From 0b7ca5a928e2271bbc225e9e1ac1f22e9fbee54f Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Tue, 22 Mar 2011 22:37:47 +0000
Subject: mlx4: Changing interrupt scheme

Adding a pool of MSI-X vectors and EQs that can be used explicitly by mlx4_core
customers (mlx4_ib, mlx4_en). The consumers will assign their own names to the
interrupt vectors. Those vectors are not opened at mlx4 device initialization,
opened by demand.
Changed the max number of possible EQs according to the new scheme, no longer relies on
on number of cores.
The new functionality is exposed through mlx4_assign_eq() and mlx4_release_eq().
Customers that do not use the new API will get completion vectors as before.

Signed-off-by: Markuze Alex <markuze@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/mlx4/device.h | 8 ++++++++
 1 file changed, 8 insertions(+)

(limited to 'include/linux/mlx4/device.h')

diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 049214642036..78380823d827 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -39,6 +39,11 @@
 
 #include <asm/atomic.h>
 
+#define MAX_MSIX_P_PORT		17
+#define MAX_MSIX		64
+#define MSIX_LEGACY_SZ		4
+#define MIN_MSIX_P_PORT		5
+
 enum {
 	MLX4_FLAG_MSI_X		= 1 << 0,
 	MLX4_FLAG_OLD_PORT_CMDS	= 1 << 1,
@@ -223,6 +228,7 @@ struct mlx4_caps {
 	int			num_eqs;
 	int			reserved_eqs;
 	int			num_comp_vectors;
+	int			comp_pool;
 	int			num_mpts;
 	int			num_mtt_segs;
 	int			mtts_per_seg;
@@ -526,5 +532,7 @@ void mlx4_fmr_unmap(struct mlx4_dev *dev, struct mlx4_fmr *fmr,
 int mlx4_fmr_free(struct mlx4_dev *dev, struct mlx4_fmr *fmr);
 int mlx4_SYNC_TPT(struct mlx4_dev *dev);
 int mlx4_test_interrupts(struct mlx4_dev *dev);
+int mlx4_assign_eq(struct mlx4_dev *dev, char* name , int* vector);
+void mlx4_release_eq(struct mlx4_dev *dev, int vec);
 
 #endif /* MLX4_DEVICE_H */
-- 
cgit v1.2.3


From 14c07b1358ede1664652bb9b28d9ace5fe6f7f92 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Tue, 22 Mar 2011 22:37:59 +0000
Subject: mlx4: Wake on LAN support

The driver queries the FW for WOL support.
Ethtool get/set_wol is implemented accordingly.
Only magic packets are supported at the time.

Signed-off-by: Igor Yarovinsky <igory@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/mlx4/en_ethtool.c | 62 +++++++++++++++++++++++++++++++++++++++++--
 drivers/net/mlx4/fw.c         | 20 ++++++++++++++
 drivers/net/mlx4/fw.h         |  1 +
 drivers/net/mlx4/main.c       |  1 +
 drivers/net/mlx4/mlx4_en.h    |  7 +++++
 include/linux/mlx4/device.h   |  4 +++
 6 files changed, 93 insertions(+), 2 deletions(-)

(limited to 'include/linux/mlx4/device.h')

diff --git a/drivers/net/mlx4/en_ethtool.c b/drivers/net/mlx4/en_ethtool.c
index 8cfe8586ed2d..c1f351f6ae57 100644
--- a/drivers/net/mlx4/en_ethtool.c
+++ b/drivers/net/mlx4/en_ethtool.c
@@ -131,8 +131,65 @@ static void mlx4_en_set_msglevel(struct net_device *dev, u32 val)
 static void mlx4_en_get_wol(struct net_device *netdev,
 			    struct ethtool_wolinfo *wol)
 {
-	wol->supported = 0;
-	wol->wolopts = 0;
+	struct mlx4_en_priv *priv = netdev_priv(netdev);
+	int err = 0;
+	u64 config = 0;
+
+	if (!priv->mdev->dev->caps.wol) {
+		wol->supported = 0;
+		wol->wolopts = 0;
+		return;
+	}
+
+	err = mlx4_wol_read(priv->mdev->dev, &config, priv->port);
+	if (err) {
+		en_err(priv, "Failed to get WoL information\n");
+		return;
+	}
+
+	if (config & MLX4_EN_WOL_MAGIC)
+		wol->supported = WAKE_MAGIC;
+	else
+		wol->supported = 0;
+
+	if (config & MLX4_EN_WOL_ENABLED)
+		wol->wolopts = WAKE_MAGIC;
+	else
+		wol->wolopts = 0;
+}
+
+static int mlx4_en_set_wol(struct net_device *netdev,
+			    struct ethtool_wolinfo *wol)
+{
+	struct mlx4_en_priv *priv = netdev_priv(netdev);
+	u64 config = 0;
+	int err = 0;
+
+	if (!priv->mdev->dev->caps.wol)
+		return -EOPNOTSUPP;
+
+	if (wol->supported & ~WAKE_MAGIC)
+		return -EINVAL;
+
+	err = mlx4_wol_read(priv->mdev->dev, &config, priv->port);
+	if (err) {
+		en_err(priv, "Failed to get WoL info, unable to modify\n");
+		return err;
+	}
+
+	if (wol->wolopts & WAKE_MAGIC) {
+		config |= MLX4_EN_WOL_DO_MODIFY | MLX4_EN_WOL_ENABLED |
+				MLX4_EN_WOL_MAGIC;
+	} else {
+		config &= ~(MLX4_EN_WOL_ENABLED | MLX4_EN_WOL_MAGIC);
+		config |= MLX4_EN_WOL_DO_MODIFY;
+	}
+
+	err = mlx4_wol_write(priv->mdev->dev, config, priv->port);
+	if (err)
+		en_err(priv, "Failed to set WoL information\n");
+
+	return err;
 }
 
 static int mlx4_en_get_sset_count(struct net_device *dev, int sset)
@@ -442,6 +499,7 @@ const struct ethtool_ops mlx4_en_ethtool_ops = {
 	.get_ethtool_stats = mlx4_en_get_ethtool_stats,
 	.self_test = mlx4_en_self_test,
 	.get_wol = mlx4_en_get_wol,
+	.set_wol = mlx4_en_set_wol,
 	.get_msglevel = mlx4_en_get_msglevel,
 	.set_msglevel = mlx4_en_set_msglevel,
 	.get_coalesce = mlx4_en_get_coalesce,
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index 5de1db897835..fd1c51b6197a 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -276,6 +276,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev_cap->udp_rss = field & 0x1;
 	MLX4_GET(field, outbox, QUERY_DEV_CAP_ETH_UC_LOOPBACK_OFFSET);
 	dev_cap->loopback_support = field & 0x1;
+	dev_cap->wol = field & 0x40;
 	MLX4_GET(dev_cap->flags, outbox, QUERY_DEV_CAP_FLAGS_OFFSET);
 	MLX4_GET(field, outbox, QUERY_DEV_CAP_RSVD_UAR_OFFSET);
 	dev_cap->reserved_uars = field >> 4;
@@ -908,3 +909,22 @@ int mlx4_NOP(struct mlx4_dev *dev)
 	/* Input modifier of 0x1f means "finish as soon as possible." */
 	return mlx4_cmd(dev, 0, 0x1f, 0, MLX4_CMD_NOP, 100);
 }
+
+#define MLX4_WOL_SETUP_MODE (5 << 28)
+int mlx4_wol_read(struct mlx4_dev *dev, u64 *config, int port)
+{
+	u32 in_mod = MLX4_WOL_SETUP_MODE | port << 8;
+
+	return mlx4_cmd_imm(dev, 0, config, in_mod, 0x3,
+			    MLX4_CMD_MOD_STAT_CFG, MLX4_CMD_TIME_CLASS_A);
+}
+EXPORT_SYMBOL_GPL(mlx4_wol_read);
+
+int mlx4_wol_write(struct mlx4_dev *dev, u64 config, int port)
+{
+	u32 in_mod = MLX4_WOL_SETUP_MODE | port << 8;
+
+	return mlx4_cmd(dev, config, in_mod, 0x1, MLX4_CMD_MOD_STAT_CFG,
+					MLX4_CMD_TIME_CLASS_A);
+}
+EXPORT_SYMBOL_GPL(mlx4_wol_write);
diff --git a/drivers/net/mlx4/fw.h b/drivers/net/mlx4/fw.h
index 65cc72eb899d..f7b9cc2d1b2a 100644
--- a/drivers/net/mlx4/fw.h
+++ b/drivers/net/mlx4/fw.h
@@ -80,6 +80,7 @@ struct mlx4_dev_cap {
 	u16 stat_rate_support;
 	int udp_rss;
 	int loopback_support;
+	int wol;
 	u32 flags;
 	int reserved_uars;
 	int uar_size;
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index 517ca34f5b37..42d4fb477870 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -227,6 +227,7 @@ 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.udp_rss	     = dev_cap->udp_rss;
 	dev->caps.loopback_support   = dev_cap->loopback_support;
+	dev->caps.wol		     = dev_cap->wol;
 	dev->caps.max_gso_sz	     = dev_cap->max_gso_sz;
 
 	dev->caps.log_num_macs  = log_num_mac;
diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h
index 2db245fcd84b..07aea8d0beeb 100644
--- a/drivers/net/mlx4/mlx4_en.h
+++ b/drivers/net/mlx4/mlx4_en.h
@@ -479,6 +479,13 @@ struct mlx4_en_priv {
 	int mc_addrs_cnt;
 	struct mlx4_en_stat_out_mbox hw_stats;
 	int vids[128];
+	bool wol;
+};
+
+enum mlx4_en_wol {
+	MLX4_EN_WOL_MAGIC = (1ULL << 61),
+	MLX4_EN_WOL_ENABLED = (1ULL << 62),
+	MLX4_EN_WOL_DO_MODIFY = (1ULL << 63),
 };
 
