* [PATCH 10/17] mlx4: Add support for promiscuous mode in the new steering model.
From: Yevgeny Petrilin @ 2011-03-07 10:17 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp, vlad
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: Vladimir Sokolovsky <vlad@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
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(-)
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 */
--
1.6.0.2
^ permalink raw reply related
* [PATCH 15/17] mlx4_core: reserve UARs for userspace consumers
From: Yevgeny Petrilin @ 2011-03-07 10:18 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp, eli
From: Eli Cohen <eli@mellanox.co.il>
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 <eli@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
drivers/net/mlx4/pd.c | 8 ++++++++
1 files changed, 8 insertions(+), 0 deletions(-)
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;
--
1.6.0.2
^ permalink raw reply related
* Re: [patch net-next-2.6] net: reinject arps into bonding slave instead of master
From: Andy Gospodarek @ 2011-03-07 14:32 UTC (permalink / raw)
To: Jiri Pirko
Cc: Nicolas de Pesloüan, netdev, davem, shemminger, kaber, fubar,
eric.dumazet, andy
In-Reply-To: <20110307125059.GA6053@psychotron.brq.redhat.com>
On Mon, Mar 07, 2011 at 01:51:00PM +0100, Jiri Pirko wrote:
> Recent patch "bonding: move processing of recv handlers into
> handle_frame()" caused a regression on following net scheme:
>
> eth0 - bond0 - bond0.5
>
> where arp monitoring is happening over vlan. This patch fixes it by
> reinjecting the arp packet into bonding slave device so the bonding
> rx_handler can pickup and process it.
>
> Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Thanks, Jiri. I will test and make sure it works correctly now. I know
several users who would be quite disappointed if this feature was
removed.
^ permalink raw reply
* [PATCH 01/17] mlx4_en: going out of range of TX rings when reporting stats
From: Yevgeny Petrilin @ 2011-03-07 10:16 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
drivers/net/mlx4/en_port.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
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;
}
--
1.6.0.2
^ permalink raw reply related
* [PATCH 17/17] mlx4_en: updated driver version to 1.5.4.1
From: Yevgeny Petrilin @ 2011-03-07 10:18 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
drivers/net/mlx4/mlx4_en.h | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
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)
--
1.6.0.2
^ permalink raw reply related
* [PATCH 09/17] mlx4: generalization of multicast steering.
From: Yevgeny Petrilin @ 2011-03-07 10:17 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp, vlad
>From 56d6974c830f34ab0652be3f46607df25f12e2bd Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Thu, 3 Mar 2011 11:45:34 +0200
Subject: [PATCH] mlx4: generalization of multicast steering.
The same packet steering mechanism would be used both for IB and Ethernet,
Both multicats and unicasts.
This commit prepares the general infrastructure for this.
Signed-off-by: Vladimir Sokolovsky <vlad@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
drivers/infiniband/hw/mlx4/main.c | 10 ++--
drivers/net/mlx4/en_main.c | 2 +-
drivers/net/mlx4/fw.c | 2 +
drivers/net/mlx4/fw.h | 2 +
drivers/net/mlx4/main.c | 2 +
drivers/net/mlx4/mcg.c | 113 +++++++++++++++++++++++++-----------
include/linux/mlx4/device.h | 14 ++++-
7 files changed, 102 insertions(+), 43 deletions(-)
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index 66e3eec..fbe1973 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -625,7 +625,7 @@ static int mlx4_ib_mcg_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
err = mlx4_multicast_attach(mdev->dev, &mqp->mqp, gid->raw,
!!(mqp->flags & MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK),
- MLX4_PROTOCOL_IB);
+ MLX4_PROT_IB_IPV6);
if (err)
return err;
@@ -636,7 +636,7 @@ static int mlx4_ib_mcg_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
return 0;
err_add:
- mlx4_multicast_detach(mdev->dev, &mqp->mqp, gid->raw, MLX4_PROTOCOL_IB);
+ mlx4_multicast_detach(mdev->dev, &mqp->mqp, gid->raw, MLX4_PROT_IB_IPV6);
return err;
}
@@ -666,7 +666,7 @@ static int mlx4_ib_mcg_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
struct mlx4_ib_gid_entry *ge;
err = mlx4_multicast_detach(mdev->dev,
- &mqp->mqp, gid->raw, MLX4_PROTOCOL_IB);
+ &mqp->mqp, gid->raw, MLX4_PROT_IB_IPV6);
if (err)
return err;
@@ -953,7 +953,7 @@ static int mlx4_ib_netdev_event(struct notifier_block *this, unsigned long event
mlx4_foreach_ib_transport_port(port, ibdev->dev) {
oldnd = iboe->netdevs[port - 1];
iboe->netdevs[port - 1] =
- mlx4_get_protocol_dev(ibdev->dev, MLX4_PROTOCOL_EN, port);
+ mlx4_get_protocol_dev(ibdev->dev, MLX4_PROT_ETH, port);
if (oldnd != iboe->netdevs[port - 1]) {
if (iboe->netdevs[port - 1])
netdev_added(ibdev, port);
@@ -1206,7 +1206,7 @@ static struct mlx4_interface mlx4_ib_interface = {
.add = mlx4_ib_add,
.remove = mlx4_ib_remove,
.event = mlx4_ib_event,
- .protocol = MLX4_PROTOCOL_IB
+ .protocol = MLX4_PROT_IB_IPV6
};
static int __init mlx4_ib_init(void)
diff --git a/drivers/net/mlx4/en_main.c b/drivers/net/mlx4/en_main.c
index 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 <linux/string.h>
+#include <linux/etherdevice.h>
#include <linux/mlx4/cmd.h>
@@ -50,28 +51,28 @@ struct mlx4_mgm {
static const u8 zero_gid[16]; /* automatically initialized to 0 */
-static int mlx4_READ_MCG(struct mlx4_dev *dev, int index,
- struct mlx4_cmd_mailbox *mailbox)
+static int mlx4_READ_ENTRY(struct mlx4_dev *dev, int index,
+ struct mlx4_cmd_mailbox *mailbox)
{
return mlx4_cmd_box(dev, 0, mailbox->dma, index, 0, MLX4_CMD_READ_MCG,
MLX4_CMD_TIME_CLASS_A);
}
-static int mlx4_WRITE_MCG(struct mlx4_dev *dev, int index,
- struct mlx4_cmd_mailbox *mailbox)
+static int mlx4_WRITE_ENTRY(struct mlx4_dev *dev, int index,
+ struct mlx4_cmd_mailbox *mailbox)
{
return mlx4_cmd(dev, mailbox->dma, index, 0, MLX4_CMD_WRITE_MCG,
MLX4_CMD_TIME_CLASS_A);
}
-static int mlx4_MGID_HASH(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox,
- u16 *hash)
+static int mlx4_GID_HASH(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox,
+ u16 *hash, u8 op_mod)
{
u64 imm;
int err;
- err = mlx4_cmd_imm(dev, mailbox->dma, &imm, 0, 0, MLX4_CMD_MGID_HASH,
- MLX4_CMD_TIME_CLASS_A);
+ err = mlx4_cmd_imm(dev, mailbox->dma, &imm, 0, op_mod,
+ MLX4_CMD_MGID_HASH, MLX4_CMD_TIME_CLASS_A);
if (!err)
*hash = imm;
@@ -94,15 +95,17 @@ static int mlx4_MGID_HASH(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox
* If no AMGM exists for given gid, *index = -1, *prev = index of last
* entry in hash chain and *mgm holds end of hash chain.
*/
-static int find_mgm(struct mlx4_dev *dev,
- u8 *gid, enum mlx4_protocol protocol,
- struct mlx4_cmd_mailbox *mgm_mailbox,
- u16 *hash, int *prev, int *index)
+static int find_entry(struct mlx4_dev *dev, u8 port,
+ u8 *gid, enum mlx4_protocol prot,
+ enum mlx4_steer_type steer,
+ struct mlx4_cmd_mailbox *mgm_mailbox,
+ u16 *hash, int *prev, int *index)
{
struct mlx4_cmd_mailbox *mailbox;
struct mlx4_mgm *mgm = mgm_mailbox->buf;
u8 *mgid;
int err;
+ u8 op_mod = (prot == MLX4_PROT_ETH) ? !!(dev->caps.vep_mc_steering) : 0;
mailbox = mlx4_alloc_cmd_mailbox(dev);
if (IS_ERR(mailbox))
@@ -111,7 +114,7 @@ static int find_mgm(struct mlx4_dev *dev,
memcpy(mgid, gid, 16);
- err = mlx4_MGID_HASH(dev, mailbox, hash);
+ err = mlx4_GID_HASH(dev, mailbox, hash, op_mod);
mlx4_free_cmd_mailbox(dev, mailbox);
if (err)
return err;
@@ -123,11 +126,11 @@ static int find_mgm(struct mlx4_dev *dev,
*prev = -1;
do {
- err = mlx4_READ_MCG(dev, *index, mgm_mailbox);
+ err = mlx4_READ_ENTRY(dev, *index, mgm_mailbox);
if (err)
return err;
- if (!memcmp(mgm->gid, zero_gid, 16)) {
+ if (!(be32_to_cpu(mgm->members_count) & 0xffffff)) {
if (*index != *hash) {
mlx4_err(dev, "Found zero MGID in AMGM.\n");
err = -EINVAL;
@@ -136,7 +139,7 @@ static int find_mgm(struct mlx4_dev *dev,
}
if (!memcmp(mgm->gid, gid, 16) &&
- be32_to_cpu(mgm->members_count) >> 30 == protocol)
+ be32_to_cpu(mgm->members_count) >> 30 == prot)
return err;
*prev = *index;
@@ -147,8 +150,9 @@ static int find_mgm(struct mlx4_dev *dev,
return err;
}
-int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
- int block_mcast_loopback, enum mlx4_protocol protocol)
+int mlx4_qp_attach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
+ int block_mcast_loopback, enum mlx4_protocol prot,
+ enum mlx4_steer_type steer)
{
struct mlx4_priv *priv = mlx4_priv(dev);
struct mlx4_cmd_mailbox *mailbox;
@@ -159,6 +163,7 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
int link = 0;
int i;
int err;
+ u8 port = gid[5];
mailbox = mlx4_alloc_cmd_mailbox(dev);
if (IS_ERR(mailbox))
@@ -166,13 +171,13 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
mgm = mailbox->buf;
mutex_lock(&priv->mcg_table.mutex);
-
- err = find_mgm(dev, gid, protocol, mailbox, &hash, &prev, &index);
+ err = find_entry(dev, port, gid, prot, steer,
+ mailbox, &hash, &prev, &index);
if (err)
goto out;
if (index != -1) {
- if (!memcmp(mgm->gid, zero_gid, 16))
+ if (!(be32_to_cpu(mgm->members_count) & 0xffffff))
memcpy(mgm->gid, gid, 16);
} else {
link = 1;
@@ -209,22 +214,22 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
else
mgm->qp[members_count++] = cpu_to_be32(qp->qpn & MGM_QPN_MASK);
- mgm->members_count = cpu_to_be32(members_count | (u32) protocol << 30);
+ mgm->members_count = cpu_to_be32(members_count | (u32) prot << 30);
- err = mlx4_WRITE_MCG(dev, index, mailbox);
+ err = mlx4_WRITE_ENTRY(dev, index, mailbox);
if (err)
goto out;
if (!link)
goto out;
- err = mlx4_READ_MCG(dev, prev, mailbox);
+ err = mlx4_READ_ENTRY(dev, prev, mailbox);
if (err)
