From nobody Thu Sep 24 22:56:29 2026 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (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 C431D51E44D; Fri, 18 Sep 2026 18:38:43 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789756731; cv=none; b=NJCj11vcF1DE1QX6ri7LL2pJQ9QJXpSIU0gKb1jE35MwLbPTweuAo1lhiojIOuQAv6TG7gg+J5axBJZCtTGUAzKZlbMJsWu5ne08EP90Kd3DV+I7OLgL227GIlFEQ52/U+N6okgCmYNcYDg+CEbI6XoCxgb2i37ajI/r10zvbfk= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789756731; c=relaxed/simple; bh=9fFB322ofigbhZ4AsHIL6Rjd777i8DmaY1KKlXdI0h4=; h=From:Date:Subject:MIME-Version:Content-Type:Message-Id:References: In-Reply-To:To:Cc; b=tFNXh88NWp9okJ3uB5J+4pzyeYjPkEsHZqEgi0KdVy1ZUrf2cAH9OPH3ZYL9e5NTog9/jrvkVvvhhRt7qaIdjLminE3hW3LF76IDtW83q/bmUb0Rans7kteDe7DXeBy/cx3WiyBqMBzGjiV0MqPZsAkE41xGc78lVunyev/S2dw= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=WhZxrER2; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="WhZxrER2" Received: by smtp.kernel.org (Postfix) with ESMTPSA id D6F5D1F00898; Fri, 18 Sep 2026 18:38:38 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1789756720; bh=HYDkWofAVJqwkkU1bY6hCgIZKtFJ2cyzVvCbMtn/XgE=; h=From:Date:Subject:References:In-Reply-To:To:Cc; b=WhZxrER2DvffppFwB5uP833/hiFEOC+Q5+20QlkGqkWNREoFWA7WQnmXVd+JoUBjh QOdENaTZ9rwlIYUYf2gJlD95VGPnQSzCWnT5ogqV6NrVKi/EjakXj/dj/QgumOQLEY K3+vdnYhQ6zKjKWU/d6scEkjKHunCEj05QZIhaCTOKK6Ts0Zfsar3TAKflppNJFGVg sbaQjZqSpsfFxrNLlbQ0FTI9v5nohoj34I03RIl2hUclqp9FLv3t1PcQ3VEQgFYIzN QC594049JRt3Wo2cYs7goSW+0hkK1CT5m0gwjM9hJvxk92rgithw1KFtmubBb64ZWE nPM2tNXd1pUxg== From: "Matthieu Baerts (NGI0)" Date: Fri, 18 Sep 2026 20:38:25 +0200 Subject: [PATCH net-next 1/4] mptcp: support MSG_ERRQUEUE on the parent socket Precedence: bulk X-Mailing-List: mptcp@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Message-Id: <20260918-net-next-mptcp-msg_errqueue-v1-1-dd77e1738248@kernel.org> References: <20260918-net-next-mptcp-msg_errqueue-v1-0-dd77e1738248@kernel.org> In-Reply-To: <20260918-net-next-mptcp-msg_errqueue-v1-0-dd77e1738248@kernel.org> To: Mat Martineau , Geliang Tang , "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , Simon Horman Cc: netdev@vger.kernel.org, mptcp@lists.linux.dev, linux-kernel@vger.kernel.org, "Matthieu Baerts (NGI0)" , David Carlier X-Mailer: b4 0.16.0 X-Developer-Signature: v=1; a=openpgp-sha256; l=4587; i=matttbe@kernel.org; h=from:subject:message-id; bh=6DCezqlDoG5LBUOfMLMJZ1aRCH4iRGtce+o1bVRCW28=; b=owGbwMvMwCVWo/Th0Gd3rumMp9WSGLLWtmowMl+5LNmx4/HegP+5gUV/FC/Jz/GbviFfljfVj GOLdVV7RykLgxgXg6yYIot0W2T+zOdVvCVefhYwc1iZQIYwcHEKwERyQxn+O7xSs0yYInWTtzVL Nv+J7/e0lAsnQnlWnmn/+OLQatl3OxgZJlR4TRZnPdyZd+aUFXfertA5lQlHmRd42U/VLmzffXg jPwA= X-Developer-Key: i=matttbe@kernel.org; a=openpgp; fpr=E8CB85F76877057A6E27F77AF6B7824F4269A073 From: David Carlier Splice pending err skbs from each subflow's error queue onto the parent msk's error queue at error-report time, so poll() and recvmsg(MSG_ERRQUEUE) on the parent socket observe TX timestamps through the standard inet ABI. The splice filters by SO_EE_ORIGIN: TIMESTAMPING events forward to the parent because they are tied to user-handed data, not to a specific path; subflow-level ICMP errors are dropped because the legacy RECVERR ABI cannot meaningfully convey their per-subflow peer identity to single-path-aware userspace. Such events will be carried by a future MPTCP_RECERR channel. Forwarded events all go through sock_queue_err_skb(), which re-homes skb->sk onto