From mboxrd@z Thu Jan 1 00:00:00 1970 Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.remote-ip=100.103.45.18 ARC-Seal: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; cv=none; t=1793827008; b=CjmiJKyNlNWfxuEStKTBbwE9iXHhhS9ADUuL+xfutAXbkTMUi0re3I37UwihC1gDw1zf cGp98fFAnwVZLJjz/8MPxbEUb0oq/+Sb+M39eEwpL/oaeIT+8m1UmViO7EehzUkWEIfDy LCDwTn/bd4xPzXQBX/d43XQ9k2qkx6l5E1DQP3tJYM6u2mAtKiDcp1p68/UubYa9TKeL2 Vbbhn9QTpE6Eyp7GnSXYiIizncMBZ/iU4J29gLhr2OYdzFVMoxQcrgpigsSjqCtNK0u9s 6f4OHx1+edHrVAiB/XAE1fbzChdBiX6rtQReky+gbNyWp3GYj3H5PbNChmxGL5e5GIQ== ARC-Message-Signature: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; c=relaxed/relaxed; t=1793827008; h=Received:DKIM-Signature:From:To:Cc:Subject:Date:Message-ID:X-Mailer: In-Reply-To:References:User-Agent:X-stable:X-Patchwork-Hint: MIME-Version:Content-Transfer-Encoding; bh=KeoAW+s9ihLxyyeN48y06pT3oO30GsPKMJqL/WK7i/4=; b=u7yjOwONZvh/A23DX3kNCS3f7bM5amcWdaKN7v+WziNRXyG4qy2+SEWWbv1CCivGnSKu L8H9taW4dEY5ntR2vYrqCFoCz8hZodwwNnfjgEbgK2jDfY+gUgbyovfmff9PoNHKAviGi cvrFclAXjt7UOZM5P5c7EhWJEEw6nFmDpxe/VU419eB1bNzqkYqfGKF3MWtAF9SVghea4 9SSV7m0Nqq4uflMrS0qEr90Uyi4YdzLZ3k/q2vDrcAVsvb33hyNtS9l2zX3wmcWKScRwq lMNSaXlS9AfPQnypj/Xd8RGd5Kt5waY3wpcwvStkytEnFMoPzZ7c/CJ9Yr2Ua0fPafg== ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass header.d=linuxfoundation.org header.i=@linuxfoundation.org header.a=rsa-sha256 header.s=korg header.b="kuPmay/y"; arc=none smtp.remote-ip=100.103.45.18 Received: from smtp.kernel.org (quasi.space.kernel.org [100.103.45.18]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (prime256v1)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 3539E607492C; Wed, 4 Nov 2026 21:16:46 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key, unprotected) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.a=rsa-sha256 header.s=korg header.b=kuPmay/y Received: by smtp.kernel.org (Postfix) with ESMTPSA id 8E9DC1F00A02; Wed, 4 Nov 2026 21:16:46 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linuxfoundation.org; s=korg; t=1793827006; bh=KeoAW+s9ihLxyyeN48y06pT3oO30GsPKMJqL/WK7i/4=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=kuPmay/y9PxycWK0m0PJINpi4HW/8a9vNl7XPHISjXM6YK+TVfUsnH8HLjNkxl1Ls gnkj9wek2nvxuZlCmW8q48TWl1foFgm3u77zEJ02im1lDu08lpxw9WCnDjnZ3pK/hy 5T9nhZkhAPprD0DslQYloS/DtVsTlHNL2f4lwLaE= From: Greg Kroah-Hartman To: stable@vger.kernel.org Cc: Greg Kroah-Hartman , patches@lists.linux.dev, Xin Liu , Yucheng Lu , Yuhang Zheng , Ren Wei , Jakub Kicinski Subject: [PATCH 7.3 099/547] net/sched: cls_flower: avoid stale mask references after delete Date: Wed, 4 Nov 2026 16:04:43 -0500 Message-ID: <20261104211041.037570832@linuxfoundation.org> X-Mailer: git-send-email 2.56.0 In-Reply-To: <20261104211038.406938024@linuxfoundation.org> References: <20261104211038.406938024@linuxfoundation.org> User-Agent: quilt/0.69 X-stable: review X-Patchwork-Hint: ignore X-Mailing-List: patches@lists.linux.dev List-Id: List-Post: List-Help: List-Subscribe: List-Unsubscribe: List-Owner: Precedence: list MIME-Version: 1.0 Content-Transfer-Encoding: 8bit 7.3-stable review patch. If anyone has any objections, please let me know. ------------------ From: Yuhang Zheng commit 58f7989da6e715ce834ac861b7ba6ae9f9bdb098 upstream. Flower releases a mask reference when a filter is removed from the classifier. The filter object can remain alive while dump, walk, or hardware reoffload paths hold a reference to it. Those paths can then dereference f->mask after the mask has already been released. Keep the mask reference until the filter reference count reaches zero. Release the mask synchronously from the final __fl_put() while the caller still holds the tcf_proto reference, and leave the existing RCU work to free the mask. This keeps the mask available to every reader that still holds the filter. The filter is still removed from the classifier indexes immediately. Therefore, a deleted filter can temporarily retain an otherwise empty mask; fl_delete() uses the handle IDR, rather than the mask list, to determine whether the classifier has become empty. Failed mask assignment can leave fnew->mask as an error or borrowed pointer without an owned mask reference. Clear it before the common failure path so the final put cannot release an unowned reference. This