sfc: Rework MAC, PHY and board event handling

From: Steve Hodgson <shodgson@solarflare.com>

MAC, PHY and board events may be separately enabled and signalled.
Our current arrangement of chaining the polling functions can result
in events being missed.  Change them to be more independent.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/sfc/efx.c

@@ -612,17 +612,26 @@ void efx_reconfigure_port(struct efx_nic *efx)
 /* Asynchronous efx_reconfigure_port work item. To speed up efx_flush_all()
  * we don't efx_reconfigure_port() if the port is disabled. Care is taken
  * in efx_stop_all() and efx_start_port() to prevent PHY events being lost */
-static void efx_reconfigure_work(struct work_struct *data)
+static void efx_phy_work(struct work_struct *data)
 {
-	struct efx_nic *efx = container_of(data, struct efx_nic,
-					   reconfigure_work);
+	struct efx_nic *efx = container_of(data, struct efx_nic, phy_work);
 
 	mutex_lock(&efx->mac_lock);
 	if (efx->port_enabled)
 		__efx_reconfigure_port(efx);
 	mutex_unlock(&efx->mac_lock);
 }
 
+static void efx_mac_work(struct work_struct *data)
+{
+	struct efx_nic *efx = container_of(data, struct efx_nic, mac_work);
+
+	mutex_lock(&efx->mac_lock);
+	if (efx->port_enabled)
+		efx->mac_op->irq(efx);
+	mutex_unlock(&efx->mac_lock);
+}
+
 static int efx_probe_port(struct efx_nic *efx)
 {
 	int rc;
@@ -688,7 +697,7 @@ static int efx_init_port(struct efx_nic *efx)
 
 /* Allow efx_reconfigure_port() to be scheduled, and close the window
  * between efx_stop_port and efx_flush_all whereby a previously scheduled
- * efx_reconfigure_port() may have been cancelled */
+ * efx_phy_work()/efx_mac_work() may have been cancelled */
 static void efx_start_port(struct efx_nic *efx)
 {
 	EFX_LOG(efx, "start port\n");
@@ -697,13 +706,14 @@ static void efx_start_port(struct efx_nic *efx)
 	mutex_lock(&efx->mac_lock);
 	efx->port_enabled = true;
 	__efx_reconfigure_port(efx);
+	efx->mac_op->irq(efx);
 	mutex_unlock(&efx->mac_lock);
 }
 
-/* Prevent efx_reconfigure_work and efx_monitor() from executing, and
- * efx_set_multicast_list() from scheduling efx_reconfigure_work.
- * efx_reconfigure_work can still be scheduled via NAPI processing
- * until efx_flush_all() is called */
+/* Prevent efx_phy_work, efx_mac_work, and efx_monitor() from executing,
+ * and efx_set_multicast_list() from scheduling efx_phy_work. efx_phy_work
+ * and efx_mac_work may still be scheduled via NAPI processing until
+ * efx_flush_all() is called */
 static void efx_stop_port(struct efx_nic *efx)
 {
 	EFX_LOG(efx, "stop port\n");
@@ -1094,7 +1104,8 @@ static void efx_flush_all(struct efx_nic *efx)
 		cancel_delayed_work_sync(&rx_queue->work);
 
 	/* Stop scheduled port reconfigurations */
-	cancel_work_sync(&efx->reconfigure_work);
+	cancel_work_sync(&efx->mac_work);
+	cancel_work_sync(&efx->phy_work);
 
 }
 
@@ -1131,7 +1142,7 @@ static void efx_stop_all(struct efx_nic *efx)
 	 * window to loose phy events */
 	efx_stop_port(efx);
 
-	/* Flush reconfigure_work, refill_workqueue, monitor_work */
+	/* Flush efx_phy_work, efx_mac_work, refill_workqueue, monitor_work */
 	efx_flush_all(efx);
 
 	/* Isolate the MAC from the TX and RX engines, so that queue
@@ -1203,24 +1214,31 @@ static void efx_monitor(struct work_struct *data)
 {
 	struct efx_nic *efx = container_of(data, struct efx_nic,
 					   monitor_work.work);
+	int rc;
 
 	EFX_TRACE(efx, "hardware monitor executing on CPU %d\n",
 		  raw_smp_processor_id());
 
