* [PATCH v2 08/16] mlx4_en: Reporting HW revision in ethtool -i
From: Yevgeny Petrilin @ 2011-03-23 8:38 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp, eugenia
HW revision is derived from device ID and rev id.
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
drivers/infiniband/hw/mlx4/main.c | 1 -
drivers/net/mlx4/en_ethtool.c | 15 ++++++++++++++-
drivers/net/mlx4/main.c | 2 ++
drivers/net/mlx4/mlx4_en.h | 3 +++
include/linux/mlx4/device.h | 2 +-
5 files changed, 20 insertions(+), 3 deletions(-)
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index c7a6213..66e3eec 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -721,7 +721,6 @@ static int init_node_data(struct mlx4_ib_dev *dev)
if (err)
goto out;
- dev->dev->rev_id = be32_to_cpup((__be32 *) (out_mad->data + 32));
memcpy(&dev->ib_dev.node_guid, out_mad->data + 12, 8);
out:
diff --git a/drivers/net/mlx4/en_ethtool.c b/drivers/net/mlx4/en_ethtool.c
index c1f351f..62ace6c 100644
--- a/drivers/net/mlx4/en_ethtool.c
+++ b/drivers/net/mlx4/en_ethtool.c
@@ -45,7 +45,20 @@ mlx4_en_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *drvinfo)
struct mlx4_en_priv *priv = netdev_priv(dev);
struct mlx4_en_dev *mdev = priv->mdev;
- sprintf(drvinfo->driver, DRV_NAME " (%s)", mdev->dev->board_id);
+ switch (mdev->dev->rev_id) {
+ case 0xa0:
+ if (dev->dev_id >= MLX4_EN_CX3_LOW_ID && dev->dev_id <= MLX4_EN_CX3_HIGH_ID)
+ sprintf(drvinfo->driver, DRV_NAME " (%s_CX-3)", mdev->dev->board_id);
+ else
+ sprintf(drvinfo->driver, DRV_NAME " (%s_CX)", mdev->dev->board_id);
+ break;
+ case 0xb0:
+ sprintf(drvinfo->driver, DRV_NAME " (%s_CX-2)", mdev->dev->board_id);
+ break;
+ default:
+ sprintf(drvinfo->driver, DRV_NAME " (%s)", mdev->dev->board_id);
+ break;
+ }
strncpy(drvinfo->version, DRV_VERSION " (" DRV_RELDATE ")", 32);
sprintf(drvinfo->fw_version, "%d.%d.%d",
(u16) (mdev->dev->caps.fw_ver >> 32),
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index 42d4fb4..5bebb88 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -1139,6 +1139,8 @@ static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
INIT_LIST_HEAD(&priv->pgdir_list);
mutex_init(&priv->pgdir_mutex);
+ pci_read_config_byte(pdev, PCI_REVISION_ID, &dev->rev_id);
+
/*
* Now reset the HCA before we touch the PCI capabilities or
* attempt a firmware command, since a boot ROM may have left
diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h
index 07aea8d..ad4df66 100644
--- a/drivers/net/mlx4/mlx4_en.h
+++ b/drivers/net/mlx4/mlx4_en.h
@@ -216,6 +216,9 @@ struct mlx4_en_tx_desc {
#define MLX4_EN_USE_SRQ 0x01000000
+#define MLX4_EN_CX3_LOW_ID 0x1000
+#define MLX4_EN_CX3_HIGH_ID 0x1005
+
struct mlx4_en_rx_alloc {
struct page *page;
u16 offset;
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 2460356..fe2a3a3 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -422,7 +422,7 @@ struct mlx4_dev {
unsigned long flags;
struct mlx4_caps caps;
struct radix_tree_root qp_table_tree;
- u32 rev_id;
+ u8 rev_id;
char board_id[MLX4_BOARD_ID_LEN];
};
--
1.6.0.2
^ permalink raw reply related
* [PATCH v2 09/16] mlx4: generalization of multicast steering.
From: Yevgeny Petrilin @ 2011-03-23 8:38 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp
The same packet steering mechanism would be used both for IB and Ethernet,
Both multicasts and unicasts.
This commit prepares the general infrastructure for this.
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
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 v2 10/16] mlx4: Add support for promiscuous mode in the new steering model.
From: Yevgeny Petrilin @ 2011-03-23 8:38 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp
For Ethernet mode only,
When we want to register QP as promiscuous, it must be added to all the
existing steering entries and also to the default one.
The promiscuous QP might also be on of "real" QPs,
which means we need to monitor every entry to avoid duplicates and ensure
we close an entry when all it has is promiscuous QPs.
Same mechanism both for unicast and multicast.
Signed-off-by: Yevgeny Petrilin <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 v2 11/16] mlx4_en: Enabling new steering
From: Yevgeny Petrilin @ 2011-03-23 8:38 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp
The mlx4_en module now uses the new steering mechanism.
The RX packets are now steered through the MCG table instead
of Mac table for unicast, and default entry for multicast.
The feature is enabled through INIT_HCA
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
drivers/net/mlx4/en_netdev.c | 133 +++++++++++++++++++++++++++-------
drivers/net/mlx4/en_port.c | 11 +++-
drivers/net/mlx4/en_port.h | 19 ++++--
drivers/net/mlx4/en_rx.c | 11 +---
drivers/net/mlx4/fw.c | 3 +
drivers/net/mlx4/mlx4.h | 6 ++
drivers/net/mlx4/mlx4_en.h | 1 +
drivers/net/mlx4/port.c | 165 ++++++++++++++++++++++++++++++++++++++----
include/linux/mlx4/device.h | 12 +++-
9 files changed, 300 insertions(+), 61 deletions(-)
diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c
index f6ed315..08e6801 100644
--- a/drivers/net/mlx4/en_netdev.c
+++ b/drivers/net/mlx4/en_netdev.c
@@ -156,9 +156,8 @@ static void mlx4_en_do_set_mac(struct work_struct *work)
mutex_lock(&mdev->state_lock);
if (priv->port_up) {
/* Remove old MAC and insert the new one */
- mlx4_unregister_mac(mdev->dev, priv->port, priv->mac_index);
- err = mlx4_register_mac(mdev->dev, priv->port,
- priv->mac, &priv->mac_index);
+ err = mlx4_replace_mac(mdev->dev, priv->port,
+ priv->base_qpn, priv->mac, 0);
if (err)
en_err(priv, "Failed changing HW MAC address\n");
} else
@@ -214,6 +213,7 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
struct mlx4_en_dev *mdev = priv->mdev;
struct net_device *dev = priv->dev;
u64 mcast_addr = 0;
+ u8 mc_list[16] = {0};
int err;
mutex_lock(&mdev->state_lock);
@@ -239,8 +239,12 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
priv->flags |= MLX4_EN_FLAG_PROMISC;
/* Enable promiscouos mode */
- err = mlx4_SET_PORT_qpn_calc(mdev->dev, priv->port,
- priv->base_qpn, 1);
+ if (!mdev->dev->caps.vep_uc_steering)
+ err = mlx4_SET_PORT_qpn_calc(mdev->dev, priv->port,
+ priv->base_qpn, 1);
+ else
+ err = mlx4_unicast_promisc_add(mdev->dev, priv->base_qpn,
+ priv->port);
if (err)
en_err(priv, "Failed enabling "
"promiscous mode\n");
@@ -252,10 +256,21 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
en_err(priv, "Failed disabling "
"multicast filter\n");
- /* Disable port VLAN filter */
- err = mlx4_SET_VLAN_FLTR(mdev->dev, priv->port, NULL);
- if (err)
- en_err(priv, "Failed disabling VLAN filter\n");
+ /* Add the default qp number as multicast promisc */
+ if (!(priv->flags & MLX4_EN_FLAG_MC_PROMISC)) {
+ err = mlx4_multicast_promisc_add(mdev->dev, priv->base_qpn,
+ priv->port);
+ if (err)
+ en_err(priv, "Failed entering multicast promisc mode\n");
+ priv->flags |= MLX4_EN_FLAG_MC_PROMISC;
+ }
+
+ if (priv->vlgrp) {
+ /* Disable port VLAN filter */
+ err = mlx4_SET_VLAN_FLTR(mdev->dev, priv->port, NULL);
+ if (err)
+ en_err(priv, "Failed disabling VLAN filter\n");
+ }
}
goto out;
}
@@ -270,11 +285,24 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
priv->flags &= ~MLX4_EN_FLAG_PROMISC;
/* Disable promiscouos mode */
- err = mlx4_SET_PORT_qpn_calc(mdev->dev, priv->port,
- priv->base_qpn, 0);
+ if (!mdev->dev->caps.vep_uc_steering)
+ err = mlx4_SET_PORT_qpn_calc(mdev->dev, priv->port,
+ priv->base_qpn, 0);
+ else
+ err = mlx4_unicast_promisc_remove(mdev->dev, priv->base_qpn,
+ priv->port);
if (err)
en_err(priv, "Failed disabling promiscous mode\n");
+ /* Disable Multicast promisc */
+ if (priv->flags & MLX4_EN_FLAG_MC_PROMISC) {
+ err = mlx4_multicast_promisc_remove(mdev->dev, priv->base_qpn,
+ priv->port);
+ if (err)
+ en_err(priv, "Failed disabling multicast promiscous mode\n");
+ priv->flags &= ~MLX4_EN_FLAG_MC_PROMISC;
+ }
+
/* Enable port VLAN filter */
err = mlx4_SET_VLAN_FLTR(mdev->dev, priv->port, priv->vlgrp);
if (err)
@@ -287,14 +315,38 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
0, MLX4_MCAST_DISABLE);
if (err)
en_err(priv, "Failed disabling multicast filter\n");
+
+ /* Add the default qp number as multicast promisc */
+ if (!(priv->flags & MLX4_EN_FLAG_MC_PROMISC)) {
+ err = mlx4_multicast_promisc_add(mdev->dev, priv->base_qpn,
+ priv->port);
+ if (err)
+ en_err(priv, "Failed entering multicast promisc mode\n");
+ priv->flags |= MLX4_EN_FLAG_MC_PROMISC;
+ }
} else {
int i;
+ /* Disable Multicast promisc */
+ if (priv->flags & MLX4_EN_FLAG_MC_PROMISC) {
+ err = mlx4_multicast_promisc_remove(mdev->dev, priv->base_qpn,
+ priv->port);
+ if (err)
+ en_err(priv, "Failed disabling multicast promiscous mode\n");
+ priv->flags &= ~MLX4_EN_FLAG_MC_PROMISC;
+ }
err = mlx4_SET_MCAST_FLTR(mdev->dev, priv->port, 0,
0, MLX4_MCAST_DISABLE);
if (err)
