From patchwork Wed Aug 24 04:41:16 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Shmulik Ladkani X-Patchwork-Id: 12952903 X-Patchwork-Delegate: bpf@iogearbox.net 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 DBCBCC38145 for ; Wed, 24 Aug 2022 04:41:37 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S234479AbiHXElg (ORCPT ); Wed, 24 Aug 2022 00:41:36 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:48268 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S234533AbiHXEle (ORCPT ); Wed, 24 Aug 2022 00:41:34 -0400 Received: from mail-wm1-x330.google.com (mail-wm1-x330.google.com [IPv6:2a00:1450:4864:20::330]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 94B5A8C456 for ; Tue, 23 Aug 2022 21:41:33 -0700 (PDT) Received: by mail-wm1-x330.google.com with SMTP id d5so8160033wms.5 for ; Tue, 23 Aug 2022 21:41:33 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=metanetworks.com; s=google; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc; bh=0NOhQ+9Eg1hYodEfeL4gY+mnmGuJvDz8XKjUQSHxK6g=; b=Yjuebj+YhyDRyRFpl3rlNEoWoSGqxs6V/h0zoDuNJtJn+hJeUT0+20dBhHeYBK5+Y+ t+9RCDiBtiOrP+BP7t98k16rYnyhYP2a8Cei+UY6Zy85L84AGL8Fi5xcK0i0DERZEA44 bdf4yJbYccbsbX33JxEuCs4Xr1UtMqxn5Etk8= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc; bh=0NOhQ+9Eg1hYodEfeL4gY+mnmGuJvDz8XKjUQSHxK6g=; b=TmJg5YKMjnGVexZNC/xu/NXrtBjL7TaKzt5x73j7a0MVTpFoo6f8gteb8gbpTvXb6J dUb+bFTt4X3bNrIdheAa64Zf7gbUGsZ8VlqbxZ96+Zf+H0DQ6RNBeHHx1GyNzo4EKikQ 5iJbOvbyP7Z0wYPLdn2+0xsNA5sMwhng078ZHkO3xmGiSXjRQA/ivg+LT5qCjgBfodLb vuTtWW1sU+/ZrZv8kp4rmDcsEZaqr2hwPE0t6XtoP4ZCSCM41wVX24gGaNA5QiYeRXiA dPpB9TyMWWL3djTVitlEMAKJM3xRnSzP/+vS0M8x2kz2Q+5DWknr5OHsfzLQmvcpEZ5Z 737Q== X-Gm-Message-State: ACgBeo33qKxt94DXK7WwI0DCTlUgW+xezD73QAkh2bGIRTj+A9qr/QIK gtttWe7iBLLho3vJHgo7tugZauLB9/bzW5Ac9NlZkgQXTY6fk291LYm5LojpB/2VrhI1GvIw5ha GoxbLdn0j+OiRL1bG6z33yasqD95EXKSVh6v1zytcwb6Tw2bRm0bWu6GbavPJ5UkG2zWEtLJT X-Google-Smtp-Source: AA6agR5KEwo2goWQhkJi4bf1SjIDfZRrgsN4uGc5S0n1oVsDZzi/5rGc8ouNUtz51192LDzaxbNK2g== X-Received: by 2002:a05:600c:3b16:b0:3a6:169:d7f8 with SMTP id m22-20020a05600c3b1600b003a60169d7f8mr3924332wms.129.1661316091773; Tue, 23 Aug 2022 21:41:31 -0700 (PDT) Received: from blondie.home ([94.230.83.151]) by smtp.gmail.com with ESMTPSA id m9-20020adfe0c9000000b00225206dd595sm15572735wri.86.2022.08.23.21.41.30 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 23 Aug 2022 21:41:31 -0700 (PDT) From: Shmulik Ladkani X-Google-Original-From: Shmulik Ladkani To: bpf@vger.kernel.org, Alexei Starovoitov , Daniel Borkmann , John Fastabend , Joanne Koong Cc: Andrii Nakryiko , Paul Chaignon , Shmulik Ladkani Subject: [PATCH v5 bpf-next 3/4] selftests/bpf: Simplify test_tunnel setup for allowing non-local tunnel traffic Date: Wed, 24 Aug 2022 07:41:16 +0300 Message-Id: <20220824044117.137658-4-shmulik.ladkani@gmail.com> X-Mailer: git-send-email 2.37.2 In-Reply-To: <20220824044117.137658-1-shmulik.ladkani@gmail.com> References: <20220824044117.137658-1-shmulik.ladkani@gmail.com> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: bpf@vger.kernel.org X-Patchwork-Delegate: bpf@iogearbox.net Commit 1115169f47ae ("selftests/bpf: Don't assign outer source IP to host") removed the secondary IP (IP4_ADDR2_VETH1) assigned to veth1, in order to test bpf_skb_set_tunnel_key's functionality when tunnel destination isn't assigned to an interface. The chosen setup for testing the "tunnel to unassigned outer IP" scenario was rather complex: (1) static ARP entries in order to bypass ARP (o/w requests will fail as the target address isn't assigned locally), and (2) a BPF program running on veth1 ingress which manipulates the IP header's daddr to the actual IP assigned to the interface (o/w tunnel traffic won't be accepted locally). This is complex, and adds a dependency on this hidden "dnat"-like eBPF program, that needs to be replicated when new tunnel tests are added. Instead, we can have a much simpler setup: Add the secondary IP as a *local route* in a table pointed by a custom fib rule. No static arp entries are needed, and the special eBPF program that "dnats" the outer destination can be removed. This commit is a revert of 1115169f47ae, with the addition of the local route of IP4_ADDR2_VETH1 (instead of the original address assignment). Signed-off-by: Shmulik Ladkani --- v2: Place the local route for the secondary IP in a custom table pointed by a custom fib rule; this ensures the IP is not considered assigned to a device. --- .../selftests/bpf/prog_tests/test_tunnel.c | 23 ++---- .../selftests/bpf/progs/test_tunnel_kern.c | 80 +++---------------- 2 files changed, 17 insertions(+), 86 deletions(-) diff --git a/tools/testing/selftests/bpf/prog_tests/test_tunnel.c b/tools/testing/selftests/bpf/prog_tests/test_tunnel.c index eea274110267..852da04ff281 100644 --- a/tools/testing/selftests/bpf/prog_tests/test_tunnel.c +++ b/tools/testing/selftests/bpf/prog_tests/test_tunnel.c @@ -82,7 +82,6 @@ #define MAC_TUNL_DEV0 "52:54:00:d9:01:00" #define MAC_TUNL_DEV1 "52:54:00:d9:02:00" -#define MAC_VETH1 "52:54:00:d9:03:00" #define VXLAN_TUNL_DEV0 "vxlan00" #define VXLAN_TUNL_DEV1 "vxlan11" @@ -109,9 +108,15 @@ static int config_device(void) { SYS("ip netns add at_ns0"); - SYS("ip link add veth0 address " MAC_VETH1 " type veth peer name veth1"); + SYS("ip link add veth0 type veth peer name veth1"); SYS("ip link set veth0 netns at_ns0"); SYS("ip addr add " IP4_ADDR1_VETH1 "/24 dev veth1"); + /* Create a custom rule routing IP4_ADDR2_VETH1 as local. + * Do not place it in "local" table, to avoid this IP being considered + * assigned to a device. + */ + SYS("ip rule add to " IP4_ADDR2_VETH1 " table 20"); + SYS("ip route add local " IP4_ADDR2_VETH1 "/32 dev veth1 table 20"); SYS("ip link set dev veth1 up mtu 1500"); SYS("ip netns exec at_ns0 ip addr add " IP4_ADDR_VETH0 "/24 dev veth0"); SYS("ip netns exec at_ns0 ip link set dev veth0 up mtu 1500"); @@ -125,6 +130,7 @@ static void cleanup(void) { SYS_NOFAIL("test -f /var/run/netns/at_ns0 && ip netns delete at_ns0"); SYS_NOFAIL("ip link del veth1 2> /dev/null"); + SYS_NOFAIL("ip rule del to %s table 20 2> /dev/null", IP4_ADDR2_VETH1); SYS_NOFAIL("ip link del %s 2> /dev/null", VXLAN_TUNL_DEV1); SYS_NOFAIL("ip link del %s 2> /dev/null", IP6VXLAN_TUNL_DEV1); } @@ -140,8 +146,6 @@ static int add_vxlan_tunnel(void) VXLAN_TUNL_DEV0, IP4_ADDR_TUNL_DEV0); SYS("ip netns exec at_ns0 ip neigh add %s lladdr %s dev %s", IP4_ADDR_TUNL_DEV1, MAC_TUNL_DEV1, VXLAN_TUNL_DEV0); - SYS("ip netns exec at_ns0 ip neigh add %s lladdr %s dev veth0", - IP4_ADDR2_VETH1, MAC_VETH1); /* root namespace */ SYS("ip link add dev %s type vxlan external gbp dstport 4789", @@ -279,17 +283,6 @@ static void test_vxlan_tunnel(void) if (attach_tc_prog(&tc_hook, get_src_prog_fd, set_src_prog_fd)) goto done; - /* load and attach bpf prog to veth dev tc hook point */ - ifindex = if_nametoindex("veth1"); - if (!ASSERT_NEQ(ifindex, 0, "veth1 ifindex")) - goto done; - tc_hook.ifindex = ifindex; - set_dst_prog_fd = bpf_program__fd(skel->progs.veth_set_outer_dst); - if (!ASSERT_GE(set_dst_prog_fd, 0, "bpf_program__fd")) - goto done; - if (attach_tc_prog(&tc_hook, set_dst_prog_fd, -1)) - goto done; - /* load and attach prog set_md to tunnel dev tc hook point at_ns0 */ nstoken = open_netns("at_ns0"); if (!ASSERT_OK_PTR(nstoken, "setns src")) diff --git a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c index df0673c4ecbe..17f2f325b3f3 100644 --- a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c +++ b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c @@ -14,24 +14,15 @@ #include #include #include -#include #include #include #include #include -#include #include #include #define log_err(__ret) bpf_printk("ERROR line:%d ret:%d\n", __LINE__, __ret) -#define VXLAN_UDP_PORT 4789 - -/* Only IPv4 address assigned to veth1. - * 172.16.1.200 - */ -#define ASSIGNED_ADDR_VETH1 0xac1001c8 - struct geneve_opt { __be16 opt_class; __u8 type; @@ -42,11 +33,6 @@ struct geneve_opt { __u8 opt_data[8]; /* hard-coded to 8 byte */ }; -struct vxlanhdr { - __be32 vx_flags; - __be32 vx_vni; -} __attribute__((packed)); - struct vxlan_metadata { __u32 gbp; }; @@ -383,8 +369,14 @@ int vxlan_get_tunnel_src(struct __sk_buff *skb) int ret; struct bpf_tunnel_key key; struct vxlan_metadata md; - __u32 orig_daddr; __u32 index = 0; + __u32 *local_ip = NULL; + + local_ip = bpf_map_lookup_elem(&local_ip_map, &index); + if (!local_ip) { + log_err(ret); + return TC_ACT_SHOT; + } ret = bpf_skb_get_tunnel_key(skb, &key, sizeof(key), 0); if (ret < 0) { @@ -398,10 +390,11 @@ int vxlan_get_tunnel_src(struct __sk_buff *skb) return TC_ACT_SHOT; } - if (key.local_ipv4 != ASSIGNED_ADDR_VETH1 || md.gbp != 0x800FF) { + if (key.local_ipv4 != *local_ip || md.gbp != 0x800FF) { bpf_printk("vxlan key %d local ip 0x%x remote ip 0x%x gbp 0x%x\n", key.tunnel_id, key.local_ipv4, key.remote_ipv4, md.gbp); + bpf_printk("local_ip 0x%x\n", *local_ip); log_err(ret); return TC_ACT_SHOT; } @@ -409,61 +402,6 @@ int vxlan_get_tunnel_src(struct __sk_buff *skb) return TC_ACT_OK; } -SEC("tc") -int veth_set_outer_dst(struct __sk_buff *skb) -{ - struct ethhdr *eth = (struct ethhdr *)(long)skb->data; - __u32 assigned_ip = bpf_htonl(ASSIGNED_ADDR_VETH1); - void *data_end = (void *)(long)skb->data_end; - struct udphdr *udph; - struct iphdr *iph; - __u32 index = 0; - int ret = 0; - int shrink; - __s64 csum; - - if ((void *)eth + sizeof(*eth) > data_end) { - log_err(ret); - return TC_ACT_SHOT; - } - - if (eth->h_proto != bpf_htons(ETH_P_IP)) - return TC_ACT_OK; - - iph = (struct iphdr *)(eth + 1); - if ((void *)iph + sizeof(*iph) > data_end) { - log_err(ret); - return TC_ACT_SHOT; - } - if (iph->protocol != IPPROTO_UDP) - return TC_ACT_OK; - - udph = (struct udphdr *)(iph + 1); - if ((void *)udph + sizeof(*udph) > data_end) { - log_err(ret); - return TC_ACT_SHOT; - } - if (udph->dest != bpf_htons(VXLAN_UDP_PORT)) - return TC_ACT_OK; - - if (iph->daddr != assigned_ip) { - csum = bpf_csum_diff(&iph->daddr, sizeof(__u32), &assigned_ip, - sizeof(__u32), 0); - if (bpf_skb_store_bytes(skb, ETH_HLEN + offsetof(struct iphdr, daddr), - &assigned_ip, sizeof(__u32), 0) < 0) { - log_err(ret); - return TC_ACT_SHOT; - } - if (bpf_l3_csum_replace(skb, ETH_HLEN + offsetof(struct iphdr, check), - 0, csum, 0) < 0) { - log_err(ret); - return TC_ACT_SHOT; - } - bpf_skb_change_type(skb, PACKET_HOST); - } - return TC_ACT_OK; -} - SEC("tc") int ip6vxlan_set_tunnel_dst(struct __sk_buff *skb) {