From a3cdcbfa8fb1fccfe48d359da86e99546610c562 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Fri, 10 Oct 2008 12:01:37 -0700
Subject: mlx4_core: Add QP range reservation support

To allow allocating an aligned range of consecutive QP numbers, add an
interface to reserve an aligned range of QP numbers and have the QP
allocation function always take a QP number.

This will be used for RSS support in the mlx4_en Ethernet driver and
also potentially by IPoIB RSS support.

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/mlx4/qp.c | 21 ++++++++++--
 drivers/net/mlx4/alloc.c        | 74 ++++++++++++++++++++++++++++++++++++++++-
 drivers/net/mlx4/mlx4.h         |  2 ++
 drivers/net/mlx4/qp.c           | 45 +++++++++++++++++--------
 4 files changed, 125 insertions(+), 17 deletions(-)

(limited to 'drivers')

diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index baa01deb243..39167a797f9 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -451,6 +451,7 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 			    struct ib_qp_init_attr *init_attr,
 			    struct ib_udata *udata, int sqpn, struct mlx4_ib_qp *qp)
 {
+	int qpn;
 	int err;
 
 	mutex_init(&qp->mutex);
@@ -545,9 +546,17 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 		}
 	}
 
-	err = mlx4_qp_alloc(dev->dev, sqpn, &qp->mqp);
+	if (sqpn) {
+		qpn = sqpn;
+	} else {
+		err = mlx4_qp_reserve_range(dev->dev, 1, 1, &qpn);
+		if (err)
+			goto err_wrid;
+	}
+
+	err = mlx4_qp_alloc(dev->dev, qpn, &qp->mqp);
 	if (err)
-		goto err_wrid;
+		goto err_qpn;
 
 	/*
 	 * Hardware wants QPN written in big-endian order (after
@@ -560,6 +569,10 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 
 	return 0;
 
+err_qpn:
+	if (!sqpn)
+		mlx4_qp_release_range(dev->dev, qpn, 1);
+
 err_wrid:
 	if (pd->uobject) {
 		if (!init_attr->srq)
@@ -655,6 +668,10 @@ static void destroy_qp_common(struct mlx4_ib_dev *dev, struct mlx4_ib_qp *qp,
 	mlx4_ib_unlock_cqs(send_cq, recv_cq);
 
 	mlx4_qp_free(dev->dev, &qp->mqp);
+
+	if (!is_sqp(dev, qp))
+		mlx4_qp_release_range(dev->dev, qp->mqp.qpn, 1);
+
 	mlx4_mtt_cleanup(dev->dev, &qp->mtt);
 
 	if (is_user) {
diff --git a/drivers/net/mlx4/alloc.c b/drivers/net/mlx4/alloc.c
index 096bca54bcf..e6c0d5bb5dc 100644
--- a/drivers/net/mlx4/alloc.c
+++ b/drivers/net/mlx4/alloc.c
@@ -65,10 +65,82 @@ u32 mlx4_bitmap_alloc(struct mlx4_bitmap *bitmap)
 
 void mlx4_bitmap_free(struct mlx4_bitmap *bitmap, u32 obj)
 {
+	mlx4_bitmap_free_range(bitmap, obj, 1);
+}
+
+static unsigned long find_aligned_range(unsigned long *bitmap,
+					u32 start, u32 nbits,
+					int len, int align)
+{
+	unsigned long end, i;
+
+again:
+	start = ALIGN(start, align);
+
+	while ((start < nbits) && test_bit(start, bitmap))
+		start += align;
+
+	if (start >= nbits)
+		return -1;
+
+	end = start+len;
+	if (end > nbits)
+		return -1;
+
+	for (i = start + 1; i < end; i++) {
+		if (test_bit(i, bitmap)) {
+			start = i + 1;
+			goto again;
+		}
+	}
+
+	return start;
+}
+
+u32 mlx4_bitmap_alloc_range(struct mlx4_bitmap *bitmap, int cnt, int align)
+{
+	u32 obj, i;
+
+	if (likely(cnt == 1 && align == 1))
+		return mlx4_bitmap_alloc(bitmap);
+
+	spin_lock(&bitmap->lock);
+
+	obj = find_aligned_range(bitmap->table, bitmap->last,
+				 bitmap->max, cnt, align);
+	if (obj >= bitmap->max) {
+		bitmap->top = (bitmap->top + bitmap->max) & bitmap->mask;
+		obj = find_aligned_range(bitmap->table, 0,
+					 bitmap->max,
+					 cnt, align);
+	}
+
+	if (obj < bitmap->max) {
+		for (i = 0; i < cnt; i++)
+			set_bit(obj + i, bitmap->table);
+		if (obj == bitmap->last) {
+			bitmap->last = (obj + cnt);
+			if (bitmap->last >= bitmap->max)
+				bitmap->last = 0;
+		}
+		obj |= bitmap->top;
+	} else
+		obj = -1;
+
+	spin_unlock(&bitmap->lock);
+
+	return obj;
+}
+
+void mlx4_bitmap_free_range(struct mlx4_bitmap *bitmap, u32 obj, int cnt)
+{
+	u32 i;
+
 	obj &= bitmap->max - 1;
 
 	spin_lock(&bitmap->lock);
-	clear_bit(obj, bitmap->table);
+	for (i = 0; i < cnt; i++)
+		clear_bit(obj + i, bitmap->table);
 	bitmap->last = min(bitmap->last, obj);
 	bitmap->top = (bitmap->top + bitmap->max) & bitmap->mask;
 	spin_unlock(&bitmap->lock);
diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h
index 5337e3ac3e7..b55ddab73f6 100644
--- a/drivers/net/mlx4/mlx4.h
+++ b/drivers/net/mlx4/mlx4.h
@@ -288,6 +288,8 @@ static inline struct mlx4_priv *mlx4_priv(struct mlx4_dev *dev)
 
 u32 mlx4_bitmap_alloc(struct mlx4_bitmap *bitmap);
 void mlx4_bitmap_free(struct mlx4_bitmap *bitmap, u32 obj);
+u32 mlx4_bitmap_alloc_range(struct mlx4_bitmap *bitmap, int cnt, int align);
+void mlx4_bitmap_free_range(struct mlx4_bitmap *bitmap, u32 obj, int cnt);
 int mlx4_bitmap_init(struct mlx4_bitmap *bitmap, u32 num, u32 mask, u32 reserved);
 void mlx4_bitmap_cleanup(struct mlx4_bitmap *bitmap);
 
diff --git a/drivers/net/mlx4/qp.c b/drivers/net/mlx4/qp.c
index c49a86044bf..98e0c40ba36 100644
--- a/drivers/net/mlx4/qp.c
+++ b/drivers/net/mlx4/qp.c
@@ -147,19 +147,42 @@ int mlx4_qp_modify(struct mlx4_dev *dev, struct mlx4_mtt *mtt,
 }
 EXPORT_SYMBOL_GPL(mlx4_qp_modify);
 
-int mlx4_qp_alloc(struct mlx4_dev *dev, int sqpn, struct mlx4_qp *qp)
+int mlx4_qp_reserve_range(struct mlx4_dev *dev, int cnt, int align, int *base)
+{
+	struct mlx4_priv *priv = mlx4_priv(dev);
+	struct mlx4_qp_table *qp_table = &priv->qp_table;
+	int qpn;
+
+	qpn = mlx4_bitmap_alloc_range(&qp_table->bitmap, cnt, align);
+	if (qpn == -1)
+		return -ENOMEM;
+
+	*base = qpn;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(mlx4_qp_reserve_range);
+
+void mlx4_qp_release_range(struct mlx4_dev *dev, int base_qpn, int cnt)
+{
+	struct mlx4_priv *priv = mlx4_priv(dev);
+	struct mlx4_qp_table *qp_table = &priv->qp_table;
+	if (base_qpn < dev->caps.sqp_start + 8)
+		return;
+
+	mlx4_bitmap_free_range(&qp_table->bitmap, base_qpn, cnt);
+}
+EXPORT_SYMBOL_GPL(mlx4_qp_release_range);
+
+int mlx4_qp_alloc(struct mlx4_dev *dev, int qpn, struct mlx4_qp *qp)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct mlx4_qp_table *qp_table = &priv->qp_table;
 	int err;
 
-	if (sqpn)
-		qp->qpn = sqpn;
-	else {
-		qp->qpn = mlx4_bitmap_alloc(&qp_table->bitmap);
-		if (qp->qpn == -1)
-			return -ENOMEM;
-	}
+	if (!qpn)
+		return -EINVAL;
+
+	qp->qpn = qpn;
 
 	err = mlx4_table_get(dev, &qp_table->qp_table, qp->qpn);
 	if (err)
@@ -208,9 +231,6 @@ err_put_qp:
 	mlx4_table_put(dev, &qp_table->qp_table, qp->qpn);
 
 err_out:
-	if (!sqpn)
-		mlx4_bitmap_free(&qp_table->bitmap, qp->qpn);
-
 	return err;
 }
 EXPORT_SYMBOL_GPL(mlx4_qp_alloc);
@@ -239,9 +259,6 @@ void mlx4_qp_free(struct mlx4_dev *dev, struct mlx4_qp *qp)
 	mlx4_table_put(dev, &qp_table->altc_table, qp->qpn);
 	mlx4_table_put(dev, &qp_table->auxc_table, qp->qpn);
 	mlx4_table_put(dev, &qp_table->qp_table, qp->qpn);
-
-	if (qp->qpn >= dev->caps.sqp_start + 8)
-		mlx4_bitmap_free(&qp_table->bitmap, qp->qpn);
 }
 EXPORT_SYMBOL_GPL(mlx4_qp_free);
 
-- 
cgit v1.2.3-18-g5258


From 0540bbbe455e123a1692d26205ad1a29983883b0 Mon Sep 17 00:00:00 2001
From: Hoang-Nam Nguyen <hnguyen@linux.vnet.ibm.com>
Date: Fri, 10 Oct 2008 14:40:39 -0700
Subject: IB/ehca: Don't allow creating UC QP with SRQ

This patch prevents a UC QP to be created attached to an SRQ, since
current firmware does not support this feature.