goto out;
mgm->next_gid_index = cpu_to_be32(index << 6);
- err = mlx4_WRITE_MCG(dev, prev, mailbox);
+ err = mlx4_WRITE_ENTRY(dev, prev, mailbox);
if (err)
goto out;
@@ -242,10 +247,9 @@ out:
mlx4_free_cmd_mailbox(dev, mailbox);
return err;
}
-EXPORT_SYMBOL_GPL(mlx4_multicast_attach);
-int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
- enum mlx4_protocol protocol)
+int mlx4_qp_detach_common(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
+ enum mlx4_protocol prot, enum mlx4_steer_type steer)
{
struct mlx4_priv *priv = mlx4_priv(dev);
struct mlx4_cmd_mailbox *mailbox;
@@ -255,6 +259,7 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
int prev, index;
int i, loc;
int err;
+ u8 port = gid[5];
mailbox = mlx4_alloc_cmd_mailbox(dev);
if (IS_ERR(mailbox))
@@ -263,7 +268,8 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
mutex_lock(&priv->mcg_table.mutex);
- err = find_mgm(dev, gid, protocol, mailbox, &hash, &prev, &index);
+ err = find_entry(dev, port, gid, prot, steer,
+ mailbox, &hash, &prev, &index);
if (err)
goto out;
@@ -285,12 +291,12 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
}
- mgm->members_count = cpu_to_be32(--members_count | (u32) protocol << 30);
+ mgm->members_count = cpu_to_be32(--members_count | (u32) prot << 30);
mgm->qp[loc] = mgm->qp[i - 1];
mgm->qp[i - 1] = 0;
if (i != 1) {
- err = mlx4_WRITE_MCG(dev, index, mailbox);
+ err = mlx4_WRITE_ENTRY(dev, index, mailbox);
goto out;
}
@@ -298,13 +304,13 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
/* Remove entry from MGM */
int amgm_index = be32_to_cpu(mgm->next_gid_index) >> 6;
if (amgm_index) {
- err = mlx4_READ_MCG(dev, amgm_index, mailbox);
+ err = mlx4_READ_ENTRY(dev, amgm_index, mailbox);
if (err)
goto out;
} else
memset(mgm->gid, 0, 16);
- err = mlx4_WRITE_MCG(dev, index, mailbox);
+ err = mlx4_WRITE_ENTRY(dev, index, mailbox);
if (err)
goto out;
@@ -319,13 +325,13 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
} else {
/* Remove entry from AMGM */
int cur_next_index = be32_to_cpu(mgm->next_gid_index) >> 6;
- err = mlx4_READ_MCG(dev, prev, mailbox);
+ err = mlx4_READ_ENTRY(dev, prev, mailbox);
if (err)
goto out;
mgm->next_gid_index = cpu_to_be32(cur_next_index << 6);
- err = mlx4_WRITE_MCG(dev, prev, mailbox);
+ err = mlx4_WRITE_ENTRY(dev, prev, mailbox);
if (err)
goto out;
@@ -343,6 +349,43 @@ out:
mlx4_free_cmd_mailbox(dev, mailbox);
return err;
}
+
+
+int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
+ int block_mcast_loopback, enum mlx4_protocol prot)
+{
+ enum mlx4_steer_type steer;
+
+ steer = (is_valid_ether_addr(&gid[10])) ? MLX4_UC_STEER : MLX4_MC_STEER;
+
+ if (prot == MLX4_PROT_ETH && !dev->caps.vep_mc_steering)
+ return 0;
+
+ if (prot == MLX4_PROT_ETH)
+ gid[7] |= (steer << 1);
+
+ return mlx4_qp_attach_common(dev, qp, gid,
+ block_mcast_loopback, prot,
+ steer);
+}
+EXPORT_SYMBOL_GPL(mlx4_multicast_attach);
+
+int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
+ enum mlx4_protocol prot)
+{
+ enum mlx4_steer_type steer;
+
+ steer = (is_valid_ether_addr(&gid[10])) ? MLX4_UC_STEER : MLX4_MC_STEER;
+
+ if (prot == MLX4_PROT_ETH && !dev->caps.vep_mc_steering)
+ return 0;
+
+ if (prot == MLX4_PROT_ETH) {
+ gid[7] |= (steer << 1);
+ }
+
+ return mlx4_qp_detach_common(dev, qp, gid, prot, steer);
+}
EXPORT_SYMBOL_GPL(mlx4_multicast_detach);
int mlx4_init_mcg_table(struct mlx4_dev *dev)
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index fe2a3a3..ebda939 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -150,8 +150,10 @@ enum {
};
enum mlx4_protocol {
- MLX4_PROTOCOL_IB,
- MLX4_PROTOCOL_EN,
+ MLX4_PROT_IB_IPV6 = 0,
+ MLX4_PROT_ETH,
+ MLX4_PROT_IB_IPV4,
+ MLX4_PROT_FCOE
};
enum {
@@ -178,6 +180,12 @@ enum mlx4_special_vlan_idx {
MLX4_VLAN_REGULAR
};
+enum mlx4_steer_type {
+ MLX4_MC_STEER = 0,
+ MLX4_UC_STEER,
+ MLX4_NUM_STEERS
+};
+
enum {
MLX4_NUM_FEXCH = 64 * 1024,
};
@@ -251,6 +259,8 @@ struct mlx4_caps {
u16 stat_rate_support;
int udp_rss;
int loopback_support;
+ int vep_uc_steering;
+ int vep_mc_steering;
int wol;
u8 port_width_cap[MLX4_MAX_PORTS + 1];
int max_gso_sz;
--
1.6.0.2
^ permalink raw reply related
* [PATCH 05/17] mlx4: Changing interrupt scheme
From: Yevgeny Petrilin @ 2011-03-07 10:17 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp, markuze
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 only
on number of cores.
The new functionality is exposed through mlx4_assign_eq() and mlx4_release_eq().
Customers that do not use the new API will get completion vectors as before.
Signed-off-by: Markuze Alex <markuze@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
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 +-
include/linux/mlx4/device.h | 8 +++
6 files changed, 134 insertions(+), 13 deletions(-)
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 0da5bb7..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;
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 0492146..7838082 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -39,6 +39,11 @@
#include <asm/atomic.h>
+#define MAX_MSIX_P_PORT 17
+#define MAX_MSIX 64
+#define MSIX_LEGACY_SZ 4
+#define MIN_MSIX_P_PORT 5
+
enum {
MLX4_FLAG_MSI_X = 1 << 0,
MLX4_FLAG_OLD_PORT_CMDS = 1 << 1,
@@ -223,6 +228,7 @@ struct mlx4_caps {
int num_eqs;
int reserved_eqs;
int num_comp_vectors;
+ int comp_pool;
int num_mpts;
int num_mtt_segs;
int mtts_per_seg;
@@ -526,5 +532,7 @@ void mlx4_fmr_unmap(struct mlx4_dev *dev, struct mlx4_fmr *fmr,
int mlx4_fmr_free(struct mlx4_dev *dev, struct mlx4_fmr *fmr);
int mlx4_SYNC_TPT(struct mlx4_dev *dev);
int mlx4_test_interrupts(struct mlx4_dev *dev);
+int mlx4_assign_eq(struct mlx4_dev *dev, char* name , int* vector);
+void mlx4_release_eq(struct mlx4_dev *dev, int vec);
#endif /* MLX4_DEVICE_H */
--
1.6.0.2
^ permalink raw reply related
* [PATCH 11/17] mlx4_en: Enabling new steering
From: Yevgeny Petrilin @ 2011-03-07 10:17 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp, vlad
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: Vladimir Sokolovsky <vlad@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
drivers/infiniband/hw/mlx4/main.c | 16 ++++
drivers/net/mlx4/en_netdev.c | 133 ++++++++++++++++++++++++------
drivers/net/mlx4/en_port.c | 11 ++-
drivers/net/mlx4/en_port.h | 19 +++-
drivers/net/mlx4/en_rx.c | 11 +--
drivers/net/mlx4/fw.c | 3 +
drivers/net/mlx4/mlx4.h | 6 ++
drivers/net/mlx4/mlx4_en.h | 1 +
drivers/net/mlx4/port.c | 165 +++++++++++++++++++++++++++++++++---
include/linux/mlx4/device.h | 12 ++-
10 files changed, 316 insertions(+), 61 deletions(-)
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index fbe1973..ecd7c95 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -623,6 +623,14 @@ static int mlx4_ib_mcg_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
struct mlx4_ib_dev *mdev = to_mdev(ibqp->device);
struct mlx4_ib_qp *mqp = to_mqp(ibqp);
+ if (mdev->dev->caps.vep_mc_steering && ibqp->qp_type == IB_QPT_RAW_ETHERTYPE) {
+ gid->raw[5] = mqp->port;
+
+ if (!gid->raw[5]) {
+ printk(KERN_WARNING "0 port num for raw_eth attach\n");
+ return -EINVAL;
+ }
+ }
err = mlx4_multicast_attach(mdev->dev, &mqp->mqp, gid->raw,
!!(mqp->flags & MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK),
MLX4_PROT_IB_IPV6);
@@ -665,6 +673,14 @@ static int mlx4_ib_mcg_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid)
struct net_device *ndev;
struct mlx4_ib_gid_entry *ge;
+ if (mdev->dev->caps.vep_mc_steering && ibqp->qp_type == IB_QPT_RAW_ETHERTYPE) {
+ gid->raw[5] = mqp->port;
+
+ if (!gid->raw[5]) {
+ printk(KERN_WARNING "0 port num for raw_eth attach\n");
+ return -EINVAL;
+ }
+ }
err = mlx4_multicast_detach(mdev->dev,
&mqp->mqp, gid->raw, MLX4_PROT_IB_IPV6);
if (err)
diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c
index f0d06a0..0da4b7b 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)
{
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index ebda939..56fa5e1 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -525,9 +525,15 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
int block_mcast_loopback, enum mlx4_protocol protocol);
int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
enum mlx4_protocol protocol);
-
-int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *index);
-void mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, int index);
+int mlx4_multicast_promisc_add(struct mlx4_dev *dev, u32 qpn, u8 port);
+int mlx4_multicast_promisc_remove(struct mlx4_dev *dev, u32 qpn, u8 port);
+int mlx4_unicast_promisc_add(struct mlx4_dev *dev, u32 qpn, u8 port);
+int mlx4_unicast_promisc_remove(struct mlx4_dev *dev, u32 qpn, u8 port);
+int mlx4_SET_MCAST_FLTR(struct mlx4_dev *dev, u8 port, u64 mac, u64 clear, u8 mode);
+
+int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *qpn, u8 wrap);
+void mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, int qpn);
+int mlx4_replace_mac(struct mlx4_dev *dev, u8 port, int qpn, u64 new_mac, u8 wrap);
int mlx4_find_cached_vlan(struct mlx4_dev *dev, u8 port, u16 vid, int *idx);
int mlx4_register_vlan(struct mlx4_dev *dev, u8 port, u16 vlan, int *index);
--
1.6.0.2
^ permalink raw reply related
* [Patch] bonding: fix netpoll in active-backup mode
From: Amerigo Wang @ 2011-03-07 14:11 UTC (permalink / raw)
To: linux-kernel
Cc: WANG Cong, Jay Vosburgh, David S. Miller, Herbert Xu,
Paul E. McKenney, Neil Horman, John W. Linville, Eric Dumazet,
netdev
netconsole doesn't work in active-backup mode, because we don't do anything
for nic failover in active-backup mode. This patch fixes the problem by:
1) make slave_enable_netpoll() and slave_disable_netpoll() callable in softirq
context, that is, moving code after synchronize_rcu_bh() into call_rcu_bh()
callback function, teaching kzalloc() to use GFP_ATOMIC.