en_err(priv, "Failed disabling multicast filter\n");
+ /* Detach our qp from all the multicast addresses */
+ for (i = 0; i < priv->mc_addrs_cnt; i++) {
+ memcpy(&mc_list[10], priv->mc_addrs + i * ETH_ALEN, ETH_ALEN);
+ mc_list[5] = priv->port;
+ mlx4_multicast_detach(mdev->dev, &priv->rss_map.indir_qp,
+ mc_list, MLX4_PROT_ETH);
+ }
/* Flush mcast filter and init it with broadcast address */
mlx4_SET_MCAST_FLTR(mdev->dev, priv->port, ETH_BCAST,
1, MLX4_MCAST_CONFIG);
@@ -307,6 +359,10 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
for (i = 0; i < priv->mc_addrs_cnt; i++) {
mcast_addr =
mlx4_en_mac_to_u64(priv->mc_addrs + i * ETH_ALEN);
+ memcpy(&mc_list[10], priv->mc_addrs + i * ETH_ALEN, ETH_ALEN);
+ mc_list[5] = priv->port;
+ mlx4_multicast_attach(mdev->dev, &priv->rss_map.indir_qp,
+ mc_list, 0, MLX4_PROT_ETH);
mlx4_SET_MCAST_FLTR(mdev->dev, priv->port,
mcast_addr, 0, MLX4_MCAST_CONFIG);
}
@@ -314,8 +370,6 @@ static void mlx4_en_do_set_multicast(struct work_struct *work)
0, MLX4_MCAST_ENABLE);
if (err)
en_err(priv, "Failed enabling multicast filter\n");
-
- mlx4_en_clear_list(dev);
}
out:
mutex_unlock(&mdev->state_lock);
@@ -557,6 +611,7 @@ int mlx4_en_start_port(struct net_device *dev)
int err = 0;
int i;
int j;
+ u8 mc_list[16] = {0};
char name[32];
if (priv->port_up) {
@@ -596,10 +651,20 @@ int mlx4_en_start_port(struct net_device *dev)
++rx_index;
}
+ /* Set port mac number */
+ en_dbg(DRV, priv, "Setting mac for port %d\n", priv->port);
+ err = mlx4_register_mac(mdev->dev, priv->port,
+ priv->mac, &priv->base_qpn, 0);
+ if (err) {
+ en_err(priv, "Failed setting port mac\n");
+ goto cq_err;
+ }
+ mdev->mac_removed[priv->port] = 0;
+
err = mlx4_en_config_rss_steer(priv);
if (err) {
en_err(priv, "Failed configuring rss steering\n");
- goto cq_err;
+ goto mac_err;
}
if (mdev->dev->caps.comp_pool && !priv->tx_vector) {
@@ -661,24 +726,22 @@ int mlx4_en_start_port(struct net_device *dev)
en_err(priv, "Failed setting default qp numbers\n");
goto tx_err;
}
- /* Set port mac number */
- en_dbg(DRV, priv, "Setting mac for port %d\n", priv->port);
- err = mlx4_register_mac(mdev->dev, priv->port,
- priv->mac, &priv->mac_index);
- if (err) {
- en_err(priv, "Failed setting port mac\n");
- goto tx_err;
- }
- mdev->mac_removed[priv->port] = 0;
/* Init port */
en_dbg(HW, priv, "Initializing port\n");
err = mlx4_INIT_PORT(mdev->dev, priv->port);
if (err) {
en_err(priv, "Failed Initializing port\n");
- goto mac_err;
+ goto tx_err;
}
+ /* Attach rx QP to bradcast address */
+ memset(&mc_list[10], 0xff, ETH_ALEN);
+ mc_list[5] = priv->port;
+ if (mlx4_multicast_attach(mdev->dev, &priv->rss_map.indir_qp, mc_list,
+ 0, MLX4_PROT_ETH))
+ mlx4_warn(mdev, "Failed Attaching Broadcast\n");
+
/* Schedule multicast task to populate multicast list */
queue_work(mdev->workqueue, &priv->mcast_task);
@@ -686,8 +749,6 @@ int mlx4_en_start_port(struct net_device *dev)
netif_tx_start_all_queues(dev);
return 0;
-mac_err:
- mlx4_unregister_mac(mdev->dev, priv->port, priv->mac_index);
tx_err:
while (tx_index--) {
mlx4_en_deactivate_tx_ring(priv, &priv->tx_ring[tx_index]);
@@ -695,6 +756,8 @@ tx_err:
}
mlx4_en_release_rss_steer(priv);
+mac_err:
+ mlx4_unregister_mac(mdev->dev, priv->port, priv->base_qpn);
cq_err:
while (rx_index--)
mlx4_en_deactivate_cq(priv, &priv->rx_cq[rx_index]);
@@ -710,6 +773,7 @@ void mlx4_en_stop_port(struct net_device *dev)
struct mlx4_en_priv *priv = netdev_priv(dev);
struct mlx4_en_dev *mdev = priv->mdev;
int i;
+ u8 mc_list[16] = {0};
if (!priv->port_up) {
en_dbg(DRV, priv, "stop port called while port already down\n");
@@ -724,8 +788,23 @@ void mlx4_en_stop_port(struct net_device *dev)
/* Set port as not active */
priv->port_up = false;
+ /* Detach All multicasts */
+ memset(&mc_list[10], 0xff, ETH_ALEN);
+ mc_list[5] = priv->port;
+ mlx4_multicast_detach(mdev->dev, &priv->rss_map.indir_qp, mc_list,
+ MLX4_PROT_ETH);
+ for (i = 0; i < priv->mc_addrs_cnt; i++) {
+ memcpy(&mc_list[10], priv->mc_addrs + i * ETH_ALEN, ETH_ALEN);
+ mc_list[5] = priv->port;
+ mlx4_multicast_detach(mdev->dev, &priv->rss_map.indir_qp,
+ mc_list, MLX4_PROT_ETH);
+ }
+ mlx4_en_clear_list(dev);
+ /* Flush multicast filter */
+ mlx4_SET_MCAST_FLTR(mdev->dev, priv->port, 0, 1, MLX4_MCAST_CONFIG);
+
/* Unregister Mac address for the port */
- mlx4_unregister_mac(mdev->dev, priv->port, priv->mac_index);
+ mlx4_unregister_mac(mdev->dev, priv->port, priv->base_qpn);
mdev->mac_removed[priv->port] = 1;
/* Free TX Rings */
diff --git a/drivers/net/mlx4/en_port.c b/drivers/net/mlx4/en_port.c
index a2b0bc4..f2a4f5d 100644
--- a/drivers/net/mlx4/en_port.c
+++ b/drivers/net/mlx4/en_port.c
@@ -119,6 +119,10 @@ int mlx4_SET_PORT_qpn_calc(struct mlx4_dev *dev, u8 port, u32 base_qpn,
struct mlx4_set_port_rqp_calc_context *context;
int err;
u32 in_mod;
+ u32 m_promisc = (dev->caps.vep_mc_steering) ? MCAST_DIRECT : MCAST_DEFAULT;
+
+ if (dev->caps.vep_mc_steering && dev->caps.vep_uc_steering)
+ return 0;
mailbox = mlx4_alloc_cmd_mailbox(dev);
if (IS_ERR(mailbox))
@@ -127,8 +131,11 @@ int mlx4_SET_PORT_qpn_calc(struct mlx4_dev *dev, u8 port, u32 base_qpn,
memset(context, 0, sizeof *context);
context->base_qpn = cpu_to_be32(base_qpn);
- context->promisc = cpu_to_be32(promisc << SET_PORT_PROMISC_EN_SHIFT | base_qpn);
- context->mcast = cpu_to_be32(1 << SET_PORT_PROMISC_MODE_SHIFT | base_qpn);
+ context->n_mac = 0x7;
+ context->promisc = cpu_to_be32(promisc << SET_PORT_PROMISC_SHIFT |
+ base_qpn);
+ context->mcast = cpu_to_be32(m_promisc << SET_PORT_MC_PROMISC_SHIFT |
+ base_qpn);
context->intra_no_vlan = 0;
context->no_vlan = MLX4_NO_VLAN_IDX;
context->intra_vlan_miss = 0;
diff --git a/drivers/net/mlx4/en_port.h b/drivers/net/mlx4/en_port.h
index 092e814..e3d73e4 100644
--- a/drivers/net/mlx4/en_port.h
+++ b/drivers/net/mlx4/en_port.h
@@ -36,8 +36,8 @@
#define SET_PORT_GEN_ALL_VALID 0x7
-#define SET_PORT_PROMISC_EN_SHIFT 31
-#define SET_PORT_PROMISC_MODE_SHIFT 30
+#define SET_PORT_PROMISC_SHIFT 31
+#define SET_PORT_MC_PROMISC_SHIFT 30
enum {
MLX4_CMD_SET_VLAN_FLTR = 0x47,
@@ -45,6 +45,12 @@ enum {
MLX4_CMD_DUMP_ETH_STATS = 0x49,
};
+enum {
+ MCAST_DIRECT_ONLY = 0,
+ MCAST_DIRECT = 1,
+ MCAST_DEFAULT = 2
+};
+
struct mlx4_set_port_general_context {
u8 reserved[3];
u8 flags;
@@ -60,14 +66,17 @@ struct mlx4_set_port_general_context {
struct mlx4_set_port_rqp_calc_context {
__be32 base_qpn;
- __be32 flags;
- u8 reserved[3];
+ u8 rererved;
+ u8 n_mac;
+ u8 n_vlan;
+ u8 n_prio;
+ u8 reserved2[3];
u8 mac_miss;
u8 intra_no_vlan;
u8 no_vlan;
u8 intra_vlan_miss;
u8 vlan_miss;
- u8 reserved2[3];
+ u8 reserved3[3];
u8 no_vlan_prio;
__be32 promisc;
__be32 mcast;
diff --git a/drivers/net/mlx4/en_rx.c b/drivers/net/mlx4/en_rx.c
index 570f250..05998ee 100644
--- a/drivers/net/mlx4/en_rx.c
+++ b/drivers/net/mlx4/en_rx.c
@@ -845,16 +845,10 @@ int mlx4_en_config_rss_steer(struct mlx4_en_priv *priv)
}
/* Configure RSS indirection qp */
- err = mlx4_qp_reserve_range(mdev->dev, 1, 1, &priv->base_qpn);
- if (err) {
- en_err(priv, "Failed to reserve range for RSS "
- "indirection qp\n");
- goto rss_err;
- }
err = mlx4_qp_alloc(mdev->dev, priv->base_qpn, &rss_map->indir_qp);
if (err) {
en_err(priv, "Failed to allocate RSS indirection QP\n");
- goto reserve_err;
+ goto rss_err;
}
rss_map->indir_qp.event = mlx4_en_sqp_event;
mlx4_en_fill_qp_context(priv, 0, 0, 0, 1, priv->base_qpn,
@@ -881,8 +875,6 @@ indir_err:
MLX4_QP_STATE_RST, NULL, 0, 0, &rss_map->indir_qp);
mlx4_qp_remove(mdev->dev, &rss_map->indir_qp);
mlx4_qp_free(mdev->dev, &rss_map->indir_qp);
-reserve_err:
- mlx4_qp_release_range(mdev->dev, priv->base_qpn, 1);
rss_err:
for (i = 0; i < good_qps; i++) {
mlx4_qp_modify(mdev->dev, NULL, rss_map->state[i],
@@ -904,7 +896,6 @@ void mlx4_en_release_rss_steer(struct mlx4_en_priv *priv)
MLX4_QP_STATE_RST, NULL, 0, 0, &rss_map->indir_qp);
mlx4_qp_remove(mdev->dev, &rss_map->indir_qp);
mlx4_qp_free(mdev->dev, &rss_map->indir_qp);
- mlx4_qp_release_range(mdev->dev, priv->base_qpn, 1);
for (i = 0; i < priv->rx_ring_num; i++) {
mlx4_qp_modify(mdev->dev, NULL, rss_map->state[i],
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index 7813913..67a209b 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -740,6 +740,7 @@ int mlx4_INIT_HCA(struct mlx4_dev *dev, struct mlx4_init_hca_param *param)
#define INIT_HCA_MC_BASE_OFFSET (INIT_HCA_MCAST_OFFSET + 0x00)
#define INIT_HCA_LOG_MC_ENTRY_SZ_OFFSET (INIT_HCA_MCAST_OFFSET + 0x12)
#define INIT_HCA_LOG_MC_HASH_SZ_OFFSET (INIT_HCA_MCAST_OFFSET + 0x16)
+#define INIT_HCA_UC_STEERING_OFFSET (INIT_HCA_MCAST_OFFSET + 0x18)
#define INIT_HCA_LOG_MC_TABLE_SZ_OFFSET (INIT_HCA_MCAST_OFFSET + 0x1b)
#define INIT_HCA_TPT_OFFSET 0x0f0
#define INIT_HCA_DMPT_BASE_OFFSET (INIT_HCA_TPT_OFFSET + 0x00)
@@ -800,6 +801,8 @@ int mlx4_INIT_HCA(struct mlx4_dev *dev, struct mlx4_init_hca_param *param)
