netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Vladimir Oltean <olteanv@gmail.com>
To: "David S . Miller" <davem@davemloft.net>,
	Jakub Kicinski <kuba@kernel.org>
Cc: netdev <netdev@vger.kernel.org>,
	Microchip Linux Driver Support <UNGLinuxDriver@microchip.com>,
	Alexandre Belloni <alexandre.belloni@bootlin.com>,
	Andrew Lunn <andrew@lunn.ch>,
	Florian Fainelli <f.fainelli@gmail.com>,
	Vivien Didelot <vivien.didelot@gmail.com>,
	Claudiu Manoil <claudiu.manoil@nxp.com>
Subject: [PATCH RESEND v3 net-next 08/12] net: mscc: ocelot: drop the use of the "lags" array
Date: Sat,  6 Feb 2021 00:02:17 +0200	[thread overview]
Message-ID: <20210205220221.255646-9-olteanv@gmail.com> (raw)
In-Reply-To: <20210205220221.255646-1-olteanv@gmail.com>

From: Vladimir Oltean <vladimir.oltean@nxp.com>

We can now simplify the implementation by always using ocelot_get_bond_mask
to look up the other ports that are offloading the same bonding interface
as us.

In ocelot_set_aggr_pgids, the code had a way to uniquely iterate through
LAGs. We need to achieve the same behavior by marking each LAG as visited,
which we do now by using a temporary 32-bit "visited" bitmask. This is
ok and we do not need dynamic memory allocation, because we know that
this switch architecture will not have more than 32 ports (the PGID port
masks are 32-bit anyway).

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
Changes in v3:
Using a "visited" bit mask to avoid memory allocation.

Changes in v2:
Context looks a bit different.

 drivers/net/ethernet/mscc/ocelot.c | 95 ++++++++++++------------------
 include/soc/mscc/ocelot.h          |  2 -
 2 files changed, 39 insertions(+), 58 deletions(-)

diff --git a/drivers/net/ethernet/mscc/ocelot.c b/drivers/net/ethernet/mscc/ocelot.c
index 5d765245c6d3..c906c449d2dd 100644
--- a/drivers/net/ethernet/mscc/ocelot.c
+++ b/drivers/net/ethernet/mscc/ocelot.c
@@ -957,21 +957,11 @@ void ocelot_apply_bridge_fwd_mask(struct ocelot *ocelot)
 			mask = GENMASK(ocelot->num_phys_ports - 1, 0);
 			mask &= ~cpu_fwd_mask;
 		} else if (ocelot->bridge_fwd_mask & BIT(port)) {
-			int lag;
+			struct net_device *bond = ocelot_port->bond;
 
 			mask = ocelot->bridge_fwd_mask & ~BIT(port);
-
-			for (lag = 0; lag < ocelot->num_phys_ports; lag++) {
-				unsigned long bond_mask = ocelot->lags[lag];
-
-				if (!bond_mask)
-					continue;
-
-				if (bond_mask & BIT(port)) {
-					mask &= ~bond_mask;
-					break;
-				}
-			}
+			if (bond)
+				mask &= ~ocelot_get_bond_mask(ocelot, bond);
 		} else {
 			/* Standalone ports forward only to DSA tag_8021q CPU
 			 * ports (if those exist), or to the hardware CPU port
@@ -1277,6 +1267,7 @@ EXPORT_SYMBOL(ocelot_port_bridge_leave);
 
 static void ocelot_set_aggr_pgids(struct ocelot *ocelot)
 {
+	unsigned long visited = GENMASK(ocelot->num_phys_ports - 1, 0);
 	int i, port, lag;
 
 	/* Reset destination and aggregation PGIDS */
@@ -1287,16 +1278,35 @@ static void ocelot_set_aggr_pgids(struct ocelot *ocelot)
 		ocelot_write_rix(ocelot, GENMASK(ocelot->num_phys_ports - 1, 0),
 				 ANA_PGID_PGID, i);
 