2) disable netpoll on old slave and enable netpoll on the new slave.
Tested by ifdown the current active slave and ifup it again for several times,
netconsole works well.
Signed-off-by: WANG Cong <amwang@redhat.com>
---
drivers/net/bonding/bond_main.c | 236 +++++++++++++++++++++------------------
include/linux/netpoll.h | 2 +
net/core/netpoll.c | 22 ++--
3 files changed, 140 insertions(+), 120 deletions(-)
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 0592e6d..2d6ec1b 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -907,6 +907,121 @@ static void bond_mc_list_flush(struct net_device *bond_dev,
}
}
+/*--------------------------- Netpoll code ---------------------------*/
+#ifdef CONFIG_NET_POLL_CONTROLLER
+static inline int slave_enable_netpoll(struct slave *slave)
+{
+ struct netpoll *np;
+ int err = 0;
+
+ np = kzalloc(sizeof(*np), GFP_ATOMIC);
+ err = -ENOMEM;
+ if (!np)
+ goto out;
+
+ np->dev = slave->dev;
+ err = __netpoll_setup(np);
+ if (err) {
+ kfree(np);
+ goto out;
+ }
+ slave->np = np;
+out:
+ return err;
+}
+static void slave_netpoll_reclaim(struct rcu_head *rp)
+{
+ struct netpoll *np = container_of(rp, struct netpoll, rcu);
+ __netpoll_cleanup(np);
+ kfree(np);
+}
+static inline void slave_disable_netpoll(struct slave *slave)
+{
+ struct netpoll *np = slave->np;
+
+ if (!np)
+ return;
+
+ slave->np = NULL;
+ call_rcu_bh(&np->rcu, slave_netpoll_reclaim);
+}
+static inline bool slave_dev_support_netpoll(struct net_device *slave_dev)
+{
+ if (slave_dev->priv_flags & IFF_DISABLE_NETPOLL)
+ return false;
+ if (!slave_dev->netdev_ops->ndo_poll_controller)
+ return false;
+ return true;
+}
+
+static void bond_poll_controller(struct net_device *bond_dev)
+{
+}
+
+static void __bond_netpoll_cleanup(struct bonding *bond)
+{
+ struct slave *slave;
+ int i;
+
+ bond_for_each_slave(bond, slave, i)
+ if (IS_UP(slave->dev))
+ slave_disable_netpoll(slave);
+}
+static void bond_netpoll_cleanup(struct net_device *bond_dev)
+{
+ struct bonding *bond = netdev_priv(bond_dev);
+
+ read_lock(&bond->lock);
+ __bond_netpoll_cleanup(bond);
+ read_unlock(&bond->lock);
+}
+
+static int bond_netpoll_setup(struct net_device *dev, struct netpoll_info *ni)
+{
+ struct bonding *bond = netdev_priv(dev);
+ struct slave *slave;
+ int i, err = 0;
+
+ read_lock(&bond->lock);
+ bond_for_each_slave(bond, slave, i) {
+ if (!IS_UP(slave->dev))
+ continue;
+ err = slave_enable_netpoll(slave);
+ if (err) {
+ __bond_netpoll_cleanup(bond);
+ break;
+ }
+ }
+ read_unlock(&bond->lock);
+ return err;
+}
+
+static struct netpoll_info *bond_netpoll_info(struct bonding *bond)
+{
+ return bond->dev->npinfo;
+}
+
+#else
+static inline int slave_enable_netpoll(struct slave *slave)
+{
+ return 0;
+}
+static inline void slave_disable_netpoll(struct slave *slave)
+{
+}
+static void bond_netpoll_cleanup(struct net_device *bond_dev)
+{
+}
+static int bond_netpoll_setup(struct net_device *dev, struct netpoll_info *ni)
+{
+ return 0;
+}
+static struct netpoll_info *bond_netpoll_info(struct bonding *bond)
+{
+ return NULL;
+}
+#endif
+
/*--------------------------- Active slave change ---------------------------*/
/*
@@ -1155,11 +1270,20 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active)
}
if (bond->params.mode == BOND_MODE_ACTIVEBACKUP) {
- if (old_active)
+ if (old_active) {
bond_set_slave_inactive_flags(old_active);
+ if (bond_netpoll_info(bond))
+ slave_disable_netpoll(old_active);
+ }
if (new_active) {
+ struct netpoll_info *ni;
bond_set_slave_active_flags(new_active);
+ ni = bond_netpoll_info(bond);
+ if (ni) {
+ new_active->dev->npinfo = ni;
+ slave_enable_netpoll(new_active);
+ }
if (bond->params.fail_over_mac)
bond_do_fail_over_mac(bond, new_active,
@@ -1280,116 +1404,6 @@ static void bond_detach_slave(struct bonding *bond, struct slave *slave)
bond->slave_cnt--;
}
-#ifdef CONFIG_NET_POLL_CONTROLLER
-static inline int slave_enable_netpoll(struct slave *slave)
-{
- struct netpoll *np;
- int err = 0;
-
- np = kzalloc(sizeof(*np), GFP_KERNEL);
- err = -ENOMEM;
- if (!np)
- goto out;
-
- np->dev = slave->dev;
- err = __netpoll_setup(np);
- if (err) {
- kfree(np);
- goto out;
- }
- slave->np = np;
-out:
- return err;
-}
-static inline void slave_disable_netpoll(struct slave *slave)
-{
- struct netpoll *np = slave->np;
-
- if (!np)
- return;
-
- slave->np = NULL;
- synchronize_rcu_bh();
- __netpoll_cleanup(np);
- kfree(np);
-}
-static inline bool slave_dev_support_netpoll(struct net_device *slave_dev)
-{
- if (slave_dev->priv_flags & IFF_DISABLE_NETPOLL)
- return false;
- if (!slave_dev->netdev_ops->ndo_poll_controller)
- return false;
- return true;
-}
-
-static void bond_poll_controller(struct net_device *bond_dev)
-{
-}
-
-static void __bond_netpoll_cleanup(struct bonding *bond)
-{
- struct slave *slave;
- int i;
-
- bond_for_each_slave(bond, slave, i)
- if (IS_UP(slave->dev))
- slave_disable_netpoll(slave);
-}
-static void bond_netpoll_cleanup(struct net_device *bond_dev)
-{
- struct bonding *bond = netdev_priv(bond_dev);
-
- read_lock(&bond->lock);
- __bond_netpoll_cleanup(bond);
- read_unlock(&bond->lock);
-}
-
-static int bond_netpoll_setup(struct net_device *dev, struct netpoll_info *ni)
-{
- struct bonding *bond = netdev_priv(dev);
- struct slave *slave;
- int i, err = 0;
-
- read_lock(&bond->lock);
- bond_for_each_slave(bond, slave, i) {
- if (!IS_UP(slave->dev))
- continue;
- err = slave_enable_netpoll(slave);
- if (err) {
- __bond_netpoll_cleanup(bond);
- break;
- }
- }
- read_unlock(&bond->lock);
- return err;
-}
-
-static struct netpoll_info *bond_netpoll_info(struct bonding *bond)
-{
- return bond->dev->npinfo;
-}
-
-#else
-static inline int slave_enable_netpoll(struct slave *slave)
-{
- return 0;
-}
-static inline void slave_disable_netpoll(struct slave *slave)
-{
-}
-static void bond_netpoll_cleanup(struct net_device *bond_dev)
-{
-}
-static int bond_netpoll_setup(struct net_device *dev, struct netpoll_info *ni)
-{
- return 0;
-}
-static struct netpoll_info *bond_netpoll_info(struct bonding *bond)
-{
- return NULL;
-}
-#endif
-
/*---------------------------------- IOCTL ----------------------------------*/
static int bond_sethwaddr(struct net_device *bond_dev,
diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h
index 79358bb..9412aa5 100644
--- a/include/linux/netpoll.h
+++ b/include/linux/netpoll.h
@@ -23,6 +23,7 @@ struct netpoll {
u8 remote_mac[ETH_ALEN];
struct list_head rx; /* rx_np list element */
+ struct rcu_head rcu;
};
struct netpoll_info {
@@ -38,6 +39,7 @@ struct netpoll_info {
struct delayed_work tx_work;
struct netpoll *netpoll;
+ struct rcu_head rcu;
};
void netpoll_poll_dev(struct net_device *dev);
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index 06be243..9870dac 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -901,6 +901,18 @@ static int __init netpoll_init(void)
}
core_initcall(netpoll_init);
+static void netpoll_reclaim(struct rcu_head *rp)
+{
+ struct netpoll_info *npinfo = container_of(rp, struct netpoll_info, rcu);
+ skb_queue_purge(&npinfo->arp_tx);
+ skb_queue_purge(&npinfo->txq);
+ cancel_delayed_work_sync(&npinfo->tx_work);
+
+ /* clean after last, unfinished work */
+ __skb_queue_purge(&npinfo->txq);
+ kfree(npinfo);
+}
+
void __netpoll_cleanup(struct netpoll *np)
{
struct netpoll_info *npinfo;
@@ -928,15 +940,7 @@ void __netpoll_cleanup(struct netpoll *np)
rcu_assign_pointer(np->dev->npinfo, NULL);
/* avoid racing with NAPI reading npinfo */
- synchronize_rcu_bh();
-
- skb_queue_purge(&npinfo->arp_tx);
- skb_queue_purge(&npinfo->txq);
- cancel_delayed_work_sync(&npinfo->tx_work);
-
- /* clean after last, unfinished work */
- __skb_queue_purge(&npinfo->txq);
- kfree(npinfo);
+ call_rcu_bh(&npinfo->rcu, netpoll_reclaim);
}
}
EXPORT_SYMBOL_GPL(__netpoll_cleanup);
^ permalink raw reply related
* [PATCH 06/17 V2] mlx4_en: using new mlx4 interrupt scheme
From: Yevgeny Petrilin @ 2011-03-07 13:27 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp, markuze
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 <markuze@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
diff from V1: using <ifname>-tx instead of <ifname>_tx
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(-)
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..f0d06a0 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);
--
1.6.0.2
^ permalink raw reply related
* RE: [PATCH 06/17] mlx4_en: using new mlx4 interrupt scheme
From: Yevgeny Petrilin @ 2011-03-07 13:22 UTC (permalink / raw)
To: Ben Hutchings; +Cc: davem@davemloft.net, netdev@vger.kernel.org, Alex Markuze
In-Reply-To: <1299501715.3134.49.camel@localhost>
>
> If this string is being copied to the IRQ name, it should be
> "<ifname>-tx" not "<ifname>_tx" for consistency with other drivers.