MLX4_PUT(inbox, param->mc_base, INIT_HCA_MC_BASE_OFFSET);
MLX4_PUT(inbox, param->log_mc_entry_sz, INIT_HCA_LOG_MC_ENTRY_SZ_OFFSET);
MLX4_PUT(inbox, param->log_mc_hash_sz, INIT_HCA_LOG_MC_HASH_SZ_OFFSET);
+ if (dev->caps.vep_mc_steering)
+ MLX4_PUT(inbox, (u8) (1 << 3), INIT_HCA_UC_STEERING_OFFSET);
MLX4_PUT(inbox, param->log_mc_table_sz, INIT_HCA_LOG_MC_TABLE_SZ_OFFSET);
/* TPT attributes */
diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h
index 1513a91..d8bb441 100644
--- a/drivers/net/mlx4/mlx4.h
+++ b/drivers/net/mlx4/mlx4.h
@@ -286,6 +286,10 @@ struct mlx4_vlan_table {
int max;
};
+struct mlx4_mac_entry {
+ u64 mac;
+};
+
struct mlx4_port_info {
struct mlx4_dev *dev;
int port;
@@ -293,7 +297,9 @@ struct mlx4_port_info {
struct device_attribute port_attr;
enum mlx4_port_type tmp_type;
struct mlx4_mac_table mac_table;
+ struct radix_tree_root mac_tree;
struct mlx4_vlan_table vlan_table;
+ int base_qpn;
};
struct mlx4_sense {
diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h
index ad4df66..5a2c560 100644
--- a/drivers/net/mlx4/mlx4_en.h
+++ b/drivers/net/mlx4/mlx4_en.h
@@ -458,6 +458,7 @@ struct mlx4_en_priv {
struct mlx4_en_rss_map rss_map;
u32 flags;
#define MLX4_EN_FLAG_PROMISC 0x1
+#define MLX4_EN_FLAG_MC_PROMISC 0x2
u32 tx_ring_num;
u32 rx_ring_num;
u32 rx_skb_size;
diff --git a/drivers/net/mlx4/port.c b/drivers/net/mlx4/port.c
index 4513395..eca7d85 100644
--- a/drivers/net/mlx4/port.c
+++ b/drivers/net/mlx4/port.c
@@ -90,12 +90,79 @@ static int mlx4_set_port_mac_table(struct mlx4_dev *dev, u8 port,
return err;
}
-int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *index)
+static int mlx4_uc_steer_add(struct mlx4_dev *dev, u8 port,
+ u64 mac, int *qpn, u8 reserve)
{
- struct mlx4_mac_table *table = &mlx4_priv(dev)->port[port].mac_table;
+ struct mlx4_qp qp;
+ u8 gid[16] = {0};
+ int err;
+
+ if (reserve) {
+ err = mlx4_qp_reserve_range(dev, 1, 1, qpn);
+ if (err) {
+ mlx4_err(dev, "Failed to reserve qp for mac registration\n");
+ return err;
+ }
+ }
+ qp.qpn = *qpn;
+
+ mac &= 0xffffffffffffULL;
+ mac = cpu_to_be64(mac << 16);
+ memcpy(&gid[10], &mac, ETH_ALEN);
+ gid[5] = port;
+ gid[7] = MLX4_UC_STEER << 1;
+
+ err = mlx4_qp_attach_common(dev, &qp, gid, 0,
+ MLX4_PROT_ETH, MLX4_UC_STEER);
+ if (err && reserve)
+ mlx4_qp_release_range(dev, *qpn, 1);
+
+ return err;
+}
+
+static void mlx4_uc_steer_release(struct mlx4_dev *dev, u8 port,
+ u64 mac, int qpn, u8 free)
+{
+ struct mlx4_qp qp;
+ u8 gid[16] = {0};
+
+ qp.qpn = qpn;
+ mac &= 0xffffffffffffULL;
+ mac = cpu_to_be64(mac << 16);
+ memcpy(&gid[10], &mac, ETH_ALEN);
+ gid[5] = port;
+ gid[7] = MLX4_UC_STEER << 1;
+
+ mlx4_qp_detach_common(dev, &qp, gid, MLX4_PROT_ETH, MLX4_UC_STEER);
+ if (free)
+ mlx4_qp_release_range(dev, qpn, 1);
+}
+
+int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *qpn, u8 wrap)
+{
+ struct mlx4_port_info *info = &mlx4_priv(dev)->port[port];
+ struct mlx4_mac_table *table = &info->mac_table;
+ struct mlx4_mac_entry *entry;
int i, err = 0;
int free = -1;
+ if (dev->caps.vep_uc_steering) {
+ err = mlx4_uc_steer_add(dev, port, mac, qpn, 1);
+ if (!err) {
+ entry = kmalloc(sizeof *entry, GFP_KERNEL);
+ if (!entry) {
+ mlx4_uc_steer_release(dev, port, mac, *qpn, 1);
+ return -ENOMEM;
+ }
+ entry->mac = mac;
+ err = radix_tree_insert(&info->mac_tree, *qpn, entry);
+ if (err) {
+ mlx4_uc_steer_release(dev, port, mac, *qpn, 1);
+ return err;
+ }
+ } else
+ return err;
+ }
mlx4_dbg(dev, "Registering MAC: 0x%llx\n", (unsigned long long) mac);
mutex_lock(&table->mutex);
for (i = 0; i < MLX4_MAX_MAC_NUM - 1; i++) {
@@ -106,7 +173,6 @@ int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *index)
if (mac == (MLX4_MAC_MASK & be64_to_cpu(table->entries[i]))) {
/* MAC already registered, increase refernce count */
- *index = i;
++table->refs[i];
goto out;
}
@@ -137,7 +203,8 @@ int mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac, int *index)
goto out;
}
- *index = free;
+ if (!dev->caps.vep_uc_steering)
+ *qpn = info->base_qpn + free;
++table->total;
out:
mutex_unlock(&table->mutex);
@@ -145,20 +212,52 @@ out:
}
EXPORT_SYMBOL_GPL(mlx4_register_mac);
-void mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, int index)
+static int validate_index(struct mlx4_dev *dev,
+ struct mlx4_mac_table *table, int index)
{
- struct mlx4_mac_table *table = &mlx4_priv(dev)->port[port].mac_table;
+ int err = 0;
- mutex_lock(&table->mutex);
- if (!table->refs[index]) {
- mlx4_warn(dev, "No MAC entry for index %d\n", index);
- goto out;
+ if (index < 0 || index >= table->max || !table->entries[index]) {
+ mlx4_warn(dev, "No valid Mac entry for the given index\n");
+ err = -EINVAL;
}
- if (--table->refs[index]) {
- mlx4_warn(dev, "Have more references for index %d,"
- "no need to modify MAC table\n", index);
- goto out;
+ return err;
+}
+
+static int find_index(struct mlx4_dev *dev,
+ struct mlx4_mac_table *table, u64 mac)
+{
+ int i;
+ for (i = 0; i < MLX4_MAX_MAC_NUM; i++) {
+ if (mac == (MLX4_MAC_MASK & be64_to_cpu(table->entries[i])))
+ return i;
}
+ /* Mac not found */
+ return -EINVAL;
+}
+
+void mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, int qpn)
+{
+ struct mlx4_port_info *info = &mlx4_priv(dev)->port[port];
+ struct mlx4_mac_table *table = &info->mac_table;
+ int index = qpn - info->base_qpn;
+ struct mlx4_mac_entry *entry;
+
+ if (dev->caps.vep_uc_steering) {
+ entry = radix_tree_lookup(&info->mac_tree, qpn);
+ if (entry) {
+ mlx4_uc_steer_release(dev, port, entry->mac, qpn, 1);
+ radix_tree_delete(&info->mac_tree, qpn);
+ index = find_index(dev, table, entry->mac);
+ kfree(entry);
+ }
+ }
+
+ mutex_lock(&table->mutex);
+
+ if (validate_index(dev, table, index))
+ goto out;
+
table->entries[index] = 0;
mlx4_set_port_mac_table(dev, port, table->entries);
--table->total;
@@ -167,6 +266,44 @@ out:
}
EXPORT_SYMBOL_GPL(mlx4_unregister_mac);
+int mlx4_replace_mac(struct mlx4_dev *dev, u8 port, int qpn, u64 new_mac, u8 wrap)
+{
+ struct mlx4_port_info *info = &mlx4_priv(dev)->port[port];
+ struct mlx4_mac_table *table = &info->mac_table;
+ int index = qpn - info->base_qpn;
+ struct mlx4_mac_entry *entry;
+ int err;
+
+ if (dev->caps.vep_uc_steering) {
+ entry = radix_tree_lookup(&info->mac_tree, qpn);
+ if (!entry)
+ return -EINVAL;
+ index = find_index(dev, table, entry->mac);
+ mlx4_uc_steer_release(dev, port, entry->mac, qpn, 0);
+ entry->mac = new_mac;
+ err = mlx4_uc_steer_add(dev, port, entry->mac, &qpn, 0);
+ if (err || index < 0)
+ return err;
+ }
+
+ mutex_lock(&table->mutex);
+
+ err = validate_index(dev, table, index);
+ if (err)
+ goto out;
+
+ table->entries[index] = cpu_to_be64(new_mac | MLX4_MAC_VALID);
+
+ err = mlx4_set_port_mac_table(dev, port, table->entries);
+ if (unlikely(err)) {
+ mlx4_err(dev, "Failed adding MAC: 0x%llx\n", (unsigned long long) new_mac);
+ table->entries[index] = 0;
+ }
+out:
+ mutex_unlock(&table->mutex);
+ return err;
+}
+EXPORT_SYMBOL_GPL(mlx4_replace_mac);
static int mlx4_set_port_vlan_table(struct mlx4_dev *dev, u8 port,
__be32 *entries)
{
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 v2 12/16] mlx4: Add blue flame support for kernel consumers
From: Yevgeny Petrilin @ 2011-03-23 8:38 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp, eli
From: Eli Cohen <eli@mellanox.co.il>
Using blue flame can improve latency by allowing the HW to more efficiently
access the WQE. This patch presents two functions that are used to allocate or
release HW resources for using blue flame; the caller need to supply a struct
mlx4_bf object when allocating resources. Consumers that make use of this API
should post doorbells to the UAR object pointed by the initialized struct
mlx4_bf;
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
---
drivers/net/mlx4/main.c | 31 ++++++++++++++
drivers/net/mlx4/mlx4.h | 3 +
drivers/net/mlx4/pd.c | 94 +++++++++++++++++++++++++++++++++++++++++++
include/linux/mlx4/device.h | 13 ++++++
include/linux/mlx4/qp.h | 1 +
5 files changed, 142 insertions(+), 0 deletions(-)
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index c8e2761..05c5671 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -39,6 +39,7 @@
#include <linux/pci.h>
#include <linux/dma-mapping.h>
#include <linux/slab.h>
+#include <linux/io-mapping.h>
#include <linux/mlx4/device.h>
#include <linux/mlx4/doorbell.h>
@@ -721,8 +722,31 @@ static void mlx4_free_icms(struct mlx4_dev *dev)
mlx4_free_icm(dev, priv->fw.aux_icm, 0);
}
+static int map_bf_area(struct mlx4_dev *dev)
+{
+ struct mlx4_priv *priv = mlx4_priv(dev);
+ resource_size_t bf_start;
+ resource_size_t bf_len;
+ int err = 0;
+
+ bf_start = pci_resource_start(dev->pdev, 2) + (dev->caps.num_uars << PAGE_SHIFT);
+ bf_len = pci_resource_len(dev->pdev, 2) - (dev->caps.num_uars << PAGE_SHIFT);
