From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (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 8809B509F07; Tue, 13 Oct 2026 12:11:41 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1791893504; cv=none; b=dIib7OueLNykkguQyFH5diWE2TAQau/sibUWoZeqmS+YWKTUnRzAPsDNNldyeHK8qcLGBh2FEegNM3RiiFOCeLt7tEDldTbM87lOLaY6G4ySKHCcj5HP3kfiMOuXV90xv8TrkBEvnPRfVXYplpdQj8Zx3E+AStetuaWa7662q18= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1791893504; c=relaxed/simple; bh=ML6LiziaYhnfNQifjFMAYWS6qSrTvqxDN+aYZzLc4YA=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=N+N0b/NGvszOItKPF70EUSI0as11SZVDOX/2cSUYn6jDPX44BXMwZ+CRuz67fuJ+maifyS47KCJkJx+DrVvVAP+rJpRM6ZPQC1Nnv6eC0thYkyX0/elXAmx+fVI9+Lq5PnluM6Pg+vZ18lJrbkwNYy4usgy1/vkslUeTZdnaK2Y= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b=1iUp6/93; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b="1iUp6/93" Received: by smtp.kernel.org (Postfix) with ESMTPSA id D89141F000FF; Tue, 13 Oct 2026 12:11:40 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linuxfoundation.org; s=korg; t=1791893501; bh=C2HnY7bjlrmO0OS3zno5qM/Pe2f9hiiyHl22eJyQOYY=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=1iUp6/93/RdpjCXLYgOFNH9P8ArPx7QvHorIvcZdlumTCDqWAFIiUIc/AUSYn732Z mb0DffxSFYROaBzprWOYzA6Vq+qCrRUDlrtKUsyJKnOGKawMQmj/b8ouNUZMJGIPbr eI7VS5YrtBzF0w/K+CFJMechynPKBO3VS67eiwLM= From: Greg Kroah-Hartman To: stable@vger.kernel.org Cc: Greg Kroah-Hartman , patches@lists.linux.dev, Tristan Madani , Florian Westphal , Pablo Neira Ayuso , Sasha Levin Subject: [PATCH 6.18 245/621] netfilter: x_tables: allocate hook ops while under mutex Date: Tue, 13 Oct 2026 13:34:00 +0200 Message-ID: <20261013112926.676575145@linuxfoundation.org> X-Mailer: git-send-email 2.56.0 In-Reply-To: <20261013112920.855194808@linuxfoundation.org> References: <20261013112920.855194808@linuxfoundation.org> User-Agent: quilt/0.69 X-stable: review X-Patchwork-Hint: ignore Precedence: bulk X-Mailing-List: patches@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit 6.18-stable review patch. If anyone has any objections, please let me know. ------------------ From: Florian Westphal [ Upstream commit b62eb8dcf2c47d4d676a434efbd57c4f776f7829 ] arp/ip(6)t_register_table() add the table to the per-netns list via xt_register_table() before allocating the per-netns hook ops copy via kmemdup_array(). This leaves a window where the table is visible in the list with ops=NULL. If the pernet exit happens runs concurrently the pre_exit callback finds the table via xt_find_table() and passes the NULL ops pointer to nf_unregister_net_hooks(), causing a NULL dereference: general protection fault in nf_unregister_net_hooks+0xbc/0x150 RIP: nf_unregister_net_hooks (net/netfilter/core.c:613) Call Trace: ipt_unregister_table_pre_exit iptable_mangle_net_pre_exit ops_pre_exit_list cleanup_net Fix by moving the ops allocation into the xtables core so the table is never in the list without valid ops. Also ensure the table is no longer processing packets before its torn down on error unwind. nf_register_net_hooks might have published at least one hook; call synchronize_rcu() if there was an error. audit log register message gets deferred until all operations have passed, this avoids need to emit another ureg message in case of error unwinding. Based on earlier patch by Tristan Madani. Fixes: f9006acc8dfe5 ("netfilter: arp_tables: pass table pointer via nf_hook_ops") Fixes: ee177a54413a ("netfilter: ip6_tables: pass table pointer via nf_hook_ops") Fixes: ae689334225f ("netfilter: ip_tables: pass table pointer via nf_hook_ops") Link: https://lore.kernel.org/netfilter-devel/20260429175613.1459342-1-tristmd@gmail.com/ Signed-off-by: Tristan Madani Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso Signed-off-by: Sasha Levin --- include/linux/netfilter/x_tables.h | 1 + net/ipv4/netfilter/arp_tables.c | 35 +++------------------ net/ipv4/netfilter/ip_tables.c | 41 +++--------------------- net/ipv6/netfilter/ip6_tables.c | 38 +++-------------------- net/netfilter/x_tables.c | 50 +++++++++++++++++++++++++----- 5 files changed, 55 insertions(+), 110 deletions(-) diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index 3aef60abd362c..6fd365f7b35b0 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -297,6 +297,7 @@ struct xt_counters *xt_counters_alloc(unsigned int counters); struct xt_table *xt_register_table(struct net *net, const struct xt_table *table, + const struct nf_hook_ops *template_ops, struct xt_table_info *bootstrap, struct xt_table_info *newinfo); void xt_unregister_table_pre_exit(struct net *net, u8 af, const char *name); diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 1490466b146e4..0ea513bf77fb6 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -1518,13 +1518,11 @@ int arpt_register_table(struct net *net, const struct arpt_replace *repl, const struct nf_hook_ops *template_ops) { - struct nf_hook_ops *ops; - unsigned int num_ops; - int ret, i; - struct xt_table_info *newinfo; struct xt_table_info bootstrap = {0}; - void *loc_cpu_entry; + struct xt_table_info *newinfo; struct xt_table *new_table; + void *loc_cpu_entry; + int ret; newinfo = xt_alloc_table_info(repl->size); if (!newinfo) @@ -1539,7 +1537,7 @@ int arpt_register_table(struct net *net, return ret; } - new_table = xt_register_table(net, table, &bootstrap, newinfo); + new_table = xt_register_table(net, table, template_ops, &bootstrap, newinfo); if (IS_ERR(new_table)) { struct arpt_entry *iter; @@ -1549,31 +1547,6 @@ int arpt_register_table(struct net *net, return PTR_ERR(new_table); } - num_ops = hweight32(table->valid_hooks); - if (num_ops == 0) { - ret = -EINVAL; - goto out_free; - } - - ops = kmemdup_array(template_ops, num_ops, sizeof(*ops), GFP_KERNEL); - if (!ops) { - ret = -ENOMEM; - goto out_free; - } - - for (i = 0; i < num_ops; i++) - ops[i].priv = new_table; - - new_table->ops = ops; - - ret = nf_register_net_hooks(net, ops, num_ops); - if (ret != 0) - goto out_free; - - return ret; - -out_free: - __arpt_unregister_table(net, new_table); return ret; } diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 0549fad53c036..ca8ff0ae6cdb9 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -1720,13 +1720,11 @@ int ipt_register_table(struct net *net, const struct xt_table *table, const struct ipt_replace *repl, const struct nf_hook_ops *template_ops) { - struct nf_hook_ops *ops; - unsigned int num_ops; - int ret, i; - struct xt_table_info *newinfo; struct xt_table_info bootstrap = {0}; - void *loc_cpu_entry; + struct xt_table_info *newinfo; struct xt_table *new_table; + void *loc_cpu_entry; + int ret; newinfo = xt_alloc_table_info(repl->size); if (!newinfo) @@ -1741,7 +1739,7 @@ int ipt_register_table(struct net *net, const struct xt_table *table, return ret; } - new_table = xt_register_table(net, table, &bootstrap, newinfo); + new_table = xt_register_table(net, table, template_ops, &bootstrap, newinfo); if (IS_ERR(new_table)) { struct ipt_entry *iter; @@ -1751,37 +1749,6 @@ int ipt_register_table(struct net *net, const struct xt_table *table, return PTR_ERR(new_table); } - /* No template? No need to do anything. This is used by 'nat' table, it registers - * with the nat core instead of the netfilter core. - */ - if (!template_ops) - return 0; - - num_ops = hweight32(table->valid_hooks); - if (num_ops == 0) { - ret = -EINVAL; - goto out_free; - } - - ops = kmemdup_array(template_ops, num_ops, sizeof(*ops), GFP_KERNEL); - if (!ops) { - ret = -ENOMEM; - goto out_free; - } - - for (i = 0; i < num_ops; i++) - ops[i].priv = new_table; - - new_table->ops = ops; - - ret = nf_register_net_hooks(net, ops, num_ops); - if (ret != 0) - goto out_free; - - return ret; - -out_free: - __ipt_unregister_table(net, new_table); return ret; } diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 7efc2a19662f9..22db263682bda 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -1734,13 +1734,11 @@ int