 
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 78380823d827..2460356d2c72 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -251,6 +251,7 @@ struct mlx4_caps {
 	u16			stat_rate_support;
 	int			udp_rss;
 	int			loopback_support;
+	int			wol;
 	u8			port_width_cap[MLX4_MAX_PORTS + 1];
 	int			max_gso_sz;
 	int                     reserved_qps_cnt[MLX4_NUM_QP_REGION];
@@ -535,4 +536,7 @@ int mlx4_test_interrupts(struct mlx4_dev *dev);
 int mlx4_assign_eq(struct mlx4_dev *dev, char* name , int* vector);
 void mlx4_release_eq(struct mlx4_dev *dev, int vec);
 
+int mlx4_wol_read(struct mlx4_dev *dev, u64 *config, int port);
+int mlx4_wol_write(struct mlx4_dev *dev, u64 config, int port);
+
 #endif /* MLX4_DEVICE_H */
-- 
cgit v1.2.3


From 725c89997e03d71b09ea3c17c997da0712b9d835 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Tue, 22 Mar 2011 22:38:07 +0000
Subject: mlx4_en: Reporting HW revision in ethtool -i

HW revision is derived from device ID and rev id.

Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/infiniband/hw/mlx4/main.c |  1 -
 drivers/net/mlx4/en_ethtool.c     | 15 ++++++++++++++-
 drivers/net/mlx4/main.c           |  2 ++
 drivers/net/mlx4/mlx4_en.h        |  3 +++
 include/linux/mlx4/device.h       |  2 +-
 5 files changed, 20 insertions(+), 3 deletions(-)

(limited to 'include/linux/mlx4/device.h')

diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index c7a6213c6996..66e3eecfd0db 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -721,7 +721,6 @@ static int init_node_data(struct mlx4_ib_dev *dev)
 	if (err)
 		goto out;
 
-	dev->dev->rev_id = be32_to_cpup((__be32 *) (out_mad->data + 32));
 	memcpy(&dev->ib_dev.node_guid, out_mad->data + 12, 8);
 
 out:
diff --git a/drivers/net/mlx4/en_ethtool.c b/drivers/net/mlx4/en_ethtool.c
index c1f351f6ae57..62ace6c72b42 100644
--- a/drivers/net/mlx4/en_ethtool.c
+++ b/drivers/net/mlx4/en_ethtool.c
@@ -45,7 +45,20 @@ mlx4_en_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *drvinfo)
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_en_dev *mdev = priv->mdev;
 
-	sprintf(drvinfo->driver, DRV_NAME " (%s)", mdev->dev->board_id);
+	switch (mdev->dev->rev_id) {
+	case 0xa0:
+		if (dev->dev_id >= MLX4_EN_CX3_LOW_ID && dev->dev_id <= MLX4_EN_CX3_HIGH_ID)
+			sprintf(drvinfo->driver, DRV_NAME " (%s_CX-3)", mdev->dev->board_id);
+		else
+			sprintf(drvinfo->driver, DRV_NAME " (%s_CX)", mdev->dev->board_id);
+		break;
+	case 0xb0:
+		sprintf(drvinfo->driver, DRV_NAME " (%s_CX-2)", mdev->dev->board_id);
+		break;
+	default:
+		sprintf(drvinfo->driver, DRV_NAME " (%s)", mdev->dev->board_id);
+		break;
+	}
 	strncpy(drvinfo->version, DRV_VERSION " (" DRV_RELDATE ")", 32);
 	sprintf(drvinfo->fw_version, "%d.%d.%d",
 		(u16) (mdev->dev->caps.fw_ver >> 32),
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index 42d4fb477870..5bebb8800ab2 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -1139,6 +1139,8 @@ static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	INIT_LIST_HEAD(&priv->pgdir_list);
 	mutex_init(&priv->pgdir_mutex);
 
+	pci_read_config_byte(pdev, PCI_REVISION_ID, &dev->rev_id);
+
 	/*
 	 * Now reset the HCA before we touch the PCI capabilities or
 	 * attempt a firmware command, since a boot ROM may have left
diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h
index 07aea8d0beeb..ad4df66750bc 100644
--- a/drivers/net/mlx4/mlx4_en.h
+++ b/drivers/net/mlx4/mlx4_en.h
@@ -216,6 +216,9 @@ struct mlx4_en_tx_desc {
 
 #define MLX4_EN_USE_SRQ		0x01000000
 
+#define MLX4_EN_CX3_LOW_ID	0x1000
+#define MLX4_EN_CX3_HIGH_ID	0x1005
+
 struct mlx4_en_rx_alloc {
 	struct page *page;
 	u16 offset;
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 2460356d2c72..fe2a3a3f046c 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -422,7 +422,7 @@ struct mlx4_dev {
 	unsigned long		flags;
 	struct mlx4_caps	caps;
 	struct radix_tree_root	qp_table_tree;
-	u32			rev_id;
+	u8			rev_id;
 	char			board_id[MLX4_BOARD_ID_LEN];
 };
 
-- 
cgit v1.2.3


From 0345584e0b8be3735a950d17c7e463db20c6ce27 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Tue, 22 Mar 2011 22:38:17 +0000
Subject: mlx4: generalization of multicast steering.

The same packet steering mechanism would be used both for IB and Ethernet,
Both multicasts and unicasts.
This commit prepares the general infrastructure for this.

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/infiniband/hw/mlx4/main.c |  10 ++--
 drivers/net/mlx4/en_main.c        |   2 +-
 drivers/net/mlx4/fw.c             |   2 +
 drivers/net/mlx4/fw.h             |   2 +
 drivers/net/mlx4/main.c           |   2 +
 drivers/net/mlx4/mcg.c            | 113 ++++++++++++++++++++++++++------------
 include/linux/mlx4/device.h       |  14 ++++-
 7 files changed, 102 insertions(+), 43 deletions(-)

(limited to 'include/linux/mlx4/device.h')

diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index 66e3eecfd0db..fbe1973f77b0 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -625,7 +625,7 @@ static int mlx4_ib_mcg_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
 
 	err = mlx4_multicast_attach(mdev->dev, &mqp->mqp, gid->raw,
 				    !!(mqp->flags & MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK),
-				    MLX4_PROTOCOL_IB);
+				    MLX4_PROT_IB_IPV6);
 	if (err)
 		return err;
 
@@ -636,7 +636,7 @@ static int mlx4_ib_mcg_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
 	return 0;
 
 err_add:
-	mlx4_multicast_detach(mdev->dev, &mqp->mqp, gid->raw, MLX4_PROTOCOL_IB);
+	mlx4_multicast_detach(mdev->dev, &mqp->mqp, gid->raw, MLX4_PROT_IB_IPV6);
 	return err;
 }
 
@@ -666,7 +666,7 @@ static int mlx4_ib_mcg_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
 	struct mlx4_ib_gid_entry *ge;
 
 	err = mlx4_multicast_detach(mdev->dev,
-				    &mqp->mqp, gid->raw, MLX4_PROTOCOL_IB);
+				    &mqp->mqp, gid->raw, MLX4_PROT_IB_IPV6);
 	if (err)
 		return err;
 
@@ -953,7 +953,7 @@ static int mlx4_ib_netdev_event(struct notifier_block *this, unsigned long event
 	mlx4_foreach_ib_transport_port(port, ibdev->dev) {
 		oldnd = iboe->netdevs[port - 1];
 		iboe->netdevs[port - 1] =
-			mlx4_get_protocol_dev(ibdev->dev, MLX4_PROTOCOL_EN, port);
+			mlx4_get_protocol_dev(ibdev->dev, MLX4_PROT_ETH, port);
 		if (oldnd != iboe->netdevs[port - 1]) {
 			if (iboe->netdevs[port - 1])
 				netdev_added(ibdev, port);
@@ -1206,7 +1206,7 @@ static struct mlx4_interface mlx4_ib_interface = {
 	.add		= mlx4_ib_add,
 	.remove		= mlx4_ib_remove,
 	.event		= mlx4_ib_event,
-	.protocol	= MLX4_PROTOCOL_IB
+	.protocol	= MLX4_PROT_IB_IPV6
 };
 
 static int __init mlx4_ib_init(void)
diff --git a/drivers/net/mlx4/en_main.c b/drivers/net/mlx4/en_main.c
index 29aaa4303991..9317b61a75b8 100644
--- a/drivers/net/mlx4/en_main.c
+++ b/drivers/net/mlx4/en_main.c
@@ -296,7 +296,7 @@ static struct mlx4_interface mlx4_en_interface = {
 	.remove		= mlx4_en_remove,
 	.event		= mlx4_en_event,
 	.get_dev	= mlx4_en_get_netdev,
-	.protocol	= MLX4_PROTOCOL_EN,
+	.protocol	= MLX4_PROT_ETH,
 };
 