+ priv->bf_mapping = io_mapping_create_wc(bf_start, bf_len);
+ if (!priv->bf_mapping)
+ err = -ENOMEM;
+
+ return err;
+}
+
+static void unmap_bf_area(struct mlx4_dev *dev)
+{
+ if (mlx4_priv(dev)->bf_mapping)
+ io_mapping_free(mlx4_priv(dev)->bf_mapping);
+}
+
static void mlx4_close_hca(struct mlx4_dev *dev)
{
+ unmap_bf_area(dev);
mlx4_CLOSE_HCA(dev, 0);
mlx4_free_icms(dev);
mlx4_UNMAP_FA(dev);
@@ -775,6 +799,9 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
goto err_stop_fw;
}
+ if (map_bf_area(dev))
+ mlx4_dbg(dev, "Failed to map blue flame area\n");
+
init_hca.log_uar_sz = ilog2(dev->caps.num_uars);
err = mlx4_init_icm(dev, &dev_cap, &init_hca, icm_size);
@@ -805,6 +832,7 @@ err_free_icm:
mlx4_free_icms(dev);
err_stop_fw:
+ unmap_bf_area(dev);
mlx4_UNMAP_FA(dev);
mlx4_free_icm(dev, priv->fw.fw_icm, 0);
@@ -1196,6 +1224,9 @@ static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
pci_read_config_byte(pdev, PCI_REVISION_ID, &dev->rev_id);
+ INIT_LIST_HEAD(&priv->bf_list);
+ mutex_init(&priv->bf_mutex);
+
/*
* Now reset the HCA before we touch the PCI capabilities or
* attempt a firmware command, since a boot ROM may have left
diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h
index d8bb441..bc9a216 100644
--- a/drivers/net/mlx4/mlx4.h
+++ b/drivers/net/mlx4/mlx4.h
@@ -353,6 +353,9 @@ struct mlx4_priv {
struct mutex port_mutex;
struct mlx4_msix_ctl msix_ctl;
struct mlx4_steer *steer;
+ struct list_head bf_list;
+ struct mutex bf_mutex;
+ struct io_mapping *bf_mapping;
};
static inline struct mlx4_priv *mlx4_priv(struct mlx4_dev *dev)
diff --git a/drivers/net/mlx4/pd.c b/drivers/net/mlx4/pd.c
index c4988d6..5210a0f 100644
--- a/drivers/net/mlx4/pd.c
+++ b/drivers/net/mlx4/pd.c
@@ -32,6 +32,7 @@
*/
#include <linux/errno.h>
+#include <linux/io-mapping.h>
#include <asm/page.h>
@@ -77,6 +78,7 @@ int mlx4_uar_alloc(struct mlx4_dev *dev, struct mlx4_uar *uar)
return -ENOMEM;
uar->pfn = (pci_resource_start(dev->pdev, 2) >> PAGE_SHIFT) + uar->index;
+ uar->map = NULL;
return 0;
}
@@ -88,6 +90,98 @@ void mlx4_uar_free(struct mlx4_dev *dev, struct mlx4_uar *uar)
}
EXPORT_SYMBOL_GPL(mlx4_uar_free);
+int mlx4_bf_alloc(struct mlx4_dev *dev, struct mlx4_bf *bf)
+{
+ struct mlx4_priv *priv = mlx4_priv(dev);
+ struct mlx4_uar *uar;
+ int err = 0;
+ int idx;
+
+ if (!priv->bf_mapping)
+ return -ENOMEM;
+
+ mutex_lock(&priv->bf_mutex);
+ if (!list_empty(&priv->bf_list))
+ uar = list_entry(priv->bf_list.next, struct mlx4_uar, bf_list);
+ else {
+ uar = kmalloc(sizeof *uar, GFP_KERNEL);
+ if (!uar) {
+ err = -ENOMEM;
+ goto out;
+ }
+ err = mlx4_uar_alloc(dev, uar);
+ if (err)
+ goto free_kmalloc;
+
+ uar->map = ioremap(uar->pfn << PAGE_SHIFT, PAGE_SIZE);
+ if (!uar->map) {
+ err = -ENOMEM;
+ goto free_uar;
+ }
+
+ uar->bf_map = io_mapping_map_wc(priv->bf_mapping, uar->index << PAGE_SHIFT);
+ if (!uar->bf_map) {
+ err = -ENOMEM;
+ goto unamp_uar;
+ }
+ uar->free_bf_bmap = 0;
+ list_add(&uar->bf_list, &priv->bf_list);
+ }
+
+ bf->uar = uar;
+ idx = ffz(uar->free_bf_bmap);
+ uar->free_bf_bmap |= 1 << idx;
+ bf->uar = uar;
+ bf->offset = 0;
+ bf->buf_size = dev->caps.bf_reg_size / 2;
+ bf->reg = uar->bf_map + idx * dev->caps.bf_reg_size;
+ if (uar->free_bf_bmap == (1 << dev->caps.bf_regs_per_page) - 1)
+ list_del_init(&uar->bf_list);
+
+ goto out;
+
+unamp_uar:
+ bf->uar = NULL;
+ iounmap(uar->map);
+
+free_uar:
+ mlx4_uar_free(dev, uar);
+
+free_kmalloc:
+ kfree(uar);
+
+out:
+ mutex_unlock(&priv->bf_mutex);
+ return err;
+}
+EXPORT_SYMBOL_GPL(mlx4_bf_alloc);
+
+void mlx4_bf_free(struct mlx4_dev *dev, struct mlx4_bf *bf)
+{
+ struct mlx4_priv *priv = mlx4_priv(dev);
+ int idx;
+
+ if (!bf->uar || !bf->uar->bf_map)
+ return;
+
+ mutex_lock(&priv->bf_mutex);
+ idx = (bf->reg - bf->uar->bf_map) / dev->caps.bf_reg_size;
+ bf->uar->free_bf_bmap &= ~(1 << idx);
+ if (!bf->uar->free_bf_bmap) {
+ if (!list_empty(&bf->uar->bf_list))
+ list_del(&bf->uar->bf_list);
+
+ io_mapping_unmap(bf->uar->bf_map);
+ iounmap(bf->uar->map);
+ mlx4_uar_free(dev, bf->uar);
+ kfree(bf->uar);
+ } else if (list_empty(&bf->uar->bf_list))
+ list_add(&bf->uar->bf_list, &priv->bf_list);
+
+ mutex_unlock(&priv->bf_mutex);
+}
+EXPORT_SYMBOL_GPL(mlx4_bf_free);
+
int mlx4_init_uar_table(struct mlx4_dev *dev)
{
if (dev->caps.num_uars <= 128) {
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 56fa5e1..8985768 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -351,6 +351,17 @@ struct mlx4_fmr {
struct mlx4_uar {
unsigned long pfn;
int index;
+ struct list_head bf_list;
+ unsigned free_bf_bmap;
+ void __iomem *map;
+ void __iomem *bf_map;
+};
+
+struct mlx4_bf {
+ unsigned long offset;
+ int buf_size;
+ struct mlx4_uar *uar;
+ void __iomem *reg;
};
struct mlx4_cq {
@@ -478,6 +489,8 @@ void mlx4_pd_free(struct mlx4_dev *dev, u32 pdn);
int mlx4_uar_alloc(struct mlx4_dev *dev, struct mlx4_uar *uar);
void mlx4_uar_free(struct mlx4_dev *dev, struct mlx4_uar *uar);
+int mlx4_bf_alloc(struct mlx4_dev *dev, struct mlx4_bf *bf);
+void mlx4_bf_free(struct mlx4_dev *dev, struct mlx4_bf *bf);
int mlx4_mtt_init(struct mlx4_dev *dev, int npages, int page_shift,
struct mlx4_mtt *mtt);
diff --git a/include/linux/mlx4/qp.h b/include/linux/mlx4/qp.h
index 0eeb2a1..9e9eb21 100644
--- a/include/linux/mlx4/qp.h
+++ b/include/linux/mlx4/qp.h
@@ -303,6 +303,7 @@ struct mlx4_wqe_data_seg {
enum {
MLX4_INLINE_ALIGN = 64,
+ MLX4_INLINE_SEG = 1 << 31,
};
struct mlx4_wqe_inline_seg {
--
1.6.0.2
^ permalink raw reply related
* [PATCH v2 13/16] mlx4_core: maintain available field in bitmap allocator
From: Yevgeny Petrilin @ 2011-03-23 8:38 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp, eli
From: Eli Cohen <eli@mellanox.co.il>
Add mlx4_bitmap_avail() to give the number of available resources. We want to
use this as a hint to whether to allocate a resources or not. This patch is
introduced to be used with allocation blue flame registers.
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
---
drivers/net/mlx4/alloc.c | 13 +++++++++++++
drivers/net/mlx4/mlx4.h | 2 ++
2 files changed, 15 insertions(+), 0 deletions(-)
diff --git a/drivers/net/mlx4/alloc.c b/drivers/net/mlx4/alloc.c
index 3a4277f..116cae3 100644
--- a/drivers/net/mlx4/alloc.c
+++ b/drivers/net/mlx4/alloc.c
@@ -62,6 +62,9 @@ u32 mlx4_bitmap_alloc(struct mlx4_bitmap *bitmap)
} else
obj = -1;
+ if (obj != -1)
+ --bitmap->avail;
+
spin_unlock(&bitmap->lock);
return obj;
@@ -101,11 +104,19 @@ u32 mlx4_bitmap_alloc_range(struct mlx4_bitmap *bitmap, int cnt, int align)
} else
obj = -1;
+ if (obj != -1)
+ bitmap->avail -= cnt;
+
spin_unlock(&bitmap->lock);
return obj;
}
+u32 mlx4_bitmap_avail(struct mlx4_bitmap *bitmap)
+{
+ return bitmap->avail;
+}
+
void mlx4_bitmap_free_range(struct mlx4_bitmap *bitmap, u32 obj, int cnt)
{
obj &= bitmap->max + bitmap->reserved_top - 1;
@@ -115,6 +126,7 @@ void mlx4_bitmap_free_range(struct mlx4_bitmap *bitmap, u32 obj, int cnt)
bitmap->last = min(bitmap->last, obj);
bitmap->top = (bitmap->top + bitmap->max + bitmap->reserved_top)
& bitmap->mask;
+ bitmap->avail += cnt;
spin_unlock(&bitmap->lock);
}
@@ -130,6 +142,7 @@ int mlx4_bitmap_init(struct mlx4_bitmap *bitmap, u32 num, u32 mask,
bitmap->max = num - reserved_top;
bitmap->mask = mask;
bitmap->reserved_top = reserved_top;
+ bitmap->avail = num - reserved_top - reserved_bot;
spin_lock_init(&bitmap->lock);
bitmap->table = kzalloc(BITS_TO_LONGS(bitmap->max) *
sizeof (long), GFP_KERNEL);
diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h
index bc9a216..c1e0e5f 100644
--- a/drivers/net/mlx4/mlx4.h
+++ b/drivers/net/mlx4/mlx4.h
@@ -105,6 +105,7 @@ struct mlx4_bitmap {
u32 max;
u32 reserved_top;
u32 mask;
+ u32 avail;
spinlock_t lock;
unsigned long *table;
};
@@ -371,6 +372,7 @@ u32 mlx4_bitmap_alloc(struct mlx4_bitmap *bitmap);
void mlx4_bitmap_free(struct mlx4_bitmap *bitmap, u32 obj);
u32 mlx4_bitmap_alloc_range(struct mlx4_bitmap *bitmap, int cnt, int align);
void mlx4_bitmap_free_range(struct mlx4_bitmap *bitmap, u32 obj, int cnt);
+u32 mlx4_bitmap_avail(struct mlx4_bitmap *bitmap);
int mlx4_bitmap_init(struct mlx4_bitmap *bitmap, u32 num, u32 mask,
u32 reserved_bot, u32 resetrved_top);
void mlx4_bitmap_cleanup(struct mlx4_bitmap *bitmap);
--
1.6.0.2
^ permalink raw reply related
* [PATCH v2 14/16] mlx4_core: reserve UARs for userspace consumers
From: Yevgeny Petrilin @ 2011-03-23 8:38 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>
---
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
* [PATCH v2 15/16] mlx4_en: Using blue flame support
From: Yevgeny Petrilin @ 2011-03-23 8:38 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp
Doorbell is used according to usage of BlueFlame.
For Blue Flame to work in Ethernet mode QP number should have 0
at bits 6,7.
Allocating range of QPs accordingly.