ip6t_register_table(struct net *net, const struct xt_table *table, const struct ip6t_replace *repl, const struct nf_hook_ops *template_ops) { - struct nf_hook_ops *ops; - unsigned int num_ops; - int ret, i; - struct xt_table_info *newinfo; struct xt_table_info bootstrap = {0}; - void *loc_cpu_entry; + struct xt_table_info *newinfo; struct xt_table *new_table; + void *loc_cpu_entry; + int ret; newinfo = xt_alloc_table_info(repl->size); if (!newinfo) @@ -1755,7 +1753,7 @@ int ip6t_register_table(struct net *net, const struct xt_table *table, return ret; } - new_table = xt_register_table(net, table, &bootstrap, newinfo); + new_table = xt_register_table(net, table, template_ops, &bootstrap, newinfo); if (IS_ERR(new_table)) { struct ip6t_entry *iter; @@ -1765,34 +1763,6 @@ int ip6t_register_table(struct net *net, const struct xt_table *table, return PTR_ERR(new_table); } - if (!template_ops) - return 0; - - num_ops = hweight32(table->valid_hooks); - if (num_ops == 0) { - ret = -EINVAL; - goto out_free; - } - - ops = kmemdup_array(template_ops, num_ops, sizeof(*ops), GFP_KERNEL); - if (!ops) { - ret = -ENOMEM; - goto out_free; - } - - for (i = 0; i < num_ops; i++) - ops[i].priv = new_table; - - new_table->ops = ops; - - ret = nf_register_net_hooks(net, ops, num_ops); - if (ret != 0) - goto out_free; - - return ret; - -out_free: - __ip6t_unregister_table(net, new_table); return ret; } diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index 492b1086157b2..f4b32b6c70bc6 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -1478,7 +1478,6 @@ xt_replace_table(struct xt_table *table, unsigned int num_counters, private = do_replace_table(table, num_counters, newinfo, error); if (private) audit_log_nfcfg(table->name, table->af, private->number, - !private->number ? AUDIT_XT_OP_REGISTER : AUDIT_XT_OP_REPLACE, GFP_KERNEL); @@ -1488,20 +1487,32 @@ EXPORT_SYMBOL_GPL(xt_replace_table); struct xt_table *xt_register_table(struct net *net, const struct xt_table *input_table, + const struct nf_hook_ops *template_ops, struct xt_table_info *bootstrap, struct xt_table_info *newinfo) { struct xt_pernet *xt_net = net_generic(net, xt_pernet_id); + struct xt_table *t, *table = NULL; + struct nf_hook_ops *ops = NULL; struct xt_table_info *private; - struct xt_table *t, *table; - int ret; + unsigned int num_ops; + int ret = -EINVAL; + + num_ops = hweight32(input_table->valid_hooks); + if (num_ops == 0) + goto out; + + ret = -ENOMEM; + if (template_ops) { + ops = kmemdup_array(template_ops, num_ops, sizeof(*ops), GFP_KERNEL); + if (!ops) + goto out; + } /* Don't add one object to multiple lists. */ table = kmemdup(input_table, sizeof(struct xt_table), GFP_KERNEL); - if (!table) { - ret = -ENOMEM; + if (!table) goto out; - } mutex_lock(&xt[table->af].mutex); /* Don't autoload: we'd eat our tail... */ @@ -1515,7 +1526,7 @@ struct xt_table *xt_register_table(struct net *net, /* Simplifies replace_table code. */ table->private = bootstrap; - if (!xt_replace_table(table, 0, newinfo, &ret)) + if (!do_replace_table(table, 0, newinfo, &ret)) goto unlock; private = table->private; @@ -1524,14 +1535,37 @@ struct xt_table *xt_register_table(struct net *net, /* save number of initial entries */ private->initial_entries = private->number; + if (ops) { + int i; + + for (i = 0; i < num_ops; i++) + ops[i].priv = table; + + ret = nf_register_net_hooks(net, ops, num_ops); + if (ret != 0) { + mutex_unlock(&xt[table->af].mutex); + /* nf_register_net_hooks() might have published a + * base chain before internal error unwind. + */ + synchronize_rcu(); + goto out; + } + + table->ops = ops; + } + + audit_log_nfcfg(table->name, table->af, private->number, + AUDIT_XT_OP_REGISTER, GFP_KERNEL); + list_add(&table->list, &xt_net->tables[table->af]); mutex_unlock(&xt[table->af].mutex); return table; unlock: mutex_unlock(&xt[table->af].mutex); - kfree(table); out: + kfree(table); + kfree(ops); return ERR_PTR(ret); } EXPORT_SYMBOL_GPL(xt_register_table); -- 2.53.0