the parent and charges sk_rmem_alloc, so the parent's error queue stays bounded by sk_rcvbuf and is dropped under rmem pressure (sk_rmem_alloc + truesize >=3D sk_rcvbuf), matching tcp's sk_rcvbuf-gated tx-timestamp path and ip_icmp_error() / ipv6_icmp_error(). The MSG_ERRQUEUE branch of mptcp_recvmsg() forwards to inet_recv_error() directly, and poll() advertises EPOLLERR purely on the parent's sk_err / sk_error_queue, matching tcp_poll(). Signed-off-by: David Carlier Acked-by: Paolo Abeni Signed-off-by: Matthieu Baerts (NGI0) --- net/mptcp/protocol.c | 55 +++++++++++++++++++++++++++++++++++++++++++-----= ---- 1 file changed, 46 insertions(+), 9 deletions(-) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index e89a69ab927c..15ae270ee6fd 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include @@ -919,21 +920,52 @@ static bool __mptcp_ofo_queue(struct mptcp_sock *msk) return moved; } =20 +static bool mptcp_errqueue_skb_forwardable(const struct sk_buff *skb) +{ + /* Subflow-level ICMP errors are dropped: the legacy RECVERR ABI + * cannot convey their per-subflow peer identity. + */ + return SKB_EXT_ERR(skb)->ee.ee_origin =3D=3D SO_EE_ORIGIN_TIMESTAMPING; +} + +static bool __mptcp_subflow_splice_errqueue(struct sock *sk, struct sock *= ssk) +{ + struct sk_buff *skb; + bool moved =3D false; + + while ((skb =3D skb_dequeue(&ssk->sk_error_queue))) { + /* sock_queue_err_skb() re-homes skb->sk onto the parent and + * charges sk_rmem_alloc, bounding the queue by sk_rcvbuf. + */ + if (!mptcp_errqueue_skb_forwardable(skb) || + sock_queue_err_skb(sk, skb)) { + kfree_skb(skb); + continue; + } + moved =3D true; + } + + return moved; +} + static bool __mptcp_subflow_error_report(struct sock *sk, struct sock *ssk) { + bool propagated =3D false; int ssk_state; + bool report; int err; =20 + report =3D __mptcp_subflow_splice_errqueue(sk, ssk); + /* only propagate errors on fallen-back sockets or * on MPC connect */ if (sk->sk_state !=3D TCP_SYN_SENT && !__mptcp_check_fallback(mptcp_sk(sk= ))) - return false; + goto out; =20 err =3D sock_error(ssk); if (!err) - return false; - + goto out; /* We need to propagate only transition to CLOSE state. * Orphaned socket will see such state change via * subflow_sched_work_if_closed() and that path will properly @@ -943,11 +975,16 @@ static bool __mptcp_subflow_error_report(struct sock = *sk, struct sock *ssk) if (ssk_state =3D=3D TCP_CLOSE && !sock_flag(sk, SOCK_DEAD)) mptcp_set_state(sk, ssk_state); WRITE_ONCE(sk->sk_err, -err); + report =3D true; + propagated =3D true; =20 - /* This barrier is coupled with smp_rmb() in mptcp_poll() */ - smp_wmb(); - sk_error_report(sk); - return true; +out: + if (report) { + /* This barrier is coupled with smp_rmb() in mptcp_poll() */ + smp_wmb(); + sk_error_report(sk); + } + return propagated; } =20 void __mptcp_error_report(struct sock *sk) @@ -2385,7 +2422,6 @@ static int mptcp_recvmsg(struct sock *sk, struct msgh= dr *msg, size_t len, int target; long timeo; =20 - /* MSG_ERRQUEUE is really a no-op till we support IP_RECVERR */ if (unlikely(flags & MSG_ERRQUEUE)) return inet_recv_error(sk, msg, len); =20 @@ -4523,7 +4559,8 @@ static __poll_t mptcp_poll(struct file *file, struct = socket *sock, =20 /* This barrier is coupled with smp_wmb() in __mptcp_error_report() */ smp_rmb(); - if (READ_ONCE(sk->sk_err)) + if (READ_ONCE(sk->sk_err) || + !skb_queue_empty_lockless(&sk->sk_error_queue)) mask |=3D EPOLLERR; =20 return mask; --=20 