From mboxrd@z Thu Jan 1 00:00:00 1970 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 31A0E2AD00; Sun, 7 Jun 2026 10:16:08 +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=1780827369; cv=none; b=FP923J/otkaYKx2SJv5Exvr6jALeQRVKm+ZT/8y5mEJD3uAZ2J4yS7v6rzloOm1BEgTLOZJdyj0w/MGY8uN3lPYWkcBJbed40RDH1kUXC8lHe/J0Bgscb4gvSgW97OKEAQlCtUMz4cIFai5sQti8yT3My7Hjk/FRJCc77sZl+dI= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1780827369; c=relaxed/simple; bh=vQYZ6E8+QBB/WuUBxc8FnJ7p0HfGG8TaMlFaIHavgGo=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=QAV2fL7FwlpmNW5cAFao3DWEVhXzFkR2aeIx4VmtXpa8cyos6frPLrW/MzgN740Z+zhc9xIoKaDoKtUxvsq8zuozmJDmRiV2liLGiB94mkHmVq1fyMEafWgUd/8qUT2ynmNbjILDKg4HZz/4hu7tnRiB+U5wnVNao2E85TLuA70= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b=MSFSY5JO; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b="MSFSY5JO" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 75FC61F00893; Sun, 7 Jun 2026 10:16:07 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linuxfoundation.org; s=korg; t=1780827368; bh=oY411OjsxxUELtRMfZJJJd94dEEV4TALC+hNjUJH/nU=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=MSFSY5JOGAwx3g+rGTUDDX/6+x/8P6TAqQP2xVLSEOk1dj4tSSRCFyI805yi+sFAk Rg3838FuWuVBfv6PCyLx/RKX0Q5Xi8Ngg8nVSP/IKIvbEalA7cMuExLY/4ZEfuqOyv WXfeq9+XvXeTZfOZ98d3iwZUoReFTVHIzBT9pxQs= From: Greg Kroah-Hartman To: stable@vger.kernel.org Cc: Greg Kroah-Hartman , patches@lists.linux.dev, Chuck Lever , Hannes Reinecke , Paolo Abeni , Sasha Levin Subject: [PATCH 7.0 087/332] net/handshake: Use spin_lock_bh for hn_lock Date: Sun, 7 Jun 2026 11:57:36 +0200 Message-ID: <20260607095731.342058596@linuxfoundation.org> X-Mailer: git-send-email 2.54.0 In-Reply-To: <20260607095728.031258202@linuxfoundation.org> References: <20260607095728.031258202@linuxfoundation.org> User-Agent: quilt/0.69 X-stable: review X-Patchwork-Hint: ignore Precedence: bulk X-Mailing-List: patches@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit 7.0-stable review patch. If anyone has any objections, please let me know. ------------------ From: Chuck Lever [ Upstream commit cc993e0927ec8bd98ea33377ada03295fcda0f24 ] nvmet_tcp_state_change(), a socket callback that runs in BH context, can reach handshake_req_cancel() via nvmet_tcp_schedule_release_queue() and tls_handshake_cancel(). handshake_req_cancel() acquires hn->hn_lock with plain spin_lock(). If a process-context thread on the same CPU holds hn->hn_lock when a softirq invokes the cancel path, the lock attempt deadlocks. This is the only caller that invokes tls_handshake_cancel() from BH context; every other consumer calls it from process context. Deferring the cancel to process context in the NVMe target is not straightforward: nvmet_tcp_schedule_release_queue() must call tls_handshake_cancel() atomically with its state transition to DISCONNECTING. If the cancel were deferred, the handshake completion callback could fire in the window before the cancel runs, observe the unexpected state, and return without dropping its kref on the queue. Reworking that interlock is considerably more invasive than hardening the handshake lock. Convert all hn->hn_lock acquisitions from spin_lock/spin_unlock to spin_lock_bh/spin_unlock_bh so the lock is never taken with softirqs enabled. Fixes: 675b453e0241 ("nvmet-tcp: enable TLS handshake upcall") Signed-off-by: Chuck Lever Reviewed-by: Hannes Reinecke Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-1-66c616906ead@oracle.com Signed-off-by: Paolo Abeni Signed-off-by: Sasha Levin --- net/handshake/netlink.c | 4 ++-- net/handshake/request.c | 14 +++++++------- net/handshake/tlshd.c | 2 ++ 3 files changed, 11 insertions(+), 9 deletions(-) diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index b989456fc4c5ff..97114ec8027a5a 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -202,10 +202,10 @@ static void __net_exit handshake_net_exit(struct net *net) * accepted and are in progress will be destroyed when * the socket is closed. */ - spin_lock(&hn->hn_lock); + spin_lock_bh(&hn->hn_lock); set_bit(HANDSHAKE_F_NET_DRAINING, &hn->hn_flags); list_splice_init(&requests, &hn->hn_requests); - spin_unlock(&hn->hn_lock); + spin_unlock_bh(&hn->hn_lock); while (!list_empty(&requests)) { req = list_first_entry(&requests, struct handshake_req, hr_list); diff --git a/net/handshake/request.c b/net/handshake/request.c index 2829adbeb149b0..5d4a17f902d201 100644 --- a/net/handshake/request.c +++ b/net/handshake/request.c @@ -167,12 +167,12 @@ static bool remove_pending(struct handshake_net *hn, struct handshake_req *req) { bool ret = false; - spin_lock(&hn->hn_lock); + spin_lock_bh(&hn->hn_lock); if (!list_empty(&req->hr_list)) { __remove_pending_locked(hn, req); ret = true; } - spin_unlock(&hn->hn_lock); + spin_unlock_bh(&hn->hn_lock); return ret; } @@ -182,7 +182,7 @@ struct handshake_req *handshake_req_next(struct handshake_net *hn, int class) struct handshake_req *req, *pos; req = NULL; - spin_lock(&hn->hn_lock); + spin_lock_bh(&hn->hn_lock); list_for_each_entry(pos, &hn->hn_requests, hr_list) { if (pos->hr_proto->hp_handler_class != class) continue; @@ -190,7 +190,7 @@ struct handshake_req *handshake_req_next(struct handshake_net *hn, int class) req = pos; break; } - spin_unlock(&hn->hn_lock); + spin_unlock_bh(&hn->hn_lock); return req; } @@ -249,7 +249,7 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, if (READ_ONCE(hn->hn_pending) >= hn->hn_pending_max) goto out_err; - spin_lock(&hn->hn_lock); + spin_lock_bh(&hn->hn_lock); ret = -EOPNOTSUPP; if (test_bit(HANDSHAKE_F_NET_DRAINING, &hn->hn_flags)) goto out_unlock; @@ -258,7 +258,7 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, goto out_unlock; if (!__add_pending_locked(hn, req)) goto out_unlock; - spin_unlock(&hn->hn_lock); + spin_unlock_bh(&hn->hn_lock); ret = handshake_genl_notify(net, req->hr_proto, flags); if (ret) { @@ -274,7 +274,7 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, return 0; out_unlock: - spin_unlock(&hn->hn_lock); + spin_unlock_bh(&hn->hn_lock); out_err: /* Restore original destructor so socket teardown still runs on failure */ req->hr_sk->sk_destruct = req->hr_odestruct; diff --git a/net/handshake/tlshd.c b/net/handshake/tlshd.c index 8f9532a15f43f9..af294c6cc71731 100644 --- a/net/handshake/tlshd.c +++ b/net/handshake/tlshd.c @@ -425,6 +425,8 @@ EXPORT_SYMBOL(tls_server_hello_psk); * Request cancellation races with request completion. To determine * who won, callers examine the return value from this function. * + * Context: May be called from process or softirq context. + * * Return values: * %true - Uncompleted handshake request was canceled * %false - Handshake request already completed or not found -- 2.53.0