From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-qk1-f197.google.com (mail-qk1-f197.google.com [209.85.222.197]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 6C47235295C for ; Thu, 27 Aug 2026 09:59:43 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.222.197 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1787824798; cv=none; b=l4+E/cyOHsKRj7d73oLt5BwJCfVrn7ebAivl6I3b5MqNgW+TKr3R5+DrSzih16dAO040ZArZU6rV5dINijSeL0YB7hOxwy8hDVnM24mT90Kk66NTtw/7ahdbRad4tVzGEUl053M7X1MQ3FlZhgsvZ/dCdDtKGy/4xnomlSil0UU= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1787824798; c=relaxed/simple; bh=h7G3/kC+p8D1kUbhgBcyAsgXcspLGJsAuz7vfvbqZ0w=; h=Date:Mime-Version:Message-ID:Subject:From:To:Cc:Content-Type; b=rWYcmuW0uJLQ4AjOK8to2AjEtJjCQWno/lZYK2cS8OzwYz86preTGHi6u3s9tzvDoGOv5weHMtoiVUYXehKgDHxmt5dzqOkxAxanti2pmJYGPmIesS2xDu4e91wbuScEPdU1rHxyD6s2PuJZH2jmMbz63pxWV74PiF9a756aEw4= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com; spf=pass smtp.mailfrom=flex--edumazet.bounces.google.com; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b=CG2BSFet; arc=none smtp.client-ip=209.85.222.197 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=flex--edumazet.bounces.google.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="CG2BSFet" Received: by mail-qk1-f197.google.com with SMTP id af79cd13be357-92ec3146553so312993585a.1 for ; Thu, 27 Aug 2026 02:59:42 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20251104; t=1787824778; x=1788429578; darn=vger.kernel.org; h=content-type:cc:to:from:subject:message-id:mime-version:date:from :to:cc:subject:date:message-id:reply-to:content-type; bh=FBIzrg8ZFKoNTtfDO8edGrIoQ09TXt9xuNZA8cBP54Q=; b=CG2BSFetK9MyOU601BDRfJdGMrJ4J6i9Dw6hErb6cG2Cvp/vYUAWAZg1tTzOSipBij XA7BBdwCkyoIXO164XGyxCuauEnJKX3aNLUivor0vA+k6HVrgsfonMGz3cD7rUY082l+ 4Wyih7zNmIvQhYq1ZNL+wUdTvUnizpQy8W1GimpmWaCB/tYQnScSiJQdlHH2DDJWuBON 4gT+o2vwhpUG4sa7gZpHp+UYy+S+8OZPCOiXXzimrSlkyIolV2tur3RsSBJKcPCXF1KM 0QCAbhmUsP+/4Jl0wmz5F55PUyNAjWzNseAEHuvgLlKS2ZYYYuRlMvyyx/GAFwcDktWk 2KnA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1787824778; x=1788429578; h=content-type:cc:to:from:subject:message-id:mime-version:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to :content-type; bh=FBIzrg8ZFKoNTtfDO8edGrIoQ09TXt9xuNZA8cBP54Q=; b=nN0n5n1XOhvalgF8nGbKCN/Ol44OXvtteDxfH+tocdE3oXLg8CWqfhDbHZY8ZoReC7 wYvuKT/RZgCaOis7HIsQBW31fTM1DDO2WsKjVmhIPrv/RTd5QcQjOcH9UY7ROzf2BdS7 kW9yri6pe7WgOySCzd1Oy2wf9+Yg64FthoYiQRbuSBDWx9WnoW8GGMNftazHhlBfLKry p1mhq5eGREoBUukdsMlz03eumrXozSfFxPSJxDkG5B7Pts+xnm90cyqZN5yn1zl1b8he dYQe0hmmrwUhR116Q1gqHfzlNpRx56PEsXunU/mlq1c7YYL+jFhXsU+KBkMXAD8zMekV oi2w== X-Forwarded-Encrypted: i=1; AHgh+Ro28NzK/GrgHbgI3ZKGTr2aKjbSksrjY7yseZwDlzJ0z/v8DT+cvJSHT6ZAQ4UlzV+qYoxdHP8=@vger.kernel.org X-Gm-Message-State: AFuF++nBQKuBufd/vvgUO8CJ+ZJcTGxFkYRpQ0JI5NmsoOXBhDceBdE/ CLqBXpPCX31WjvaX6KYwPfZnpgOrOl8FAu5t2Ux4zTDEL3c4QVo5NsjisXh0vt8i5VZqCiY1H+J fcTxo4SEI8E7xeA== X-Received: from qkac8.prod.google.com ([2002:a05:620a:2008:b0:934:95c3:41d6]) (user=edumazet job=prod-delivery.src-stubby-dispatcher) by 2002:a05:620a:708c:b0:92e:f380:dabf with SMTP id af79cd13be357-937803ebd11mr1266668485a.30.1787824777658; Thu, 27 Aug 2026 02:59:37 -0700 (PDT) Date: Thu, 27 Aug 2026 09:59:36 +0000 Precedence: bulk X-Mailing-List: netdev@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 X-Mailer: git-send-email 2.55.0.887.g758fc8c411-goog Message-ID: <20260827095936.551524-1-edumazet@google.com> Subject: [PATCH v2 net] tcp: use GFP_ATOMIC in tcp_send_active_reset() From: Eric Dumazet To: "David S . Miller" , Jakub Kicinski , Paolo Abeni Cc: Simon Horman , Neal Cardwell , Kuniyuki Iwashima , netdev@vger.kernel.org, eric.dumazet@gmail.com, Eric Dumazet