From patchwork Tue Jun 21 08:56:14 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Pablo Neira Ayuso X-Patchwork-Id: 12888867 X-Patchwork-Delegate: kuba@kernel.org Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 6A74ACCA483 for ; Tue, 21 Jun 2022 08:56:33 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1347666AbiFUI4b (ORCPT ); Tue, 21 Jun 2022 04:56:31 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41430 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1346557AbiFUI42 (ORCPT ); Tue, 21 Jun 2022 04:56:28 -0400 Received: from mail.netfilter.org (mail.netfilter.org [217.70.188.207]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id 97CDA26551; Tue, 21 Jun 2022 01:56:27 -0700 (PDT) From: Pablo Neira Ayuso To: netfilter-devel@vger.kernel.org Cc: davem@davemloft.net, netdev@vger.kernel.org, kuba@kernel.org, pabeni@redhat.com, edumazet@google.com Subject: [PATCH net 1/5] netfilter: use get_random_u32 instead of prandom Date: Tue, 21 Jun 2022 10:56:14 +0200 Message-Id: <20220621085618.3975-2-pablo@netfilter.org> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20220621085618.3975-1-pablo@netfilter.org> References: <20220621085618.3975-1-pablo@netfilter.org> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org X-Patchwork-Delegate: kuba@kernel.org From: Florian Westphal bh might occur while updating per-cpu rnd_state from user context, ie. local_out path. BUG: using smp_processor_id() in preemptible [00000000] code: nginx/2725 caller is nft_ng_random_eval+0x24/0x54 [nft_numgen] Call Trace: check_preemption_disabled+0xde/0xe0 nft_ng_random_eval+0x24/0x54 [nft_numgen] Use the random driver instead, this also avoids need for local prandom state. Moreover, prandom now uses the random driver since d4150779e60f ("random32: use real rng for non-deterministic randomness"). Based on earlier patch from Pablo Neira. Fixes: 6b2faee0ca91 ("netfilter: nft_meta: place prandom handling in a helper") Fixes: 978d8f9055c3 ("netfilter: nft_numgen: add map lookups for numgen random operations") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_meta.c | 13 ++----------- net/netfilter/nft_numgen.c | 12 +++--------- 2 files changed, 5 insertions(+), 20 deletions(-) diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c index ac4859241e17..55d2d49c3425 100644 --- a/net/netfilter/nft_meta.c +++ b/net/netfilter/nft_meta.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -32,8 +33,6 @@ #define NFT_META_SECS_PER_DAY 86400 #define NFT_META_DAYS_PER_WEEK 7 -static DEFINE_PER_CPU(struct rnd_state, nft_prandom_state); - static u8 nft_meta_weekday(void) { time64_t secs = ktime_get_real_seconds(); @@ -271,13 +270,6 @@ static bool nft_meta_get_eval_ifname(enum nft_meta_keys key, u32 *dest, return true; } -static noinline u32 nft_prandom_u32(void) -{ - struct rnd_state *state = this_cpu_ptr(&nft_prandom_state); - - return prandom_u32_state(state); -} - #ifdef CONFIG_IP_ROUTE_CLASSID static noinline bool nft_meta_get_eval_rtclassid(const struct sk_buff *skb, u32 *dest) @@ -389,7 +381,7 @@ void nft_meta_get_eval(const struct nft_expr *expr, break; #endif case NFT_META_PRANDOM: - *dest = nft_prandom_u32(); + *dest = get_random_u32(); break; #ifdef CONFIG_XFRM case NFT_META_SECPATH: @@ -518,7 +510,6 @@ int nft_meta_get_init(const struct nft_ctx *ctx, len = IFNAMSIZ; break; case NFT_META_PRANDOM: - prandom_init_once(&nft_prandom_state); len = sizeof(u32); break; #ifdef CONFIG_XFRM diff --git a/net/netfilter/nft_numgen.c b/net/netfilter/nft_numgen.c index 81b40c663d86..45d3dc9e96f2 100644 --- a/net/netfilter/nft_numgen.c +++ b/net/netfilter/nft_numgen.c @@ -9,12 +9,11 @@ #include #include #include +#include #include #include #include -static DEFINE_PER_CPU(struct rnd_state, nft_numgen_prandom_state); - struct nft_ng_inc { u8 dreg; u32 modulus; @@ -135,12 +134,9 @@ struct nft_ng_random { u32 offset; }; -static u32 nft_ng_random_gen(struct