-
 	/* If the mac_lock is already held then it is likely a port
 	 * reconfiguration is already in place, which will likely do
 	 * most of the work of check_hw() anyway. */
-	if (!mutex_trylock(&efx->mac_lock)) {
-		queue_delayed_work(efx->workqueue, &efx->monitor_work,
-				   efx_monitor_interval);
-		return;
+	if (!mutex_trylock(&efx->mac_lock))
+		goto out_requeue;
+	if (!efx->port_enabled)
+		goto out_unlock;
+	rc = efx->board_info.monitor(efx);
+	if (rc) {
+		EFX_ERR(efx, "Board sensor %s; shutting down PHY\n",
+			(rc == -ERANGE) ? "reported fault" : "failed");
+		efx->phy_mode |= PHY_MODE_LOW_POWER;
+		falcon_sim_phy_event(efx);
 	}
+	efx->phy_op->poll(efx);
+	efx->mac_op->poll(efx);
 
-	if (efx->port_enabled)
-		efx->mac_op->check_hw(efx);
+out_unlock:
 	mutex_unlock(&efx->mac_lock);
-
+out_requeue:
 	queue_delayed_work(efx->workqueue, &efx->monitor_work,
 			   efx_monitor_interval);
 }
@@ -1477,7 +1495,7 @@ static void efx_set_multicast_list(struct net_device *net_dev)
 		return;
 
 	if (changed)
-		queue_work(efx->workqueue, &efx->reconfigure_work);
+		queue_work(efx->workqueue, &efx->phy_work);
 
 	/* Create and activate new global multicast hash table */
 	falcon_set_multicast_hash(efx);
@@ -1800,12 +1818,14 @@ void efx_port_dummy_op_blink(struct efx_nic *efx, bool blink) {}
 
 static struct efx_mac_operations efx_dummy_mac_operations = {
 	.reconfigure	= efx_port_dummy_op_void,
+	.poll		= efx_port_dummy_op_void,
+	.irq		= efx_port_dummy_op_void,
 };
 
 static struct efx_phy_operations efx_dummy_phy_operations = {
 	.init		 = efx_port_dummy_op_int,
 	.reconfigure	 = efx_port_dummy_op_void,
-	.check_hw        = efx_port_dummy_op_int,
+	.poll		 = efx_port_dummy_op_void,
 	.fini		 = efx_port_dummy_op_void,
 	.clear_interrupt = efx_port_dummy_op_void,
 };
@@ -1857,7 +1877,8 @@ static int efx_init_struct(struct efx_nic *efx, struct efx_nic_type *type,
 	efx->mac_op = &efx_dummy_mac_operations;
 	efx->phy_op = &efx_dummy_phy_operations;
 	efx->mii.dev = net_dev;
-	INIT_WORK(&efx->reconfigure_work, efx_reconfigure_work);
+	INIT_WORK(&efx->phy_work, efx_phy_work);
+	INIT_WORK(&efx->mac_work, efx_mac_work);
 	atomic_set(&efx->netif_stop_count, 1);
 
 	for (i = 0; i < EFX_MAX_CHANNELS; i++) {

drivers/net/sfc/falcon.c

@@ -910,22 +910,20 @@ static void falcon_handle_global_event(struct efx_channel *channel,
 				       efx_qword_t *event)
 {
 	struct efx_nic *efx = channel->efx;
-	bool is_phy_event = false, handled = false;
+	bool handled = false;
 
-	/* Check for interrupt on either port.  Some boards have a
-	 * single PHY wired to the interrupt line for port 1. */
 	if (EFX_QWORD_FIELD(*event, G_PHY0_INTR) ||
 	    EFX_QWORD_FIELD(*event, G_PHY1_INTR) ||
-	    EFX_QWORD_FIELD(*event, XG_PHY_INTR))
-		is_phy_event = true;
+	    EFX_QWORD_FIELD(*event, XG_PHY_INTR) ||
+	    EFX_QWORD_FIELD(*event, XFP_PHY_INTR)) {
+		efx->phy_op->clear_interrupt(efx);
+		queue_work(efx->workqueue, &efx->phy_work);
+		handled = true;
+	}
 
 	if ((falcon_rev(efx) >= FALCON_REV_B0) &&
-	    EFX_QWORD_FIELD(*event, XG_MNT_INTR_B0))
-		is_phy_event = true;
-
-	if (is_phy_event) {
-		efx->phy_op->clear_interrupt(efx);
-		queue_work(efx->workqueue, &efx->reconfigure_work);
+	    EFX_QWORD_FIELD(*event, XG_MNT_INTR_B0)) {
+		queue_work(efx->workqueue, &efx->mac_work);
 		handled = true;
 	}
 

drivers/net/sfc/falcon_gmac.c

@@ -221,13 +221,9 @@ static void falcon_update_stats_gmac(struct efx_nic *efx)
 	mac_stats->rx_lt64 = mac_stats->rx_good_lt64 + mac_stats->rx_bad_lt64;
 }
 