>
Thanks,
Will send an updated patch
Yevgeny
^ permalink raw reply
* [PATCH net-next-2.6 7/7] be2net: Add multicast filter capability for Lancer
From: Padmanabh Ratnakar @ 2011-03-07 13:09 UTC (permalink / raw)
To: netdev, davem; +Cc: Padmanabh Ratnakar, Sathya Perla, Subramanian Seetharaman
Lancer requires multicast capability flag set during IFACE_CREATE
for adding multicast filters.
Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@emulex.com>
Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: Subramanian Seetharaman <subbu.seetharaman@emulex.com>
---
drivers/net/benet/be_cmds.h | 3 ++-
drivers/net/benet/be_main.c | 4 +++-
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/drivers/net/benet/be_cmds.h b/drivers/net/benet/be_cmds.h
index 5bdaa22..b4ac393 100644
--- a/drivers/net/benet/be_cmds.h
+++ b/drivers/net/benet/be_cmds.h
@@ -518,7 +518,8 @@ enum be_if_flags {
BE_IF_FLAGS_VLAN = 0x100,
BE_IF_FLAGS_MCAST_PROMISCUOUS = 0x200,
BE_IF_FLAGS_PASS_L2_ERRORS = 0x400,
- BE_IF_FLAGS_PASS_L3L4_ERRORS = 0x800
+ BE_IF_FLAGS_PASS_L3L4_ERRORS = 0x800,
+ BE_IF_FLAGS_MULTICAST = 0x1000
};
/* An RX interface is an object with one or more MAC addresses and
diff --git a/drivers/net/benet/be_main.c b/drivers/net/benet/be_main.c
index a7fd833..68f1078 100644
--- a/drivers/net/benet/be_main.c
+++ b/drivers/net/benet/be_main.c
@@ -2263,7 +2263,9 @@ static int be_setup(struct be_adapter *adapter)
int status;
u8 mac[ETH_ALEN];
- cap_flags = en_flags = BE_IF_FLAGS_UNTAGGED | BE_IF_FLAGS_BROADCAST;
+ cap_flags = en_flags = BE_IF_FLAGS_UNTAGGED |
+ BE_IF_FLAGS_BROADCAST |
+ BE_IF_FLAGS_MULTICAST;
if (be_physfn(adapter)) {
cap_flags |= BE_IF_FLAGS_MCAST_PROMISCUOUS |
--
1.6.0.2
^ permalink raw reply related
* [PATCH net-next-2.6 6/7] be2net: Disarm CQ and EQ to disable interrupt in Lancer
From: Padmanabh Ratnakar @ 2011-03-07 13:09 UTC (permalink / raw)
To: netdev, davem; +Cc: Padmanabh Ratnakar, Sathya Perla, Subramanian Seetharaman
For Lancer disable interrupts in close by disarming CQs and EQs.
Change the order of calls in be_close to achieve the correct result.
Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@emulex.com>
Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: Subramanian Seetharaman <subbu.seetharaman@emulex.com>
---
drivers/net/benet/be_main.c | 17 ++++++++++++-----
1 files changed, 12 insertions(+), 5 deletions(-)
diff --git a/drivers/net/benet/be_main.c b/drivers/net/benet/be_main.c
index 68b9d17..a7fd833 100644
--- a/drivers/net/benet/be_main.c
+++ b/drivers/net/benet/be_main.c
@@ -2088,6 +2088,18 @@ static int be_close(struct net_device *netdev)
if (!lancer_chip(adapter))
be_intr_set(adapter, false);
+ for_all_rx_queues(adapter, rxo, i)
+ napi_disable(&rxo->rx_eq.napi);
+
+ napi_disable(&tx_eq->napi);
+
+ if (lancer_chip(adapter)) {
+ be_cq_notify(adapter, adapter->tx_obj.cq.id, false, 0);
+ be_cq_notify(adapter, adapter->mcc_obj.cq.id, false, 0);
+ for_all_rx_queues(adapter, rxo, i)
+ be_cq_notify(adapter, rxo->cq.id, false, 0);
+ }
+
if (adapter->msix_enabled) {
vec = be_msix_vec_get(adapter, tx_eq);
synchronize_irq(vec);
@@ -2101,11 +2113,6 @@ static int be_close(struct net_device *netdev)
}
be_irq_unregister(adapter);
- for_all_rx_queues(adapter, rxo, i)
- napi_disable(&rxo->rx_eq.napi);
-
- napi_disable(&tx_eq->napi);
-
/* Wait for all pending tx completions to arrive so that
* all tx skbs are freed.
*/
--
1.6.0.2
^ permalink raw reply related
* [PATCH net-next-2.6 5/7] be2net: Remove TX Queue stop in close
From: Padmanabh Ratnakar @ 2011-03-07 13:09 UTC (permalink / raw)
To: netdev, davem; +Cc: Padmanabh Ratnakar, Sathya Perla, Subramanian Seetharaman
Remove TX Queue stop in close
Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@emulex.com>
Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: Subramanian Seetharaman <subbu.seetharaman@emulex.com>
---
drivers/net/benet/be_main.c | 1 -
1 files changed, 0 insertions(+), 1 deletions(-)
diff --git a/drivers/net/benet/be_main.c b/drivers/net/benet/be_main.c
index 665a9b7..68b9d17 100644
--- a/drivers/net/benet/be_main.c
+++ b/drivers/net/benet/be_main.c
@@ -2082,7 +2082,6 @@ static int be_close(struct net_device *netdev)
be_async_mcc_disable(adapter);
- netif_stop_queue(netdev);
netif_carrier_off(netdev);
adapter->link_up = false;
--
1.6.0.2
^ permalink raw reply related
* [PATCH net-next-2.6 3/7] be2net: Add error recovery during load for Lancer
From: Padmanabh Ratnakar @ 2011-03-07 13:08 UTC (permalink / raw)
To: netdev, davem; +Cc: Padmanabh Ratnakar, Sathya Perla, Subramanian Seetharaman
Add error recovery during load for Lancer
Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@emulex.com>
Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: Subramanian Seetharaman <subbu.seetharaman@emulex.com>
---
drivers/net/benet/be_hw.h | 12 +++++++++
drivers/net/benet/be_main.c | 56 +++++++++++++++++++++++++++++++++++++++++++
2 files changed, 68 insertions(+), 0 deletions(-)
diff --git a/drivers/net/benet/be_hw.h b/drivers/net/benet/be_hw.h
index 3f459f7..dbe67f3 100644
--- a/drivers/net/benet/be_hw.h
+++ b/drivers/net/benet/be_hw.h
@@ -44,6 +44,18 @@
#define POST_STAGE_BE_RESET 0x3 /* Host wants to reset chip */
#define POST_STAGE_ARMFW_RDY 0xc000 /* FW is done with POST */
+
+/* Lancer SLIPORT_CONTROL SLIPORT_STATUS registers */
+#define SLIPORT_STATUS_OFFSET 0x404
+#define SLIPORT_CONTROL_OFFSET 0x408
+
+#define SLIPORT_STATUS_ERR_MASK 0x80000000
+#define SLIPORT_STATUS_RN_MASK 0x01000000
+#define SLIPORT_STATUS_RDY_MASK 0x00800000
+
+
+#define SLI_PORT_CONTROL_IP_MASK 0x08000000
+
/********* Memory BAR register ************/
#define PCICFG_MEMBAR_CTRL_INT_CTRL_OFFSET 0xfc
/* Host Interrupt Enable, if set interrupts are enabled although "PCI Interrupt
diff --git a/drivers/net/benet/be_main.c b/drivers/net/benet/be_main.c
index ac7ae21..665a9b7 100644
--- a/drivers/net/benet/be_main.c
+++ b/drivers/net/benet/be_main.c
@@ -2901,6 +2901,54 @@ static int be_dev_family_check(struct be_adapter *adapter)
return 0;
}
+static int lancer_wait_ready(struct be_adapter *adapter)
+{
+#define SLIPORT_READY_TIMEOUT 500
+ u32 sliport_status;
+ int status = 0, i;
+
+ for (i = 0; i < SLIPORT_READY_TIMEOUT; i++) {
+ sliport_status = ioread32(adapter->db + SLIPORT_STATUS_OFFSET);
+ if (sliport_status & SLIPORT_STATUS_RDY_MASK)
+ break;
+
+ msleep(20);
+ }
+
+ if (i == SLIPORT_READY_TIMEOUT)
+ status = -1;
+
+ return status;
+}
+
+static int lancer_test_and_set_rdy_state(struct be_adapter *adapter)
+{
+ int status;
+ u32 sliport_status, err, reset_needed;
+ status = lancer_wait_ready(adapter);
+ if (!status) {
+ sliport_status = ioread32(adapter->db + SLIPORT_STATUS_OFFSET);
+ err = sliport_status & SLIPORT_STATUS_ERR_MASK;
+ reset_needed = sliport_status & SLIPORT_STATUS_RN_MASK;
+ if (err && reset_needed) {
+ iowrite32(SLI_PORT_CONTROL_IP_MASK,
+ adapter->db + SLIPORT_CONTROL_OFFSET);
+
+ /* check adapter has corrected the error */
+ status = lancer_wait_ready(adapter);
+ sliport_status = ioread32(adapter->db +
+ SLIPORT_STATUS_OFFSET);
+ sliport_status &= (SLIPORT_STATUS_ERR_MASK |
+ SLIPORT_STATUS_RN_MASK);
+ if (status || sliport_status)
+ status = -1;
+ } else if (err || reset_needed) {
+ status = -1;
+ }
+ }
+ return status;
+}
+
static int __devinit be_probe(struct pci_dev *pdev,
const struct pci_device_id *pdev_id)
{
@@ -2950,6 +2998,14 @@ static int __devinit be_probe(struct pci_dev *pdev,
if (status)
goto free_netdev;
+ if (lancer_chip(adapter)) {
+ status = lancer_test_and_set_rdy_state(adapter);
+ if (status) {
+ dev_err(&pdev->dev, "Adapter in non recoverable error\n");
+ goto free_netdev;
+ }
+ }
+
/* sync up with fw's ready state */
if (be_physfn(adapter)) {
status = be_cmd_POST(adapter);
--
1.6.0.2
^ permalink raw reply related
* [PATCH net-next-2.6 4/7] be2net: Change f/w command versions for Lancer
From: Padmanabh Ratnakar @ 2011-03-07 13:08 UTC (permalink / raw)
To: netdev, davem; +Cc: Padmanabh Ratnakar, Sathya Perla, Subramanian Seetharaman
Change f/w command versions for Lancer
Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@emulex.com>
Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: Subramanian Seetharaman <subbu.seetharaman@emulex.com>
---
drivers/net/benet/be_cmds.c | 8 +++++++-
drivers/net/benet/be_cmds.h | 2 +-