Signed-off-by: Michael Faath <micfaath@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/ehca/ehca_qp.c | 6 ++++++
 1 file changed, 6 insertions(+)

(limited to 'drivers')

diff --git a/drivers/infiniband/hw/ehca/ehca_qp.c b/drivers/infiniband/hw/ehca/ehca_qp.c
index 4dbe2870e01..40b578d601c 100644
--- a/drivers/infiniband/hw/ehca/ehca_qp.c
+++ b/drivers/infiniband/hw/ehca/ehca_qp.c
@@ -502,6 +502,12 @@ static struct ehca_qp *internal_create_qp(
 	if (init_attr->srq) {
 		my_srq = container_of(init_attr->srq, struct ehca_qp, ib_srq);
 
+		if (qp_type == IB_QPT_UC) {
+			ehca_err(pd->device, "UC with SRQ not supported");
+			atomic_dec(&shca->num_qps);
+			return ERR_PTR(-EINVAL);
+		}
+
 		has_srq = 1;
 		parms.ext_type = EQPT_SRQBASE;
 		parms.srq_qpn = my_srq->real_qp_num;
-- 
cgit v1.2.3-18-g5258


From 2767840a5ca73fde62b25e0209aa9269ec4fa7c7 Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Fri, 10 Oct 2008 15:58:52 -0700
Subject: IPoIB: Always initialize poll_timer to avoid crash on unload

ipoib_ib_dev_stop() does del_timer_sync(&priv->poll_timer), but if a
P_key for an interface is not found, poll_timer is not initialized, so
this leads to a crash or hang.  Fix this by moving where poll_timer is
initialized to ipoib_ib_dev_init(), which is always called.

This fixes <https://bugs.openfabrics.org/show_bug.cgi?id=1172>.

Debugged-by: Yosef Etigin <yosefe@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/ulp/ipoib/ipoib_ib.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

(limited to 'drivers')

diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
index 0e748aeeae9..28eb6f03c58 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
@@ -685,10 +685,6 @@ int ipoib_ib_dev_open(struct net_device *dev)
 	queue_delayed_work(ipoib_workqueue, &priv->ah_reap_task,
 			   round_jiffies_relative(HZ));
 
-	init_timer(&priv->poll_timer);
-	priv->poll_timer.function = ipoib_ib_tx_timer_func;
-	priv->poll_timer.data = (unsigned long)dev;
-
 	set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags);
 
 	return 0;
@@ -906,6 +902,9 @@ int ipoib_ib_dev_init(struct net_device *dev, struct ib_device *ca, int port)
 		return -ENODEV;
 	}
 
+	setup_timer(&priv->poll_timer, ipoib_ib_tx_timer_func,
+		    (unsigned long) dev);
+
 	if (dev->flags & IFF_UP) {
 		if (ipoib_ib_dev_open(dev)) {
 			ipoib_transport_dev_cleanup(dev);
-- 
cgit v1.2.3-18-g5258


From 528051746b24dd214883db11bcbb0e667f60447d Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Tue, 14 Oct 2008 14:05:36 -0700
Subject: IB/mad: Use krealloc() to resize snoop table

Use krealloc() instead of kmalloc() followed by memcpy() when resizing
the MAD module's snoop table.

Also put parentheses around the new table size to avoid calculating
the wrong size to allocate, which fixes a bug pointed out by Haven
Hash <haven.hash@isilon.com>.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/mad.c | 14 +++++---------
 1 file changed, 5 insertions(+), 9 deletions(-)

(limited to 'drivers')

diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c
index 49c45feccd5..5c54fc2350b 100644
--- a/drivers/infiniband/core/mad.c
+++ b/drivers/infiniband/core/mad.c
@@ -406,19 +406,15 @@ static int register_snoop_agent(struct ib_mad_qp_info *qp_info,
 
 	if (i == qp_info->snoop_table_size) {
 		/* Grow table. */
-		new_snoop_table = kmalloc(sizeof mad_snoop_priv *
-					  qp_info->snoop_table_size + 1,
-					  GFP_ATOMIC);
+		new_snoop_table = krealloc(qp_info->snoop_table,
+					   sizeof mad_snoop_priv *
+					   (qp_info->snoop_table_size + 1),
+					   GFP_ATOMIC);
 		if (!new_snoop_table) {
 			i = -ENOMEM;
 			goto out;
 		}
-		if (qp_info->snoop_table) {
-			memcpy(new_snoop_table, qp_info->snoop_table,
-			       sizeof mad_snoop_priv *
-			       qp_info->snoop_table_size);
-			kfree(qp_info->snoop_table);
-		}
+
 		qp_info->snoop_table = new_snoop_table;
 		qp_info->snoop_table_size++;
 	}
-- 
cgit v1.2.3-18-g5258


From dc35fac9e936c6cc6ad825fc7e4455468d10adc6 Mon Sep 17 00:00:00 2001
From: Steve Wise <swise@opengridcomputing.com>
Date: Wed, 15 Oct 2008 10:50:34 -0700
Subject: RDMA/cxgb3: Remove cmid reference on tid allocation failures

The error path in iwch_connect() can fail to drop the cmid reference,
which will cause the process to hang when destroying the cmid.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/cxgb3/iwch_cm.c | 1 +
 1 file changed, 1 insertion(+)

(limited to 'drivers')

diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.c b/drivers/infiniband/hw/cxgb3/iwch_cm.c
index c325c44807e..44e936e48a3 100644
--- a/drivers/infiniband/hw/cxgb3/iwch_cm.c
+++ b/drivers/infiniband/hw/cxgb3/iwch_cm.c
@@ -1942,6 +1942,7 @@ fail4:
 fail3:
 	cxgb3_free_atid(ep->com.tdev, ep->atid);
 fail2:
+	cm_id->rem_ref(cm_id);
 	put_ep(&ep->com);
 out:
 	return err;
-- 
cgit v1.2.3-18-g5258


From 93fc9e1bb6507dde945c2eab68c93e1066ac3691 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Wed, 22 Oct 2008 10:25:29 -0700
Subject: mlx4_core: Support multiple pre-reserved QP regions

For ethernet support, we need to reserve QPs for the ethernet and
fibre channel driver.  The QPs are reserved at the end of the QP
table.  (This way we assure that they are aligned to their size)

We need to consider these reserved ranges in bitmap creation, so we
extend the mlx4 bitmap utility functions to allow reserved ranges at
both the bottom and the top of the range.

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/net/mlx4/alloc.c | 29 +++++++++++++---------
 drivers/net/mlx4/cq.c    |  2 +-
 drivers/net/mlx4/eq.c    |  2 +-
 drivers/net/mlx4/fw.c    |  5 ++++
 drivers/net/mlx4/fw.h    |  2 ++
 drivers/net/mlx4/main.c  | 62 +++++++++++++++++++++++++++++++++++++++++++-----
 drivers/net/mlx4/mcg.c   |  4 ++--
 drivers/net/mlx4/mlx4.h  |  4 +++-
 drivers/net/mlx4/mr.c    |  2 +-
 drivers/net/mlx4/pd.c    |  4 ++--
 drivers/net/mlx4/qp.c    | 36 ++++++++++++++++++++++++++--
 drivers/net/mlx4/srq.c   |  2 +-
 12 files changed, 126 insertions(+), 28 deletions(-)

(limited to 'drivers')

diff --git a/drivers/net/mlx4/alloc.c b/drivers/net/mlx4/alloc.c
index e6c0d5bb5dc..e2bc7ecf162 100644
--- a/drivers/net/mlx4/alloc.c
+++ b/drivers/net/mlx4/alloc.c
@@ -47,13 +47,16 @@ u32 mlx4_bitmap_alloc(struct mlx4_bitmap *bitmap)
 
 	obj = find_next_zero_bit(bitmap->table, bitmap->max, bitmap->last);
 	if (obj >= bitmap->max) {
-		bitmap->top = (bitmap->top + bitmap->max) & bitmap->mask;
+		bitmap->top = (bitmap->top + bitmap->max + bitmap->reserved_top)
+				& bitmap->mask;
 		obj = find_first_zero_bit(bitmap->table, bitmap->max);
 	}
 
 	if (obj < bitmap->max) {
 		set_bit(obj, bitmap->table);
-		bitmap->last = (obj + 1) & (bitmap->max - 1);
+		bitmap->last = (obj + 1);
+		if (bitmap->last == bitmap->max)
+			bitmap->last = 0;
 		obj |= bitmap->top;
 	} else
 		obj = -1;
@@ -109,9 +112,9 @@ u32 mlx4_bitmap_alloc_range(struct mlx4_bitmap *bitmap, int cnt, int align)
 	obj = find_aligned_range(bitmap->table, bitmap->last,
 				 bitmap->max, cnt, align);
 	if (obj >= bitmap->max) {
-		bitmap->top = (bitmap->top + bitmap->max) & bitmap->mask;
-		obj = find_aligned_range(bitmap->table, 0,
-					 bitmap->max,
+		bitmap->top = (bitmap->top + bitmap->max + bitmap->reserved_top)
+				& bitmap->mask;
+		obj = find_aligned_range(bitmap->table, 0, bitmap->max,
 					 cnt, align);
 	}
 
@@ -136,17 +139,19 @@ void mlx4_bitmap_free_range(struct mlx4_bitmap *bitmap, u32 obj, int cnt)
 {
 	u32 i;
 
-	obj &= bitmap->max - 1;
+	obj &= bitmap->max + bitmap->reserved_top - 1;
 
 	spin_lock(&bitmap->lock);
 	for (i = 0; i < cnt; i++)
 		clear_bit(obj + i, bitmap->table);
 	bitmap->last = min(bitmap->last, obj);
-	bitmap->top = (bitmap->top + bitmap->max) & bitmap->mask;
+	bitmap->top = (bitmap->top + bitmap->max + bitmap->reserved_top)
+			& bitmap->mask;
 	spin_unlock(&bitmap->lock);
 }
 
-int mlx4_bitmap_init(struct mlx4_bitmap *bitmap, u32 num, u32 mask, u32 reserved)
+int mlx4_bitmap_init(struct mlx4_bitmap *bitmap, u32 num, u32 mask,
+		     u32 reserved_bot, u32 reserved_top)
 {
 	int i;
 