 static int __init mlx4_en_init(void)
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index fd1c51b6197a..7813913b6d0f 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -274,6 +274,8 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev_cap->stat_rate_support = stat_rate;
 	MLX4_GET(field, outbox, QUERY_DEV_CAP_UDP_RSS_OFFSET);
 	dev_cap->udp_rss = field & 0x1;
+	dev_cap->vep_uc_steering = field & 0x2;
+	dev_cap->vep_mc_steering = field & 0x4;
 	MLX4_GET(field, outbox, QUERY_DEV_CAP_ETH_UC_LOOPBACK_OFFSET);
 	dev_cap->loopback_support = field & 0x1;
 	dev_cap->wol = field & 0x40;
diff --git a/drivers/net/mlx4/fw.h b/drivers/net/mlx4/fw.h
index f7b9cc2d1b2a..88003ebc6185 100644
--- a/drivers/net/mlx4/fw.h
+++ b/drivers/net/mlx4/fw.h
@@ -80,6 +80,8 @@ struct mlx4_dev_cap {
 	u16 stat_rate_support;
 	int udp_rss;
 	int loopback_support;
+	int vep_uc_steering;
+	int vep_mc_steering;
 	int wol;
 	u32 flags;
 	int reserved_uars;
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index 5bebb8800ab2..da0da281d830 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -227,6 +227,8 @@ 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.udp_rss	     = dev_cap->udp_rss;
 	dev->caps.loopback_support   = dev_cap->loopback_support;
+	dev->caps.vep_uc_steering    = dev_cap->vep_uc_steering;
+	dev->caps.vep_mc_steering    = dev_cap->vep_mc_steering;
 	dev->caps.wol		     = dev_cap->wol;
 	dev->caps.max_gso_sz	     = dev_cap->max_gso_sz;
 
diff --git a/drivers/net/mlx4/mcg.c b/drivers/net/mlx4/mcg.c
index 79cf42db2ea9..b103ad23bd84 100644
--- a/drivers/net/mlx4/mcg.c
+++ b/drivers/net/mlx4/mcg.c
@@ -32,6 +32,7 @@
  */
 
 #include <linux/string.h>
+#include <linux/etherdevice.h>
 
 #include <linux/mlx4/cmd.h>
 
@@ -50,28 +51,28 @@ struct mlx4_mgm {
 
 static const u8 zero_gid[16];	/* automatically initialized to 0 */
 
-static int mlx4_READ_MCG(struct mlx4_dev *dev, int index,
-			 struct mlx4_cmd_mailbox *mailbox)
+static int mlx4_READ_ENTRY(struct mlx4_dev *dev, int index,
+			   struct mlx4_cmd_mailbox *mailbox)
 {
 	return mlx4_cmd_box(dev, 0, mailbox->dma, index, 0, MLX4_CMD_READ_MCG,
 			    MLX4_CMD_TIME_CLASS_A);
 }
 
-static int mlx4_WRITE_MCG(struct mlx4_dev *dev, int index,
-			  struct mlx4_cmd_mailbox *mailbox)
+static int mlx4_WRITE_ENTRY(struct mlx4_dev *dev, int index,
+			    struct mlx4_cmd_mailbox *mailbox)
 {
 	return mlx4_cmd(dev, mailbox->dma, index, 0, MLX4_CMD_WRITE_MCG,
 			MLX4_CMD_TIME_CLASS_A);
 }
 
-static int mlx4_MGID_HASH(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox,
-			  u16 *hash)
+static int mlx4_GID_HASH(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox,
+			 u16 *hash, u8 op_mod)
 {
 	u64 imm;
 	int err;
 
-	err = mlx4_cmd_imm(dev, mailbox->dma, &imm, 0, 0, MLX4_CMD_MGID_HASH,
-			   MLX4_CMD_TIME_CLASS_A);
+	err = mlx4_cmd_imm(dev, mailbox->dma, &imm, 0, op_mod,
+			   MLX4_CMD_MGID_HASH, MLX4_CMD_TIME_CLASS_A);
 
 	if (!err)
 		*hash = imm;
@@ -94,15 +95,17 @@ static int mlx4_MGID_HASH(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox
  * If no AMGM exists for given gid, *index = -1, *prev = index of last
  * entry in hash chain and *mgm holds end of hash chain.
  */
-static int find_mgm(struct mlx4_dev *dev,
-		    u8 *gid, enum mlx4_protocol protocol,
-		    struct mlx4_cmd_mailbox *mgm_mailbox,
-		    u16 *hash, int *prev, int *index)
+static int find_entry(struct mlx4_dev *dev, u8 port,
+		      u8 *gid, enum mlx4_protocol prot,
+		      enum mlx4_steer_type steer,
+		      struct mlx4_cmd_mailbox *mgm_mailbox,
+		      u16 *hash, int *prev, int *index)
 {
 	struct mlx4_cmd_mailbox *mailbox;
 	struct mlx4_mgm *mgm = mgm_mailbox->buf;
 	u8 *mgid;
 	int err;
+	u8 op_mod = (prot == MLX4_PROT_ETH) ? !!(dev->caps.vep_mc_steering) : 0;
 
 	mailbox = mlx4_alloc_cmd_mailbox(dev);
 	if (IS_ERR(mailbox))
@@ -111,7 +114,7 @@ static int find_mgm(struct mlx4_dev *dev,
 
 	memcpy(mgid, gid, 16);
 
-	err = mlx4_MGID_HASH(dev, mailbox, hash);
+	err = mlx4_GID_HASH(dev, mailbox, hash, op_mod);
 	mlx4_free_cmd_mailbox(dev, mailbox);
 	if (err)
 		return err;
@@ -123,11 +126,11 @@ static int find_mgm(struct mlx4_dev *dev,
 	*prev  = -1;
 
 	do {
-		err = mlx4_READ_MCG(dev, *index, mgm_mailbox);
+		err = mlx4_READ_ENTRY(dev, *index, mgm_mailbox);
 		if (err)
 			return err;
 
-		if (!memcmp(mgm->gid, zero_gid, 16)) {
+		if (!(be32_to_cpu(mgm->members_count) & 0xffffff)) {
 			if (*index != *hash) {
 				mlx4_err(dev, "Found zero MGID in AMGM.\n");
 				err = -EINVAL;
@@ -136,7 +139,7 @@ static int find_mgm(struct mlx4_dev *dev,
 		}
 
 		if (!memcmp(mgm->gid, gid, 16) &&
-		    be32_to_cpu(mgm->members_count) >> 30 == protocol)
+		    be32_to_cpu(mgm->members_count) >> 30 == prot)
 			return err;
 
 		*prev = *index;
@@ -147,8 +150,9 @@ static int find_mgm(struct mlx4_dev *dev,
 	return err;
 }
 
-int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
-			  int block_mcast_loopback, enum mlx4_protocol protocol)
+int mlx4_qp_attach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
+			  int block_mcast_loopback, enum mlx4_protocol prot,
+			  enum mlx4_steer_type steer)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct mlx4_cmd_mailbox *mailbox;
@@ -159,6 +163,7 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 	int link = 0;
 	int i;
 	int err;
+	u8 port = gid[5];
 
 	mailbox = mlx4_alloc_cmd_mailbox(dev);
 	if (IS_ERR(mailbox))
@@ -166,13 +171,13 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 	mgm = mailbox->buf;
 
 	mutex_lock(&priv->mcg_table.mutex);
-
-	err = find_mgm(dev, gid, protocol, mailbox, &hash, &prev, &index);
+	err = find_entry(dev, port, gid, prot, steer,
+			 mailbox, &hash, &prev, &index);
 	if (err)
 		goto out;
 
 	if (index != -1) {
-		if (!memcmp(mgm->gid, zero_gid, 16))
+		if (!(be32_to_cpu(mgm->members_count) & 0xffffff))
 			memcpy(mgm->gid, gid, 16);
 	} else {
 		link = 1;
@@ -209,22 +214,22 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 	else
 		mgm->qp[members_count++] = cpu_to_be32(qp->qpn & MGM_QPN_MASK);
 
-	mgm->members_count = cpu_to_be32(members_count | (u32) protocol << 30);
+	mgm->members_count = cpu_to_be32(members_count | (u32) prot << 30);
 
-	err = mlx4_WRITE_MCG(dev, index, mailbox);
+	err = mlx4_WRITE_ENTRY(dev, index, mailbox);
 	if (err)
 		goto out;
 
 	if (!link)
 		goto out;
 
-	err = mlx4_READ_MCG(dev, prev, mailbox);
+	err = mlx4_READ_ENTRY(dev, prev, mailbox);
 	if (err)
 		goto out;
 
 	mgm->next_gid_index = cpu_to_be32(index << 6);
 
-	err = mlx4_WRITE_MCG(dev, prev, mailbox);
+	err = mlx4_WRITE_ENTRY(dev, prev, mailbox);
 	if (err)
 		goto out;
 
@@ -242,10 +247,9 @@ out:
 	mlx4_free_cmd_mailbox(dev, mailbox);
 	return err;
 }
-EXPORT_SYMBOL_GPL(mlx4_multicast_attach);
 
-int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
-			  enum mlx4_protocol protocol)
+int mlx4_qp_detach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
+			  enum mlx4_protocol prot, enum mlx4_steer_type steer)
 {
 	struct mlx4_priv *priv = mlx4_priv(dev);
 	struct mlx4_cmd_mailbox *mailbox;
@@ -255,6 +259,7 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 	int prev, index;
 	int i, loc;
 	int err;
+	u8 port = gid[5];
 
 	mailbox = mlx4_alloc_cmd_mailbox(dev);
 	if (IS_ERR(mailbox))
@@ -263,7 +268,8 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 
 	mutex_lock(&priv->mcg_table.mutex);
 