-static int falcon_check_gmac(struct efx_nic *efx)
-{
-	return efx->phy_op->check_hw(efx);
-}
-
 struct efx_mac_operations falcon_gmac_operations = {
 	.reconfigure	= falcon_reconfigure_gmac,
 	.update_stats	= falcon_update_stats_gmac,
-	.check_hw	= falcon_check_gmac,
+	.irq		= efx_port_dummy_op_void,
+	.poll		= efx_port_dummy_op_void,
 };

drivers/net/sfc/falcon_hwdefs.h

@@ -1051,6 +1051,8 @@
 #define XG_MNT_INTR_B0_WIDTH 1
 #define RX_RECOVERY_A1_LBN 11
 #define RX_RECOVERY_A1_WIDTH 1
+#define XFP_PHY_INTR_LBN 10
+#define XFP_PHY_INTR_WIDTH 1
 #define XG_PHY_INTR_LBN 9
 #define XG_PHY_INTR_WIDTH 1
 #define G_PHY1_INTR_LBN 8

drivers/net/sfc/falcon_xmac.c

@@ -342,33 +342,35 @@ static void falcon_update_stats_xmac(struct efx_nic *efx)
 		 mac_stats->rx_control * 64);
 }
 
-static int falcon_check_xmac(struct efx_nic *efx)
+static void falcon_xmac_irq(struct efx_nic *efx)
 {
-	bool xaui_link_ok;
-	int rc;
+	/* The XGMII link has a transient fault, which indicates either:
+	 *   - there's a transient xgmii fault
+	 *   - falcon's end of the xaui link may need a kick
+	 *   - the wire-side link may have gone down, but the lasi/poll()
+	 *     hasn't noticed yet.
+	 *
+	 * We only want to even bother polling XAUI if we're confident it's
+	 * not (1) or (3). In both cases, the only reliable way to spot this
+	 * is to wait a bit. We do this here by forcing the mac link state
+	 * to down, and waiting for the mac poll to come round and check
+	 */
+	efx->mac_up = false;
+}
 
-	if ((efx->loopback_mode == LOOPBACK_NETWORK) ||
-	    efx_phy_mode_disabled(efx->phy_mode))
-		return 0;
+static void falcon_poll_xmac(struct efx_nic *efx)
+{
+	if (!EFX_WORKAROUND_5147(efx) || !efx->link_up || efx->mac_up)
+		return;
 
 	falcon_mask_status_intr(efx, false);
-	xaui_link_ok = falcon_xaui_link_ok(efx);
-
-	if (EFX_WORKAROUND_5147(efx) && !xaui_link_ok)
-		falcon_reset_xaui(efx);
-
-	/* Call the PHY check_hw routine */
-	rc = efx->phy_op->check_hw(efx);
-
-	/* Unmask interrupt if everything was (and still is) ok */
-	if (xaui_link_ok && efx->link_up)
-		falcon_mask_status_intr(efx, true);
-
-	return rc;
+	falcon_check_xaui_link_up(efx, 1);
+	falcon_mask_status_intr(efx, true);
 }
 
 struct efx_mac_operations falcon_xmac_operations = {
 	.reconfigure	= falcon_reconfigure_xmac,
 	.update_stats	= falcon_update_stats_xmac,
-	.check_hw	= falcon_check_xmac,
+	.irq		= falcon_xmac_irq,
+	.poll		= falcon_poll_xmac,
 };

drivers/net/sfc/net_driver.h

@@ -544,12 +544,14 @@ static inline enum efx_fc_type efx_fc_resolve(enum efx_fc_type wanted_fc,
  * struct efx_mac_operations - Efx MAC operations table
  * @reconfigure: Reconfigure MAC. Serialised by the mac_lock
  * @update_stats: Update statistics
- * @check_hw: Check hardware. Serialised by the mac_lock
+ * @irq: Hardware MAC event callback. Serialised by the mac_lock
+ * @poll: Poll for hardware state. Serialised by the mac_lock
  */
 struct efx_mac_operations {
 	void (*reconfigure) (struct efx_nic *efx);
 	void (*update_stats) (struct efx_nic *efx);
-	int (*check_hw) (struct efx_nic *efx);
+	void (*irq) (struct efx_nic *efx);
+	void (*poll) (struct efx_nic *efx);
 };
 
