From: Vladimir Oltean <olteanv@gmail.com>
To: f.fainelli@gmail.com, vivien.didelot@gmail.com, andrew@lunn.ch,
davem@davemloft.net
Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
Vladimir Oltean <olteanv@gmail.com>
Subject: [PATCH net-next 8/9] net: dsa: sja1105: Add support for Spanning Tree Protocol
Date: Sat, 4 May 2019 04:18:25 +0300 [thread overview]
Message-ID: <20190504011826.30477-9-olteanv@gmail.com> (raw)
In-Reply-To: <20190504011826.30477-1-olteanv@gmail.com>
While not explicitly documented as supported in UM10944, compliance with
the STP states can be obtained by manipulating 3 settings at the
(per-port) MAC config level: dynamic learning, inhibiting reception of
regular traffic, and inhibiting transmission of regular traffic.
In all these modes, transmission and reception of special BPDU frames
from the stack is still enabled (not inhibited by the MAC-level
settings).
On ingress, BPDUs are classified by the MAC filter as link-local
(01-80-C2-00-00-00) and forwarded to the CPU port. This mechanism works
under all conditions (even without the custom 802.1Q tagging) because
the switch hardware inserts the source port and switch ID into bytes 4
and 5 of the MAC-filtered frames. Then the DSA .rcv handler needs to put
back zeroes into the MAC address after decoding the source port
information.
On egress, BPDUs are transmitted using management routes from the xmit
worker thread. Again this does not require switch tagging, as the switch
port is programmed through SPI to hold a temporary (single-fire) route
for a frame with the programmed destination MAC (01-80-C2-00-00-00).
STP is activated using the following commands and was tested by
connecting two front-panel ports together and noticing that switching
loops were prevented (one port remains in the blocking state):
$ ip link add name br0 type bridge stp_state 1 && ip link set br0 up
$ for eth in $(ls /sys/devices/platform/soc/2100000.spi/spi_master/spi0/spi0.1/net/);
do ip link set ${eth} master br0 && ip link set ${eth} up; done
Signed-off-by: Vladimir Oltean <olteanv@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
---
drivers/net/dsa/sja1105/sja1105_main.c | 108 ++++++++++++++++++++++---
1 file changed, 99 insertions(+), 9 deletions(-)
diff --git a/drivers/net/dsa/sja1105/sja1105_main.c b/drivers/net/dsa/sja1105/sja1105_main.c
index 785bb42cb993..50ff625c85d6 100644
--- a/drivers/net/dsa/sja1105/sja1105_main.c
+++ b/drivers/net/dsa/sja1105/sja1105_main.c
@@ -92,8 +92,10 @@ static int sja1105_init_mac_settings(struct sja1105_private *priv)
.drpuntag = false,
/* Don't retag 802.1p (VID 0) traffic with the pvid */
.retag = false,
- /* Enable learning and I/O on user ports by default. */
- .dyn_learn = true,
+ /* Disable learning and I/O on user ports by default -
+ * STP will enable it.
+ */
+ .dyn_learn = false,
.egress = false,
.ingress = false,
};
@@ -119,8 +121,17 @@ static int sja1105_init_mac_settings(struct sja1105_private *priv)
mac = table->entries;
- for (i = 0; i < SJA1105_NUM_PORTS; i++)
+ for (i = 0; i < SJA1105_NUM_PORTS; i++) {
mac[i] = default_mac;
+ if (i == dsa_upstream_port(priv->ds, i)) {
+ /* STP doesn't get called for CPU port, so we need to
+ * set the I/O parameters statically.
+ */
+ mac[i].dyn_learn = true;
+ mac[i].ingress = true;
+ mac[i].egress = true;
+ }
+ }
return 0;
}
@@ -655,12 +666,14 @@ static sja1105_speed_t sja1105_get_speed_cfg(unsigned int speed_mbps)
* for a specific port.
*
* @speed_mbps: If 0, leave the speed unchanged, else adapt MAC to PHY speed.
- * @enabled: Manage Rx and Tx settings for this port. Overrides the static
- * configuration settings.
+ * @enabled: Manage Rx and Tx settings for this port. If false, overrides the
+ * settings from the STP state, but not persistently (does not
+ * overwrite the static MAC info for this port).