-	err = find_mgm(dev, gid, protocol, mailbox, &hash, &prev, &index);
+	err = find_entry(dev, port, gid, prot, steer,
+			 mailbox, &hash, &prev, &index);
 	if (err)
 		goto out;
 
@@ -285,12 +291,12 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 	}
 
 
-	mgm->members_count = cpu_to_be32(--members_count | (u32) protocol << 30);
+	mgm->members_count = cpu_to_be32(--members_count | (u32) prot << 30);
 	mgm->qp[loc]       = mgm->qp[i - 1];
 	mgm->qp[i - 1]     = 0;
 
 	if (i != 1) {
-		err = mlx4_WRITE_MCG(dev, index, mailbox);
+		err = mlx4_WRITE_ENTRY(dev, index, mailbox);
 		goto out;
 	}
 
@@ -298,13 +304,13 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 		/* Remove entry from MGM */
 		int amgm_index = be32_to_cpu(mgm->next_gid_index) >> 6;
 		if (amgm_index) {
-			err = mlx4_READ_MCG(dev, amgm_index, mailbox);
+			err = mlx4_READ_ENTRY(dev, amgm_index, mailbox);
 			if (err)
 				goto out;
 		} else
 			memset(mgm->gid, 0, 16);
 
-		err = mlx4_WRITE_MCG(dev, index, mailbox);
+		err = mlx4_WRITE_ENTRY(dev, index, mailbox);
 		if (err)
 			goto out;
 
@@ -319,13 +325,13 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 	} else {
 		/* Remove entry from AMGM */
 		int cur_next_index = be32_to_cpu(mgm->next_gid_index) >> 6;
-		err = mlx4_READ_MCG(dev, prev, mailbox);
+		err = mlx4_READ_ENTRY(dev, prev, mailbox);
 		if (err)
 			goto out;
 
 		mgm->next_gid_index = cpu_to_be32(cur_next_index << 6);
 
-		err = mlx4_WRITE_MCG(dev, prev, mailbox);
+		err = mlx4_WRITE_ENTRY(dev, prev, mailbox);
 		if (err)
 			goto out;
 
@@ -343,6 +349,43 @@ out:
 	mlx4_free_cmd_mailbox(dev, mailbox);
 	return err;
 }
+
+
+int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
+			  int block_mcast_loopback, enum mlx4_protocol prot)
+{
+	enum mlx4_steer_type steer;
+
+	steer = (is_valid_ether_addr(&gid[10])) ? MLX4_UC_STEER : MLX4_MC_STEER;
+
+	if (prot == MLX4_PROT_ETH && !dev->caps.vep_mc_steering)
+		return 0;
+
+	if (prot == MLX4_PROT_ETH)
+		gid[7] |= (steer << 1);
+
+	return mlx4_qp_attach_common(dev, qp, gid,
+				     block_mcast_loopback, prot,
+				     steer);
+}
+EXPORT_SYMBOL_GPL(mlx4_multicast_attach);
+
+int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
+			  enum mlx4_protocol prot)
+{
+	enum mlx4_steer_type steer;
+
+	steer = (is_valid_ether_addr(&gid[10])) ? MLX4_UC_STEER : MLX4_MC_STEER;
+
+	if (prot == MLX4_PROT_ETH && !dev->caps.vep_mc_steering)
+		return 0;
+
+	if (prot == MLX4_PROT_ETH) {
+		gid[7] |= (steer << 1);
+	}
+
+	return mlx4_qp_detach_common(dev, qp, gid, prot, steer);
+}
 EXPORT_SYMBOL_GPL(mlx4_multicast_detach);
 
 int mlx4_init_mcg_table(struct mlx4_dev *dev)
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index fe2a3a3f046c..ebda939b9fc3 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -150,8 +150,10 @@ enum {
 };
 
 enum mlx4_protocol {
-	MLX4_PROTOCOL_IB,
-	MLX4_PROTOCOL_EN,
+	MLX4_PROT_IB_IPV6 = 0,
+	MLX4_PROT_ETH,
+	MLX4_PROT_IB_IPV4,
+	MLX4_PROT_FCOE
 };
 
 enum {
@@ -178,6 +180,12 @@ enum mlx4_special_vlan_idx {
 	MLX4_VLAN_REGULAR
 };
 
+enum mlx4_steer_type {
+	MLX4_MC_STEER = 0,
+	MLX4_UC_STEER,
+	MLX4_NUM_STEERS
+};
+
 enum {
 	MLX4_NUM_FEXCH          = 64 * 1024,
 };
@@ -251,6 +259,8 @@ struct mlx4_caps {
 	u16			stat_rate_support;
 	int			udp_rss;
 	int			loopback_support;
+	int			vep_uc_steering;
+	int			vep_mc_steering;
 	int			wol;
 	u8			port_width_cap[MLX4_MAX_PORTS + 1];
 	int			max_gso_sz;
-- 
cgit v1.2.3


From 1679200f91da6a054b06954c9bd3eeed29b6731f Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Tue, 22 Mar 2011 22:38:31 +0000
Subject: mlx4_en: Enabling new steering

The mlx4_en module now uses the new steering mechanism.
The RX packets are now steered through the MCG table instead
of Mac table for unicast, and default entry for multicast.
The feature is enabled through INIT_HCA

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/mlx4/en_netdev.c | 133 +++++++++++++++++++++++++++-------
 drivers/net/mlx4/en_port.c   |  11 ++-
 drivers/net/mlx4/en_port.h   |  19 +++--
 drivers/net/mlx4/en_rx.c     |  11 +--
 drivers/net/mlx4/fw.c        |   3 +
 drivers/net/mlx4/mlx4.h      |   6 ++
 drivers/net/mlx4/mlx4_en.h   |   1 +
 drivers/net/mlx4/port.c      | 165 +++++++++++++++++++++++++++++++++++++++----
 include/linux/mlx4/device.h  |  12 +++-
 9 files changed, 300 insertions(+), 61 deletions(-)

(limited to 'include/linux/mlx4/device.h')

diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c
index f6ed315b4b89..08e680100db8 100644
--- a/drivers/net/mlx4/en_netdev.c
+++ b/drivers/net/mlx4/en_netdev.c
@@ -156,9 +156,8 @@ static void mlx4_en_do_set_mac(struct work_struct *work)
 	mutex_lock(&mdev->state_lock);
 	if (priv->port_up) {
 		/* Remove old MAC and insert the new one */
-		mlx4_unregister_mac(mdev->dev, priv->port, priv->mac_index);
-		err = mlx4_register_mac(mdev->dev, priv->port,
-					priv->mac, &priv->mac_index);
+		err = mlx4_replace_mac(mdev->dev, priv->port,
+				       priv->base_qpn, priv->mac, 0);
 		if (err)
 			en_err(priv, "Failed changing HW MAC address\n");
 	} else
@@ -214,6 +213,7 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
 	struct mlx4_en_dev *mdev = priv->mdev;
 	struct net_device *dev = priv->dev;
 	u64 mcast_addr = 0;
+	u8 mc_list[16] = {0};
 	int err;
 
 	mutex_lock(&mdev->state_lock);
@@ -239,8 +239,12 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
 			priv->flags |= MLX4_EN_FLAG_PROMISC;
 
 			/* Enable promiscouos mode */
-			err = mlx4_SET_PORT_qpn_calc(mdev->dev, priv->port,
-						     priv->base_qpn, 1);
+			if (!mdev->dev->caps.vep_uc_steering)
+				err = mlx4_SET_PORT_qpn_calc(mdev->dev, priv->port,
+							     priv->base_qpn, 1);
+			else
+				err = mlx4_unicast_promisc_add(mdev->dev, priv->base_qpn,
+							       priv->port);
 			if (err)
 				en_err(priv, "Failed enabling "
 					     "promiscous mode\n");
@@ -252,10 +256,21 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
 				en_err(priv, "Failed disabling "
 					     "multicast filter\n");
 
-			/* Disable port VLAN filter */
-			err = mlx4_SET_VLAN_FLTR(mdev->dev, priv->port, NULL);
-			if (err)
-				en_err(priv, "Failed disabling VLAN filter\n");
+			/* Add the default qp number as multicast promisc */
+			if (!(priv->flags & MLX4_EN_FLAG_MC_PROMISC)) {
+				err = mlx4_multicast_promisc_add(mdev->dev, priv->base_qpn,
+								 priv->port);
+				if (err)
+					en_err(priv, "Failed entering multicast promisc mode\n");
+				priv->flags |= MLX4_EN_FLAG_MC_PROMISC;
+			}
+
+			if (priv->vlgrp) {
+				/* Disable port VLAN filter */
+				err = mlx4_SET_VLAN_FLTR(mdev->dev, priv->port, NULL);
+				if (err)
+					en_err(priv, "Failed disabling VLAN filter\n");
+			}
 		}
 		goto out;
 	}
@@ -270,11 +285,24 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
 		priv->flags &= ~MLX4_EN_FLAG_PROMISC;
 
