Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com>
To:     nhorman@tuxdriver.com
Cc:     davem@davemloft.net,
        Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com>,
        Daniel Borkmann <daniel@iogearbox.net>,
        Christian Brauner <brauner@kernel.org>,
        Stanislav Fomichev <sdf@google.com>,
        Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>,
        Xin Long <lucien.xin@gmail.com>, linux-sctp@vger.kernel.org,
        linux-kernel@vger.kernel.org, netdev@vger.kernel.org
Subject: [PATCH net-next v2] sctp: add bpf_bypass_getsockopt proto callback
Date:   Thu, 11 May 2023 14:31:48 +0200
Message-Id: <20230511123148.332043-1-aleksandr.mikhalitsyn@canonical.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

Add bpf_bypass_getsockopt proto callback and filter out
SCTP_SOCKOPT_PEELOFF and SCTP_SOCKOPT_PEELOFF_FLAGS socket options
from running eBPF hook on them.

These options do fd_install(), and if BPF_CGROUP_RUN_PROG_GETSOCKOPT
hook returns an error after success of the original handler
sctp_getsockopt(...), userspace will receive an error from getsockopt
syscall and will be not aware that fd was successfully installed into fdtable.

This patch was born as a result of discussion around a new SCM_PIDFD interface:
https://lore.kernel.org/all/20230413133355.350571-3-aleksandr.mikhalitsyn@canonical.com/

Fixes: 0d01da6afc54 ("bpf: implement getsockopt and setsockopt hooks")
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Stanislav Fomichev <sdf@google.com>
Cc: Neil Horman <nhorman@tuxdriver.com>
Cc: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Cc: Xin Long <lucien.xin@gmail.com>
Cc: linux-sctp@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: netdev@vger.kernel.org
Suggested-by: Stanislav Fomichev <sdf@google.com>
Acked-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com>
---
 net/sctp/socket.c | 31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index cda8c2874691..a211a203003c 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -8281,6 +8281,35 @@ static int sctp_getsockopt(struct sock *sk, int level, int optname,
 	return retval;
 }
 
+static bool sctp_bpf_bypass_getsockopt(int level, int optname)
+{
+	if (level == SOL_SCTP) {
+		switch (optname) {
+		/*
+		 * These options do fd_install(), and if BPF_CGROUP_RUN_PROG_GETSOCKOPT
+		 * hook returns an error after success of the original handler
+		 * sctp_getsockopt(...), userspace will receive an error from getsockopt
+		 * syscall and will be not aware that fd was successfully installed into fdtable.
+		 *
+		 * Let's prevent bpf cgroup hook from running on them.
+		 */
+		case SCTP_SOCKOPT_PEELOFF:
+		case SCTP_SOCKOPT_PEELOFF_FLAGS:
+		/*
+		 * As pointed by Marcelo Ricardo Leitner it seems reasonable to skip
+		 * bpf getsockopt hook for this sockopt too. Because internaly, it
+		 * triggers connect() and if error will be masked userspace can be confused.
+		 */
+		case SCTP_SOCKOPT_CONNECTX3:
+			return true;
+		default:
+			return false;
+		}
+	}
+
+	return false;
+}
+
 static int sctp_hash(struct sock *sk)
 {
 	/* STUB */
@@ -9650,6 +9679,7 @@ struct proto sctp_prot = {
 	.shutdown    =	sctp_shutdown,
 	.setsockopt  =	sctp_setsockopt,
 	.getsockopt  =	sctp_getsockopt,
+	.bpf_bypass_getsockopt	= sctp_bpf_bypass_getsockopt,
 	.sendmsg     =	sctp_sendmsg,
 	.recvmsg     =	sctp_recvmsg,
 	.bind        =	sctp_bind,
@@ -9705,6 +9735,7 @@ struct proto sctpv6_prot = {
 	.shutdown	= sctp_shutdown,
 	.setsockopt	= sctp_setsockopt,
 	.getsockopt	= sctp_getsockopt,
+	.bpf_bypass_getsockopt	= sctp_bpf_bypass_getsockopt,
 	.sendmsg	= sctp_sendmsg,
 	.recvmsg	= sctp_recvmsg,
 	.bind		= sctp_bind,
-- 
2.34.1