-	/* Now, set PGIDs for each LAG */
+	/* The visited ports bitmask holds the list of ports offloading any
+	 * bonding interface. Initially we mark all these ports as unvisited,
+	 * then every time we visit a port in this bitmask, we know that it is
+	 * the lowest numbered port, i.e. the one whose logical ID == physical
+	 * port ID == LAG ID. So we mark as visited all further ports in the
+	 * bitmask that are offloading the same bonding interface. This way,
+	 * we set up the aggregation PGIDs only once per bonding interface.
+	 */
+	for (port = 0; port < ocelot->num_phys_ports; port++) {
+		struct ocelot_port *ocelot_port = ocelot->ports[port];
+
+		if (!ocelot_port || !ocelot_port->bond)
+			continue;
+
+		visited &= ~BIT(port);
+	}
+
+	/* Now, set PGIDs for each active LAG */
 	for (lag = 0; lag < ocelot->num_phys_ports; lag++) {
+		struct net_device *bond = ocelot->ports[lag]->bond;
 		unsigned long bond_mask;
 		int aggr_count = 0;
 		u8 aggr_idx[16];
 
-		bond_mask = ocelot->lags[lag];
-		if (!bond_mask)
+		if (!bond || (visited & BIT(lag)))
 			continue;
 
+		bond_mask = ocelot_get_bond_mask(ocelot, bond);
+
 		for_each_set_bit(port, &bond_mask, ocelot->num_phys_ports) {
 			// Destination mask
 			ocelot_write_rix(ocelot, bond_mask,
@@ -1313,6 +1323,19 @@ static void ocelot_set_aggr_pgids(struct ocelot *ocelot)
 			ac |= BIT(aggr_idx[i % aggr_count]);
 			ocelot_write_rix(ocelot, ac, ANA_PGID_PGID, i);
 		}
+
+		/* Mark all ports in the same LAG as visited to avoid applying
+		 * the same config again.
+		 */
+		for (port = lag; port < ocelot->num_phys_ports; port++) {
+			struct ocelot_port *ocelot_port = ocelot->ports[port];
+
+			if (!ocelot_port)
+				continue;
+
+			if (ocelot_port->bond == bond)
+				visited |= BIT(port);
+		}
 	}
 }
 
@@ -1353,30 +1376,11 @@ int ocelot_port_lag_join(struct ocelot *ocelot, int port,
 			 struct net_device *bond,
 			 struct netdev_lag_upper_info *info)
 {
-	u32 bond_mask = 0;
-	int lag;
-
 	if (info->tx_type != NETDEV_LAG_TX_TYPE_HASH)
 		return -EOPNOTSUPP;
 
 	ocelot->ports[port]->bond = bond;
 
-	bond_mask = ocelot_get_bond_mask(ocelot, bond);
-
-	lag = __ffs(bond_mask);
-
-	/* If the new port is the lowest one, use it as the logical port from
-	 * now on
-	 */
-	if (port == lag) {
-		ocelot->lags[port] = bond_mask;
-		bond_mask &= ~BIT(port);
-		if (bond_mask)
-			ocelot->lags[__ffs(bond_mask)] = 0;
-	} else {
-		ocelot->lags[lag] |= BIT(port);
-	}
-
 	ocelot_setup_logical_port_ids(ocelot);
 	ocelot_apply_bridge_fwd_mask(ocelot);
 	ocelot_set_aggr_pgids(ocelot);
@@ -1388,24 +1392,8 @@ EXPORT_SYMBOL(ocelot_port_lag_join);
 void ocelot_port_lag_leave(struct ocelot *ocelot, int port,
 			   struct net_device *bond)
 {
-	int i;
-
 	ocelot->ports[port]->bond = NULL;
 
-	/* Remove port from any lag */
-	for (i = 0; i < ocelot->num_phys_ports; i++)
-		ocelot->lags[i] &= ~BIT(port);
-
-	/* if it was the logical port of the lag, move the lag config to the
-	 * next port
-	 */
-	if (ocelot->lags[port]) {
-		int n = __ffs(ocelot->lags[port]);
-
-		ocelot->lags[n] = ocelot->lags[port];
-		ocelot->lags[port] = 0;
-	}
-
 	ocelot_setup_logical_port_ids(ocelot);
 	ocelot_apply_bridge_fwd_mask(ocelot);
 	ocelot_set_aggr_pgids(ocelot);
@@ -1587,11 +1575,6 @@ int ocelot_init(struct ocelot *ocelot)
 		}
 	}
 
