Merge branch 'DSA-mtu'

Vladimir Oltean says:

====================
Configure the MTU on DSA switches

This series adds support for configuring the MTU on front-panel switch
ports, while seamlessly adapting the CPU port and the DSA master to the
largest value plus the tagger overhead.

It also implements bridge MTU auto-normalization within the DSA core, as
resulted after the feedback of the implementation of this feature inside
the bridge driver in v2.

Support was added for quite a number of switches, in the hope that this
series would gain some traction:
 - sja1105
 - felix
 - vsc73xx
 - b53 and rest of the platform

V3 of this series was submitted here:
https://patchwork.ozlabs.org/cover/1262394/

V2 of this series was submitted here:
https://patchwork.ozlabs.org/cover/1261471/

V1 of this series was submitted here:
https://patchwork.ozlabs.org/cover/1199868/
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/dsa/b53/b53_common.c

@@ -699,9 +699,6 @@ int b53_configure_vlan(struct dsa_switch *ds)
 		b53_write16(dev, B53_VLAN_PAGE,
 			    B53_VLAN_PORT_DEF_TAG(i), def_vid);
 
-	if (!is5325(dev) && !is5365(dev))
-		b53_set_jumbo(dev, dev->enable_jumbo, false);
-
 	return 0;
 }
 EXPORT_SYMBOL(b53_configure_vlan);
@@ -807,8 +804,6 @@ static int b53_phy_write16(struct dsa_switch *ds, int addr, int reg, u16 val)
 static int b53_reset_switch(struct b53_device *priv)
 {
 	/* reset vlans */
-	priv->enable_jumbo = false;
-
 	memset(priv->vlans, 0, sizeof(*priv->vlans) * priv->num_vlans);
 	memset(priv->ports, 0, sizeof(*priv->ports) * priv->num_ports);
 
@@ -2065,6 +2060,26 @@ int b53_set_mac_eee(struct dsa_switch *ds, int port, struct ethtool_eee *e)
 }
 EXPORT_SYMBOL(b53_set_mac_eee);
 
+static int b53_change_mtu(struct dsa_switch *ds, int port, int mtu)
+{
+	struct b53_device *dev = ds->priv;
+	bool enable_jumbo;
+	bool allow_10_100;
+
+	if (is5325(dev) || is5365(dev))
+		return -EOPNOTSUPP;
+
+	enable_jumbo = (mtu >= JMS_MIN_SIZE);
+	allow_10_100 = (dev->chip_id == BCM583XX_DEVICE_ID);
+
+	return b53_set_jumbo(dev, enable_jumbo, allow_10_100);
+}
+
+static int b53_get_max_mtu(struct dsa_switch *ds, int port)
+{
+	return JMS_MAX_SIZE;
+}
+
 static const struct dsa_switch_ops b53_switch_ops = {
 	.get_tag_protocol	= b53_get_tag_protocol,
 	.setup			= b53_setup,
@@ -2102,6 +2117,8 @@ static const struct dsa_switch_ops b53_switch_ops = {
 	.port_mdb_prepare	= b53_mdb_prepare,
 	.port_mdb_add		= b53_mdb_add,
 	.port_mdb_del		= b53_mdb_del,
+	.port_max_mtu		= b53_get_max_mtu,
+	.port_change_mtu	= b53_change_mtu,
 };
 
 struct b53_chip_data {

drivers/net/dsa/ocelot/felix.c

@@ -532,6 +532,7 @@ static int felix_setup(struct dsa_switch *ds)
 			 ANA_PGID_PGID_PGID(GENMASK(ocelot->num_phys_ports, 0)),
 			 ANA_PGID_PGID, PGID_UC);
 
+	ds->mtu_enforcement_ingress = true;
 	/* It looks like the MAC/PCS interrupt register - PM0_IEVENT (0x8040)
 	 * isn't instantiated for the Felix PF.
 	 * In-band AN may take a few ms to complete, so we need to poll.
@@ -609,6 +610,22 @@ static bool felix_txtstamp(struct dsa_switch *ds, int port,
 	return false;
 }
 
