Received-SPF: pass (google.com: domain of linux-kernel+bounces-9175-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
From: =?UTF-8?q?J=C3=B6rn-Thorben=20Hinz?= <jthinz@mailbox.tu-berlin.de>
To: <netdev@vger.kernel.org>, <linux-kernel@vger.kernel.org>
CC: =?UTF-8?q?J=C3=B6rn-Thorben=20Hinz?= <jthinz@mailbox.tu-berlin.de>, "David
 S. Miller" <davem@davemloft.net>, Eric Dumazet <edumazet@google.com>, Jakub
 Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>, Willem de Bruijn
	<willemb@google.com>, Deepa Dinamani <deepa.kernel@gmail.com>
Subject: [PATCH net] net: Implement missing getsockopt(SO_TIMESTAMPING_NEW)
Date: Fri, 22 Dec 2023 00:19:01 +0100
Message-ID: <20231221231901.67003-1-jthinz@mailbox.tu-berlin.de>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8bit

Commit 9718475e6908 ("socket: Add SO_TIMESTAMPING_NEW") added the new
socket option SO_TIMESTAMPING_NEW. Setting the option is handled in
sk_setsockopt(), querying it was not handled in sk_getsockopt(), though.

Following remarks on an earlier submission of this patch, keep the old
behavior of getsockopt(SO_TIMESTAMPING_OLD) which returns the active
flags even if they actually have been set through SO_TIMESTAMPING_NEW.

The new getsockopt(SO_TIMESTAMPING_NEW) is stricter, returning flags
only if they have been set through the same option.

Fixes: 9718475e6908 ("socket: Add SO_TIMESTAMPING_NEW")
Link: https://lore.kernel.org/lkml/20230703175048.151683-1-jthinz@mailbox.tu-berlin.de/
Link: https://lore.kernel.org/netdev/0d7cddc9-03fa-43db-a579-14f3e822615b@app.fastmail.com/
Signed-off-by: Jörn-Thorben Hinz <jthinz@mailbox.tu-berlin.de>
---
 net/core/sock.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/net/core/sock.c b/net/core/sock.c
index fef349dd72fa..51d52859e942 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1711,9 +1711,16 @@ int sk_getsockopt(struct sock *sk, int level, int optname,
 		break;
 
 	case SO_TIMESTAMPING_OLD:
+	case SO_TIMESTAMPING_NEW:
 		lv = sizeof(v.timestamping);
-		v.timestamping.flags = READ_ONCE(sk->sk_tsflags);
-		v.timestamping.bind_phc = READ_ONCE(sk->sk_bind_phc);
+		/* For the later-added case SO_TIMESTAMPING_NEW: Be strict about only
+		 * returning the flags when they were set through the same option.
+		 * Don't change the beviour for the old case SO_TIMESTAMPING_OLD.
+		 */
+		if (optname == SO_TIMESTAMPING_OLD || sock_flag(sk, SOCK_TSTAMP_NEW)) {
+			v.timestamping.flags = READ_ONCE(sk->sk_tsflags);
+			v.timestamping.bind_phc = READ_ONCE(sk->sk_bind_phc);
+		}
 		break;
 
 	case SO_RCVTIMEO_OLD:
-- 
2.39.2