 /**
@@ -559,7 +561,7 @@ struct efx_mac_operations {
  * @reconfigure: Reconfigure PHY (e.g. for new link parameters)
  * @clear_interrupt: Clear down interrupt
  * @blink: Blink LEDs
- * @check_hw: Check hardware
+ * @poll: Poll for hardware state. Serialised by the mac_lock.
  * @get_settings: Get ethtool settings. Serialised by the mac_lock.
  * @set_settings: Set ethtool settings. Serialised by the mac_lock.
  * @set_xnp_advertise: Set abilities advertised in Extended Next Page
@@ -573,7 +575,7 @@ struct efx_phy_operations {
 	void (*fini) (struct efx_nic *efx);
 	void (*reconfigure) (struct efx_nic *efx);
 	void (*clear_interrupt) (struct efx_nic *efx);
-	int (*check_hw) (struct efx_nic *efx);
+	void (*poll) (struct efx_nic *efx);
 	int (*test) (struct efx_nic *efx);
 	void (*get_settings) (struct efx_nic *efx,
 			      struct ethtool_cmd *ecmd);
@@ -728,10 +730,10 @@ union efx_multicast_hash {
  * @mac_lock: MAC access lock. Protects @port_enabled, @phy_mode,
  *	@port_inhibited, efx_monitor() and efx_reconfigure_port()
  * @port_enabled: Port enabled indicator.
- *	Serialises efx_stop_all(), efx_start_all() and efx_monitor() and
- *	efx_reconfigure_work with kernel interfaces. Safe to read under any
- *	one of the rtnl_lock, mac_lock, or netif_tx_lock, but all three must
- *	be held to modify it.
+ *	Serialises efx_stop_all(), efx_start_all(), efx_monitor(),
+ *	efx_phy_work(), and efx_mac_work() with kernel interfaces. Safe to read
+ *	under any one of the rtnl_lock, mac_lock, or netif_tx_lock, but all
+ *	three must be held to modify it.
  * @port_inhibited: If set, the netif_carrier is always off. Hold the mac_lock
  * @port_initialized: Port initialized?
  * @net_dev: Operating system network device. Consider holding the rtnl lock
@@ -762,7 +764,8 @@ union efx_multicast_hash {
  * @promiscuous: Promiscuous flag. Protected by netif_tx_lock.
  * @multicast_hash: Multicast hash table
  * @wanted_fc: Wanted flow control flags
- * @reconfigure_work: work item for dealing with PHY events
+ * @phy_work: work item for dealing with PHY events
+ * @mac_work: work item for dealing with MAC events
  * @loopback_mode: Loopback status
  * @loopback_modes: Supported loopback mode bitmask
  * @loopback_selftest: Offline self-test private state
@@ -810,6 +813,7 @@ struct efx_nic {
 	struct falcon_nic_data *nic_data;
 
 	struct mutex mac_lock;
+	struct work_struct mac_work;
 	bool port_enabled;
 	bool port_inhibited;
 
@@ -830,6 +834,7 @@ struct efx_nic {
 
 	enum phy_type phy_type;
 	spinlock_t phy_lock;
+	struct work_struct phy_work;
 	struct efx_phy_operations *phy_op;
 	void *phy_data;
 	struct mii_if_info mii;
@@ -845,7 +850,6 @@ struct efx_nic {
 	bool promiscuous;
 	union efx_multicast_hash multicast_hash;
 	enum efx_fc_type wanted_fc;
-	struct work_struct reconfigure_work;
 
 	atomic_t rx_reset;
 	enum efx_loopback_mode loopback_mode;

drivers/net/sfc/selftest.c

@@ -594,12 +594,14 @@ static int efx_test_loopbacks(struct efx_nic *efx, struct ethtool_cmd ecmd,
 		efx->loopback_mode = mode;
 		efx_reconfigure_port(efx);
 
-		/* Wait for the PHY to signal the link is up */
+		/* Wait for the PHY to signal the link is up. Interrupts
+		 * are enabled for PHY's using LASI, otherwise we poll()
+		 * quickly */
 		count = 0;
 		do {
 			struct efx_channel *channel = &efx->channel[0];
 
-			efx->mac_op->check_hw(efx);
+			efx->phy_op->poll(efx);
 			schedule_timeout_uninterruptible(HZ / 10);
 			if (channel->work_pending)
 				efx_process_channel_now(channel);