@@ -156,14 +161,16 @@ int mlx4_bitmap_init(struct mlx4_bitmap *bitmap, u32 num, u32 mask, u32 reserved
 
 	bitmap->last = 0;
 	bitmap->top  = 0;
-	bitmap->max  = num;
+	bitmap->max  = num - reserved_top;
 	bitmap->mask = mask;
+	bitmap->reserved_top = reserved_top;
 	spin_lock_init(&bitmap->lock);
-	bitmap->table = kzalloc(BITS_TO_LONGS(num) * sizeof (long), GFP_KERNEL);
+	bitmap->table = kzalloc(BITS_TO_LONGS(bitmap->max) *
+				sizeof (long), GFP_KERNEL);
 	if (!bitmap->table)
 		return -ENOMEM;
 
-	for (i = 0; i < reserved; ++i)
+	for (i = 0; i < reserved_bot; ++i)
 		set_bit(i, bitmap->table);
 
 	return 0;
diff --git a/drivers/net/mlx4/cq.c b/drivers/net/mlx4/cq.c
index 9bb50e3f897..b7ad2829d67 100644
--- a/drivers/net/mlx4/cq.c
+++ b/drivers/net/mlx4/cq.c
@@ -300,7 +300,7 @@ int mlx4_init_cq_table(struct mlx4_dev *dev)
 	INIT_RADIX_TREE(&cq_table->tree, GFP_ATOMIC);
 
 	err = mlx4_bitmap_init(&cq_table->bitmap, dev->caps.num_cqs,
-			       dev->caps.num_cqs - 1, dev->caps.reserved_cqs);
+			       dev->caps.num_cqs - 1, dev->caps.reserved_cqs, 0);
 	if (err)
 		return err;
 
diff --git a/drivers/net/mlx4/eq.c b/drivers/net/mlx4/eq.c
index 8a8b56135a5..de169338cd9 100644
--- a/drivers/net/mlx4/eq.c
+++ b/drivers/net/mlx4/eq.c
@@ -558,7 +558,7 @@ int mlx4_init_eq_table(struct mlx4_dev *dev)
 	int i;
 
 	err = mlx4_bitmap_init(&priv->eq_table.bitmap, dev->caps.num_eqs,
-			       dev->caps.num_eqs - 1, dev->caps.reserved_eqs);
+			       dev->caps.num_eqs - 1, dev->caps.reserved_eqs, 0);
 	if (err)
 		return err;
 
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index 7e32955da98..40d8142c23b 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -357,6 +357,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 #define QUERY_PORT_MTU_OFFSET			0x01
 #define QUERY_PORT_WIDTH_OFFSET			0x06
 #define QUERY_PORT_MAX_GID_PKEY_OFFSET		0x07
+#define QUERY_PORT_MAX_MACVLAN_OFFSET		0x0a
 #define QUERY_PORT_MAX_VL_OFFSET		0x0b
 
 		for (i = 1; i <= dev_cap->num_ports; ++i) {
@@ -374,6 +375,10 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 			dev_cap->max_pkeys[i]	   = 1 << (field & 0xf);
 			MLX4_GET(field, outbox, QUERY_PORT_MAX_VL_OFFSET);
 			dev_cap->max_vl[i]	   = field & 0xf;
+			MLX4_GET(field, outbox, QUERY_PORT_MAX_MACVLAN_OFFSET);
+			dev_cap->log_max_macs[i]  = field & 0xf;
+			dev_cap->log_max_vlans[i] = field >> 4;
+
 		}
 	}
 
diff --git a/drivers/net/mlx4/fw.h b/drivers/net/mlx4/fw.h
index decbb5c2ad4..c34e726d66e 100644
--- a/drivers/net/mlx4/fw.h
+++ b/drivers/net/mlx4/fw.h
@@ -102,6 +102,8 @@ struct mlx4_dev_cap {
 	u32 reserved_lkey;
 	u64 max_icm_sz;
 	int max_gso_sz;
+	u8  log_max_macs[MLX4_MAX_PORTS + 1];
+	u8  log_max_vlans[MLX4_MAX_PORTS + 1];
 };
 
 struct mlx4_adapter {
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index 1252a919de2..560e1962212 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -85,6 +85,19 @@ static struct mlx4_profile default_profile = {
 	.num_mtt	= 1 << 20,
 };
 
+static int log_num_mac = 2;
+module_param_named(log_num_mac, log_num_mac, int, 0444);
+MODULE_PARM_DESC(log_num_mac, "Log2 max number of MACs per ETH port (1-7)");
+
+static int log_num_vlan;
+module_param_named(log_num_vlan, log_num_vlan, int, 0444);
+MODULE_PARM_DESC(log_num_vlan, "Log2 max number of VLANs per ETH port (0-7)");
+
+static int use_prio;
+module_param_named(use_prio, use_prio, bool, 0444);
+MODULE_PARM_DESC(use_prio, "Enable steering by VLAN priority on ETH ports "
+		  "(0/1, default 0)");
+
 static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 {
 	int err;
@@ -134,7 +147,6 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev->caps.max_rq_sg	     = dev_cap->max_rq_sg;
 	dev->caps.max_wqes	     = dev_cap->max_qp_sz;
 	dev->caps.max_qp_init_rdma   = dev_cap->max_requester_per_qp;
-	dev->caps.reserved_qps	     = dev_cap->reserved_qps;
 	dev->caps.max_srq_wqes	     = dev_cap->max_srq_sz;
 	dev->caps.max_srq_sge	     = dev_cap->max_rq_sg - 1;
 	dev->caps.reserved_srqs	     = dev_cap->reserved_srqs;
@@ -163,6 +175,39 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev->caps.stat_rate_support  = dev_cap->stat_rate_support;
 	dev->caps.max_gso_sz	     = dev_cap->max_gso_sz;
 
+	dev->caps.log_num_macs  = log_num_mac;
+	dev->caps.log_num_vlans = log_num_vlan;
+	dev->caps.log_num_prios = use_prio ? 3 : 0;
+
+	for (i = 1; i <= dev->caps.num_ports; ++i) {
+		if (dev->caps.log_num_macs > dev_cap->log_max_macs[i]) {
+			dev->caps.log_num_macs = dev_cap->log_max_macs[i];
+			mlx4_warn(dev, "Requested number of MACs is too much "
+				  "for port %d, reducing to %d.\n",
+				  i, 1 << dev->caps.log_num_macs);
+		}
+		if (dev->caps.log_num_vlans > dev_cap->log_max_vlans[i]) {
+			dev->caps.log_num_vlans = dev_cap->log_max_vlans[i];
+			mlx4_warn(dev, "Requested number of VLANs is too much "
+				  "for port %d, reducing to %d.\n",
+				  i, 1 << dev->caps.log_num_vlans);
+		}
+	}
+
+	dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW] = dev_cap->reserved_qps;
+	dev->caps.reserved_qps_cnt[MLX4_QP_REGION_ETH_ADDR] =
+		dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FC_ADDR] =
+		(1 << dev->caps.log_num_macs) *
+		(1 << dev->caps.log_num_vlans) *
+		(1 << dev->caps.log_num_prios) *
+		dev->caps.num_ports;
+	dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FC_EXCH] = MLX4_NUM_FEXCH;
+
+	dev->caps.reserved_qps = dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW] +
+		dev->caps.reserved_qps_cnt[MLX4_QP_REGION_ETH_ADDR] +
+		dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FC_ADDR] +
+		dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FC_EXCH];
+
 	return 0;
 }
 
@@ -211,7 +256,8 @@ static int mlx4_init_cmpt_table(struct mlx4_dev *dev, u64 cmpt_base,
 				  ((u64) (MLX4_CMPT_TYPE_QP *
 					  cmpt_entry_sz) << MLX4_CMPT_SHIFT),
 				  cmpt_entry_sz, dev->caps.num_qps,
-				  dev->caps.reserved_qps, 0, 0);
+				  dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW],
+				  0, 0);
 	if (err)
 		goto err;
 
@@ -336,7 +382,8 @@ static int mlx4_init_icm(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap,
 				  init_hca->qpc_base,
 				  dev_cap->qpc_entry_sz,
 				  dev->caps.num_qps,
-				  dev->caps.reserved_qps, 0, 0);
+				  dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW],
+				  0, 0);
 	if (err) {
 		mlx4_err(dev, "Failed to map QP context memory, aborting.\n");
 		goto err_unmap_dmpt;
@@ -346,7 +393,8 @@ static int mlx4_init_icm(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap,
 				  init_hca->auxc_base,
 				  dev_cap->aux_entry_sz,
 				  dev->caps.num_qps,
-				  dev->caps.reserved_qps, 0, 0);
+				  dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW],
+				  0, 0);
 	if (err) {
 		mlx4_err(dev, "Failed to map AUXC context memory, aborting.\n");
 		goto err_unmap_qp;
@@ -356,7 +404,8 @@ static int mlx4_init_icm(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap,
 				  init_hca->altc_base,
 				  dev_cap->altc_entry_sz,
 				  dev->caps.num_qps,
-				  dev->caps.reserved_qps, 0, 0);
+				  dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW],
+				  0, 0);
 	if (err) {
 		mlx4_err(dev, "Failed to map ALTC context memory, aborting.\n");
 		goto err_unmap_auxc;
@@ -366,7 +415,8 @@ static int mlx4_init_icm(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap,
 				  init_hca->rdmarc_base,
 				  dev_cap->rdmarc_entry_sz << priv->qp_table.rdmarc_shift,
 				  dev->caps.num_qps,
-				  dev->caps.reserved_qps, 0, 0);
+				  dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW],
+				  0, 0);
 	if (err) {
 		mlx4_err(dev, "Failed to map RDMARC context memory, aborting\n");
 		goto err_unmap_altc;
diff --git a/drivers/net/mlx4/mcg.c b/drivers/net/mlx4/mcg.c
index c83f88ce073..592c01ae2c5 100644
--- a/drivers/net/mlx4/mcg.c
+++ b/drivers/net/mlx4/mcg.c
@@ -368,8 +368,8 @@ int mlx4_init_mcg_table(struct mlx4_dev *dev)
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	int err;
 