nft_ng_random *priv) +static u32 nft_ng_random_gen(const struct nft_ng_random *priv) { - struct rnd_state *state = this_cpu_ptr(&nft_numgen_prandom_state); - - return reciprocal_scale(prandom_u32_state(state), priv->modulus) + - priv->offset; + return reciprocal_scale(get_random_u32(), priv->modulus) + priv->offset; } static void nft_ng_random_eval(const struct nft_expr *expr, @@ -168,8 +164,6 @@ static int nft_ng_random_init(const struct nft_ctx *ctx, if (priv->offset + priv->modulus - 1 < priv->offset) return -EOVERFLOW; - prandom_init_once(&nft_numgen_prandom_state); - return nft_parse_register_store(ctx, tb[NFTA_NG_DREG], &priv->dreg, NULL, NFT_DATA_VALUE, sizeof(u32)); } From patchwork Tue Jun 21 08:56:15 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Pablo Neira Ayuso X-Patchwork-Id: 12888868 X-Patchwork-Delegate: kuba@kernel.org Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 71372CCA480 for ; Tue, 21 Jun 2022 08:56:46 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1348453AbiFUI4n (ORCPT ); Tue, 21 Jun 2022 04:56:43 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41546 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1347109AbiFUI4b (ORCPT ); Tue, 21 Jun 2022 04:56:31 -0400 Received: from mail.netfilter.org (mail.netfilter.org [217.70.188.207]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id 7E03E26AC0; Tue, 21 Jun 2022 01:56:29 -0700 (PDT) From: Pablo Neira Ayuso To: netfilter-devel@vger.kernel.org Cc: davem@davemloft.net, netdev@vger.kernel.org, kuba@kernel.org, pabeni@redhat.com, edumazet@google.com Subject: [PATCH net 2/5] netfilter: cttimeout: fix slab-out-of-bounds read typo in cttimeout_net_exit Date: Tue, 21 Jun 2022 10:56:15 +0200 Message-Id: <20220621085618.3975-3-pablo@netfilter.org> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20220621085618.3975-1-pablo@netfilter.org> References: <20220621085618.3975-1-pablo@netfilter.org> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org X-Patchwork-Delegate: kuba@kernel.org From: Florian Westphal syzbot reports: BUG: KASAN: slab-out-of-bounds in __list_del_entry_valid+0xcc/0xf0 lib/list_debug.c:42 [..] list_del include/linux/list.h:148 [inline] cttimeout_net_exit+0x211/0x540 net/netfilter/nfnetlink_cttimeout.c:617 Problem is the wrong name of the list member, so container_of() result is wrong. Reported-by: Fixes: 78222bacfca9 ("netfilter: cttimeout: decouple unlink and free on netns destruction") Signed-off-by: Florian Westphal --- net/netfilter/nfnetlink_cttimeout.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c index af15102bc696..f466af4f8531 100644 --- a/net/netfilter/nfnetlink_cttimeout.c +++ b/net/netfilter/nfnetlink_cttimeout.c @@ -614,7 +614,7 @@ static void __net_exit cttimeout_net_exit(struct net *net) nf_ct_untimeout(net, NULL); - list_for_each_entry_safe(cur, tmp, &pernet->nfct_timeout_freelist, head) { + list_for_each_entry_safe(cur, tmp, &pernet->nfct_timeout_freelist, free_head) { list_del(&cur->free_head); if (refcount_dec_and_test(&cur->refcnt)) From patchwork Tue Jun 21 08:56:16 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Pablo Neira Ayuso X-Patchwork-Id: 12888871 X-Patchwork-Delegate: kuba@kernel.org Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id AC8D1CCA483 for ; Tue, 21 Jun 2022 08:56:50 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1347828AbiFUI4s (ORCPT ); Tue, 21 Jun 2022 04:56:48 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41534 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1347849AbiFUI4l (ORCPT ); Tue, 21 Jun 2022 04:56:41 -0400 Received: from mail.netfilter.org (mail.netfilter.org [217.70.188.207]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id 57E8D240BB; Tue, 21 Jun 2022 01:56:30 -0700 (PDT) From: Pablo Neira Ayuso To: netfilter-devel@vger.kernel.org Cc: davem@davemloft.net, netdev@vger.kernel.org, kuba@kernel.org, pabeni@redhat.com, edumazet@google.com