2.55.0 From nobody Thu Sep 24 22:56:29 2026 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (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 C58ED51DE18; Fri, 18 Sep 2026 18:38:48 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789756732; cv=none; b=UsxXNnsVwVy65hzaMkwU3OB5HwMe7xGmb+dxBoG3SIvR2TdyUnoW4kemFSWgUUvR7dljglT+5wUUXcXjae78QQYTISLISlVaVOXoLkYJ79mU7qMI0VpQgR+OPQ4ZU629CEq7VGawG2mvEMf8o3Kl8/F5PHfAvQVYLKNVW0npfUI= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789756732; c=relaxed/simple; bh=wBV9w9NEHxkHdKdv+pY/PQzJ9oBFaLgpjdOOAi9u9lg=; h=From:Date:Subject:MIME-Version:Content-Type:Message-Id:References: In-Reply-To:To:Cc; b=nBiuLpPegC0g46DKR6uArbh+CbWPOEb5+esc6NeTH/g4QjERnVcjAnW3Y0o5uA4LrQs1Sx2neC+6bZjbCN0vCOpeEvuvRSaF/tSx5R1o4CX7IqawRg1H9xSBrbqh4z0uiXvbJTu+UJjuVfCPl+3oi6/X07yZ++kV5F7dZ/1E/vk= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=oP3xccBq; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="oP3xccBq" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 470D41F00899; Fri, 18 Sep 2026 18:38:41 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1789756723; bh=h/DLk8aAp4ApzIpkXH1VE7IRsVU6O2wuBRPsq9VNpy8=; h=From:Date:Subject:References:In-Reply-To:To:Cc; b=oP3xccBqHxIHLQKrg1BzfFMGrFps+h6DfFOPibWGk0LaOFHCGoT+JWreu2jSuy1K3 9lFs1Ucdn7xMs+shJr1Shpm+yLs81qmJauLvt1YimHO4V93fO/0GW8iqVPE7e3pXKK 1Je/YnR3sejQWF9zHnKjST22hnAAUjdJ7guU5DTZBdRFNQb7ESZpzACmVALISgE7kk yvBkNOpVnS4wI96D0awd9KtOEh8TJ6ZJU3DETYgf6PdCA7+Ss6JSpjBdRXJmao70wg ldzlYbRAzsDSjYWaE2fTzZDuIkHia0h63qotEMERoZIMI7CaJUny+PBiM/XX1jr/+G oKEk42nHZbCjA== From: "Matthieu Baerts (NGI0)" Date: Fri, 18 Sep 2026 20:38:26 +0200 Subject: [PATCH net-next 2/4] mptcp: sockopt: factor inet_flags propagation into a mask Precedence: bulk X-Mailing-List: mptcp@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Message-Id: <20260918-net-next-mptcp-msg_errqueue-v1-2-dd77e1738248@kernel.org> References: <20260918-net-next-mptcp-msg_errqueue-v1-0-dd77e1738248@kernel.org> In-Reply-To: <20260918-net-next-mptcp-msg_errqueue-v1-0-dd77e1738248@kernel.org> To: Mat Martineau , Geliang Tang , "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , Simon Horman Cc: netdev@vger.kernel.org, mptcp@lists.linux.dev, linux-kernel@vger.kernel.org, "Matthieu Baerts (NGI0)" , David Carlier X-Mailer: b4 0.16.0 X-Developer-Signature: v=1; a=openpgp-sha256; l=2231; i=matttbe@kernel.org; h=from:subject:message-id; bh=MZb1Th3F+8BZ7Pe907g7jydsYXsj2hesnGvFASzCFGY=; b=owGbwMvMwCVWo/Th0Gd3rumMp9WSGLLWtmoGBD9+/1X035YFCxUW35Y4/jLaf36frc9x+/h1u +be+dF/o6OUhUGMi0FWTJFFui0yf+bzKt4SLz8LmDmsTCBDGLg4BWAi2yoY/rtrLJk3OydGyoH3 VdJXi8KrZp9fXfSNN9x8PuPqik2VmswMf7gjd+5tmNoxmffBsZfn+881fktacEnjjtaG88rm6gI HjbgA X-Developer-Key: i=matttbe@kernel.org; a=openpgp; fpr=E8CB85F76877057A6E27F77AF6B7824F4269A073 From: David Carlier Introduce MPTCP_INET_FLAGS_MASK and replace the per-flag inet_assign_bit() calls in sync_socket_options() with a loop driven by the mask that calls assign_bit() per set bit, preserving the per-bit atomicity of the original. Further flags propagated by MPTCP can be added by extending the mask rather than touching the call site. No functional change. Signed-off-by: David Carlier Acked-by: Paolo Abeni Signed-off-by: Matthieu Baerts (NGI0) --- net/mptcp/sockopt.