*/
static int sja1105_adjust_port_config(struct sja1105_private *priv, int port,
int speed_mbps, bool enabled)
{
+ struct sja1105_mac_config_entry dyn_mac;
struct sja1105_xmii_params_entry *mii;
struct sja1105_mac_config_entry *mac;
struct device *dev = priv->ds->dev;
@@ -693,12 +706,13 @@ static int sja1105_adjust_port_config(struct sja1105_private *priv, int port,
* the code common, we'll use the static configuration tables as a
* reasonable approximation for both E/T and P/Q/R/S.
*/
- mac[port].ingress = enabled;
- mac[port].egress = enabled;
+ dyn_mac = mac[port];
+ dyn_mac.ingress = enabled && mac[port].ingress;
+ dyn_mac.egress = enabled && mac[port].egress;
/* Write to the dynamic reconfiguration tables */
rc = sja1105_dynamic_config_write(priv, BLK_IDX_MAC_CONFIG,
- port, &mac[port], true);
+ port, &dyn_mac, true);
if (rc < 0) {
dev_err(dev, "Failed to write MAC config: %d\n", rc);
return rc;
@@ -986,6 +1000,50 @@ static int sja1105_bridge_member(struct dsa_switch *ds, int port,
port, &l2_fwd[port], true);
}
+static void sja1105_bridge_stp_state_set(struct dsa_switch *ds, int port,
+ u8 state)
+{
+ struct sja1105_private *priv = ds->priv;
+ struct sja1105_mac_config_entry *mac;
+
+ mac = priv->static_config.tables[BLK_IDX_MAC_CONFIG].entries;
+
+ switch (state) {
+ case BR_STATE_DISABLED:
+ case BR_STATE_BLOCKING:
+ /* From UM10944 description of DRPDTAG (why put this there?):
+ * "Management traffic flows to the port regardless of the state
+ * of the INGRESS flag". So BPDUs are still be allowed to pass.
+ * At the moment no difference between DISABLED and BLOCKING.
+ */
+ mac[port].ingress = false;
+ mac[port].egress = false;
+ mac[port].dyn_learn = false;
+ break;
+ case BR_STATE_LISTENING:
+ mac[port].ingress = true;
+ mac[port].egress = false;
+ mac[port].dyn_learn = false;
+ break;
+ case BR_STATE_LEARNING:
+ mac[port].ingress = true;
+ mac[port].egress = false;
+ mac[port].dyn_learn = true;
+ break;
+ case BR_STATE_FORWARDING:
+ mac[port].ingress = true;
+ mac[port].egress = true;
+ mac[port].dyn_learn = true;
+ break;
+ default:
+ dev_err(ds->dev, "invalid STP state: %d\n", state);
+ return;
+ }
+
+ sja1105_dynamic_config_write(priv, BLK_IDX_MAC_CONFIG, port,
+ &mac[port], true);
+}
+
static int sja1105_bridge_join(struct dsa_switch *ds, int port,
struct net_device *br)
{
@@ -998,6 +1056,23 @@ static void sja1105_bridge_leave(struct dsa_switch *ds, int port,
sja1105_bridge_member(ds, port, br, false);
}
+static u8 sja1105_stp_state_get(struct sja1105_private *priv, int port)
+{
+ struct sja1105_mac_config_entry *mac;
+
+ mac = priv->static_config.tables[BLK_IDX_MAC_CONFIG].entries;
+
+ if (!mac[port].ingress && !mac[port].egress && !mac[port].dyn_learn)
+ return BR_STATE_BLOCKING;
+ if (mac[port].ingress && !mac[port].egress && !mac[port].dyn_learn)
+ return BR_STATE_LISTENING;
+ if (mac[port].ingress && !mac[port].egress && mac[port].dyn_learn)
+ return BR_STATE_LEARNING;
+ if (mac[port].ingress && mac[port].egress && mac[port].dyn_learn)
+ return BR_STATE_FORWARDING;
+ return -EINVAL;
+}
+
/* For situations where we need to change a setting at runtime that is only
* available through the static configuration, resetting the switch in order
* to upload the new static config is unavoidable. Back up the settings we
@@ -1008,16 +1083,27 @@ static int sja1105_static_config_reload(struct sja1105_private *priv)
{
struct sja1105_mac_config_entry *mac;
int speed_mbps[SJA1105_NUM_PORTS];
+ u8 stp_state[SJA1105_NUM_PORTS];
int rc, i;
mac = priv->static_config.tables[BLK_IDX_MAC_CONFIG].entries;
/* Back up settings changed by sja1105_adjust_port_config and
- * and restore their defaults.