fix does not add fields, snapshots, or per-filter memory. Fixes: 92149190067d ("net: sched: flower: set unlocked flag for flower proto ops") Cc: stable@vger.kernel.org Reported-by: Xin Liu Tested-by: Yucheng Lu Signed-off-by: Yuhang Zheng Signed-off-by: Ren Wei Link: https://patch.msgid.link/20261012010720.2248662-2-n05ec@lzu.edu.cn Signed-off-by: Jakub Kicinski Signed-off-by: Greg Kroah-Hartman --- net/sched/cls_flower.c | 50 +++++++++++++++++++++++++++---------------------- 1 file changed, 28 insertions(+), 22 deletions(-) --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -518,15 +518,22 @@ static void fl_hw_update_stats(struct tc tcf_exts_hw_stats_update(&f->exts, &cls_flower.stats, cls_flower.use_act_stats); } -static void __fl_put(struct cls_fl_filter *f) +static bool __fl_put(struct tcf_proto *tp, struct cls_fl_filter *f) { + bool last = false; + if (!refcount_dec_and_test(&f->refcnt)) - return; + return false; + + if (f->mask) + last = fl_mask_put(fl_head_dereference(tp), f->mask); if (tcf_exts_get_net(&f->exts)) tcf_queue_work(&f->rwork, fl_destroy_filter_work); else __fl_destroy_filter(f); + + return last; } static struct cls_fl_filter *__fl_get(struct cls_fl_head *head, u32 handle) @@ -572,11 +579,10 @@ static int __fl_delete(struct tcf_proto list_del_rcu(&f->list); spin_unlock(&tp->lock); - *last = fl_mask_put(head, f->mask); if (!tc_skip_hw(f->flags)) fl_hw_destroy_filter(tp, f, rtnl_held, extack); tcf_unbind_filter(tp, &f->res); - __fl_put(f); + *last = __fl_put(tp, f); return 0; } @@ -617,7 +623,7 @@ static void fl_put(struct tcf_proto *tp, { struct cls_fl_filter *f = arg; - __fl_put(f); + __fl_put(tp, f); } static void *fl_get(struct tcf_proto *tp, u32 handle) @@ -2488,12 +2494,14 @@ static int fl_change(struct net *net, st } err = fl_check_assign_mask(head, fnew, fold, mask); - if (err) + if (err) { + fnew->mask = NULL; goto unbind_filter; + } err = fl_ht_insert_unique(fnew, fold, &in_ht); if (err) - goto errout_mask; + goto unbind_filter; if (!tc_skip_hw(fnew->flags)) { err = fl_hw_replace_filter(tp, fnew, rtnl_held, extack); @@ -2547,7 +2555,6 @@ static int fl_change(struct net *net, st spin_unlock(&tp->lock); - fl_mask_put(head, fold->mask); if (!tc_skip_hw(fold->flags)) fl_hw_destroy_filter(tp, fold, rtnl_held, NULL); tcf_unbind_filter(tp, &fold->res); @@ -2555,7 +2562,7 @@ static int fl_change(struct net *net, st * after this. */ refcount_dec(&fold->refcnt); - __fl_put(fold); + __fl_put(tp, fold); } else { idr_replace(&head->handle_idr, fnew, fnew->handle); @@ -2580,9 +2587,6 @@ errout_hw: if (in_ht) rhashtable_remove_fast(&fnew->mask->ht, &fnew->ht_node, fnew->mask->filter_ht_params); -errout_mask: - fl_mask_put(head, fnew->mask); - unbind_filter: if (bound_to_filter) { if (flags & TCA_ACT_FLAGS_NO_RTNL) @@ -2598,14 +2602,14 @@ errout_idr: idr_remove(&head->handle_idr, fnew->handle); spin_unlock(&tp->lock); } - __fl_put(fnew); + __fl_put(tp, fnew); errout_tb: kfree(tb); errout_mask_alloc: tcf_queue_work(&mask->rwork, fl_uninit_mask_free_work); errout_fold: if (fold) - __fl_put(fold); + __fl_put(tp, fold); return err; } @@ -2618,8 +2622,10 @@ static int fl_delete(struct tcf_proto *t int err = 0; err = __fl_delete(tp, f, &last_on_mask, rtnl_held, extack); - *last = list_empty(&head->masks); - __fl_put(f); + spin_lock(&tp->lock); + *last = idr_is_empty(&head->handle_idr); + spin_unlock(&tp->lock); + __fl_put(tp, f); return err; } @@ -2641,12 +2647,12 @@ static void fl_walk(struct tcf_proto *tp rcu_read_unlock(); if (arg->fn(tp, f, arg) < 0) { - __fl_put(f); + __fl_put(tp, f); arg->stop = 1; rcu_read_lock(); break; } - __fl_put(f); + __fl_put(tp, f); arg->count++; rcu_read_lock(); } @@ -2697,7 +2703,7 @@ static int fl_reoffload(struct tcf_proto cls_flower.rule = flow_rule_alloc(tcf_exts_num_actions(&f->exts)); if (!cls_flower.rule) { - __fl_put(f); + __fl_put(tp, f); return -ENOMEM; } @@ -2715,7 +2721,7 @@ static int fl_reoffload(struct tcf_proto if (err) { kfree(cls_flower.rule); if (tc_skip_sw(f->flags)) { - __fl_put(f); + __fl_put(tp, f); return err; } goto next_flow; @@ -2731,11 +2737,11 @@ static int fl_reoffload(struct tcf_proto kfree(cls_flower.rule); if (err) { - __fl_put(f); + __fl_put(tp, f); return err; } next_flow: - __fl_put(f); + __fl_put(tp, f); } return 0;