From b6055006b26de15aaa66870809b3d63b334737ce Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin Date: Tue, 22 Mar 2011 22:37:23 +0000 Subject: mlx4_en: going out of range of TX rings when reporting stats Signed-off-by: Yevgeny Petrilin Signed-off-by: David S. Miller --- drivers/net/mlx4/en_port.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/en_port.c b/drivers/net/mlx4/en_port.c index 7f5a322..a2b0bc4 100644 --- a/drivers/net/mlx4/en_port.c +++ b/drivers/net/mlx4/en_port.c @@ -206,7 +206,7 @@ int mlx4_en_DUMP_ETH_STATS(struct mlx4_en_dev *mdev, u8 port, u8 reset) } stats->tx_packets = 0; stats->tx_bytes = 0; - for (i = 0; i <= priv->tx_ring_num; i++) { + for (i = 0; i < priv->tx_ring_num; i++) { stats->tx_packets += priv->tx_ring[i].packets; stats->tx_bytes += priv->tx_ring[i].bytes; } -- cgit v1.1 From 39f17b44aaaa6bc47ded0bbb17fc617ec6cf08bb Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin Date: Tue, 22 Mar 2011 22:37:28 +0000 Subject: mlx4_en: moderation parameters are not reseted. Instead of reseting the module parameters each ifup or mtu change, they are being set once at device initialization Signed-off-by: Yevgeny Petrilin Signed-off-by: David S. Miller --- drivers/net/mlx4/en_netdev.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c index 897f576..56627c2 100644 --- a/drivers/net/mlx4/en_netdev.c +++ b/drivers/net/mlx4/en_netdev.c @@ -801,7 +801,6 @@ static int mlx4_en_open(struct net_device *dev) priv->rx_ring[i].packets = 0; } - mlx4_en_set_default_moderation(priv); err = mlx4_en_start_port(dev); if (err) en_err(priv, "Failed starting port:%d\n", priv->port); @@ -932,7 +931,6 @@ static int mlx4_en_change_mtu(struct net_device *dev, int new_mtu) en_dbg(DRV, priv, "Change MTU called with card down!?\n"); } else { mlx4_en_stop_port(dev); - mlx4_en_set_default_moderation(priv); err = mlx4_en_start_port(dev); if (err) { en_err(priv, "Failed restarting port:%d\n", @@ -1080,6 +1078,7 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, en_warn(priv, "Using %d RX rings\n", prof->rx_ring_num); priv->registered = 1; + mlx4_en_set_default_moderation(priv); queue_delayed_work(mdev->workqueue, &priv->stats_task, STATS_DELAY); return 0; -- cgit v1.1 From 46afd0fb01d62ee217f142ab9c26054304ab7454 Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin Date: Tue, 22 Mar 2011 22:37:36 +0000 Subject: mlx4_en: optimize adaptive moderation algorithm for better latency Signed-off-by: Yevgeny Petrilin Signed-off-by: David S. Miller --- drivers/net/mlx4/en_netdev.c | 18 +++++------------- drivers/net/mlx4/mlx4_en.h | 1 + 2 files changed, 6 insertions(+), 13 deletions(-) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c index 56627c2..b1180dd 100644 --- a/drivers/net/mlx4/en_netdev.c +++ b/drivers/net/mlx4/en_netdev.c @@ -417,7 +417,6 @@ static void mlx4_en_auto_moderation(struct mlx4_en_priv *priv) unsigned long avg_pkt_size; unsigned long rx_packets; unsigned long rx_bytes; - unsigned long rx_byte_diff; unsigned long tx_packets; unsigned long tx_pkt_diff; unsigned long rx_pkt_diff; @@ -441,25 +440,20 @@ static void mlx4_en_auto_moderation(struct mlx4_en_priv *priv) rx_pkt_diff = ((unsigned long) (rx_packets - priv->last_moder_packets)); packets = max(tx_pkt_diff, rx_pkt_diff); - rx_byte_diff = rx_bytes - priv->last_moder_bytes; - rx_byte_diff = rx_byte_diff ? rx_byte_diff : 1; rate = packets * HZ / period; avg_pkt_size = packets ? ((unsigned long) (rx_bytes - priv->last_moder_bytes)) / packets : 0; /* Apply auto-moderation only when packet rate exceeds a rate that * it matters */ - if (rate > MLX4_EN_RX_RATE_THRESH) { + if (rate > MLX4_EN_RX_RATE_THRESH && avg_pkt_size > MLX4_EN_AVG_PKT_SMALL) { /* If tx and rx packet rates are not balanced, assume that * traffic is mainly BW bound and apply maximum moderation. * Otherwise, moderate according to packet rate */ - if (2 * tx_pkt_diff > 3 * rx_pkt_diff && - rx_pkt_diff / rx_byte_diff < - MLX4_EN_SMALL_PKT_SIZE) - moder_time = priv->rx_usecs_low; - else if (2 * rx_pkt_diff > 3 * tx_pkt_diff) + if (2 * tx_pkt_diff > 3 * rx_pkt_diff || + 2 * rx_pkt_diff > 3 * tx_pkt_diff) { moder_time = priv->rx_usecs_high; - else { + } else { if (rate < priv->pkt_rate_low) moder_time = priv->rx_usecs_low; else if (rate > priv->pkt_rate_high) @@ -471,9 +465,7 @@ static void mlx4_en_auto_moderation(struct mlx4_en_priv *priv) priv->rx_usecs_low; } } else { - /* When packet rate is low, use default moderation rather than - * 0 to prevent interrupt storms if traffic suddenly increases */ - moder_time = priv->rx_usecs; + moder_time = priv->rx_usecs_low; } en_dbg(INTR, priv, "tx rate:%lu rx_rate:%lu\n", diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h index dfed6a0..5387c4e 100644 --- a/drivers/net/mlx4/mlx4_en.h +++ b/drivers/net/mlx4/mlx4_en.h @@ -124,6 +124,7 @@ enum { #define MLX4_EN_RX_SIZE_THRESH 1024 #define MLX4_EN_RX_RATE_THRESH (1000000 / MLX4_EN_RX_COAL_TIME_HIGH) #define MLX4_EN_SAMPLE_INTERVAL 0 +#define MLX4_EN_AVG_PKT_SMALL 256 #define MLX4_EN_AUTO_CONF 0xffff -- cgit v1.1 From 908222655b38005483d64385e06341a38647fdf1 Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin Date: Tue, 22 Mar 2011 22:37:41 +0000 Subject: mlx4_en: bringing link up when registering netdevice Signed-off-by: Yevgeny Petrilin Signed-off-by: David S. Miller --- drivers/net/mlx4/en_netdev.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c index b1180dd..5727bf5 100644 --- a/drivers/net/mlx4/en_netdev.c +++ b/drivers/net/mlx4/en_netdev.c @@ -1069,6 +1069,23 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, en_warn(priv, "Using %d TX rings\n", prof->tx_ring_num); en_warn(priv, "Using %d RX rings\n", prof->rx_ring_num); + /* Configure port */ + err = mlx4_SET_PORT_general(mdev->dev, priv->port, + MLX4_EN_MIN_MTU, + 0, 0, 0, 0); + if (err) { + en_err(priv, "Failed setting port general configurations " + "for port %d, with error %d\n", priv->port, err); + goto out; + } + + /* Init port */ + en_warn(priv, "Initializing port\n"); + err = mlx4_INIT_PORT(mdev->dev, priv->port); + if (err) { + en_err(priv, "Failed Initializing port\n"); + goto out; + } priv->registered = 1; mlx4_en_set_default_moderation(priv); queue_delayed_work(mdev->workqueue, &priv->stats_task, STATS_DELAY); -- cgit v1.1 From 0b7ca5a928e2271bbc225e9e1ac1f22e9fbee54f Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin 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 Signed-off-by: Yevgeny Petrilin Signed-off-by: David S. Miller --- drivers/net/mlx4/cq.c | 2 +- drivers/net/mlx4/eq.c | 107 ++++++++++++++++++++++++++++++++++++++++++--- drivers/net/mlx4/main.c | 20 +++++++-- drivers/net/mlx4/mlx4.h | 6 +++ drivers/net/mlx4/profile.c | 4 +- 5 files changed, 126 insertions(+), 13 deletions(-) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/cq.c b/drivers/net/mlx4/cq.c index 7cd34e9..bd8ef9f 100644 --- a/drivers/net/mlx4/cq.c +++ b/drivers/net/mlx4/cq.c @@ -198,7 +198,7 @@ int mlx4_cq_alloc(struct mlx4_dev *dev, int nent, struct mlx4_mtt *mtt, u64 mtt_addr; int err; - if (vector >= dev->caps.num_comp_vectors) + if (vector > dev->caps.num_comp_vectors + dev->caps.comp_pool) return -EINVAL; cq->vector = vector; diff --git a/drivers/net/mlx4/eq.c b/drivers/net/mlx4/eq.c index 552d0fc..506cfd0 100644 --- a/drivers/net/mlx4/eq.c +++ b/drivers/net/mlx4/eq.c @@ -42,7 +42,7 @@ #include "fw.h" enum { - MLX4_IRQNAME_SIZE = 64 + MLX4_IRQNAME_SIZE = 32 }; enum { @@ -317,8 +317,8 @@ static int mlx4_num_eq_uar(struct mlx4_dev *dev) * we need to map, take the difference of highest index and * the lowest index we'll use and add 1. */ - return (dev->caps.num_comp_vectors + 1 + dev->caps.reserved_eqs) / 4 - - dev->caps.reserved_eqs / 4 + 1; + return (dev->caps.num_comp_vectors + 1 + dev->caps.reserved_eqs + + dev->caps.comp_pool)/4 - dev->caps.reserved_eqs/4 + 1; } static void __iomem *mlx4_get_eq_uar(struct mlx4_dev *dev, struct mlx4_eq *eq) @@ -496,16 +496,32 @@ static void mlx4_free_eq(struct mlx4_dev *dev, static void mlx4_free_irqs(struct mlx4_dev *dev) { struct mlx4_eq_table *eq_table = &mlx4_priv(dev)->eq_table; - int i; + struct mlx4_priv *priv = mlx4_priv(dev); + int i, vec; if (eq_table->have_irq) free_irq(dev->pdev->irq, dev); + for (i = 0; i < dev->caps.num_comp_vectors + 1; ++i) if (eq_table->eq[i].have_irq) { free_irq(eq_table->eq[i].irq, eq_table->eq + i); eq_table->eq[i].have_irq = 0; } + for (i = 0; i < dev->caps.comp_pool; i++) { + /* + * Freeing the assigned irq's + * all bits should be 0, but we need to validate + */ + if (priv->msix_ctl.pool_bm & 1ULL << i) { + /* NO need protecting*/ + vec = dev->caps.num_comp_vectors + 1 + i; + free_irq(priv->eq_table.eq[vec].irq, + &priv->eq_table.eq[vec]); + } + } + + kfree(eq_table->irq_names); } @@ -578,7 +594,8 @@ int mlx4_init_eq_table(struct mlx4_dev *dev) (priv->eq_table.inta_pin < 32 ? 