* [net 1/8] net/mlx5e: Don't override vport admin link state in switchdev mode
From: Saeed Mahameed @ 2018-03-23 22:05 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Jianbo Liu, Saeed Mahameed
In-Reply-To: <20180323220534.19353-1-saeedm@mellanox.com>
From: Jianbo Liu <jianbol@mellanox.com>
The vport admin original link state will be re-applied after returning
back to legacy mode, it is not right to change the admin link state value
when in switchdev mode.
Use direct vport commands to alter logical vport state in netdev
representor open/close flows rather than the administrative eswitch API.
Fixes: 20a1ea674783 ('net/mlx5e: Support VF vport link state control for SRIOV switchdev mode')
Signed-off-by: Jianbo Liu <jianbol@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 11 ++++++-----
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 363d8dcb7f17..0273c233bc85 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -668,7 +668,6 @@ static int mlx5e_rep_open(struct net_device *dev)
struct mlx5e_priv *priv = netdev_priv(dev);
struct mlx5e_rep_priv *rpriv = priv->ppriv;
struct mlx5_eswitch_rep *rep = rpriv->rep;
- struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
int err;
mutex_lock(&priv->state_lock);
@@ -676,8 +675,9 @@ static int mlx5e_rep_open(struct net_device *dev)
if (err)
goto unlock;
- if (!mlx5_eswitch_set_vport_state(esw, rep->vport,
- MLX5_ESW_VPORT_ADMIN_STATE_UP))
+ if (!mlx5_modify_vport_admin_state(priv->mdev,
+ MLX5_QUERY_VPORT_STATE_IN_OP_MOD_ESW_VPORT,
+ rep->vport, MLX5_ESW_VPORT_ADMIN_STATE_UP))
netif_carrier_on(dev);
unlock:
@@ -690,11 +690,12 @@ static int mlx5e_rep_close(struct net_device *dev)
struct mlx5e_priv *priv = netdev_priv(dev);
struct mlx5e_rep_priv *rpriv = priv->ppriv;
struct mlx5_eswitch_rep *rep = rpriv->rep;
- struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
int ret;
mutex_lock(&priv->state_lock);
- (void)mlx5_eswitch_set_vport_state(esw, rep->vport, MLX5_ESW_VPORT_ADMIN_STATE_DOWN);
+ mlx5_modify_vport_admin_state(priv->mdev,
+ MLX5_QUERY_VPORT_STATE_IN_OP_MOD_ESW_VPORT,
+ rep->vport, MLX5_ESW_VPORT_ADMIN_STATE_DOWN);
ret = mlx5e_close_locked(dev);
mutex_unlock(&priv->state_lock);
return ret;
--
2.14.3
^ permalink raw reply related
* [pull request][net 0/8] Mellanox, mlx5 fixes 2018-03-23
From: Saeed Mahameed @ 2018-03-23 22:05 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Saeed Mahameed
Hi Dave,
The following series includes fixes for mlx5 netdev and eswitch.
For -stable v4.12
('net/mlx5e: Avoid using the ipv6 stub in the TC offload neigh update path')
('net/mlx5e: Fix traffic being dropped on VF representor')
For -stable v4.13
('net/mlx5e: Fix memory usage issues in offloading TC flows')
('net/mlx5e: Verify coalescing parameters in range')
For -stable v4.14
('net/mlx5e: Don't override vport admin link state in switchdev mode')
For -stable v4.15
('108b2b6d5c02 net/mlx5e: Sync netdev vxlan ports at open')
Please pull and let me know if there's any problem.
Thanks,
Saeed.