2 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/drivers/net/benet/be_cmds.c b/drivers/net/benet/be_cmds.c
index 1822ecd..cc3a235 100644
--- a/drivers/net/benet/be_cmds.c
+++ b/drivers/net/benet/be_cmds.c
@@ -726,7 +726,7 @@ int be_cmd_cq_create(struct be_adapter *adapter,
req->num_pages = cpu_to_le16(PAGES_4K_SPANNED(q_mem->va, q_mem->size));
if (lancer_chip(adapter)) {
- req->hdr.version = 1;
+ req->hdr.version = 2;
req->page_size = 1; /* 1 for 4K */
AMAP_SET_BITS(struct amap_cq_context_lancer, coalescwm, ctxt,
coalesce_wm);
@@ -862,6 +862,12 @@ int be_cmd_txq_create(struct be_adapter *adapter,
be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ETH, OPCODE_ETH_TX_CREATE,
sizeof(*req));
+ if (lancer_chip(adapter)) {
+ req->hdr.version = 1;
+ AMAP_SET_BITS(struct amap_tx_context, if_id, ctxt,
+ adapter->if_handle);
+ }
+
req->num_pages = PAGES_4K_SPANNED(q_mem->va, q_mem->size);
req->ulp_num = BE_ULP1_NUM;
req->type = BE_ETH_TX_RING_TYPE_STANDARD;
diff --git a/drivers/net/benet/be_cmds.h b/drivers/net/benet/be_cmds.h
index 93e5768..5bdaa22 100644
--- a/drivers/net/benet/be_cmds.h
+++ b/drivers/net/benet/be_cmds.h
@@ -430,7 +430,7 @@ struct be_cmd_resp_mcc_create {
/* Pseudo amap definition in which each bit of the actual structure is defined
* as a byte: used to calculate offset/shift/mask of each field */
struct amap_tx_context {
- u8 rsvd0[16]; /* dword 0 */
+ u8 if_id[16]; /* dword 0 */
u8 tx_ring_size[4]; /* dword 0 */
u8 rsvd1[26]; /* dword 0 */
u8 pci_func_id[8]; /* dword 1 */
--
1.6.0.2
^ permalink raw reply related
* [PATCH net-next-2.6 2/7] be2net: Checksum field valid only for TCP/UDP
From: Padmanabh Ratnakar @ 2011-03-07 13:08 UTC (permalink / raw)
To: netdev, davem; +Cc: Padmanabh Ratnakar, Sathya Perla, Subramanian Seetharaman
L4 checksum field is valid only for TCP/UDP packets in Lancer
Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@emulex.com>
Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: Subramanian Seetharaman <subbu.seetharaman@emulex.com>
---
drivers/net/benet/be_main.c | 9 ++++++---
1 files changed, 6 insertions(+), 3 deletions(-)
diff --git a/drivers/net/benet/be_main.c b/drivers/net/benet/be_main.c
index bf34434..ac7ae21 100644
--- a/drivers/net/benet/be_main.c
+++ b/drivers/net/benet/be_main.c
@@ -865,14 +865,17 @@ static void be_rx_stats_update(struct be_rx_obj *rxo,
static inline bool csum_passed(struct be_eth_rx_compl *rxcp)
{
- u8 l4_cksm, ipv6, ipcksm;
+ u8 l4_cksm, ipv6, ipcksm, tcpf, udpf;
l4_cksm = AMAP_GET_BITS(struct amap_eth_rx_compl, l4_cksm, rxcp);
ipcksm = AMAP_GET_BITS(struct amap_eth_rx_compl, ipcksm, rxcp);
ipv6 = AMAP_GET_BITS(struct amap_eth_rx_compl, ip_version, rxcp);
+ tcpf = AMAP_GET_BITS(struct amap_eth_rx_compl, tcpf, rxcp);
+ udpf = AMAP_GET_BITS(struct amap_eth_rx_compl, udpf, rxcp);
- /* Ignore ipcksm for ipv6 pkts */
- return l4_cksm && (ipcksm || ipv6);
+ /* L4 checksum is not reliable for non TCP/UDP packets.
+ * Also ignore ipcksm for ipv6 pkts */
+ return (tcpf || udpf) && l4_cksm && (ipcksm || ipv6);
}
static struct be_rx_page_info *
--
1.6.0.2
^ permalink raw reply related
* [PATCH net-next-2.6 1/7] be2net: Remove ERR compl workaround for Lancer
From: Padmanabh Ratnakar @ 2011-03-07 13:07 UTC (permalink / raw)
To: netdev, davem; +Cc: Padmanabh Ratnakar, Sathya Perla, Subramanian Seetharaman
Workaround added for Lancer in handling RX ERR completion received
when no RX buffers are posted is not needed.
Signed-off-by: Padmanabh Ratnakar <padmanabh.ratnakar@emulex.com>
Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: Subramanian Seetharaman <subbu.seetharaman@emulex.com>
---
drivers/net/benet/be.h | 4 +---
drivers/net/benet/be_main.c | 31 ++++++++-----------------------
2 files changed, 9 insertions(+), 26 deletions(-)
diff --git a/drivers/net/benet/be.h b/drivers/net/benet/be.h
index ed709a5..4ac0d72 100644
--- a/drivers/net/benet/be.h
+++ b/drivers/net/benet/be.h
@@ -220,9 +220,7 @@ struct be_rx_obj {
struct be_rx_stats stats;
u8 rss_id;
bool rx_post_starved; /* Zero rx frags have been posted to BE */
- u16 last_frag_index;
- u16 rsvd;
- u32 cache_line_barrier[15];
+ u32 cache_line_barrier[16];
};
struct be_drv_stats {
diff --git a/drivers/net/benet/be_main.c b/drivers/net/benet/be_main.c
index ef66dc6..bf34434 100644
--- a/drivers/net/benet/be_main.c
+++ b/drivers/net/benet/be_main.c
@@ -909,17 +909,11 @@ static void be_rx_compl_discard(struct be_adapter *adapter,
rxq_idx = AMAP_GET_BITS(struct amap_eth_rx_compl, fragndx, rxcp);
num_rcvd = AMAP_GET_BITS(struct amap_eth_rx_compl, numfrags, rxcp);
- /* Skip out-of-buffer compl(lancer) or flush compl(BE) */
- if (likely(rxq_idx != rxo->last_frag_index && num_rcvd != 0)) {
-
- rxo->last_frag_index = rxq_idx;
-
- for (i = 0; i < num_rcvd; i++) {
- page_info = get_rx_page_info(adapter, rxo, rxq_idx);
- put_page(page_info->page);
- memset(page_info, 0, sizeof(*page_info));
- index_inc(&rxq_idx, rxq->len);
- }
+ for (i = 0; i < num_rcvd; i++) {
+ page_info = get_rx_page_info(adapter, rxo, rxq_idx);
+ put_page(page_info->page);
+ memset(page_info, 0, sizeof(*page_info));
+ index_inc(&rxq_idx, rxq->len);
}
}
@@ -1579,9 +1573,6 @@ static int be_rx_queues_create(struct be_adapter *adapter)
adapter->big_page_size = (1 << get_order(rx_frag_size)) * PAGE_SIZE;
for_all_rx_queues(adapter, rxo, i) {
rxo->adapter = adapter;
- /* Init last_frag_index so that the frag index in the first
- * completion will never match */
- rxo->last_frag_index = 0xffff;
rxo->rx_eq.max_eqd = BE_MAX_EQD;
rxo->rx_eq.enable_aic = true;
@@ -1722,7 +1713,7 @@ static int be_poll_rx(struct napi_struct *napi, int budget)
struct be_queue_info *rx_cq = &rxo->cq;
struct be_eth_rx_compl *rxcp;
u32 work_done;
- u16 frag_index, num_rcvd;
+ u16 num_rcvd;
u8 err;
rxo->stats.rx_polls++;
@@ -1732,16 +1723,10 @@ static int be_poll_rx(struct napi_struct *napi, int budget)
break;
err = AMAP_GET_BITS(struct amap_eth_rx_compl, err, rxcp);
- frag_index = AMAP_GET_BITS(struct amap_eth_rx_compl, fragndx,
- rxcp);
num_rcvd = AMAP_GET_BITS(struct amap_eth_rx_compl, numfrags,
rxcp);
-
- /* Skip out-of-buffer compl(lancer) or flush compl(BE) */
- if (likely(frag_index != rxo->last_frag_index &&
- num_rcvd != 0)) {
- rxo->last_frag_index = frag_index;
-
+ /* Ignore flush completions */
+ if (num_rcvd) {
if (do_gro(rxo, rxcp, err))
be_rx_compl_process_gro(adapter, rxo, rxcp);
else
--
1.6.0.2
^ permalink raw reply related
* [PATCH net-next-2.6 0/7] be2net: Patch series for supporting Lancer
From: Padmanabh Ratnakar @ 2011-03-07 13:05 UTC (permalink / raw)
To: netdev, davem; +Cc: Padmanabh Ratnakar, Sathya Perla, Subramanian Seetharaman
Hi David,
Following are patches to support Lancer family of CNA.
Please apply.
Thanks,
Padmanabh
Signed-off-by: Sathya Perla <sathya.perla@emulex.com>
Signed-off-by: Subramanian Seetharaman <subbu.seetharaman@emulex.com>
Padmanabh Ratnakar (7):
be2net: Remove ERR compl workaround for Lancer
be2net: Checksum field valid only for TCP/UDP
be2net: Add error recovery during load for Lancer
be2net: Change f/w command versions for Lancer
be2net: Remove TX Queue stop in close
be2net: Disarm CQ and EQ to disable interrupt in Lancer
be2net: Add multicast filter capability for Lancer
drivers/net/benet/be.h | 4 +-
drivers/net/benet/be_cmds.c | 8 +++-
drivers/net/benet/be_cmds.h | 5 +-
drivers/net/benet/be_hw.h | 12 ++++
drivers/net/benet/be_main.c | 118 +++++++++++++++++++++++++++++++------------
5 files changed, 108 insertions(+), 39 deletions(-)
^ permalink raw reply
* Re: tun routing is broken
From: Jiri Kosina @ 2011-03-07 12:54 UTC (permalink / raw)
To: Jiri Slaby, Andrew Morton
Cc: David Miller, Jiri Slaby, linux-kernel, mm-commits, netdev, maxk
In-Reply-To: <4D71F194.9010402@suse.cz>
On Sat, 5 Mar 2011, Jiri Slaby wrote:
> > Date: Sat, 05 Mar 2011 09:04:22 +0100
> >
> >> Ok, so I booted the new kernel, and tun is broken there completely. If I
> >> try to ping a vpn peer:
> >
> > -mm tree is missing this fix which went in yesterday. Please if
> > you are going to be testing networking a lot, test against net-next-2.6
> > instead of Andrew's tree which invariable lags behind:
>
> I'm not testing networking, I'm just using my desktop :). (And
> next/master is mostly unusable for these needs. In comparison to mmotm
> which is some kind of a "release".)
