From nobody Tue Jun 16 04:57:57 2026 Received: from out-185.mta0.migadu.com (out-185.mta0.migadu.com [91.218.175.185]) (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 A2E4739B96D for ; Thu, 16 Apr 2026 11:24:24 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=91.218.175.185 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1776338666; cv=none; b=IEQJ0mT7VSJcfnb3CLdvaaCNZ/yXBk21NNZkEDUVZe643+VUEt7ZfafXgwHMDUDHsMXneU47iOTm7Rudja/Hphk5XdPyVdvENQdA9Bjt4+lG69jmGQ5F3Na4I9rk/DG2xa3gFFGI92yckRHfLF2wfukt7A6O2WomPHzKQPA4YR8= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1776338666; c=relaxed/simple; bh=tGOhrptT9PJ5/cvzp6EKffTU5Kr88yNrPL4dsjtp0/E=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=uxJN+goL3je3jp/He4ACsZUNvqiz8kjvF9dyxigX81Fb2MhNLY2BrU6pkz6GU9MmgAid6dWFmcNIJP1sPL/xPajZxHzEmrKVVTiGPKk2jjyGOKadq407fmXFBsppRw9incxWF1Cxc1SYXQCHLUimprYTBLwvqkCQF8PVrdvEcoU= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.dev; spf=pass smtp.mailfrom=linux.dev; dkim=pass (1024-bit key) header.d=linux.dev header.i=@linux.dev header.b=UF1KpSD0; arc=none smtp.client-ip=91.218.175.185 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.dev Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.dev Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux.dev header.i=@linux.dev header.b="UF1KpSD0" X-Report-Abuse: Please report any abuse attempt to abuse@migadu.com and include these headers. DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linux.dev; s=key1; t=1776338662; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=8AldNZMVHS+FDrWsqQWGH6HMzCXTBhC6l0kEDtweIFA=; b=UF1KpSD0RSSqrsSf0EPQ+pdlhKbgUzEbBpVt3z0hBlQ9KX33ed7Ig5M4GtdluxgiCN2NdE NtRnLLXQ+AAm4swHPVKFRRkpN+J2ZGW/ycKIQwiQNJzvTko3F5O3GFRrx+5/FKPmFEY917 a0VVUGPFpWNVuzdxNaUDTio4cDNPofg= From: KaFai Wan To: martin.lau@linux.dev, daniel@iogearbox.net, john.fastabend@gmail.com, sdf@fomichev.me, ast@kernel.org, andrii@kernel.org, eddyz87@gmail.com, memxor@gmail.com, song@kernel.org, yonghong.song@linux.dev, jolsa@kernel.org, davem@davemloft.net, edumazet@google.com, kuba@kernel.org, pabeni@redhat.com, horms@kernel.org, shuah@kernel.org, jiayuan.chen@linux.dev, kafai.wan@linux.dev, bpf@vger.kernel.org, netdev@vger.kernel.org, linux-kernel@vger.kernel.org, linux-kselftest@vger.kernel.org Cc: Quan Sun <2022090917019@std.uestc.edu.cn>, Yinhao Hu , Kaiyan Mei Subject: [PATCH bpf v2 1/2] bpf: Reject TCP_NODELAY in TCP header option callbacks Date: Thu, 16 Apr 2026 19:23:07 +0800 Message-ID: <20260416112308.1820332-2-kafai.wan@linux.dev> In-Reply-To: <20260416112308.1820332-1-kafai.wan@linux.dev> References: <20260416112308.1820332-1-kafai.wan@linux.dev> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Migadu-Flow: FLOW_OUT Content-Type: text/plain; charset="utf-8" A BPF_SOCK_OPS program can enable BPF_SOCK_OPS_WRITE_HDR_OPT_CB_FLAG and then call bpf_setsockopt(TCP_NODELAY) from BPF_SOCK_OPS_HDR_OPT_LEN_CB or BPF_SOCK_OPS_WRITE_HDR_OPT_CB. In these callbacks, bpf_setsockopt(TCP_NODELAY) can reach __tcp_sock_set_nodelay(), which can call tcp_push_pending_frames(). From BPF_SOCK_OPS_HDR_OPT_LEN_CB, tcp_push_pending_frames() can call tcp_current_mss(), which calls tcp_established_options() and re-enters bpf_skops_hdr_opt_len(). BPF_SOCK_OPS_HDR_OPT_LEN_CB -> bpf_setsockopt(TCP_NODELAY) -> tcp_push_pending_frames() -> tcp_current_mss() -> tcp_established_options() -> bpf_skops_hdr_opt_len() -> BPF_SOCK_OPS_HDR_OPT_LEN_CB From BPF_SOCK_OPS_WRITE_HDR_OPT_CB, tcp_push_pending_frames() can call tcp_write_xmit(), which calls tcp_transmit_skb(). That path recomputes header option length through tcp_established_options() and bpf_skops_hdr_opt_len() before re-entering bpf_skops_write_hdr_opt(). BPF_SOCK_OPS_WRITE_HDR_OPT_CB -> bpf_setsockopt(TCP_NODELAY) -> tcp_push_pending_frames() -> tcp_write_xmit() -> tcp_transmit_skb() -> tcp_established_options() -> bpf_skops_hdr_opt_len() -> bpf_skops_write_hdr_opt() -> BPF_SOCK_OPS_WRITE_HDR_OPT_CB This leads to unbounded recursion and can overflow the kernel stack. Reject TCP_NODELAY with -EOPNOTSUPP in bpf_sock_ops_setsockopt() when bpf_setsockopt() is called from BPF_SOCK_OPS_HDR_OPT_LEN_CB or BPF_SOCK_OPS_WRITE_HDR_OPT_CB. Reported-by: Quan Sun <2022090917019@std.uestc.edu.cn> Reported-by: Yinhao Hu Reported-by: Kaiyan Mei Closes: https://lore.kernel.org/bpf/d1d523c9-6901-4454-a183-94462b8f3e4e@st= d.uestc.edu.cn/ Fixes: 7e41df5dbba2 ("bpf: Add a few optnames to bpf_setsockopt") Signed-off-by: KaFai Wan --- net/core/filter.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/net/core/filter.c b/net/core/filter.c index fcfcb72663ca..911ff04bca5a 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -5833,6 +5833,11 @@ BPF_CALL_5(bpf_sock_ops_setsockopt, struct bpf_sock_= ops_kern *, bpf_sock, if (!is_locked_tcp_sock_ops(bpf_sock)) return -EOPNOTSUPP; =20 + if ((bpf_sock->op =3D=3D BPF_SOCK_OPS_HDR_OPT_LEN_CB || + bpf_sock->op =3D=3D BPF_SOCK_OPS_WRITE_HDR_OPT_CB) && + IS_ENABLED(CONFIG_INET) && level =3D=3D SOL_TCP && optname =3D=3D TCP= _NODELAY) + return -EOPNOTSUPP; + return _bpf_setsockopt(bpf_sock->sk, level, optname, optval, optlen); } =20 --=20 2.43.0 From nobody Tue Jun 16 04:57:57 2026 Received: from out-184.mta0.migadu.com (out-184.mta0.migadu.com [91.218.175.184]) (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 6672C39FCAE for ; Thu, 16 Apr 2026 11:24:35 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=91.218.175.184 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1776338677; cv=none; b=Q4PHPa3RSz3YUYgckT4+jkFftYtv6KifiMBPuEC5q934fjP65TsYRhF5QaG/K1MY7rn4jYIE1XF44BXuCnGrrsH/o3iivNFPAsbgJBBwZ8Gtv14Ilo2Zi9NLEOPNfiiT0PWgXMsW9uiynrt5KyJ0bpmb7bZk8o3fso9rVkivFKg= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1776338677; c=relaxed/simple; bh=dKUQGdzSULQLFpjO4woDfNNas3V9RQeU8cHdUhNQEJU=; h=From:To:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=gxxnA7fhCjsNNEOxGUT85l6vpQeOpWJenarFMhmdZYsBDa2NryQwDKF/RGBmhni8sLfx61E/2X00HnkRhFdl96Nw8EriFyGEFj65MY4lmMqy1rF3xd1IgHL5Ld9qjejy3T2fQQbkEWoGsu8T8e0WGf7cCddhTO6KEVO/Gbg/qgc= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.dev; spf=pass smtp.mailfrom=linux.dev; dkim=pass (1024-bit key) header.d=linux.dev header.i=@linux.dev header.b=tPxOlj1m; arc=none smtp.client-ip=91.218.175.184 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=linux.dev Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=linux.dev Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux.dev header.i=@linux.dev header.b="tPxOlj1m" X-Report-Abuse: Please report any abuse