---
The following changes since commit 1bfa26ff8c4b7512f4e4efa6df211239223033d4:
ipv6: fix possible deadlock in rt6_age_examine_exception() (2018-03-23 13:40:34 -0400)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git tags/mlx5-fixes-2018-03-23
for you to fetch changes up to dcbfde231207f7b6e1f5628d59b182198e1cfae2:
net/mlx5e: Sync netdev vxlan ports at open (2018-03-23 14:23:50 -0700)
----------------------------------------------------------------
mlx5-fixes-2018-03-23
----------------------------------------------------------------
Jianbo Liu (2):
net/mlx5e: Don't override vport admin link state in switchdev mode
net/mlx5e: Fix memory usage issues in offloading TC flows
Moshe Shemesh (1):
net/mlx5e: Verify coalescing parameters in range
Or Gerlitz (3):
net/mlx5e: Use 32 bits to store VF representor SQ number
net/mlx5: Make eswitch support to depend on switchdev
net/mlx5e: Avoid using the ipv6 stub in the TC offload neigh update path
Roi Dayan (1):
net/mlx5e: Fix traffic being dropped on VF representor
Shahar Klein (1):
net/mlx5e: Sync netdev vxlan ports at open
drivers/net/ethernet/mellanox/mlx5/core/Kconfig | 2 +-
.../net/ethernet/mellanox/mlx5/core/en_ethtool.c | 17 +++++++++++
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 13 ++++-----
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 34 ++++++++++++----------
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 18 ++++++------
5 files changed, 51 insertions(+), 33 deletions(-)
^ permalink raw reply
* [net 3/8] net/mlx5: Make eswitch support to depend on switchdev
From: Saeed Mahameed @ 2018-03-23 22:05 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Or Gerlitz, Saeed Mahameed
In-Reply-To: <20180323220534.19353-1-saeedm@mellanox.com>
From: Or Gerlitz <ogerlitz@mellanox.com>
Add dependancy for switchdev to be congfigured as any user-space control
plane SW is expected to use the HW switchdev ID to locate the representors
related to VFs of a certain PF and apply SW/offloaded switching on them.
Fixes: e80541ecabd5 ('net/mlx5: Add CONFIG_MLX5_ESWITCH Kconfig')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/Kconfig | 2 +-
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 4 ++--
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 2 --
3 files changed, 3 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
index 25deaa5a534c..c032319f1cb9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
@@ -46,7 +46,7 @@ config MLX5_MPFS
config MLX5_ESWITCH
bool "Mellanox Technologies MLX5 SRIOV E-Switch support"
- depends on MLX5_CORE_EN
+ depends on MLX5_CORE_EN && NET_SWITCHDEV
default y
---help---
Mellanox Technologies Ethernet SRIOV E-Switch support in ConnectX NIC.
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index da94c8cba5ee..e35859657217 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -4069,7 +4069,7 @@ static void mlx5e_set_netdev_dev_addr(struct net_device *netdev)
}
}
-#if IS_ENABLED(CONFIG_NET_SWITCHDEV) && IS_ENABLED(CONFIG_MLX5_ESWITCH)
+#if IS_ENABLED(CONFIG_MLX5_ESWITCH)
static const struct switchdev_ops mlx5e_switchdev_ops = {
.switchdev_port_attr_get = mlx5e_attr_get,
};
@@ -4175,7 +4175,7 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
mlx5e_set_netdev_dev_addr(netdev);
-#if IS_ENABLED(CONFIG_NET_SWITCHDEV) && IS_ENABLED(CONFIG_MLX5_ESWITCH)
+#if IS_ENABLED(CONFIG_MLX5_ESWITCH)
if (MLX5_VPORT_MANAGER(mdev))
netdev->switchdev_ops = &mlx5e_switchdev_ops;
#endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 738554a6c69f..5ece289548ad 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -900,9 +900,7 @@ static void mlx5e_build_rep_netdev(struct net_device *netdev)
netdev->ethtool_ops = &mlx5e_rep_ethtool_ops;
-#ifdef CONFIG_NET_SWITCHDEV
netdev->switchdev_ops = &mlx5e_rep_switchdev_ops;
-#endif
netdev->features |= NETIF_F_VLAN_CHALLENGED | NETIF_F_HW_TC | NETIF_F_NETNS_LOCAL;
netdev->hw_features |= NETIF_F_HW_TC;
--
2.14.3
^ permalink raw reply related
* [net 6/8] net/mlx5e: Fix memory usage issues in offloading TC flows
From: Saeed Mahameed @ 2018-03-23 22:05 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Jianbo Liu, Saeed Mahameed
In-Reply-To: <20180323220534.19353-1-saeedm@mellanox.com>
From: Jianbo Liu <jianbol@mellanox.com>
For NIC flows, the parsed attributes are not freed when we exit
successfully from mlx5e_configure_flower().
