++) { - obj =3D llist_del_first(&c->waiting_for_gp_ttrace); - if (!obj) - break; - add_obj_to_free_list(c, obj); + for (; i < cnt; i++) { + obj =3D llist_del_first(&c->waiting_for_gp_ttrace); + if (!obj) + break; + add_obj_to_free_list(c, obj); + } } if (i >=3D cnt) return; @@ -279,8 +279,12 @@ static int free_all(struct bpf_mem_cache *c, struct ll= ist_node *llnode, bool per static void __free_rcu(struct rcu_head *head) { struct bpf_mem_cache *c =3D container_of(head, struct bpf_mem_cache, rcu_= ttrace); + struct llist_node *llnode; + + scoped_guard(raw_spinlock_irqsave, &c->lock) + llnode =3D llist_del_all(&c->waiting_for_gp_ttrace); =20 - free_all(c, llist_del_all(&c->waiting_for_gp_ttrace), !!c->percpu_size); + free_all(c, llnode, !!c->percpu_size); atomic_set(&c->call_rcu_ttrace_in_progress, 0); } =20 @@ -300,7 +304,8 @@ static void do_call_rcu_ttrace(struct bpf_mem_cache *c) =20 if (atomic_xchg(&c->call_rcu_ttrace_in_progress, 1)) { if (unlikely(READ_ONCE(c->draining))) { - llnode =3D llist_del_all(&c->free_by_rcu_ttrace); + scoped_guard(raw_spinlock_irqsave, &c->lock) + llnode =3D llist_del_all(&c->free_by_rcu_ttrace); free_all(c, llnode, !!c->percpu_size); } return; @@ -535,6 +540,7 @@ int bpf_mem_alloc_init(struct bpf_mem_alloc *ma, int si= ze, bool percpu) c->objcg =3D objcg; c->percpu_size =3D percpu_size; c->tgt =3D c; + raw_spin_lock_init(&c->lock); init_refill_work(c); prefill_mem_cache(c, cpu); } @@ -557,7 +563,7 @@ int bpf_mem_alloc_init(struct bpf_mem_alloc *ma, int si= ze, bool percpu) c->objcg =3D objcg; c->percpu_size =3D percpu_size; c->tgt =3D c; - + raw_spin_lock_init(&c->lock); init_refill_work(c); prefill_mem_cache(c, cpu); } @@ -609,7 +615,7 @@ int bpf_mem_alloc_percpu_unit_init(struct bpf_mem_alloc= *ma, int size) c->objcg =3D objcg; c->percpu_size =3D percpu_size; c->tgt =3D c; - + raw_spin_lock_init(&c->lock); init_refill_work(c); prefill_mem_cache(c, cpu); } --=20 2.34.1