Content-Type: text/plain; charset="UTF-8" tcp_send_active_reset() can be called from contexts where gfp_any() (in tcp_disconnect()) or sk->sk_allocation (in __tcp_close() and mptcp_do_fastclose()) evaluates to GFP_KERNEL, which includes __GFP_FS and __GFP_DIRECT_RECLAIM. Allocating with GFP_KERNEL while holding the socket lock (sk_lock) creates a lockdep dependency: sk_lock -> fs_reclaim This causes false-positive lockdep circular locking warnings with storage subsystems (such as nvme-tcp) that acquire socket locks in block I/O paths and invoke tcp_disconnect() or close sockets upon teardown: set->srcu -> sk_lock -> fs_reclaim -> elevator_lock -> set->srcu Active resets are small RST packet headers that should never enter direct reclaim or block while holding socket locks. Use sk_gfp_mask(sk, GFP_ATOMIC | __GFP_NOWARN) inside tcp_send_active_reset() and remove its priority argument. This preserves __GFP_MEMALLOC access for SOCK_MEMALLOC sockets, suppresses allocation failure warnings, and aligns with other control packet allocations (e.g. tcp_send_fin(), __tcp_send_ack(), tcp_xmit_probe_skb()). Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Eric Dumazet --- v2: addressed actionnable Sashiko's feedback (https://netdev-ai.bots.linux.dev/sashiko/#/patchset/20260825023614.1228551-1-edumazet%40google.com) v1: https://lore.kernel.org/netdev/20260825023614.1228551-1-edumazet@google.com/ include/net/tcp.h | 3 +-- net/ipv4/tcp.c | 14 ++++++-------- net/ipv4/tcp_output.c | 4 ++-- net/ipv4/tcp_timer.c | 6 +++--- net/mptcp/protocol.c | 3 +-- net/mptcp/protocol.h | 2 +- 6 files changed, 14 insertions(+), 18 deletions(-) diff --git a/include/net/tcp.h b/include/net/tcp.h index 670c20876f265c14504c26f45b87763ae47d3127..436495ff2271de047423dbe33036b7c6d1556584 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -765,8 +765,7 @@ int tcp_fragment(struct sock *sk, enum tcp_queue tcp_queue, void tcp_send_probe0(struct sock *); int tcp_write_wakeup(struct sock *, int mib); void tcp_send_fin(struct sock *sk); -void tcp_send_active_reset(struct sock *sk, gfp_t priority, - enum sk_rst_reason reason); +void tcp_send_active_reset(struct sock *sk, enum sk_rst_reason reason); int tcp_send_synack(struct sock *); void tcp_push_one(struct sock *, unsigned int mss_now); void __tcp_send_ack(struct sock *sk, u32 rcv_nxt, u16 flags); diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index b4237d0e994d6f9d754d2167023e3981a40b58f4..93d723d8c1098e421cb7e3596318acd20fd80233 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -3182,8 +3182,7 @@ void __tcp_close(struct sock *sk, long timeout) /* Unread data was tossed, zap the connection. */ NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTONCLOSE); tcp_set_state(sk, TCP_CLOSE); - tcp_send_active_reset(sk, sk->sk_allocation, - SK_RST_REASON_TCP_ABORT_ON_CLOSE); + tcp_send_active_reset(sk, SK_RST_REASON_TCP_ABORT_ON_CLOSE); } else if (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime) { /* Check zero linger _after_ checking for unread data. */ sk->sk_prot->disconnect(sk, 0); @@ -3257,7 +3256,7 @@ void __tcp_close(struct sock *sk, long timeout) struct tcp_sock *tp = tcp_sk(sk); if (READ_ONCE(tp->linger2) < 0) { tcp_set_state(sk, TCP_CLOSE); - tcp_send_active_reset(sk, GFP_ATOMIC, + tcp_send_active_reset(sk, SK_RST_REASON_TCP_ABORT_ON_LINGER); __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTONLINGER); @@ -3276,7 +3275,7 @@ void __tcp_close(struct sock *sk, long timeout) if (sk->sk_state != TCP_CLOSE) { if (tcp_check_oom(sk, 0)) { tcp_set_state(sk, TCP_CLOSE); - tcp_send_active_reset(sk, GFP_ATOMIC, + tcp_send_active_reset(sk, SK_RST_REASON_TCP_ABORT_ON_MEMORY); __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTONMEMORY); @@ -3377,14 +3376,14 @@ int tcp_disconnect(struct sock *sk, int flags) } else if (unlikely(tp->repair)) { WRITE_ONCE(sk->sk_err, ECONNABORTED); } else if (tcp_need_reset(old_state)) { - tcp_send_active_reset(sk, gfp_any(), SK_RST_REASON_TCP_STATE); + tcp_send_active_reset(sk, SK_RST_REASON_TCP_STATE); WRITE_ONCE(sk->sk_err, ECONNRESET); } else if (tp->snd_nxt != tp->write_seq && (1 << old_state) & (TCPF_CLOSING | TCPF_LAST_ACK)) { /* The last check adjusts for discrepancy of Linux wrt. RFC * states */ - tcp_send_active_reset(sk, gfp_any(), + tcp_send_active_reset(sk, SK_RST_REASON_TCP_DISCONNECT_WITH_DATA); WRITE_ONCE(sk->sk_err, ECONNRESET); } else if (old_state == TCP_SYN_SENT) @@ -5147,8 +5146,7 @@ int tcp_abort(struct sock *sk, int err) bh_lock_sock(sk); if (tcp_need_reset(sk->sk_state)) - tcp_send_active_reset(sk, GFP_ATOMIC, - SK_RST_REASON_TCP_STATE); + tcp_send_active_reset(sk, SK_RST_REASON_TCP_STATE); tcp_done_with_error(sk, err); bh_unlock_sock(sk); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index f2709d585edbd9d9fef97953920edfcb7c45e61c..3581384d097feea4cddc3c1f30404367aa93aaa8 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -3849,9 +3849,9 @@ void tcp_send_fin(struct sock *sk) * was unread data in the receive queue. This behavior is recommended * by RFC 2525, section 2.17. -DaveM */ -void tcp_send_active_reset(struct sock *sk, gfp_t priority, - enum sk_rst_reason reason) +void tcp_send_active_reset(struct sock *sk, enum sk_rst_reason reason) { + gfp_t priority = sk_gfp_mask(sk, GFP_ATOMIC | __GFP_NOWARN); struct sk_buff *skb; TCP_INC_STATS(sock_net(sk), TCP_MIB_OUTRSTS); diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 1038e7ba9c2eb19279b431249b56f8a8e4ffaf74..e56eae4bc341e94880bf0d6d1435094dfaaf879f 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -126,7 +126,7 @@ static int tcp_out_of_resources(struct sock *sk, bool do_reset) (!tp->snd_wnd && !tp->packets_out)) do_reset = true; if (do_reset) - tcp_send_active_reset(sk, GFP_ATOMIC, + tcp_send_active_reset(sk, SK_RST_REASON_TCP_ABORT_ON_MEMORY); tcp_done(sk); __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTONMEMORY); @@ -809,7 +809,7 @@ static void tcp_keepalive_timer(struct timer_list *t) goto out; } } - tcp_send_active_reset(sk, GFP_ATOMIC, SK_RST_REASON_TCP_STATE); + tcp_send_active_reset(sk, SK_RST_REASON_TCP_STATE); goto death; } @@ -836,7 +836,7 @@ static void tcp_keepalive_timer(struct timer_list *t) icsk->icsk_probes_out > 0) || (user_timeout == 0 && icsk->icsk_probes_out >= keepalive_probes(tp))) { - tcp_send_active_reset(sk, GFP_ATOMIC, + tcp_send_active_reset(sk, SK_RST_REASON_TCP_KEEPALIVE_TIMEOUT); tcp_write_err(sk); goto out; diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index b474d03620a75d3df26fcae1a84901b965c6954e..e1f08f71cdb16b2bbecd5630bca895d651c17b4c 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -3109,8 +3109,7 @@ static void mptcp_do_fastclose(struct sock *sk) */ inet_csk(ssk)->icsk_ack.rcv_mss = TCP_MIN_MSS; - tcp_send_active_reset(ssk, ssk->sk_allocation, - SK_RST_REASON_TCP_ABORT_ON_CLOSE); + tcp_send_active_reset(ssk, SK_RST_REASON_TCP_ABORT_ON_CLOSE); unlock: release_sock(ssk); } diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 06a107d4e8392269b42f3af7ac531764619107c1..87ccb84e9927ccb23b242c11d34eb69427362702 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -690,7 +690,7 @@ mptcp_send_active_reset_reason(struct sock *sk) enum sk_rst_reason reason; reason = sk_rst_convert_mptcp_reason(subflow->reset_reason); - tcp_send_active_reset(sk, GFP_ATOMIC, reason); + tcp_send_active_reset(sk, reason); } /* Made the fwd mem carried by the given skb available to the msk, -- 2.55.0.887.g758fc8c411-goog