[PATCH net v2 0/2] Fix TCP retransmissions on NS/NSP SoCs

[PATCH net v2 0/2] Fix TCP retransmissions on NS/NSP SoCs

[Florian Fainelli]

On Northstar and Northstar Plus SoCs we would see about 1000 TCP
retransmissions per second due to a rate mismatch between bgmac and b53,
make sure that we run at the same speed on both side to avoid dropping
packets on the floor.

This issue was found during testing on a Northstar Plus reference board
(BCM958625HR) while running iperf3 to a test machine and noticing
retranmsisions:

# iperf3 -c 192.168.3.1
Connecting to host 192.168.3.1, port 5201
[  4] local 192.168.3.204 port 49856 connected to 192.168.3.1 port 5201
[ ID] Interval           Transfer     Bitrate         Retr  Cwnd
[  4]   0.00-1.00   sec   111 MBytes   933 Mbits/sec  915   99.0 KBytes
[  4]   1.00-2.00   sec   111 MBytes   928 Mbits/sec  909    109 KBytes
[  4]   2.00-3.00   sec   110 MBytes   925 Mbits/sec  1097   99.0 KBytes
[  4]   3.00-4.00   sec   111 MBytes   928 Mbits/sec  814   58.0 KBytes
[  4]   4.00-5.00   sec   111 MBytes   930 Mbits/sec  1187    109 KBytes
[  4]   5.00-6.00   sec   110 MBytes   923 Mbits/sec  893   72.1 KBytes
[  4]   6.00-7.00   sec   111 MBytes   929 Mbits/sec  843   67.9 KBytes
[  4]   7.00-8.00   sec   111 MBytes   928 Mbits/sec  920   79.2 KBytes
[  4]   8.00-9.00   sec   111 MBytes   929 Mbits/sec  1109   66.5 KBytes
[  4]   9.00-10.00  sec   111 MBytes   930 Mbits/sec  1055   63.6 KBytes
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate         Retr
[  4]   0.00-10.00  sec  1.09 GBytes   933 Mbits/sec  9742
sender
[  4]   0.00-10.01  sec  1.09 GBytes   933 Mbits/sec
receiver

Changes in v2:

- make it apply cleanly to net/main

Florian Fainelli (2):
  net: ethernet: bgmac: preserve forced 2.5 Gbps link speed
  net: dsa: b53: restore the NSP IMP port 2 Gbps mode

 drivers/net/dsa/b53/b53_common.c      | 10 +++++++---
 drivers/net/ethernet/broadcom/bgmac.c |  8 ++++++--
 2 files changed, 13 insertions(+), 5 deletions(-)

-- 
2.34.1

[PATCH net v2 1/2] net: ethernet: bgmac: preserve forced 2.5 Gbps link speed

[Florian Fainelli]

Northstar and Northstar Plus use a 2.5 Gbps internal link between the
AMAC and the switch.  BGMAC_FEAT_FORCE_SPEED_2500 makes bgmac_miiconfig()
configure that speed, but bgmac_adjust_link() subsequently replaces it
with the synthetic fixed PHY's 1 Gbps speed.

Use the forced speed when processing link changes so the MAC remains at
2.5 Gbps.  Together with configuring the switch IMP port for its 2 Gbps
mode, this eliminates the output queue discards that caused approximately
1000 TCP retransmissions per second during an iperf3 transmit test.

Fixes: f6a95a24957a ("net: ethernet: bgmac: Add platform device support")
Signed-off-by: Florian Fainelli <florian.fainelli@broadcom.com>
---
 drivers/net/ethernet/broadcom/bgmac.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bgmac.c b/drivers/net/ethernet/broadcom/bgmac.c
index 3e9c57196a39..11487ff48e0d 100644
--- a/drivers/net/ethernet/broadcom/bgmac.c
+++ b/drivers/net/ethernet/broadcom/bgmac.c
@@ -1415,11 +1415,15 @@ void bgmac_adjust_link(struct net_device *net_dev)
 {
 	struct bgmac *bgmac = netdev_priv(net_dev);
 	struct phy_device *phy_dev = net_dev->phydev;
+	int speed = phy_dev->speed;
 	bool update = false;
 
+	if (bgmac->feature_flags & BGMAC_FEAT_FORCE_SPEED_2500)
+		speed = SPEED_2500;
+
 	if (phy_dev->link) {
-		if (phy_dev->speed != bgmac->mac_speed) {
-			bgmac->mac_speed = phy_dev->speed;
+		if (speed != bgmac->mac_speed) {
+			bgmac->mac_speed = speed;
 			update = true;
 		}
 
-- 
2.34.1

[PATCH net v2 2/2] net: dsa: b53: restore the NSP IMP port 2 Gbps mode

[Florian Fainelli]

The Northstar Plus IMP port uses the switch's proprietary 2 Gbps mode for
its internal link to the AMAC.  Its fixed-link description reports 1 Gbps,
so applying that speed clears PORT_OVERRIDE_SPEED_2000M and leaves the two
ends of the internal link configured at different speeds.

Force the 2 Gbps mode and flow control for BCM5301x and NSP CPU ports.  The
mismatch caused switch output queue discards on both copper and SerDes
ports, with one discard for every TCP retransmission.  With both ends
restored to their high-speed modes, iperf3 reports no retransmissions.

Fixes: b6a8a5477fe9 ("net: dsa: b53: fix resetting speed and pause on forced link")
Signed-off-by: Florian Fainelli <florian.fainelli@broadcom.com>
---
 drivers/net/dsa/b53/b53_common.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c
index 0880310c9ce3..1f4964c28fc5 100644
--- a/drivers/net/dsa/b53/b53_common.c
+++ b/drivers/net/dsa/b53/b53_common.c
@@ -1663,9 +1663,13 @@ static void b53_phylink_mac_link_up(struct phylink_config *config,
 	}
 
 	if (mode == MLO_AN_FIXED) {
-		/* Force flow control on BCM5301x's CPU port */
-		if (is5301x(dev) && dsa_is_cpu_port(ds, port))
-			tx_pause = rx_pause = true;
+		/* Force flow control on BCM5301x and NSP CPU ports */
+		if ((is5301x(dev) || dev->chip_id == BCM58XX_DEVICE_ID) &&
+		    dsa_is_cpu_port(ds, port)) {
+			tx_pause = true;
+			rx_pause = true;
+			speed = 2000;
+		}
 
 		b53_force_port_config(dev, port, speed, duplex,
 				      tx_pause, rx_pause);
-- 
2.34.1