-	err = mlx4_bitmap_init(&priv->mcg_table.bitmap,
-			       dev->caps.num_amgms, dev->caps.num_amgms - 1, 0);
+	err = mlx4_bitmap_init(&priv->mcg_table.bitmap, dev->caps.num_amgms,
+			       dev->caps.num_amgms - 1, 0, 0);
 	if (err)
 		return err;
 
diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h
index b55ddab73f6..9e2f44c3181 100644
--- a/drivers/net/mlx4/mlx4.h
+++ b/drivers/net/mlx4/mlx4.h
@@ -111,6 +111,7 @@ struct mlx4_bitmap {
 	u32			last;
 	u32			top;
 	u32			max;
+	u32                     reserved_top;
 	u32			mask;
 	spinlock_t		lock;
 	unsigned long	       *table;
@@ -290,7 +291,8 @@ u32 mlx4_bitmap_alloc(struct mlx4_bitmap *bitmap);
 void mlx4_bitmap_free(struct mlx4_bitmap *bitmap, u32 obj);
 u32 mlx4_bitmap_alloc_range(struct mlx4_bitmap *bitmap, int cnt, int align);
 void mlx4_bitmap_free_range(struct mlx4_bitmap *bitmap, u32 obj, int cnt);
-int mlx4_bitmap_init(struct mlx4_bitmap *bitmap, u32 num, u32 mask, u32 reserved);
+int mlx4_bitmap_init(struct mlx4_bitmap *bitmap, u32 num, u32 mask,
+		     u32 reserved_bot, u32 resetrved_top);
 void mlx4_bitmap_cleanup(struct mlx4_bitmap *bitmap);
 
 int mlx4_reset(struct mlx4_dev *dev);
diff --git a/drivers/net/mlx4/mr.c b/drivers/net/mlx4/mr.c
index d1dd5b48dbd..0caf74cae8b 100644
--- a/drivers/net/mlx4/mr.c
+++ b/drivers/net/mlx4/mr.c
@@ -461,7 +461,7 @@ int mlx4_init_mr_table(struct mlx4_dev *dev)
 	int err;
 
 	err = mlx4_bitmap_init(&mr_table->mpt_bitmap, dev->caps.num_mpts,
-			       ~0, dev->caps.reserved_mrws);
+			       ~0, dev->caps.reserved_mrws, 0);
 	if (err)
 		return err;
 
diff --git a/drivers/net/mlx4/pd.c b/drivers/net/mlx4/pd.c
index aa616892d09..26d1a7a9e37 100644
--- a/drivers/net/mlx4/pd.c
+++ b/drivers/net/mlx4/pd.c
@@ -62,7 +62,7 @@ int mlx4_init_pd_table(struct mlx4_dev *dev)
 	struct mlx4_priv *priv = mlx4_priv(dev);
 
 	return mlx4_bitmap_init(&priv->pd_bitmap, dev->caps.num_pds,
-				(1 << 24) - 1, dev->caps.reserved_pds);
+				(1 << 24) - 1, dev->caps.reserved_pds, 0);
 }
 
 void mlx4_cleanup_pd_table(struct mlx4_dev *dev)
@@ -100,7 +100,7 @@ int mlx4_init_uar_table(struct mlx4_dev *dev)
 
 	return mlx4_bitmap_init(&mlx4_priv(dev)->uar_table.bitmap,
 				dev->caps.num_uars, dev->caps.num_uars - 1,
-				max(128, dev->caps.reserved_uars));
+				max(128, dev->caps.reserved_uars), 0);
 }
 
 void mlx4_cleanup_uar_table(struct mlx4_dev *dev)
diff --git a/drivers/net/mlx4/qp.c b/drivers/net/mlx4/qp.c
index 98e0c40ba36..1c565ef8d17 100644
--- a/drivers/net/mlx4/qp.c
+++ b/drivers/net/mlx4/qp.c
@@ -272,6 +272,7 @@ int mlx4_init_qp_table(struct mlx4_dev *dev)
 {
 	struct mlx4_qp_table *qp_table = &mlx4_priv(dev)->qp_table;
 	int err;
+	int reserved_from_top = 0;
 
 	spin_lock_init(&qp_table->lock);
 	INIT_RADIX_TREE(&dev->qp_table_tree, GFP_ATOMIC);
@@ -281,9 +282,40 @@ int mlx4_init_qp_table(struct mlx4_dev *dev)
 	 * block of special QPs must be aligned to a multiple of 8, so
 	 * round up.
 	 */
-	dev->caps.sqp_start = ALIGN(dev->caps.reserved_qps, 8);
+	dev->caps.sqp_start =
+		ALIGN(dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW], 8);
+
+	{
+		int sort[MLX4_NUM_QP_REGION];
+		int i, j, tmp;
+		int last_base = dev->caps.num_qps;
+
+		for (i = 1; i < MLX4_NUM_QP_REGION; ++i)
+			sort[i] = i;
+
+		for (i = MLX4_NUM_QP_REGION; i > 0; --i) {
+			for (j = 2; j < i; ++j) {
+				if (dev->caps.reserved_qps_cnt[sort[j]] >
+				    dev->caps.reserved_qps_cnt[sort[j - 1]]) {
+					tmp             = sort[j];
+					sort[j]         = sort[j - 1];
+					sort[j - 1]     = tmp;
+				}
+			}
+		}
+
+		for (i = 1; i < MLX4_NUM_QP_REGION; ++i) {
+			last_base -= dev->caps.reserved_qps_cnt[sort[i]];
+			dev->caps.reserved_qps_base[sort[i]] = last_base;
+			reserved_from_top +=
+				dev->caps.reserved_qps_cnt[sort[i]];
+		}
+
+	}
+
 	err = mlx4_bitmap_init(&qp_table->bitmap, dev->caps.num_qps,
-			       (1 << 24) - 1, dev->caps.sqp_start + 8);
+			       (1 << 23) - 1, dev->caps.sqp_start + 8,
+			       reserved_from_top);
 	if (err)
 		return err;
 
diff --git a/drivers/net/mlx4/srq.c b/drivers/net/mlx4/srq.c
index 533eb6db24b..fe9f218691f 100644
--- a/drivers/net/mlx4/srq.c
+++ b/drivers/net/mlx4/srq.c
@@ -245,7 +245,7 @@ int mlx4_init_srq_table(struct mlx4_dev *dev)
 	INIT_RADIX_TREE(&srq_table->tree, GFP_ATOMIC);
 
 	err = mlx4_bitmap_init(&srq_table->bitmap, dev->caps.num_srqs,
-			       dev->caps.num_srqs - 1, dev->caps.reserved_srqs);
+			       dev->caps.num_srqs - 1, dev->caps.reserved_srqs, 0);
 	if (err)
 		return err;
 
-- 
cgit v1.2.3-18-g5258


From b79acb49de6c2ab9ff0245f0f2b573d48b9a2d93 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Wed, 22 Oct 2008 10:56:48 -0700
Subject: mlx4_core: Get ethernet MTU and default address from firmware

Get maximum ethernet MTU and default MAC address from the firmware
QUERY_DEV_CAP command.

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/net/mlx4/fw.c   | 13 ++++++++-----
 drivers/net/mlx4/fw.h   |  4 +++-
 drivers/net/mlx4/main.c |  4 +++-
 3 files changed, 14 insertions(+), 7 deletions(-)

(limited to 'drivers')

diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index 40d8142c23b..8d402db9a03 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -346,7 +346,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 			MLX4_GET(field, outbox, QUERY_DEV_CAP_VL_PORT_OFFSET);
 			dev_cap->max_vl[i]	   = field >> 4;
 			MLX4_GET(field, outbox, QUERY_DEV_CAP_MTU_WIDTH_OFFSET);
-			dev_cap->max_mtu[i]	   = field >> 4;
+			dev_cap->ib_mtu[i]	   = field >> 4;
 			dev_cap->max_port_width[i] = field & 0xf;
 			MLX4_GET(field, outbox, QUERY_DEV_CAP_MAX_GID_OFFSET);
 			dev_cap->max_gids[i]	   = 1 << (field & 0xf);
@@ -355,8 +355,10 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 		}
 	} else {
 #define QUERY_PORT_MTU_OFFSET			0x01
+#define QUERY_PORT_ETH_MTU_OFFSET		0x02
 #define QUERY_PORT_WIDTH_OFFSET			0x06
 #define QUERY_PORT_MAX_GID_PKEY_OFFSET		0x07
+#define QUERY_PORT_MAC_OFFSET			0x08
 #define QUERY_PORT_MAX_MACVLAN_OFFSET		0x0a
 #define QUERY_PORT_MAX_VL_OFFSET		0x0b
 
@@ -367,7 +369,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 				goto out;
 
 			MLX4_GET(field, outbox, QUERY_PORT_MTU_OFFSET);
-			dev_cap->max_mtu[i]	   = field & 0xf;
+			dev_cap->ib_mtu[i]	   = field & 0xf;
 			MLX4_GET(field, outbox, QUERY_PORT_WIDTH_OFFSET);
 			dev_cap->max_port_width[i] = field & 0xf;
 			MLX4_GET(field, outbox, QUERY_PORT_MAX_GID_PKEY_OFFSET);
@@ -378,7 +380,8 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 			MLX4_GET(field, outbox, QUERY_PORT_MAX_MACVLAN_OFFSET);
 			dev_cap->log_max_macs[i]  = field & 0xf;
 			dev_cap->log_max_vlans[i] = field >> 4;
-
+			MLX4_GET(dev_cap->eth_mtu[i], outbox, QUERY_PORT_ETH_MTU_OFFSET);
+			MLX4_GET(dev_cap->def_mac[i], outbox, QUERY_PORT_MAC_OFFSET);
 		}
 	}
 
@@ -412,7 +415,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	mlx4_dbg(dev, "Max CQEs: %d, max WQEs: %d, max SRQ WQEs: %d\n",
 		 dev_cap->max_cq_sz, dev_cap->max_qp_sz, dev_cap->max_srq_sz);
 	mlx4_dbg(dev, "Local CA ACK delay: %d, max MTU: %d, port width cap: %d\n",
-		 dev_cap->local_ca_ack_delay, 128 << dev_cap->max_mtu[1],
+		 dev_cap->local_ca_ack_delay, 128 << dev_cap->ib_mtu[1],
 		 dev_cap->max_port_width[1]);
 	mlx4_dbg(dev, "Max SQ desc size: %d, max SQ S/G: %d\n",
 		 dev_cap->max_sq_desc_sz, dev_cap->max_sq_sg);
@@ -824,7 +827,7 @@ int mlx4_INIT_PORT(struct mlx4_dev *dev, int port)
 		flags |= (dev->caps.port_width_cap[port] & 0xf) << INIT_PORT_PORT_WIDTH_SHIFT;
 		MLX4_PUT(inbox, flags,		  INIT_PORT_FLAGS_OFFSET);
 
