Merge tag 'rxrpc-rewrite-20160824-2' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs
David Howells says:
====================
rxrpc: Add better client conn management strategy
These two patches add a better client connection management strategy. They
need to be applied on top of the just-posted fixes.
(1) Duplicate the connection list and separate out procfs iteration from
garbage collection. This is necessary for the next patch as with that
client connections no longer appear on a single list and may not
appear on a list at all - and really don't want to be exposed to the
old garbage collector.
(Note that client conns aren't left dangling, they're also in a tree
rooted in the local endpoint so that they can be found by a user
wanting to make a new client call. Service conns do not appear in
this tree.)
(2) Implement a better lifetime management and garbage collection strategy
for client connections.
In this, a client connection can be in one of five cache states
(inactive, waiting, active, culled and idle). Limits are set on the
number of client conns that may be active at any one time and makes
users wait if they want to start a new call when there isn't capacity
available.
To make capacity available, active and idle connections can be culled,
after a short delay (to allow for retransmission). The delay is
reduced if the capacity exceeds a tunable threshold.
If there is spare capacity, client conns are permitted to hang around
a fair bit longer (tunable) so as to allow reuse of negotiated
security contexts.
After this patch, the client conn strategy is separate from that of
service conns (which continues to use the old code for the moment).
This difference in strategy is because the client side retains control
over when it allows a connection to become active, whereas the service
side has no control over when it sees a new connection or a new call
on an old connection.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c
index 6fb77d7..f02a9753 100644
--- a/drivers/infiniband/hw/mlx5/main.c
+++ b/drivers/infiniband/hw/mlx5/main.c
@@ -748,8 +748,7 @@
&props->active_width);
if (err)
goto out;
- err = mlx5_query_port_proto_oper(mdev, &props->active_speed, MLX5_PTYS_IB,
- port);
+ err = mlx5_query_port_ib_proto_oper(mdev, &props->active_speed, port);
if (err)
goto out;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 05cc1ef..dad326c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -3,7 +3,7 @@
mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \
health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o \
mad.o transobj.o vport.o sriov.o fs_cmd.o fs_core.o \
- fs_counters.o rl.o
+ fs_counters.o rl.o lag.o
mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o eswitch_offloads.o \
en_main.o en_common.o en_fs.o en_ethtool.o en_tx.o \
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
index e9f6afb..bb89f04 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
@@ -285,6 +285,8 @@
case MLX5_CMD_OP_DEALLOC_TRANSPORT_DOMAIN:
case MLX5_CMD_OP_DELETE_VXLAN_UDP_DPORT:
case MLX5_CMD_OP_DELETE_L2_TABLE_ENTRY:
+ case MLX5_CMD_OP_DESTROY_LAG:
+ case MLX5_CMD_OP_DESTROY_VPORT_LAG:
case MLX5_CMD_OP_DESTROY_TIR:
case MLX5_CMD_OP_DESTROY_SQ:
case MLX5_CMD_OP_DESTROY_RQ:
@@ -376,6 +378,10 @@
case MLX5_CMD_OP_ADD_VXLAN_UDP_DPORT:
case MLX5_CMD_OP_SET_L2_TABLE_ENTRY:
case MLX5_CMD_OP_QUERY_L2_TABLE_ENTRY:
+ case MLX5_CMD_OP_CREATE_LAG:
+ case MLX5_CMD_OP_MODIFY_LAG:
+ case MLX5_CMD_OP_QUERY_LAG:
+ case MLX5_CMD_OP_CREATE_VPORT_LAG:
case MLX5_CMD_OP_CREATE_TIR:
case MLX5_CMD_OP_MODIFY_TIR:
case MLX5_CMD_OP_QUERY_TIR:
@@ -514,6 +520,12 @@
MLX5_COMMAND_STR_CASE(DELETE_L2_TABLE_ENTRY);
MLX5_COMMAND_STR_CASE(SET_WOL_ROL);
MLX5_COMMAND_STR_CASE(QUERY_WOL_ROL);
+ MLX5_COMMAND_STR_CASE(CREATE_LAG);
+ MLX5_COMMAND_STR_CASE(MODIFY_LAG);
+ MLX5_COMMAND_STR_CASE(QUERY_LAG);
+ MLX5_COMMAND_STR_CASE(DESTROY_LAG);
+ MLX5_COMMAND_STR_CASE(CREATE_VPORT_LAG);
+ MLX5_COMMAND_STR_CASE(DESTROY_VPORT_LAG);
MLX5_COMMAND_STR_CASE(CREATE_TIR);
MLX5_COMMAND_STR_CASE(MODIFY_TIR);
MLX5_COMMAND_STR_CASE(DESTROY_TIR);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 1b495ef..61902b1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -651,40 +651,6 @@
void *ppriv;
};
-enum mlx5e_link_mode {
- MLX5E_1000BASE_CX_SGMII = 0,
- MLX5E_1000BASE_KX = 1,
- MLX5E_10GBASE_CX4 = 2,
- MLX5E_10GBASE_KX4 = 3,
- MLX5E_10GBASE_KR = 4,
- MLX5E_20GBASE_KR2 = 5,
- MLX5E_40GBASE_CR4 = 6,
- MLX5E_40GBASE_KR4 = 7,
- MLX5E_56GBASE_R4 = 8,
- MLX5E_10GBASE_CR = 12,
- MLX5E_10GBASE_SR = 13,
- MLX5E_10GBASE_ER = 14,
- MLX5E_40GBASE_SR4 = 15,
- MLX5E_40GBASE_LR4 = 16,
- MLX5E_50GBASE_SR2 = 18,
- MLX5E_100GBASE_CR4 = 20,
- MLX5E_100GBASE_SR4 = 21,
- MLX5E_100GBASE_KR4 = 22,
- MLX5E_100GBASE_LR4 = 23,
- MLX5E_100BASE_TX = 24,
- MLX5E_1000BASE_T = 25,
- MLX5E_10GBASE_T = 26,
- MLX5E_25GBASE_CR = 27,
- MLX5E_25GBASE_KR = 28,
- MLX5E_25GBASE_SR = 29,
- MLX5E_50GBASE_CR2 = 30,
- MLX5E_50GBASE_KR2 = 31,
- MLX5E_LINK_MODES_NUMBER,
-};
-
-#define MLX5E_PROT_MASK(link_mode) (1 << link_mode)
-
-
void mlx5e_build_ptys2ethtool_map(void);
void mlx5e_send_nop(struct mlx5e_sq *sq, bool notify_hw);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 10fa12a..fa40414 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -489,7 +489,8 @@
rqc = MLX5_ADDR_OF(modify_rq_in, in, ctx);
MLX5_SET(modify_rq_in, in, rq_state, MLX5_RQC_STATE_RDY);
- MLX5_SET64(modify_rq_in, in, modify_bitmask, MLX5_RQ_BITMASK_VSD);
+ MLX5_SET64(modify_rq_in, in, modify_bitmask,
+ MLX5_MODIFY_RQ_IN_MODIFY_BITMASK_VSD);
MLX5_SET(rqc, rqc, vsd, vsd);
MLX5_SET(rqc, rqc, state, MLX5_RQC_STATE_RDY);
@@ -2024,6 +2025,10 @@
MLX5_SET(tisc, tisc, prio, tc << 1);
MLX5_SET(tisc, tisc, transport_domain, mdev->mlx5e_res.td.tdn);
+
+ if (mlx5_lag_is_lacp_owner(mdev))
+ MLX5_SET(tisc, tisc, strict_lag_tx_port_affinity, 1);
+
return mlx5_core_create_tis(mdev, in, sizeof(in), &priv->tisn[tc]);
}
@@ -3368,6 +3373,8 @@
struct mlx5_eswitch *esw = mdev->priv.eswitch;
struct mlx5_eswitch_rep rep;
+ mlx5_lag_add(mdev, netdev);
+
if (mlx5e_vxlan_allowed(mdev)) {
rtnl_lock();
udp_tunnel_get_rx_info(netdev);
@@ -3390,6 +3397,7 @@
{
queue_work(priv->wq, &priv->set_rx_mode_work);
mlx5e_disable_async_events(priv);
+ mlx5_lag_remove(priv->mdev);
}
static const struct mlx5e_profile mlx5e_nic_profile = {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
index 7aaefa9..7a0415e6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
@@ -58,6 +58,7 @@
int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev,
u16 vport,
+ enum fs_flow_table_op_mod op_mod,
enum fs_flow_table_type type, unsigned int level,
unsigned int log_size, struct mlx5_flow_table
*next_ft, unsigned int *table_id)
@@ -69,10 +70,6 @@
MLX5_SET(create_flow_table_in, in, opcode,
MLX5_CMD_OP_CREATE_FLOW_TABLE);
- if (next_ft) {
- MLX5_SET(create_flow_table_in, in, table_miss_mode, 1);
- MLX5_SET(create_flow_table_in, in, table_miss_id, next_ft->id);
- }
MLX5_SET(create_flow_table_in, in, table_type, type);
MLX5_SET(create_flow_table_in, in, level, level);
MLX5_SET(create_flow_table_in, in, log_size, log_size);
@@ -81,6 +78,22 @@
MLX5_SET(create_flow_table_in, in, other_vport, 1);
}
+ switch (op_mod) {
+ case FS_FT_OP_MOD_NORMAL:
+ if (next_ft) {
+ MLX5_SET(create_flow_table_in, in, table_miss_mode, 1);
+ MLX5_SET(create_flow_table_in, in, table_miss_id, next_ft->id);
+ }
+ break;
+
+ case FS_FT_OP_MOD_LAG_DEMUX:
+ MLX5_SET(create_flow_table_in, in, op_mod, 0x1);
+ if (next_ft)
+ MLX5_SET(create_flow_table_in, in, lag_master_next_table_id,
+ next_ft->id);
+ break;
+ }
+
err = mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
if (!err)
*table_id = MLX5_GET(create_flow_table_out, out,
@@ -117,17 +130,32 @@
MLX5_CMD_OP_MODIFY_FLOW_TABLE);
MLX5_SET(modify_flow_table_in, in, table_type, ft->type);
MLX5_SET(modify_flow_table_in, in, table_id, ft->id);
- if (ft->vport) {
- MLX5_SET(modify_flow_table_in, in, vport_number, ft->vport);
- MLX5_SET(modify_flow_table_in, in, other_vport, 1);
- }
- MLX5_SET(modify_flow_table_in, in, modify_field_select,
- MLX5_MODIFY_FLOW_TABLE_MISS_TABLE_ID);
- if (next_ft) {
- MLX5_SET(modify_flow_table_in, in, table_miss_mode, 1);
- MLX5_SET(modify_flow_table_in, in, table_miss_id, next_ft->id);