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
drivers/net/mlx4/en_netdev.c | 10 +++++-
drivers/net/mlx4/en_tx.c | 72 ++++++++++++++++++++++++++++++-----------
drivers/net/mlx4/mlx4_en.h | 4 ++-
3 files changed, 64 insertions(+), 22 deletions(-)
diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c
index 08e6801..5762ebd 100644
--- a/drivers/net/mlx4/en_netdev.c
+++ b/drivers/net/mlx4/en_netdev.c
@@ -931,6 +931,13 @@ int mlx4_en_alloc_resources(struct mlx4_en_priv *priv)
{
struct mlx4_en_port_profile *prof = priv->prof;
int i;
+ int base_tx_qpn, err;
+
+ err = mlx4_qp_reserve_range(priv->mdev->dev, priv->tx_ring_num, 256, &base_tx_qpn);
+ if (err) {
+ en_err(priv, "failed reserving range for TX rings\n");
+ return err;
+ }
/* Create tx Rings */
for (i = 0; i < priv->tx_ring_num; i++) {
@@ -938,7 +945,7 @@ int mlx4_en_alloc_resources(struct mlx4_en_priv *priv)
prof->tx_ring_size, i, TX))
goto err;
- if (mlx4_en_create_tx_ring(priv, &priv->tx_ring[i],
+ if (mlx4_en_create_tx_ring(priv, &priv->tx_ring[i], base_tx_qpn + i,
prof->tx_ring_size, TXBB_SIZE))
goto err;
}
@@ -958,6 +965,7 @@ int mlx4_en_alloc_resources(struct mlx4_en_priv *priv)
err:
en_err(priv, "Failed to allocate NIC resources\n");
+ mlx4_qp_release_range(priv->mdev->dev, base_tx_qpn, priv->tx_ring_num);
return -ENOMEM;
}
diff --git a/drivers/net/mlx4/en_tx.c b/drivers/net/mlx4/en_tx.c
index a680cd4..01feb8f 100644
--- a/drivers/net/mlx4/en_tx.c
+++ b/drivers/net/mlx4/en_tx.c
@@ -44,6 +44,7 @@
enum {
MAX_INLINE = 104, /* 128 - 16 - 4 - 4 */
+ MAX_BF = 256,
};
static int inline_thold __read_mostly = MAX_INLINE;
@@ -52,7 +53,7 @@ module_param_named(inline_thold, inline_thold, int, 0444);
MODULE_PARM_DESC(inline_thold, "threshold for using inline data");
int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv,
- struct mlx4_en_tx_ring *ring, u32 size,
+ struct mlx4_en_tx_ring *ring, int qpn, u32 size,
u16 stride)
{
struct mlx4_en_dev *mdev = priv->mdev;
@@ -103,23 +104,25 @@ int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv,
"buf_size:%d dma:%llx\n", ring, ring->buf, ring->size,
ring->buf_size, (unsigned long long) ring->wqres.buf.direct.map);
- err = mlx4_qp_reserve_range(mdev->dev, 1, 1, &ring->qpn);
- if (err) {
- en_err(priv, "Failed reserving qp for tx ring.\n");
- goto err_map;
- }
-
+ ring->qpn = qpn;
err = mlx4_qp_alloc(mdev->dev, ring->qpn, &ring->qp);
if (err) {
en_err(priv, "Failed allocating qp %d\n", ring->qpn);
- goto err_reserve;
+ goto err_map;
}
ring->qp.event = mlx4_en_sqp_event;
+ err = mlx4_bf_alloc(mdev->dev, &ring->bf);
+ if (err) {
+ en_dbg(DRV, priv, "working without blueflame (%d)", err);
+ ring->bf.uar = &mdev->priv_uar;
+ ring->bf.uar->map = mdev->uar_map;
+ ring->bf_enabled = false;
+ } else
+ ring->bf_enabled = true;
+
return 0;
-err_reserve:
- mlx4_qp_release_range(mdev->dev, ring->qpn, 1);
err_map:
mlx4_en_unmap_buffer(&ring->wqres.buf);
err_hwq_res:
@@ -139,6 +142,8 @@ void mlx4_en_destroy_tx_ring(struct mlx4_en_priv *priv,
struct mlx4_en_dev *mdev = priv->mdev;
en_dbg(DRV, priv, "Destroying tx ring, qpn: %d\n", ring->qpn);
+ if (ring->bf_enabled)
+ mlx4_bf_free(mdev->dev, &ring->bf);
mlx4_qp_remove(mdev->dev, &ring->qp);
mlx4_qp_free(mdev->dev, &ring->qp);
mlx4_qp_release_range(mdev->dev, ring->qpn, 1);
@@ -171,6 +176,8 @@ int mlx4_en_activate_tx_ring(struct mlx4_en_priv *priv,
mlx4_en_fill_qp_context(priv, ring->size, ring->stride, 1, 0, ring->qpn,
ring->cqn, &ring->context);
+ if (ring->bf_enabled)
+ ring->context.usr_page = cpu_to_be32(ring->bf.uar->index);
err = mlx4_qp_to_ready(mdev->dev, &ring->wqres.mtt, &ring->context,
&ring->qp, &ring->qp_state);
@@ -591,6 +598,11 @@ u16 mlx4_en_select_queue(struct net_device *dev, struct sk_buff *skb)
return skb_tx_hash(dev, skb);
}
+static void mlx4_bf_copy(unsigned long *dst, unsigned long *src, unsigned bytecnt)
+{
+ __iowrite64_copy(dst, src, bytecnt / 8);
+}
+
netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
{
struct mlx4_en_priv *priv = netdev_priv(dev);
@@ -609,12 +621,13 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
int desc_size;
int real_size;
dma_addr_t dma;
- u32 index;
+ u32 index, bf_index;
__be32 op_own;
u16 vlan_tag = 0;
int i;
int lso_header_size;
void *fragptr;
+ bool bounce = false;
if (!priv->port_up)
goto tx_drop;
@@ -657,13 +670,16 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
/* Packet is good - grab an index and transmit it */
index = ring->prod & ring->size_mask;
+ bf_index = ring->prod;
/* See if we have enough space for whole descriptor TXBB for setting
* SW ownership on next descriptor; if not, use a bounce buffer. */
if (likely(index + nr_txbb <= ring->size))
tx_desc = ring->buf + index * TXBB_SIZE;
- else
+ else {
tx_desc = (struct mlx4_en_tx_desc *) ring->bounce_buf;
+ bounce = true;
+ }
/* Save skb in tx_info ring */
tx_info = &ring->tx_info[index];
@@ -768,21 +784,37 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
ring->prod += nr_txbb;
/* If we used a bounce buffer then copy descriptor back into place */
- if (tx_desc == (struct mlx4_en_tx_desc *) ring->bounce_buf)
+ if (bounce)
tx_desc = mlx4_en_bounce_to_desc(priv, ring, index, desc_size);
/* Run destructor before passing skb to HW */
if (likely(!skb_shared(skb)))
skb_orphan(skb);
- /* Ensure new descirptor hits memory
- * before setting ownership of this descriptor to HW */
- wmb();
- tx_desc->ctrl.owner_opcode = op_own;
+ if (ring->bf_enabled && desc_size <= MAX_BF && !bounce && !vlan_tag) {
+ *(u32 *) (&tx_desc->ctrl.vlan_tag) |= ring->doorbell_qpn;
+ op_own |= htonl((bf_index & 0xffff) << 8);
+ /* Ensure new descirptor hits memory
+ * before setting ownership of this descriptor to HW */
+ wmb();
+ tx_desc->ctrl.owner_opcode = op_own;
- /* Ring doorbell! */
- wmb();
- writel(ring->doorbell_qpn, mdev->uar_map + MLX4_SEND_DOORBELL);
+ wmb();
+
+ mlx4_bf_copy(ring->bf.reg + ring->bf.offset, (unsigned long *) &tx_desc->ctrl,
+ desc_size);
+
+ wmb();
+
+ ring->bf.offset ^= ring->bf.buf_size;
+ } else {
+ /* Ensure new descirptor hits memory
+ * before setting ownership of this descriptor to HW */
+ wmb();
+ tx_desc->ctrl.owner_opcode = op_own;
+ wmb();
+ writel(ring->doorbell_qpn, ring->bf.uar->map + MLX4_SEND_DOORBELL);
+ }
/* Poll CQ here */
mlx4_en_xmit_poll(priv, tx_ind);
diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h
index 5a2c560..edcb535 100644
--- a/drivers/net/mlx4/mlx4_en.h
+++ b/drivers/net/mlx4/mlx4_en.h
@@ -248,6 +248,8 @@ struct mlx4_en_tx_ring {
unsigned long bytes;
unsigned long packets;
spinlock_t comp_lock;
+ struct mlx4_bf bf;
+ bool bf_enabled;
};
struct mlx4_en_rx_desc {
@@ -518,7 +520,7 @@ u16 mlx4_en_select_queue(struct net_device *dev, struct sk_buff *skb);
netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev);
int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv, struct mlx4_en_tx_ring *ring,
- u32 size, u16 stride);
+ int qpn, u32 size, u16 stride);
void mlx4_en_destroy_tx_ring(struct mlx4_en_priv *priv, struct mlx4_en_tx_ring *ring);
int mlx4_en_activate_tx_ring(struct mlx4_en_priv *priv,
struct mlx4_en_tx_ring *ring,
--
1.6.0.2
^ permalink raw reply related
* [PATCH v2 16/16] mlx4_en: updated driver version to 1.5.4.1
From: Yevgeny Petrilin @ 2011-03-23 8:39 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
* Re: can: c_can: TX echo
From: Kurt Van Dijck @ 2011-03-23 8:51 UTC (permalink / raw)
To: Jan Altenberg; +Cc: bhupesh.sharma, wg, b.spranger, netdev
In-Reply-To: <16a340801622a96218c76dbbabc7a23f.squirrel@www.linutronix.de>
I'm not too familiar with the c_can chip.
So I have a few questions inside my reply.
On Tue, Mar 22, 2011 at 04:59:23PM +0100, Jan Altenberg wrote:
> Hi all,
>
> I did some more testing on the SocketCAN driver for the Bosch c_can
> controller and I observed some strange behaviour for the TX handling.
> First of all the TX bytes are not accounted correctly. The reason for that
> seems to be quite obvious if we look into c_can_do_tx():
>
> [...]
>
> c_can_inval_msg_object(dev, 0, msg_obj_no);
> val = c_can_read_reg32(priv, &priv->regs->txrqst1);
> if (!(val & (1 << msg_obj_no))) {
> can_get_echo_skb(dev,
> msg_obj_no - C_CAN_MSG_OBJ_TX_FIRST);
> stats->tx_bytes += priv->read_reg(priv,
> &priv->regs->ifregs[0].msg_cntrl)
> & IF_MCONT_DLC_MASK;
>
> So, we first invalidate the message object and afterwards we read the DLC
> value from the msg_cntrl (which is 0 after invalidating the
> message object) to account the TX bytes. So tx_bytes will always be 0. The
> fix should be easy, I think, we can just move
> c_can_inval_msg_object to the end of that loop.
IMO, it looks necessary to call c_can_inval_msg_object inside the if (),
after it has been transmitted. Otherwise, if for some other (TX) reason you
get in this loop, you may clear a pending transmission?
Again, I haven't read this one's datasheet. I was familiar with its predecessor.
>
> Cheers,
> Jan
Regards,
Kurt
^ permalink raw reply
* Re: can: c_can: TX delivery
From: Kurt Van Dijck @ 2011-03-23 8:53 UTC (permalink / raw)
To: Jan Altenberg; +Cc: bhupesh.sharma, wg, b.spranger, netdev
In-Reply-To: <16a340801622a96218c76dbbabc7a23f.squirrel@www.linutronix.de>
Jan,
I split your 2 questions in 2 replies.
On Tue, Mar 22, 2011 at 04:59:23PM +0100, Jan Altenberg wrote:
> Hi all,
>
> The second problem is related to tx_next, which should hold the number of
> the oldest CAN frame, which was not on the line:
>
> for (/* nix */; (priv->tx_next - priv->tx_echo) > 0; priv->tx_echo++) {
> msg_obj_no = get_tx_echo_msg_obj(priv);
> c_can_inval_msg_object(dev, 0, msg_obj_no);
> val = c_can_read_reg32(priv, &priv->regs->txrqst1);
> if (!(val & (1 << msg_obj_no))) {
> can_get_echo_skb(dev,
> msg_obj_no - C_CAN_MSG_OBJ_TX_FIRST);
> stats->tx_bytes += priv->read_reg(priv,
> &priv->regs->ifregs[0].msg_cntrl)
> & IF_MCONT_DLC_MASK;
> stats->tx_packets++;
> }
> }
>
> But tx_echo is incremented unconditionally and we don't actually track the
> number of the oldest unsent frame.
> Let's assume the following scenario: We bring up can0 and send 3
> frames: TX object: 0, 1, 2; 1 and 2 make it on the line, but 0 is
> still pending. If we go through the above loop in that situation, we will
> skip message object 0, because the txrqst bit is still set. We will
> account message object 1 and 2. That's correct, but afterwards tx_echo is
> set to 2, BUT the oldest message which is pending is 0. Am I right or did
> I get something wrong?
> The operation of c_can_do_tx() is described as follows: "We iterate from
> priv->tx_echo to priv->tx_next and check if the packet has been
> transmitted, echo it back to the CAN framework. If we discover a not yet
> transmitted package, stop looking for more." The actual
> implementation doesn't seem to stop if we discover a not yet
> transmitted package. But I'm not sure if just stopping might be a
> good idea, because in that case, the echo skb for already transmitted
> messages might be delayed by not yet transmitted messages.
It is better to deliver the echo's in the order they were delivered
on CAN.
For that, we can replace the tx_next/tx_echo pair with a single tx_bitmask.
In the tx_bitmask, we could set a bit when the TX is queued, and test for
those bits in the tx interrupt.
When tx is done, clear the bit in tx_bitmask & stuff.
16 tx objects fit inside a single int.
c_can_start_xmit would look a bit different. I think of something like this:
/* find free object */
for (j = 0; j < TX_MAX; ++j) {
if (!(priv->tx_bitmask & (1 << j)))
break;
}
if (j >= TX_MAX)
/* no free objects */
return NETDEV_TX_BUSY;
c_can_write_...(j, ...)
can_put_echo_skb(, j, ...)
priv->tx_bitmask |= (1 << j);
return NETDEV_TX_OK;
not sure if I made my point.
Note that this will eliminate the need for explicit wrap-around. It's
done implicitely.
>
> Cheers,
> Jan
>
Regards,
Kurt
^ permalink raw reply
* Re: [BUG] VPN broken in net-next
From: Julian Anastasov @ 2011-03-23 9:05 UTC (permalink / raw)
To: David Miller; +Cc: shemminger, netdev
In-Reply-To: <20110322.215655.245381151.davem@davemloft.net>
Hello,
On Tue, 22 Mar 2011, David Miller wrote:
> I just got back to this, sorry for taking so long :-)
>
> Here is the patch I've come up with and will commit to
> net-2.6, thanks!
>
> --------------------
> ipv4: Fallback to FIB local table in __ip_dev_find().
>
> In commit 9435eb1cf0b76b323019cebf8d16762a50a12a19
> ("ipv4: Implement __ip_dev_find using new interface address hash.")
> we reimplemented __ip_dev_find() so that it doesn't have to
> do a full FIB table lookup.