-		field = 128 << dev->caps.mtu_cap[port];
+		field = 128 << dev->caps.ib_mtu_cap[port];
 		MLX4_PUT(inbox, field, INIT_PORT_MTU_OFFSET);
 		field = dev->caps.gid_table_len[port];
 		MLX4_PUT(inbox, field, INIT_PORT_MAX_GID_OFFSET);
diff --git a/drivers/net/mlx4/fw.h b/drivers/net/mlx4/fw.h
index c34e726d66e..d0913d4d262 100644
--- a/drivers/net/mlx4/fw.h
+++ b/drivers/net/mlx4/fw.h
@@ -66,11 +66,13 @@ struct mlx4_dev_cap {
 	int local_ca_ack_delay;
 	int num_ports;
 	u32 max_msg_sz;
-	int max_mtu[MLX4_MAX_PORTS + 1];
+	int ib_mtu[MLX4_MAX_PORTS + 1];
 	int max_port_width[MLX4_MAX_PORTS + 1];
 	int max_vl[MLX4_MAX_PORTS + 1];
 	int max_gids[MLX4_MAX_PORTS + 1];
 	int max_pkeys[MLX4_MAX_PORTS + 1];
+	u64 def_mac[MLX4_MAX_PORTS + 1];
+	u16 eth_mtu[MLX4_MAX_PORTS + 1];
 	u16 stat_rate_support;
 	u32 flags;
 	int reserved_uars;
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index 560e1962212..28f36b88de3 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -133,10 +133,12 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev->caps.num_ports	     = dev_cap->num_ports;
 	for (i = 1; i <= dev->caps.num_ports; ++i) {
 		dev->caps.vl_cap[i]	    = dev_cap->max_vl[i];
-		dev->caps.mtu_cap[i]	    = dev_cap->max_mtu[i];
+		dev->caps.ib_mtu_cap[i]	    = dev_cap->ib_mtu[i];
 		dev->caps.gid_table_len[i]  = dev_cap->max_gids[i];
 		dev->caps.pkey_table_len[i] = dev_cap->max_pkeys[i];
 		dev->caps.port_width_cap[i] = dev_cap->max_port_width[i];
+		dev->caps.eth_mtu_cap[i]    = dev_cap->eth_mtu[i];
+		dev->caps.def_mac[i]        = dev_cap->def_mac[i];
 	}
 
 	dev->caps.num_uars	     = dev_cap->uar_size / PAGE_SIZE;
-- 
cgit v1.2.3-18-g5258


From 2a2336f8228292b8197f4187e54b0748903e6645 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Wed, 22 Oct 2008 11:44:46 -0700
Subject: mlx4_core: Ethernet MAC/VLAN management

Add support for managing MAC and VLAN filters for each port.

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: Oren Duer <oren@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/net/mlx4/Makefile |   2 +-
 drivers/net/mlx4/main.c   |  14 +++
 drivers/net/mlx4/mlx4.h   |  33 ++++++
 drivers/net/mlx4/port.c   | 259 ++++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 307 insertions(+), 1 deletion(-)
 create mode 100644 drivers/net/mlx4/port.c

(limited to 'drivers')

diff --git a/drivers/net/mlx4/Makefile b/drivers/net/mlx4/Makefile
index 0952a6528f5..9f493666e27 100644
--- a/drivers/net/mlx4/Makefile
+++ b/drivers/net/mlx4/Makefile
@@ -1,4 +1,4 @@
 obj-$(CONFIG_MLX4_CORE)		+= mlx4_core.o
 
 mlx4_core-y :=	alloc.o catas.o cmd.o cq.o eq.o fw.o icm.o intf.o main.o mcg.o \
-		mr.o pd.o profile.o qp.o reset.o srq.o
+		mr.o pd.o port.o profile.o qp.o reset.o srq.o
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index 28f36b88de3..0a5c8bfb3f1 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -780,11 +780,22 @@ no_msi:
 		priv->eq_table.eq[i].irq = dev->pdev->irq;
 }
 
+static void mlx4_init_port_info(struct mlx4_dev *dev, int port)
+{
+	struct mlx4_port_info *info = &mlx4_priv(dev)->port[port];
+
+	info->dev = dev;
+	info->port = port;
+	mlx4_init_mac_table(dev, &info->mac_table);
+	mlx4_init_vlan_table(dev, &info->vlan_table);
+}
+
 static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	struct mlx4_priv *priv;
 	struct mlx4_dev *dev;
 	int err;
+	int port;
 
 	printk(KERN_INFO PFX "Initializing %s\n",
 	       pci_name(pdev));
@@ -894,6 +905,9 @@ static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (err)
 		goto err_close;
 
+	for (port = 1; port <= dev->caps.num_ports; port++)
+		mlx4_init_port_info(dev, port);
+
 	err = mlx4_register_device(dev);
 	if (err)
 		goto err_cleanup;
diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h
index 9e2f44c3181..23309f381ee 100644
--- a/drivers/net/mlx4/mlx4.h
+++ b/drivers/net/mlx4/mlx4.h
@@ -252,6 +252,35 @@ struct mlx4_catas_err {
 	struct list_head	list;
 };
 
+#define MLX4_MAX_MAC_NUM	128
+#define MLX4_MAC_TABLE_SIZE	(MLX4_MAX_MAC_NUM << 3)
+
+struct mlx4_mac_table {
+	__be64			entries[MLX4_MAX_MAC_NUM];
+	int			refs[MLX4_MAX_MAC_NUM];
+	struct mutex		mutex;
+	int			total;
+	int			max;
+};
+
+#define MLX4_MAX_VLAN_NUM	128
+#define MLX4_VLAN_TABLE_SIZE	(MLX4_MAX_VLAN_NUM << 2)
+
+struct mlx4_vlan_table {
+	__be32			entries[MLX4_MAX_VLAN_NUM];
+	int			refs[MLX4_MAX_VLAN_NUM];
+	struct mutex		mutex;
+	int			total;
+	int			max;
+};
+
+struct mlx4_port_info {
+	struct mlx4_dev	       *dev;
+	int			port;
+	struct mlx4_mac_table	mac_table;
+	struct mlx4_vlan_table	vlan_table;
+};
+
 struct mlx4_priv {
 	struct mlx4_dev		dev;
 
@@ -280,6 +309,7 @@ struct mlx4_priv {
 
 	struct mlx4_uar		driver_uar;
 	void __iomem	       *kar;
+	struct mlx4_port_info	port[MLX4_MAX_PORTS + 1];
 };
 
 static inline struct mlx4_priv *mlx4_priv(struct mlx4_dev *dev)
@@ -350,4 +380,7 @@ void mlx4_srq_event(struct mlx4_dev *dev, u32 srqn, int event_type);
 
 void mlx4_handle_catas_err(struct mlx4_dev *dev);
 