+
+ if (ft->op_mod == FS_FT_OP_MOD_LAG_DEMUX) {
+ MLX5_SET(modify_flow_table_in, in, modify_field_select,
+ MLX5_MODIFY_FLOW_TABLE_LAG_NEXT_TABLE_ID);
+ if (next_ft) {
+ MLX5_SET(modify_flow_table_in, in,
+ lag_master_next_table_id, next_ft->id);
+ } else {
+ MLX5_SET(modify_flow_table_in, in,
+ lag_master_next_table_id, 0);
+ }
} else {
- MLX5_SET(modify_flow_table_in, in, table_miss_mode, 0);
+ if (ft->vport) {
+ MLX5_SET(modify_flow_table_in, in, vport_number,
+ ft->vport);
+ MLX5_SET(modify_flow_table_in, in, other_vport, 1);
+ }
+ MLX5_SET(modify_flow_table_in, in, modify_field_select,
+ MLX5_MODIFY_FLOW_TABLE_MISS_TABLE_ID);
+ if (next_ft) {
+ MLX5_SET(modify_flow_table_in, in, table_miss_mode, 1);
+ MLX5_SET(modify_flow_table_in, in, table_miss_id,
+ next_ft->id);
+ } else {
+ MLX5_SET(modify_flow_table_in, in, table_miss_mode, 0);
+ }
}
return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h
index ac52fdf..c5bc468 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h
@@ -35,6 +35,7 @@
int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev,
u16 vport,
+ enum fs_flow_table_op_mod op_mod,
enum fs_flow_table_type type, unsigned int level,
unsigned int log_size, struct mlx5_flow_table
*next_ft, unsigned int *table_id);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index 243efc5..f5571a5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -96,6 +96,10 @@
#define OFFLOADS_NUM_PRIOS 1
#define OFFLOADS_MIN_LEVEL (ANCHOR_MIN_LEVEL + 1)
+#define LAG_PRIO_NUM_LEVELS 1
+#define LAG_NUM_PRIOS 1
+#define LAG_MIN_LEVEL (OFFLOADS_MIN_LEVEL + 1)
+
struct node_caps {
size_t arr_sz;
long *caps;
@@ -111,12 +115,16 @@
int num_levels;
} root_fs = {
.type = FS_TYPE_NAMESPACE,
- .ar_size = 6,
+ .ar_size = 7,
.children = (struct init_tree_node[]) {
ADD_PRIO(0, BY_PASS_MIN_LEVEL, 0,
FS_CHAINING_CAPS,
ADD_NS(ADD_MULTIPLE_PRIO(MLX5_BY_PASS_NUM_PRIOS,
BY_PASS_PRIO_NUM_LEVELS))),
+ ADD_PRIO(0, LAG_MIN_LEVEL, 0,
+ FS_CHAINING_CAPS,
+ ADD_NS(ADD_MULTIPLE_PRIO(LAG_NUM_PRIOS,
+ LAG_PRIO_NUM_LEVELS))),
ADD_PRIO(0, OFFLOADS_MIN_LEVEL, 0, {},
ADD_NS(ADD_MULTIPLE_PRIO(OFFLOADS_NUM_PRIOS, OFFLOADS_MAX_FT))),
ADD_PRIO(0, ETHTOOL_MIN_LEVEL, 0,
@@ -477,7 +485,8 @@
}
static struct mlx5_flow_table *alloc_flow_table(int level, u16 vport, int max_fte,
- enum fs_flow_table_type table_type)
+ enum fs_flow_table_type table_type,
+ enum fs_flow_table_op_mod op_mod)
{
struct mlx5_flow_table *ft;
@@ -487,6 +496,7 @@
ft->level = level;
ft->node.type = FS_TYPE_FLOW_TABLE;
+ ft->op_mod = op_mod;
ft->type = table_type;
ft->vport = vport;
ft->max_fte = max_fte;
@@ -724,6 +734,7 @@
}
static struct mlx5_flow_table *__mlx5_create_flow_table(struct mlx5_flow_namespace *ns,
+ enum fs_flow_table_op_mod op_mod,
u16 vport, int prio,
int max_fte, u32 level)
{
@@ -756,18 +767,19 @@
level += fs_prio->start_level;
ft = alloc_flow_table(level,
vport,
- roundup_pow_of_two(max_fte),
- root->table_type);
+ max_fte ? roundup_pow_of_two(max_fte) : 0,
+ root->table_type,
+ op_mod);
if (!ft) {
err = -ENOMEM;
goto unlock_root;
}
tree_init_node(&ft->node, 1, del_flow_table);
- log_table_sz = ilog2(ft->max_fte);
+ log_table_sz = ft->max_fte ? ilog2(ft->max_fte) : 0;
next_ft = find_next_chained_ft(fs_prio);
- err = mlx5_cmd_create_flow_table(root->dev, ft->vport, ft->type, ft->level,
- log_table_sz, next_ft, &ft->id);
+ err = mlx5_cmd_create_flow_table(root->dev, ft->vport, ft->op_mod, ft->type,
+ ft->level, log_table_sz, next_ft, &ft->id);
if (err)
goto free_ft;
@@ -794,16 +806,27 @@
int prio, int max_fte,
u32 level)
{
- return __mlx5_create_flow_table(ns, 0, prio, max_fte, level);
+ return __mlx5_create_flow_table(ns, FS_FT_OP_MOD_NORMAL, 0, prio,
+ max_fte, level);
}
struct mlx5_flow_table *mlx5_create_vport_flow_table(struct mlx5_flow_namespace *ns,
int prio, int max_fte,
u32 level, u16 vport)
{
- return __mlx5_create_flow_table(ns, vport, prio, max_fte, level);
+ return __mlx5_create_flow_table(ns, FS_FT_OP_MOD_NORMAL, vport, prio,
+ max_fte, level);
}
+struct mlx5_flow_table *mlx5_create_lag_demux_flow_table(
+ struct mlx5_flow_namespace *ns,
+ int prio, u32 level)
+{
+ return __mlx5_create_flow_table(ns, FS_FT_OP_MOD_LAG_DEMUX, 0, prio, 0,
+ level);
+}
+EXPORT_SYMBOL(mlx5_create_lag_demux_flow_table);
+
struct mlx5_flow_table *mlx5_create_auto_grouped_flow_table(struct mlx5_flow_namespace *ns,
int prio,
int num_flow_table_entries,
@@ -1381,6 +1404,7 @@
switch (type) {
case MLX5_FLOW_NAMESPACE_BYPASS:
+ case MLX5_FLOW_NAMESPACE_LAG:
case MLX5_FLOW_NAMESPACE_OFFLOADS:
case MLX5_FLOW_NAMESPACE_ETHTOOL:
case MLX5_FLOW_NAMESPACE_KERNEL:
@@ -1403,6 +1427,16 @@
return &steering->esw_ingress_root_ns->ns;
else
return NULL;
+ case MLX5_FLOW_NAMESPACE_SNIFFER_RX:
+ if (steering->sniffer_rx_root_ns)
+ return &steering->sniffer_rx_root_ns->ns;
+ else
+ return NULL;
+ case MLX5_FLOW_NAMESPACE_SNIFFER_TX:
+ if (steering->sniffer_tx_root_ns)
+ return &steering->sniffer_tx_root_ns->ns;
+ else
+ return NULL;
default:
return NULL;
}
@@ -1702,10 +1736,46 @@
cleanup_root_ns(steering->esw_egress_root_ns);
cleanup_root_ns(steering->esw_ingress_root_ns);
cleanup_root_ns(steering->fdb_root_ns);
+ cleanup_root_ns(steering->sniffer_rx_root_ns);
+ cleanup_root_ns(steering->sniffer_tx_root_ns);
mlx5_cleanup_fc_stats(dev);
kfree(steering);
}
+static int init_sniffer_tx_root_ns(struct mlx5_flow_steering *steering)
+{
+ struct fs_prio *prio;
+
+ steering->sniffer_tx_root_ns = create_root_ns(steering, FS_FT_SNIFFER_TX);
+ if (!steering->sniffer_tx_root_ns)
+ return -ENOMEM;
+
+ /* Create single prio */
+ prio = fs_create_prio(&steering->sniffer_tx_root_ns->ns, 0, 1);
+ if (IS_ERR(prio)) {
+ cleanup_root_ns(steering->sniffer_tx_root_ns);
+ return PTR_ERR(prio);
+ }
+ return 0;
+}
+
+static int init_sniffer_rx_root_ns(struct mlx5_flow_steering *steering)
+{
+ struct fs_prio *prio;
+
+ steering->sniffer_rx_root_ns = create_root_ns(steering, FS_FT_SNIFFER_RX);
+ if (!steering->sniffer_rx_root_ns)
+ return -ENOMEM;
+
+ /* Create single prio */
+ prio = fs_create_prio(&steering->sniffer_rx_root_ns->ns, 0, 1);
+ if (IS_ERR(prio)) {
+ cleanup_root_ns(steering->sniffer_rx_root_ns);
+ return PTR_ERR(prio);
+ }
+ return 0;
+}
+
static int init_fdb_root_ns(struct mlx5_flow_steering *steering)
{
struct fs_prio *prio;
@@ -1802,6 +1872,18 @@
}
}
+ if (MLX5_CAP_FLOWTABLE_SNIFFER_RX(dev, ft_support)) {
+ err = init_sniffer_rx_root_ns(steering);
+ if (err)
+ goto err;
+ }
+
+ if (MLX5_CAP_FLOWTABLE_SNIFFER_TX(dev, ft_support)) {
+ err = init_sniffer_tx_root_ns(steering);
+ if (err)
+ goto err;
+ }
+
return 0;
err:
mlx5_cleanup_fs(dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
index 9cffb6a..71ff03b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
@@ -49,6 +49,13 @@
FS_FT_ESW_EGRESS_ACL = 0x2,
FS_FT_ESW_INGRESS_ACL = 0x3,
FS_FT_FDB = 0X4,
+ FS_FT_SNIFFER_RX = 0X5,
+ FS_FT_SNIFFER_TX = 0X6,
+};
+
+enum fs_flow_table_op_mod {
+ FS_FT_OP_MOD_NORMAL,
+ FS_FT_OP_MOD_LAG_DEMUX,
};
enum fs_fte_status {
@@ -61,6 +68,8 @@
struct mlx5_flow_root_namespace *fdb_root_ns;
struct mlx5_flow_root_namespace *esw_egress_root_ns;
struct mlx5_flow_root_namespace *esw_ingress_root_ns;
+ struct mlx5_flow_root_namespace *sniffer_tx_root_ns;
+ struct mlx5_flow_root_namespace *sniffer_rx_root_ns;
};
struct fs_node {
@@ -93,6 +102,7 @@
unsigned int max_fte;
unsigned int level;
enum fs_flow_table_type type;
+ enum fs_flow_table_op_mod op_mod;
struct {
bool active;
unsigned int required_groups;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag.c b/drivers/net/ethernet/mellanox/mlx5/core/lag.c
new file mode 100644
index 0000000..92c3e0d
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag.c
@@ -0,0 +1,602 @@
+/*
+ * Copyright (c) 2016, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/netdevice.h>
+#include <linux/mlx5/driver.h>
+#include <linux/mlx5/vport.h>
+#include "mlx5_core.h"
+
+enum {
+ MLX5_LAG_FLAG_BONDED = 1 << 0,
+};
+
+struct lag_func {
+ struct mlx5_core_dev *dev;
+ struct net_device *netdev;
+};
+
+/* Used for collection of netdev event info. */
+struct lag_tracker {
+ enum netdev_lag_tx_type tx_type;
+ struct netdev_lag_lower_state_info netdev_state[MLX5_MAX_PORTS];
+ bool is_bonded;
+};
+
+/* LAG data of a ConnectX card.