There is possible double free for eswitch flows. If error is returned
from rhashtable_insert_fast(), the parse attrs will be freed in
mlx5e_tc_del_flow(), but they will be freed again before exiting
mlx5e_configure_flower().
To fix both issues we do the following:
(1) change the condition that determines if to issue the free call to
check if this flow is NIC flow, or it does not have encap action.
(2) reorder the code such that that the check and free calls are done
before we attempt to add into the hash table.
Fixes: 232c001398ae ('net/mlx5e: Add support to neighbour update flow')
Signed-off-by: Jianbo Liu <jianbol@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 16 ++++++++--------
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index fa86a1466718..ae11678a31e8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -2608,19 +2608,19 @@ int mlx5e_configure_flower(struct mlx5e_priv *priv,
if (err != -EAGAIN)
flow->flags |= MLX5E_TC_FLOW_OFFLOADED;
+ if (!(flow->flags & MLX5E_TC_FLOW_ESWITCH) ||
+ !(flow->esw_attr->action & MLX5_FLOW_CONTEXT_ACTION_ENCAP))
+ kvfree(parse_attr);
+
err = rhashtable_insert_fast(&tc->ht, &flow->node,
tc->ht_params);
- if (err)
- goto err_del_rule;
+ if (err) {
+ mlx5e_tc_del_flow(priv, flow);
+ kfree(flow);
+ }
- if (flow->flags & MLX5E_TC_FLOW_ESWITCH &&
- !(flow->esw_attr->action & MLX5_FLOW_CONTEXT_ACTION_ENCAP))
- kvfree(parse_attr);
return err;
-err_del_rule:
- mlx5e_tc_del_flow(priv, flow);
-
err_free:
kvfree(parse_attr);
kfree(flow);
--
2.14.3
^ permalink raw reply related
* [net 2/8] net/mlx5e: Use 32 bits to store VF representor SQ number
From: Saeed Mahameed @ 2018-03-23 22:05 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Or Gerlitz, Saeed Mahameed
In-Reply-To: <20180323220534.19353-1-saeedm@mellanox.com>
From: Or Gerlitz <ogerlitz@mellanox.com>
SQs are 32 and not 16 bits, hence it's wrong to use only 16 bits to
store the sq number for which are going to set steering rule, fix that.
Fixes: cb67b832921c ('net/mlx5e: Introduce SRIOV VF representors')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 0273c233bc85..738554a6c69f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -209,7 +209,7 @@ static void mlx5e_sqs2vport_stop(struct mlx5_eswitch *esw,
static int mlx5e_sqs2vport_start(struct mlx5_eswitch *esw,
struct mlx5_eswitch_rep *rep,
- u16 *sqns_array, int sqns_num)
+ u32 *sqns_array, int sqns_num)
{
struct mlx5_flow_handle *flow_rule;
struct mlx5e_rep_priv *rpriv;
@@ -255,9 +255,9 @@ int mlx5e_add_sqs_fwd_rules(struct mlx5e_priv *priv)
struct mlx5e_channel *c;
int n, tc, num_sqs = 0;
int err = -ENOMEM;
- u16 *sqs;
+ u32 *sqs;
- sqs = kcalloc(priv->channels.num * priv->channels.params.num_tc, sizeof(u16), GFP_KERNEL);
+ sqs = kcalloc(priv->channels.num * priv->channels.params.num_tc, sizeof(*sqs), GFP_KERNEL);
if (!sqs)
goto out;
--
2.14.3
^ permalink raw reply related
* [net 5/8] net/mlx5e: Fix traffic being dropped on VF representor
From: Saeed Mahameed @ 2018-03-23 22:05 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Roi Dayan, Saeed Mahameed
In-Reply-To: <20180323220534.19353-1-saeedm@mellanox.com>
From: Roi Dayan <roid@mellanox.com>
Increase representor netdev RQ size to avoid dropped packets.
The current size (two) is just too small to keep up with
conventional slow path traffic patterns.
Also match the SQ size to the RQ size.
