[PATCH net-next v3 1/2] tcp: keep the route window metric when SO_RCVBUF is set before connect

[Shashank Mohan Jain <jain.sm@gmail.com>]

Since commit b025461303d8 ("tcp: update window_clamp when SO_RCVBUF is
set"), setsockopt(SO_RCVBUF) writes tp->window_clamp in every socket
state through tcp_set_window_clamp().

tcp_connect_init() and tcp_openreq_init_rwin() treat any non-zero
window_clamp as a clamp the user asked for and, in that case, skip the
route window metric (RTAX_WINDOW, "ip route ... window N"). A clamp left
behind by an SO_RCVBUF call therefore has two observable effects:

 (1) on a route that carries an RTAX_WINDOW metric, SO_RCVBUF set before
     connect() (or on a listener) changes the window and window scale
     advertised in the SYN / SYN-ACK, because the route metric is no
     longer applied;

 (2) a window clamp the application set with setsockopt(TCP_WINDOW_CLAMP)
     is overwritten by a later SO_RCVBUF on the still unconnected socket
     or on a listener.

Without an RTAX_WINDOW metric there is no change: both before and after
this patch the advertised window and scale are derived from sk_rcvbuf
(tcp_full_space()), so only (1) and (2) are affected.

Keying the decision off the socket state is not enough, because the
socket is already in TCP_SYN_SENT before tcp_connect_init() runs.
SO_RCVBUF applied in that gap - through TCP_FASTOPEN_CONNECT before the
first write(), or through bpf_setsockopt() in a
BPF_SOCK_OPS_TCP_CONNECT_CB program (as samples/bpf/tcp_clamp_kern.c and
friends do) - still lands before tcp_connect_init() and still overrides
the metric.

Record instead whether window_clamp was set explicitly by the user via
setsockopt(TCP_WINDOW_CLAMP), in a new tp->window_clamp_user bit, and key
tcp_connect_init() and tcp_openreq_init_rwin() off that bit rather than
off a non-zero window_clamp. The route metric is then applied on all
pre-connection paths, regardless of socket state, whenever the clamp was
not set by the user; a passive socket inherits the decision from its
listener. SO_RCVBUF no longer overwrites a genuine TCP_WINDOW_CLAMP
either. When SOCK_RCVBUF_LOCK is set both functions still clamp
window_clamp to tcp_full_space(), and without a user clamp SO_RCVBUF
keeps updating window_clamp, so SO_RCVBUF on an established socket still
grows the advertised window as b025461303d8 intended.

Example: two netns joined by a veth pair, "ip route ... window 4096" on
the route to the peer, SO_RCVBUF set to 1 MiB before connect():

  before:  SYN win 65535, wscale 5
  after:   SYN win 2920,  wscale 0

(the window scale option is still sent, with a shift of 0.)

Fixes: b025461303d8 ("tcp: update window_clamp when SO_RCVBUF is set")
Assisted-by: LLM
Signed-off-by: Shashank Mohan Jain <jain.sm@gmail.com>
---
 include/linux/tcp.h      |  3 ++-
 net/ipv4/tcp.c           | 12 ++++++++++++
 net/ipv4/tcp_minisocks.c |  7 +++++--
 net/ipv4/tcp_output.c    |  2 +-
 4 files changed, 20 insertions(+), 4 deletions(-)

diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index e4d1720f4..2324841e5 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -301,7 +301,8 @@ struct tcp_sock {
  *	0x5?10 << 16 + snd_wnd in net byte order
  */
 	u8	nonagle     : 4,/* Disable Nagle algorithm?             */
-		rate_app_limited:1;  /* rate_{delivered,interval_us} limited? */
+		rate_app_limited:1,  /* rate_{delivered,interval_us} limited? */
+		window_clamp_user:1; /* TCP_WINDOW_CLAMP set by the user */
 	u8	received_ce_pending:4, /* Not yet transmit cnt of received_ce */
 		accecn_opt_sent_w_dsack:1,/* Sent ACCECN opt in previous ACK w/ D-SACK */
 		unused2:3;
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index f4c9101e5..5f364928b 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1858,6 +1858,15 @@ int tcp_set_rcvlowat(struct sock *sk, int val)
 
 void tcp_set_rcvbuf(struct sock *sk, int val)
 {
+	/* A window clamp set explicitly via setsockopt(TCP_WINDOW_CLAMP)
+	 * takes precedence: SO_RCVBUF must not overwrite it.  Without such
+	 * a user clamp, keep updating window_clamp from the buffer size so
+	 * that SO_RCVBUF on an established socket still grows the advertised
+	 * window (commit b025461303d8).
+	 */
+	if (tcp_sk(sk)->window_clamp_user)
+		return;
+
 	tcp_set_window_clamp(sk, tcp_win_from_space(sk, val));
 }
 
@@ -3427,6 +3436,7 @@ int tcp_disconnect(struct sock *sk, int flags)
 	tp->is_cwnd_limited = 0;
 	tp->max_packets_out = 0;
 	tp->window_clamp = 0;
+	tp->window_clamp_user = 0;
 	tp->delivered = 0;
 	tp->delivered_ce = 0;
 	tp->accecn_fail_mode = 0;
@@ -4056,6 +4066,8 @@ int do_tcp_setsockopt(struct sock *sk, int level, int optname,
 
 	case TCP_WINDOW_CLAMP:
 		err = tcp_set_window_clamp(sk, val);
+		if (!err)
+			tp->window_clamp_user = !!val;
 		break;
 
 	case TCP_QUICKACK:
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 0ddfd5af6..c24dc975a 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -442,8 +442,11 @@ void tcp_openreq_init_rwin(struct request_sock *req,
 
 	mss = tcp_mss_clamp(tp, tcp_dst_advmss(dst));
 	window_clamp = READ_ONCE(tp->window_clamp);
-	/* Set this up on the first call only */
-	req->rsk_window_clamp = window_clamp ? : dst_metric(dst, RTAX_WINDOW);
+	/* Set this up on the first call only.  Honour a window clamp the
+	 * listener set via TCP_WINDOW_CLAMP; otherwise use the route metric.
+	 */
+	req->rsk_window_clamp = tp->window_clamp_user ? window_clamp :
+				dst_metric(dst, RTAX_WINDOW);
 
 	/* limit the window selection if the user enforce a smaller rx buffer */
 	if (sk_listener->sk_userlocks & SOCK_RCVBUF_LOCK &&
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index c8865205c..21b4ae303 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -4137,7 +4137,7 @@ static void tcp_connect_init(struct sock *sk)
 
 	tcp_ca_dst_init(sk, dst);
 
-	if (!tp->window_clamp)
+	if (!tp->window_clamp_user)
 		WRITE_ONCE(tp->window_clamp, dst_metric(dst, RTAX_WINDOW));
 	tp->advmss = tcp_mss_clamp(tp, tcp_dst_advmss(dst));
 
-- 
2.43.0