+ * It serves both its phys functions.
+ */
+struct mlx5_lag {
+ u8 flags;
+ u8 v2p_map[MLX5_MAX_PORTS];
+ struct lag_func pf[MLX5_MAX_PORTS];
+ struct lag_tracker tracker;
+ struct delayed_work bond_work;
+ struct notifier_block nb;
+};
+
+/* General purpose, use for short periods of time.
+ * Beware of lock dependencies (preferably, no locks should be acquired
+ * under it).
+ */
+static DEFINE_MUTEX(lag_mutex);
+
+static int mlx5_cmd_create_lag(struct mlx5_core_dev *dev, u8 remap_port1,
+ u8 remap_port2)
+{
+ u32 in[MLX5_ST_SZ_DW(create_lag_in)] = {0};
+ u32 out[MLX5_ST_SZ_DW(create_lag_out)] = {0};
+ void *lag_ctx = MLX5_ADDR_OF(create_lag_in, in, ctx);
+
+ MLX5_SET(create_lag_in, in, opcode, MLX5_CMD_OP_CREATE_LAG);
+
+ MLX5_SET(lagc, lag_ctx, tx_remap_affinity_1, remap_port1);
+ MLX5_SET(lagc, lag_ctx, tx_remap_affinity_2, remap_port2);
+
+ return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+}
+
+static int mlx5_cmd_modify_lag(struct mlx5_core_dev *dev, u8 remap_port1,
+ u8 remap_port2)
+{
+ u32 in[MLX5_ST_SZ_DW(modify_lag_in)] = {0};
+ u32 out[MLX5_ST_SZ_DW(modify_lag_out)] = {0};
+ void *lag_ctx = MLX5_ADDR_OF(modify_lag_in, in, ctx);
+
+ MLX5_SET(modify_lag_in, in, opcode, MLX5_CMD_OP_MODIFY_LAG);
+ MLX5_SET(modify_lag_in, in, field_select, 0x1);
+
+ MLX5_SET(lagc, lag_ctx, tx_remap_affinity_1, remap_port1);
+ MLX5_SET(lagc, lag_ctx, tx_remap_affinity_2, remap_port2);
+
+ return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+}
+
+static int mlx5_cmd_destroy_lag(struct mlx5_core_dev *dev)
+{
+ u32 in[MLX5_ST_SZ_DW(destroy_lag_in)] = {0};
+ u32 out[MLX5_ST_SZ_DW(destroy_lag_out)] = {0};
+
+ MLX5_SET(destroy_lag_in, in, opcode, MLX5_CMD_OP_DESTROY_LAG);
+
+ return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+}
+
+int mlx5_cmd_create_vport_lag(struct mlx5_core_dev *dev)
+{
+ u32 in[MLX5_ST_SZ_DW(create_vport_lag_in)] = {0};
+ u32 out[MLX5_ST_SZ_DW(create_vport_lag_out)] = {0};
+
+ MLX5_SET(create_vport_lag_in, in, opcode, MLX5_CMD_OP_CREATE_VPORT_LAG);
+
+ return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+}
+EXPORT_SYMBOL(mlx5_cmd_create_vport_lag);
+
+int mlx5_cmd_destroy_vport_lag(struct mlx5_core_dev *dev)
+{
+ u32 in[MLX5_ST_SZ_DW(destroy_vport_lag_in)] = {0};
+ u32 out[MLX5_ST_SZ_DW(destroy_vport_lag_out)] = {0};
+
+ MLX5_SET(destroy_vport_lag_in, in, opcode, MLX5_CMD_OP_DESTROY_VPORT_LAG);
+
+ return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
+}
+EXPORT_SYMBOL(mlx5_cmd_destroy_vport_lag);
+
+static struct mlx5_lag *mlx5_lag_dev_get(struct mlx5_core_dev *dev)
+{
+ return dev->priv.lag;
+}
+
+static int mlx5_lag_dev_get_netdev_idx(struct mlx5_lag *ldev,
+ struct net_device *ndev)
+{
+ int i;
+
+ for (i = 0; i < MLX5_MAX_PORTS; i++)
+ if (ldev->pf[i].netdev == ndev)
+ return i;
+
+ return -1;
+}
+
+static bool mlx5_lag_is_bonded(struct mlx5_lag *ldev)
+{
+ return !!(ldev->flags & MLX5_LAG_FLAG_BONDED);
+}
+
+static void mlx5_infer_tx_affinity_mapping(struct lag_tracker *tracker,
+ u8 *port1, u8 *port2)
+{
+ if (tracker->tx_type == NETDEV_LAG_TX_TYPE_ACTIVEBACKUP) {
+ if (tracker->netdev_state[0].tx_enabled) {
+ *port1 = 1;
+ *port2 = 1;
+ } else {
+ *port1 = 2;
+ *port2 = 2;
+ }
+ } else {
+ *port1 = 1;
+ *port2 = 2;
+ if (!tracker->netdev_state[0].link_up)
+ *port1 = 2;
+ else if (!tracker->netdev_state[1].link_up)
+ *port2 = 1;
+ }
+}
+
+static void mlx5_activate_lag(struct mlx5_lag *ldev,
+ struct lag_tracker *tracker)
+{
+ struct mlx5_core_dev *dev0 = ldev->pf[0].dev;
+ int err;
+
+ ldev->flags |= MLX5_LAG_FLAG_BONDED;
+
+ mlx5_infer_tx_affinity_mapping(tracker, &ldev->v2p_map[0],
+ &ldev->v2p_map[1]);
+
+ err = mlx5_cmd_create_lag(dev0, ldev->v2p_map[0], ldev->v2p_map[1]);
+ if (err)
+ mlx5_core_err(dev0,
+ "Failed to create LAG (%d)\n",
+ err);
+}
+
+static void mlx5_deactivate_lag(struct mlx5_lag *ldev)
+{
+ struct mlx5_core_dev *dev0 = ldev->pf[0].dev;
+ int err;
+
+ ldev->flags &= ~MLX5_LAG_FLAG_BONDED;
+
+ err = mlx5_cmd_destroy_lag(dev0);
+ if (err)
+ mlx5_core_err(dev0,
+ "Failed to destroy LAG (%d)\n",
+ err);
+}
+
+static void mlx5_do_bond(struct mlx5_lag *ldev)
+{
+ struct mlx5_core_dev *dev0 = ldev->pf[0].dev;
+ struct mlx5_core_dev *dev1 = ldev->pf[1].dev;
+ struct lag_tracker tracker;
+ u8 v2p_port1, v2p_port2;
+ int i, err;
+
+ if (!dev0 || !dev1)
+ return;
+
+ mutex_lock(&lag_mutex);
+ tracker = ldev->tracker;
+ mutex_unlock(&lag_mutex);
+
+ if (tracker.is_bonded && !mlx5_lag_is_bonded(ldev)) {
+ if (mlx5_sriov_is_enabled(dev0) ||
+ mlx5_sriov_is_enabled(dev1)) {
+ mlx5_core_warn(dev0, "LAG is not supported with SRIOV");
+ return;
+ }
+
+ for (i = 0; i < MLX5_MAX_PORTS; i++)
+ mlx5_remove_dev_by_protocol(ldev->pf[i].dev,
+ MLX5_INTERFACE_PROTOCOL_IB);
+
+ mlx5_activate_lag(ldev, &tracker);
+
+ mlx5_add_dev_by_protocol(dev0, MLX5_INTERFACE_PROTOCOL_IB);
+ mlx5_nic_vport_enable_roce(dev1);
+ } else if (tracker.is_bonded && mlx5_lag_is_bonded(ldev)) {
+ mlx5_infer_tx_affinity_mapping(&tracker, &v2p_port1,
+ &v2p_port2);
+
+ if ((v2p_port1 != ldev->v2p_map[0]) ||
+ (v2p_port2 != ldev->v2p_map[1])) {
+ ldev->v2p_map[0] = v2p_port1;
+ ldev->v2p_map[1] = v2p_port2;
+
+ err = mlx5_cmd_modify_lag(dev0, v2p_port1, v2p_port2);
+ if (err)
+ mlx5_core_err(dev0,
+ "Failed to modify LAG (%d)\n",
+ err);
+ }
+ } else if (!tracker.is_bonded && mlx5_lag_is_bonded(ldev)) {
+ mlx5_remove_dev_by_protocol(dev0, MLX5_INTERFACE_PROTOCOL_IB);
+ mlx5_nic_vport_disable_roce(dev1);
+
+ mlx5_deactivate_lag(ldev);
+
+ for (i = 0; i < MLX5_MAX_PORTS; i++)
+ if (ldev->pf[i].dev)
+ mlx5_add_dev_by_protocol(ldev->pf[i].dev,
+ MLX5_INTERFACE_PROTOCOL_IB);
+ }
+}
+
+static void mlx5_queue_bond_work(struct mlx5_lag *ldev, unsigned long delay)
+{
+ schedule_delayed_work(&ldev->bond_work, delay);
+}
+
+static void mlx5_do_bond_work(struct work_struct *work)
+{
+ struct delayed_work *delayed_work = to_delayed_work(work);
+ struct mlx5_lag *ldev = container_of(delayed_work, struct mlx5_lag,
+ bond_work);
+ int status;
+
+ status = mutex_trylock(&mlx5_intf_mutex);
+ if (!status) {
+ /* 1 sec delay. */
+ mlx5_queue_bond_work(ldev, HZ);
+ return;
+ }
+
+ mlx5_do_bond(ldev);
+ mutex_unlock(&mlx5_intf_mutex);
+}
+
+static int mlx5_handle_changeupper_event(struct mlx5_lag *ldev,
+ struct lag_tracker *tracker,
+ struct net_device *ndev,
+ struct netdev_notifier_changeupper_info *info)
+{
+ struct net_device *upper = info->upper_dev, *ndev_tmp;
+ struct netdev_lag_upper_info *lag_upper_info;
+ bool is_bonded;
+ int bond_status = 0;
+ int num_slaves = 0;
+ int idx;
+
+ if (!netif_is_lag_master(upper))
+ return 0;
+
+ lag_upper_info = info->upper_info;
+
+ /* The event may still be of interest if the slave does not belong to
+ * us, but is enslaved to a master which has one or more of our netdevs
+ * as slaves (e.g., if a new slave is added to a master that bonds two
+ * of our netdevs, we should unbond).
+ */
+ rcu_read_lock();
+ for_each_netdev_in_bond_rcu(upper, ndev_tmp) {
+ idx = mlx5_lag_dev_get_netdev_idx(ldev, ndev_tmp);
+ if (idx > -1)
+ bond_status |= (1 << idx);
+
+ num_slaves++;
+ }
+ rcu_read_unlock();
+
+ /* None of this lagdev's netdevs are slaves of this master. */
+ if (!(bond_status & 0x3))
+ return 0;
+
+ if (lag_upper_info)
+ tracker->tx_type = lag_upper_info->tx_type;
+
+ /* Determine bonding status:
+ * A device is considered bonded if both its physical ports are slaves
+ * of the same lag master, and only them.
+ * Lag mode must be activebackup or hash.
+ */
+ is_bonded = (num_slaves == MLX5_MAX_PORTS) &&
+ (bond_status == 0x3) &&
+ ((tracker->tx_type == NETDEV_LAG_TX_TYPE_ACTIVEBACKUP) ||
+ (tracker->tx_type == NETDEV_LAG_TX_TYPE_HASH));
+
+ if (tracker->is_bonded != is_bonded) {
+ tracker->is_bonded = is_bonded;
+ return 1;
+ }
+
+ return 0;
+}
+
+static int mlx5_handle_changelowerstate_event(struct mlx5_lag *ldev,
+ struct lag_tracker *tracker,
+ struct net_device *ndev,
+ struct netdev_notifier_changelowerstate_info *info)
+{
+ struct netdev_lag_lower_state_info *lag_lower_info;
+ int idx;
+
+ if (!netif_is_lag_port(ndev))
+ return 0;
+
+ idx = mlx5_lag_dev_get_netdev_idx(ldev, ndev);
+ if (idx == -1)
+ return 0;
+
+ /* This information is used to determine virtual to physical
+ * port mapping.
+ */
+ lag_lower_info = info->lower_state_info;
+ if (!lag_lower_info)
+ return 0;
+
+ tracker->netdev_state[idx] = *lag_lower_info;
+
+ return 1;
+}
+
+static int mlx5_lag_netdev_event(struct notifier_block *this,
+ unsigned long event, void *ptr)
+{
+ struct net_device *ndev = netdev_notifier_info_to_dev(ptr);
+ struct lag_tracker tracker;
+ struct mlx5_lag *ldev;
+ int changed = 0;
+
+ if (!net_eq(dev_net(ndev), &init_net))
+ return NOTIFY_DONE;
+
+ if ((event != NETDEV_CHANGEUPPER) && (event != NETDEV_CHANGELOWERSTATE))
+ return NOTIFY_DONE;
+
+ ldev = container_of(this, struct mlx5_lag, nb);
+ tracker = ldev->tracker;
+
+ switch (event) {
+ case NETDEV_CHANGEUPPER:
+ changed = mlx5_handle_changeupper_event(ldev, &tracker, ndev,
+ ptr);
+ break;
+ case NETDEV_CHANGELOWERSTATE:
+ changed = mlx5_handle_changelowerstate_event(ldev, &tracker,
+ ndev, ptr);
+ break;
+ }
+
+ mutex_lock(&lag_mutex);
+ ldev->tracker = tracker;
+ mutex_unlock(&lag_mutex);
+
+ if (changed)
+ mlx5_queue_bond_work(ldev, 0);
+
+ return NOTIFY_DONE;
+}
+
+static struct mlx5_lag *mlx5_lag_dev_alloc(void)
+{
+ struct mlx5_lag *ldev;
+
+ ldev = kzalloc(sizeof(*ldev), GFP_KERNEL);
+ if (!ldev)
+ return NULL;
+
+ INIT_DELAYED_WORK(&ldev->bond_work, mlx5_do_bond_work);
+
+ return ldev;
+}
+
+static void mlx5_lag_dev_free(struct mlx5_lag *ldev)
+{
+ kfree(ldev);
+}
+
+static void mlx5_lag_dev_add_pf(struct mlx5_lag *ldev,
+ struct mlx5_core_dev *dev,
+ struct net_device *netdev)
+{
+ unsigned int fn = PCI_FUNC(dev->pdev->devfn);
+
+ if (fn >= MLX5_MAX_PORTS)
+ return;
+
+ mutex_lock(&lag_mutex);
+ ldev->pf[fn].dev = dev;
+ ldev->pf[fn].netdev = netdev;
+ ldev->tracker.netdev_state[fn].link_up = 0;
+ ldev->tracker.netdev_state[fn].tx_enabled = 0;
+
+ dev->priv.lag = ldev;
+ mutex_unlock(&lag_mutex);
+}
+
+static void mlx5_lag_dev_remove_pf(struct mlx5_lag *ldev,
+ struct mlx5_core_dev *dev)
+{
+ int i;
+
+ for (i = 0; i < MLX5_MAX_PORTS; i++)
+ if (ldev->pf[i].dev == dev)
+ break;
+
+ if (i == MLX5_MAX_PORTS)
+ return;
+
+ mutex_lock(&lag_mutex);
+ memset(&ldev->pf[i], 0, sizeof(*ldev->pf));
+
+ dev->priv.lag = NULL;
+ mutex_unlock(&lag_mutex);
+}
+
+static u16 mlx5_gen_pci_id(struct mlx5_core_dev *dev)
+{
+ return (u16)((dev->pdev->bus->number << 8) |
+ PCI_SLOT(dev->pdev->devfn));
+}
+
+/* Must be called with intf_mutex held */
+void mlx5_lag_add(struct mlx5_core_dev *dev, struct net_device *netdev)
+{
+ struct mlx5_lag *ldev = NULL;
+ struct mlx5_core_dev *tmp_dev;
+ struct mlx5_priv *priv;
+ u16 pci_id;
+
+ if (!MLX5_CAP_GEN(dev, vport_group_manager) ||
+ !MLX5_CAP_GEN(dev, lag_master) ||
+ (MLX5_CAP_GEN(dev, num_lag_ports) != MLX5_MAX_PORTS))
+ return;
+
+ pci_id = mlx5_gen_pci_id(dev);
+
+ mlx5_core_for_each_priv(priv) {
+ tmp_dev = container_of(priv, struct mlx5_core_dev, priv);
+ if ((dev != tmp_dev) &&
+ (mlx5_gen_pci_id(tmp_dev) == pci_id)) {
+ ldev = tmp_dev->priv.lag;
+ break;
+ }
+ }
+
+ if (!ldev) {
+ ldev = mlx5_lag_dev_alloc();
+ if (!ldev) {
+ mlx5_core_err(dev, "Failed to alloc lag dev\n");
+ return;
+ }
+ }
+
+ mlx5_lag_dev_add_pf(ldev, dev, netdev);
+
+ if (!ldev->nb.notifier_call) {
+ ldev->nb.notifier_call = mlx5_lag_netdev_event;
+ if (register_netdevice_notifier(&ldev->nb)) {
+ ldev->nb.notifier_call = NULL;
+ mlx5_core_err(dev, "Failed to register LAG netdev notifier\n");
+ }
+ }
+}
+
+/* Must be called with intf_mutex held */
+void mlx5_lag_remove(struct mlx5_core_dev *dev)
+{
+ struct mlx5_lag *ldev;
+ int i;
+
+ ldev = mlx5_lag_dev_get(dev);
+ if (!ldev)
+ return;
+
+ if (mlx5_lag_is_bonded(ldev))
+ mlx5_deactivate_lag(ldev);
+
+ mlx5_lag_dev_remove_pf(ldev, dev);
+
+ for (i = 0; i < MLX5_MAX_PORTS; i++)
+ if (ldev->pf[i].dev)
+ break;
+
+ if (i == MLX5_MAX_PORTS) {
+ if (ldev->nb.notifier_call)
+ unregister_netdevice_notifier(&ldev->nb);
+ cancel_delayed_work_sync(&ldev->bond_work);
+ mlx5_lag_dev_free(ldev);
+ }
+}
+
+bool mlx5_lag_is_active(struct mlx5_core_dev *dev)
+{
+ struct mlx5_lag *ldev;
+ bool res;
+
+ mutex_lock(&lag_mutex);
+ ldev = mlx5_lag_dev_get(dev);
+ res = ldev && mlx5_lag_is_bonded(ldev);
+ mutex_unlock(&lag_mutex);
+
+ return res;
+}
+EXPORT_SYMBOL(mlx5_lag_is_active);
+
+struct net_device *mlx5_lag_get_roce_netdev(struct mlx5_core_dev *dev)
+{
+ struct net_device *ndev = NULL;
+ struct mlx5_lag *ldev;
+
+ mutex_lock(&lag_mutex);
+ ldev = mlx5_lag_dev_get(dev);
+
+ if (!(ldev && mlx5_lag_is_bonded(ldev)))
+ goto unlock;
+
+ if (ldev->tracker.tx_type == NETDEV_LAG_TX_TYPE_ACTIVEBACKUP) {
+ ndev = ldev->tracker.netdev_state[0].tx_enabled ?
+ ldev->pf[0].netdev : ldev->pf[1].netdev;
+ } else {
+ ndev = ldev->pf[0].netdev;
+ }
+ if (ndev)
+ dev_hold(ndev);
+
+unlock:
+ mutex_unlock(&lag_mutex);
+
+ return ndev;
+}
+EXPORT_SYMBOL(mlx5_lag_get_roce_netdev);
+
+bool mlx5_lag_intf_add(struct mlx5_interface *intf, struct mlx5_priv *priv)
+{
+ struct mlx5_core_dev *dev = container_of(priv, struct mlx5_core_dev,
+ priv);
+ struct mlx5_lag *ldev;
+
+ if (intf->protocol != MLX5_INTERFACE_PROTOCOL_IB)
+ return true;
+
+ ldev = mlx5_lag_dev_get(dev);
+ if (!ldev || !mlx5_lag_is_bonded(ldev) || ldev->pf[0].dev == dev)
+ return true;
+
+ /* If bonded, we do not add an IB device for PF1. */
+ return false;
+}
+
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index d6446a3..28e3ce6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -73,8 +73,9 @@
MODULE_PARM_DESC(prof_sel, "profile selector. Valid range 0 - 2");
static LIST_HEAD(intf_list);
-static LIST_HEAD(dev_list);
-static DEFINE_MUTEX(intf_mutex);
+
+LIST_HEAD(mlx5_dev_list);
+DEFINE_MUTEX(mlx5_intf_mutex);
struct mlx5_device_context {
struct list_head list;
@@ -782,6 +783,9 @@
struct mlx5_device_context *dev_ctx;
struct mlx5_core_dev *dev = container_of(priv, struct mlx5_core_dev, priv);
+ if (!mlx5_lag_intf_add(intf, priv))
+ return;
+
dev_ctx = kmalloc(sizeof(*dev_ctx), GFP_KERNEL);
if (!dev_ctx)
return;
@@ -820,11 +824,11 @@
struct mlx5_priv *priv = &dev->priv;
struct mlx5_interface *intf;
- mutex_lock(&intf_mutex);
- list_add_tail(&priv->dev_list, &dev_list);
+ mutex_lock(&mlx5_intf_mutex);
+ list_add_tail(&priv->dev_list, &mlx5_dev_list);
list_for_each_entry(intf, &intf_list, list)
mlx5_add_device(intf, priv);
- mutex_unlock(&intf_mutex);
+ mutex_unlock(&mlx5_intf_mutex);
return 0;
}
@@ -834,11 +838,11 @@
struct mlx5_priv *priv = &dev->priv;
struct mlx5_interface *intf;
- mutex_lock(&intf_mutex);
+ mutex_lock(&mlx5_intf_mutex);
list_for_each_entry(intf, &intf_list, list)
mlx5_remove_device(intf, priv);
list_del(&priv->dev_list);
- mutex_unlock(&intf_mutex);
+ mutex_unlock(&mlx5_intf_mutex);
}
int mlx5_register_interface(struct mlx5_interface *intf)
@@ -848,11 +852,11 @@
if (!intf->add || !intf->remove)
return -EINVAL;
- mutex_lock(&intf_mutex);
+ mutex_lock(&mlx5_intf_mutex);
list_add_tail(&intf->list, &intf_list);
- list_for_each_entry(priv, &dev_list, dev_list)
+ list_for_each_entry(priv, &mlx5_dev_list, dev_list)
mlx5_add_device(intf, priv);
- mutex_unlock(&intf_mutex);
+ mutex_unlock(&mlx5_intf_mutex);
return 0;
}
@@ -862,11 +866,11 @@
{
struct mlx5_priv *priv;
- mutex_lock(&intf_mutex);
- list_for_each_entry(priv, &dev_list, dev_list)
+ mutex_lock(&mlx5_intf_mutex);
+ list_for_each_entry(priv, &mlx5_dev_list, dev_list)
mlx5_remove_device(intf, priv);
list_del(&intf->list);
- mutex_unlock(&intf_mutex);
+ mutex_unlock(&mlx5_intf_mutex);
}
EXPORT_SYMBOL(mlx5_unregister_interface);
@@ -892,6 +896,30 @@
}
EXPORT_SYMBOL(mlx5_get_protocol_dev);
+/* Must be called with intf_mutex held */
+void mlx5_add_dev_by_protocol(struct mlx5_core_dev *dev, int protocol)
+{
+ struct mlx5_interface *intf;
+
+ list_for_each_entry(intf, &intf_list, list)
+ if (intf->protocol == protocol) {
+ mlx5_add_device(intf, &dev->priv);
+ break;
+ }
+}
+
+/* Must be called with intf_mutex held */
+void mlx5_remove_dev_by_protocol(struct mlx5_core_dev *dev, int protocol)
+{
+ struct mlx5_interface *intf;
+
+ list_for_each_entry(intf, &intf_list, list)
+ if (intf->protocol == protocol) {
+ mlx5_remove_device(intf, &dev->priv);
+ break;
+ }
+}
+
static int mlx5_pci_init(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
{
struct pci_dev *pdev = dev->pdev;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index ef0eeed..714b71b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -46,6 +46,9 @@
extern int mlx5_core_debug_mask;
+extern struct list_head mlx5_dev_list;
+extern struct mutex mlx5_intf_mutex;
+
#define mlx5_core_dbg(__dev, format, ...) \
dev_dbg(&(__dev)->pdev->dev, "%s:%s:%d:(pid %d): " format, \
(__dev)->priv.name, __func__, __LINE__, current->pid, \
@@ -70,6 +73,9 @@
#define mlx5_core_info(__dev, format, ...) \
dev_info(&(__dev)->pdev->dev, format, ##__VA_ARGS__)
+#define mlx5_core_for_each_priv(__priv) \
+ list_for_each_entry(__priv, &mlx5_dev_list, dev_list)
+
enum {
MLX5_CMD_DATA, /* print command payload only */
MLX5_CMD_TIME, /* print command execution time */
@@ -84,6 +90,7 @@
void mlx5_enter_error_state(struct mlx5_core_dev *dev);
void mlx5_disable_device(struct mlx5_core_dev *dev);
int mlx5_core_sriov_configure(struct pci_dev *dev, int num_vfs);
+bool mlx5_sriov_is_enabled(struct mlx5_core_dev *dev);
int mlx5_core_enable_hca(struct mlx5_core_dev *dev, u16 func_id);
int mlx5_core_disable_hca(struct mlx5_core_dev *dev, u16 func_id);
int mlx5_wait_for_vf_pages(struct mlx5_core_dev *dev);
@@ -92,7 +99,27 @@
struct mlx5_eq *mlx5_eqn2eq(struct mlx5_core_dev *dev, int eqn);
void mlx5_cq_tasklet_cb(unsigned long data);
+void mlx5_lag_add(struct mlx5_core_dev *dev, struct net_device *netdev);
+void mlx5_lag_remove(struct mlx5_core_dev *dev);
+
+void mlx5_add_dev_by_protocol(struct mlx5_core_dev *dev, int protocol);
+void mlx5_remove_dev_by_protocol(struct mlx5_core_dev *dev, int protocol);
+
+bool mlx5_lag_intf_add(struct mlx5_interface *intf, struct mlx5_priv *priv);
+
void mlx5e_init(void);
void mlx5e_cleanup(void);
+static inline int mlx5_lag_is_lacp_owner(struct mlx5_core_dev *dev)
+{
+ /* LACP owner conditions:
+ * 1) Function is physical.
+ * 2) LAG is supported by FW.
+ * 3) LAG is managed by driver (currently the only option).
+ */
+ return MLX5_CAP_GEN(dev, vport_group_manager) &&
+ (MLX5_CAP_GEN(dev, num_lag_ports) > 1) &&
+ MLX5_CAP_GEN(dev, lag_master);
+}
+
#endif /* __MLX5_CORE_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c
index 2ee28ad..34e7184 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/port.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c
@@ -175,25 +175,39 @@
}
EXPORT_SYMBOL_GPL(mlx5_query_port_link_width_oper);
-int mlx5_query_port_proto_oper(struct mlx5_core_dev *dev,
- u8 *proto_oper, int proto_mask,
- u8 local_port)
+int mlx5_query_port_eth_proto_oper(struct mlx5_core_dev *dev,
+ u32 *proto_oper, u8 local_port)
{
u32 out[MLX5_ST_SZ_DW(ptys_reg)];
int err;
- err = mlx5_query_port_ptys(dev, out, sizeof(out), proto_mask, local_port);
+ err = mlx5_query_port_ptys(dev, out, sizeof(out), MLX5_PTYS_EN,
+ local_port);
if (err)
return err;
- if (proto_mask == MLX5_PTYS_EN)
- *proto_oper = MLX5_GET(ptys_reg, out, eth_proto_oper);
- else
- *proto_oper = MLX5_GET(ptys_reg, out, ib_proto_oper);
+ *proto_oper = MLX5_GET(ptys_reg, out, eth_proto_oper);
return 0;
}
-EXPORT_SYMBOL_GPL(mlx5_query_port_proto_oper);
+EXPORT_SYMBOL(mlx5_query_port_eth_proto_oper);
+
+int mlx5_query_port_ib_proto_oper(struct mlx5_core_dev *dev,
+ u8 *proto_oper, u8 local_port)
+{
+ u32 out[MLX5_ST_SZ_DW(ptys_reg)];
+ int err;
+
+ err = mlx5_query_port_ptys(dev, out, sizeof(out), MLX5_PTYS_IB,
+ local_port);
+ if (err)
+ return err;
+
+ *proto_oper = MLX5_GET(ptys_reg, out, ib_proto_oper);
+
+ return 0;
+}
+EXPORT_SYMBOL(mlx5_query_port_ib_proto_oper);
int mlx5_set_port_ptys(struct mlx5_core_dev *dev, bool an_disable,
u32 proto_admin, int proto_mask)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sriov.c b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c
index 0680dfb..78e7892 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sriov.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sriov.c
@@ -37,6 +37,13 @@
#include "eswitch.h"
#endif
+bool mlx5_sriov_is_enabled(struct mlx5_core_dev *dev)
+{
+ struct mlx5_core_sriov *sriov = &dev->priv.sriov;
+
+ return !!sriov->num_vfs;
+}
+
static void enable_vfs(struct mlx5_core_dev *dev, int num_vfs)
{
struct mlx5_core_sriov *sriov = &dev->priv.sriov;
@@ -144,6 +151,11 @@
if (!mlx5_core_is_pf(dev))
return -EPERM;
+ if (num_vfs && mlx5_lag_is_active(dev)) {
+ mlx5_core_warn(dev, "can't turn sriov on while LAG is active");
+ return -EINVAL;
+ }
+
mlx5_core_cleanup_vfs(dev);
if (!num_vfs) {
diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h
index 1721098..b83d0a7 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/reg.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h
@@ -591,6 +591,12 @@
*/
MLXSW_ITEM32(reg, sfn, swid, 0x00, 24, 8);
+/* reg_sfn_end
+ * Forces the current session to end.
+ * Access: OP
+ */
+MLXSW_ITEM32(reg, sfn, end, 0x04, 20, 1);
+
/* reg_sfn_num_rec
* Request: Number of learned notifications and aged-out notification
* records requested.
@@ -605,6 +611,7 @@
{
MLXSW_REG_ZERO(sfn, payload);
mlxsw_reg_sfn_swid_set(payload, 0);
+ mlxsw_reg_sfn_end_set(payload, 1);
mlxsw_reg_sfn_num_rec_set(payload, MLXSW_REG_SFN_REC_MAX_COUNT);
}
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 1f81689..2713a64 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -555,8 +555,9 @@
return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(svfa), svfa_pl);
}
-static int mlxsw_sp_port_vid_learning_set(struct mlxsw_sp_port *mlxsw_sp_port,
- u16 vid, bool learn_enable)
+int __mlxsw_sp_port_vid_learning_set(struct mlxsw_sp_port *mlxsw_sp_port,
+ u16 vid_begin, u16 vid_end,
+ bool learn_enable)
{
struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
char *spvmlr_pl;
@@ -565,13 +566,20 @@
spvmlr_pl = kmalloc(MLXSW_REG_SPVMLR_LEN, GFP_KERNEL);
if (!spvmlr_pl)
return -ENOMEM;
- mlxsw_reg_spvmlr_pack(spvmlr_pl, mlxsw_sp_port->local_port, vid, vid,
- learn_enable);
+ mlxsw_reg_spvmlr_pack(spvmlr_pl, mlxsw_sp_port->local_port, vid_begin,
+ vid_end, learn_enable);
err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(spvmlr), spvmlr_pl);
kfree(spvmlr_pl);
return err;
}
+static int mlxsw_sp_port_vid_learning_set(struct mlxsw_sp_port *mlxsw_sp_port,
+ u16 vid, bool learn_enable)
+{
+ return __mlxsw_sp_port_vid_learning_set(mlxsw_sp_port, vid, vid,
+ learn_enable);
+}
+
static int
mlxsw_sp_port_system_port_mapping_set(struct mlxsw_sp_port *mlxsw_sp_port)
{
@@ -973,10 +981,6 @@
goto err_port_vp_mode_trans;
}
- err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, false);
- if (err)
- goto err_port_vid_learning_set;
-
err = mlxsw_sp_port_vlan_set(mlxsw_sp_vport, vid, vid, true, untagged);
if (err)
goto err_port_add_vid;
@@ -984,8 +988,6 @@
return 0;
err_port_add_vid:
- mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, true);
-err_port_vid_learning_set:
if (list_is_singular(&mlxsw_sp_port->vports_list))
mlxsw_sp_port_vlan_mode_trans(mlxsw_sp_port);
err_port_vp_mode_trans:
@@ -1012,8 +1014,6 @@
mlxsw_sp_port_vlan_set(mlxsw_sp_vport, vid, vid, false, false);
- mlxsw_sp_port_vid_learning_set(mlxsw_sp_vport, vid, true);
-
/* Drop FID reference. If this was the last reference the
* resources will be freed.
*/
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index ab3feb8..01537d3 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -558,6 +558,9 @@
int mlxsw_sp_port_ets_maxrate_set(struct mlxsw_sp_port *mlxsw_sp_port,
enum mlxsw_reg_qeec_hr hr, u8 index,
u8 next_index, u32 maxrate);
+int __mlxsw_sp_port_vid_learning_set(struct mlxsw_sp_port *mlxsw_sp_port,
+ u16 vid_begin, u16 vid_end,
+ bool learn_enable);
#ifdef CONFIG_MLXSW_SPECTRUM_DCB
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
index d1b59cd..0c3fbbc 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
@@ -261,12 +261,40 @@
false);
}
+static int mlxsw_sp_port_learning_set(struct mlxsw_sp_port *mlxsw_sp_port,
+ bool set)
+{
+ u16 vid;
+ int err;
+
+ if (mlxsw_sp_port_is_vport(mlxsw_sp_port)) {
+ vid = mlxsw_sp_vport_vid_get(mlxsw_sp_port);
+
+ return __mlxsw_sp_port_vid_learning_set(mlxsw_sp_port, vid, vid,
+ set);
+ }
+
+ for_each_set_bit(vid, mlxsw_sp_port->active_vlans, VLAN_N_VID) {
+ err = __mlxsw_sp_port_vid_learning_set(mlxsw_sp_port, vid, vid,
+ set);
+ if (err)
+ goto err_port_vid_learning_set;
+ }
+
+ return 0;
+
+err_port_vid_learning_set:
+ for_each_set_bit(vid, mlxsw_sp_port->active_vlans, VLAN_N_VID)
+ __mlxsw_sp_port_vid_learning_set(mlxsw_sp_port, vid, vid, !set);
+ return err;
+}
+
static int mlxsw_sp_port_attr_br_flags_set(struct mlxsw_sp_port *mlxsw_sp_port,
struct switchdev_trans *trans,
unsigned long brport_flags)
{
+ unsigned long learning = mlxsw_sp_port->learning ? BR_LEARNING : 0;
unsigned long uc_flood = mlxsw_sp_port->uc_flood ? BR_FLOOD : 0;
- bool set;
int err;
if (!mlxsw_sp_port->bridged)
@@ -276,17 +304,30 @@
return 0;
if ((uc_flood ^ brport_flags) & BR_FLOOD) {
- set = mlxsw_sp_port->uc_flood ? false : true;
- err = mlxsw_sp_port_uc_flood_set(mlxsw_sp_port, set);
+ err = mlxsw_sp_port_uc_flood_set(mlxsw_sp_port,
+ !mlxsw_sp_port->uc_flood);
if (err)
return err;
}
+ if ((learning ^ brport_flags) & BR_LEARNING) {
+ err = mlxsw_sp_port_learning_set(mlxsw_sp_port,
+ !mlxsw_sp_port->learning);
+ if (err)
+ goto err_port_learning_set;
+ }
+
mlxsw_sp_port->uc_flood = brport_flags & BR_FLOOD ? 1 : 0;
mlxsw_sp_port->learning = brport_flags & BR_LEARNING ? 1 : 0;
mlxsw_sp_port->learning_sync = brport_flags & BR_LEARNING_SYNC ? 1 : 0;
return 0;
+
+err_port_learning_set:
+ if ((uc_flood ^ brport_flags) & BR_FLOOD)
+ mlxsw_sp_port_uc_flood_set(mlxsw_sp_port,
+ mlxsw_sp_port->uc_flood);
+ return err;
}
static int mlxsw_sp_ageing_set(struct mlxsw_sp *mlxsw_sp, u32 ageing_time)
@@ -635,6 +676,27 @@
return 0;
}
+static int mlxsw_sp_port_vid_learning_set(struct mlxsw_sp_port *mlxsw_sp_port,
+ u16 vid_begin, u16 vid_end,
+ bool learn_enable)
+{
+ u16 vid, vid_e;
+ int err;
+
+ for (vid = vid_begin; vid <= vid_end;
+ vid += MLXSW_REG_SPVMLR_REC_MAX_COUNT) {
+ vid_e = min((u16) (vid + MLXSW_REG_SPVMLR_REC_MAX_COUNT - 1),
+ vid_end);
+
+ err = __mlxsw_sp_port_vid_learning_set(mlxsw_sp_port, vid,
+ vid_e, learn_enable);
+ if (err)
+ return err;
+ }
+
+ return 0;
+}
+
static int __mlxsw_sp_port_vlans_add(struct mlxsw_sp_port *mlxsw_sp_port,
u16 vid_begin, u16 vid_end,
bool flag_untagged, bool flag_pvid)
@@ -675,6 +737,14 @@
}
}
+ err = mlxsw_sp_port_vid_learning_set(mlxsw_sp_port, vid_begin, vid_end,
+ mlxsw_sp_port->learning);
+ if (err) {
+ netdev_err(dev, "Failed to set learning for VIDs %d-%d\n",
+ vid_begin, vid_end);
+ goto err_port_vid_learning_set;
+ }
+
/* Changing activity bits only if HW operation succeded */
for (vid = vid_begin; vid <= vid_end; vid++) {
set_bit(vid, mlxsw_sp_port->active_vlans);
@@ -697,6 +767,9 @@
err_port_stp_state_set:
for (vid = vid_begin; vid <= vid_end; vid++)
clear_bit(vid, mlxsw_sp_port->active_vlans);
+ mlxsw_sp_port_vid_learning_set(mlxsw_sp_port, vid_begin, vid_end,
+ false);
+err_port_vid_learning_set:
if (old_pvid != mlxsw_sp_port->pvid)
mlxsw_sp_port_pvid_set(mlxsw_sp_port, old_pvid);
err_port_pvid_set:
@@ -1001,29 +1074,20 @@
static int __mlxsw_sp_port_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port,
u16 vid_begin, u16 vid_end)
{
- struct net_device *dev = mlxsw_sp_port->dev;
u16 vid, pvid;
- int err;
if (!mlxsw_sp_port->bridged)
return -EINVAL;
- err = __mlxsw_sp_port_vlans_set(mlxsw_sp_port, vid_begin, vid_end,
- false, false);
- if (err) {
- netdev_err(dev, "Unable to del VIDs %d-%d\n", vid_begin,
- vid_end);
- return err;
- }
+ mlxsw_sp_port_vid_learning_set(mlxsw_sp_port, vid_begin, vid_end,
+ false);
pvid = mlxsw_sp_port->pvid;
- if (pvid >= vid_begin && pvid <= vid_end) {
- err = mlxsw_sp_port_pvid_set(mlxsw_sp_port, 0);
- if (err) {
- netdev_err(dev, "Unable to del PVID %d\n", pvid);
- return err;
- }
- }
+ if (pvid >= vid_begin && pvid <= vid_end)
+ mlxsw_sp_port_pvid_set(mlxsw_sp_port, 0);
+
+ __mlxsw_sp_port_vlans_set(mlxsw_sp_port, vid_begin, vid_end, false,
+ false);
mlxsw_sp_port_fid_leave(mlxsw_sp_port, vid_begin, vid_end);
@@ -1366,8 +1430,6 @@
vid = fid;
}
- adding = adding && mlxsw_sp_port->learning;
-
do_fdb_op:
err = mlxsw_sp_port_fdb_uc_op(mlxsw_sp, local_port, mac, fid,
adding, true);
@@ -1429,8 +1491,6 @@
vid = fid;
}
- adding = adding && mlxsw_sp_port->learning;
-
do_fdb_op:
err = mlxsw_sp_port_fdb_uc_lag_op(mlxsw_sp, lag_id, mac, fid, lag_vid,
adding, true);
@@ -1496,20 +1556,18 @@
mlxsw_sp = container_of(work, struct mlxsw_sp, fdb_notify.dw.work);
rtnl_lock();
- do {
- mlxsw_reg_sfn_pack(sfn_pl);
- err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(sfn), sfn_pl);
- if (err) {
- dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Failed to get FDB notifications\n");
- break;
- }
- num_rec = mlxsw_reg_sfn_num_rec_get(sfn_pl);
- for (i = 0; i < num_rec; i++)
- mlxsw_sp_fdb_notify_rec_process(mlxsw_sp, sfn_pl, i);
+ mlxsw_reg_sfn_pack(sfn_pl);
+ err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(sfn), sfn_pl);
+ if (err) {
+ dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Failed to get FDB notifications\n");
+ goto out;
+ }
+ num_rec = mlxsw_reg_sfn_num_rec_get(sfn_pl);
+ for (i = 0; i < num_rec; i++)
+ mlxsw_sp_fdb_notify_rec_process(mlxsw_sp, sfn_pl, i);
- } while (num_rec);
+out:
rtnl_unlock();
-
kfree(sfn_pl);
mlxsw_sp_fdb_notify_work_schedule(mlxsw_sp);
}
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index 2575070..77c1417 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -964,6 +964,18 @@
#define MLX5_CAP_FLOWTABLE_NIC_RX_MAX(mdev, cap) \
MLX5_CAP_FLOWTABLE_MAX(mdev, flow_table_properties_nic_receive.cap)
+#define MLX5_CAP_FLOWTABLE_SNIFFER_RX(mdev, cap) \
+ MLX5_CAP_FLOWTABLE(mdev, flow_table_properties_nic_receive_sniffer.cap)
+
+#define MLX5_CAP_FLOWTABLE_SNIFFER_RX_MAX(mdev, cap) \
+ MLX5_CAP_FLOWTABLE_MAX(mdev, flow_table_properties_nic_receive_sniffer.cap)
+
+#define MLX5_CAP_FLOWTABLE_SNIFFER_TX(mdev, cap) \
+ MLX5_CAP_FLOWTABLE(mdev, flow_table_properties_nic_transmit_sniffer.cap)
+
+#define MLX5_CAP_FLOWTABLE_SNIFFER_TX_MAX(mdev, cap) \
+ MLX5_CAP_FLOWTABLE_MAX(mdev, flow_table_properties_nic_transmit_sniffer.cap)
+
#define MLX5_CAP_ESW_FLOWTABLE(mdev, cap) \
MLX5_GET(flow_table_eswitch_cap, \
mdev->hca_caps_cur[MLX5_CAP_ESWITCH_FLOW_TABLE], cap)
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index ebe57ab..5cb9fa7 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -49,10 +49,6 @@
#include <linux/mlx5/srq.h>
enum {
- MLX5_RQ_BITMASK_VSD = 1 << 1,
-};
-
-enum {
MLX5_BOARD_ID_LEN = 64,
MLX5_MAX_NAME_LEN = 16,
};
@@ -481,6 +477,7 @@
};
struct mlx5_eswitch;
+struct mlx5_lag;
struct mlx5_rl_entry {
u32 rate;
@@ -554,6 +551,7 @@
struct mlx5_flow_steering *steering;
struct mlx5_eswitch *eswitch;
struct mlx5_core_sriov sriov;
+ struct mlx5_lag *lag;
unsigned long pci_dev_data;
struct mlx5_fc_stats fc_stats;
struct mlx5_rl_table rl_table;
@@ -946,6 +944,11 @@
void mlx5_unregister_interface(struct mlx5_interface *intf);
int mlx5_core_query_vendor_id(struct mlx5_core_dev *mdev, u32 *vendor_id);
+int mlx5_cmd_create_vport_lag(struct mlx5_core_dev *dev);
+int mlx5_cmd_destroy_vport_lag(struct mlx5_core_dev *dev);
+bool mlx5_lag_is_active(struct mlx5_core_dev *dev);
+struct net_device *mlx5_lag_get_roce_netdev(struct mlx5_core_dev *dev);
+
struct mlx5_profile {
u64 mask;
u8 log_max_qp;
diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h
index e036d60..93ebc5e 100644
--- a/include/linux/mlx5/fs.h
+++ b/include/linux/mlx5/fs.h
@@ -54,6 +54,7 @@
enum mlx5_flow_namespace_type {
MLX5_FLOW_NAMESPACE_BYPASS,
+ MLX5_FLOW_NAMESPACE_LAG,
MLX5_FLOW_NAMESPACE_OFFLOADS,
MLX5_FLOW_NAMESPACE_ETHTOOL,
MLX5_FLOW_NAMESPACE_KERNEL,
@@ -62,6 +63,8 @@
MLX5_FLOW_NAMESPACE_FDB,
MLX5_FLOW_NAMESPACE_ESW_EGRESS,
MLX5_FLOW_NAMESPACE_ESW_INGRESS,
+ MLX5_FLOW_NAMESPACE_SNIFFER_RX,
+ MLX5_FLOW_NAMESPACE_SNIFFER_TX,
};
struct mlx5_flow_table;
@@ -106,6 +109,9 @@
int prio,
int num_flow_table_entries,
u32 level, u16 vport);
+struct mlx5_flow_table *mlx5_create_lag_demux_flow_table(
+ struct mlx5_flow_namespace *ns,
+ int prio, u32 level);
int mlx5_destroy_flow_table(struct mlx5_flow_table *ft);
/* inbox should be set with the following values:
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 3766110..73a720f 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -174,6 +174,12 @@
MLX5_CMD_OP_DELETE_L2_TABLE_ENTRY = 0x82b,
MLX5_CMD_OP_SET_WOL_ROL = 0x830,
MLX5_CMD_OP_QUERY_WOL_ROL = 0x831,
+ MLX5_CMD_OP_CREATE_LAG = 0x840,
+ MLX5_CMD_OP_MODIFY_LAG = 0x841,
+ MLX5_CMD_OP_QUERY_LAG = 0x842,
+ MLX5_CMD_OP_DESTROY_LAG = 0x843,
+ MLX5_CMD_OP_CREATE_VPORT_LAG = 0x844,
+ MLX5_CMD_OP_DESTROY_VPORT_LAG = 0x845,
MLX5_CMD_OP_CREATE_TIR = 0x900,
MLX5_CMD_OP_MODIFY_TIR = 0x901,
MLX5_CMD_OP_DESTROY_TIR = 0x902,
@@ -477,7 +483,9 @@
struct mlx5_ifc_flow_table_nic_cap_bits {
u8 nic_rx_multi_path_tirs[0x1];
- u8 reserved_at_1[0x1ff];
+ u8 nic_rx_multi_path_tirs_fts[0x1];
+ u8 allow_sniffer_and_nic_rx_shared_tir[0x1];
+ u8 reserved_at_3[0x1fd];
struct mlx5_ifc_flow_table_prop_layout_bits flow_table_properties_nic_receive;
@@ -779,7 +787,9 @@
u8 out_of_seq_cnt[0x1];
u8 vport_counters[0x1];
u8 retransmission_q_counters[0x1];
- u8 reserved_at_183[0x3];
+ u8 reserved_at_183[0x1];
+ u8 modify_rq_counter_set_id[0x1];
+ u8 reserved_at_185[0x1];
u8 max_qp_cnt[0xa];
u8 pkey_table_size[0x10];
@@ -882,7 +892,10 @@
u8 pad_tx_eth_packet[0x1];
u8 reserved_at_263[0x8];
u8 log_bf_reg_size[0x5];
- u8 reserved_at_270[0x10];
+
+ u8 reserved_at_270[0xb];
+ u8 lag_master[0x1];
+ u8 num_lag_ports[0x4];
u8 reserved_at_280[0x10];
u8 max_wqe_sz_sq[0x10];
@@ -1916,7 +1929,7 @@
struct mlx5_ifc_qpc_bits {
u8 state[0x4];
- u8 reserved_at_4[0x4];
+ u8 lag_tx_port_affinity[0x4];
u8 st[0x8];
u8 reserved_at_10[0x3];
u8 pm_state[0x2];
@@ -2165,7 +2178,11 @@
};
struct mlx5_ifc_tisc_bits {
- u8 reserved_at_0[0xc];
+ u8 strict_lag_tx_port_affinity[0x1];
+ u8 reserved_at_1[0x3];
+ u8 lag_tx_port_affinity[0x04];
+
+ u8 reserved_at_8[0x4];
u8 prio[0x4];
u8 reserved_at_10[0x10];
@@ -2827,7 +2844,7 @@
struct mlx5_ifc_tag_matching_topology_context_bits tag_matching_topology_context;
- u8 reserved_at_180[0x180];
+ u8 reserved_at_180[0x200];
struct mlx5_ifc_wq_bits wq;
};
@@ -4615,7 +4632,9 @@
struct mlx5_ifc_modify_tis_bitmask_bits {
u8 reserved_at_0[0x20];
- u8 reserved_at_20[0x1f];
+ u8 reserved_at_20[0x1d];
+ u8 lag_tx_port_affinity[0x1];
+ u8 strict_lag_tx_port_affinity[0x1];
u8 prio[0x1];
};
@@ -4750,6 +4769,11 @@
u8 reserved_at_40[0x40];
};
+enum {
+ MLX5_MODIFY_RQ_IN_MODIFY_BITMASK_VSD = 1ULL << 1,
+ MLX5_MODIFY_RQ_IN_MODIFY_BITMASK_MODIFY_RQ_COUNTER_SET_ID = 1ULL << 3,
+};
+
struct mlx5_ifc_modify_rq_in_bits {
u8 opcode[0x10];
u8 reserved_at_10[0x10];
@@ -6208,7 +6232,10 @@
u8 reserved_at_e0[0x8];
u8 table_miss_id[0x18];
- u8 reserved_at_100[0x100];
+ u8 reserved_at_100[0x8];
+ u8 lag_master_next_table_id[0x18];
+
+ u8 reserved_at_120[0x80];
};
struct mlx5_ifc_create_flow_group_out_bits {
@@ -7662,7 +7689,8 @@
};
enum {
- MLX5_MODIFY_FLOW_TABLE_MISS_TABLE_ID = 0x1,
+ MLX5_MODIFY_FLOW_TABLE_MISS_TABLE_ID = (1UL << 0),
+ MLX5_MODIFY_FLOW_TABLE_LAG_NEXT_TABLE_ID = (1UL << 15),
};
struct mlx5_ifc_modify_flow_table_out_bits {
@@ -7701,7 +7729,10 @@
u8 reserved_at_e0[0x8];
u8 table_miss_id[0x18];
- u8 reserved_at_100[0x100];
+ u8 reserved_at_100[0x8];
+ u8 lag_master_next_table_id[0x18];
+
+ u8 reserved_at_120[0x80];
};
struct mlx5_ifc_ets_tcn_config_reg_bits {
@@ -7809,4 +7840,134 @@
u8 error[0x8];
u8 reserved_at_a0[0x160];
};
+
+struct mlx5_ifc_lagc_bits {
+ u8 reserved_at_0[0x1d];
+ u8 lag_state[0x3];
+
+ u8 reserved_at_20[0x14];
+ u8 tx_remap_affinity_2[0x4];
+ u8 reserved_at_38[0x4];
+ u8 tx_remap_affinity_1[0x4];
+};
+
+struct mlx5_ifc_create_lag_out_bits {
+ u8 status[0x8];
+ u8 reserved_at_8[0x18];
+
+ u8 syndrome[0x20];
+
+ u8 reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_create_lag_in_bits {
+ u8 opcode[0x10];
+ u8 reserved_at_10[0x10];
+
+ u8 reserved_at_20[0x10];
+ u8 op_mod[0x10];
+
+ struct mlx5_ifc_lagc_bits ctx;
+};
+
+struct mlx5_ifc_modify_lag_out_bits {
+ u8 status[0x8];
+ u8 reserved_at_8[0x18];
+
+ u8 syndrome[0x20];
+
+ u8 reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_modify_lag_in_bits {
+ u8 opcode[0x10];
+ u8 reserved_at_10[0x10];
+
+ u8 reserved_at_20[0x10];
+ u8 op_mod[0x10];
+
+ u8 reserved_at_40[0x20];
+ u8 field_select[0x20];
+
+ struct mlx5_ifc_lagc_bits ctx;
+};
+
+struct mlx5_ifc_query_lag_out_bits {
+ u8 status[0x8];
+ u8 reserved_at_8[0x18];
+
+ u8 syndrome[0x20];
+
+ u8 reserved_at_40[0x40];
+
+ struct mlx5_ifc_lagc_bits ctx;
+};
+
+struct mlx5_ifc_query_lag_in_bits {
+ u8 opcode[0x10];
+ u8 reserved_at_10[0x10];
+
+ u8 reserved_at_20[0x10];
+ u8 op_mod[0x10];
+
+ u8 reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_destroy_lag_out_bits {
+ u8 status[0x8];
+ u8 reserved_at_8[0x18];
+
+ u8 syndrome[0x20];
+
+ u8 reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_destroy_lag_in_bits {
+ u8 opcode[0x10];
+ u8 reserved_at_10[0x10];
+
+ u8 reserved_at_20[0x10];
+ u8 op_mod[0x10];
+
+ u8 reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_create_vport_lag_out_bits {
+ u8 status[0x8];
+ u8 reserved_at_8[0x18];
+
+ u8 syndrome[0x20];
+
+ u8 reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_create_vport_lag_in_bits {
+ u8 opcode[0x10];
+ u8 reserved_at_10[0x10];
+
+ u8 reserved_at_20[0x10];
+ u8 op_mod[0x10];
+
+ u8 reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_destroy_vport_lag_out_bits {
+ u8 status[0x8];
+ u8 reserved_at_8[0x18];
+
+ u8 syndrome[0x20];
+
+ u8 reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_destroy_vport_lag_in_bits {
+ u8 opcode[0x10];
+ u8 reserved_at_10[0x10];
+
+ u8 reserved_at_20[0x10];
+ u8 op_mod[0x10];
+
+ u8 reserved_at_40[0x40];
+};
+
#endif /* MLX5_IFC_H */
diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h
index e3012cc..b3065ac 100644
--- a/include/linux/mlx5/port.h
+++ b/include/linux/mlx5/port.h
@@ -61,6 +61,39 @@
#define MLX5_I2C_ADDR_HIGH 0x51
#define MLX5_EEPROM_PAGE_LENGTH 256
+enum mlx5e_link_mode {
+ MLX5E_1000BASE_CX_SGMII = 0,
+ MLX5E_1000BASE_KX = 1,
+ MLX5E_10GBASE_CX4 = 2,
+ MLX5E_10GBASE_KX4 = 3,
+ MLX5E_10GBASE_KR = 4,
+ MLX5E_20GBASE_KR2 = 5,
+ MLX5E_40GBASE_CR4 = 6,
+ MLX5E_40GBASE_KR4 = 7,
+ MLX5E_56GBASE_R4 = 8,
+ MLX5E_10GBASE_CR = 12,
+ MLX5E_10GBASE_SR = 13,
+ MLX5E_10GBASE_ER = 14,
+ MLX5E_40GBASE_SR4 = 15,
+ MLX5E_40GBASE_LR4 = 16,
+ MLX5E_50GBASE_SR2 = 18,
+ MLX5E_100GBASE_CR4 = 20,
+ MLX5E_100GBASE_SR4 = 21,
+ MLX5E_100GBASE_KR4 = 22,
+ MLX5E_100GBASE_LR4 = 23,
+ MLX5E_100BASE_TX = 24,
+ MLX5E_1000BASE_T = 25,
+ MLX5E_10GBASE_T = 26,
+ MLX5E_25GBASE_CR = 27,
+ MLX5E_25GBASE_KR = 28,
+ MLX5E_25GBASE_SR = 29,
+ MLX5E_50GBASE_CR2 = 30,
+ MLX5E_50GBASE_KR2 = 31,
+ MLX5E_LINK_MODES_NUMBER,
+};
+
+#define MLX5E_PROT_MASK(link_mode) (1 << link_mode)
+
int mlx5_set_port_caps(struct mlx5_core_dev *dev, u8 port_num, u32 caps);
int mlx5_query_port_ptys(struct mlx5_core_dev *dev, u32 *ptys,
int ptys_size, int proto_mask, u8 local_port);
@@ -70,9 +103,10 @@
u32 *proto_admin, int proto_mask);
int mlx5_query_port_link_width_oper(struct mlx5_core_dev *dev,
u8 *link_width_oper, u8 local_port);
-int mlx5_query_port_proto_oper(struct mlx5_core_dev *dev,
- u8 *proto_oper, int proto_mask,
- u8 local_port);
+int mlx5_query_port_ib_proto_oper(struct mlx5_core_dev *dev,
+ u8 *proto_oper, u8 local_port);
+int mlx5_query_port_eth_proto_oper(struct mlx5_core_dev *dev,
+ u32 *proto_oper, u8 local_port);
int mlx5_set_port_ptys(struct mlx5_core_dev *dev, bool an_disable,
u32 proto_admin, int proto_mask);
void mlx5_toggle_port_link(struct mlx5_core_dev *dev);
diff --git a/include/net/sock.h b/include/net/sock.h
index 2aab9b6..c797c57 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1020,7 +1020,6 @@
void (*unhash)(struct sock *sk);
void (*rehash)(struct sock *sk);
int (*get_port)(struct sock *sk, unsigned short snum);
- void (*clear_sk)(struct sock *sk, int size);
/* Keeping track of sockets in use */
#ifdef CONFIG_PROC_FS
@@ -1242,8 +1241,6 @@
return sk->sk_prot->hash(sk);
}
-void sk_prot_clear_portaddr_nulls(struct sock *sk, int size);
-
/* About 10 seconds */
#define SOCK_DESTROY_TIME (10*HZ)
diff --git a/net/core/sock.c b/net/core/sock.c
index 25dab8b..51a7304 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1315,24 +1315,6 @@
#endif
}
-void sk_prot_clear_portaddr_nulls(struct sock *sk, int size)
-{
- unsigned long nulls1, nulls2;
-
- nulls1 = offsetof(struct sock, __sk_common.skc_node.next);
- nulls2 = offsetof(struct sock, __sk_common.skc_portaddr_node.next);
- if (nulls1 > nulls2)
- swap(nulls1, nulls2);
-
- if (nulls1 != 0)
- memset((char *)sk, 0, nulls1);
- memset((char *)sk + nulls1 + sizeof(void *), 0,
- nulls2 - nulls1 - sizeof(void *));
- memset((char *)sk + nulls2 + sizeof(void *), 0,
- size - nulls2 - sizeof(void *));
-}
-EXPORT_SYMBOL(sk_prot_clear_portaddr_nulls);
-
static struct sock *sk_prot_alloc(struct proto *prot, gfp_t priority,
int family)
{
@@ -1344,12 +1326,8 @@
sk = kmem_cache_alloc(slab, priority & ~__GFP_ZERO);
if (!sk)
return sk;
- if (priority & __GFP_ZERO) {
- if (prot->clear_sk)
- prot->clear_sk(sk, prot->obj_size);
- else
- sk_prot_clear_nulls(sk, prot->obj_size);
- }
+ if (priority & __GFP_ZERO)
+ sk_prot_clear_nulls(sk, prot->obj_size);
} else
sk = kmalloc(prot->obj_size, priority);
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index e9ffc27b..f0ebb0b 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -2237,7 +2237,6 @@
.compat_setsockopt = compat_udp_setsockopt,
.compat_getsockopt = compat_udp_getsockopt,
#endif
- .clear_sk = sk_prot_clear_portaddr_nulls,
.diag_destroy = udp_abort,
};
EXPORT_SYMBOL(udp_prot);
diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c
index 3b3efbd..67fc9d9 100644
--- a/net/ipv4/udplite.c
+++ b/net/ipv4/udplite.c
@@ -61,7 +61,6 @@
.compat_setsockopt = compat_udp_setsockopt,
.compat_getsockopt = compat_udp_getsockopt,
#endif
- .clear_sk = sk_prot_clear_portaddr_nulls,
};
EXPORT_SYMBOL(udplite_prot);
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 33df8b8..e0f4643 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1862,17 +1862,6 @@
}
#endif
-static void tcp_v6_clear_sk(struct sock *sk, int size)
-{
- struct inet_sock *inet = inet_sk(sk);
-
- /* we do not want to clear pinet6 field, because of RCU lookups */
- sk_prot_clear_nulls(sk, offsetof(struct inet_sock, pinet6));
-
- size -= offsetof(struct inet_sock, pinet6) + sizeof(inet->pinet6);
- memset(&inet->pinet6 + 1, 0, size);
-}
-
struct proto tcpv6_prot = {
.name = "TCPv6",
.owner = THIS_MODULE,
@@ -1914,7 +1903,6 @@
.compat_setsockopt = compat_tcp_setsockopt,
.compat_getsockopt = compat_tcp_getsockopt,
#endif
- .clear_sk = tcp_v6_clear_sk,
.diag_destroy = tcp_abort,
};
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 16512cf..9efe740f 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1424,17 +1424,6 @@
}
#endif /* CONFIG_PROC_FS */
-void udp_v6_clear_sk(struct sock *sk, int size)
-{
- struct inet_sock *inet = inet_sk(sk);
-
- /* we do not want to clear pinet6 field, because of RCU lookups */
- sk_prot_clear_portaddr_nulls(sk, offsetof(struct inet_sock, pinet6));
-
- size -= offsetof(struct inet_sock, pinet6) + sizeof(inet->pinet6);
- memset(&inet->pinet6 + 1, 0, size);
-}
-
/* ------------------------------------------------------------------------ */
struct proto udpv6_prot = {
@@ -1466,7 +1455,6 @@
.compat_setsockopt = compat_udpv6_setsockopt,
.compat_getsockopt = compat_udpv6_getsockopt,
#endif
- .clear_sk = udp_v6_clear_sk,
.diag_destroy = udp_abort,
};
diff --git a/net/ipv6/udp_impl.h b/net/ipv6/udp_impl.h
index 0682c03..f6eb1ab 100644
--- a/net/ipv6/udp_impl.h
+++ b/net/ipv6/udp_impl.h
@@ -29,8 +29,6 @@
int udpv6_queue_rcv_skb(struct sock *sk, struct sk_buff *skb);
void udpv6_destroy_sock(struct sock *sk);
-void udp_v6_clear_sk(struct sock *sk, int size);
-
#ifdef CONFIG_PROC_FS
int udp6_seq_show(struct seq_file *seq, void *v);
#endif
diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c
index 9cf097e..5cf0099 100644
--- a/net/ipv6/udplite.c
+++ b/net/ipv6/udplite.c
@@ -56,7 +56,6 @@
.compat_setsockopt = compat_udpv6_setsockopt,
.compat_getsockopt = compat_udpv6_getsockopt,
#endif
- .clear_sk = udp_v6_clear_sk,
};
static struct inet_protosw udplite6_protosw = {
