From patchwork Mon Jan 11 23:09:39 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Borkmann X-Patchwork-Id: 362760 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=-16.8 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER,INCLUDES_PATCH, MAILING_LIST_MULTI, SPF_HELO_NONE, SPF_PASS, URIBL_BLOCKED, USER_AGENT_GIT autolearn=unavailable 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 3A97CC433DB for ; Tue, 12 Jan 2021 00:35:38 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 0C21D22D6E for ; Tue, 12 Jan 2021 00:35:38 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S2405372AbhALAZ1 (ORCPT ); Mon, 11 Jan 2021 19:25:27 -0500 Received: from www62.your-server.de ([213.133.104.62]:41986 "EHLO www62.your-server.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S2403812AbhAKXK0 (ORCPT ); Mon, 11 Jan 2021 18:10:26 -0500 Received: from 30.101.7.85.dynamic.wline.res.cust.swisscom.ch ([85.7.101.30] helo=localhost) by www62.your-server.de with esmtpsa (TLSv1.3:TLS_AES_256_GCM_SHA384:256) (Exim 4.92.3) (envelope-from ) id 1kz6Jk-000Dkf-Sm; Tue, 12 Jan 2021 00:09:44 +0100 From: Daniel Borkmann To: ast@kernel.org Cc: yhs@fb.com, bpf@vger.kernel.org, netdev@vger.kernel.org, Daniel Borkmann Subject: [PATCH bpf-next v2 1/2] bpf: allow to retrieve sol_socket opts from sock_addr progs Date: Tue, 12 Jan 2021 00:09:39 +0100 Message-Id: X-Mailer: git-send-email 2.21.0 MIME-Version: 1.0 X-Authenticated-Sender: daniel@iogearbox.net X-Virus-Scanned: Clear (ClamAV 0.102.4/26046/Mon Jan 11 13:34:14 2021) Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org The _bpf_setsockopt() is able to set some of the SOL_SOCKET level options, however, _bpf_getsockopt() has little support to actually retrieve them. This small patch adds few misc options such as SO_MARK, SO_PRIORITY and SO_BINDTOIFINDEX. For the latter getter and setter are added. The mark and priority in particular allow to retrieve the options from BPF cgroup hooks to then implement custom behavior / settings on the syscall hooks compared to other sockets that stick to the defaults, for example. Signed-off-by: Daniel Borkmann Acked-by: Yonghong Song --- net/core/filter.c | 25 +++++++++++++++++++++++-- 1 file changed, 23 insertions(+), 2 deletions(-) diff --git a/net/core/filter.c b/net/core/filter.c index 255aeee72402..9ab94e90d660 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -4770,6 +4770,10 @@ static int _bpf_setsockopt(struct sock *sk, int level, int optname, ifindex = dev->ifindex; dev_put(dev); } + fallthrough; + case SO_BINDTOIFINDEX: + if (optname == SO_BINDTOIFINDEX) + ifindex = val; ret = sock_bindtoindex(sk, ifindex, false); break; case SO_KEEPALIVE: @@ -4932,8 +4936,25 @@ static int _bpf_getsockopt(struct sock *sk, int level, int optname, sock_owned_by_me(sk); + if (level == SOL_SOCKET) { + if (optlen != sizeof(int)) + goto err_clear; + + switch (optname) { + case SO_MARK: + *((int *)optval) = sk->sk_mark; + break; + case SO_PRIORITY: + *((int *)optval) = sk->sk_priority; + break; + case SO_BINDTOIFINDEX: + *((int *)optval) = sk->sk_bound_dev_if; + break; + default: + goto err_clear; + } #ifdef CONFIG_INET - if (level == SOL_TCP && sk->sk_prot->getsockopt == tcp_getsockopt) { + } else if (level == SOL_TCP && sk->sk_prot->getsockopt == tcp_getsockopt) { struct inet_connection_sock *icsk; struct tcp_sock *tp; @@ -4986,12 +5007,12 @@ static int _bpf_getsockopt(struct sock *sk, int level, int optname, default: goto err_clear; } +#endif #endif } else { goto err_clear; } return 0; -#endif err_clear: memset(optval, 0, optlen); return -EINVAL; From patchwork Mon Jan 11 23:09:40 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Borkmann X-Patchwork-Id: 361723 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=-16.8 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER,INCLUDES_PATCH, MAILING_LIST_MULTI, SPF_HELO_NONE, SPF_PASS, URIBL_BLOCKED, USER_AGENT_GIT autolearn=unavailable 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 AF855C433E6 for ; Tue, 12 Jan 2021 00:35:32 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 7B6F422D6E for ; Tue, 12 Jan 2021 00:35:32 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S2405379AbhALAZ2 (ORCPT ); Mon, 11 Jan 2021 19:25:28 -0500 Received: from www62.your-server.de ([213.133.104.62]:41996 "EHLO www62.your-server.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S2403813AbhAKXK1 (ORCPT ); Mon, 11 Jan 2021 18:10:27 -0500 Received: from 30.101.7.85.dynamic.wline.res.cust.swisscom.ch ([85.7.101.30] helo=localhost) by www62.your-server.de with esmtpsa (TLSv1.3:TLS_AES_256_GCM_SHA384:256) (Exim 4.92.3) (envelope-from ) id 1kz6Jl-000Dkk-D8; Tue, 12 Jan 2021 00:09:45 +0100 From: Daniel Borkmann To: ast@kernel.org Cc: yhs@fb.com, bpf@vger.kernel.org, netdev@vger.kernel.org, Daniel Borkmann Subject: [PATCH bpf-next v2 2/2] bpf: extend bind v4/v6 selftests for mark/prio/bindtoifindex Date: Tue, 12 Jan 2021 00:09:40 +0100 Message-Id: <384fdc90e5fa83f8335a37aa90fa2f5f3661929c.1610406333.git.daniel@iogearbox.net> X-Mailer: git-send-email 2.21.0 In-Reply-To: References: MIME-Version: 1.0 X-Authenticated-Sender: daniel@iogearbox.net X-Virus-Scanned: Clear (ClamAV 0.102.4/26046/Mon Jan 11 13:34:14 2021) Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org Extend existing cgroup bind4/bind6 tests to add coverage for setting and retrieving SO_MARK, SO_PRIORITY and SO_BINDTOIFINDEX at the bind hook. Signed-off-by: Daniel Borkmann Acked-by: Yonghong Song --- v1 -> v2: - Add comment on old vs new optval (Yonghong), rest as-is .../testing/selftests/bpf/progs/bind4_prog.c | 42 +++++++++++++++++-- .../testing/selftests/bpf/progs/bind6_prog.c | 42 +++++++++++++++++-- 2 files changed, 76 insertions(+), 8 deletions(-) diff --git a/tools/testing/selftests/bpf/progs/bind4_prog.c b/tools/testing/selftests/bpf/progs/bind4_prog.c index c6520f21f5f5..115a3b0ad984 100644 --- a/tools/testing/selftests/bpf/progs/bind4_prog.c +++ b/tools/testing/selftests/bpf/progs/bind4_prog.c @@ -29,18 +29,48 @@ static __inline int bind_to_device(struct bpf_sock_addr *ctx) char veth2[IFNAMSIZ] = "test_sock_addr2"; char missing[IFNAMSIZ] = "nonexistent_dev"; char del_bind[IFNAMSIZ] = ""; + int veth1_idx, veth2_idx; if (bpf_setsockopt(ctx, SOL_SOCKET, SO_BINDTODEVICE, - &veth1, sizeof(veth1))) + &veth1, sizeof(veth1))) + return 1; + if (bpf_getsockopt(ctx, SOL_SOCKET, SO_BINDTOIFINDEX, + &veth1_idx, sizeof(veth1_idx)) || !veth1_idx) return 1; if (bpf_setsockopt(ctx, SOL_SOCKET, SO_BINDTODEVICE, - &veth2, sizeof(veth2))) + &veth2, sizeof(veth2))) + return 1; + if (bpf_getsockopt(ctx, SOL_SOCKET, SO_BINDTOIFINDEX, + &veth2_idx, sizeof(veth2_idx)) || !veth2_idx || + veth1_idx == veth2_idx) return 1; if (bpf_setsockopt(ctx, SOL_SOCKET, SO_BINDTODEVICE, - &missing, sizeof(missing)) != -ENODEV) + &missing, sizeof(missing)) != -ENODEV) + return 1; + if (bpf_setsockopt(ctx, SOL_SOCKET, SO_BINDTOIFINDEX, + &veth1_idx, sizeof(veth1_idx))) return 1; if (bpf_setsockopt(ctx, SOL_SOCKET, SO_BINDTODEVICE, - &del_bind, sizeof(del_bind))) + &del_bind, sizeof(del_bind))) + return 1; + + return 0; +} + +static __inline int misc_opts(struct bpf_sock_addr *ctx, int opt) +{ + int old, tmp, new = 0xeb9f; + + /* Socket in test case has guarantee that old never equals to new. */ + if (bpf_getsockopt(ctx, SOL_SOCKET, opt, &old, sizeof(old)) || + old == new) + return 1; + if (bpf_setsockopt(ctx, SOL_SOCKET, opt, &new, sizeof(new))) + return 1; + if (bpf_getsockopt(ctx, SOL_SOCKET, opt, &tmp, sizeof(tmp)) || + tmp != new) + return 1; + if (bpf_setsockopt(ctx, SOL_SOCKET, opt, &old, sizeof(old))) return 1; return 0; @@ -93,6 +123,10 @@ int bind_v4_prog(struct bpf_sock_addr *ctx) if (bind_to_device(ctx)) return 0; + /* Test for misc socket options. */ + if (misc_opts(ctx, SO_MARK) || misc_opts(ctx, SO_PRIORITY)) + return 0; + ctx->user_ip4 = bpf_htonl(SERV4_REWRITE_IP); ctx->user_port = bpf_htons(SERV4_REWRITE_PORT); diff --git a/tools/testing/selftests/bpf/progs/bind6_prog.c b/tools/testing/selftests/bpf/progs/bind6_prog.c index 4358e44dcf47..4c0d348034b9 100644 --- a/tools/testing/selftests/bpf/progs/bind6_prog.c +++ b/tools/testing/selftests/bpf/progs/bind6_prog.c @@ -35,18 +35,48 @@ static __inline int bind_to_device(struct bpf_sock_addr *ctx) char veth2[IFNAMSIZ] = "test_sock_addr2"; char missing[IFNAMSIZ] = "nonexistent_dev"; char del_bind[IFNAMSIZ] = ""; + int veth1_idx, veth2_idx; if (bpf_setsockopt(ctx, SOL_SOCKET, SO_BINDTODEVICE, - &veth1, sizeof(veth1))) + &veth1, sizeof(veth1))) + return 1; + if (bpf_getsockopt(ctx, SOL_SOCKET, SO_BINDTOIFINDEX, + &veth1_idx, sizeof(veth1_idx)) || !veth1_idx) return 1; if (bpf_setsockopt(ctx, SOL_SOCKET, SO_BINDTODEVICE, - &veth2, sizeof(veth2))) + &veth2, sizeof(veth2))) + return 1; + if (bpf_getsockopt(ctx, SOL_SOCKET, SO_BINDTOIFINDEX, + &veth2_idx, sizeof(veth2_idx)) || !veth2_idx || + veth1_idx == veth2_idx) return 1; if (bpf_setsockopt(ctx, SOL_SOCKET, SO_BINDTODEVICE, - &missing, sizeof(missing)) != -ENODEV) + &missing, sizeof(missing)) != -ENODEV) + return 1; + if (bpf_setsockopt(ctx, SOL_SOCKET, SO_BINDTOIFINDEX, + &veth1_idx, sizeof(veth1_idx))) return 1; if (bpf_setsockopt(ctx, SOL_SOCKET, SO_BINDTODEVICE, - &del_bind, sizeof(del_bind))) + &del_bind, sizeof(del_bind))) + return 1; + + return 0; +} + +static __inline int misc_opts(struct bpf_sock_addr *ctx, int opt) +{ + int old, tmp, new = 0xeb9f; + + /* Socket in test case has guarantee that old never equals to new. */ + if (bpf_getsockopt(ctx, SOL_SOCKET, opt, &old, sizeof(old)) || + old == new) + return 1; + if (bpf_setsockopt(ctx, SOL_SOCKET, opt, &new, sizeof(new))) + return 1; + if (bpf_getsockopt(ctx, SOL_SOCKET, opt, &tmp, sizeof(tmp)) || + tmp != new) + return 1; + if (bpf_setsockopt(ctx, SOL_SOCKET, opt, &old, sizeof(old))) return 1; return 0; @@ -107,6 +137,10 @@ int bind_v6_prog(struct bpf_sock_addr *ctx) if (bind_to_device(ctx)) return 0; + /* Test for misc socket options. */ + if (misc_opts(ctx, SO_MARK) || misc_opts(ctx, SO_PRIORITY)) + return 0; + ctx->user_ip6[0] = bpf_htonl(SERV6_REWRITE_IP_0); ctx->user_ip6[1] = bpf_htonl(SERV6_REWRITE_IP_1); ctx->user_ip6[2] = bpf_htonl(SERV6_REWRITE_IP_2);