4 : 0); priv->eq_table.irq_names = - kmalloc(MLX4_IRQNAME_SIZE * (dev->caps.num_comp_vectors + 1), + kmalloc(MLX4_IRQNAME_SIZE * (dev->caps.num_comp_vectors + 1 + + dev->caps.comp_pool), GFP_KERNEL); if (!priv->eq_table.irq_names) { err = -ENOMEM; @@ -601,6 +618,22 @@ int mlx4_init_eq_table(struct mlx4_dev *dev) if (err) goto err_out_comp; + /*if additional completion vectors poolsize is 0 this loop will not run*/ + for (i = dev->caps.num_comp_vectors + 1; + i < dev->caps.num_comp_vectors + dev->caps.comp_pool + 1; ++i) { + + err = mlx4_create_eq(dev, dev->caps.num_cqs - + dev->caps.reserved_cqs + + MLX4_NUM_SPARE_EQE, + (dev->flags & MLX4_FLAG_MSI_X) ? i : 0, + &priv->eq_table.eq[i]); + if (err) { + --i; + goto err_out_unmap; + } + } + + if (dev->flags & MLX4_FLAG_MSI_X) { const char *eq_name; @@ -686,7 +719,7 @@ void mlx4_cleanup_eq_table(struct mlx4_dev *dev) mlx4_free_irqs(dev); - for (i = 0; i < dev->caps.num_comp_vectors + 1; ++i) + for (i = 0; i < dev->caps.num_comp_vectors + dev->caps.comp_pool + 1; ++i) mlx4_free_eq(dev, &priv->eq_table.eq[i]); mlx4_unmap_clr_int(dev); @@ -743,3 +776,65 @@ int mlx4_test_interrupts(struct mlx4_dev *dev) return err; } EXPORT_SYMBOL(mlx4_test_interrupts); + +int mlx4_assign_eq(struct mlx4_dev *dev, char* name, int * vector) +{ + + struct mlx4_priv *priv = mlx4_priv(dev); + int vec = 0, err = 0, i; + + spin_lock(&priv->msix_ctl.pool_lock); + for (i = 0; !vec && i < dev->caps.comp_pool; i++) { + if (~priv->msix_ctl.pool_bm & 1ULL << i) { + priv->msix_ctl.pool_bm |= 1ULL << i; + vec = dev->caps.num_comp_vectors + 1 + i; + snprintf(priv->eq_table.irq_names + + vec * MLX4_IRQNAME_SIZE, + MLX4_IRQNAME_SIZE, "%s", name); + err = request_irq(priv->eq_table.eq[vec].irq, + mlx4_msi_x_interrupt, 0, + &priv->eq_table.irq_names[vec<<5], + priv->eq_table.eq + vec); + if (err) { + /*zero out bit by fliping it*/ + priv->msix_ctl.pool_bm ^= 1 << i; + vec = 0; + continue; + /*we dont want to break here*/ + } + eq_set_ci(&priv->eq_table.eq[vec], 1); + } + } + spin_unlock(&priv->msix_ctl.pool_lock); + + if (vec) { + *vector = vec; + } else { + *vector = 0; + err = (i == dev->caps.comp_pool) ? -ENOSPC : err; + } + return err; +} +EXPORT_SYMBOL(mlx4_assign_eq); + +void mlx4_release_eq(struct mlx4_dev *dev, int vec) +{ + struct mlx4_priv *priv = mlx4_priv(dev); + /*bm index*/ + int i = vec - dev->caps.num_comp_vectors - 1; + + if (likely(i >= 0)) { + /*sanity check , making sure were not trying to free irq's + Belonging to a legacy EQ*/ + spin_lock(&priv->msix_ctl.pool_lock); + if (priv->msix_ctl.pool_bm & 1ULL << i) { + free_irq(priv->eq_table.eq[vec].irq, + &priv->eq_table.eq[vec]); + priv->msix_ctl.pool_bm &= ~(1ULL << i); + } + spin_unlock(&priv->msix_ctl.pool_lock); + } + +} +EXPORT_SYMBOL(mlx4_release_eq); + diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c index 2765a3c..517ca34 100644 --- a/drivers/net/mlx4/main.c +++ b/drivers/net/mlx4/main.c @@ -969,13 +969,15 @@ static void mlx4_enable_msi_x(struct mlx4_dev *dev) { struct mlx4_priv *priv = mlx4_priv(dev); struct msix_entry *entries; - int nreq; + int nreq = min_t(int, dev->caps.num_ports * + min_t(int, num_online_cpus() + 1, MAX_MSIX_P_PORT) + + MSIX_LEGACY_SZ, MAX_MSIX); int err; int i; if (msi_x) { nreq = min_t(int, dev->caps.num_eqs - dev->caps.reserved_eqs, - num_possible_cpus() + 1); + nreq); entries = kcalloc(nreq, sizeof *entries, GFP_KERNEL); if (!entries) goto no_msi; @@ -998,7 +1000,15 @@ static void mlx4_enable_msi_x(struct mlx4_dev *dev) goto no_msi; } - dev->caps.num_comp_vectors = nreq - 1; + if (nreq < + MSIX_LEGACY_SZ + dev->caps.num_ports * MIN_MSIX_P_PORT) { + /*Working in legacy mode , all EQ's shared*/ + dev->caps.comp_pool = 0; + dev->caps.num_comp_vectors = nreq - 1; + } else { + dev->caps.comp_pool = nreq - MSIX_LEGACY_SZ; + dev->caps.num_comp_vectors = MSIX_LEGACY_SZ - 1; + } for (i = 0; i < nreq; ++i) priv->eq_table.eq[i].irq = entries[i].vector; @@ -1010,6 +1020,7 @@ static void mlx4_enable_msi_x(struct mlx4_dev *dev) no_msi: dev->caps.num_comp_vectors = 1; + dev->caps.comp_pool = 0; for (i = 0; i < 2; ++i) priv->eq_table.eq[i].irq = dev->pdev->irq; @@ -1151,6 +1162,9 @@ static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id) if (err) goto err_close; + priv->msix_ctl.pool_bm = 0; + spin_lock_init(&priv->msix_ctl.pool_lock); + mlx4_enable_msi_x(dev); err = mlx4_setup_hca(dev); diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h index 0da5bb72..67ee8da 100644 --- a/drivers/net/mlx4/mlx4.h +++ b/drivers/net/mlx4/mlx4.h @@ -282,6 +282,11 @@ struct mlx4_sense { struct delayed_work sense_poll; }; +struct mlx4_msix_ctl { + u64 pool_bm; + spinlock_t pool_lock; +}; + struct mlx4_priv { struct mlx4_dev dev; @@ -313,6 +318,7 @@ struct mlx4_priv { struct mlx4_port_info port[MLX4_MAX_PORTS + 1]; struct mlx4_sense sense; struct mutex port_mutex; + struct mlx4_msix_ctl msix_ctl; }; static inline struct mlx4_priv *mlx4_priv(struct mlx4_dev *dev) diff --git a/drivers/net/mlx4/profile.c b/drivers/net/mlx4/profile.c index e749f82..b967647 100644 --- a/drivers/net/mlx4/profile.c +++ b/drivers/net/mlx4/profile.c @@ -107,9 +107,7 @@ u64 mlx4_make_profile(struct mlx4_dev *dev, profile[MLX4_RES_AUXC].num = request->num_qp; profile[MLX4_RES_SRQ].num = request->num_srq; profile[MLX4_RES_CQ].num = request->num_cq; - profile[MLX4_RES_EQ].num = min_t(unsigned, dev_cap->max_eqs, - dev_cap->reserved_eqs + - num_possible_cpus() + 1); + profile[MLX4_RES_EQ].num = min_t(unsigned, dev_cap->max_eqs, MAX_MSIX); profile[MLX4_RES_DMPT].num = request->num_mpt; profile[MLX4_RES_CMPT].num = MLX4_NUM_CMPTS; profile[MLX4_RES_MTT].num = request->num_mtt; -- cgit v1.1 From 1fb9876e9bf895ea4127ff17180f1b2ab37771b6 Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin Date: Tue, 22 Mar 2011 22:37:52 +0000 Subject: mlx4_en: using new mlx4 interrupt scheme Each RX ring will have its own interrupt vector, and TX rings will share one (we mostly use polling for TX completions). The vectors are assigned first time device is opened, and its name includes the interface name and ring number. Signed-off-by: Markuze Alex Signed-off-by: Yevgeny Petrilin Signed-off-by: David S. Miller --- drivers/net/mlx4/en_cq.c | 38 +++++++++++++++++++++++++++++++------- drivers/net/mlx4/en_ethtool.c | 2 +- drivers/net/mlx4/en_main.c | 20 +++++++++++--------- drivers/net/mlx4/en_netdev.c | 18 ++++++++++++++---- drivers/net/mlx4/mlx4_en.h | 7 +++++-- 5 files changed, 62 insertions(+), 23 deletions(-) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/en_cq.c b/drivers/net/mlx4/en_cq.c index 21786ad..ec4b6d0 100644 --- a/drivers/net/mlx4/en_cq.c +++ b/drivers/net/mlx4/en_cq.c @@ -51,13 +51,10 @@ int mlx4_en_create_cq(struct mlx4_en_priv *priv, int err; cq->size = entries; - if (mode == RX) { + if (mode == RX) cq->buf_size = cq->size * sizeof(struct mlx4_cqe); - cq->vector = ring % mdev->dev->caps.num_comp_vectors; - } else { + else cq->buf_size = sizeof(struct mlx4_cqe); - cq->vector = 0; - } cq->ring = ring; cq->is_tx = mode; @@ -80,7 +77,8 @@ int mlx4_en_create_cq(struct mlx4_en_priv *priv, int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq) { struct mlx4_en_dev *mdev = priv->mdev; - int err; + int err = 0; + char name[25]; cq->dev = mdev->pndev[priv->port]; cq->mcq.set_ci_db = cq->wqres.db.db; @@ -89,6 +87,29 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq) *cq->mcq.arm_db = 0; memset(cq->buf, 0, cq->buf_size); + if (cq->is_tx == RX) { + if (mdev->dev->caps.comp_pool) { + if (!cq->vector) { + sprintf(name , "%s-rx-%d", priv->dev->name, cq->ring); + if (mlx4_assign_eq(mdev->dev, name, &cq->vector)) { + cq->vector = (cq->ring + 1 + priv->port) % + mdev->dev->caps.num_comp_vectors; + mlx4_warn(mdev, "Failed Assigning an EQ to " + "%s_rx-%d ,Falling back to legacy EQ's\n", + priv->dev->name, cq->ring); + } + } + } else { + cq->vector = (cq->ring + 1 + priv->port) % + mdev->dev->caps.num_comp_vectors; + } + } else { + if (!cq->vector || !mdev->dev->caps.comp_pool) { + /*Fallback to legacy pool in case of error*/ + cq->vector = 0; + } + } + if (!cq->is_tx) cq->size = priv->rx_ring[cq->ring].actual_size; @@ -112,12 +133,15 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq) return 0; } -void mlx4_en_destroy_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq) +void mlx4_en_destroy_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq, + bool reserve_vectors) { struct mlx4_en_dev *mdev = priv->mdev; mlx4_en_unmap_buffer(&cq->wqres.buf); mlx4_free_hwq_res(mdev->dev, &cq->wqres, cq->buf_size); + if (priv->mdev->dev->caps.comp_pool && cq->vector && !reserve_vectors) + mlx4_release_eq(priv->mdev->dev, cq->vector); cq->buf_size = 0; cq->buf = NULL; } diff --git a/drivers/net/mlx4/en_ethtool.c b/drivers/net/mlx4/en_ethtool.c index 056152b..8cfe858 100644 --- a/drivers/net/mlx4/en_ethtool.c +++ b/drivers/net/mlx4/en_ethtool.c @@ -388,7 +388,7 @@ static int mlx4_en_set_ringparam(struct net_device *dev, mlx4_en_stop_port(dev); } - mlx4_en_free_resources(priv); + mlx4_en_free_resources(priv, true); priv->prof->tx_ring_size = tx_size; priv->prof->rx_ring_size = rx_size; diff --git a/drivers/net/mlx4/en_main.c b/drivers/net/mlx4/en_main.c index 1ff6ca6..29aaa43 100644 --- a/drivers/net/mlx4/en_main.c +++ b/drivers/net/mlx4/en_main.c @@ -241,16 +241,18 @@ static void *mlx4_en_add(struct mlx4_dev *dev) mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_ETH) mdev->port_cnt++; - /* If we did not receive an explicit number of Rx rings, default to - * the number of completion vectors populated by the mlx4_core */ mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_ETH) { - mlx4_info(mdev, "Using %d tx rings for port:%d\n", - mdev->profile.prof[i].tx_ring_num, i); - mdev->profile.prof[i].rx_ring_num = min_t(int, - roundup_pow_of_two(dev->caps.num_comp_vectors), - MAX_RX_RINGS); - mlx4_info(mdev, "Defaulting to %d rx rings for port:%d\n", - mdev->profile.prof[i].rx_ring_num, i); + if (!dev->caps.comp_pool) { + mdev->profile.prof[i].rx_ring_num = + rounddown_pow_of_two(max_t(int, MIN_RX_RINGS, + min_t(int, + dev->caps.num_comp_vectors, + MAX_RX_RINGS))); + } else { + mdev->profile.prof[i].rx_ring_num = rounddown_pow_of_two( + min_t(int, dev->caps.comp_pool/ + dev->caps.num_ports - 1 , MAX_MSIX_P_PORT - 1)); + } } /* Create our own workqueue for reset/multicast tasks diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c index 5727bf5..f6ed315 100644 --- a/drivers/net/mlx4/en_netdev.c +++ b/drivers/net/mlx4/en_netdev.c @@ -557,6 +557,7 @@ int mlx4_en_start_port(struct net_device *dev) int err = 0; int i; int j; + char name[32]; if (priv->port_up) { en_dbg(DRV, priv, "start port called while port already up\n"); @@ -601,10 +602,19 @@ int mlx4_en_start_port(struct net_device *dev) goto cq_err; } + if (mdev->dev->caps.comp_pool && !priv->tx_vector) { + sprintf(name , "%s-tx", priv->dev->name); + if (mlx4_assign_eq(mdev->dev , name, &priv->tx_vector)) { + mlx4_warn(mdev, "Failed Assigning an EQ to " + "%s_tx ,Falling back to legacy " + "EQ's\n", priv->dev->name); + } + } /* Configure tx cq's and rings */ for (i = 0; i < priv->tx_ring_num; i++) { /* Configure cq */ cq = &priv->tx_cq[i]; + cq->vector = priv->tx_vector; err = mlx4_en_activate_cq(priv, cq); if (err) { en_err(priv, "Failed allocating Tx CQ\n"); @@ -819,7 +829,7 @@ static int mlx4_en_close(struct net_device *dev) return 0; } -void mlx4_en_free_resources(struct mlx4_en_priv *priv) +void mlx4_en_free_resources(struct mlx4_en_priv *priv, bool reserve_vectors) { int i; @@ -827,14 +837,14 @@ void mlx4_en_free_resources(struct mlx4_en_priv *priv) if (priv->tx_ring[i].tx_info) mlx4_en_destroy_tx_ring(priv, &priv->tx_ring[i]); if (priv->tx_cq[i].buf) - mlx4_en_destroy_cq(priv, &priv->tx_cq[i]); + mlx4_en_destroy_cq(priv, &priv->tx_cq[i], reserve_vectors); } for (i = 0; i < priv->rx_ring_num; i++) { if (priv->rx_ring[i].rx_info) mlx4_en_destroy_rx_ring(priv, &priv->rx_ring[i]); if (priv->rx_cq[i].buf) - mlx4_en_destroy_cq(priv, &priv->rx_cq[i]); + mlx4_en_destroy_cq(priv, &priv->rx_cq[i], reserve_vectors); } } @@ -896,7 +906,7 @@ void mlx4_en_destroy_netdev(struct net_device *dev) mdev->pndev[priv->port] = NULL; mutex_unlock(&mdev->state_lock); - mlx4_en_free_resources(priv); + mlx4_en_free_resources(priv, false); free_netdev(dev); } diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h index 5387c4e..2db245f 100644 --- a/drivers/net/mlx4/mlx4_en.h +++ b/drivers/net/mlx4/mlx4_en.h @@ -62,6 +62,7 @@ #define MLX4_EN_PAGE_SHIFT 12 #define MLX4_EN_PAGE_SIZE (1 << MLX4_EN_PAGE_SHIFT) #define MAX_RX_RINGS 16 +#define MIN_RX_RINGS 4 #define TXBB_SIZE 64 #define HEADROOM (2048 / TXBB_SIZE + 1) #define STAMP_STRIDE 64 @@ -462,6 +463,7 @@ struct mlx4_en_priv { u16 log_rx_info; struct mlx4_en_tx_ring tx_ring[MAX_TX_RINGS]; + int tx_vector; struct mlx4_en_rx_ring rx_ring[MAX_RX_RINGS]; struct mlx4_en_cq tx_cq[MAX_TX_RINGS]; struct mlx4_en_cq rx_cq[MAX_RX_RINGS]; @@ -487,12 +489,13 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, int mlx4_en_start_port(struct net_device *dev); void mlx4_en_stop_port(struct net_device *dev); -void mlx4_en_free_resources(struct mlx4_en_priv *priv); +void mlx4_en_free_resources(struct mlx4_en_priv *priv, bool reserve_vectors); int mlx4_en_alloc_resources(struct mlx4_en_priv *priv); int mlx4_en_create_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq, int entries, int ring, enum cq_type mode); -void mlx4_en_destroy_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq); +void mlx4_en_destroy_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq, + bool reserve_vectors); int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq); void mlx4_en_deactivate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq); int mlx4_en_set_cq_moder(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq); -- cgit v1.1 From 14c07b1358ede1664652bb9b28d9ace5fe6f7f92 Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin 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 Signed-off-by: Yevgeny Petrilin Signed-off-by: David S. Miller --- 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 +++++ 5 files changed, 89 insertions(+), 2 deletions(-) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/en_ethtool.c b/drivers/net/mlx4/en_ethtool.c index 8cfe858..c1f351f 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 