+ * sja1105_bridge_stp_state_set and restore their defaults.
*/
for (i = 0; i < SJA1105_NUM_PORTS; i++) {
speed_mbps[i] = sja1105_speed[mac[i].speed];
mac[i].speed = SJA1105_SPEED_AUTO;
+ if (i == dsa_upstream_port(priv->ds, i)) {
+ mac[i].ingress = true;
+ mac[i].egress = true;
+ mac[i].dyn_learn = true;
+ } else {
+ stp_state[i] = sja1105_stp_state_get(priv, i);
+ mac[i].ingress = false;
+ mac[i].egress = false;
+ mac[i].dyn_learn = false;
+ }
}
/* Reset switch and send updated static configuration */
@@ -1036,6 +1122,9 @@ static int sja1105_static_config_reload(struct sja1105_private *priv)
for (i = 0; i < SJA1105_NUM_PORTS; i++) {
bool enabled = (speed_mbps[i] != 0);
+ if (i != dsa_upstream_port(priv->ds, i))
+ sja1105_bridge_stp_state_set(priv->ds, i, stp_state[i]);
+
rc = sja1105_adjust_port_config(priv, i, speed_mbps[i],
enabled);
if (rc < 0)
@@ -1433,6 +1522,7 @@ static const struct dsa_switch_ops sja1105_switch_ops = {
.port_fdb_del = sja1105_fdb_del,
.port_bridge_join = sja1105_bridge_join,
.port_bridge_leave = sja1105_bridge_leave,
+ .port_stp_state_set = sja1105_bridge_stp_state_set,
.port_vlan_prepare = sja1105_vlan_prepare,
.port_vlan_filtering = sja1105_vlan_filtering,
.port_vlan_add = sja1105_vlan_add,
--
2.17.1
next prev parent reply other threads:[~2019-05-04 1:19 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-05-04 1:18 [PATCH net-next 0/9] Traffic support for SJA1105 DSA driver Vladimir Oltean
2019-05-04 1:18 ` [PATCH net-next 1/9] net: dsa: Call driver's setup callback after setting up its switchdev notifier Vladimir Oltean
2019-05-04 1:18 ` [PATCH net-next 2/9] net: dsa: Optional VLAN-based port separation for switches without tagging Vladimir Oltean
2019-05-04 1:18 ` [PATCH net-next 3/9] net: dsa: Allow drivers to filter packets they can decode source port from Vladimir Oltean
2019-05-04 2:00 ` Florian Fainelli
2019-05-04 1:18 ` [PATCH net-next 4/9] net: dsa: Keep private info in the skb->cb Vladimir Oltean
2019-05-04 2:04 ` Florian Fainelli
2019-05-04 2:23 ` Vladimir Oltean
2019-05-04 1:18 ` [PATCH net-next 5/9] net: dsa: Add support for deferred xmit Vladimir Oltean
2019-05-04 2:07 ` Florian Fainelli
2019-05-04 2:10 ` Vladimir Oltean
2019-05-04 1:18 ` [PATCH net-next 6/9] net: dsa: Add a private structure pointer to dsa_port Vladimir Oltean
2019-05-04 2:08 ` Florian Fainelli
2019-05-04 1:18 ` [PATCH net-next 7/9] net: dsa: sja1105: Add support for traffic through standalone ports Vladimir Oltean
2019-05-04 1:18 ` Vladimir Oltean [this message]
2019-05-04 2:10 ` [PATCH net-next 8/9] net: dsa: sja1105: Add support for Spanning Tree Protocol Florian Fainelli
2019-05-04 1:18 ` [PATCH net-next 9/9] Documentation: net: dsa: sja1105: Add info about supported traffic modes Vladimir Oltean
2019-05-04 2:17 ` Florian Fainelli
2019-05-04 2:36 ` Vladimir Oltean
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190504011826.30477-9-olteanv@gmail.com \
--to=olteanv@gmail.com \
--cc=andrew@lunn.ch \
--cc=davem@davemloft.net \
--cc=f.fainelli@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=vivien.didelot@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).