attempt to abuse@migadu.com and include these headers. DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linux.dev; s=key1; t=1776338673; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=0g3FoFi5R4RZ3NBwYcu3bG8S2Vu+l0yHuaglflknQ3U=; b=tPxOlj1mIfymjjjGNuzOmCJiDGCniTAj4kwYlQ8s5hYU3z+RzIFlU/5PRo3Bw3fdyPSXev 5+MFfyZmOSWCpGGtNPd71quIwsYBQ2Qv+jl3PK/QUegVHxdIwAtC0dFApp/2bMhlXcqL/D 577U9NAVmpVV88aCmkm/2FhWXxvMzKQ= From: KaFai Wan To: martin.lau@linux.dev, daniel@iogearbox.net, john.fastabend@gmail.com, sdf@fomichev.me, ast@kernel.org, andrii@kernel.org, eddyz87@gmail.com, memxor@gmail.com, song@kernel.org, yonghong.song@linux.dev, jolsa@kernel.org, davem@davemloft.net, edumazet@google.com, kuba@kernel.org, pabeni@redhat.com, horms@kernel.org, shuah@kernel.org, jiayuan.chen@linux.dev, kafai.wan@linux.dev, bpf@vger.kernel.org, netdev@vger.kernel.org, linux-kernel@vger.kernel.org, linux-kselftest@vger.kernel.org Subject: [PATCH bpf v2 2/2] selftests/bpf: Test TCP_NODELAY in TCP hdr opt callbacks Date: Thu, 16 Apr 2026 19:23:08 +0800 Message-ID: <20260416112308.1820332-3-kafai.wan@linux.dev> In-Reply-To: <20260416112308.1820332-1-kafai.wan@linux.dev> References: <20260416112308.1820332-1-kafai.wan@linux.dev> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Migadu-Flow: FLOW_OUT Content-Type: text/plain; charset="utf-8" Add a sockops selftest for the TCP_NODELAY restriction in BPF_SOCK_OPS_HDR_OPT_LEN_CB and BPF_SOCK_OPS_WRITE_HDR_OPT_CB. The test program calls bpf_setsockopt(TCP_NODELAY) from ACTIVE_ESTABLISHED_CB and PASSIVE_ESTABLISHED_CB to verify that it is still allowed outside the TCP header option callbacks. It then enables BPF_SOCK_OPS_WRITE_HDR_OPT_CB_FLAG, sends data to exercise the TCP header option path, and checks that bpf_setsockopt(TCP_NODELAY) returns -EOPNOTSUPP from both BPF_SOCK_OPS_HDR_OPT_LEN_CB and BPF_SOCK_OPS_WRITE_HDR_OPT_CB while the connection continues to make forward progress. Signed-off-by: KaFai Wan --- .../bpf/prog_tests/tcp_hdr_options.c | 54 +++++++++++++++++++ .../bpf/progs/test_misc_tcp_hdr_options.c | 40 ++++++++++++++ 2 files changed, 94 insertions(+) diff --git a/tools/testing/selftests/bpf/prog_tests/tcp_hdr_options.c b/too= ls/testing/selftests/bpf/prog_tests/tcp_hdr_options.c index 56685fc03c7e..2d738c0c4259 100644 --- a/tools/testing/selftests/bpf/prog_tests/tcp_hdr_options.c +++ b/tools/testing/selftests/bpf/prog_tests/tcp_hdr_options.c @@ -513,6 +513,59 @@ static void misc(void) bpf_link__destroy(link); } =20 +static void hdr_sockopt(void) +{ + const char send_msg[] =3D "MISC!!!"; + char recv_msg[sizeof(send_msg)]; + const unsigned int nr_data =3D 2; + struct bpf_link *link; + struct sk_fds sk_fds; + int i, ret, true_val =3D 1; + + lport_linum_map_fd =3D bpf_map__fd(misc_skel->maps.lport_linum_map); + + link =3D bpf_program__attach_cgroup(misc_skel->progs.misc_hdr_sockopt, cg= _fd); + if (!ASSERT_OK_PTR(link, "attach_cgroup(misc_hdr_sockopt)")) + return; + + if (sk_fds_connect(&sk_fds, false)) { + bpf_link__destroy(link); + return; + } + + ret =3D setsockopt(sk_fds.active_fd, SOL_TCP, TCP_NODELAY, &true_val, siz= eof(true_val)); + if (!ASSERT_OK(ret, "setsockopt(TCP_NODELAY) active")) + goto check_linum; + + ret =3D