>
> Instead, it consults a hash table of addresses configured to
> interfaces.
>
> This works identically to the old code in all except one case,
> and that is for loopback subnets.
>
> The old code would match the loopback device for any IP address
> that falls within a subnet configured to the loopback device.
>
> Handle this corner case by doing the FIB lookup.
>
> We could implement this via inet_addr_onlink() but:
>
> 1) Someone could configure many addresses to loopback and
> inet_addr_onlink() is a simple list traversal.
>
> 2) We know the old code works.
>
> Reported-by: Julian Anastasov <ja@ssi.bg>
> Signed-off-by: David S. Miller <davem@davemloft.net>
> ---
> net/ipv4/devinet.c | 16 ++++++++++++++++
> 1 files changed, 16 insertions(+), 0 deletions(-)
>
> diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
> index d5a4553..5345b0b 100644
> --- a/net/ipv4/devinet.c
> +++ b/net/ipv4/devinet.c
> @@ -64,6 +64,8 @@
> #include <net/rtnetlink.h>
> #include <net/net_namespace.h>
>
> +#include "fib_lookup.h"
> +
> static struct ipv4_devconf ipv4_devconf = {
> .data = {
> [IPV4_DEVCONF_ACCEPT_REDIRECTS - 1] = 1,
> @@ -151,6 +153,20 @@ struct net_device *__ip_dev_find(struct net *net, __be32 addr, bool devref)
> break;
> }
> }
> + if (!result) {
> + struct flowi4 fl4 = { .daddr = addr };
> + struct fib_result res = { 0 };
> + struct fib_table *local;
> +
> + /* Fallback to FIB local table so that communication
> + * over loopback subnets work.
> + */
> + local = fib_get_table(net, RT_TABLE_LOCAL);
> + if (local &&
> + !fib_table_lookup(local, &fl4, &res, FIB_LOOKUP_NOREF) &&
> + res.type == RTN_LOCAL)
> + result = FIB_RES_DEV(res);
> + }
> if (result && devref)
> dev_hold(result);
> rcu_read_unlock();
> --
> 1.7.4.1
Not tested but looks ok for me.
Regards
--
Julian Anastasov <ja@ssi.bg>
^ permalink raw reply
* Re: 2.6.39-rc: WARNING: at __ip_select_ident+0xd3/0xf0()
From: Eric Dumazet @ 2011-03-23 9:49 UTC (permalink / raw)
To: Alexander Beregalov; +Cc: Linux Kernel Mailing List, netdev, David Miller
In-Reply-To: <AANLkTinacRCOx+z02n1OwBjGfVL5itzZs7Jqv_4PRXZK@mail.gmail.com>
Le mercredi 23 mars 2011 à 10:14 +0300, Alexander Beregalov a écrit :
> Hi
>
> I do not know how to reproduce it.
>
> WARNING: at include/net/inetpeer.h:102 __ip_select_ident+0xd3/0xf0()
> Hardware name:
> Modules linked in: hwmon_vid sata_sil i2c_nforce2
> Pid: 0, comm: swapper Not tainted 2.6.38-07035-g6447f55d #1
> Call Trace:
> [<c102ad5d>] warn_slowpath_common+0x6d/0xa0
> [<c13070c3>] ? __ip_select_ident+0xd3/0xf0
> [<c13070c3>] ? __ip_select_ident+0xd3/0xf0
> [<c102adad>] warn_slowpath_null+0x1d/0x20
> [<c13070c3>] __ip_select_ident+0xd3/0xf0
> [<c130df4c>] __ip_make_skb+0x25c/0x360
> [<c130d807>] ? ip_append_data+0x67/0xa0
> [<c132f980>] ? icmp_glue_bits+0x0/0x60
> [<c130e0b4>] ip_push_pending_frames+0x14/0x30
> [<c132f95c>] icmp_push_reply+0xdc/0x100
> [<c13306a8>] icmp_send+0x408/0x4a0
> [<c132ca61>] ? __udp4_lib_lookup.clone.49+0x171/0x2d0
> [<c132d877>] __udp4_lib_rcv+0x487/0x6a0
> [<c13058c0>] ? ip_route_input_common+0x0/0xa00
> [<c132daa2>] udp_rcv+0x12/0x20
> [<c13084bc>] ip_local_deliver+0x9c/0x140
> [<c130846d>] ? ip_local_deliver+0x4d/0x140
> [<c13087c6>] ip_rcv+0x266/0x560
> [<c12e92b3>] __netif_receive_skb+0x393/0x400
> [<c12e8fe4>] ? __netif_receive_skb+0xc4/0x400
> [<c12e9472>] netif_receive_skb+0x42/0x50
> [<c12e94b7>] napi_skb_finish+0x37/0x50
> [<c12eadf3>] napi_gro_receive+0xb3/0xc0
> [<c128d76f>] e1000_receive_skb.clone.34+0x3f/0x60
> [<c128dfcf>] e1000_clean_rx_irq+0x27f/0x4b0
> [<c1055be1>] ? __lock_acquire+0x441/0x19a0
> [<c128d249>] e1000_clean+0x1e9/0x570
> [<c13697fc>] ? _raw_spin_lock+0x5c/0x70
> [<c12e9778>] net_rx_action+0xe8/0x170
> [<c102ffe7>] __do_softirq+0x77/0x110
> [<c102ff70>] ? __do_softirq+0x0/0x110
> <IRQ> [<c1030246>] ? irq_exit+0x76/0x90
> [<c1003823>] ? do_IRQ+0x43/0xb0
> [<c136f5ee>] ? common_interrupt+0x2e/0x34
> [<c100755f>] ? default_idle+0x2f/0x60
> [<c10015d6>] ? cpu_idle+0x36/0x60
> [<c135f2e5>] ? rest_init+0xa5/0xb0
> [<c135f240>] ? rest_init+0x0/0xb0
> [<c163c663>] ? start_kernel+0x277/0x27e
> [<c163c174>] ? unknown_bootoption+0x0/0x19f
> [<c163c071>] ? i386_start_kernel+0x71/0x77
> --
Thanks Alexander for this bug report.
Here is a patch to fix this issue (and other crash you mentioned in next
mail)
[PATCH] ipv4: fix ip_rt_update_pmtu()
commit 2c8cec5c10bc (Cache learned PMTU information in inetpeer) added
an extra inet_putpeer() call in ip_rt_update_pmtu().
This results in various problems, since we can free one inetpeer, while
it is still in use.
Ref: http://www.spinics.net/lists/netdev/msg159121.html
Reported-by: Alexander Beregalov <a.beregalov@gmail.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
---
net/ipv4/route.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 870b518..34921b0 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1593,8 +1593,6 @@ static void ip_rt_update_pmtu(struct dst_entry *dst, u32 mtu)
rt->rt_peer_genid = rt_peer_genid();
}
check_peer_pmtu(dst, peer);
-
- inet_putpeer(peer);
}
}
^ permalink raw reply related
* [PATCH] net: fix ethtool->set_flags not intended -EINVAL return value
From: Stanislaw Gruszka @ 2011-03-23 9:54 UTC (permalink / raw)
To: netdev
Cc: Ben Hutchings, Eric Dumazet, Jesse Gross, Amit Kumar Salecha,
Jon Mason, Shreyas Bhatewara, Jesper Dangaard Brouer
After commit d5dbda23804156ae6f35025ade5307a49d1db6d7 "ethtool: Add
support for vlan accleration.", drivers that have NETIF_F_HW_VLAN_TX,
and/or NETIF_F_HW_VLAN_RX feature, but do not allow enable/disable vlan
acceleration via ethtool set_flags, always return -EINVAL from that
function. Fix by returning -EINVAL only if requested features do not
match current settings and can not be changed by driver.
Change any driver that define ethtool->set_flags to use
ethtool_invalid_flags() to avoid similar problems in the future
(also on drivers that do not have the problem).
Tested with modified (to reproduce this bug) myri10ge driver.
Cc: stable@kernel.org # 2.6.37+
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
---
drivers/net/netxen/netxen_nic_ethtool.c | 2 +-
drivers/net/qlcnic/qlcnic_ethtool.c | 2 +-
drivers/net/s2io.c | 2 +-
drivers/net/vmxnet3/vmxnet3_ethtool.c | 4 ++--
drivers/net/vxge/vxge-ethtool.c | 4 ++--
include/linux/ethtool.h | 1 +
net/core/ethtool.c | 17 ++++++++++++++++-
7 files changed, 24 insertions(+), 8 deletions(-)
diff --git a/drivers/net/netxen/netxen_nic_ethtool.c b/drivers/net/netxen/netxen_nic_ethtool.c
index 653d308..3bdcc80 100644
--- a/drivers/net/netxen/netxen_nic_ethtool.c
+++ b/drivers/net/netxen/netxen_nic_ethtool.c
@@ -871,7 +871,7 @@ static int netxen_nic_set_flags(struct net_device *netdev, u32 data)
struct netxen_adapter *adapter = netdev_priv(netdev);
int hw_lro;
- if (data & ~ETH_FLAG_LRO)
+ if (ethtool_invalid_flags(netdev, data, ETH_FLAG_LRO))
return -EINVAL;
if (!(adapter->capabilities & NX_FW_CAPABILITY_HW_LRO))
diff --git a/drivers/net/qlcnic/qlcnic_ethtool.c b/drivers/net/qlcnic/qlcnic_ethtool.c
index 4c14510..45b2755 100644
--- a/drivers/net/qlcnic/qlcnic_ethtool.c
+++ b/drivers/net/qlcnic/qlcnic_ethtool.c
@@ -1003,7 +1003,7 @@ static int qlcnic_set_flags(struct net_device *netdev, u32 data)
struct qlcnic_adapter *adapter = netdev_priv(netdev);
int hw_lro;
- if (data & ~ETH_FLAG_LRO)
+ if (ethtool_invalid_flags(netdev, data, ETH_FLAG_LRO))
return -EINVAL;
if (!(adapter->capabilities & QLCNIC_FW_CAPABILITY_HW_LRO))
diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c
index 2ad6364..356e74d 100644
--- a/drivers/net/s2io.c
+++ b/drivers/net/s2io.c
@@ -6726,7 +6726,7 @@ static int s2io_ethtool_set_flags(struct net_device *dev, u32 data)
int rc = 0;
int changed = 0;
- if (data & ~ETH_FLAG_LRO)
+ if (ethtool_invalid_flags(dev, data, ETH_FLAG_LRO))
return -EINVAL;
if (data & ETH_FLAG_LRO) {
diff --git a/drivers/net/vmxnet3/vmxnet3_ethtool.c b/drivers/net/vmxnet3/vmxnet3_ethtool.c
index 81254be..51f2ef1 100644
--- a/drivers/net/vmxnet3/vmxnet3_ethtool.c
+++ b/drivers/net/vmxnet3/vmxnet3_ethtool.c
@@ -304,8 +304,8 @@ vmxnet3_set_flags(struct net_device *netdev, u32 data)
u8 lro_present = (netdev->features & NETIF_F_LRO) == 0 ? 0 : 1;
unsigned long flags;
- if (data & ~ETH_FLAG_LRO)
- return -EOPNOTSUPP;
+ if (ethtool_invalid_flags(netdev, data, ETH_FLAG_LRO))
+ return -EINVAL;
if (lro_requested ^ lro_present) {
/* toggle the LRO feature*/
diff --git a/drivers/net/vxge/vxge-ethtool.c b/drivers/net/vxge/vxge-ethtool.c
index 1dd3a21..c5eb034 100644
--- a/drivers/net/vxge/vxge-ethtool.c
+++ b/drivers/net/vxge/vxge-ethtool.c
@@ -1117,8 +1117,8 @@ static int vxge_set_flags(struct net_device *dev, u32 data)
struct vxgedev *vdev = netdev_priv(dev);
enum vxge_hw_status status;
- if (data & ~ETH_FLAG_RXHASH)
- return -EOPNOTSUPP;
+ if (ethtool_invalid_flags(dev, data, ETH_FLAG_RXHASH))
+ return -EINVAL;
if (!!(data & ETH_FLAG_RXHASH) == vdev->devh->config.rth_en)
return 0;
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index aac3e2e..e22b046 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -643,6 +643,7 @@ int ethtool_op_set_ufo(struct net_device *dev, u32 data);
u32 ethtool_op_get_flags(struct net_device *dev);
int ethtool_op_set_flags(struct net_device *dev, u32 data, u32 supported);
void ethtool_ntuple_flush(struct net_device *dev);
+bool ethtool_invalid_flags(struct net_device *dev, u32 data, u32 supported);
/**
* ðtool_ops - Alter and report network device settings
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index c1a71bb..25d1ea2 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -141,9 +141,24 @@ u32 ethtool_op_get_flags(struct net_device *dev)
}
EXPORT_SYMBOL(ethtool_op_get_flags);
+/* Check if device can enable (or disable) particular feature coded in "data"
+ * argument. Flags "supported" describe features that can be toggled by device.
+ * If feature can not be toggled, it state (enabled or disabled) must match
+ * hardcoded device features state, otherwise flags are marked as invalid.
+ */
+bool ethtool_invalid_flags(struct net_device *dev, u32 data, u32 supported)
+{
+ u32 features = dev->features & flags_dup_features;
+ /* "data" can contain only flags_dup_features bits,
+ * see __ethtool_set_flags */
+
+ return (features & ~supported) != (data & ~supported);
+}
+EXPORT_SYMBOL(ethtool_invalid_flags);
+
int ethtool_op_set_flags(struct net_device *dev, u32 data, u32 supported)
{
- if (data & ~supported)
+ if (ethtool_invalid_flags(dev, data, supported))
return -EINVAL;
dev->features = ((dev->features & ~flags_dup_features) |
--
1.7.1
^ permalink raw reply related
* Re: [PATCH] via-velocity: fix the WOL bug on 1000M full duplex forced mode.
From: David Lv @ 2011-03-23 10:07 UTC (permalink / raw)
To: David Miller, netdev, DavidLv
In-Reply-To: <20110120.165634.257520752.davem@davemloft.net>
Hi, David. I haven't found out this patch code in the mainline of 2.6.38.
Could you help me check it?
Thanks!
Best regards
2011/1/21 David Miller <davem@davemloft.net>:
> From: Francois Romieu <romieu@fr.zoreil.com>
> Date: Thu, 20 Jan 2011 15:59:33 +0100
>
>> The VIA velocity card can't be waken up by WOL tool on 1000M full
>> duplex forced mode. This patch fixes the bug.
>>
>> Signed-off-by: David Lv <DavidLv@viatech.com.cn>
>> Acked-by: Francois Romieu <romieu@fr.zoreil.com>
>
> Applied.
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
^ permalink raw reply
* Re: bridge not routing packets via source bridgeport
From: Sebastian J. Bronner @ 2011-03-23 10:12 UTC (permalink / raw)
To: netdev; +Cc: Daniel Kraft
In-Reply-To: <4D22D990.7030409@d9t.de>
I have submitted this bug on the kernel bugtracker. Further discussion
should probably take place there:
https://bugzilla.kernel.org/show_bug.cgi?id=31722
--
*Sebastian J. Bronner*
Administrator
D9T GmbH - Magirusstr. 39/1 - D-89077 Ulm
Tel: +49 731 1411 696-0 - Fax: +49 731 3799-220
Geschäftsführer: Daniel Kraft
Sitz und Register: Ulm, HRB 722416
Ust.IdNr: DE 260484638
http://d9t.de - D9T High Performance Hosting
info@d9t.de
^ permalink raw reply
* [PATCH] ip_fragment:kernel may panic when replay big packet with RST flag
From: Feng Gao @ 2011-03-23 10:49 UTC (permalink / raw)
To: netdev
Hello everyone:
PC(A)-linux(B)-PC(C)
computer(linux B) with two net interface,eth0 and eth1.
PC(A) send syn to PC(C) though linux B.
then PC(C) replay a big packet with RST flag(use tcpsic or other tools).
This RST packet(1480) come in from eth0(mtu 1500) and go out from
eth1(mtu 700), so this RST packet should fragment.
BUT in tcp_packet func: if the connection has no reply packet,and
receive the RST packet.ip_conntrack should destroy.
if (!test_bit(IPS_SEEN_REPLY_BIT, &ct->status)) {
/* If only reply is a RST, we can consider ourselves not to
have an established connection: this is a fairly common
problem case, so we can delete the conntrack
immediately. --RR */
if (th->rst) {
nf_ct_kill_acct(ct, ctinfo, skb);
return NF_ACCEPT;
}
}
BUT the skb->nfct is not set NULL in func nf_ct_kill_acct.
so when this RST packet goto ip_fragment,ip_fragment call nf_copy, in
__nf_copy func
the fragment skb->nfct point to the destory mem.
dst->nfct = src->nfct;
nf_conntrack_get(src->nfct);
SO finally.kfree_skb call destroy_conntrack again. this may result in
LINUX B kernel panic.
here is the patch,sorry ,i dont know how to use diff to generate patch.
:-D
/* This func for ip_fragment deal with RST packet */
static inline void nf_copy_rst(struct sk_buff *dst, const struct sk_buff *src)
{
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
nf_conntrack_put(dst->nfct);
nf_conntrack_put_reasm(dst->nfct_reasm);
#endif
#ifdef CONFIG_BRIDGE_NETFILTER
nf_bridge_put(dst->nf_bridge);
dst->nf_bridge = src->nf_bridge;
nf_bridge_get(src->nf_bridge);
#endif
}
static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from)
{
to->pkt_type = from->pkt_type;
to->priority = from->priority;
to->protocol = from->protocol;
dst_release(to->dst);
to->dst = dst_clone(from->dst);
to->dev = from->dev;
to->mark = from->mark;
/* Copy the flags to each fragment. */
IPCB(to)->flags = IPCB(from)->flags;
#ifdef CONFIG_NET_SCHED
to->tc_index = from->tc_index;
#endif
struct tcphdr *th = (struct tcphdr *)((char *)iph + (iph->ihl << 2));
if(iph->protocol == IPPROTO_TCP && th->rst){
/*RST packet*/
nf_copy_rst(to, from);
}else{
/*Other packet*/
nf_copy(to, from);
}
#if defined(CONFIG_NETFILTER_XT_
TARGET_TRACE) || \
defined(CONFIG_NETFILTER_XT_TARGET_TRACE_MODULE)
to->nf_trace = from->nf_trace;
#endif
#if defined(CONFIG_IP_VS) || defined(CONFIG_IP_VS_MODULE)
to->ipvs_property = from->ipvs_property;
#endif
skb_copy_secmark(to, from);
}
^ permalink raw reply
* [PATCH] sctp: Make some local functions static
From: Holger Hans Peter Freyther @ 2011-03-23 10:48 UTC (permalink / raw)
To: netdev; +Cc: linux-sctp
sparse proposed to mark these functions as static as they are
only accessed locally. This patch is doing that.
Signed-off-by: Holger Hans Peter Freyther <zecke@selfish.org>
Acked-by: Wei Yongjun <yjwei@cn.fujitsu.com>
---
net/sctp/probe.c | 10 +++++-----
net/sctp/sm_make_chunk.c | 4 ++--
2 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/net/sctp/probe.c b/net/sctp/probe.c
index bc6cd75..4616548 100644
--- a/net/sctp/probe.c
+++ b/net/sctp/probe.c
@@ -122,11 +122,11 @@ static const struct file_operations sctpprobe_fops = {
.llseek = noop_llseek,
};
-sctp_disposition_t jsctp_sf_eat_sack(const struct sctp_endpoint *ep,
- const struct sctp_association *asoc,
- const sctp_subtype_t type,
- void *arg,
- sctp_cmd_seq_t *commands)
+static sctp_disposition_t jsctp_sf_eat_sack(const struct sctp_endpoint *ep,
+ const struct sctp_association *asoc,
+ const sctp_subtype_t type,
+ void *arg,
+ sctp_cmd_seq_t *commands)
{
struct sctp_transport *sp;
static __u32 lcwnd = 0;
diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index de98665..a8c0798 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -132,8 +132,8 @@ void sctp_init_cause(struct sctp_chunk *chunk, __be16
cause_code,
* abort chunk. Differs from sctp_init_cause in that it won't oops
* if there isn't enough space in the op error chunk
*/
-int sctp_init_cause_fixed(struct sctp_chunk *chunk, __be16 cause_code,
- size_t paylen)
+static int sctp_init_cause_fixed(struct sctp_chunk *chunk, __be16 cause_code,
+ size_t paylen)
{
sctp_errhdr_t err;
__u16 len;
--
1.7.4
^ permalink raw reply related
* Re: [PATCH] ip_fragment:kernel may panic when replay big packet with RST flag
From: Eric Dumazet @ 2011-03-23 11:01 UTC (permalink / raw)
To: Feng Gao; +Cc: netdev
In-Reply-To: <AANLkTikoSxVd3kTggsCGpjXz-5CFuVNxjnpuZNgWd3eV@mail.gmail.com>
Le mercredi 23 mars 2011 à 18:49 +0800, Feng Gao a écrit :
> Hello everyone:
>
> PC(A)-linux(B)-PC(C)
> computer(linux B) with two net interface,eth0 and eth1.
> PC(A) send syn to PC(C) though linux B.
> then PC(C) replay a big packet with RST flag(use tcpsic or other tools).
>
> This RST packet(1480) come in from eth0(mtu 1500) and go out from
> eth1(mtu 700), so this RST packet should fragment.
>
> BUT in tcp_packet func: if the connection has no reply packet,and
> receive the RST packet.ip_conntrack should destroy.
> if (!test_bit(IPS_SEEN_REPLY_BIT, &ct->status)) {
> /* If only reply is a RST, we can consider ourselves not to
> have an established connection: this is a fairly common
> problem case, so we can delete the conntrack
> immediately. --RR */
> if (th->rst) {
> nf_ct_kill_acct(ct, ctinfo, skb);
> return NF_ACCEPT;
> }
> }
>
> BUT the skb->nfct is not set NULL in func nf_ct_kill_acct.
> so when this RST packet goto ip_fragment,ip_fragment call nf_copy, in
> __nf_copy func
> the fragment skb->nfct point to the destory mem.
> dst->nfct = src->nfct;
> nf_conntrack_get(src->nfct);
>
> SO finally.kfree_skb call destroy_conntrack again. this may result in
> LINUX B kernel panic.
>
>
> here is the patch,sorry ,i dont know how to use diff to generate patch.
> :-D
>
Hi Feng
Oh well, please learn how to do this. If you are able to make such
analysis, you sure can learn how to use diff !!!
What is the version of linux you use ?
^ permalink raw reply
* [PATCH] sctp: Remove store to ret as it is not used
From: Holger Freyther @ 2011-03-23 10:56 UTC (permalink / raw)
To: netdev; +Cc: linux-sctp, Holger Hans Peter Freyther
From: Holger Hans Peter Freyther <zecke@selfish.org>
The store to ret is not needed. In case it is 0 the method
will be exited directly, if not ret will be assigned again
and then returned to the caller.
Signed-off-by: Holger Hans Peter Freyther <zecke@selfish.org>
Acked-by: Wei Yongjun <yjwei@cn.fujitsu.com>
---
net/sctp/socket.c | 3 +--
1 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index b2bed35..5f72996 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -393,9 +393,8 @@ SCTP_STATIC int sctp_do_bind(struct sock *sk, union sctp_addr *addr, int len)
* detection.
*/
addr->v4.sin_port = htons(snum);
- if ((ret = sctp_get_port_local(sk, addr))) {
+ if (sctp_get_port_local(sk, addr))
return -EADDRINUSE;
- }
/* Refresh ephemeral port. */
if (!bp->port)
--
1.7.4
^ permalink raw reply related
* Re: 2.6.39-rc: WARNING: at __ip_select_ident+0xd3/0xf0()
From: Alexander Beregalov @ 2011-03-23 11:22 UTC (permalink / raw)
To: Eric Dumazet; +Cc: Linux Kernel Mailing List, netdev, David Miller
In-Reply-To: <1300873779.3063.39.camel@edumazet-laptop>
> Thanks Alexander for this bug report.
>
> Here is a patch to fix this issue (and other crash you mentioned in next
> mail)
>
Hi Eric,
Thank you for the fix, have not it checked however.
I did not know that the second problem
(https://lkml.org/lkml/2011/3/23/38) had the same reason.
I did not notice inetpeer_free_rcu() in its calltrace.
Thanks for that.
>
> [PATCH] ipv4: fix ip_rt_update_pmtu()
>
> commit 2c8cec5c10bc (Cache learned PMTU information in inetpeer) added
> an extra inet_putpeer() call in ip_rt_update_pmtu().
>
> This results in various problems, since we can free one inetpeer, while
> it is still in use.
>
> Ref: http://www.spinics.net/lists/netdev/msg159121.html
>
> Reported-by: Alexander Beregalov <a.beregalov@gmail.com>
> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
> ---
> net/ipv4/route.c | 2 --
> 1 file changed, 2 deletions(-)
>
> diff --git a/net/ipv4/route.c b/net/ipv4/route.c
> index 870b518..34921b0 100644
> --- a/net/ipv4/route.c
> +++ b/net/ipv4/route.c
> @@ -1593,8 +1593,6 @@ static void ip_rt_update_pmtu(struct dst_entry *dst, u32 mtu)
> rt->rt_peer_genid = rt_peer_genid();
> }
> check_peer_pmtu(dst, peer);
> -
> - inet_putpeer(peer);
> }
> }
>
>
>
>
^ permalink raw reply
* Re: [PATCH 02/36] scsi,rcu: convert call_rcu(fc_rport_free_rcu) to kfree_rcu()
From: Matthew Wilcox @ 2011-03-23 11:22 UTC (permalink / raw)
To: Paul E. McKenney
Cc: Robert Love, Lai Jiangshan, Ingo Molnar, Jens Axboe,
James E.J. Bottomley, Neil Horman, David S. Miller,
Alexey Kuznetsov, Pekka Savola (ipv6), James Morris,
Hideaki YOSHIFUJI, Patrick McHardy, Eric Dumazet,
Stephen Hemminger, Tejun Heo, Jarek Poplawski,
linux-kernel@vger.kernel.org, devel@open-fcoe.org,
linux-scsi@vger.kernel.org, netdev@vger.kernel.org
In-Reply-To: <20110323065014.GU2322@linux.vnet.ibm.com>
On Tue, Mar 22, 2011 at 11:50:14PM -0700, Paul E. McKenney wrote:
> On Tue, Mar 22, 2011 at 10:28:33AM -0700, Robert Love wrote:
> > On Thu, 2011-03-17 at 20:41 -0700, Lai Jiangshan wrote:
> > > - call_rcu(&rdata->rcu, fc_rport_free_rcu);
> > > + kfree_rcu(rdata, rcu);
> >
> > I think this last line should be:
> >
> > kfree_rcu(rdata, &rdata->rcu);
>
> Hello, Robert,
>
> I believe that it is correct as is. The kfree_rcu() definition is as
> follows:
>
> #define kfree_rcu(ptr, rcu_head) \
> __kfree_rcu(&((ptr)->rcu_head), offsetof(typeof(*(ptr)), rcu_head))
>
> Then __kfree_rcu() encodes the offset into the rcu_head so that it can
> be handled properly at the end of the grace period.
To be fair to Robert, I had the same thought, but decided to check the
definition of kfree_rcu before I made the same comment. Unfortunately,
the definition of kfree_rcu is still not in Linus' tree (and there's no
indication in the patch series about where to find the definition). It
would have been better if kfree_rcu were patch 1/n in the series, then
we'd've been able to review it more effectively.
Any idea when kfree_rcu will be submitted to Linus?
--
Matthew Wilcox Intel Open Source Technology Centre
"Bill, look, we understand that you're interested in selling us this
operating system, but compare it to ours. We can't possibly take such
a retrograde step."
^ permalink raw reply
* Re: [PATCH] ip_fragment:kernel may panic when replay big packet with RST flag
From: Feng Gao @ 2011-03-23 11:34 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev
In-Reply-To: <1300878116.3063.101.camel@edumazet-laptop>
Sorry, I will google it.
I find this problem in 2.4.35. and then I read the Latest code,maybe
it has the same problem too.
diff -Nur skbuff.h skbuff.h.frag
--- skbuff.h 2008-08-21 02:16:14.000000000 +0800
+++ skbuff.h.frag 2011-03-23 19:30:20.000000000 +0800
@@ -1754,6 +1754,20 @@
__nf_copy(dst, src);
}
+/* This func for ip_fragment deal with RST packet */
+static inline void nf_copy_rst(struct sk_buff *dst, const struct sk_buff *src)
+{
+#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
+ nf_conntrack_put(dst->nfct);
+ nf_conntrack_put_reasm(dst->nfct_reasm);
+#endif
+#ifdef CONFIG_BRIDGE_NETFILTER
+ nf_bridge_put(dst->nf_bridge);
+ dst->nf_bridge = src->nf_bridge;
+ nf_bridge_get(src->nf_bridge);
+#endif
+}
+
#ifdef CONFIG_NETWORK_SECMARK
static inline void skb_copy_secmark(struct sk_buff *to, const struct
sk_buff *from)
{
diff -Nur ip_output.c ip_output.c.frag
--- ip_output.c 2008-08-21 02:16:14.000000000 +0800
+++ ip_output.c.frag 2011-03-23 19:27:57.000000000 +0800
@@ -413,7 +413,14 @@
#ifdef CONFIG_NET_SCHED
to->tc_index = from->tc_index;
#endif
- nf_copy(to, from);
+ struct tcphdr *th = (struct tcphdr *)((char *)iph + (iph->ihl << 2));
+ if(iph->protocol == IPPROTO_TCP && th->rst){
+ /*RST packet*/
+ nf_copy_rst(to, from);
+ }else{
+ /*Other packet*/
+ nf_copy(to, from);
+ }
#if defined(CONFIG_NETFILTER_XT_TARGET_TRACE) || \
defined(CONFIG_NETFILTER_XT_TARGET_TRACE_MODULE)
to->nf_trace = from->nf_trace;
2011/3/23 Eric Dumazet <eric.dumazet@gmail.com>:
> Le mercredi 23 mars 2011 à 18:49 +0800, Feng Gao a écrit :
>> Hello everyone:
>>
>> PC(A)-linux(B)-PC(C)
>> computer(linux B) with two net interface,eth0 and eth1.
>> PC(A) send syn to PC(C) though linux B.
>> then PC(C) replay a big packet with RST flag(use tcpsic or other tools).
>>
>> This RST packet(1480) come in from eth0(mtu 1500) and go out from
>> eth1(mtu 700), so this RST packet should fragment.
>>
>> BUT in tcp_packet func: if the connection has no reply packet,and
>> receive the RST packet.ip_conntrack should destroy.
>> if (!test_bit(IPS_SEEN_REPLY_BIT, &ct->status)) {
>> /* If only reply is a RST, we can consider ourselves not to
>> have an established connection: this is a fairly common
>> problem case, so we can delete the conntrack
>> immediately. --RR */
>> if (th->rst) {
>> nf_ct_kill_acct(ct, ctinfo, skb);
>> return NF_ACCEPT;
>> }
>> }
>>
>> BUT the skb->nfct is not set NULL in func nf_ct_kill_acct.
>> so when this RST packet goto ip_fragment,ip_fragment call nf_copy, in
>> __nf_copy func
>> the fragment skb->nfct point to the destory mem.
>> dst->nfct = src->nfct;
>> nf_conntrack_get(src->nfct);
>>
>> SO finally.kfree_skb call destroy_conntrack again. this may result in
>> LINUX B kernel panic.
>>
>>
>> here is the patch,sorry ,i dont know how to use diff to generate patch.
>> :-D
>>
>
> Hi Feng
>
> Oh well, please learn how to do this. If you are able to make such
> analysis, you sure can learn how to use diff !!!
>
> What is the version of linux you use ?
>
>
>
>
^ permalink raw reply
* Re: [PATCH] ip_fragment:kernel may panic when replay big packet with RST flag
From: Feng Gao @ 2011-03-23 11:37 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev
In-Reply-To: <AANLkTik5=zcsVHwLsFqMvRvoLyPdhK=1dRUny5+jqoxf@mail.gmail.com>
This patch is for 2.6.25.16.
2011/3/23 Feng Gao <kernel.goter@gmail.com>:
> Sorry, I will google it.
> I find this problem in 2.4.35. and then I read the Latest code,maybe
> it has the same problem too.
>
> diff -Nur skbuff.h skbuff.h.frag
> --- skbuff.h 2008-08-21 02:16:14.000000000 +0800
> +++ skbuff.h.frag 2011-03-23 19:30:20.000000000 +0800
> @@ -1754,6 +1754,20 @@
> __nf_copy(dst, src);
> }
>
> +/* This func for ip_fragment deal with RST packet */
> +static inline void nf_copy_rst(struct sk_buff *dst, const struct sk_buff *src)
> +{
> +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
> + nf_conntrack_put(dst->nfct);
> + nf_conntrack_put_reasm(dst->nfct_reasm);
> +#endif
> +#ifdef CONFIG_BRIDGE_NETFILTER
> + nf_bridge_put(dst->nf_bridge);
> + dst->nf_bridge = src->nf_bridge;
> + nf_bridge_get(src->nf_bridge);
> +#endif
> +}
> +
> #ifdef CONFIG_NETWORK_SECMARK
> static inline void skb_copy_secmark(struct sk_buff *to, const struct
> sk_buff *from)
> {
>
> diff -Nur ip_output.c ip_output.c.frag
> --- ip_output.c 2008-08-21 02:16:14.000000000 +0800
> +++ ip_output.c.frag 2011-03-23 19:27:57.000000000 +0800
> @@ -413,7 +413,14 @@
> #ifdef CONFIG_NET_SCHED
> to->tc_index = from->tc_index;
> #endif
> - nf_copy(to, from);
> + struct tcphdr *th = (struct tcphdr *)((char *)iph + (iph->ihl << 2));
> + if(iph->protocol == IPPROTO_TCP && th->rst){
> + /*RST packet*/
> + nf_copy_rst(to, from);
> + }else{
> + /*Other packet*/
> + nf_copy(to, from);
> + }
> #if defined(CONFIG_NETFILTER_XT_TARGET_TRACE) || \
> defined(CONFIG_NETFILTER_XT_TARGET_TRACE_MODULE)
> to->nf_trace = from->nf_trace;
>
> 2011/3/23 Eric Dumazet <eric.dumazet@gmail.com>:
>> Le mercredi 23 mars 2011 à 18:49 +0800, Feng Gao a écrit :
>>> Hello everyone:
>>>
>>> PC(A)-linux(B)-PC(C)
>>> computer(linux B) with two net interface,eth0 and eth1.
>>> PC(A) send syn to PC(C) though linux B.
>>> then PC(C) replay a big packet with RST flag(use tcpsic or other tools).
>>>
>>> This RST packet(1480) come in from eth0(mtu 1500) and go out from
>>> eth1(mtu 700), so this RST packet should fragment.
>>>
>>> BUT in tcp_packet func: if the connection has no reply packet,and
>>> receive the RST packet.ip_conntrack should destroy.
>>> if (!test_bit(IPS_SEEN_REPLY_BIT, &ct->status)) {
>>> /* If only reply is a RST, we can consider ourselves not to
>>> have an established connection: this is a fairly common
>>> problem case, so we can delete the conntrack
>>> immediately. --RR */
>>> if (th->rst) {
>>> nf_ct_kill_acct(ct, ctinfo, skb);
>>> return NF_ACCEPT;
>>> }
>>> }
>>>
>>> BUT the skb->nfct is not set NULL in func nf_ct_kill_acct.
>>> so when this RST packet goto ip_fragment,ip_fragment call nf_copy, in
>>> __nf_copy func
>>> the fragment skb->nfct point to the destory mem.
>>> dst->nfct = src->nfct;
>>> nf_conntrack_get(src->nfct);
>>>
>>> SO finally.kfree_skb call destroy_conntrack again. this may result in
>>> LINUX B kernel panic.
>>>
>>>
>>> here is the patch,sorry ,i dont know how to use diff to generate patch.
>>> :-D
>>>
>>
>> Hi Feng
>>
>> Oh well, please learn how to do this. If you are able to make such
>> analysis, you sure can learn how to use diff !!!
>>
>> What is the version of linux you use ?
>>
>>
>>
>>
>
^ permalink raw reply
* [PATCH] myri10ge: fix rmmod crash
From: Stanislaw Gruszka @ 2011-03-23 12:44 UTC (permalink / raw)
To: netdev; +Cc: stable, Brice Goglin, Andrew Gallatin
Rmmod myri10ge crash at free_netdev() -> netif_napi_del(), because napi
structures are already deallocated. To fix call netif_napi_del() before
kfree() at myri10ge_free_slices().
Cc: stable@kernel.org
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
---
drivers/net/myri10ge/myri10ge.c | 1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/drivers/net/myri10ge/myri10ge.c b/drivers/net/myri10ge/myri10ge.c
index ea5cfe2..24386a8 100644
--- a/drivers/net/myri10ge/myri10ge.c
+++ b/drivers/net/myri10ge/myri10ge.c
@@ -3645,6 +3645,7 @@ static void myri10ge_free_slices(struct myri10ge_priv *mgp)
dma_free_coherent(&pdev->dev, bytes,
ss->fw_stats, ss->fw_stats_bus);
ss->fw_stats = NULL;
+ netif_napi_del(&ss->napi);
}
}
kfree(mgp->ss);
--
1.7.1
_______________________________________________
stable mailing list
stable@linux.kernel.org
http://linux.kernel.org/mailman/listinfo/stable
^ 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