+static int felix_change_mtu(struct dsa_switch *ds, int port, int new_mtu)
+{
+	struct ocelot *ocelot = ds->priv;
+
+	ocelot_port_set_maxlen(ocelot, port, new_mtu);
+
+	return 0;
+}
+
+static int felix_get_max_mtu(struct dsa_switch *ds, int port)
+{
+	struct ocelot *ocelot = ds->priv;
+
+	return ocelot_get_max_mtu(ocelot, port);
+}
+
 static int felix_cls_flower_add(struct dsa_switch *ds, int port,
 				struct flow_cls_offload *cls, bool ingress)
 {
@@ -664,6 +681,8 @@ static const struct dsa_switch_ops felix_switch_ops = {
 	.port_hwtstamp_set	= felix_hwtstamp_set,
 	.port_rxtstamp		= felix_rxtstamp,
 	.port_txtstamp		= felix_txtstamp,
+	.port_change_mtu	= felix_change_mtu,
+	.port_max_mtu		= felix_get_max_mtu,
 	.cls_flower_add		= felix_cls_flower_add,
 	.cls_flower_del		= felix_cls_flower_del,
 	.cls_flower_stats	= felix_cls_flower_stats,

drivers/net/dsa/sja1105/sja1105.h

@@ -126,6 +126,7 @@ enum sja1105_reset_reason {
 	SJA1105_RX_HWTSTAMPING,
 	SJA1105_AGEING_TIME,
 	SJA1105_SCHEDULING,
+	SJA1105_BEST_EFFORT_POLICING,
 };
 
 int sja1105_static_config_reload(struct sja1105_private *priv,

drivers/net/dsa/sja1105/sja1105_main.c

@@ -519,12 +519,12 @@ static int sja1105_init_avb_params(struct sja1105_private *priv)
 #define SJA1105_RATE_MBPS(speed) (((speed) * 64000) / 1000)
 
 static void sja1105_setup_policer(struct sja1105_l2_policing_entry *policing,
-				  int index)
+				  int index, int mtu)
 {
 	policing[index].sharindx = index;
 	policing[index].smax = 65535; /* Burst size in bytes */
 	policing[index].rate = SJA1105_RATE_MBPS(1000);
-	policing[index].maxlen = ETH_FRAME_LEN + VLAN_HLEN + ETH_FCS_LEN;
+	policing[index].maxlen = mtu;
 	policing[index].partition = 0;
 }
 
@@ -556,12 +556,16 @@ static int sja1105_init_l2_policing(struct sja1105_private *priv)
 	 */
 	for (i = 0, k = 0; i < SJA1105_NUM_PORTS; i++) {
 		int bcast = (SJA1105_NUM_PORTS * SJA1105_NUM_TC) + i;
+		int mtu = VLAN_ETH_FRAME_LEN + ETH_FCS_LEN;
+
+		if (dsa_is_cpu_port(priv->ds, i))
+			mtu += VLAN_HLEN;
 
 		for (j = 0; j < SJA1105_NUM_TC; j++, k++)
-			sja1105_setup_policer(policing, k);
+			sja1105_setup_policer(policing, k, mtu);
 
 		/* Set up this port's policer for broadcast traffic */
-		sja1105_setup_policer(policing, bcast);
+		sja1105_setup_policer(policing, bcast, mtu);
 	}
 	return 0;
 }
@@ -1544,6 +1548,7 @@ static const char * const sja1105_reset_reasons[] = {
 	[SJA1105_RX_HWTSTAMPING] = "RX timestamping",
 	[SJA1105_AGEING_TIME] = "Ageing time",
 	[SJA1105_SCHEDULING] = "Time-aware scheduling",
+	[SJA1105_BEST_EFFORT_POLICING] = "Best-effort policing",
 };
 
 /* For situations where we need to change a setting at runtime that is only
@@ -1952,6 +1957,8 @@ static int sja1105_setup(struct dsa_switch *ds)
 	/* Advertise the 8 egress queues */
 	ds->num_tx_queues = SJA1105_NUM_TC;
 
