net: mana: add msix index sharing between EQs

This patch allows to assign and poll more than one EQ on the same
msix index.
It is achieved by introducing a list of attached EQs in each IRQ context.
It also removes the existing msix_index map that tried to ensure that there
is only one EQ at each msix_index.
This patch exports symbols for creating EQs from other MANA kernel modules.
Signed-off-by: Konstantin Taranov <kotaranov@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/ethernet/microsoft/mana/gdma_main.c

@@ -414,8 +414,12 @@ static void mana_gd_process_eq_events(void *arg)
 
 		old_bits = (eq->head / num_eqe - 1) & GDMA_EQE_OWNER_MASK;
 		/* No more entries */
-		if (owner_bits == old_bits)
+		if (owner_bits == old_bits) {
+			/* return here without ringing the doorbell */
+			if (i == 0)
+				return;
 			break;
+		}
 
 		new_bits = (eq->head / num_eqe) & GDMA_EQE_OWNER_MASK;
 		if (owner_bits != new_bits) {
@@ -445,42 +449,29 @@ static int mana_gd_register_irq(struct gdma_queue *queue,
 	struct gdma_dev *gd = queue->gdma_dev;
 	struct gdma_irq_context *gic;
 	struct gdma_context *gc;
-	struct gdma_resource *r;
 	unsigned int msi_index;
 	unsigned long flags;
 	struct device *dev;
 	int err = 0;
 
 	gc = gd->gdma_context;
-	r = &gc->msix_resource;
 	dev = gc->dev;
+	msi_index = spec->eq.msix_index;
 
-	spin_lock_irqsave(&r->lock, flags);
-
-	msi_index = find_first_zero_bit(r->map, r->size);
-	if (msi_index >= r->size || msi_index >= gc->num_msix_usable) {
+	if (msi_index >= gc->num_msix_usable) {
 		err = -ENOSPC;
-	} else {
-		bitmap_set(r->map, msi_index, 1);
-		queue->eq.msix_index = msi_index;
-	}
-
-	spin_unlock_irqrestore(&r->lock, flags);
-
-	if (err) {
-		dev_err(dev, "Register IRQ err:%d, msi:%u rsize:%u, nMSI:%u",
-			err, msi_index, r->size, gc->num_msix_usable);
+		dev_err(dev, "Register IRQ err:%d, msi:%u nMSI:%u",
+			err, msi_index, gc->num_msix_usable);
 
 		return err;
 	}
 
+	queue->eq.msix_index = msi_index;
 	gic = &gc->irq_contexts[msi_index];
 
-	WARN_ON(gic->handler || gic->arg);
-
-	gic->arg = queue;
-
-	gic->handler = mana_gd_process_eq_events;
+	spin_lock_irqsave(&gic->lock, flags);
+	list_add_rcu(&queue->entry, &gic->eq_list);
+	spin_unlock_irqrestore(&gic->lock, flags);
 
 	return 0;
 }
@@ -490,12 +481,11 @@ static void mana_gd_deregiser_irq(struct gdma_queue *queue)
 	struct gdma_dev *gd = queue->gdma_dev;
 	struct gdma_irq_context *gic;
 	struct gdma_context *gc;
-	struct gdma_resource *r;
 	unsigned int msix_index;
 	unsigned long flags;
+	struct gdma_queue *eq;
 
 	gc = gd->gdma_context;
-	r = &gc->msix_resource;
 
 	/* At most num_online_cpus() + 1 interrupts are used. */
 	msix_index = queue->eq.msix_index;
@@ -503,14 +493,17 @@ static void mana_gd_deregiser_irq(struct gdma_queue *queue)
 		return;
 
 	gic = &gc->irq_contexts[msix_index];
-	gic->handler = NULL;
-	gic->arg = NULL;
-
-	spin_lock_irqsave(&r->lock, flags);
-	bitmap_clear(r->map, msix_index, 1);
-	spin_unlock_irqrestore(&r->lock, flags);
+	spin_lock_irqsave(&gic->lock, flags);
+	list_for_each_entry_rcu(eq, &gic->eq_list, entry) {
+		if (queue == eq) {
+			list_del_rcu(&eq->entry);
+			break;
+		}
+	}
+	spin_unlock_irqrestore(&gic->lock, flags);
 
 	queue->eq.msix_index = INVALID_PCI_MSIX_INDEX;
+	synchronize_rcu();
 }
 
 int mana_gd_test_eq(struct gdma_context *gc, struct gdma_queue *eq)
@@ -588,6 +581,7 @@ static int mana_gd_create_eq(struct gdma_dev *gd,
 	int err;
 
 	queue->eq.msix_index = INVALID_PCI_MSIX_INDEX;
+	queue->id = INVALID_QUEUE_ID;
 
 	log2_num_entries = ilog2(queue->queue_size / GDMA_EQE_SIZE);
 
@@ -819,6 +813,7 @@ int mana_gd_create_mana_eq(struct gdma_dev *gd,
 	kfree(queue);
 	return err;
 }
+EXPORT_SYMBOL_NS(mana_gd_create_mana_eq, NET_MANA);
 
 int mana_gd_create_mana_wq_cq(struct gdma_dev *gd,
 			      const struct gdma_queue_spec *spec,
@@ -895,6 +890,7 @@ void mana_gd_destroy_queue(struct gdma_context *gc, struct gdma_queue *queue)
 	mana_gd_free_memory(gmi);
 	kfree(queue);
 }
+EXPORT_SYMBOL_NS(mana_gd_destroy_queue, NET_MANA);
 
 int mana_gd_verify_vf_version(struct pci_dev *pdev)
 {
@@ -1217,9 +1213,14 @@ int mana_gd_poll_cq(struct gdma_queue *cq, struct gdma_comp *comp, int num_cqe)
 static irqreturn_t mana_gd_intr(int irq, void *arg)
 {
 	struct gdma_irq_context *gic = arg;
+	struct list_head *eq_list = &gic->eq_list;
+	struct gdma_queue *eq;
 
-	if (gic->handler)
-		gic->handler(gic->arg);
+	rcu_read_lock();
+	list_for_each_entry_rcu(eq, eq_list, entry) {
+		gic->handler(eq);
+	}
+	rcu_read_unlock();
 
 	return IRQ_HANDLED;
 }
@@ -1271,8 +1272,9 @@ static int mana_gd_setup_irqs(struct pci_dev *pdev)
 
 	for (i = 0; i < nvec; i++) {
 		gic = &gc->irq_contexts[i];
-		gic->handler = NULL;
-		gic->arg = NULL;
+		gic->handler = mana_gd_process_eq_events;
+		INIT_LIST_HEAD(&gic->eq_list);
+		spin_lock_init(&gic->lock);
 
 		if (!i)
 			snprintf(gic->name, MANA_IRQ_NAME_SZ, "mana_hwc@pci:%s",
@@ -1295,10 +1297,6 @@ static int mana_gd_setup_irqs(struct pci_dev *pdev)
 		irq_set_affinity_and_hint(irq, cpumask_of(cpu));
 	}
 
-	err = mana_gd_alloc_res_map(nvec, &gc->msix_resource);
-	if (err)
-		goto free_irq;
-
 	gc->max_num_msix = nvec;
 	gc->num_msix_usable = nvec;
 
@@ -1329,8 +1327,6 @@ static void mana_gd_remove_irqs(struct pci_dev *pdev)
 	if (gc->max_num_msix < 1)
 		return;
 
-	mana_gd_free_res_map(&gc->msix_resource);
-
 	for (i = 0; i < gc->max_num_msix; i++) {
 		irq = pci_irq_vector(pdev, i);
 		if (irq < 0)

drivers/net/ethernet/microsoft/mana/hw_channel.c

@@ -300,6 +300,7 @@ static int mana_hwc_create_gdma_eq(struct hw_channel_context *hwc,
 	spec.eq.context = ctx;
 	spec.eq.callback = cb;
 	spec.eq.log2_throttle_limit = DEFAULT_LOG2_THROTTLING_FOR_ERROR_EQ;
+	spec.eq.msix_index = 0;
 
 	return mana_gd_create_hwc_queue(hwc->gdma_dev, &spec, queue);
 }

drivers/net/ethernet/microsoft/mana/mana_en.c

@@ -1244,6 +1244,7 @@ static int mana_create_eq(struct mana_context *ac)
 	spec.eq.log2_throttle_limit = LOG2_EQ_THROTTLE;
 
 	for (i = 0; i < gc->max_num_queues; i++) {
+		spec.eq.msix_index = (i + 1) % gc->num_msix_usable;
 		err = mana_gd_create_mana_eq(gd, &spec, &ac->eqs[i].eq);
 		if (err)
 			goto out;

include/net/mana/gdma.h

@@ -293,6 +293,7 @@ struct gdma_queue {
 
 	u32 head;
 	u32 tail;
+	struct list_head entry;
 
 	/* Extra fields specific to EQ/CQ. */
 	union {
@@ -328,6 +329,7 @@ struct gdma_queue_spec {
 			void *context;
 
 			unsigned long log2_throttle_limit;
+			unsigned int msix_index;
 		} eq;
 
 		struct {
@@ -344,7 +346,9 @@ struct gdma_queue_spec {
 
 struct gdma_irq_context {
 	void (*handler)(void *arg);
-	void *arg;
+	/* Protect the eq_list */
+	spinlock_t lock;
+	struct list_head eq_list;
 	char name[MANA_IRQ_NAME_SZ];
 };
 
@@ -355,7 +359,6 @@ struct gdma_context {
 	unsigned int		max_num_queues;
 	unsigned int		max_num_msix;
 	unsigned int		num_msix_usable;
-	struct gdma_resource	msix_resource;
 	struct gdma_irq_context	*irq_contexts;
 
 	/* L2 MTU */