 		/* Disable promiscouos mode */
-		err = mlx4_SET_PORT_qpn_calc(mdev->dev, priv->port,
-					     priv->base_qpn, 0);
+		if (!mdev->dev->caps.vep_uc_steering)
+			err = mlx4_SET_PORT_qpn_calc(mdev->dev, priv->port,
+						     priv->base_qpn, 0);
+		else
+			err = mlx4_unicast_promisc_remove(mdev->dev, priv->base_qpn,
+							  priv->port);
 		if (err)
 			en_err(priv, "Failed disabling promiscous mode\n");
 
+		/* Disable Multicast promisc */
+		if (priv->flags & MLX4_EN_FLAG_MC_PROMISC) {
+			err = mlx4_multicast_promisc_remove(mdev->dev, priv->base_qpn,
+							    priv->port);
+			if (err)
+				en_err(priv, "Failed disabling multicast promiscous mode\n");
+			priv->flags &= ~MLX4_EN_FLAG_MC_PROMISC;
+		}
+
 		/* Enable port VLAN filter */
 		err = mlx4_SET_VLAN_FLTR(mdev->dev, priv->port, priv->vlgrp);
 		if (err)
@@ -287,14 +315,38 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
 					  0, MLX4_MCAST_DISABLE);
 		if (err)
 			en_err(priv, "Failed disabling multicast filter\n");
+
+		/* Add the default qp number as multicast promisc */
+		if (!(priv->flags & MLX4_EN_FLAG_MC_PROMISC)) {
+			err = mlx4_multicast_promisc_add(mdev->dev, priv->base_qpn,
+							 priv->port);
+			if (err)
+				en_err(priv, "Failed entering multicast promisc mode\n");
+			priv->flags |= MLX4_EN_FLAG_MC_PROMISC;
+		}
 	} else {
 		int i;
+		/* Disable Multicast promisc */
+		if (priv->flags & MLX4_EN_FLAG_MC_PROMISC) {
+			err = mlx4_multicast_promisc_remove(mdev->dev, priv->base_qpn,
+							    priv->port);
+			if (err)
+				en_err(priv, "Failed disabling multicast promiscous mode\n");
+			priv->flags &= ~MLX4_EN_FLAG_MC_PROMISC;
+		}
 
 		err = mlx4_SET_MCAST_FLTR(mdev->dev, priv->port, 0,
 					  0, MLX4_MCAST_DISABLE);
 		if (err)
 			en_err(priv, "Failed disabling multicast filter\n");
 
+		/* Detach our qp from all the multicast addresses */
+		for (i = 0; i < priv->mc_addrs_cnt; i++) {
+			memcpy(&mc_list[10], priv->mc_addrs + i * ETH_ALEN, ETH_ALEN);
+			mc_list[5] = priv->port;
+			mlx4_multicast_detach(mdev->dev, &priv->rss_map.indir_qp,
+					      mc_list, MLX4_PROT_ETH);
+		}
 		/* Flush mcast filter and init it with broadcast address */
 		mlx4_SET_MCAST_FLTR(mdev->dev, priv->port, ETH_BCAST,
 				    1, MLX4_MCAST_CONFIG);
@@ -307,6 +359,10 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
 		for (i = 0; i < priv->mc_addrs_cnt; i++) {
 			mcast_addr =
 			      mlx4_en_mac_to_u64(priv->mc_addrs + i * ETH_ALEN);
+			memcpy(&mc_list[10], priv->mc_addrs + i * ETH_ALEN, ETH_ALEN);
+			mc_list[5] = priv->port;
+			mlx4_multicast_attach(mdev->dev, &priv->rss_map.indir_qp,
+					      mc_list, 0, MLX4_PROT_ETH);
 			mlx4_SET_MCAST_FLTR(mdev->dev, priv->port,
 					    mcast_addr, 0, MLX4_MCAST_CONFIG);
 		}
@@ -314,8 +370,6 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
 					  0, MLX4_MCAST_ENABLE);
 		if (err)
 			en_err(priv, "Failed enabling multicast filter\n");
-
-		mlx4_en_clear_list(dev);
 	}
 out:
 	mutex_unlock(&mdev->state_lock);
@@ -557,6 +611,7 @@ int mlx4_en_start_port(struct net_device *dev)
 	int err = 0;
 	int i;
 	int j;
+	u8 mc_list[16] = {0};
 	char name[32];
 
 	if (priv->port_up) {
@@ -596,10 +651,20 @@ int mlx4_en_start_port(struct net_device *dev)
 		++rx_index;
 	}
 
+	/* Set port mac number */
+	en_dbg(DRV, priv, "Setting mac for port %d\n", priv->port);
+	err = mlx4_register_mac(mdev->dev, priv->port,
+				priv->mac, &priv->base_qpn, 0);
+	if (err) {
+		en_err(priv, "Failed setting port mac\n");
+		goto cq_err;
+	}
+	mdev->mac_removed[priv->port] = 0;
+
 	err = mlx4_en_config_rss_steer(priv);
 	if (err) {
 		en_err(priv, "Failed configuring rss steering\n");
-		goto cq_err;
+		goto mac_err;
 	}
 
 	if (mdev->dev->caps.comp_pool && !priv->tx_vector) {
@@ -661,24 +726,22 @@ int mlx4_en_start_port(struct net_device *dev)
 		en_err(priv, "Failed setting default qp numbers\n");
 		goto tx_err;
 	}
-	/* Set port mac number */
-	en_dbg(DRV, priv, "Setting mac for port %d\n", priv->port);
-	err = mlx4_register_mac(mdev->dev, priv->port,
-				priv->mac, &priv->mac_index);
-	if (err) {
-		en_err(priv, "Failed setting port mac\n");
-		goto tx_err;
-	}
-	mdev->mac_removed[priv->port] = 0;
 
 	/* Init port */
 	en_dbg(HW, priv, "Initializing port\n");
 	err = mlx4_INIT_PORT(mdev->dev, priv->port);
 	if (err) {
 		en_err(priv, "Failed Initializing port\n");
-		goto mac_err;
+		goto tx_err;
 	}
 
+	/* Attach rx QP to bradcast address */
+	memset(&mc_list[10], 0xff, ETH_ALEN);
+	mc_list[5] = priv->port;
+	if (mlx4_multicast_attach(mdev->dev, &priv->rss_map.indir_qp, mc_list,
+				  0, MLX4_PROT_ETH))
+		mlx4_warn(mdev, "Failed Attaching Broadcast\n");
+
 	/* Schedule multicast task to populate multicast list */
 	queue_work(mdev->workqueue, &priv->mcast_task);
 
@@ -686,8 +749,6 @@ int mlx4_en_start_port(struct net_device *dev)
 	netif_tx_start_all_queues(dev);
 	return 0;
 
-mac_err:
-	mlx4_unregister_mac(mdev->dev, priv->port, priv->mac_index);
 tx_err:
 	while (tx_index--) {
 		mlx4_en_deactivate_tx_ring(priv, &priv->tx_ring[tx_index]);
@@ -695,6 +756,8 @@ tx_err:
 	}
 
 	mlx4_en_release_rss_steer(priv);
+mac_err:
+	mlx4_unregister_mac(mdev->dev, priv->port, priv->base_qpn);
 cq_err:
 	while (rx_index--)
 		mlx4_en_deactivate_cq(priv, &priv->rx_cq[rx_index]);
@@ -710,6 +773,7 @@ void mlx4_en_stop_port(struct net_device *dev)
 	struct mlx4_en_priv *priv = netdev_priv(dev);
 	struct mlx4_en_dev *mdev = priv->mdev;
 	int i;
+	u8 mc_list[16] = {0};
 
 	if (!priv->port_up) {
 		en_dbg(DRV, priv, "stop port called while port already down\n");
@@ -724,8 +788,23 @@ void mlx4_en_stop_port(struct net_device *dev)
 	/* Set port as not active */
 	priv->port_up = false;
 
+	/* Detach All multicasts */
+	memset(&mc_list[10], 0xff, ETH_ALEN);
+	mc_list[5] = priv->port;
+	mlx4_multicast_detach(mdev->dev, &priv->rss_map.indir_qp, mc_list,
+			      MLX4_PROT_ETH);
+	for (i = 0; i < priv->mc_addrs_cnt; i++) {
+		memcpy(&mc_list[10], priv->mc_addrs + i * ETH_ALEN, ETH_ALEN);
+		mc_list[5] = priv->port;
+		mlx4_multicast_detach(mdev->dev, &priv->rss_map.indir_qp,
+				      mc_list, MLX4_PROT_ETH);
+	}
+	mlx4_en_clear_list(dev);
+	/* Flush multicast filter */
+	mlx4_SET_MCAST_FLTR(mdev->dev, priv->port, 0, 1, MLX4_MCAST_CONFIG);
+
 	/* Unregister Mac address for the port */
-	mlx4_unregister_mac(mdev->dev, priv->port, priv->mac_index);
+	mlx4_unregister_mac(mdev->dev, priv->port, priv->base_qpn);
 	mdev->mac_removed[priv->port] = 1;
 