5de1db8..fd1c51b 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 65cc72e..f7b9cc2 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 517ca34..42d4fb4 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 2db245f..07aea8d 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), }; -- cgit v1.1 From 725c89997e03d71b09ea3c17c997da0712b9d835 Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin 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 Signed-off-by: Yevgeny Petrilin Signed-off-by: David S. Miller --- drivers/net/mlx4/en_ethtool.c | 15 ++++++++++++++- drivers/net/mlx4/main.c | 2 ++ drivers/net/mlx4/mlx4_en.h | 3 +++ 3 files changed, 19 insertions(+), 1 deletion(-) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/en_ethtool.c b/drivers/net/mlx4/en_ethtool.c index c1f351f..62ace6c 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 42d4fb4..5bebb88 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 07aea8d..ad4df66 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; -- cgit v1.1 From 0345584e0b8be3735a950d17c7e463db20c6ce27 Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin 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 Signed-off-by: David S. Miller --- 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 +++++++++++++++++++++++++++++++-------------- 5 files changed, 85 insertions(+), 36 deletions(-) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/en_main.c b/drivers/net/mlx4/en_main.c index 29aaa43..9317b61 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 fd1c51b..7813913 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 f7b9cc2..88003eb 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 5bebb88..da0da28 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 79cf42d..b103ad2 100644 --- a/drivers/net/mlx4/mcg.c +++ b/drivers/net/mlx4/mcg.c @@ -32,6 +32,7 @@ */ #include +#include #include @@ -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) -- cgit v1.1 From b12d93d63c3217f0ec923ff938b12a744e242ffa Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin Date: Tue, 22 Mar 2011 22:38:24 +0000 Subject: mlx4: Add support for promiscuous mode in the new steering model. For Ethernet mode only, When we want to register QP as promiscuous, it must be added to all the existing steering entries and also to the default one. The promiscuous QP might also be on of "real" QPs, which means we need to monitor every entry to avoid duplicates and ensure we close an entry when all it has is promiscuous QPs. Same mechanism both for unicast and multicast. Signed-off-by: Yevgeny Petrilin Signed-off-by: David S. Miller --- drivers/net/mlx4/main.c | 63 +++++- drivers/net/mlx4/mcg.c | 535 +++++++++++++++++++++++++++++++++++++++++++++++- drivers/net/mlx4/mlx4.h | 33 +++ 3 files changed, 620 insertions(+), 11 deletions(-) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c index da0da28..c8e2761 100644 --- a/drivers/net/mlx4/main.c +++ b/drivers/net/mlx4/main.c @@ -1063,6 +1063,59 @@ static void mlx4_cleanup_port_info(struct mlx4_port_info *info) device_remove_file(&info->dev->pdev->dev, &info->port_attr); } +static int mlx4_init_steering(struct mlx4_dev *dev) +{ + struct mlx4_priv *priv = mlx4_priv(dev); + int num_entries = dev->caps.num_ports; + int i, j; + + priv->steer = kzalloc(sizeof(struct mlx4_steer) * num_entries, GFP_KERNEL); + if (!priv->steer) + return -ENOMEM; + + for (i = 0; i < num_entries; i++) { + for (j = 0; j < MLX4_NUM_STEERS; j++) { + INIT_LIST_HEAD(&priv->steer[i].promisc_qps[j]); + INIT_LIST_HEAD(&priv->steer[i].steer_entries[j]); + } + INIT_LIST_HEAD(&priv->steer[i].high_prios); + } + return 0; +} + +static void mlx4_clear_steering(struct mlx4_dev *dev) +{ + struct mlx4_priv *priv = mlx4_priv(dev); + struct mlx4_steer_index *entry, *tmp_entry; + struct mlx4_promisc_qp *pqp, *tmp_pqp; + int num_entries = dev->caps.num_ports; + int i, j; + + for (i = 0; i < num_entries; i++) { + for (j = 0; j < MLX4_NUM_STEERS; j++) { + list_for_each_entry_safe(pqp, tmp_pqp, + &priv->steer[i].promisc_qps[j], + list) { + list_del(&pqp->list); + kfree(pqp); + } + list_for_each_entry_safe(entry, tmp_entry, + &priv->steer[i].steer_entries[j], + list) { + list_del(&entry->list); + list_for_each_entry_safe(pqp, tmp_pqp, + &entry->duplicates, + list) { + list_del(&pqp->list); + kfree(pqp); + } + kfree(entry); + } + } + } + kfree(priv->steer); +} + static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id) { struct mlx4_priv *priv; @@ -1172,6 +1225,10 @@ static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id) mlx4_enable_msi_x(dev); + err = mlx4_init_steering(dev); + if (err) + goto err_free_eq; + err = mlx4_setup_hca(dev); if (err == -EBUSY && (dev->flags & MLX4_FLAG_MSI_X)) { dev->flags &= ~MLX4_FLAG_MSI_X; @@ -1180,7 +1237,7 @@ static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id) } if (err) - goto err_free_eq; + goto err_steer; for (port = 1; port <= dev->caps.num_ports; port++) { err = mlx4_init_port_info(dev, port); @@ -1213,6 +1270,9 @@ err_port: mlx4_cleanup_pd_table(dev); mlx4_cleanup_uar_table(dev); +err_steer: + mlx4_clear_steering(dev); + err_free_eq: mlx4_free_eq_table(dev); @@ -1272,6 +1332,7 @@ static void mlx4_remove_one(struct pci_dev *pdev) iounmap(priv->kar); mlx4_uar_free(dev, &priv->driver_uar); mlx4_cleanup_uar_table(dev); + mlx4_clear_steering(dev); mlx4_free_eq_table(dev); mlx4_close_hca(dev); mlx4_cmd_cleanup(dev); diff --git a/drivers/net/mlx4/mcg.c b/drivers/net/mlx4/mcg.c index b103ad2..e71372a 100644 --- a/drivers/net/mlx4/mcg.c +++ b/drivers/net/mlx4/mcg.c @@ -41,14 +41,6 @@ #define MGM_QPN_MASK 0x00FFFFFF #define MGM_BLCK_LB_BIT 30 -struct mlx4_mgm { - __be32 next_gid_index; - __be32 members_count; - u32 reserved[2]; - u8 gid[16]; - __be32 qp[MLX4_QP_PER_MGM]; -}; - static const u8 zero_gid[16]; /* automatically initialized to 0 */ static int mlx4_READ_ENTRY(struct mlx4_dev *dev, int index, @@ -65,6 +57,16 @@ static int mlx4_WRITE_ENTRY(struct mlx4_dev *dev, int index, MLX4_CMD_TIME_CLASS_A); } +static int mlx4_WRITE_PROMISC(struct mlx4_dev *dev, u8 vep_num, u8 port, u8 steer, + struct mlx4_cmd_mailbox *mailbox) +{ + u32 in_mod; + + in_mod = (u32) vep_num << 24 | (u32) port << 16 | steer << 1; + return mlx4_cmd(dev, mailbox->dma, in_mod, 0x1, + MLX4_CMD_WRITE_MCG, MLX4_CMD_TIME_CLASS_A); +} + static int mlx4_GID_HASH(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox, u16 *hash, u8 op_mod) { @@ -80,6 +82,457 @@ static int mlx4_GID_HASH(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox, return err; } +static struct mlx4_promisc_qp *get_promisc_qp(struct mlx4_dev *dev, u8 pf_num, + enum mlx4_steer_type steer, + u32 qpn) +{ + struct mlx4_steer *s_steer = &mlx4_priv(dev)->steer[pf_num]; + struct mlx4_promisc_qp *pqp; + + list_for_each_entry(pqp, &s_steer->promisc_qps[steer], list) { + if (pqp->qpn == qpn) + return pqp; + } + /* not found */ + return NULL; +} + +/* + * Add new entry to steering data structure. + * All promisc QPs should be added as well + */ +static int new_steering_entry(struct mlx4_dev *dev, u8 vep_num, u8 port, + enum mlx4_steer_type steer, + unsigned int index, u32 qpn) +{ + struct mlx4_steer *s_steer; + struct mlx4_cmd_mailbox *mailbox; + struct mlx4_mgm *mgm; + u32 members_count; + struct mlx4_steer_index *new_entry; + struct mlx4_promisc_qp *pqp; + struct mlx4_promisc_qp *dqp; + u32 prot; + int err; + u8 pf_num; + + pf_num = (dev->caps.num_ports == 1) ? vep_num : (vep_num << 1) | (port - 1); + s_steer = &mlx4_priv(dev)->steer[pf_num]; + new_entry = kzalloc(sizeof *new_entry, GFP_KERNEL); + if (!new_entry) + return -ENOMEM; + + INIT_LIST_HEAD(&new_entry->duplicates); + new_entry->index = index; + list_add_tail(&new_entry->list, &s_steer->steer_entries[steer]); + + /* If