+	ds->mtu_enforcement_ingress = true;
+
 	/* The DSA/switchdev model brings up switch ports in standalone mode by
 	 * default, and that means vlan_filtering is 0 since they're not under
 	 * a bridge, so it's safe to set up switch tagging at this time.
@@ -2120,6 +2127,39 @@ static int sja1105_set_ageing_time(struct dsa_switch *ds,
 	return sja1105_static_config_reload(priv, SJA1105_AGEING_TIME);
 }
 
+static int sja1105_change_mtu(struct dsa_switch *ds, int port, int new_mtu)
+{
+	int bcast = (SJA1105_NUM_PORTS * SJA1105_NUM_TC) + port;
+	struct sja1105_l2_policing_entry *policing;
+	struct sja1105_private *priv = ds->priv;
+	int tc;
+
+	new_mtu += VLAN_ETH_HLEN + ETH_FCS_LEN;
+
+	if (dsa_is_cpu_port(ds, port))
+		new_mtu += VLAN_HLEN;
+
+	policing = priv->static_config.tables[BLK_IDX_L2_POLICING].entries;
+
+	/* We set all 9 port policers to the same value, so just checking the
+	 * broadcast one is fine.
+	 */
+	if (policing[bcast].maxlen == new_mtu)
+		return 0;
+
+	for (tc = 0; tc < SJA1105_NUM_TC; tc++)
+		policing[port * SJA1105_NUM_TC + tc].maxlen = new_mtu;
+
+	policing[bcast].maxlen = new_mtu;
+
+	return sja1105_static_config_reload(priv, SJA1105_BEST_EFFORT_POLICING);
+}
+
+static int sja1105_get_max_mtu(struct dsa_switch *ds, int port)
+{
+	return 2043 - VLAN_ETH_HLEN - ETH_FCS_LEN;
+}
+
 static int sja1105_port_setup_tc(struct dsa_switch *ds, int port,
 				 enum tc_setup_type type,
 				 void *type_data)
@@ -2215,6 +2255,8 @@ static const struct dsa_switch_ops sja1105_switch_ops = {
 	.setup			= sja1105_setup,
 	.teardown		= sja1105_teardown,
 	.set_ageing_time	= sja1105_set_ageing_time,
+	.port_change_mtu	= sja1105_change_mtu,
+	.port_max_mtu		= sja1105_get_max_mtu,
 	.phylink_validate	= sja1105_phylink_validate,
 	.phylink_mac_link_state	= sja1105_mac_pcs_get_state,
 	.phylink_mac_config	= sja1105_mac_config,

drivers/net/dsa/vitesse-vsc73xx-core.c

@@ -664,16 +664,6 @@ static void vsc73xx_init_port(struct vsc73xx *vsc, int port)
 		      VSC73XX_MAC_CFG_TX_EN |
 		      VSC73XX_MAC_CFG_RX_EN);
 
-	/* Max length, we can do up to 9.6 KiB, so allow that.
-	 * According to application not "VSC7398 Jumbo Frames" setting
-	 * up the MTU to 9.6 KB does not affect the performance on standard
-	 * frames, so just enable it. It is clear from the application note
-	 * that "9.6 kilobytes" == 9600 bytes.
-	 */
-	vsc73xx_write(vsc, VSC73XX_BLOCK_MAC,
-		      port,
-		      VSC73XX_MAXLEN, 9600);
-
 	/* Flow control for the CPU port:
 	 * Use a zero delay pause frame when pause condition is left
 	 * Obey pause control frames
@@ -1030,6 +1020,24 @@ static void vsc73xx_get_ethtool_stats(struct dsa_switch *ds, int port,
 	}
 }
 
+static int vsc73xx_change_mtu(struct dsa_switch *ds, int port, int new_mtu)
+{
+	struct vsc73xx *vsc = ds->priv;
+
+	return vsc73xx_write(vsc, VSC73XX_BLOCK_MAC, port,
+			     VSC73XX_MAXLEN, new_mtu);
+}
+
+/* According to application not "VSC7398 Jumbo Frames" setting
+ * up the MTU to 9.6 KB does not affect the performance on standard
+ * frames. It is clear from the application note that
+ * "9.6 kilobytes" == 9600 bytes.
+ */
+static int vsc73xx_get_max_mtu(struct dsa_switch *ds, int port)
+{
+	return 9600;
+}
+
 static const struct dsa_switch_ops vsc73xx_ds_ops = {
 	.get_tag_protocol = vsc73xx_get_tag_protocol,
 	.setup = vsc73xx_setup,
@@ -1041,6 +1049,8 @@ static const struct dsa_switch_ops vsc73xx_ds_ops = {
 	.get_sset_count = vsc73xx_get_sset_count,
 	.port_enable = vsc73xx_port_enable,
 	.port_disable = vsc73xx_port_disable,
+	.port_change_mtu = vsc73xx_change_mtu,
+	.port_max_mtu = vsc73xx_get_max_mtu,
 };
 