+void mlx4_init_mac_table(struct mlx4_dev *dev, struct mlx4_mac_table *table);
+void mlx4_init_vlan_table(struct mlx4_dev *dev, struct mlx4_vlan_table *table);
+
 #endif /* MLX4_H */
diff --git a/drivers/net/mlx4/port.c b/drivers/net/mlx4/port.c
new file mode 100644
index 00000000000..8644f3d978e
--- /dev/null
+++ b/drivers/net/mlx4/port.c
@@ -0,0 +1,259 @@
+/*
+ * Copyright (c) 2007 Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/errno.h>
+#include <linux/if_ether.h>
+
+#include <linux/mlx4/cmd.h>
+
+#include "mlx4.h"
+
+#define MLX4_MAC_VALID		(1ull << 63)
+#define MLX4_MAC_MASK		0xffffffffffffULL
+
+#define MLX4_VLAN_VALID		(1u << 31)
+#define MLX4_VLAN_MASK		0xfff
+
+void mlx4_init_mac_table(struct mlx4_dev *dev, struct mlx4_mac_table *table)
+{
+	int i;
+
+	mutex_init(&table->mutex);
+	for (i = 0; i < MLX4_MAX_MAC_NUM; i++) {
+		table->entries[i] = 0;
+		table->refs[i]	 = 0;
+	}
+	table->max   = 1 << dev->caps.log_num_macs;
+	table->total = 0;
+}
+
+void mlx4_init_vlan_table(struct mlx4_dev *dev, struct mlx4_vlan_table *table)
+{
+	int i;
+
+	mutex_init(&table->mutex);
+	for (i = 0; i < MLX4_MAX_VLAN_NUM; i++) {
+		table->entries[i] = 0;
+		table->refs[i]	 = 0;
+	}
+	table->max   = 1 << dev->caps.log_num_vlans;
+	table->total = 0;
+}
+
+static int mlx4_set_port_mac_table(struct mlx4_dev *dev, u8 port,
+				   __be64 *entries)
+{
+	struct mlx4_cmd_mailbox *mailbox;
+	u32 in_mod;
+	int err;
+
+	mailbox = mlx4_alloc_cmd_mailbox(dev);
+	if (IS_ERR(mailbox))
+		return PTR_ERR(mailbox);
+
+	memcpy(mailbox->buf, entries, MLX4_MAC_TABLE_SIZE);
+
+	in_mod = MLX4_SET_PORT_MAC_TABLE << 8 | port;
+	err = mlx4_cmd(dev, mailbox->dma, in_mod, 1, MLX4_CMD_SET_PORT,
+		       MLX4_CMD_TIME_CLASS_B);
+
+	mlx4_free_cmd_mailbox(dev, mailbox);
+	return err;
+}
+
+int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *index)
+{
+	struct mlx4_mac_table *table = &mlx4_priv(dev)->port[port].mac_table;
+	int i, err = 0;
+	int free = -1;
+
+	mlx4_dbg(dev, "Registering MAC: 0x%llx\n", (unsigned long long) mac);
+	mutex_lock(&table->mutex);
+	for (i = 0; i < MLX4_MAX_MAC_NUM - 1; i++) {
+		if (free < 0 && !table->refs[i]) {
+			free = i;
+			continue;
+		}
+
+		if (mac == (MLX4_MAC_MASK & be64_to_cpu(table->entries[i]))) {
+			/* MAC already registered, increase refernce count */
+			*index = i;
+			++table->refs[i];
+			goto out;
+		}
+	}
+	mlx4_dbg(dev, "Free MAC index is %d\n", free);
+
+	if (table->total == table->max) {
+		/* No free mac entries */
+		err = -ENOSPC;
+		goto out;
+	}
+
+	/* Register new MAC */
+	table->refs[free] = 1;
+	table->entries[free] = cpu_to_be64(mac | MLX4_MAC_VALID);
+
+	err = mlx4_set_port_mac_table(dev, port, table->entries);
+	if (unlikely(err)) {
+		mlx4_err(dev, "Failed adding MAC: 0x%llx\n", (unsigned long long) mac);
+		table->refs[free] = 0;
+		table->entries[free] = 0;
+		goto out;
+	}
+
+	*index = free;
+	++table->total;
+out:
+	mutex_unlock(&table->mutex);
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx4_register_mac);
+
+void mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, int index)
+{
+	struct mlx4_mac_table *table = &mlx4_priv(dev)->port[port].mac_table;
+
+	mutex_lock(&table->mutex);
+	if (!table->refs[index]) {
+		mlx4_warn(dev, "No MAC entry for index %d\n", index);
+		goto out;
+	}
+	if (--table->refs[index]) {
+		mlx4_warn(dev, "Have more references for index %d,"
+			  "no need to modify MAC table\n", index);
+		goto out;
+	}
+	table->entries[index] = 0;
+	mlx4_set_port_mac_table(dev, port, table->entries);
+	--table->total;
+out:
+	mutex_unlock(&table->mutex);
+}
+EXPORT_SYMBOL_GPL(mlx4_unregister_mac);
+
+static int mlx4_set_port_vlan_table(struct mlx4_dev *dev, u8 port,
+				    __be32 *entries)
+{
+	struct mlx4_cmd_mailbox *mailbox;
+	u32 in_mod;
+	int err;
+
+	mailbox = mlx4_alloc_cmd_mailbox(dev);
+	if (IS_ERR(mailbox))
+		return PTR_ERR(mailbox);
+
+	memcpy(mailbox->buf, entries, MLX4_VLAN_TABLE_SIZE);
+	in_mod = MLX4_SET_PORT_VLAN_TABLE << 8 | port;
+	err = mlx4_cmd(dev, mailbox->dma, in_mod, 1, MLX4_CMD_SET_PORT,
+		       MLX4_CMD_TIME_CLASS_B);
+
+	mlx4_free_cmd_mailbox(dev, mailbox);
+
+	return err;
+}
+
+int mlx4_register_vlan(struct mlx4_dev *dev, u8 port, u16 vlan, int *index)
+{
+	struct mlx4_vlan_table *table = &mlx4_priv(dev)->port[port].vlan_table;
+	int i, err = 0;
+	int free = -1;
+
+	mutex_lock(&table->mutex);
+	for (i = MLX4_VLAN_REGULAR; i < MLX4_MAX_VLAN_NUM; i++) {
+		if (free < 0 && (table->refs[i] == 0)) {
+			free = i;
+			continue;
+		}
+
+		if (table->refs[i] &&
+		    (vlan == (MLX4_VLAN_MASK &
+			      be32_to_cpu(table->entries[i])))) {
+			/* Vlan already registered, increase refernce count */
+			*index = i;
+			++table->refs[i];
+			goto out;
+		}
+	}
+
+	if (table->total == table->max) {
+		/* No free vlan entries */
+		err = -ENOSPC;
+		goto out;
+	}
+
+	/* Register new MAC */
+	table->refs[free] = 1;
+	table->entries[free] = cpu_to_be32(vlan | MLX4_VLAN_VALID);
+
+	err = mlx4_set_port_vlan_table(dev, port, table->entries);
+	if (unlikely(err)) {
+		mlx4_warn(dev, "Failed adding vlan: %u\n", vlan);
+		table->refs[free] = 0;
+		table->entries[free] = 0;
+		goto out;
+	}
+
+	*index = free;
+	++table->total;
+out:
+	mutex_unlock(&table->mutex);
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx4_register_vlan);
+
+void mlx4_unregister_vlan(struct mlx4_dev *dev, u8 port, int index)
+{
+	struct mlx4_vlan_table *table = &mlx4_priv(dev)->port[port].vlan_table;
+
+	if (index < MLX4_VLAN_REGULAR) {
+		mlx4_warn(dev, "Trying to free special vlan index %d\n", index);
+		return;
+	}
+
+	mutex_lock(&table->mutex);
+	if (!table->refs[index]) {
+		mlx4_warn(dev, "No vlan entry for index %d\n", index);
+		goto out;
+	}
+	if (--table->refs[index]) {
+		mlx4_dbg(dev, "Have more references for index %d,"
+			 "no need to modify vlan table\n", index);
+		goto out;
+	}
+	table->entries[index] = 0;
+	mlx4_set_port_vlan_table(dev, port, table->entries);
+	--table->total;
+out:
+	mutex_unlock(&table->mutex);
+}
+EXPORT_SYMBOL_GPL(mlx4_unregister_vlan);
-- 
cgit v1.2.3-18-g5258


From 7ff93f8b7ecbc36e7ffc5c11a61643821c1bfee5 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Wed, 22 Oct 2008 15:38:42 -0700
Subject: mlx4_core: Multiple port type support

Multi-protocol adapters support different port types.  Each consumer
of mlx4_core queries for supported port types; in particular mlx4_ib
can no longer assume that all physical ports belong to it.  Port type
is configured through a sysfs interface.  When the type of a port is
changed, all mlx4 interfaces are unregistered, and then registered
again with the new port types.

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/mlx4/mad.c     |   6 +-
 drivers/infiniband/hw/mlx4/main.c    |  11 +-
 drivers/infiniband/hw/mlx4/mlx4_ib.h |   1 +
 drivers/net/mlx4/fw.c                |   4 +
 drivers/net/mlx4/fw.h                |   1 +
 drivers/net/mlx4/main.c              | 211 ++++++++++++++++++++++++++++++++++-
 drivers/net/mlx4/mlx4.h              |   6 +
 drivers/net/mlx4/port.c              |  23 ++++
 8 files changed, 252 insertions(+), 11 deletions(-)

(limited to 'drivers')

diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
index cdca3a511e1..606f1e2ef28 100644
--- a/drivers/infiniband/hw/mlx4/mad.c
+++ b/drivers/infiniband/hw/mlx4/mad.c
@@ -298,7 +298,7 @@ int mlx4_ib_mad_init(struct mlx4_ib_dev *dev)
 	int p, q;
 	int ret;
 
-	for (p = 0; p < dev->dev->caps.num_ports; ++p)
+	for (p = 0; p < dev->num_ports; ++p)
 		for (q = 0; q <= 1; ++q) {
 			agent = ib_register_mad_agent(&dev->ib_dev, p + 1,
 						      q ? IB_QPT_GSI : IB_QPT_SMI,
@@ -314,7 +314,7 @@ int mlx4_ib_mad_init(struct mlx4_ib_dev *dev)
 	return 0;
 
 err:
-	for (p = 0; p < dev->dev->caps.num_ports; ++p)
+	for (p = 0; p < dev->num_ports; ++p)
 		for (q = 0; q <= 1; ++q)
 			if (dev->send_agent[p][q])
 				ib_unregister_mad_agent(dev->send_agent[p][q]);
@@ -327,7 +327,7 @@ void mlx4_ib_mad_cleanup(struct mlx4_ib_dev *dev)
 	struct ib_mad_agent *agent;
 	int p, q;
 
-	for (p = 0; p < dev->dev->caps.num_ports; ++p) {
+	for (p = 0; p < dev->num_ports; ++p) {
 		for (q = 0; q <= 1; ++q) {
 			agent = dev->send_agent[p][q];
 			dev->send_agent[p][q] = NULL;
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index a3c2851c054..2e80f8f47b0 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -574,7 +574,10 @@ static void *mlx4_ib_add(struct mlx4_dev *dev)
 	ibdev->ib_dev.owner		= THIS_MODULE;
 	ibdev->ib_dev.node_type		= RDMA_NODE_IB_CA;
 	ibdev->ib_dev.local_dma_lkey	= dev->caps.reserved_lkey;
-	ibdev->ib_dev.phys_port_cnt	= dev->caps.num_ports;
+	ibdev->num_ports = 0;
+	mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_IB)
+		ibdev->num_ports++;
+	ibdev->ib_dev.phys_port_cnt     = ibdev->num_ports;
 	ibdev->ib_dev.num_comp_vectors	= 1;
 	ibdev->ib_dev.dma_device	= &dev->pdev->dev;
 
@@ -691,7 +694,7 @@ static void mlx4_ib_remove(struct mlx4_dev *dev, void *ibdev_ptr)
 	struct mlx4_ib_dev *ibdev = ibdev_ptr;
 	int p;
 