 	/* Free TX Rings */
diff --git a/drivers/net/mlx4/en_port.c b/drivers/net/mlx4/en_port.c
index a2b0bc4ada5d..f2a4f5dd313d 100644
--- a/drivers/net/mlx4/en_port.c
+++ b/drivers/net/mlx4/en_port.c
@@ -119,6 +119,10 @@ int mlx4_SET_PORT_qpn_calc(struct mlx4_dev *dev, u8 port, u32 base_qpn,
 	struct mlx4_set_port_rqp_calc_context *context;
 	int err;
 	u32 in_mod;
+	u32 m_promisc = (dev->caps.vep_mc_steering) ? MCAST_DIRECT : MCAST_DEFAULT;
+
+	if (dev->caps.vep_mc_steering && dev->caps.vep_uc_steering)
+		return 0;
 
 	mailbox = mlx4_alloc_cmd_mailbox(dev);
 	if (IS_ERR(mailbox))
@@ -127,8 +131,11 @@ int mlx4_SET_PORT_qpn_calc(struct mlx4_dev *dev, u8 port, u32 base_qpn,
 	memset(context, 0, sizeof *context);
 
 	context->base_qpn = cpu_to_be32(base_qpn);
-	context->promisc = cpu_to_be32(promisc << SET_PORT_PROMISC_EN_SHIFT | base_qpn);
-	context->mcast = cpu_to_be32(1 << SET_PORT_PROMISC_MODE_SHIFT | base_qpn);
+	context->n_mac = 0x7;
+	context->promisc = cpu_to_be32(promisc << SET_PORT_PROMISC_SHIFT |
+				       base_qpn);
+	context->mcast = cpu_to_be32(m_promisc << SET_PORT_MC_PROMISC_SHIFT |
+				     base_qpn);
 	context->intra_no_vlan = 0;
 	context->no_vlan = MLX4_NO_VLAN_IDX;
 	context->intra_vlan_miss = 0;
diff --git a/drivers/net/mlx4/en_port.h b/drivers/net/mlx4/en_port.h
index 092e814b1981..e3d73e41c567 100644
--- a/drivers/net/mlx4/en_port.h
+++ b/drivers/net/mlx4/en_port.h
@@ -36,8 +36,8 @@
 
 
 #define SET_PORT_GEN_ALL_VALID	0x7
-#define SET_PORT_PROMISC_EN_SHIFT	31
-#define SET_PORT_PROMISC_MODE_SHIFT	30
+#define SET_PORT_PROMISC_SHIFT	31
+#define SET_PORT_MC_PROMISC_SHIFT	30
 
 enum {
 	MLX4_CMD_SET_VLAN_FLTR  = 0x47,
@@ -45,6 +45,12 @@ enum {
 	MLX4_CMD_DUMP_ETH_STATS = 0x49,
 };
 
+enum {
+	MCAST_DIRECT_ONLY       = 0,
+	MCAST_DIRECT            = 1,
+	MCAST_DEFAULT           = 2
+};
+
 struct mlx4_set_port_general_context {
 	u8 reserved[3];
 	u8 flags;
@@ -60,14 +66,17 @@ struct mlx4_set_port_general_context {
 
 struct mlx4_set_port_rqp_calc_context {
 	__be32 base_qpn;
-	__be32 flags;
-	u8 reserved[3];
+	u8 rererved;
+	u8 n_mac;
+	u8 n_vlan;
+	u8 n_prio;
+	u8 reserved2[3];
 	u8 mac_miss;
 	u8 intra_no_vlan;
 	u8 no_vlan;
 	u8 intra_vlan_miss;
 	u8 vlan_miss;
-	u8 reserved2[3];
+	u8 reserved3[3];
 	u8 no_vlan_prio;
 	__be32 promisc;
 	__be32 mcast;
diff --git a/drivers/net/mlx4/en_rx.c b/drivers/net/mlx4/en_rx.c
index 570f2508fb30..05998ee297c9 100644
--- a/drivers/net/mlx4/en_rx.c
+++ b/drivers/net/mlx4/en_rx.c
@@ -845,16 +845,10 @@ int mlx4_en_config_rss_steer(struct mlx4_en_priv *priv)
 	}
 
 	/* Configure RSS indirection qp */
-	err = mlx4_qp_reserve_range(mdev->dev, 1, 1, &priv->base_qpn);
-	if (err) {
-		en_err(priv, "Failed to reserve range for RSS "
-			     "indirection qp\n");
-		goto rss_err;
-	}
 	err = mlx4_qp_alloc(mdev->dev, priv->base_qpn, &rss_map->indir_qp);
 	if (err) {
 		en_err(priv, "Failed to allocate RSS indirection QP\n");
-		goto reserve_err;
+		goto rss_err;
 	}
 	rss_map->indir_qp.event = mlx4_en_sqp_event;
 	mlx4_en_fill_qp_context(priv, 0, 0, 0, 1, priv->base_qpn,
@@ -881,8 +875,6 @@ indir_err:
 		       MLX4_QP_STATE_RST, NULL, 0, 0, &rss_map->indir_qp);
 	mlx4_qp_remove(mdev->dev, &rss_map->indir_qp);
 	mlx4_qp_free(mdev->dev, &rss_map->indir_qp);
-reserve_err:
-	mlx4_qp_release_range(mdev->dev, priv->base_qpn, 1);
 rss_err:
 	for (i = 0; i < good_qps; i++) {
 		mlx4_qp_modify(mdev->dev, NULL, rss_map->state[i],
@@ -904,7 +896,6 @@ void mlx4_en_release_rss_steer(struct mlx4_en_priv *priv)
 		       MLX4_QP_STATE_RST, NULL, 0, 0, &rss_map->indir_qp);
 	mlx4_qp_remove(mdev->dev, &rss_map->indir_qp);
 	mlx4_qp_free(mdev->dev, &rss_map->indir_qp);
-	mlx4_qp_release_range(mdev->dev, priv->base_qpn, 1);
 
 	for (i = 0; i < priv->rx_ring_num; i++) {
 		mlx4_qp_modify(mdev->dev, NULL, rss_map->state[i],
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index 7813913b6d0f..67a209ba939d 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -740,6 +740,7 @@ int mlx4_INIT_HCA(struct mlx4_dev *dev, struct mlx4_init_hca_param *param)
 #define	 INIT_HCA_MC_BASE_OFFSET	 (INIT_HCA_MCAST_OFFSET + 0x00)
 #define	 INIT_HCA_LOG_MC_ENTRY_SZ_OFFSET (INIT_HCA_MCAST_OFFSET + 0x12)
 #define	 INIT_HCA_LOG_MC_HASH_SZ_OFFSET	 (INIT_HCA_MCAST_OFFSET + 0x16)
+#define  INIT_HCA_UC_STEERING_OFFSET	 (INIT_HCA_MCAST_OFFSET + 0x18)
 #define	 INIT_HCA_LOG_MC_TABLE_SZ_OFFSET (INIT_HCA_MCAST_OFFSET + 0x1b)
 #define INIT_HCA_TPT_OFFSET		 0x0f0
 #define	 INIT_HCA_DMPT_BASE_OFFSET	 (INIT_HCA_TPT_OFFSET + 0x00)
@@ -800,6 +801,8 @@ int mlx4_INIT_HCA(struct mlx4_dev *dev, struct mlx4_init_hca_param *param)
 	MLX4_PUT(inbox, param->mc_base,		INIT_HCA_MC_BASE_OFFSET);
 	MLX4_PUT(inbox, param->log_mc_entry_sz, INIT_HCA_LOG_MC_ENTRY_SZ_OFFSET);
 	MLX4_PUT(inbox, param->log_mc_hash_sz,  INIT_HCA_LOG_MC_HASH_SZ_OFFSET);
+	if (dev->caps.vep_mc_steering)
+		MLX4_PUT(inbox, (u8) (1 << 3),	INIT_HCA_UC_STEERING_OFFSET);
 	MLX4_PUT(inbox, param->log_mc_table_sz, INIT_HCA_LOG_MC_TABLE_SZ_OFFSET);
 
 	/* TPT attributes */
diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h
index 1513a91b4bd4..d8bb4418581b 100644
--- a/drivers/net/mlx4/mlx4.h
+++ b/drivers/net/mlx4/mlx4.h
@@ -286,6 +286,10 @@ struct mlx4_vlan_table {
 	int			max;
 };
 
+struct mlx4_mac_entry {
+	u64 mac;
+};
+
 struct mlx4_port_info {
 	struct mlx4_dev	       *dev;
 	int			port;
@@ -293,7 +297,9 @@ struct mlx4_port_info {
 	struct device_attribute port_attr;
 	enum mlx4_port_type	tmp_type;
 	struct mlx4_mac_table	mac_table;
+	struct radix_tree_root	mac_tree;
 	struct mlx4_vlan_table	vlan_table;
+	int			base_qpn;
 };
 
 struct mlx4_sense {
diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h
index ad4df66750bc..5a2c56023668 100644
--- a/drivers/net/mlx4/mlx4_en.h
+++ b/drivers/net/mlx4/mlx4_en.h
@@ -458,6 +458,7 @@ struct mlx4_en_priv {
 	struct mlx4_en_rss_map rss_map;
 	u32 flags;
 #define MLX4_EN_FLAG_PROMISC	0x1
+#define MLX4_EN_FLAG_MC_PROMISC	0x2
 	u32 tx_ring_num;
 	u32 rx_ring_num;
 	u32 rx_skb_size;
diff --git a/drivers/net/mlx4/port.c b/drivers/net/mlx4/port.c
index 451339559bdc..eca7d8596f87 100644
--- a/drivers/net/mlx4/port.c
+++ b/drivers/net/mlx4/port.c
@@ -90,12 +90,79 @@ static int mlx4_set_port_mac_table(struct mlx4_dev *dev, u8 port,
 	return err;
 }
 