 static int vsc73xx_gpio_get(struct gpio_chip *chip, unsigned int offset)

drivers/net/ethernet/broadcom/bgmac.c

@@ -1248,6 +1248,14 @@ static int bgmac_set_mac_address(struct net_device *net_dev, void *addr)
 	return 0;
 }
 
+static int bgmac_change_mtu(struct net_device *net_dev, int mtu)
+{
+	struct bgmac *bgmac = netdev_priv(net_dev);
+
+	bgmac_write(bgmac, BGMAC_RXMAX_LENGTH, 32 + mtu);
+	return 0;
+}
+
 static const struct net_device_ops bgmac_netdev_ops = {
 	.ndo_open		= bgmac_open,
 	.ndo_stop		= bgmac_stop,
@@ -1256,6 +1264,7 @@ static const struct net_device_ops bgmac_netdev_ops = {
 	.ndo_set_mac_address	= bgmac_set_mac_address,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_do_ioctl           = phy_do_ioctl_running,
+	.ndo_change_mtu		= bgmac_change_mtu,
 };
 
 /**************************************************
@@ -1530,6 +1539,9 @@ int bgmac_enet_probe(struct bgmac *bgmac)
 	net_dev->hw_features = net_dev->features;
 	net_dev->vlan_features = net_dev->features;
 
+	/* Omit FCS from max MTU size */
+	net_dev->max_mtu = BGMAC_RX_MAX_FRAME_SIZE - ETH_FCS_LEN;
+
 	err = register_netdev(bgmac->net_dev);
 	if (err) {
 		dev_err(bgmac->dev, "Cannot register net device\n");

drivers/net/ethernet/broadcom/bgmac.h

@@ -351,7 +351,7 @@
 #define BGMAC_DESC_CTL0_IOC			0x20000000	/* IRQ on complete */
 #define BGMAC_DESC_CTL0_EOF			0x40000000	/* End of frame */
 #define BGMAC_DESC_CTL0_SOF			0x80000000	/* Start of frame */
-#define BGMAC_DESC_CTL1_LEN			0x00001FFF
+#define BGMAC_DESC_CTL1_LEN			0x00003FFF
 
 #define BGMAC_PHY_NOREGS			BRCM_PSEUDO_PHY_ADDR
 #define BGMAC_PHY_MASK				0x1F
@@ -366,7 +366,8 @@
 #define BGMAC_RX_FRAME_OFFSET			30		/* There are 2 unused bytes between header and real data */
 #define BGMAC_RX_BUF_OFFSET			(NET_SKB_PAD + NET_IP_ALIGN - \
 						 BGMAC_RX_FRAME_OFFSET)
-#define BGMAC_RX_MAX_FRAME_SIZE			1536		/* Copied from b44/tg3 */
+/* Jumbo frame size with FCS */
+#define BGMAC_RX_MAX_FRAME_SIZE			9724
 #define BGMAC_RX_BUF_SIZE			(BGMAC_RX_FRAME_OFFSET + BGMAC_RX_MAX_FRAME_SIZE)
 #define BGMAC_RX_ALLOC_SIZE			(SKB_DATA_ALIGN(BGMAC_RX_BUF_SIZE + BGMAC_RX_BUF_OFFSET) + \
 						 SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))

drivers/net/ethernet/mscc/ocelot.c

@@ -2185,13 +2185,25 @@ static int ocelot_init_timestamp(struct ocelot *ocelot)
 