I have been hit by similar issues several times already as well.
I guess that question of including -mm into linux-next should be raised
again here ... Andrew, if I remember correctly, on last kernel summit you
were stating that you'd stop eating untill you have -mm included in -next.
Seems like you must be quite hungry these days already :)
Thanks,
--
Jiri Kosina
SUSE Labs, Novell Inc.
^ permalink raw reply
* [patch net-next-2.6] net: reinject arps into bonding slave instead of master
From: Jiri Pirko @ 2011-03-07 12:51 UTC (permalink / raw)
To: Nicolas de Pesloüan
Cc: netdev, davem, shemminger, kaber, fubar, eric.dumazet, andy
In-Reply-To: <20110306133413.GB2795@psychotron.redhat.com>
Recent patch "bonding: move processing of recv handlers into
handle_frame()" caused a regression on following net scheme:
eth0 - bond0 - bond0.5
where arp monitoring is happening over vlan. This patch fixes it by
reinjecting the arp packet into bonding slave device so the bonding
rx_handler can pickup and process it.
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
---
net/core/dev.c | 8 ++++----
1 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/net/core/dev.c b/net/core/dev.c
index c71bd18..3d88458 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3094,12 +3094,12 @@ void netdev_rx_handler_unregister(struct net_device *dev)
}
EXPORT_SYMBOL_GPL(netdev_rx_handler_unregister);
-static void vlan_on_bond_hook(struct sk_buff *skb)
+static void vlan_on_bond_hook(struct sk_buff *skb, struct net_device *orig_dev)
{
/*
* Make sure ARP frames received on VLAN interfaces stacked on
* bonding interfaces still make their way to any base bonding
- * device that may have registered for a specific ptype.
+ * device by reinjecting the frame into bonding slave (orig_dev)
*/
if (skb->dev->priv_flags & IFF_802_1Q_VLAN &&
vlan_dev_real_dev(skb->dev)->priv_flags & IFF_BONDING &&
@@ -3108,7 +3108,7 @@ static void vlan_on_bond_hook(struct sk_buff *skb)
if (!skb2)
return;
- skb2->dev = vlan_dev_real_dev(skb->dev);
+ skb2->dev = orig_dev;
netif_rx(skb2);
}
}
@@ -3202,7 +3202,7 @@ ncls:
goto out;
}
- vlan_on_bond_hook(skb);
+ vlan_on_bond_hook(skb, orig_dev);
/* deliver only exact match when indicated */
null_or_dev = deliver_exact ? skb->dev : NULL;
--
1.7.4
^ permalink raw reply related
* Re: [PATCH 06/17] mlx4_en: using new mlx4 interrupt scheme
From: Ben Hutchings @ 2011-03-07 12:41 UTC (permalink / raw)
To: Yevgeny Petrilin; +Cc: davem, netdev, markuze
In-Reply-To: <4D74B0AE.5070403@mellanox.co.il>
On Mon, 2011-03-07 at 12:17 +0200, Yevgeny Petrilin wrote:
> 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.
[...]
> --- a/drivers/net/mlx4/en_main.c
> +++ b/drivers/net/mlx4/en_main.c
[...]
> @@ -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);
> + }
> + }
[...]
If this string is being copied to the IRQ name, it should be
"<ifname>-tx" not "<ifname>_tx" for consistency with other drivers.
Ben.
--
Ben Hutchings, Senior Software Engineer, Solarflare Communications
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply
* [PATCH 8/8] Phonet: kill the ST-Ericsson pipe controller Kconfig
From: Rémi Denis-Courmont @ 2011-03-07 10:46 UTC (permalink / raw)
To: netdev
In-Reply-To: <201103071245.47722.remi.denis-courmont@nokia.com>
This is now a run-time choice so that a single kernel can support both
old and new generation ISI modems.
Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
---
Documentation/networking/phonet.txt | 13 -------------
include/linux/phonet.h | 2 --
net/phonet/Kconfig | 12 ------------
net/phonet/pep.c | 25 -------------------------
4 files changed, 0 insertions(+), 52 deletions(-)
diff --git a/Documentation/networking/phonet.txt b/Documentation/networking/phonet.txt
index 3d12779..8100358 100644
--- a/Documentation/networking/phonet.txt
+++ b/Documentation/networking/phonet.txt
@@ -205,19 +205,6 @@ The pipe protocol provides two socket options at the SOL_PNPIPE level:
socket descriptors that are already connected or being connected.
-Phonet Pipe-controller Implementation
--------------------------------------
-
-Phonet Pipe-controller is enabled by selecting the CONFIG_PHONET_PIPECTRLR
-Kconfig option.
-
-The implementation adds socket options at SOL_PNPIPE level:
-
- PNPIPE_ENABLE accepts one integer value (int). If set to zero, the pipe
- is disabled. If the value is non-zero, the pipe is enabled. If the pipe
- is not (yet) connected, ENOTCONN is error is returned.
-
-
Authors
-------
diff --git a/include/linux/phonet.h b/include/linux/phonet.h
index 32a0965..6fb1384 100644
--- a/include/linux/phonet.h
+++ b/include/linux/phonet.h
@@ -37,8 +37,6 @@
#define PNPIPE_ENCAP 1
#define PNPIPE_IFINDEX 2
#define PNPIPE_HANDLE 3
-#define PNPIPE_ENABLE 4
-/* unused slot */
#define PNADDR_ANY 0
#define PNADDR_BROADCAST 0xFC
diff --git a/net/phonet/Kconfig b/net/phonet/Kconfig
index 0d9b8a2..6ec7d55 100644
--- a/net/phonet/Kconfig
+++ b/net/phonet/Kconfig
@@ -14,15 +14,3 @@ config PHONET
To compile this driver as a module, choose M here: the module
will be called phonet. If unsure, say N.
-
-config PHONET_PIPECTRLR
- bool "Phonet Pipe Controller (EXPERIMENTAL)"
- depends on PHONET && EXPERIMENTAL
- default N
- help
- The Pipe Controller implementation in Phonet stack to support Pipe
- data with Nokia Slim modems like WG2.5 used on ST-Ericsson U8500
- platform.
-
- This option is incompatible with older Nokia modems.
- Say N here unless you really know what you are doing.
diff --git a/net/phonet/pep.c b/net/phonet/pep.c
index 671effb..68e635f 100644
--- a/net/phonet/pep.c
+++ b/net/phonet/pep.c
@@ -167,15 +167,6 @@ static int pipe_handler_send_created_ind(struct sock *sk)
data, 4, GFP_ATOMIC);
}
-#ifdef CONFIG_PHONET_PIPECTRLR
-static int pipe_handler_enable_pipe(struct sock *sk, int enable)
-{
- u8 id = enable ? PNS_PEP_ENABLE_REQ : PNS_PEP_DISABLE_REQ;
-
- return pipe_handler_request(sk, id, PAD, NULL, 0);
-}
-#endif
-
static int pep_accept_conn(struct sock *sk, struct sk_buff *skb)
{
static const u8 data[20] = {
@@ -968,16 +959,6 @@ static int pep_setsockopt(struct sock *sk, int level, int optname,
}
goto out_norel;
-#ifdef CONFIG_PHONET_PIPECTRLR
- case PNPIPE_ENABLE:
- if ((1 << sk->sk_state) & ~(TCPF_SYN_RECV|TCPF_ESTABLISHED)) {
- err = -ENOTCONN;
- break;
- }
- err = pipe_handler_enable_pipe(sk, val);
- break;
-#endif
-
default:
err = -ENOPROTOOPT;
}
@@ -1013,12 +994,6 @@ static int pep_getsockopt(struct sock *sk, int level, int optname,
return -EINVAL;
break;
-#ifdef CONFIG_PHONET_PIPECTRLR
- case PNPIPE_ENABLE:
- val = sk->sk_state == TCP_ESTABLISHED;
- break;
-#endif
-
default:
return -ENOPROTOOPT;
}
--
1.7.1
^ permalink raw reply related
* [PATCH 6/8] Phonet: provide pipe socket option to retrieve the pipe identifier
From: Rémi Denis-Courmont @ 2011-03-07 10:46 UTC (permalink / raw)
To: netdev
In-Reply-To: <201103071245.47722.remi.denis-courmont@nokia.com>
User-space sometimes needs this information.
Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
---
Documentation/networking/phonet.txt | 7 ++++---
include/linux/phonet.h | 2 +-
net/phonet/pep.c | 15 +++++++--------
3 files changed, 12 insertions(+), 12 deletions(-)
diff --git a/Documentation/networking/phonet.txt b/Documentation/networking/phonet.txt
index 24ad2ad..cacac96 100644
--- a/Documentation/networking/phonet.txt
+++ b/Documentation/networking/phonet.txt
@@ -181,6 +181,10 @@ The pipe protocol provides two socket options at the SOL_PNPIPE level:
interface index of the network interface created by PNPIPE_ENCAP,
or zero if encapsulation is off.
+ PNPIPE_HANDLE is a read-only integer value. It contains the underlying
+ identifier ("pipe handle") of the pipe. This is only defined for
+ socket descriptors that are already connected or being connected.
+
Phonet Pipe-controller Implementation
-------------------------------------
@@ -199,9 +203,6 @@ between itself and a remote pipe-end point (e.g. modem).
The implementation adds socket options at SOL_PNPIPE level:
- PNPIPE_PIPE_HANDLE
- It accepts an integer argument for setting value of pipe handle.
-
PNPIPE_ENABLE accepts one integer value (int). If set to zero, the pipe
is disabled. If the value is non-zero, the pipe is enabled. If the pipe
is not (yet) connected, ENOTCONN is error is returned.
diff --git a/include/linux/phonet.h b/include/linux/phonet.h
index 26c8df7..32a0965 100644
--- a/include/linux/phonet.h
+++ b/include/linux/phonet.h
@@ -36,7 +36,7 @@
/* Socket options for SOL_PNPIPE level */
#define PNPIPE_ENCAP 1
#define PNPIPE_IFINDEX 2
-#define PNPIPE_PIPE_HANDLE 3
+#define PNPIPE_HANDLE 3
#define PNPIPE_ENABLE 4
/* unused slot */
diff --git a/net/phonet/pep.c b/net/phonet/pep.c
index c0fab4c..abfb795 100644
--- a/net/phonet/pep.c
+++ b/net/phonet/pep.c
@@ -853,6 +853,7 @@ static int pep_sock_connect(struct sock *sk, struct sockaddr *addr, int len)
pn->pn_sk.dobject = pn_sockaddr_get_object(spn);
pn->pn_sk.resource = pn_sockaddr_get_resource(spn);
+ pn->pipe_handle = 1; /* anything but INVALID_HANDLE */
return pipe_handler_request(sk, PNS_PEP_CONNECT_REQ,
PN_PIPE_DISABLE, data, 4);
}
@@ -909,14 +910,6 @@ static int pep_setsockopt(struct sock *sk, int level, int optname,
lock_sock(sk);
switch (optname) {
-#ifdef CONFIG_PHONET_PIPECTRLR
- case PNPIPE_PIPE_HANDLE:
- if (val) {
- pn->pipe_handle = val;
- break;
- }
-#endif
-
case PNPIPE_ENCAP:
if (val && val != PNPIPE_ENCAP_IP) {
err = -EINVAL;
@@ -982,6 +975,12 @@ static int pep_getsockopt(struct sock *sk, int level, int optname,
val = pn->ifindex;
break;
+ case PNPIPE_HANDLE:
+ val = pn->pipe_handle;
+ if (val == PN_PIPE_INVALID_HANDLE)
+ return -EINVAL;
+ break;
+
#ifdef CONFIG_PHONET_PIPECTRLR
case PNPIPE_ENABLE:
val = sk->sk_state == TCP_ESTABLISHED;
--
1.7.1
^ permalink raw reply related
* [PATCH 7/8] Phonet: support active connection without pipe controller on modem
From: Rémi Denis-Courmont @ 2011-03-07 10:46 UTC (permalink / raw)
To: netdev
In-Reply-To: <201103071245.47722.remi.denis-courmont@nokia.com>
This provides support for newer ISI modems with no need for the
earlier experimental compile-time alternative choice. This also fixes
a number of remaining bugs in the old ST-E code.
Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
---
Documentation/networking/phonet.txt | 53 +++++------
net/phonet/pep.c | 172 ++++++++++++++++++++--------------
net/phonet/socket.c | 102 ++++++++-------------
3 files changed, 165 insertions(+), 162 deletions(-)
diff --git a/Documentation/networking/phonet.txt b/Documentation/networking/phonet.txt
index cacac96..3d12779 100644
--- a/Documentation/networking/phonet.txt
+++ b/Documentation/networking/phonet.txt
@@ -154,9 +154,28 @@ connections, one per accept()'d socket.
write(cfd, msg, msglen);
}
-Connections are established between two endpoints by a "third party"
-application. This means that both endpoints are passive; so connect()
-is not possible.
+Connections are traditionally established between two endpoints by a
+"third party" application. This means that both endpoints are passive.
+
+
+As of Linux kernel version 2.6.39, it is also possible to connect
+two endpoints directly, using connect() on the active side. This is
+intended to support the newer Nokia Wireless Modem API, as found in
+e.g. the Nokia Slim Modem in the ST-Ericsson U8500 platform:
+
+ struct sockaddr_spn spn;
+ int fd;
+
+ fd = socket(PF_PHONET, SOCK_SEQPACKET, PN_PROTO_PIPE);
+ memset(&spn, 0, sizeof(spn));
+ spn.spn_family = AF_PHONET;
+ spn.spn_obj = ...;
+ spn.spn_dev = ...;
+ spn.spn_resource = 0xD9;
+ connect(fd, (struct sockaddr *)&spn, sizeof(spn));
+ /* normal I/O here ... */
+ close(fd);
+
WARNING:
When polling a connected pipe socket for writability, there is an
@@ -189,17 +208,8 @@ The pipe protocol provides two socket options at the SOL_PNPIPE level:
Phonet Pipe-controller Implementation
-------------------------------------
-Phonet Pipe-controller is enabled by selecting the CONFIG_PHONET_PIPECTRLR Kconfig
-option. It is useful when communicating with those Nokia Modems which do not
-implement Pipe controller in them e.g. Nokia Slim Modem used in ST-Ericsson
-U8500 platform.
-
-The implementation is based on the Data Connection Establishment Sequence
-depicted in 'Nokia Wireless Modem API - Wireless_modem_user_guide.pdf'
-document.
-
-It allows a phonet sequenced socket (host-pep) to initiate a Pipe connection
-between itself and a remote pipe-end point (e.g. modem).
+Phonet Pipe-controller is enabled by selecting the CONFIG_PHONET_PIPECTRLR
+Kconfig option.
The implementation adds socket options at SOL_PNPIPE level:
@@ -207,21 +217,6 @@ The implementation adds socket options at SOL_PNPIPE level:
is disabled. If the value is non-zero, the pipe is enabled. If the pipe
is not (yet) connected, ENOTCONN is error is returned.
-The implementation also adds socket 'connect'. On calling the 'connect', pipe
-will be created between the source socket and the destination, and the pipe
-state will be set to PIPE_DISABLED.
-
-After a pipe has been created and enabled successfully, the Pipe data can be
-exchanged between the host-pep and remote-pep (modem).
-
-User-space would typically follow below sequence with Pipe controller:-
--socket
--bind
--setsockopt for PNPIPE_PIPE_HANDLE
--connect
--setsockopt for PNPIPE_ENCAP_IP
--setsockopt for PNPIPE_ENABLE
-
Authors
-------
diff --git a/net/phonet/pep.c b/net/phonet/pep.c
index abfb795..671effb 100644
--- a/net/phonet/pep.c
+++ b/net/phonet/pep.c
@@ -136,7 +136,6 @@ static int pep_indicate(struct sock *sk, u8 id, u8 code,
#define PAD 0x00
-#ifdef CONFIG_PHONET_PIPECTRLR
static int pipe_handler_request(struct sock *sk, u8 id, u8 code,
const void *data, int len)
{
@@ -168,11 +167,7 @@ static int pipe_handler_send_created_ind(struct sock *sk)
data, 4, GFP_ATOMIC);
}
-static int pipe_handler_send_ind(struct sock *sk, u8 id)
-{
- return pep_indicate(sk, id, PAD, NULL, 0, GFP_ATOMIC);
-}
-
+#ifdef CONFIG_PHONET_PIPECTRLR
static int pipe_handler_enable_pipe(struct sock *sk, int enable)
{
u8 id = enable ? PNS_PEP_ENABLE_REQ : PNS_PEP_DISABLE_REQ;
@@ -376,32 +371,11 @@ static int pipe_do_rcv(struct sock *sk, struct sk_buff *skb)
sk->sk_state_change(sk);
break;
-#ifdef CONFIG_PHONET_PIPECTRLR
- case PNS_PEP_DISCONNECT_RESP:
- sk->sk_state = TCP_CLOSE;
- break;
-#endif
-
case PNS_PEP_ENABLE_REQ:
/* Wait for PNS_PIPE_(ENABLED|REDIRECTED)_IND */
pep_reply(sk, skb, PN_PIPE_NO_ERROR, NULL, 0, GFP_ATOMIC);
break;
-#ifdef CONFIG_PHONET_PIPECTRLR
- case PNS_PEP_ENABLE_RESP:
- pipe_handler_send_ind(sk, PNS_PIPE_ENABLED_IND);
-
- if (!pn_flow_safe(pn->tx_fc)) {
- atomic_set(&pn->tx_credits, 1);
- sk->sk_write_space(sk);
- }
- if (sk->sk_state == TCP_ESTABLISHED)
- break; /* Nothing to do */
- sk->sk_state = TCP_ESTABLISHED;
- pipe_grant_credits(sk, GFP_ATOMIC);
- break;
-#endif
-
case PNS_PEP_RESET_REQ:
switch (hdr->state_after_reset) {
case PN_PIPE_DISABLE:
@@ -420,15 +394,6 @@ static int pipe_do_rcv(struct sock *sk, struct sk_buff *skb)
pep_reply(sk, skb, PN_PIPE_NO_ERROR, NULL, 0, GFP_ATOMIC);
break;
-#ifdef CONFIG_PHONET_PIPECTRLR
- case PNS_PEP_DISABLE_RESP:
- atomic_set(&pn->tx_credits, 0);
- pipe_handler_send_ind(sk, PNS_PIPE_DISABLED_IND);
- sk->sk_state = TCP_SYN_RECV;
- pn->rx_credits = 0;
- break;
-#endif
-
case PNS_PEP_CTRL_REQ:
if (skb_queue_len(&pn->ctrlreq_queue) >= PNPIPE_CTRLREQ_MAX) {
atomic_inc(&sk->sk_drops);
@@ -521,7 +486,6 @@ static void pipe_destruct(struct sock *sk)
skb_queue_purge(&pn->ctrlreq_queue);
}
-#ifdef CONFIG_PHONET_PIPECTRLR
static u8 pipe_negotiate_fc(const u8 *fcs, unsigned n)
{
unsigned i;
@@ -546,6 +510,8 @@ static int pep_connresp_rcv(struct sock *sk, struct sk_buff *skb)
return -EINVAL;
hdr = pnp_hdr(skb);
+ if (hdr->error_code != PN_PIPE_NO_ERROR)
+ return -ECONNREFUSED;
/* Parse sub-blocks */
n_sb = hdr->data[4];
@@ -573,14 +539,74 @@ static int pep_connresp_rcv(struct sock *sk, struct sk_buff *skb)
n_sb--;
}
- sk->sk_state = TCP_SYN_RECV;
- sk->sk_backlog_rcv = pipe_do_rcv;
- pn->rx_credits = 0;
- sk->sk_state_change(sk);
-
return pipe_handler_send_created_ind(sk);
}
-#endif
+
+/* Queue an skb to an actively connected sock.