the given qpn is also a promisc qp, + * it should be inserted to duplicates list + */ + pqp = get_promisc_qp(dev, pf_num, steer, qpn); + if (pqp) { + dqp = kmalloc(sizeof *dqp, GFP_KERNEL); + if (!dqp) { + err = -ENOMEM; + goto out_alloc; + } + dqp->qpn = qpn; + list_add_tail(&dqp->list, &new_entry->duplicates); + } + + /* if no promisc qps for this vep, we are done */ + if (list_empty(&s_steer->promisc_qps[steer])) + return 0; + + /* now need to add all the promisc qps to the new + * steering entry, as they should also receive the packets + * destined to this address */ + mailbox = mlx4_alloc_cmd_mailbox(dev); + if (IS_ERR(mailbox)) { + err = -ENOMEM; + goto out_alloc; + } + mgm = mailbox->buf; + + err = mlx4_READ_ENTRY(dev, index, mailbox); + if (err) + goto out_mailbox; + + members_count = be32_to_cpu(mgm->members_count) & 0xffffff; + prot = be32_to_cpu(mgm->members_count) >> 30; + list_for_each_entry(pqp, &s_steer->promisc_qps[steer], list) { + /* don't add already existing qpn */ + if (pqp->qpn == qpn) + continue; + if (members_count == MLX4_QP_PER_MGM) { + /* out of space */ + err = -ENOMEM; + goto out_mailbox; + } + + /* add the qpn */ + mgm->qp[members_count++] = cpu_to_be32(pqp->qpn & MGM_QPN_MASK); + } + /* update the qps count and update the entry with all the promisc qps*/ + mgm->members_count = cpu_to_be32(members_count | (prot << 30)); + err = mlx4_WRITE_ENTRY(dev, index, mailbox); + +out_mailbox: + mlx4_free_cmd_mailbox(dev, mailbox); + if (!err) + return 0; +out_alloc: + if (dqp) { + list_del(&dqp->list); + kfree(&dqp); + } + list_del(&new_entry->list); + kfree(new_entry); + return err; +} + +/* update the data structures with existing steering entry */ +static int existing_steering_entry(struct mlx4_dev *dev, u8 vep_num, u8 port, + enum mlx4_steer_type steer, + unsigned int index, u32 qpn) +{ + struct mlx4_steer *s_steer; + struct mlx4_steer_index *tmp_entry, *entry = NULL; + struct mlx4_promisc_qp *pqp; + struct mlx4_promisc_qp *dqp; + u8 pf_num; + + pf_num = (dev->caps.num_ports == 1) ? vep_num : (vep_num << 1) | (port - 1); + s_steer = &mlx4_priv(dev)->steer[pf_num]; + + pqp = get_promisc_qp(dev, pf_num, steer, qpn); + if (!pqp) + return 0; /* nothing to do */ + + list_for_each_entry(tmp_entry, &s_steer->steer_entries[steer], list) { + if (tmp_entry->index == index) { + entry = tmp_entry; + break; + } + } + if (unlikely(!entry)) { + mlx4_warn(dev, "Steering entry at index %x is not registered\n", index); + return -EINVAL; + } + + /* the given qpn is listed as a promisc qpn + * we need to add it as a duplicate to this entry + * for future refernce */ + list_for_each_entry(dqp, &entry->duplicates, list) { + if (qpn == dqp->qpn) + return 0; /* qp is already duplicated */ + } + + /* add the qp as a duplicate on this index */ + dqp = kmalloc(sizeof *dqp, GFP_KERNEL); + if (!dqp) + return -ENOMEM; + dqp->qpn = qpn; + list_add_tail(&dqp->list, &entry->duplicates); + + return 0; +} + +/* Check whether a qpn is a duplicate on steering entry + * If so, it should not be removed from mgm */ +static bool check_duplicate_entry(struct mlx4_dev *dev, u8 vep_num, u8 port, + enum mlx4_steer_type steer, + unsigned int index, u32 qpn) +{ + struct mlx4_steer *s_steer; + struct mlx4_steer_index *tmp_entry, *entry = NULL; + struct mlx4_promisc_qp *dqp, *tmp_dqp; + u8 pf_num; + + pf_num = (dev->caps.num_ports == 1) ? vep_num : (vep_num << 1) | (port - 1); + s_steer = &mlx4_priv(dev)->steer[pf_num]; + + /* if qp is not promisc, it cannot be duplicated */ + if (!get_promisc_qp(dev, pf_num, steer, qpn)) + return false; + + /* The qp is promisc qp so it is a duplicate on this index + * Find the index entry, and remove the duplicate */ + list_for_each_entry(tmp_entry, &s_steer->steer_entries[steer], list) { + if (tmp_entry->index == index) { + entry = tmp_entry; + break; + } + } + if (unlikely(!entry)) { + mlx4_warn(dev, "Steering entry for index %x is not registered\n", index); + return false; + } + list_for_each_entry_safe(dqp, tmp_dqp, &entry->duplicates, list) { + if (dqp->qpn == qpn) { + list_del(&dqp->list); + kfree(dqp); + } + } + return true; +} + +/* I a steering entry contains only promisc QPs, it can be removed. */ +static bool can_remove_steering_entry(struct mlx4_dev *dev, u8 vep_num, u8 port, + enum mlx4_steer_type steer, + unsigned int index, u32 tqpn) +{ + struct mlx4_steer *s_steer; + struct mlx4_cmd_mailbox *mailbox; + struct mlx4_mgm *mgm; + struct mlx4_steer_index *entry = NULL, *tmp_entry; + u32 qpn; + u32 members_count; + bool ret = false; + int i; + u8 pf_num; + + pf_num = (dev->caps.num_ports == 1) ? vep_num : (vep_num << 1) | (port - 1); + s_steer = &mlx4_priv(dev)->steer[pf_num]; + + mailbox = mlx4_alloc_cmd_mailbox(dev); + if (IS_ERR(mailbox)) + return false; + mgm = mailbox->buf; + + if (mlx4_READ_ENTRY(dev, index, mailbox)) + goto out; + members_count = be32_to_cpu(mgm->members_count) & 0xffffff; + for (i = 0; i < members_count; i++) { + qpn = be32_to_cpu(mgm->qp[i]) & MGM_QPN_MASK; + if (!get_promisc_qp(dev, pf_num, steer, qpn) && qpn != tqpn) { + /* the qp is not promisc, the entry can't be removed */ + goto out; + } + } + /* All the qps currently registered for this entry are promiscuous, + * Checking for duplicates */ + ret = true; + list_for_each_entry_safe(entry, tmp_entry, &s_steer->steer_entries[steer], list) { + if (entry->index == index) { + if (list_empty(&entry->duplicates)) { + list_del(&entry->list); + kfree(entry); + } else { + /* This entry contains duplicates so it shouldn't be removed */ + ret = false; + goto out; + } + } + } + +out: + mlx4_free_cmd_mailbox(dev, mailbox); + return ret; +} + +static int add_promisc_qp(struct mlx4_dev *dev, u8 vep_num, u8 port, + enum mlx4_steer_type steer, u32 qpn) +{ + struct mlx4_steer *s_steer; + struct mlx4_cmd_mailbox *mailbox; + struct mlx4_mgm *mgm; + struct mlx4_steer_index *entry; + struct mlx4_promisc_qp *pqp; + struct mlx4_promisc_qp *dqp; + u32 members_count; + u32 prot; + int i; + bool found; + int last_index; + int err; + u8 pf_num; + struct mlx4_priv *priv = mlx4_priv(dev); + pf_num = (dev->caps.num_ports == 1) ? vep_num : (vep_num << 1) | (port - 1); + s_steer = &mlx4_priv(dev)->steer[pf_num]; + + mutex_lock(&priv->mcg_table.mutex); + + if (get_promisc_qp(dev, pf_num, steer, qpn)) { + err = 0; /* Noting to do, already exists */ + goto out_mutex; + } + + pqp = kmalloc(sizeof *pqp, GFP_KERNEL); + if (!pqp) { + err = -ENOMEM; + goto out_mutex; + } + pqp->qpn = qpn; + + mailbox = mlx4_alloc_cmd_mailbox(dev); + if (IS_ERR(mailbox)) { + err = -ENOMEM; + goto out_alloc; + } + mgm = mailbox->buf; + + /* the promisc qp needs to be added for each one of the steering + * entries, if it already exists, needs to be added as a duplicate + * for this entry */ + list_for_each_entry(entry, &s_steer->steer_entries[steer], list) { + err = mlx4_READ_ENTRY(dev, entry->index, mailbox); + if (err) + goto out_mailbox; + + members_count = be32_to_cpu(mgm->members_count) & 0xffffff; + prot = be32_to_cpu(mgm->members_count) >> 30; + found = false; + for (i = 0; i < members_count; i++) { + if ((be32_to_cpu(mgm->qp[i]) & MGM_QPN_MASK) == qpn) { + /* Entry already exists, add to duplicates */ + dqp = kmalloc(sizeof *dqp, GFP_KERNEL); + if (!dqp) + goto out_mailbox; + dqp->qpn = qpn; + list_add_tail(&dqp->list, &entry->duplicates); + found = true; + } + } + if (!found) { + /* Need to add the qpn to mgm */ + if (members_count == MLX4_QP_PER_MGM) { + /* entry is full */ + err = -ENOMEM; + goto out_mailbox; + } + mgm->qp[members_count++] = cpu_to_be32(qpn & MGM_QPN_MASK); + mgm->members_count = cpu_to_be32(members_count | (prot << 30)); + err = mlx4_WRITE_ENTRY(dev, entry->index, mailbox); + if (err) + goto out_mailbox; + } + last_index = entry->index; + } + + /* add the new qpn to list of promisc qps */ + list_add_tail(&pqp->list, &s_steer->promisc_qps[steer]); + /* now need to add all the promisc qps to default entry */ + memset(mgm, 0, sizeof *mgm); + members_count = 0; + list_for_each_entry(dqp, &s_steer->promisc_qps[steer], list) + mgm->qp[members_count++] = cpu_to_be32(dqp->qpn & MGM_QPN_MASK); + mgm->members_count = cpu_to_be32(members_count | MLX4_PROT_ETH << 30); + + err = mlx4_WRITE_PROMISC(dev, vep_num, port, steer, mailbox); + if (err) + goto out_list; + + mlx4_free_cmd_mailbox(dev, mailbox); + mutex_unlock(&priv->mcg_table.mutex); + return 0; + +out_list: + list_del(&pqp->list); +out_mailbox: + mlx4_free_cmd_mailbox(dev, mailbox); +out_alloc: + kfree(pqp); +out_mutex: + mutex_unlock(&priv->mcg_table.mutex); + return err; +} + +static int remove_promisc_qp(struct mlx4_dev *dev, u8 vep_num, u8 port, + enum mlx4_steer_type steer, u32 qpn) +{ + struct mlx4_priv *priv = mlx4_priv(dev); + struct mlx4_steer *s_steer; + struct mlx4_cmd_mailbox *mailbox; + struct mlx4_mgm *mgm; + struct mlx4_steer_index *entry; + struct mlx4_promisc_qp *pqp; + struct mlx4_promisc_qp *dqp; + u32 members_count; + bool found; + bool back_to_list = false; + int loc, i; + int err; + u8 pf_num; + + pf_num = (dev->caps.num_ports == 1) ? vep_num : (vep_num << 1) | (port - 1); + s_steer = &mlx4_priv(dev)->steer[pf_num]; + mutex_lock(&priv->mcg_table.mutex); + + pqp = get_promisc_qp(dev, pf_num, steer, qpn); + if (unlikely(!pqp)) { + mlx4_warn(dev, "QP %x is not promiscuous QP\n", qpn); + /* nothing to do */ + err = 0; + goto out_mutex; + } + + /*remove from list of promisc qps */ + list_del(&pqp->list); + kfree(pqp); + + /* set the default entry not to include the removed one */ + mailbox = mlx4_alloc_cmd_mailbox(dev); + if (IS_ERR(mailbox)) { + err = -ENOMEM; + back_to_list = true; + goto out_list; + } + mgm = mailbox->buf; + members_count = 0; + list_for_each_entry(dqp, &s_steer->promisc_qps[steer], list) + mgm->qp[members_count++] = cpu_to_be32(dqp->qpn & MGM_QPN_MASK); + mgm->members_count = cpu_to_be32(members_count | MLX4_PROT_ETH << 30); + + err = mlx4_WRITE_PROMISC(dev, vep_num, port, steer, mailbox); + if (err) + goto out_mailbox; + + /* remove the qp from all the steering entries*/ + list_for_each_entry(entry, &s_steer->steer_entries[steer], list) { + found = false; + list_for_each_entry(dqp, &entry->duplicates, list) { + if (dqp->qpn == qpn) { + found = true; + break; + } + } + if (found) { + /* a duplicate, no need to change the mgm, + * only update the duplicates list */ + list_del(&dqp->list); + kfree(dqp); + } else { + err = mlx4_READ_ENTRY(dev, entry->index, mailbox); + if (err) + goto out_mailbox; + members_count = be32_to_cpu(mgm->members_count) & 0xffffff; + for (loc = -1, i = 0; i < members_count; ++i) + if ((be32_to_cpu(mgm->qp[i]) & MGM_QPN_MASK) == qpn) + loc = i; + + mgm->members_count = cpu_to_be32(--members_count | + (MLX4_PROT_ETH << 30)); + mgm->qp[loc] = mgm->qp[i - 1]; + mgm->qp[i - 1] = 0; + + err = mlx4_WRITE_ENTRY(dev, entry->index, mailbox); + if (err) + goto out_mailbox; + } + + } + +out_mailbox: + mlx4_free_cmd_mailbox(dev, mailbox); +out_list: + if (back_to_list) + list_add_tail(&pqp->list, &s_steer->promisc_qps[steer]); +out_mutex: + mutex_unlock(&priv->mcg_table.mutex); + return err; +} + /* * Caller must hold MCG table semaphore. gid and mgm parameters must * be properly aligned for command interface. @@ -164,6 +617,7 @@ int mlx4_qp_attach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16], int i; int err; u8 port = gid[5]; + u8 new_entry = 0; mailbox = mlx4_alloc_cmd_mailbox(dev); if (IS_ERR(mailbox)) @@ -177,8 +631,10 @@ int mlx4_qp_attach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16], goto out; if (index != -1) { - if (!(be32_to_cpu(mgm->members_count) & 0xffffff)) + if (!(be32_to_cpu(mgm->members_count) & 0xffffff)) { + new_entry = 1; memcpy(mgm->gid, gid, 16); + } } else { link = 1; @@ -234,6 +690,14 @@ int mlx4_qp_attach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16], goto out; out: + if (prot == MLX4_PROT_ETH) { + /* manage the steering entry for promisc mode */ + if (new_entry) + new_steering_entry(dev, 0, port, steer, index, qp->qpn); + else + existing_steering_entry(dev, 0, port, steer, + index, qp->qpn); + } if (err && link && index != -1) { if (index < dev->caps.num_mgms) mlx4_warn(dev, "Got AMGM index %d < %d", @@ -260,6 +724,7 @@ int mlx4_qp_detach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16], int i, loc; int err; u8 port = gid[5]; + bool removed_entry = false; mailbox = mlx4_alloc_cmd_mailbox(dev); if (IS_ERR(mailbox)) @@ -279,6 +744,11 @@ int mlx4_qp_detach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16], goto out; } + /* if this pq is also a promisc qp, it shouldn't be removed */ + if (prot == MLX4_PROT_ETH && + check_duplicate_entry(dev, 0, port, steer, index, qp->qpn)) + goto out; + members_count = be32_to_cpu(mgm->members_count) & 0xffffff; for (loc = -1, i = 0; i < members_count; ++i) if ((be32_to_cpu(mgm->qp[i]) & MGM_QPN_MASK) == qp->qpn) @@ -295,11 +765,16 @@ int mlx4_qp_detach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16], mgm->qp[loc] = mgm->qp[i - 1]; mgm->qp[i - 1] = 0; - if (i != 1) { + if (prot == MLX4_PROT_ETH) + removed_entry = can_remove_steering_entry(dev, 0, port, steer, index, qp->qpn); + if (i != 1 && (prot != MLX4_PROT_ETH || !removed_entry)) { err = mlx4_WRITE_ENTRY(dev, index, mailbox); goto out; } + /* We are going to delete the entry, members count should be 0 */ + mgm->members_count = cpu_to_be32((u32) prot << 30); + if (prev == -1) { /* Remove entry from MGM */ int amgm_index = be32_to_cpu(mgm->next_gid_index) >> 6; @@ -388,6 +863,46 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16], } EXPORT_SYMBOL_GPL(mlx4_multicast_detach); + +int mlx4_multicast_promisc_add(struct mlx4_dev *dev, u32 qpn, u8 port) +{ + if (!dev->caps.vep_mc_steering) + return 0; + + + return add_promisc_qp(dev, 0, port, MLX4_MC_STEER, qpn); +} +EXPORT_SYMBOL_GPL(mlx4_multicast_promisc_add); + +int mlx4_multicast_promisc_remove(struct mlx4_dev *dev, u32 qpn, u8 port) +{ + if (!dev->caps.vep_mc_steering) + return 0; + + + return remove_promisc_qp(dev, 0, port, MLX4_MC_STEER, qpn); +} +EXPORT_SYMBOL_GPL(mlx4_multicast_promisc_remove); + +int mlx4_unicast_promisc_add(struct mlx4_dev *dev, u32 qpn, u8 port) +{ + if (!dev->caps.vep_mc_steering) + return 0; + + + return add_promisc_qp(dev, 0, port, MLX4_UC_STEER, qpn); +} +EXPORT_SYMBOL_GPL(mlx4_unicast_promisc_add); + +int mlx4_unicast_promisc_remove(struct mlx4_dev *dev, u32 qpn, u8 port) +{ + if (!dev->caps.vep_mc_steering) + return 0; + + return remove_promisc_qp(dev, 0, port, MLX4_UC_STEER, qpn); +} +EXPORT_SYMBOL_GPL(mlx4_unicast_promisc_remove); + int mlx4_init_mcg_table(struct mlx4_dev *dev) { struct mlx4_priv *priv = mlx4_priv(dev); diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h index 67ee8da..1513a91 100644 --- a/drivers/net/mlx4/mlx4.h +++ b/drivers/net/mlx4/mlx4.h @@ -162,6 +162,27 @@ struct mlx4_fw { u8 catas_bar; }; +#define MGM_QPN_MASK 0x00FFFFFF +#define MGM_BLCK_LB_BIT 30 + +struct mlx4_promisc_qp { + struct list_head list; + u32 qpn; +}; + +struct mlx4_steer_index { + struct list_head list; + unsigned int index; + struct list_head duplicates; +}; + +struct mlx4_mgm { + __be32 next_gid_index; + __be32 members_count; + u32 reserved[2]; + u8 gid[16]; + __be32 qp[MLX4_QP_PER_MGM]; +}; struct mlx4_cmd { struct pci_pool *pool; void __iomem *hcr; @@ -287,6 +308,12 @@ struct mlx4_msix_ctl { spinlock_t pool_lock; }; +struct mlx4_steer { + struct list_head