-	for (p = 1; p <= dev->caps.num_ports; ++p)
+	for (p = 1; p <= ibdev->num_ports; ++p)
 		mlx4_CLOSE_PORT(dev, p);
 
 	mlx4_ib_mad_cleanup(ibdev);
@@ -706,6 +709,10 @@ static void mlx4_ib_event(struct mlx4_dev *dev, void *ibdev_ptr,
 			  enum mlx4_dev_event event, int port)
 {
 	struct ib_event ibev;
+	struct mlx4_ib_dev *ibdev = to_mdev((struct ib_device *) ibdev_ptr);
+
+	if (port > ibdev->num_ports)
+		return;
 
 	switch (event) {
 	case MLX4_DEV_EVENT_PORT_UP:
diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h
index 6e2b0dc21b6..9974e886b8d 100644
--- a/drivers/infiniband/hw/mlx4/mlx4_ib.h
+++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h
@@ -162,6 +162,7 @@ struct mlx4_ib_ah {
 struct mlx4_ib_dev {
 	struct ib_device	ib_dev;
 	struct mlx4_dev	       *dev;
+	int			num_ports;
 	void __iomem	       *uar_map;
 
 	struct mlx4_uar		priv_uar;
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index 8d402db9a03..be09fdb79cb 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -88,6 +88,7 @@ static void dump_dev_cap_flags(struct mlx4_dev *dev, u32 flags)
 		[ 8] = "P_Key violation counter",
 		[ 9] = "Q_Key violation counter",
 		[10] = "VMM",
+		[12] = "DPDP",
 		[16] = "MW support",
 		[17] = "APM support",
 		[18] = "Atomic ops support",
@@ -354,6 +355,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 			dev_cap->max_pkeys[i]	   = 1 << (field & 0xf);
 		}
 	} else {
+#define QUERY_PORT_SUPPORTED_TYPE_OFFSET	0x00
 #define QUERY_PORT_MTU_OFFSET			0x01
 #define QUERY_PORT_ETH_MTU_OFFSET		0x02
 #define QUERY_PORT_WIDTH_OFFSET			0x06
@@ -368,6 +370,8 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 			if (err)
 				goto out;
 
+			MLX4_GET(field, outbox, QUERY_PORT_SUPPORTED_TYPE_OFFSET);
+			dev_cap->supported_port_types[i] = field & 3;
 			MLX4_GET(field, outbox, QUERY_PORT_MTU_OFFSET);
 			dev_cap->ib_mtu[i]	   = field & 0xf;
 			MLX4_GET(field, outbox, QUERY_PORT_WIDTH_OFFSET);
diff --git a/drivers/net/mlx4/fw.h b/drivers/net/mlx4/fw.h
index d0913d4d262..526d7f30c04 100644
--- a/drivers/net/mlx4/fw.h
+++ b/drivers/net/mlx4/fw.h
@@ -104,6 +104,7 @@ struct mlx4_dev_cap {
 	u32 reserved_lkey;
 	u64 max_icm_sz;
 	int max_gso_sz;
+	u8  supported_port_types[MLX4_MAX_PORTS + 1];
 	u8  log_max_macs[MLX4_MAX_PORTS + 1];
 	u8  log_max_vlans[MLX4_MAX_PORTS + 1];
 };
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index 0a5c8bfb3f1..c1d447873bf 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -98,6 +98,44 @@ module_param_named(use_prio, use_prio, bool, 0444);
 MODULE_PARM_DESC(use_prio, "Enable steering by VLAN priority on ETH ports "
 		  "(0/1, default 0)");
 
+static int mlx4_check_port_params(struct mlx4_dev *dev,
+				  enum mlx4_port_type *port_type)
+{
+	int i;
+
+	for (i = 0; i < dev->caps.num_ports - 1; i++) {
+		if (port_type[i] != port_type[i+1] &&
+		    !(dev->caps.flags & MLX4_DEV_CAP_FLAG_DPDP)) {
+			mlx4_err(dev, "Only same port types supported "
+				 "on this HCA, aborting.\n");
+			return -EINVAL;
+		}
+	}
+	if ((port_type[0] == MLX4_PORT_TYPE_ETH) &&
+	    (port_type[1] == MLX4_PORT_TYPE_IB)) {
+		mlx4_err(dev, "eth-ib configuration is not supported.\n");
+		return -EINVAL;
+	}
+
+	for (i = 0; i < dev->caps.num_ports; i++) {
+		if (!(port_type[i] & dev->caps.supported_type[i+1])) {
+			mlx4_err(dev, "Requested port type for port %d is not "
+				      "supported on this HCA\n", i + 1);
+			return -EINVAL;
+		}
+	}
+	return 0;
+}
+
+static void mlx4_set_port_mask(struct mlx4_dev *dev)
+{
+	int i;
+
+	dev->caps.port_mask = 0;
+	for (i = 1; i <= dev->caps.num_ports; ++i)
+		if (dev->caps.port_type[i] == MLX4_PORT_TYPE_IB)
+			dev->caps.port_mask |= 1 << (i - 1);
+}
 static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 {
 	int err;
@@ -139,6 +177,7 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 		dev->caps.port_width_cap[i] = dev_cap->max_port_width[i];
 		dev->caps.eth_mtu_cap[i]    = dev_cap->eth_mtu[i];
 		dev->caps.def_mac[i]        = dev_cap->def_mac[i];
+		dev->caps.supported_type[i] = dev_cap->supported_port_types[i];
 	}
 
 	dev->caps.num_uars	     = dev_cap->uar_size / PAGE_SIZE;
@@ -182,6 +221,11 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev->caps.log_num_prios = use_prio ? 3 : 0;
 
 	for (i = 1; i <= dev->caps.num_ports; ++i) {
+		if (dev->caps.supported_type[i] != MLX4_PORT_TYPE_ETH)
+			dev->caps.port_type[i] = MLX4_PORT_TYPE_IB;
+		else
+			dev->caps.port_type[i] = MLX4_PORT_TYPE_ETH;
+
 		if (dev->caps.log_num_macs > dev_cap->log_max_macs[i]) {
 			dev->caps.log_num_macs = dev_cap->log_max_macs[i];
 			mlx4_warn(dev, "Requested number of MACs is too much "
@@ -196,6 +240,8 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 		}
 	}
 
+	mlx4_set_port_mask(dev);
+
 	dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW] = dev_cap->reserved_qps;
 	dev->caps.reserved_qps_cnt[MLX4_QP_REGION_ETH_ADDR] =
 		dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FC_ADDR] =
@@ -213,6 +259,95 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	return 0;
 }
 
+/*
+ * Change the port configuration of the device.
+ * Every user of this function must hold the port mutex.
+ */
+static int mlx4_change_port_types(struct mlx4_dev *dev,
+				  enum mlx4_port_type *port_types)
+{
+	int err = 0;
+	int change = 0;
+	int port;
+
+	for (port = 0; port <  dev->caps.num_ports; port++) {
+		if (port_types[port] != dev->caps.port_type[port + 1]) {
+			change = 1;
+			dev->caps.port_type[port + 1] = port_types[port];
+		}
+	}
+	if (change) {
+		mlx4_unregister_device(dev);
+		for (port = 1; port <= dev->caps.num_ports; port++) {
+			mlx4_CLOSE_PORT(dev, port);
+			err = mlx4_SET_PORT(dev, port);
+			if (err) {
+				mlx4_err(dev, "Failed to set port %d, "
+					      "aborting\n", port);
+				goto out;
+			}
+		}
+		mlx4_set_port_mask(dev);
+		err = mlx4_register_device(dev);
+	}
+
+out:
+	return err;
+}
+
+static ssize_t show_port_type(struct device *dev,
+			      struct device_attribute *attr,
+			      char *buf)
+{
+	struct mlx4_port_info *info = container_of(attr, struct mlx4_port_info,
+						   port_attr);
+	struct mlx4_dev *mdev = info->dev;
+
+	return sprintf(buf, "%s\n",
+		       mdev->caps.port_type[info->port] == MLX4_PORT_TYPE_IB ?
+		       "ib" : "eth");
+}
+
+static ssize_t set_port_type(struct device *dev,
+			     struct device_attribute *attr,
+			     const char *buf, size_t count)
+{
+	struct mlx4_port_info *info = container_of(attr, struct mlx4_port_info,
+						   port_attr);
+	struct mlx4_dev *mdev = info->dev;
+	struct mlx4_priv *priv = mlx4_priv(mdev);
+	enum mlx4_port_type types[MLX4_MAX_PORTS];
+	int i;
+	int err = 0;
+
+	if (!strcmp(buf, "ib\n"))
+		info->tmp_type = MLX4_PORT_TYPE_IB;
+	else if (!strcmp(buf, "eth\n"))
+		info->tmp_type = MLX4_PORT_TYPE_ETH;
+	else {
+		mlx4_err(mdev, "%s is not supported port type\n", buf);
+		return -EINVAL;
+	}
+
+	mutex_lock(&priv->port_mutex);
+	for (i = 0; i < mdev->caps.num_ports; i++)
+		types[i] = priv->port[i+1].tmp_type ? priv->port[i+1].tmp_type :
+					mdev->caps.port_type[i+1];
+
+	err = mlx4_check_port_params(mdev, types);
+	if (err)
+		goto out;
+
+	for (i = 1; i <= mdev->caps.num_ports; i++)
+		priv->port[i].tmp_type = 0;
+
+	err = mlx4_change_port_types(mdev, types);
+
+out:
+	mutex_unlock(&priv->port_mutex);
+	return err ? err : count;
+}
+
 static int mlx4_load_fw(struct mlx4_dev *dev)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
@@ -617,6 +752,7 @@ static int mlx4_setup_hca(struct mlx4_dev *dev)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	int err;
+	int port;
 
 	err = mlx4_init_uar_table(dev);
 	if (err) {
@@ -715,8 +851,20 @@ static int mlx4_setup_hca(struct mlx4_dev *dev)
 		goto err_qp_table_free;
 	}
 
+	for (port = 1; port <= dev->caps.num_ports; port++) {
+		err = mlx4_SET_PORT(dev, port);
+		if (err) {
+			mlx4_err(dev, "Failed to set port %d, aborting\n",
+				port);
+			goto err_mcg_table_free;
+		}
+	}
+
 	return 0;
 
+err_mcg_table_free:
+	mlx4_cleanup_mcg_table(dev);
+
 err_qp_table_free:
 	mlx4_cleanup_qp_table(dev);
 