Subject: [PATCH net 3/5] selftests: netfilter: correct PKTGEN_SCRIPT_PATHS in nft_concat_range.sh Date: Tue, 21 Jun 2022 10:56:16 +0200 Message-Id: <20220621085618.3975-4-pablo@netfilter.org> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20220621085618.3975-1-pablo@netfilter.org> References: <20220621085618.3975-1-pablo@netfilter.org> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org X-Patchwork-Delegate: kuba@kernel.org From: Jie2x Zhou Before change: make -C netfilter TEST: performance net,port [SKIP] perf not supported port,net [SKIP] perf not supported net6,port [SKIP] perf not supported port,proto [SKIP] perf not supported net6,port,mac [SKIP] perf not supported net6,port,mac,proto [SKIP] perf not supported net,mac [SKIP] perf not supported After change: net,mac [ OK ] baseline (drop from netdev hook): 2061098pps baseline hash (non-ranged entries): 1606741pps baseline rbtree (match on first field only): 1191607pps set with 1000 full, ranged entries: 1639119pps ok 8 selftests: netfilter: nft_concat_range.sh Fixes: 611973c1e06f ("selftests: netfilter: Introduce tests for sets with range concatenation") Reported-by: kernel test robot Signed-off-by: Jie2x Zhou Signed-off-by: Pablo Neira Ayuso --- tools/testing/selftests/netfilter/nft_concat_range.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/netfilter/nft_concat_range.sh b/tools/testing/selftests/netfilter/nft_concat_range.sh index b35010cc7f6a..a6991877e50c 100755 --- a/tools/testing/selftests/netfilter/nft_concat_range.sh +++ b/tools/testing/selftests/netfilter/nft_concat_range.sh @@ -31,7 +31,7 @@ BUGS="flush_remove_add reload" # List of possible paths to pktgen script from kernel tree for performance tests PKTGEN_SCRIPT_PATHS=" - ../../../samples/pktgen/pktgen_bench_xmit_mode_netif_receive.sh + ../../../../samples/pktgen/pktgen_bench_xmit_mode_netif_receive.sh pktgen/pktgen_bench_xmit_mode_netif_receive.sh" # Definition of set types: From patchwork Tue Jun 21 08:56:17 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Pablo Neira Ayuso X-Patchwork-Id: 12888869 X-Patchwork-Delegate: kuba@kernel.org Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 96E7FC43334 for ; Tue, 21 Jun 2022 08:56:47 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1347800AbiFUI4p (ORCPT ); Tue, 21 Jun 2022 04:56:45 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41500 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1346241AbiFUI4k (ORCPT ); Tue, 21 Jun 2022 04:56:40 -0400 Received: from mail.netfilter.org (mail.netfilter.org [217.70.188.207]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id DA8EA26ACB; Tue, 21 Jun 2022 01:56:29 -0700 (PDT) From: Pablo Neira Ayuso To: netfilter-devel@vger.kernel.org Cc: davem@davemloft.net, netdev@vger.kernel.org, kuba@kernel.org, pabeni@redhat.com, edumazet@google.com Subject: [PATCH net 4/5] netfilter: nf_dup_netdev: do not push mac header a second time Date: Tue, 21 Jun 2022 10:56:17 +0200 Message-Id: <20220621085618.3975-5-pablo@netfilter.org> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20220621085618.3975-1-pablo@netfilter.org> References: <20220621085618.3975-1-pablo@netfilter.org> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org X-Patchwork-Delegate: kuba@kernel.org From: Florian Westphal Eric reports skb_under_panic when using dup/fwd via bond+egress hook. Before pushing mac header, we should make sure that we're called from ingress to put back what was pulled earlier. In egress case, the MAC header is already there; we should leave skb alone. While at it be more careful here: skb might have been altered and headroom reduced, so add a skb_cow() before so that headroom is increased if necessary. nf_do_netdev_egress() assumes skb ownership (it normally ends with a call to dev_queue_xmit), so we must free the packet on error. Fixes: f87b9464d152 ("netfilter: nft_fwd_netdev: Support egress hook") Reported-by: Eric Garver Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_dup_netdev.