promisc_qps[MLX4_NUM_STEERS]; + struct list_head steer_entries[MLX4_NUM_STEERS]; + struct list_head high_prios; +}; + struct mlx4_priv { struct mlx4_dev dev; @@ -319,6 +346,7 @@ struct mlx4_priv { struct mlx4_sense sense; struct mutex port_mutex; struct mlx4_msix_ctl msix_ctl; + struct mlx4_steer *steer; }; static inline struct mlx4_priv *mlx4_priv(struct mlx4_dev *dev) @@ -409,4 +437,9 @@ void mlx4_init_vlan_table(struct mlx4_dev *dev, struct mlx4_vlan_table *table); int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port); int mlx4_get_port_ib_caps(struct mlx4_dev *dev, u8 port, __be32 *caps); +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); +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); #endif /* MLX4_H */ -- cgit v1.1 From 1679200f91da6a054b06954c9bd3eeed29b6731f Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin 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 Signed-off-by: David S. Miller --- 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 +++++++++++++++++++++++++++++++++++++++---- 8 files changed, 291 insertions(+), 58 deletions(-) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c index f6ed315..08e6801 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 a2b0bc4..f2a4f5d 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 092e814..e3d73e4 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 570f250..05998ee 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 7813913..67a209b 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 1513a91..d8bb441 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 ad4df66..5a2c560 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 4513395..eca7d85 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) { -- cgit v1.1 From c1b43dca137f2154845122417fba86d4bae67182 Mon Sep 17 00:00:00 2001 From: Eli Cohen 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 Signed-off-by: David S. Miller --- drivers/net/mlx4/main.c | 31 ++++++++++++++++ drivers/net/mlx4/mlx4.h | 3 ++ drivers/net/mlx4/pd.c | 94 +++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 128 insertions(+) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c index c8e2761..05c5671 100644 --- a/drivers/net/mlx4/main.c +++ b/drivers/net/mlx4/main.c @@ -39,6 +39,7 @@ #include #include #include +#include #include #include @@ -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 d8bb441..bc9a216 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 c4988d6..5210a0f 100644 --- a/drivers/net/mlx4/pd.c +++ b/drivers/net/mlx4/pd.c @@ -32,6 +32,7 @@ */ #include +#include #include @@ -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) { -- cgit v1.1 From 42d1e017e2f040c3e3d62b6cfa3a26351a865d04 Mon Sep 17 00:00:00 2001 From: Eli Cohen Date: Tue, 22 Mar 2011 22:38:45 +0000 Subject: mlx4_core: maintain available field in bitmap allocator Add mlx4_bitmap_avail() to give the number of available resources. We want to use this as a hint to whether to allocate a resources or not. This patch is introduced to be used with allocation blue flame registers. Signed-off-by: Eli Cohen Signed-off-by: David S. Miller --- drivers/net/mlx4/alloc.c | 13 +++++++++++++ drivers/net/mlx4/mlx4.h | 2 ++ 2 files changed, 15 insertions(+) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/alloc.c b/drivers/net/mlx4/alloc.c index 3a4277f..116cae3 100644 --- a/drivers/net/mlx4/alloc.c +++ b/drivers/net/mlx4/alloc.c @@ -62,6 +62,9 @@ u32 mlx4_bitmap_alloc(struct mlx4_bitmap *bitmap) } else obj = -1; + if (obj != -1) + --bitmap->avail; + spin_unlock(&bitmap->lock); return obj; @@ -101,11 +104,19 @@ u32 mlx4_bitmap_alloc_range(struct mlx4_bitmap *bitmap, int cnt, int align) } else obj = -1; + if (obj != -1) + bitmap->avail -= cnt; + spin_unlock(&bitmap->lock); return obj; } +u32 mlx4_bitmap_avail(struct mlx4_bitmap *bitmap) +{ + return bitmap->avail; +} + void mlx4_bitmap_free_range(struct mlx4_bitmap *bitmap, u32 obj, int cnt) { obj &= bitmap->max + bitmap->reserved_top - 1; @@ -115,6 +126,7 @@ void mlx4_bitmap_free_range(struct mlx4_bitmap *bitmap, u32 obj, int cnt) bitmap->last = min(bitmap->last, obj); bitmap->top = (bitmap->top + bitmap->max + bitmap->reserved_top) & bitmap->mask; + bitmap->avail += cnt; spin_unlock(&bitmap->lock); } @@ -130,6 +142,7 @@ int mlx4_bitmap_init(struct mlx4_bitmap *bitmap, u32 num, u32 mask, bitmap->max = num - reserved_top; bitmap->mask = mask; bitmap->reserved_top = reserved_top; + bitmap->avail = num - reserved_top - reserved_bot; spin_lock_init(&bitmap->lock); bitmap->table = kzalloc(BITS_TO_LONGS(bitmap->max) * sizeof (long), GFP_KERNEL); diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h index bc9a216..c1e0e5f 100644 --- a/drivers/net/mlx4/mlx4.h +++ b/drivers/net/mlx4/mlx4.h @@ -105,6 +105,7 @@ struct mlx4_bitmap { u32 max; u32 reserved_top; u32 mask; + u32 avail; spinlock_t lock; unsigned long *table; }; @@ -371,6 +372,7 @@ 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); +u32 mlx4_bitmap_avail(struct mlx4_bitmap *bitmap); 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); -- cgit v1.1 From 9ace5e0176f165f04537497143506621d33a0139 Mon Sep 17 00:00:00 2001 From: Eli Cohen Date: Tue, 22 Mar 2011 22:38:48 +0000 Subject: mlx4_core: reserve UARs for userspace consumers Do not allow a kernel consumer to allocate a UAR to serve for blue flame if the number of available UARs gets below MLX4_NUM_RESERVED_UARS (currently 8). This will allow userspace apps to open a device file and run things like ibv_devinfo. Signed-off-by: Eli Cohen Signed-off-by: David S. Miller --- drivers/net/mlx4/pd.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/pd.c b/drivers/net/mlx4/pd.c index 5210a0f..1286b88 100644 --- a/drivers/net/mlx4/pd.c +++ b/drivers/net/mlx4/pd.c @@ -39,6 +39,10 @@ #include "mlx4.h" #include "icm.h" +enum { + MLX4_NUM_RESERVED_UARS = 8 +}; + int mlx4_pd_alloc(struct mlx4_dev *dev, u32 *pdn) { struct mlx4_priv *priv = mlx4_priv(dev); @@ -104,6 +108,10 @@ int mlx4_bf_alloc(struct mlx4_dev *dev, struct mlx4_bf *bf) if (!list_empty(&priv->bf_list)) uar = list_entry(priv->bf_list.next, struct mlx4_uar, bf_list); else { + if (mlx4_bitmap_avail(&priv->uar_table.bitmap) < MLX4_NUM_RESERVED_UARS) { + err = -ENOMEM; + goto out; + } uar = kmalloc(sizeof *uar, GFP_KERNEL); if (!uar) { err = -ENOMEM; -- cgit v1.1 From 87a5c3896f9cf3b7f374af95b342d1089989327a Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin Date: Tue, 22 Mar 2011 22:38:52 +0000 Subject: mlx4_en: Using blue flame support Doorbell is used according to usage of BlueFlame. For Blue Flame to work in Ethernet mode QP number should have 0 at bits 6,7. Allocating range of QPs accordingly. Signed-off-by: Yevgeny Petrilin Signed-off-by: David S. Miller --- drivers/net/mlx4/en_netdev.c | 10 +++++- drivers/net/mlx4/en_tx.c | 72 ++++++++++++++++++++++++++++++++------------ drivers/net/mlx4/mlx4_en.h | 4 ++- 3 files changed, 64 insertions(+), 22 deletions(-) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c index 08e6801..5762ebd 100644 --- a/drivers/net/mlx4/en_netdev.c +++ b/drivers/net/mlx4/en_netdev.c @@ -931,6 +931,13 @@ int mlx4_en_alloc_resources(struct mlx4_en_priv *priv) { struct mlx4_en_port_profile *prof = priv->prof; int i; + int base_tx_qpn, err; + + err = mlx4_qp_reserve_range(priv->mdev->dev, priv->tx_ring_num, 256, &base_tx_qpn); + if (err) { + en_err(priv, "failed reserving range for TX rings\n"); + return err; + } /* Create tx Rings */ for (i = 0; i < priv->tx_ring_num; i++) { @@ -938,7 +945,7 @@ int mlx4_en_alloc_resources(struct mlx4_en_priv *priv) prof->tx_ring_size, i, TX)) goto err; - if (mlx4_en_create_tx_ring(priv, &priv->tx_ring[i], + if (mlx4_en_create_tx_ring(priv, &priv->tx_ring[i], base_tx_qpn + i, prof->tx_ring_size, TXBB_SIZE)) goto