+ * Socket lock must be held. */
+static int pipe_handler_do_rcv(struct sock *sk, struct sk_buff *skb)
+{
+ struct pep_sock *pn = pep_sk(sk);
+ struct pnpipehdr *hdr = pnp_hdr(skb);
+ int err = NET_RX_SUCCESS;
+
+ switch (hdr->message_id) {
+ case PNS_PIPE_ALIGNED_DATA:
+ __skb_pull(skb, 1);
+ /* fall through */
+ case PNS_PIPE_DATA:
+ __skb_pull(skb, 3); /* Pipe data header */
+ if (!pn_flow_safe(pn->rx_fc)) {
+ err = sock_queue_rcv_skb(sk, skb);
+ if (!err)
+ return NET_RX_SUCCESS;
+ err = NET_RX_DROP;
+ break;
+ }
+
+ if (pn->rx_credits == 0) {
+ atomic_inc(&sk->sk_drops);
+ err = NET_RX_DROP;
+ break;
+ }
+ pn->rx_credits--;
+ skb->dev = NULL;
+ skb_set_owner_r(skb, sk);
+ err = skb->len;
+ skb_queue_tail(&sk->sk_receive_queue, skb);
+ if (!sock_flag(sk, SOCK_DEAD))
+ sk->sk_data_ready(sk, err);
+ return NET_RX_SUCCESS;
+
+ case PNS_PEP_CONNECT_RESP:
+ if (sk->sk_state != TCP_SYN_SENT)
+ break;
+ if (!sock_flag(sk, SOCK_DEAD))
+ sk->sk_state_change(sk);
+ if (pep_connresp_rcv(sk, skb)) {
+ sk->sk_state = TCP_CLOSE_WAIT;
+ break;
+ }
+
+ sk->sk_state = TCP_ESTABLISHED;
+ if (!pn_flow_safe(pn->tx_fc)) {
+ atomic_set(&pn->tx_credits, 1);
+ sk->sk_write_space(sk);
+ }
+ pipe_grant_credits(sk, GFP_ATOMIC);
+ break;
+
+ case PNS_PEP_DISCONNECT_RESP:
+ /* sock should already be dead, nothing to do */
+ break;
+
+ case PNS_PEP_STATUS_IND:
+ pipe_rcv_status(sk, skb);
+ break;
+ }
+ kfree_skb(skb);
+ return err;
+}
/* Listening sock must be locked */
static struct sock *pep_find_pipe(const struct hlist_head *hlist,
@@ -649,12 +675,6 @@ static int pep_do_rcv(struct sock *sk, struct sk_buff *skb)
sk->sk_data_ready(sk, 0);
return NET_RX_SUCCESS;
-#ifdef CONFIG_PHONET_PIPECTRLR
- case PNS_PEP_CONNECT_RESP:
- pep_connresp_rcv(sk, skb);
- break;
-#endif
-
case PNS_PEP_DISCONNECT_REQ:
pep_reply(sk, skb, PN_PIPE_NO_ERROR, NULL, 0, GFP_ATOMIC);
break;
@@ -667,15 +687,19 @@ static int pep_do_rcv(struct sock *sk, struct sk_buff *skb)
case PNS_PEP_ENABLE_REQ:
case PNS_PEP_DISABLE_REQ:
/* invalid handle is not even allowed here! */
- default:
break;
+
+ default:
+ if ((1 << sk->sk_state)
+ & ~(TCPF_CLOSE|TCPF_LISTEN|TCPF_CLOSE_WAIT))
+ /* actively connected socket */
+ return pipe_handler_do_rcv(sk, skb);
}
drop:
kfree_skb(skb);
return NET_RX_SUCCESS;
}
-#ifndef CONFIG_PHONET_PIPECTRLR
static int pipe_do_remove(struct sock *sk)
{
struct pep_sock *pn = pep_sk(sk);
@@ -693,7 +717,6 @@ static int pipe_do_remove(struct sock *sk)
ph->data[0] = PAD;
return pn_skb_send(sk, skb, NULL);
}
-#endif
/* associated socket ceases to exist */
static void pep_sock_close(struct sock *sk, long timeout)
@@ -706,13 +729,12 @@ static void pep_sock_close(struct sock *sk, long timeout)
lock_sock(sk);
if ((1 << sk->sk_state) & (TCPF_SYN_RECV|TCPF_ESTABLISHED)) {
-#ifndef CONFIG_PHONET_PIPECTRLR
- /* Forcefully remove dangling Phonet pipe */
- pipe_do_remove(sk);
-#else
- /* send pep disconnect request */
- pipe_handler_request(sk, PNS_PEP_DISCONNECT_REQ, PAD, NULL, 0);
-#endif
+ if (sk->sk_backlog_rcv == pipe_do_rcv)
+ /* Forcefully remove dangling Phonet pipe */
+ pipe_do_remove(sk);
+ else
+ pipe_handler_request(sk, PNS_PEP_DISCONNECT_REQ, PAD,
+ NULL, 0);
}
sk->sk_state = TCP_CLOSE;
@@ -844,20 +866,22 @@ drop:
return newsk;
}
-#ifdef CONFIG_PHONET_PIPECTRLR
static int pep_sock_connect(struct sock *sk, struct sockaddr *addr, int len)
{
struct pep_sock *pn = pep_sk(sk);
- const struct sockaddr_pn *spn = (struct sockaddr_pn *)addr;
+ int err;
u8 data[4] = { 0 /* sub-blocks */, PAD, PAD, PAD };
- pn->pn_sk.dobject = pn_sockaddr_get_object(spn);
- pn->pn_sk.resource = pn_sockaddr_get_resource(spn);
pn->pipe_handle = 1; /* anything but INVALID_HANDLE */
- return pipe_handler_request(sk, PNS_PEP_CONNECT_REQ,
- PN_PIPE_DISABLE, data, 4);
+ err = pipe_handler_request(sk, PNS_PEP_CONNECT_REQ,
+ PN_PIPE_ENABLE, data, 4);
+ if (err) {
+ pn->pipe_handle = PN_PIPE_INVALID_HANDLE;
+ return err;
+ }
+ sk->sk_state = TCP_SYN_SENT;
+ return 0;
}
-#endif
static int pep_ioctl(struct sock *sk, int cmd, unsigned long arg)
{
@@ -890,8 +914,16 @@ static int pep_init(struct sock *sk)
sk->sk_destruct = pipe_destruct;
INIT_HLIST_HEAD(&pn->hlist);
+ pn->listener = NULL;
skb_queue_head_init(&pn->ctrlreq_queue);
+ atomic_set(&pn->tx_credits, 0);
+ pn->ifindex = 0;
+ pn->peer_type = 0;
pn->pipe_handle = PN_PIPE_INVALID_HANDLE;
+ pn->rx_credits = 0;
+ pn->rx_fc = pn->tx_fc = PN_LEGACY_FLOW_CONTROL;
+ pn->init_enable = 1;
+ pn->aligned = 0;
return 0;
}
@@ -1219,9 +1251,9 @@ static void pep_sock_unhash(struct sock *sk)
lock_sock(sk);
-#ifndef CONFIG_PHONET_PIPECTRLR
- if ((1 << sk->sk_state) & ~(TCPF_CLOSE|TCPF_LISTEN)) {
+ if (pn->listener != NULL) {
skparent = pn->listener;
+ pn->listener = NULL;
release_sock(sk);
pn = pep_sk(skparent);
@@ -1229,7 +1261,7 @@ static void pep_sock_unhash(struct sock *sk)
sk_del_node_init(sk);
sk = skparent;
}
-#endif
+
/* Unhash a listening sock only when it is closed
* and all of its active connected pipes are closed. */
if (hlist_empty(&pn->hlist))
@@ -1243,9 +1275,7 @@ static void pep_sock_unhash(struct sock *sk)
static struct proto pep_proto = {
.close = pep_sock_close,
.accept = pep_sock_accept,
-#ifdef CONFIG_PHONET_PIPECTRLR
.connect = pep_sock_connect,
-#endif
.ioctl = pep_ioctl,
.init = pep_init,
.setsockopt = pep_setsockopt,
diff --git a/net/phonet/socket.c b/net/phonet/socket.c
index 1eccfc3..b1adafa 100644
--- a/net/phonet/socket.c
+++ b/net/phonet/socket.c
@@ -225,15 +225,18 @@ static int pn_socket_autobind(struct socket *sock)
return 0; /* socket was already bound */
}
-#ifdef CONFIG_PHONET_PIPECTRLR
static int pn_socket_connect(struct socket *sock, struct sockaddr *addr,
int len, int flags)
{
struct sock *sk = sock->sk;
+ struct pn_sock *pn = pn_sk(sk);
struct sockaddr_pn *spn = (struct sockaddr_pn *)addr;
- long timeo;
+ struct task_struct *tsk = current;
+ long timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK);
int err;
+ if (pn_socket_autobind(sock))
+ return -ENOBUFS;
if (len < sizeof(struct sockaddr_pn))
return -EINVAL;
if (spn->spn_family != AF_PHONET)
@@ -243,82 +246,61 @@ static int pn_socket_connect(struct socket *sock, struct sockaddr *addr,
switch (sock->state) {
case SS_UNCONNECTED:
- sk->sk_state = TCP_CLOSE;
- break;
- case SS_CONNECTING:
- switch (sk->sk_state) {
- case TCP_SYN_RECV:
- sock->state = SS_CONNECTED;
- err = -EISCONN;
- goto out;
- case TCP_CLOSE:
- err = -EALREADY;
- if (flags & O_NONBLOCK)
- goto out;
- goto wait_connect;
- }
- break;
- case SS_CONNECTED:
- switch (sk->sk_state) {
- case TCP_SYN_RECV:
+ if (sk->sk_state != TCP_CLOSE) {
err = -EISCONN;
goto out;
- case TCP_CLOSE:
- sock->state = SS_UNCONNECTED;
- break;
}
break;
- case SS_DISCONNECTING:
- case SS_FREE:
- break;
+ case SS_CONNECTING:
+ err = -EALREADY;
+ goto out;
+ default:
+ err = -EISCONN;
+ goto out;
}
- sk->sk_state = TCP_CLOSE;
- sk_stream_kill_queues(sk);
+ pn->dobject = pn_sockaddr_get_object(spn);
+ pn->resource = pn_sockaddr_get_resource(spn);
sock->state = SS_CONNECTING;
+
err = sk->sk_prot->connect(sk, addr, len);
- if (err < 0) {
+ if (err) {
sock->state = SS_UNCONNECTED;
- sk->sk_state = TCP_CLOSE;
+ pn->dobject = 0;
goto out;
}
- err = -EINPROGRESS;
-wait_connect:
- if (sk->sk_state != TCP_SYN_RECV && (flags & O_NONBLOCK))
- goto out;
-
- timeo = sock_sndtimeo(sk, flags & O_NONBLOCK);
- release_sock(sk);
+ while (sk->sk_state == TCP_SYN_SENT) {
+ DEFINE_WAIT(wait);
- err = -ERESTARTSYS;
- timeo = wait_event_interruptible_timeout(*sk_sleep(sk),
- sk->sk_state != TCP_CLOSE,
- timeo);
-
- lock_sock(sk);
- if (timeo < 0)
- goto out; /* -ERESTARTSYS */
-
- err = -ETIMEDOUT;
- if (timeo == 0 && sk->sk_state != TCP_SYN_RECV)
- goto out;
+ if (!timeo) {
+ err = -EINPROGRESS;
+ goto out;
+ }
+ if (signal_pending(tsk)) {
+ err = sock_intr_errno(timeo);
+ goto out;
+ }
- if (sk->sk_state != TCP_SYN_RECV) {
- sock->state = SS_UNCONNECTED;
- err = sock_error(sk);
- if (!err)
- err = -ECONNREFUSED;
- goto out;
+ prepare_to_wait_exclusive(sk_sleep(sk), &wait,
+ TASK_INTERRUPTIBLE);
+ release_sock(sk);
+ timeo = schedule_timeout(timeo);
+ lock_sock(sk);
+ finish_wait(sk_sleep(sk), &wait);
}
- sock->state = SS_CONNECTED;
- err = 0;
+ if ((1 << sk->sk_state) & (TCPF_SYN_RECV|TCPF_ESTABLISHED))
+ err = 0;
+ else if (sk->sk_state == TCP_CLOSE_WAIT)
+ err = -ECONNRESET;
+ else
+ err = -ECONNREFUSED;
+ sock->state = err ? SS_UNCONNECTED : SS_CONNECTED;
out:
release_sock(sk);
return err;
}
-#endif
static int pn_socket_accept(struct socket *sock, struct socket *newsock,
int flags)
@@ -486,11 +468,7 @@ const struct proto_ops phonet_stream_ops = {
.owner = THIS_MODULE,
.release = pn_socket_release,
.bind = pn_socket_bind,
-#ifdef CONFIG_PHONET_PIPECTRLR
.connect = pn_socket_connect,
-#else
- .connect = sock_no_connect,
-#endif
.socketpair = sock_no_socketpair,
.accept = pn_socket_accept,
.getname = pn_socket_getname,
--
1.7.1
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox