From patchwork Sat May 8 00:29:28 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Michael Chan X-Patchwork-Id: 433003 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-19.4 required=3.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_CR_TRAILER, INCLUDES_PATCH, MAILING_LIST_MULTI, SPF_HELO_NONE, SPF_PASS, USER_AGENT_GIT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 36F48C433B4 for ; Sat, 8 May 2021 00:29:42 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 1D0E161451 for ; Sat, 8 May 2021 00:29:42 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230501AbhEHAak (ORCPT ); Fri, 7 May 2021 20:30:40 -0400 Received: from relay.smtp-ext.broadcom.com ([192.19.11.229]:35336 "EHLO relay.smtp-ext.broadcom.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230357AbhEHAaa (ORCPT ); Fri, 7 May 2021 20:30:30 -0400 Received: from localhost.swdvt.lab.broadcom.net (dhcp-10-13-253-90.swdvt.lab.broadcom.net [10.13.253.90]) by relay.smtp-ext.broadcom.com (Postfix) with ESMTP id 32EDD3893D; Fri, 7 May 2021 17:29:29 -0700 (PDT) DKIM-Filter: OpenDKIM Filter v2.11.0 relay.smtp-ext.broadcom.com 32EDD3893D DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=broadcom.com; s=dkimrelay; t=1620433769; bh=PdPVvnrlpo0BQHJxoGxquAnwZnHOKdxlAaByW6bUoTc=; h=From:To:Cc:Subject:Date:From; b=grhMumMnwP631lxeT9dav+J2EqhHxghuh++sP6nfegYDIWC/8BMCuxhoeKUGXbd38 m5xktAlTdsRZIipugoRxQwKTnI685gTkgAhGFyQTkJUA0l90Fd8vyJanyAGa+/wXHy Ns29B/XueNVWD6noFNGQuNpFG0BOChSP8WiHQ+bs= From: Michael Chan To: davem@davemloft.net Cc: netdev@vger.kernel.org, kuba@kernel.org, gospo@broadcom.com, alexander.duyck@gmail.com Subject: [PATCH net v2] bnxt_en: Fix and improve .ndo_features_check(). Date: Fri, 7 May 2021 20:29:28 -0400 Message-Id: <1620433768-31048-1-git-send-email-michael.chan@broadcom.com> X-Mailer: git-send-email 1.8.3.1 Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org Jakub Kicinski pointed out that we need to handle ipv6 extension headers and to explicitly check for supported tunnel types in .ndo_features_check(). For ipv6 extension headers, the hardware supports up to 2 ext. headers and each must be <= 64 bytes. For tunneled packets, the supported packets are UDP with supported VXLAN and Geneve ports, GRE, and IPIP. v2: Add missing step to check inner ipv6 header for UDP and GRE tunnels. Check TCP/UDP next header after skipping ipv6 ext headers for non-tunneled packets and for inner ipv6. (Both feedback from Alexander Duyck) Reviewed-by: Edwin Peer Reviewed-by: Pavan Chebbi Fixes: 1698d600b361 ("bnxt_en: Implement .ndo_features_check().") Signed-off-by: Michael Chan --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 125 +++++++++++++++++++--- 1 file changed, 108 insertions(+), 17 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 39ac9e2f5118..8e4670e37140 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -10785,37 +10785,128 @@ static int bnxt_set_features(struct net_device *dev, netdev_features_t features) return rc; } +static bool bnxt_exthdr_check(struct bnxt *bp, struct sk_buff *skb, int nw_off, + u8 *nextproto) +{ + struct ipv6hdr *ip6h = (struct ipv6hdr *)(skb->data + nw_off); + int hdr_count = 0; + u8 nexthdr; + int start; + + /* Check that there are at most 2 IPv6 extension headers, no + * fragment header, and each is <= 64 bytes. + */ + start = nw_off + sizeof(*ip6h); + nexthdr = ip6h->nexthdr; + while (ipv6_ext_hdr(nexthdr)) { + struct ipv6_opt_hdr _hdr, *hp; + int hdrlen; + + if (hdr_count >= 3 || nexthdr == NEXTHDR_NONE || + nexthdr == NEXTHDR_FRAGMENT) + return false; + hp = skb_header_pointer(skb, start, sizeof(_hdr), &_hdr); + if (!hp) + return false; + if (nexthdr == NEXTHDR_AUTH) + hdrlen = ipv6_authlen(hp); + else + hdrlen = ipv6_optlen(hp); + + if (hdrlen > 64) + return false; + nexthdr = hp->nexthdr; + start += hdrlen; + hdr_count++; + } + /* Only support TCP/UDP unless the caller checks the nextproto. */ + if (nextproto) { + *nextproto = nexthdr; + return true; + } + if (nexthdr == IPPROTO_TCP || nexthdr == IPPROTO_UDP) + return true; + return false; +} + +/* For UDP, we can only handle 1 Vxlan port and 1 Geneve port. */ +static bool bnxt_udp_tunl_check(struct bnxt *bp, struct sk_buff *skb) +{ + struct udphdr *uh = udp_hdr(skb); + __be16 udp_port = uh->dest; + + if (udp_port != bp->vxlan_port && udp_port != bp->nge_port) + return false; + if (skb->inner_protocol_type == ENCAP_TYPE_ETHER) { + struct ethhdr *eh = inner_eth_hdr(skb); + + switch (eh->h_proto) { + case htons(ETH_P_IP): + return true; + case htons(ETH_P_IPV6): + return bnxt_exthdr_check(bp, skb, + skb_inner_network_offset(skb), + NULL); + } + } + return false; +} + +static bool bnxt_tunl_check(struct bnxt *bp, struct sk_buff *skb, u8 l4_proto) +{ + switch (l4_proto) { + case IPPROTO_UDP: + return bnxt_udp_tunl_check(bp, skb); + case IPPROTO_IPIP: + return true; + case IPPROTO_GRE: { + switch (skb->inner_protocol) { + default: + return false; + case htons(ETH_P_IP): + return true; + case htons(ETH_P_IPV6): + fallthrough; + } + } + case IPPROTO_IPV6: + /* Check ext headers of inner ipv6 */ + return bnxt_exthdr_check(bp, skb, skb_inner_network_offset(skb), + NULL); + } + return false; +} + static netdev_features_t bnxt_features_check(struct sk_buff *skb, struct net_device *dev, netdev_features_t features) { - struct bnxt *bp; - __be16 udp_port; + struct bnxt *bp = netdev_priv(dev); u8 l4_proto = 0; features = vlan_features_check(skb, features); - if (!skb->encapsulation) - return features; - switch (vlan_get_protocol(skb)) { case htons(ETH_P_IP): + if (!skb->encapsulation) + return features; l4_proto = ip_hdr(skb)->protocol; - break; + if (!bnxt_tunl_check(bp, skb, l4_proto)) + goto disable_offload; + return features; case htons(ETH_P_IPV6): - l4_proto = ipv6_hdr(skb)->nexthdr; + if (!bnxt_exthdr_check(bp, skb, skb_network_offset(skb), + &l4_proto)) + goto disable_offload; + if (skb->encapsulation) { + if (bnxt_tunl_check(bp, skb, l4_proto)) + return features; + } else if (l4_proto == IPPROTO_TCP || l4_proto == IPPROTO_UDP) { + return features; + } break; - default: - return features; } - if (l4_proto != IPPROTO_UDP) - return features; - - bp = netdev_priv(dev); - /* For UDP, we can only handle 1 Vxlan port and 1 Geneve port. */ - udp_port = udp_hdr(skb)->dest; - if (udp_port == bp->vxlan_port || udp_port == bp->nge_port) - return features; +disable_offload: return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK); }