err; } @@ -958,6 +965,7 @@ int mlx4_en_alloc_resources(struct mlx4_en_priv *priv) err: en_err(priv, "Failed to allocate NIC resources\n"); + mlx4_qp_release_range(priv->mdev->dev, base_tx_qpn, priv->tx_ring_num); return -ENOMEM; } diff --git a/drivers/net/mlx4/en_tx.c b/drivers/net/mlx4/en_tx.c index a680cd4..01feb8f 100644 --- a/drivers/net/mlx4/en_tx.c +++ b/drivers/net/mlx4/en_tx.c @@ -44,6 +44,7 @@ enum { MAX_INLINE = 104, /* 128 - 16 - 4 - 4 */ + MAX_BF = 256, }; static int inline_thold __read_mostly = MAX_INLINE; @@ -52,7 +53,7 @@ module_param_named(inline_thold, inline_thold, int, 0444); MODULE_PARM_DESC(inline_thold, "threshold for using inline data"); int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv, - struct mlx4_en_tx_ring *ring, u32 size, + struct mlx4_en_tx_ring *ring, int qpn, u32 size, u16 stride) { struct mlx4_en_dev *mdev = priv->mdev; @@ -103,23 +104,25 @@ int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv, "buf_size:%d dma:%llx\n", ring, ring->buf, ring->size, ring->buf_size, (unsigned long long) ring->wqres.buf.direct.map); - err = mlx4_qp_reserve_range(mdev->dev, 1, 1, &ring->qpn); - if (err) { - en_err(priv, "Failed reserving qp for tx ring.\n"); - goto err_map; - } - + ring->qpn = qpn; err = mlx4_qp_alloc(mdev->dev, ring->qpn, &ring->qp); if (err) { en_err(priv, "Failed allocating qp %d\n", ring->qpn); - goto err_reserve; + goto err_map; } ring->qp.event = mlx4_en_sqp_event; + err = mlx4_bf_alloc(mdev->dev, &ring->bf); + if (err) { + en_dbg(DRV, priv, "working without blueflame (%d)", err); + ring->bf.uar = &mdev->priv_uar; + ring->bf.uar->map = mdev->uar_map; + ring->bf_enabled = false; + } else + ring->bf_enabled = true; + return 0; -err_reserve: - mlx4_qp_release_range(mdev->dev, ring->qpn, 1); err_map: mlx4_en_unmap_buffer(&ring->wqres.buf); err_hwq_res: @@ -139,6 +142,8 @@ void mlx4_en_destroy_tx_ring(struct mlx4_en_priv *priv, struct mlx4_en_dev *mdev = priv->mdev; en_dbg(DRV, priv, "Destroying tx ring, qpn: %d\n", ring->qpn); + if (ring->bf_enabled) + mlx4_bf_free(mdev->dev, &ring->bf); mlx4_qp_remove(mdev->dev, &ring->qp); mlx4_qp_free(mdev->dev, &ring->qp); mlx4_qp_release_range(mdev->dev, ring->qpn, 1); @@ -171,6 +176,8 @@ int mlx4_en_activate_tx_ring(struct mlx4_en_priv *priv, mlx4_en_fill_qp_context(priv, ring->size, ring->stride, 1, 0, ring->qpn, ring->cqn, &ring->context); + if (ring->bf_enabled) + ring->context.usr_page = cpu_to_be32(ring->bf.uar->index); err = mlx4_qp_to_ready(mdev->dev, &ring->wqres.mtt, &ring->context, &ring->qp, &ring->qp_state); @@ -591,6 +598,11 @@ u16 mlx4_en_select_queue(struct net_device *dev, struct sk_buff *skb) return skb_tx_hash(dev, skb); } +static void mlx4_bf_copy(unsigned long *dst, unsigned long *src, unsigned bytecnt) +{ + __iowrite64_copy(dst, src, bytecnt / 8); +} + netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) { struct mlx4_en_priv *priv = netdev_priv(dev); @@ -609,12 +621,13 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) int desc_size; int real_size; dma_addr_t dma; - u32 index; + u32 index, bf_index; __be32 op_own; u16 vlan_tag = 0; int i; int lso_header_size; void *fragptr; + bool bounce = false; if (!priv->port_up) goto tx_drop; @@ -657,13 +670,16 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) /* Packet is good - grab an index and transmit it */ index = ring->prod & ring->size_mask; + bf_index = ring->prod; /* See if we have enough space for whole descriptor TXBB for setting * SW ownership on next descriptor; if not, use a bounce buffer. */ if (likely(index + nr_txbb <= ring->size)) tx_desc = ring->buf + index * TXBB_SIZE; - else + else { tx_desc = (struct mlx4_en_tx_desc *) ring->bounce_buf; + bounce = true; + } /* Save skb in tx_info ring */ tx_info = &ring->tx_info[index]; @@ -768,21 +784,37 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) ring->prod += nr_txbb; /* If we used a bounce buffer then copy descriptor back into place */ - if (tx_desc == (struct mlx4_en_tx_desc *) ring->bounce_buf) + if (bounce) tx_desc = mlx4_en_bounce_to_desc(priv, ring, index, desc_size); /* Run destructor before passing skb to HW */ if (likely(!skb_shared(skb))) skb_orphan(skb); - /* Ensure new descirptor hits memory - * before setting ownership of this descriptor to HW */ - wmb(); - tx_desc->ctrl.owner_opcode = op_own; + if (ring->bf_enabled && desc_size <= MAX_BF && !bounce && !vlan_tag) { + *(u32 *) (&tx_desc->ctrl.vlan_tag) |= ring->doorbell_qpn; + op_own |= htonl((bf_index & 0xffff) << 8); + /* Ensure new descirptor hits memory + * before setting ownership of this descriptor to HW */ + wmb(); + tx_desc->ctrl.owner_opcode = op_own; - /* Ring doorbell! */ - wmb(); - writel(ring->doorbell_qpn, mdev->uar_map + MLX4_SEND_DOORBELL); + wmb(); + + mlx4_bf_copy(ring->bf.reg + ring->bf.offset, (unsigned long *) &tx_desc->ctrl, + desc_size); + + wmb(); + + ring->bf.offset ^= ring->bf.buf_size; + } else { + /* Ensure new descirptor hits memory + * before setting ownership of this descriptor to HW */ + wmb(); + tx_desc->ctrl.owner_opcode = op_own; + wmb(); + writel(ring->doorbell_qpn, ring->bf.uar->map + MLX4_SEND_DOORBELL); + } /* Poll CQ here */ mlx4_en_xmit_poll(priv, tx_ind); diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h index 5a2c560..edcb535 100644 --- a/drivers/net/mlx4/mlx4_en.h +++ b/drivers/net/mlx4/mlx4_en.h @@ -248,6 +248,8 @@ struct mlx4_en_tx_ring { unsigned long bytes; unsigned long packets; spinlock_t comp_lock; + struct mlx4_bf bf; + bool bf_enabled; }; struct mlx4_en_rx_desc { @@ -518,7 +520,7 @@ u16 mlx4_en_select_queue(struct net_device *dev, struct sk_buff *skb); netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev); int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv, struct mlx4_en_tx_ring *ring, - u32 size, u16 stride); + int qpn, u32 size, u16 stride); void mlx4_en_destroy_tx_ring(struct mlx4_en_priv *priv, struct mlx4_en_tx_ring *ring); int mlx4_en_activate_tx_ring(struct mlx4_en_priv *priv, struct mlx4_en_tx_ring *ring, -- cgit v1.1 From 61b85bf6060bfd205c218f5aeb71c19a24abb0cd Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin Date: Tue, 22 Mar 2011 22:39:05 +0000 Subject: mlx4_en: updated driver version to 1.5.4.1 Signed-off-by: Yevgeny Petrilin Signed-off-by: David S. Miller --- drivers/net/mlx4/mlx4_en.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h index edcb535..e30f609 100644 --- a/drivers/net/mlx4/mlx4_en.h +++ b/drivers/net/mlx4/mlx4_en.h @@ -49,8 +49,8 @@ #include "en_port.h" #define DRV_NAME "mlx4_en" -#define DRV_VERSION "1.5.1.6" -#define DRV_RELDATE "August 2010" +#define DRV_VERSION "1.5.4.1" +#define DRV_RELDATE "March 2011" #define MLX4_EN_MSG_LEVEL (NETIF_MSG_LINK | NETIF_MSG_IFDOWN) -- cgit v1.1 From 6f71d7927c9a9efb7d7e9e6d227d5e246a9c5bd7 Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin Date: Thu, 24 Mar 2011 11:47:51 -0700 Subject: mlx4_en: Removing HW info from ethtool -i report. Avoiding abuse of ethtool_drvinfo.driver field. HW specific info can be retrieved using lspci. Signed-off-by: Yevgeny Petrilin Signed-off-by: David S. Miller --- drivers/net/mlx4/en_ethtool.c | 15 +-------------- 1 file changed, 1 insertion(+), 14 deletions(-) (limited to 'drivers/net/mlx4') diff --git a/drivers/net/mlx4/en_ethtool.c b/drivers/net/mlx4/en_ethtool.c index 62ace6c..d54b7ab 100644 --- a/drivers/net/mlx4/en_ethtool.c +++ b/drivers/net/mlx4/en_ethtool.c @@ -45,20 +45,7 @@ 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; - 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->driver, DRV_NAME, 32); strncpy(drvinfo->version, DRV_VERSION " (" DRV_RELDATE ")", 32); sprintf(drvinfo->fw_version, "%d.%d.%d", (u16) (mdev->dev->caps.fw_ver >> 32), -- cgit v1.1