@@ -780,14 +928,37 @@ no_msi:
 		priv->eq_table.eq[i].irq = dev->pdev->irq;
 }
 
-static void mlx4_init_port_info(struct mlx4_dev *dev, int port)
+static int mlx4_init_port_info(struct mlx4_dev *dev, int port)
 {
 	struct mlx4_port_info *info = &mlx4_priv(dev)->port[port];
+	int err = 0;
 
 	info->dev = dev;
 	info->port = port;
 	mlx4_init_mac_table(dev, &info->mac_table);
 	mlx4_init_vlan_table(dev, &info->vlan_table);
+
+	sprintf(info->dev_name, "mlx4_port%d", port);
+	info->port_attr.attr.name = info->dev_name;
+	info->port_attr.attr.mode = S_IRUGO | S_IWUSR;
+	info->port_attr.show      = show_port_type;
+	info->port_attr.store     = set_port_type;
+
+	err = device_create_file(&dev->pdev->dev, &info->port_attr);
+	if (err) {
+		mlx4_err(dev, "Failed to create file for port %d\n", port);
+		info->port = -1;
+	}
+
+	return err;
+}
+
+static void mlx4_cleanup_port_info(struct mlx4_port_info *info)
+{
+	if (info->port < 0)
+		return;
+
+	device_remove_file(&info->dev->pdev->dev, &info->port_attr);
 }
 
 static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
@@ -870,6 +1041,8 @@ static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	INIT_LIST_HEAD(&priv->ctx_list);
 	spin_lock_init(&priv->ctx_lock);
 
+	mutex_init(&priv->port_mutex);
+
 	INIT_LIST_HEAD(&priv->pgdir_list);
 	mutex_init(&priv->pgdir_mutex);
 
@@ -905,18 +1078,24 @@ static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (err)
 		goto err_close;
 
-	for (port = 1; port <= dev->caps.num_ports; port++)
-		mlx4_init_port_info(dev, port);
+	for (port = 1; port <= dev->caps.num_ports; port++) {
+		err = mlx4_init_port_info(dev, port);
+		if (err)
+			goto err_port;
+	}
 
 	err = mlx4_register_device(dev);
 	if (err)
-		goto err_cleanup;
+		goto err_port;
 
 	pci_set_drvdata(pdev, dev);
 
 	return 0;
 
-err_cleanup:
+err_port:
+	for (port = 1; port <= dev->caps.num_ports; port++)
+		mlx4_cleanup_port_info(&priv->port[port]);
+
 	mlx4_cleanup_mcg_table(dev);
 	mlx4_cleanup_qp_table(dev);
 	mlx4_cleanup_srq_table(dev);
@@ -973,8 +1152,10 @@ static void mlx4_remove_one(struct pci_dev *pdev)
 	if (dev) {
 		mlx4_unregister_device(dev);
 
-		for (p = 1; p <= dev->caps.num_ports; ++p)
+		for (p = 1; p <= dev->caps.num_ports; p++) {
+			mlx4_cleanup_port_info(&priv->port[p]);
 			mlx4_CLOSE_PORT(dev, p);
+		}
 
 		mlx4_cleanup_mcg_table(dev);
 		mlx4_cleanup_qp_table(dev);
@@ -1026,10 +1207,28 @@ static struct pci_driver mlx4_driver = {
 	.remove		= __devexit_p(mlx4_remove_one)
 };
 
+static int __init mlx4_verify_params(void)
+{
+	if ((log_num_mac < 0) || (log_num_mac > 7)) {
+		printk(KERN_WARNING "mlx4_core: bad num_mac: %d\n", log_num_mac);
+		return -1;
+	}
+
+	if ((log_num_vlan < 0) || (log_num_vlan > 7)) {
+		printk(KERN_WARNING "mlx4_core: bad num_vlan: %d\n", log_num_vlan);
+		return -1;
+	}
+
+	return 0;
+}
+
 static int __init mlx4_init(void)
 {
 	int ret;
 
+	if (mlx4_verify_params())
+		return -EINVAL;
+
 	ret = mlx4_catas_init();
 	if (ret)
 		return ret;
diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h
index 23309f381ee..fa431fad0ee 100644
--- a/drivers/net/mlx4/mlx4.h
+++ b/drivers/net/mlx4/mlx4.h
@@ -277,6 +277,9 @@ struct mlx4_vlan_table {
 struct mlx4_port_info {
 	struct mlx4_dev	       *dev;
 	int			port;
+	char			dev_name[16];
+	struct device_attribute port_attr;
+	enum mlx4_port_type	tmp_type;
 	struct mlx4_mac_table	mac_table;
 	struct mlx4_vlan_table	vlan_table;
 };
@@ -310,6 +313,7 @@ struct mlx4_priv {
 	struct mlx4_uar		driver_uar;
 	void __iomem	       *kar;
 	struct mlx4_port_info	port[MLX4_MAX_PORTS + 1];
+	struct mutex		port_mutex;
 };
 
 static inline struct mlx4_priv *mlx4_priv(struct mlx4_dev *dev)
@@ -383,4 +387,6 @@ void mlx4_handle_catas_err(struct mlx4_dev *dev);
 void mlx4_init_mac_table(struct mlx4_dev *dev, struct mlx4_mac_table *table);
 void mlx4_init_vlan_table(struct mlx4_dev *dev, struct mlx4_vlan_table *table);
 
+int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port);
+
 #endif /* MLX4_H */
diff --git a/drivers/net/mlx4/port.c b/drivers/net/mlx4/port.c
index 8644f3d978e..e2fdab42c4c 100644
--- a/drivers/net/mlx4/port.c
+++ b/drivers/net/mlx4/port.c
@@ -257,3 +257,26 @@ out:
 	mutex_unlock(&table->mutex);
 }
 EXPORT_SYMBOL_GPL(mlx4_unregister_vlan);
+
+int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port)
+{
+	struct mlx4_cmd_mailbox *mailbox;
+	int err;
+	u8 is_eth = dev->caps.port_type[port] == MLX4_PORT_TYPE_ETH;
+
+	mailbox = mlx4_alloc_cmd_mailbox(dev);
+	if (IS_ERR(mailbox))
+		return PTR_ERR(mailbox);
+
+	memset(mailbox->buf, 0, 256);
+	if (is_eth) {
+		((u8 *) mailbox->buf)[3] = 6;
+		((__be16 *) mailbox->buf)[4] = cpu_to_be16(1 << 15);
+		((__be16 *) mailbox->buf)[6] = cpu_to_be16(1 << 15);
+	}
+	err = mlx4_cmd(dev, mailbox->dma, port, is_eth, MLX4_CMD_SET_PORT,
+		       MLX4_CMD_TIME_CLASS_B);
+
+	mlx4_free_cmd_mailbox(dev, mailbox);
+	return err;
+}
-- 
cgit v1.2.3-18-g5258


From c27a02cd94d6695fbe5858ca364b1e415af02212 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Wed, 22 Oct 2008 15:47:49 -0700
Subject: mlx4_en: Add driver for Mellanox ConnectX 10GbE NIC

The Mellanox ConnectX can operate as an InfiniBand adapter, as an
Ethernet NIC, or as a Fibre Channel (FC) HBA.  The kernel has a
low-level driver, mlx4_core, which handles multiplexing access to the
device, and there is also already an InfiniBad driver, mlx4_ib.

This patch adds a new driver, mlx4_en, which implements a standard
Ethernet NIC driver.

Signed-off-by: Liran Liss <liranl@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/net/Kconfig             |    9 +
 drivers/net/mlx4/Makefile       |    5 +
 drivers/net/mlx4/en_cq.c        |  146 ++++++
 drivers/net/mlx4/en_main.c      |  254 +++++++++
 drivers/net/mlx4/en_netdev.c    | 1088 +++++++++++++++++++++++++++++++++++++++
 drivers/net/mlx4/en_params.c    |  480 +++++++++++++++++
 drivers/net/mlx4/en_port.c      |  261 ++++++++++
 drivers/net/mlx4/en_port.h      |  570 ++++++++++++++++++++
 drivers/net/mlx4/en_resources.c |   96 ++++
 drivers/net/mlx4/en_rx.c        | 1080 ++++++++++++++++++++++++++++++++++++++
 drivers/net/mlx4/en_tx.c        |  820 +++++++++++++++++++++++++++++
 drivers/net/mlx4/mlx4_en.h      |  561 ++++++++++++++++++++
 12 files changed, 5370 insertions(+)
 create mode 100644 drivers/net/mlx4/en_cq.c
 create mode 100644 drivers/net/mlx4/en_main.c
 create mode 100644 drivers/net/mlx4/en_netdev.c
 create mode 100644 drivers/net/mlx4/en_params.c
 create mode 100644 drivers/net/mlx4/en_port.c
 create mode 100644 drivers/net/mlx4/en_port.h
 create mode 100644 drivers/net/mlx4/en_resources.c
 create mode 100644 drivers/net/mlx4/en_rx.c
 create mode 100644 drivers/net/mlx4/en_tx.c
 create mode 100644 drivers/net/mlx4/mlx4_en.h

(limited to 'drivers')

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 4a11296a951..69e2c10cb9f 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -2465,6 +2465,15 @@ config PASEMI_MAC
 	  This driver supports the on-chip 1/10Gbit Ethernet controller on
 	  PA Semi's PWRficient line of chips.
 
+config MLX4_EN
+	tristate "Mellanox Technologies 10Gbit Ethernet support"
+	depends on PCI && INET
+	select MLX4_CORE
+	select INET_LRO
+	help
+	  This driver supports Mellanox Technologies ConnectX Ethernet
+	  devices.
+
 config MLX4_CORE
 	tristate
 	depends on PCI
diff --git a/drivers/net/mlx4/Makefile b/drivers/net/mlx4/Makefile
index 9f493666e27..a7a97bf998f 100644
--- a/drivers/net/mlx4/Makefile
+++ b/drivers/net/mlx4/Makefile
@@ -2,3 +2,8 @@ obj-$(CONFIG_MLX4_CORE)		+= mlx4_core.o
 
 mlx4_core-y :=	alloc.o catas.o cmd.o cq.o eq.o fw.o icm.o intf.o main.o mcg.o \
 		mr.o pd.o port.o profile.o qp.o reset.o srq.o
+
+obj-$(CONFIG_MLX4_EN)               += mlx4_en.o
+
+mlx4_en-y := 	en_main.o en_tx.o en_rx.o en_params.o en_port.o en_cq.o \
+		en_resources.o en_netdev.o
diff --git a/drivers/net/mlx4/en_cq.c b/drivers/net/mlx4/en_cq.c
new file mode 100644
index 00000000000..1368a8010af
--- /dev/null
+++ b/drivers/net/mlx4/en_cq.c
@@ -0,0 +1,146 @@
+/*
+ * Copyright (c) 2007 Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+