 /* Configure the maximum SDU (L2 payload) on RX to the value specified in @sdu.
  * The length of VLAN tags is accounted for automatically via DEV_MAC_TAGS_CFG.
+ * In the special case that it's the NPI port that we're configuring, the
+ * length of the tag and optional prefix needs to be accounted for privately,
+ * in order to be able to sustain communication at the requested @sdu.
  */
-static void ocelot_port_set_maxlen(struct ocelot *ocelot, int port, size_t sdu)
+void ocelot_port_set_maxlen(struct ocelot *ocelot, int port, size_t sdu)
 {
 	struct ocelot_port *ocelot_port = ocelot->ports[port];
 	int maxlen = sdu + ETH_HLEN + ETH_FCS_LEN;
 	int atop_wm;
 
+	if (port == ocelot->npi) {
+		maxlen += OCELOT_TAG_LEN;
+
+		if (ocelot->inj_prefix == OCELOT_TAG_PREFIX_SHORT)
+			maxlen += OCELOT_SHORT_PREFIX_LEN;
+		else if (ocelot->inj_prefix == OCELOT_TAG_PREFIX_LONG)
+			maxlen += OCELOT_LONG_PREFIX_LEN;
+	}
+
 	ocelot_port_writel(ocelot_port, maxlen, DEV_MAC_MAXLEN_CFG);
 
 	/* Set Pause WM hysteresis
@@ -2209,6 +2221,24 @@ static void ocelot_port_set_maxlen(struct ocelot *ocelot, int port, size_t sdu)
 			 SYS_ATOP, port);
 	ocelot_write(ocelot, ocelot_wm_enc(atop_wm), SYS_ATOP_TOT_CFG);
 }
+EXPORT_SYMBOL(ocelot_port_set_maxlen);
+
+int ocelot_get_max_mtu(struct ocelot *ocelot, int port)
+{
+	int max_mtu = 65535 - ETH_HLEN - ETH_FCS_LEN;
+
+	if (port == ocelot->npi) {
+		max_mtu -= OCELOT_TAG_LEN;
+
+		if (ocelot->inj_prefix == OCELOT_TAG_PREFIX_SHORT)
+			max_mtu -= OCELOT_SHORT_PREFIX_LEN;
+		else if (ocelot->inj_prefix == OCELOT_TAG_PREFIX_LONG)
+			max_mtu -= OCELOT_LONG_PREFIX_LEN;
+	}
+
+	return max_mtu;
+}
+EXPORT_SYMBOL(ocelot_get_max_mtu);
 
 void ocelot_init_port(struct ocelot *ocelot, int port)
 {
@@ -2318,6 +2348,10 @@ void ocelot_configure_cpu(struct ocelot *ocelot, int npi,
 {
 	int cpu = ocelot->num_phys_ports;
 
+	ocelot->npi = npi;
+	ocelot->inj_prefix = injection;
+	ocelot->xtr_prefix = extraction;
+
 	/* The unicast destination PGID for the CPU port module is unused */
 	ocelot_write_rix(ocelot, 0, ANA_PGID_PGID, cpu);
 	/* Instead set up a multicast destination PGID for traffic copied to
@@ -2330,19 +2364,10 @@ void ocelot_configure_cpu(struct ocelot *ocelot, int npi,
 			 ANA_PORT_PORT_CFG, cpu);
 
 	if (npi >= 0 && npi < ocelot->num_phys_ports) {
-		int sdu = ETH_DATA_LEN + OCELOT_TAG_LEN;
-
 		ocelot_write(ocelot, QSYS_EXT_CPU_CFG_EXT_CPUQ_MSK_M |
 			     QSYS_EXT_CPU_CFG_EXT_CPU_PORT(npi),
 			     QSYS_EXT_CPU_CFG);
 
-		if (injection == OCELOT_TAG_PREFIX_SHORT)
-			sdu += OCELOT_SHORT_PREFIX_LEN;
-		else if (injection == OCELOT_TAG_PREFIX_LONG)
-			sdu += OCELOT_LONG_PREFIX_LEN;
-
-		ocelot_port_set_maxlen(ocelot, npi, sdu);
-
 		/* Enable NPI port */
 		ocelot_write_rix(ocelot,
 				 QSYS_SWITCH_PORT_MODE_INGRESS_DROP_MODE |