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index fcf6feb2a9eb..2ba56832587e 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -16,6 +16,10 @@ =20 #define MIN_INFO_OPTLEN_SIZE 16 #define MIN_FULL_INFO_OPTLEN_SIZE 40 +#define MPTCP_INET_FLAGS_MASK \ + (BIT(INET_FLAGS_TRANSPARENT) | \ + BIT(INET_FLAGS_FREEBIND) | \ + BIT(INET_FLAGS_BIND_ADDRESS_NO_PORT)) =20 static struct sock *__mptcp_tcp_fallback(struct mptcp_sock *msk) { @@ -1550,8 +1554,11 @@ int mptcp_getsockopt(struct sock *sk, int level, int= optname, static void sync_socket_options(struct mptcp_sock *msk, struct sock *ssk) { static const unsigned int tx_rx_locks =3D SOCK_RCVBUF_LOCK | SOCK_SNDBUF_= LOCK; + unsigned long mask =3D MPTCP_INET_FLAGS_MASK; struct sock *sk =3D (struct sock *)msk; + unsigned long src; bool keep_open; + int b; =20 keep_open =3D sock_flag(sk, SOCK_KEEPOPEN); if (ssk->sk_prot->keepalive) @@ -1597,9 +1604,11 @@ static void sync_socket_options(struct mptcp_sock *m= sk, struct sock *ssk) tcp_sock_set_keepcnt(ssk, msk->keepalive_cnt); tcp_sock_set_maxseg(ssk, msk->maxseg); =20 - inet_assign_bit(TRANSPARENT, ssk, inet_test_bit(TRANSPARENT, sk)); - inet_assign_bit(FREEBIND, ssk, inet_test_bit(FREEBIND, sk)); - inet_assign_bit(BIND_ADDRESS_NO_PORT, ssk, inet_test_bit(BIND_ADDRESS_NO_= PORT, sk)); + src =3D READ_ONCE(inet_sk(sk)->inet_flags); + + for_each_set_bit(b, &mask, BITS_PER_LONG) + assign_bit(b, &inet_sk(ssk)->inet_flags, src & BIT(b)); + WRITE_ONCE(inet_sk(ssk)->local_port_range, READ_ONCE(inet_sk(sk)->local_p= ort_range)); } =20 --=20 2.55.0 From nobody Thu Sep 24 22:56:29 2026 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (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 CA8BB51E43C; Fri, 18 Sep 2026 18:38:48 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789756732; cv=none; b=fPGOLcBHaKx5Sy/ZvmppiO83l9WqFF3lUy+DFP8YizyamEn2eXlQbQt9KIlfspHaTTw0+4fsCxZpeEErZOWUOhjndmFWjXGc5w4mc+oeCJOOc2K+tDY+7Efbu8rcoN1gr+8kjPMEJrFXMTueEP6zztezhzXFRROCoN21Lsoe2+A= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789756732; c=relaxed/simple; bh=fbWZZc8WwHlYza1PP7pAIgFAufWY3zIAyOEtVmk/sho=; h=From:Date:Subject:MIME-Version:Content-Type:Message-Id:References: In-Reply-To:To:Cc; b=ivDCiiGNnE/yJbc3PKQaikMt1yQbUY/TZNbZKGJ61UYNaxgY4ot8INXAgtTh0hsX3vZMZAcm3c8RrNtziowMKL0I/YDVinO3l0nj0XveHDUJTXup470f+MMOZowLy3f6GQ4XMh4fTgiauMToNiJjKnnTv/bON7FBHwy7+7kW8Fc= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=Xkj4drDw; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="Xkj4drDw" Received: by smtp.kernel.org (Postfix) with ESMTPSA id AB89C1F0089A; Fri, 18 Sep 2026 18:38:43 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1789756725; bh=2y7axseDLzTxuDDQ3sumIb+ws2cPPLcPH4O+iaVMq0Y=; h=From:Date:Subject:References:In-Reply-To:To:Cc; b=Xkj4drDw7PqDHbi8VYzXRWluum2lNGqg6HwYdY4Et1WY/ixopopuckEzuFxoFYkuy wF8ms12VczQfoaHJp5IXT4/DM3vcAvc/VWG5Nef3bj48kzV4rnLeVJygFI1zN8A3HX iLZGs0J6Yuk5o9ZpcOAuSqZBOUjM/CFCY34q5ov6HCIz7r4KfhBJ+oF5jMtNlKpRfb Ynyx3tQ2yr6NcK3rd8vpvFhI3zHY35u+IUUbqytOpUhp6/xYeeSZQ3Ajh+dATUU0Ap ZGP0KPDETIswKxGvtWkiO1e8Q+OQOdTTQoxw0/uZ1dCHon1H2sYEpOAt19ma2WIQSC dzznARva2Uo0g== From: "Matthieu Baerts (NGI0)" Date: Fri, 18 Sep 2026 20:38:27 +0200 Subject: [PATCH net-next 3/4] mptcp: propagate RECVERR sockopts to subflows Precedence: bulk X-Mailing-List: mptcp@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Message-Id: <20260918-net-next-mptcp-msg_errqueue-v1-3-dd77e1738248@kernel.org> References: <20260918-net-next-mptcp-msg_errqueue-v1-0-dd77e1738248@kernel.org> In-Reply-To: <20260918-net-next-mptcp-msg_errqueue-v1-0-dd77e1738248@kernel.org> To: Mat Martineau , Geliang Tang , "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , Simon Horman Cc: netdev@vger.kernel.org, mptcp@lists.linux.dev, linux-kernel@vger.kernel.org, "Matthieu Baerts (NGI0)" , David Carlier X-Mailer: b4 0.16.0 X-Developer-Signature: v=1; a=openpgp-sha256; l=7716; i=matttbe@kernel.org; h=from:subject:message-id; bh=ZV6SJUXmelrp3RjG7TZg+4EZ/ra6RMfmhMZOXd3tH80=; b=owGbwMvMwCVWo/Th0Gd3rumMp9WSGLLWtmpNXKFbU7v0LENczPHfZ7lq7ihkLZsn3cr30aRyq sSUSo39HaUsDGJcDLJiiizSbZH5M59X8ZZ4+VnAzGFlAhnCwMUpABOZ94uRoVXHtUpi3bqreo3T 2YSfRHy+EGC5ZenlLwFX/M4ksoUkGjD8L4ibuO/vZ+/z5zcsMs5nPX0kIHPB3L2MVwMWdtcqrN6 xjBUA X-Developer-Key: i=matttbe@kernel.org; a=openpgp; fpr=E8CB85F76877057A6E27F77AF6B7824F4269A073 From: David Carlier Propagate IP_RECVERR/IP_RECVERR_RFC4884 and IPV6_RECVERR/IPV6_RECVERR_RFC4884 from the MPTCP socket to existing and future subflows. mptcp_setsockopt_recverr() snapshots optval into a local int, applies it to the parent socket via ip_setsockopt() / ipv6_setsockopt(), bumps msk->setsockopt_seq, and forwards to every subflow via mptcp_setsockopt_all_sf(). Newly-joining subflows pick up the four RECVERR bits through sync_socket_options() now that MPTCP_INET_FLAGS_MASK covers them. mptcp_setsockopt_all_sf() skips IPv4 subflows when called with SOL_IPV6: ipv6_setsockopt() on a sock with sk_family !=3D AF_INET6 returns an error, which would abort the loop and leave the remaining subflows desynchronised. This branch was unreachable before this patch (the only caller was TCP_MAXSEG, family-agnostic); it becomes live with the new IPV6_RECVERR / IPV6_RECVERR_RFC4884 caller and the v4-subflow-on-AF_INET6-msk case (v4 MP_JOIN, or userspace PM grafting a v4 subflow onto a v6 msk). Assisted-by: Codex:gpt-5 Signed-off-by: David Carlier Acked-by: Paolo Abeni Signed-off-by: Matthieu Baerts (NGI0) --- net/mptcp/sockopt.c | 142 +++++++++++++++++++++++++++++++++++++++++++-----= ---- 1 file changed, 119 insertions(+), 23 deletions(-) diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index 2ba56832587e..922f6ae5c80c 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -8,6 +8,7 @@ =20 #include #include +#include #include #include #include @@ -19,7 +20,11 @@ #define MPTCP_INET_FLAGS_MASK \ (BIT(INET_FLAGS_TRANSPARENT) | \ BIT(INET_FLAGS_FREEBIND) | \ - BIT(INET_FLAGS_BIND_ADDRESS_NO_PORT)) + BIT(INET_FLAGS_BIND_ADDRESS_NO_PORT) | \ + BIT(INET_FLAGS_RECVERR) | \ + BIT(INET_FLAGS_RECVERR_RFC4884) | \ + BIT(INET_FLAGS_RECVERR6) | \ + BIT(INET_FLAGS_RECVERR6_RFC4884)) =20 static struct sock *__mptcp_tcp_fallback(struct mptcp_sock *msk) { @@ -398,6 +403,86 @@ static int mptcp_setsockopt_sol_socket(struct mptcp_so= ck *msk, int optname, return -EOPNOTSUPP; } =20 +static int mptcp_setsockopt_all_sf(struct mptcp_sock *msk, int level, + int optname, sockptr_t optval, + unsigned int optlen) +{ + struct mptcp_subflow_context *subflow; + int ret =3D 0; + + mptcp_for_each_subflow(msk, subflow) { + struct sock *ssk =3D mptcp_subflow_tcp_sock(subflow); + int err; + + /* SOL_IPV6 options on a v4 subflow (v4 MP_JOIN, or userspace PM + * grafting a v4 subflow onto an AF_INET6 msk) would otherwise + * abort the loop with -EAFNOSUPPORT from ipv6_setsockopt(). + */ + if (level =3D=3D SOL_IPV6 && ssk->sk_family !