setsockopt(sk_fds.passive_fd, SOL_TCP, TCP_NODELAY, &true_val, si= zeof(true_val)); + if (!ASSERT_OK(ret, "setsockopt(TCP_NODELAY) passive")) + goto check_linum; + + for (i =3D 0; i < nr_data; i++) { + ret =3D send(sk_fds.active_fd, send_msg, sizeof(send_msg), 0); + if (!ASSERT_EQ(ret, sizeof(send_msg), "send(msg)")) + goto check_linum; + + ret =3D read(sk_fds.passive_fd, recv_msg, sizeof(recv_msg)); + if (!ASSERT_EQ(ret, sizeof(send_msg), "read(msg)")) + goto check_linum; + } + + ASSERT_NEQ(misc_skel->bss->nr_hdr_sockopt_estab, 0, "nr_hdr_sockopt_estab= "); + ASSERT_EQ(misc_skel->bss->nr_hdr_sockopt_estab_err, 0, "nr_hdr_sockopt_es= tab_err"); + + ASSERT_NEQ(misc_skel->bss->nr_hdr_sockopt_len, 0, "nr_hdr_sockopt_len"); + ASSERT_EQ(misc_skel->bss->nr_hdr_sockopt_len_err, 0, "nr_hdr_sockopt_len_= err"); + + ASSERT_NEQ(misc_skel->bss->nr_hdr_sockopt_write, 0, "nr_hdr_sockopt_write= "); + ASSERT_EQ(misc_skel->bss->nr_hdr_sockopt_write_err, 0, "nr_hdr_sockopt_wr= ite_err"); + +check_linum: + ASSERT_FALSE(check_error_linum(&sk_fds), "check_error_linum"); + sk_fds_close(&sk_fds); + bpf_link__destroy(link); +} + struct test { const char *desc; void (*run)(void); @@ -526,6 +579,7 @@ static struct test tests[] =3D { DEF_TEST(fastopen_estab), DEF_TEST(fin), DEF_TEST(misc), + DEF_TEST(hdr_sockopt), }; =20 void test_tcp_hdr_options(void) diff --git a/tools/testing/selftests/bpf/progs/test_misc_tcp_hdr_options.c = b/tools/testing/selftests/bpf/progs/test_misc_tcp_hdr_options.c index d487153a839d..a8cf7c4e7ed2 100644 --- a/tools/testing/selftests/bpf/progs/test_misc_tcp_hdr_options.c +++ b/tools/testing/selftests/bpf/progs/test_misc_tcp_hdr_options.c @@ -28,6 +28,12 @@ unsigned int nr_data =3D 0; unsigned int nr_syn =3D 0; unsigned int nr_fin =3D 0; unsigned int nr_hwtstamp =3D 0; +unsigned int nr_hdr_sockopt_estab =3D 0; +unsigned int nr_hdr_sockopt_estab_err =3D 0; +unsigned int nr_hdr_sockopt_len =3D 0; +unsigned int nr_hdr_sockopt_len_err =3D 0; +unsigned int nr_hdr_sockopt_write =3D 0; +unsigned int nr_hdr_sockopt_write_err =3D 0; =20 /* Check the header received from the active side */ static int __check_active_hdr_in(struct bpf_sock_ops *skops, bool check_sy= n) @@ -326,4 +332,38 @@ int misc_estab(struct bpf_sock_ops *skops) return CG_OK; } =20 +SEC("sockops") +int misc_hdr_sockopt(struct bpf_sock_ops *skops) +{ + int true_val =3D 1; + int ret; + + switch (skops->op) { + case BPF_SOCK_OPS_ACTIVE_ESTABLISHED_CB: + case BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB: + nr_hdr_sockopt_estab++; + set_hdr_cb_flags(skops, 0); + ret =3D bpf_setsockopt(skops, SOL_TCP, TCP_NODELAY, &true_val, sizeof(tr= ue_val)); + if (ret) + nr_hdr_sockopt_estab_err++; + break; + case BPF_SOCK_OPS_HDR_OPT_LEN_CB: + nr_hdr_sockopt_len++; + ret =3D bpf_setsockopt(skops, SOL_TCP, TCP_NODELAY, &true_val, sizeof(tr= ue_val)); + if (ret !=3D -EOPNOTSUPP) + nr_hdr_sockopt_len_err++; + /* just trigger BPF_SOCK_OPS_WRITE_HDR_OPT_CB */ + bpf_reserve_hdr_opt(skops, 12, 0); + break; + case BPF_SOCK_OPS_WRITE_HDR_OPT_CB: + nr_hdr_sockopt_write++; + ret =3D bpf_setsockopt(skops, SOL_TCP, TCP_NODELAY, &true_val, sizeof(tr= ue_val)); + if (ret !=3D -EOPNOTSUPP) + nr_hdr_sockopt_write_err++; + break; + } + + return CG_OK; +} + char _license[] SEC("license") =3D "GPL"; --=20 2.43.0