-	ocelot->lags = devm_kcalloc(ocelot->dev, ocelot->num_phys_ports,
-				    sizeof(u32), GFP_KERNEL);
-	if (!ocelot->lags)
-		return -ENOMEM;
-
 	ocelot->stats = devm_kcalloc(ocelot->dev,
 				     ocelot->num_phys_ports * ocelot->num_stats,
 				     sizeof(u64), GFP_KERNEL);
diff --git a/include/soc/mscc/ocelot.h b/include/soc/mscc/ocelot.h
index e36a1ed29c01..089e552719e0 100644
--- a/include/soc/mscc/ocelot.h
+++ b/include/soc/mscc/ocelot.h
@@ -657,8 +657,6 @@ struct ocelot {
 	enum ocelot_tag_prefix		npi_inj_prefix;
 	enum ocelot_tag_prefix		npi_xtr_prefix;
 
-	u32				*lags;
-
 	struct list_head		multicast;
 	struct list_head		pgids;
 
-- 
2.25.1


  parent reply	other threads:[~2021-02-06  2:57 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-02-05 22:02 [PATCH RESEND v3 net-next 00/12] LAG offload for Ocelot DSA switches Vladimir Oltean
2021-02-05 22:02 ` [PATCH RESEND v3 net-next 01/12] net: mscc: ocelot: rename ocelot_netdevice_port_event to ocelot_netdevice_changeupper Vladimir Oltean
2021-02-05 22:02 ` [PATCH RESEND v3 net-next 02/12] net: mscc: ocelot: use a switch-case statement in ocelot_netdevice_event Vladimir Oltean
2021-02-05 22:02 ` [PATCH RESEND v3 net-next 03/12] net: mscc: ocelot: don't refuse bonding interfaces we can't offload Vladimir Oltean
2021-02-05 22:02 ` [PATCH RESEND v3 net-next 04/12] net: mscc: ocelot: use ipv6 in the aggregation code Vladimir Oltean
2021-02-05 22:02 ` [PATCH RESEND v3 net-next 05/12] net: mscc: ocelot: set up the bonding mask in a way that avoids a net_device Vladimir Oltean
2021-02-05 22:02 ` [PATCH RESEND v3 net-next 06/12] net: mscc: ocelot: avoid unneeded "lp" variable in LAG join Vladimir Oltean
2021-02-05 22:02 ` [PATCH RESEND v3 net-next 07/12] net: mscc: ocelot: set up logical port IDs centrally Vladimir Oltean
2021-02-05 22:02 ` Vladimir Oltean [this message]
2021-02-05 22:02 ` [PATCH RESEND v3 net-next 09/12] net: mscc: ocelot: rename aggr_count to num_ports_in_lag Vladimir Oltean
2021-02-05 22:02 ` [PATCH RESEND v3 net-next 10/12] net: mscc: ocelot: rebalance LAGs on link up/down events Vladimir Oltean
2021-02-05 22:02 ` [PATCH RESEND v3 net-next 11/12] net: dsa: make assisted_learning_on_cpu_port bypass offloaded LAG interfaces Vladimir Oltean
2021-02-05 22:02 ` [PATCH RESEND v3 net-next 12/12] net: dsa: felix: propagate the LAG offload ops towards the ocelot lib Vladimir Oltean
2021-02-06 23:00 ` [PATCH RESEND v3 net-next 00/12] LAG offload for Ocelot DSA switches patchwork-bot+netdevbpf

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=20210205220221.255646-9-olteanv@gmail.com \
    --to=olteanv@gmail.com \
    --cc=UNGLinuxDriver@microchip.com \
    --cc=alexandre.belloni@bootlin.com \
    --cc=andrew@lunn.ch \
    --cc=claudiu.manoil@nxp.com \
    --cc=davem@davemloft.net \
    --cc=f.fainelli@gmail.com \
    --cc=kuba@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).