-int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *index)
+static int mlx4_uc_steer_add(struct mlx4_dev *dev, u8 port,
+			     u64 mac, int *qpn, u8 reserve)
 {
-	struct mlx4_mac_table *table = &mlx4_priv(dev)->port[port].mac_table;
+	struct mlx4_qp qp;
+	u8 gid[16] = {0};
+	int err;
+
+	if (reserve) {
+		err = mlx4_qp_reserve_range(dev, 1, 1, qpn);
+		if (err) {
+			mlx4_err(dev, "Failed to reserve qp for mac registration\n");
+			return err;
+		}
+	}
+	qp.qpn = *qpn;
+
+	mac &= 0xffffffffffffULL;
+	mac = cpu_to_be64(mac << 16);
+	memcpy(&gid[10], &mac, ETH_ALEN);
+	gid[5] = port;
+	gid[7] = MLX4_UC_STEER << 1;
+
+	err = mlx4_qp_attach_common(dev, &qp, gid, 0,
+				    MLX4_PROT_ETH, MLX4_UC_STEER);
+	if (err && reserve)
+		mlx4_qp_release_range(dev, *qpn, 1);
+
+	return err;
+}
+
+static void mlx4_uc_steer_release(struct mlx4_dev *dev, u8 port,
+				  u64 mac, int qpn, u8 free)
+{
+	struct mlx4_qp qp;
+	u8 gid[16] = {0};
+
+	qp.qpn = qpn;
+	mac &= 0xffffffffffffULL;
+	mac = cpu_to_be64(mac << 16);
+	memcpy(&gid[10], &mac, ETH_ALEN);
+	gid[5] = port;
+	gid[7] = MLX4_UC_STEER << 1;
+
+	mlx4_qp_detach_common(dev, &qp, gid, MLX4_PROT_ETH, MLX4_UC_STEER);
+	if (free)
+		mlx4_qp_release_range(dev, qpn, 1);
+}
+
+int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *qpn, u8 wrap)
+{
+	struct mlx4_port_info *info = &mlx4_priv(dev)->port[port];
+	struct mlx4_mac_table *table = &info->mac_table;
+	struct mlx4_mac_entry *entry;
 	int i, err = 0;
 	int free = -1;
 
+	if (dev->caps.vep_uc_steering) {
+		err = mlx4_uc_steer_add(dev, port, mac, qpn, 1);
+		if (!err) {
+			entry = kmalloc(sizeof *entry, GFP_KERNEL);
+			if (!entry) {
+				mlx4_uc_steer_release(dev, port, mac, *qpn, 1);
+				return -ENOMEM;
+			}
+			entry->mac = mac;
+			err = radix_tree_insert(&info->mac_tree, *qpn, entry);
+			if (err) {
+				mlx4_uc_steer_release(dev, port, mac, *qpn, 1);
+				return err;
+			}
+		} else
+			return err;
+	}
 	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++) {
@@ -106,7 +173,6 @@ int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *index)
 
 		if (mac == (MLX4_MAC_MASK & be64_to_cpu(table->entries[i]))) {
 			/* MAC already registered, increase refernce count */
-			*index = i;
 			++table->refs[i];
 			goto out;
 		}
@@ -137,7 +203,8 @@ int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *index)
 		goto out;
 	}
 
-	*index = free;
+	if (!dev->caps.vep_uc_steering)
+		*qpn = info->base_qpn + free;
 	++table->total;
 out:
 	mutex_unlock(&table->mutex);
@@ -145,20 +212,52 @@ out:
 }
 EXPORT_SYMBOL_GPL(mlx4_register_mac);
 
-void mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, int index)
+static int validate_index(struct mlx4_dev *dev,
+			  struct mlx4_mac_table *table, int index)
 {
-	struct mlx4_mac_table *table = &mlx4_priv(dev)->port[port].mac_table;
+	int err = 0;
 
-	mutex_lock(&table->mutex);
-	if (!table->refs[index]) {
-		mlx4_warn(dev, "No MAC entry for index %d\n", index);
-		goto out;
+	if (index < 0 || index >= table->max || !table->entries[index]) {
+		mlx4_warn(dev, "No valid Mac entry for the given index\n");
+		err = -EINVAL;
 	}
-	if (--table->refs[index]) {
-		mlx4_warn(dev, "Have more references for index %d,"
-			  "no need to modify MAC table\n", index);
-		goto out;
+	return err;
+}
+
+static int find_index(struct mlx4_dev *dev,
+		      struct mlx4_mac_table *table, u64 mac)
+{
+	int i;
+	for (i = 0; i < MLX4_MAX_MAC_NUM; i++) {
+		if (mac == (MLX4_MAC_MASK & be64_to_cpu(table->entries[i])))
+			return i;
 	}
+	/* Mac not found */
+	return -EINVAL;
+}
+
+void mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, int qpn)
+{
+	struct mlx4_port_info *info = &mlx4_priv(dev)->port[port];
+	struct mlx4_mac_table *table = &info->mac_table;
+	int index = qpn - info->base_qpn;
+	struct mlx4_mac_entry *entry;
+
+	if (dev->caps.vep_uc_steering) {
+		entry = radix_tree_lookup(&info->mac_tree, qpn);
+		if (entry) {
+			mlx4_uc_steer_release(dev, port, entry->mac, qpn, 1);
+			radix_tree_delete(&info->mac_tree, qpn);
+			index = find_index(dev, table, entry->mac);
+			kfree(entry);
+		}
+	}
+
+	mutex_lock(&table->mutex);
+
+	if (validate_index(dev, table, index))
+		goto out;
+
 	table->entries[index] = 0;
 	mlx4_set_port_mac_table(dev, port, table->entries);
 	--table->total;
@@ -167,6 +266,44 @@ out:
 }
 EXPORT_SYMBOL_GPL(mlx4_unregister_mac);
 
+int mlx4_replace_mac(struct mlx4_dev *dev, u8 port, int qpn, u64 new_mac, u8 wrap)
+{
+	struct mlx4_port_info *info = &mlx4_priv(dev)->port[port];
+	struct mlx4_mac_table *table = &info->mac_table;
+	int index = qpn - info->base_qpn;
+	struct mlx4_mac_entry *entry;
+	int err;
+
+	if (dev->caps.vep_uc_steering) {
+		entry = radix_tree_lookup(&info->mac_tree, qpn);
+		if (!entry)
+			return -EINVAL;
+		index = find_index(dev, table, entry->mac);
+		mlx4_uc_steer_release(dev, port, entry->mac, qpn, 0);
+		entry->mac = new_mac;
+		err = mlx4_uc_steer_add(dev, port, entry->mac, &qpn, 0);
+		if (err || index < 0)
+			return err;
+	}
+
+	mutex_lock(&table->mutex);
+
+	err = validate_index(dev, table, index);
+	if (err)
+		goto out;
+
+	table->entries[index] = cpu_to_be64(new_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) new_mac);
+		table->entries[index] = 0;
+	}
+out:
+	mutex_unlock(&table->mutex);
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx4_replace_mac);
 static int mlx4_set_port_vlan_table(struct mlx4_dev *dev, u8 port,
 				    __be32 *entries)
 {
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index ebda939b9fc3..56fa5e1cd6d4 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -525,9 +525,15 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 			  int block_mcast_loopback, enum mlx4_protocol protocol);
 int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
 			  enum mlx4_protocol protocol);
-
-int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *index);
-void mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, int index);
+int mlx4_multicast_promisc_add(struct mlx4_dev *dev, u32 qpn, u8 port);
+int mlx4_multicast_promisc_remove(struct mlx4_dev *dev, u32 qpn, u8 port);
+int mlx4_unicast_promisc_add(struct mlx4_dev *dev, u32 qpn, u8 port);
+int mlx4_unicast_promisc_remove(struct mlx4_dev *dev, u32 qpn, u8 port);
+int mlx4_SET_MCAST_FLTR(struct mlx4_dev *dev, u8 port, u64 mac, u64 clear, u8 mode);
+
+int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *qpn, u8 wrap);
+void mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, int qpn);
+int mlx4_replace_mac(struct mlx4_dev *dev, u8 port, int qpn, u64 new_mac, u8 wrap);
 
 int mlx4_find_cached_vlan(struct mlx4_dev *dev, u8 port, u16 vid, int *idx);
 int mlx4_register_vlan(struct mlx4_dev *dev, u8 port, u16 vlan, int *index);
-- 
cgit v1.2.3


From c1b43dca137f2154845122417fba86d4bae67182 Mon Sep 17 00:00:00 2001
From: Eli Cohen <eli@mellanox.co.il>
Date: Tue, 22 Mar 2011 22:38:41 +0000
Subject: mlx4: Add blue flame support for kernel consumers

Using blue flame can improve latency by allowing the HW to more efficiently
access the WQE. This patch presents two functions that are used to allocate or
release HW resources for using blue flame; the caller need to supply a struct
mlx4_bf object when allocating resources. Consumers that make use of this API
should post doorbells to the UAR object pointed by the initialized struct
mlx4_bf;

Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/mlx4/main.c     | 31 +++++++++++++++
 drivers/net/mlx4/mlx4.h     |  3 ++
 drivers/net/mlx4/pd.c       | 94 +++++++++++++++++++++++++++++++++++++++++++++
 include/linux/mlx4/device.h | 13 +++++++
 include/linux/mlx4/qp.h     |  1 +
 5 files changed, 142 insertions(+)

(limited to 'include/linux/mlx4/device.h')

diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index c8e276138f81..05c5671749aa 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -39,6 +39,7 @@
 #include <linux/pci.h>
 #include <linux/dma-mapping.h>
 #include <linux/slab.h>
+#include <linux/io-mapping.h>
 