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/net/netfilter/nf_dup_netdev.c b/net/netfilter/nf_dup_netdev.c index 7873bd1389c3..13b7f6a66086 100644 --- a/net/netfilter/nf_dup_netdev.c +++ b/net/netfilter/nf_dup_netdev.c @@ -13,10 +13,16 @@ #include #include -static void nf_do_netdev_egress(struct sk_buff *skb, struct net_device *dev) +static void nf_do_netdev_egress(struct sk_buff *skb, struct net_device *dev, + enum nf_dev_hooks hook) { - if (skb_mac_header_was_set(skb)) + if (hook == NF_NETDEV_INGRESS && skb_mac_header_was_set(skb)) { + if (skb_cow_head(skb, skb->mac_len)) { + kfree_skb(skb); + return; + } skb_push(skb, skb->mac_len); + } skb->dev = dev; skb_clear_tstamp(skb); @@ -33,7 +39,7 @@ void nf_fwd_netdev_egress(const struct nft_pktinfo *pkt, int oif) return; } - nf_do_netdev_egress(pkt->skb, dev); + nf_do_netdev_egress(pkt->skb, dev, nft_hook(pkt)); } EXPORT_SYMBOL_GPL(nf_fwd_netdev_egress); @@ -48,7 +54,7 @@ void nf_dup_netdev_egress(const struct nft_pktinfo *pkt, int oif) skb = skb_clone(pkt->skb, GFP_ATOMIC); if (skb) - nf_do_netdev_egress(skb, dev); + nf_do_netdev_egress(skb, dev, nft_hook(pkt)); } EXPORT_SYMBOL_GPL(nf_dup_netdev_egress); From patchwork Tue Jun 21 08:56:18 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Pablo Neira Ayuso X-Patchwork-Id: 12888870 X-Patchwork-Delegate: kuba@kernel.org Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 0D4D1C433EF for ; Tue, 21 Jun 2022 08:56:49 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1348072AbiFUI4r (ORCPT ); Tue, 21 Jun 2022 04:56:47 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41764 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1348086AbiFUI4m (ORCPT ); Tue, 21 Jun 2022 04:56:42 -0400 Received: from mail.netfilter.org (mail.netfilter.org [217.70.188.207]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id 868C426109; Tue, 21 Jun 2022 01:56:31 -0700 (PDT) From: Pablo Neira Ayuso To: netfilter-devel@vger.kernel.org Cc: davem@davemloft.net, netdev@vger.kernel.org, kuba@kernel.org, pabeni@redhat.com, edumazet@google.com Subject: [PATCH net 5/5] netfilter: nf_dup_netdev: add and use recursion counter Date: Tue, 21 Jun 2022 10:56:18 +0200 Message-Id: <20220621085618.3975-6-pablo@netfilter.org> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20220621085618.3975-1-pablo@netfilter.org> References: <20220621085618.3975-1-pablo@netfilter.org> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org X-Patchwork-Delegate: kuba@kernel.org From: Florian Westphal Now that the egress function can be called from egress hook, we need to avoid recursive calls into the nf_tables traverser, else crash. Fixes: f87b9464d152 ("netfilter: nft_fwd_netdev: Support egress hook") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_dup_netdev.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-) diff --git a/net/netfilter/nf_dup_netdev.c b/net/netfilter/nf_dup_netdev.c index 13b7f6a66086..a8e2425e43b0 100644 --- a/net/netfilter/nf_dup_netdev.c +++ b/net/netfilter/nf_dup_netdev.c @@ -13,20 +13,31 @@ #include #include +#define NF_RECURSION_LIMIT 2 + +static DEFINE_PER_CPU(u8, nf_dup_skb_recursion); + static void nf_do_netdev_egress(struct sk_buff *skb, struct net_device *dev, enum nf_dev_hooks hook) { + if (__this_cpu_read(nf_dup_skb_recursion) > NF_RECURSION_LIMIT) + goto err; + if (hook == NF_NETDEV_INGRESS && skb_mac_header_was_set(skb)) { - if (skb_cow_head(skb, skb->mac_len)) { - kfree_skb(skb); - return; - } + if (skb_cow_head(skb, skb->mac_len)) + goto err; + skb_push(skb, skb->mac_len); } skb->dev = dev; skb_clear_tstamp(skb); + __this_cpu_inc(nf_dup_skb_recursion); dev_queue_xmit(skb); + __this_cpu_dec(nf_dup_skb_recursion); + return; +err: + kfree_skb(skb); } void nf_fwd_netdev_egress(const struct nft_pktinfo *pkt, int oif)