=3D AF_INET6) + continue; + + err =3D tcp_setsockopt(ssk, level, optname, optval, optlen); + if (err < 0 && ret =3D=3D 0) + ret =3D err; + } + + if (!ret) + sockopt_seq_inc(msk); + + return ret; +} + +static int mptcp_setsockopt_recverr(struct mptcp_sock *msk, int level, + int optname, sockptr_t optval, + unsigned int optlen) +{ + struct sock *sk =3D (struct sock *)msk; + int val =3D 0, ret; + + /* Let ip_setsockopt() / ipv6_setsockopt() validate optval and optlen + * (so 1-byte boolean writes keep the same ABI as plain TCP) and update + * the parent's RECVERR bit. Re-read that bit under lock_sock() and + * push it to the subflows: concurrent setsockopt callers cannot leave + * parent and subflows desynchronized this way. + */ + if (level =3D=3D SOL_IP) + ret =3D ip_setsockopt(sk, level, optname, optval, optlen); +#if IS_ENABLED(CONFIG_IPV6) + else if (level =3D=3D SOL_IPV6) { + if (sk->sk_family !=3D AF_INET6) + return -ENOPROTOOPT; + ret =3D ipv6_setsockopt(sk, level, optname, optval, optlen); + } +#endif + else + return -EOPNOTSUPP; + if (ret) + return ret; + + lock_sock(sk); + switch (optname) { + case IP_RECVERR: + val =3D inet_test_bit(RECVERR, sk); + break; + case IP_RECVERR_RFC4884: + val =3D inet_test_bit(RECVERR_RFC4884, sk); + break; +#if IS_ENABLED(CONFIG_IPV6) + case IPV6_RECVERR: + val =3D inet6_test_bit(RECVERR6, sk); + break; + case IPV6_RECVERR_RFC4884: + val =3D inet6_test_bit(RECVERR6_RFC4884, sk); + break; +#endif + } + + ret =3D mptcp_setsockopt_all_sf(msk, level, optname, + KERNEL_SOCKPTR(&val), sizeof(val)); + release_sock(sk); + return ret; +} + static int mptcp_setsockopt_v6(struct mptcp_sock *msk, int optname, sockptr_t optval, unsigned int optlen) { @@ -440,6 +525,11 @@ static int mptcp_setsockopt_v6(struct mptcp_sock *msk,= int optname, =20 release_sock(sk); break; + case IPV6_RECVERR: + case IPV6_RECVERR_RFC4884: + ret =3D mptcp_setsockopt_recverr(msk, SOL_IPV6, optname, optval, + optlen); + break; } =20 return ret; @@ -785,6 +875,10 @@ static int mptcp_setsockopt_v4(struct mptcp_sock *msk,= int optname, return mptcp_setsockopt_sol_ip_set(msk, optname, optval, optlen); case IP_TOS: return mptcp_setsockopt_v4_set_tos(msk, optname, optval, optlen); + case IP_RECVERR: + case IP_RECVERR_RFC4884: + return mptcp_setsockopt_recverr(msk, SOL_IP, optname, optval, + optlen); } =20 return -EOPNOTSUPP; @@ -812,28 +906,6 @@ static int mptcp_setsockopt_first_sf_only(struct mptcp= _sock *msk, int level, int return ret; } =20 -static int mptcp_setsockopt_all_sf(struct mptcp_sock *msk, int level, - int optname, sockptr_t optval, - unsigned int optlen) -{ - struct mptcp_subflow_context *subflow; - int ret =3D 0; - - mptcp_for_each_subflow(msk, subflow) { - struct sock *ssk =3D mptcp_subflow_tcp_sock(subflow); - int err; - - err =3D tcp_setsockopt(ssk, level, optname, optval, optlen); - if (err < 0 && ret =3D=3D 0) - ret =3D err; - } - - if (!ret) - sockopt_seq_inc(msk); - - return ret; -} - static int mptcp_setsockopt_sol_tcp(struct mptcp_sock *msk, int optname, sockptr_t optval, unsigned int optlen) { @@ -1478,6 +1550,12 @@ static int mptcp_getsockopt_v4(struct mptcp_sock *ms= k, int optname, case IP_LOCAL_PORT_RANGE: return mptcp_put_int_option(msk, optval, optlen, READ_ONCE(inet_sk(sk)->local_port_range)); + case IP_RECVERR: + return mptcp_put_int_option(msk, optval, optlen, + inet_test_bit(RECVERR, sk)); + case IP_RECVERR_RFC4884: + return mptcp_put_int_option(msk, optval, optlen, + inet_test_bit(RECVERR_RFC4884, sk)); } =20 return -EOPNOTSUPP; @@ -1498,6 +1576,16 @@ static int mptcp_getsockopt_v6(struct mptcp_sock *ms= k, int optname, case IPV6_FREEBIND: return mptcp_put_int_option(msk, optval, optlen, inet_test_bit(FREEBIND, sk)); + case IPV6_RECVERR: + if (sk->sk_family !=3D AF_INET6) + return -ENOPROTOOPT; + return mptcp_put_int_option(msk, optval, optlen, + inet6_test_bit(RECVERR6, sk)); + case IPV6_RECVERR_RFC4884: + if (sk->sk_family !=3D AF_INET6) + return -ENOPROTOOPT; + return mptcp_put_int_option(msk, optval, optlen, + inet6_test_bit(RECVERR6_RFC4884, sk)); } =20 return -EOPNOTSUPP; @@ -1606,6 +1694,14 @@ static void sync_socket_options(struct mptcp_sock *m= sk, struct sock *ssk) =20 src =3D READ_ONCE(inet_sk(sk)->inet_flags); =20 + /* RECVERR6 bits are only read on AF_INET6 sockets; copying them onto a + * v4 subflow is dead state and diverges from the SOL_IPV6 skip in + * mptcp_setsockopt_all_sf(). + */ + if (ssk->sk_family !=3D AF_INET6) + mask &=3D ~(BIT(INET_FLAGS_RECVERR6) | + BIT(INET_FLAGS_RECVERR6_RFC4884)); + for_each_set_bit(b, &mask, BITS_PER_LONG) assign_bit(b, &inet_sk(ssk)->inet_flags, src & BIT(b)); =20 --=20 2.55.0 From nobody Thu Sep 24 22:56:30 2026 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (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 05D6F51EDF2; Fri, 18 Sep 2026 18:38:49 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789756738; cv=none; b=uiTLLm3HOoVcw01MLb0xm38l2ORTchN1j68o/JnYhE4gXkFJKxfPAPuIummFxfPJ8R70geBieB5u6q4k0ggjoxgRHGVS5y6nUl0N25XOjBVRKlqQf+29Xs+E4R23Ue3RIZcwBPorefwaGVC2ycYUTJQre71epTnWZdeYJgC6ayY= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789756738; c=relaxed/simple; bh=g3to6dCeHn6zMM/3AMYugP+pLtURStDEBkyEbyy+1GY=; h=From:Date:Subject:MIME-Version:Content-Type:Message-Id:References: In-Reply-To:To:Cc; b=jRlbR8iV3BUAB0hixukm14heNTrJQuXXsqEfX8oBGo8n/VF7bRhih6OJkn67T4NxLTkyhsGUuXvgwkvFp2bXXfdragK6uI1dLhAb4xlOtVhnwTapEDk59SNFVLaSwEc2USScDjewByEwVEIt+GjNRAHC0KGZsXWvajWC7HCdMT0= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=O4c3uU5/; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="O4c3uU5/" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 1AC1B1F0089B; Fri, 18 Sep 2026 18:38:45 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1789756728; bh=ab97ABeWx519BjZUlq/57k57+C5FyDQmJadIWx10vY8=; h=From:Date:Subject:References:In-Reply-To:To:Cc; b=O4c3uU5/gLdDxORBoMVwmzOYdl3Ugn7dGHPDY/a9mQ8UpbhgAf21TgHGCIa2s4XKB nbwmNAXweQGTT4iWEucVptmgsb4ykWhQpC5r6nn9ZV4ETmYfc40RDciizxq1w9AxlL H4/QaG9AGtIdL2xqc2Ie9XnJlucpdNi15t+RZY3O4MRpvhjOHq0PO27N/4ioCQNBYS nNJB1nBIU8mUIg85BVk1ZqVBGYJ7D7d/Uc6uATTpZ2W29O/3HnM9ICUanH+HnzNBkQ jVnPHbE4r+M2YcidPtTQuiae+l0VyjdwH7NurYC7YpWom9rXAzYG23oj4y/h0tnIVf INLpOu5/rg4Sg== From: "Matthieu Baerts (NGI0)" Date: Fri, 18 Sep 2026 20:38:28 +0200 Subject: [PATCH net-next 4/4] selftests: mptcp: cover IP_RECVERR sockopt propagation Precedence: bulk X-Mailing-List: mptcp@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Message-Id: <20260918-net-next-mptcp-msg_errqueue-v1-4-dd77e1738248@kernel.org> References: <20260918-net-next-mptcp-msg_errqueue-v1-0-dd77e1738248@kernel.org> In-Reply-To: <20260918-net-next-mptcp-msg_errqueue-v1-0-dd77e1738248@kernel.org> To: Mat Martineau , Geliang Tang , "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , Simon Horman Cc: netdev@vger.kernel.org, mptcp@lists.linux.dev, linux-kernel@vger.kernel.org, "Matthieu Baerts (NGI0)" , David Carlier , Shuah Khan , linux-kselftest@vger.kernel.org X-Mailer: b4 0.16.0 X-Developer-Signature: v=1; a=openpgp-sha256; l=3312; i=matttbe@kernel.org; h=from:subject:message-id; bh=UW1zAcPF+yaxAXuK61+P5ERyAUnUn0OOq8Dg98C2P+w=; b=owGbwMvMwCVWo/Th0Gd3rumMp9WSGLLWtmrHmnk/dFK7+PJ7+86D7DtqfacKPQlrZJ0bMP/qJ HEv9TT1jlIWBjEuBlkxRRbptsj8mc+reEu8/Cxg5rAygQxh4OIUgInYtDMy/Pxj2rbN00/1inBz tlA7+1xB/uacCas6Fh7bsve1eNTVH4wMJwuiJk6SE1/dxfarT1u9Z3uPddnK3yFiLIcF1QSyQ55 zAQA= X-Developer-Key: i=matttbe@kernel.org; a=openpgp; fpr=E8CB85F76877057A6E27F77AF6B7824F4269A073 From: David Carlier Exercise setsockopt/getsockopt of IP_RECVERR and IPV6_RECVERR on the MPTCP parent socket, including the empty-errqueue EAGAIN contract on MSG_ERRQUEUE|MSG_DONTWAIT. End-to-end errqueue delivery (ICMP, TX timestamps, zerocopy) depends on subflow-side producers that are out of scope for this series and will be covered by follow-up work. Assisted-by: Codex:gpt-5 Signed-off-by: David Carlier Acked-by: Paolo Abeni Signed-off-by: Matthieu Baerts (NGI0) --- To: Shuah Khan Cc: linux-kselftest@vger.kernel.org --- tools/testing/selftests/net/mptcp/mptcp_sockopt.c | 70 +++++++++++++++++++= ++++ 1 file changed, 70 insertions(+) diff --git a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c b/tools/test= ing/selftests/net/mptcp/mptcp_sockopt.c index b6e58d936ebe..d68515b7903b 100644 --- a/tools/testing/selftests/net/mptcp/mptcp_sockopt.c +++ b/tools/testing/selftests/net/mptcp/mptcp_sockopt.c @@ -183,6 +183,13 @@ static void xgetaddrinfo(const char *node, const char = *service, } } =20 +static bool expect_all_features(void) +{ + char *env =3D getenv("SELFTESTS_MPTCP_LIB_EXPECT_ALL_FEATURES"); + + return env && strcmp(env, "1") =3D=3D 0; +} + static int sock_listen_mptcp(const char * const listenaddr, const char * const port) { @@ -769,6 +776,68 @@ static void test_ip_tos_sockopt(int fd) xerror("expect socklen_t =3D=3D -1"); } =20 +static void test_ip_recverr_sockopt(int fd) +{ + struct iovec iov =3D { + .iov_base =3D &(char){ 0 }, + .iov_len =3D 1, + }; + struct msghdr msg =3D { + .msg_iov =3D &iov, + .msg_iovlen =3D 1, + }; + int one =3D 1, zero =3D 0, val =3D -1; + socklen_t s =3D sizeof(val); + int level, optname, r; + + switch (pf) { + case AF_INET: + level =3D SOL_IP; + optname =3D IP_RECVERR; + break; + case AF_INET6: + level =3D SOL_IPV6; + optname =3D IPV6_RECVERR; + break; + default: + xerror("Unknown pf %d\n", pf); + } + + r =3D setsockopt(fd, level, optname, &one, sizeof(one)); + if (r) { + /* For older kernels not supporting IP(V6)_RECVERR yet */ + if (errno =3D=3D EOPNOTSUPP && !expect_all_features()) { + fprintf(stderr, "IP(V6)_RECVERR not supported, SKIP\n"); + return; + } + + die_perror("setsockopt IP(V6)_RECVERR on"); + } + + r =3D getsockopt(fd, level, optname, &val, &s); + if (r) + die_perror("getsockopt IP(V6)_RECVERR on"); + if (s !=3D sizeof(val) || val !=3D one) + xerror("IP(V6)_RECVERR on mismatch val=3D%d len=3D%u", val, s); + + r =3D recvmsg(fd, &msg, MSG_ERRQUEUE | MSG_DONTWAIT); + if (r !=3D -1 || errno !=3D EAGAIN) + xerror("expected empty errqueue to return EAGAIN, r=3D%d err=3D%d", + r, errno); + + r =3D setsockopt(fd, level, optname, &zero, sizeof(zero)); + if (r) + die_perror("setsockopt IP(V6)_RECVERR off"); + + val =3D -1; + s =3D sizeof(val); + r =3D getsockopt(fd, level, optname, &val, &s); + if (r) + die_perror("getsockopt IP(V6)_RECVERR off"); + if (s !=3D sizeof(val) || val !=3D zero) + xerror("IP(V6)_RECVERR off mismatch val=3D%d len=3D%u", val, s); +} + static int client(int pipefd) { int fd =3D -1; @@ -787,6 +856,7 @@ static int client(int pipefd) } =20 test_ip_tos_sockopt(fd); + test_ip_recverr_sockopt(fd); =20 connect_one_server(fd, pipefd); =20 --=20 2.55.0