 #include <linux/mlx4/device.h>
 #include <linux/mlx4/doorbell.h>
@@ -721,8 +722,31 @@ static void mlx4_free_icms(struct mlx4_dev *dev)
 	mlx4_free_icm(dev, priv->fw.aux_icm, 0);
 }
 
+static int map_bf_area(struct mlx4_dev *dev)
+{
+	struct mlx4_priv *priv = mlx4_priv(dev);
+	resource_size_t bf_start;
+	resource_size_t bf_len;
+	int err = 0;
+
+	bf_start = pci_resource_start(dev->pdev, 2) + (dev->caps.num_uars << PAGE_SHIFT);
+	bf_len = pci_resource_len(dev->pdev, 2) - (dev->caps.num_uars << PAGE_SHIFT);
+	priv->bf_mapping = io_mapping_create_wc(bf_start, bf_len);
+	if (!priv->bf_mapping)
+		err = -ENOMEM;
+
+	return err;
+}
+
+static void unmap_bf_area(struct mlx4_dev *dev)
+{
+	if (mlx4_priv(dev)->bf_mapping)
+		io_mapping_free(mlx4_priv(dev)->bf_mapping);
+}
+
 static void mlx4_close_hca(struct mlx4_dev *dev)
 {
+	unmap_bf_area(dev);
 	mlx4_CLOSE_HCA(dev, 0);
 	mlx4_free_icms(dev);
 	mlx4_UNMAP_FA(dev);
@@ -775,6 +799,9 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
 		goto err_stop_fw;
 	}
 
+	if (map_bf_area(dev))
+		mlx4_dbg(dev, "Failed to map blue flame area\n");
+
 	init_hca.log_uar_sz = ilog2(dev->caps.num_uars);
 
 	err = mlx4_init_icm(dev, &dev_cap, &init_hca, icm_size);
@@ -805,6 +832,7 @@ err_free_icm:
 	mlx4_free_icms(dev);
 
 err_stop_fw:
+	unmap_bf_area(dev);
 	mlx4_UNMAP_FA(dev);
 	mlx4_free_icm(dev, priv->fw.fw_icm, 0);
 
@@ -1196,6 +1224,9 @@ static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	pci_read_config_byte(pdev, PCI_REVISION_ID, &dev->rev_id);
 
+	INIT_LIST_HEAD(&priv->bf_list);
+	mutex_init(&priv->bf_mutex);
+
 	/*
 	 * Now reset the HCA before we touch the PCI capabilities or
 	 * attempt a firmware command, since a boot ROM may have left
diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h
index d8bb4418581b..bc9a21657a4f 100644
--- a/drivers/net/mlx4/mlx4.h
+++ b/drivers/net/mlx4/mlx4.h
@@ -353,6 +353,9 @@ struct mlx4_priv {
 	struct mutex		port_mutex;
 	struct mlx4_msix_ctl	msix_ctl;
 	struct mlx4_steer	*steer;
+	struct list_head	bf_list;
+	struct mutex		bf_mutex;
+	struct io_mapping	*bf_mapping;
 };
 
 static inline struct mlx4_priv *mlx4_priv(struct mlx4_dev *dev)
diff --git a/drivers/net/mlx4/pd.c b/drivers/net/mlx4/pd.c
index c4988d6bd5b2..5210a0f31413 100644
--- a/drivers/net/mlx4/pd.c
+++ b/drivers/net/mlx4/pd.c
@@ -32,6 +32,7 @@
  */
 
 #include <linux/errno.h>
+#include <linux/io-mapping.h>
 
 #include <asm/page.h>
 
@@ -77,6 +78,7 @@ int mlx4_uar_alloc(struct mlx4_dev *dev, struct mlx4_uar *uar)
 		return -ENOMEM;
 
 	uar->pfn = (pci_resource_start(dev->pdev, 2) >> PAGE_SHIFT) + uar->index;
+	uar->map = NULL;
 
 	return 0;
 }
@@ -88,6 +90,98 @@ void mlx4_uar_free(struct mlx4_dev *dev, struct mlx4_uar *uar)
 }
 EXPORT_SYMBOL_GPL(mlx4_uar_free);
 
+int mlx4_bf_alloc(struct mlx4_dev *dev, struct mlx4_bf *bf)
+{
+	struct mlx4_priv *priv = mlx4_priv(dev);
+	struct mlx4_uar *uar;
+	int err = 0;
+	int idx;
+
+	if (!priv->bf_mapping)
+		return -ENOMEM;
+
+	mutex_lock(&priv->bf_mutex);
+	if (!list_empty(&priv->bf_list))
+		uar = list_entry(priv->bf_list.next, struct mlx4_uar, bf_list);
+	else {
+		uar = kmalloc(sizeof *uar, GFP_KERNEL);
+		if (!uar) {
+			err = -ENOMEM;
+			goto out;
+		}
+		err = mlx4_uar_alloc(dev, uar);
+		if (err)
+			goto free_kmalloc;
+
+		uar->map = ioremap(uar->pfn << PAGE_SHIFT, PAGE_SIZE);
+		if (!uar->map) {
+			err = -ENOMEM;
+			goto free_uar;
+		}
+
+		uar->bf_map = io_mapping_map_wc(priv->bf_mapping, uar->index << PAGE_SHIFT);
+		if (!uar->bf_map) {
+			err = -ENOMEM;
+			goto unamp_uar;
+		}
+		uar->free_bf_bmap = 0;
+		list_add(&uar->bf_list, &priv->bf_list);
+	}
+
+	bf->uar = uar;
+	idx = ffz(uar->free_bf_bmap);
+	uar->free_bf_bmap |= 1 << idx;
+	bf->uar = uar;
+	bf->offset = 0;
+	bf->buf_size = dev->caps.bf_reg_size / 2;
+	bf->reg = uar->bf_map + idx * dev->caps.bf_reg_size;
+	if (uar->free_bf_bmap == (1 << dev->caps.bf_regs_per_page) - 1)
+		list_del_init(&uar->bf_list);
+
+	goto out;
+
+unamp_uar:
+	bf->uar = NULL;
+	iounmap(uar->map);
+
+free_uar:
+	mlx4_uar_free(dev, uar);
+
+free_kmalloc:
+	kfree(uar);
+
+out:
+	mutex_unlock(&priv->bf_mutex);
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx4_bf_alloc);
+
+void mlx4_bf_free(struct mlx4_dev *dev, struct mlx4_bf *bf)
+{
+	struct mlx4_priv *priv = mlx4_priv(dev);
+	int idx;
+
+	if (!bf->uar || !bf->uar->bf_map)
+		return;
+
+	mutex_lock(&priv->bf_mutex);
+	idx = (bf->reg - bf->uar->bf_map) / dev->caps.bf_reg_size;
+	bf->uar->free_bf_bmap &= ~(1 << idx);
+	if (!bf->uar->free_bf_bmap) {
+		if (!list_empty(&bf->uar->bf_list))
+			list_del(&bf->uar->bf_list);
+
+		io_mapping_unmap(bf->uar->bf_map);
+		iounmap(bf->uar->map);
+		mlx4_uar_free(dev, bf->uar);
+		kfree(bf->uar);
+	} else if (list_empty(&bf->uar->bf_list))
+		list_add(&bf->uar->bf_list, &priv->bf_list);
+
+	mutex_unlock(&priv->bf_mutex);
+}
+EXPORT_SYMBOL_GPL(mlx4_bf_free);
+
 int mlx4_init_uar_table(struct mlx4_dev *dev)
 {
 	if (dev->caps.num_uars <= 128) {
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 56fa5e1cd6d4..8985768e2c0d 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -351,6 +351,17 @@ struct mlx4_fmr {
 struct mlx4_uar {
 	unsigned long		pfn;
 	int			index;
+	struct list_head	bf_list;
+	unsigned		free_bf_bmap;
+	void __iomem	       *map;
+	void __iomem	       *bf_map;
+};
+
+struct mlx4_bf {
+	unsigned long		offset;
+	int			buf_size;
+	struct mlx4_uar	       *uar;
+	void __iomem	       *reg;
 };
 
 struct mlx4_cq {
@@ -478,6 +489,8 @@ void mlx4_pd_free(struct mlx4_dev *dev, u32 pdn);
 
 int mlx4_uar_alloc(struct mlx4_dev *dev, struct mlx4_uar *uar);
 void mlx4_uar_free(struct mlx4_dev *dev, struct mlx4_uar *uar);
+int mlx4_bf_alloc(struct mlx4_dev *dev, struct mlx4_bf *bf);
+void mlx4_bf_free(struct mlx4_dev *dev, struct mlx4_bf *bf);
 
 int mlx4_mtt_init(struct mlx4_dev *dev, int npages, int page_shift,
 		  struct mlx4_mtt *mtt);
diff --git a/include/linux/mlx4/qp.h b/include/linux/mlx4/qp.h
index 0eeb2a1a867c..9e9eb21056ca 100644
--- a/include/linux/mlx4/qp.h
+++ b/include/linux/mlx4/qp.h
@@ -303,6 +303,7 @@ struct mlx4_wqe_data_seg {
 
 enum {
 	MLX4_INLINE_ALIGN	= 64,
+	MLX4_INLINE_SEG		= 1 << 31,
 };
 
 struct mlx4_wqe_inline_seg {
-- 
cgit v1.2.3