Fixes: cb67b832921c ("net/mlx5e: Introduce SRIOV VF representors")
Signed-off-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Paul Blakey <paulb@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 9 +++++++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 5ece289548ad..9a5a2a7eeab3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -44,6 +44,11 @@
#include "en_tc.h"
#include "fs_core.h"
+#define MLX5E_REP_PARAMS_LOG_SQ_SIZE \
+ max(0x6, MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE)
+#define MLX5E_REP_PARAMS_LOG_RQ_SIZE \
+ max(0x6, MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE)
+
static const char mlx5e_rep_driver_name[] = "mlx5e_rep";
static void mlx5e_rep_get_drvinfo(struct net_device *dev,
@@ -878,9 +883,9 @@ static void mlx5e_build_rep_params(struct mlx5_core_dev *mdev,
MLX5_CQ_PERIOD_MODE_START_FROM_CQE :
MLX5_CQ_PERIOD_MODE_START_FROM_EQE;
- params->log_sq_size = MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE;
+ params->log_sq_size = MLX5E_REP_PARAMS_LOG_SQ_SIZE;
params->rq_wq_type = MLX5_WQ_TYPE_LINKED_LIST;
- params->log_rq_size = MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE;
+ params->log_rq_size = MLX5E_REP_PARAMS_LOG_RQ_SIZE;
params->rx_dim_enabled = MLX5_CAP_GEN(mdev, cq_moderation);
mlx5e_set_rx_cq_mode_params(params, cq_period_mode);
--
2.14.3
^ permalink raw reply related
* [net 7/8] net/mlx5e: Avoid using the ipv6 stub in the TC offload neigh update path
From: Saeed Mahameed @ 2018-03-23 22:05 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Or Gerlitz, Saeed Mahameed
In-Reply-To: <20180323220534.19353-1-saeedm@mellanox.com>
From: Or Gerlitz <ogerlitz@mellanox.com>
Currently we use the global ipv6_stub var to access the ipv6 global
nd table. This practice gets us to troubles when the stub is only partially
set e.g when ipv6 is loaded under the disabled policy. In this case, as of commit
343d60aada5a "ipv6: change ipv6_stub_impl.ipv6_dst_lookup to take net argument"
the stub is not null, but stub->nd_tbl is and we crash.
As we can access directly the ipv6 nd_tbl, the fix is just avoid the
reference through the stub. There is one place in the code were we
issue ipv6 route lookup and keep doing it through the stub, but that
mentioned commit makes sure we get -EAFNOSUPPORT from the stack.
Fixes: 232c001398ae ('net/mlx5e: Add support to neighbour update flow')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Aviv Heller <avivh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 6 +++---
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 2 +-
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
index 9a5a2a7eeab3..500d817d2b0a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c
@@ -293,7 +293,7 @@ void mlx5e_remove_sqs_fwd_rules(struct mlx5e_priv *priv)
static void mlx5e_rep_neigh_update_init_interval(struct mlx5e_rep_priv *rpriv)
{
#if IS_ENABLED(CONFIG_IPV6)
- unsigned long ipv6_interval = NEIGH_VAR(&ipv6_stub->nd_tbl->parms,
+ unsigned long ipv6_interval = NEIGH_VAR(&nd_tbl.parms,
DELAY_PROBE_TIME);
#else
unsigned long ipv6_interval = ~0UL;
@@ -429,7 +429,7 @@ static int mlx5e_rep_netevent_event(struct notifier_block *nb,
case NETEVENT_NEIGH_UPDATE:
n = ptr;
#if IS_ENABLED(CONFIG_IPV6)
- if (n->tbl != ipv6_stub->nd_tbl && n->tbl != &arp_tbl)
+ if (n->tbl != &nd_tbl && n->tbl != &arp_tbl)
#else
if (n->tbl != &arp_tbl)
#endif
@@ -477,7 +477,7 @@ static int mlx5e_rep_netevent_event(struct notifier_block *nb,
* done per device delay prob time parameter.
*/
#if IS_ENABLED(CONFIG_IPV6)
- if (!p->dev || (p->tbl != ipv6_stub->nd_tbl && p->tbl != &arp_tbl))
+ if (!p->dev || (p->tbl != &nd_tbl && p->tbl != &arp_tbl))
#else
if (!p->dev || p->tbl != &arp_tbl)
#endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index ae11678a31e8..43234cabf444 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -963,7 +963,7 @@ void mlx5e_tc_update_neigh_used_value(struct mlx5e_neigh_hash_entry *nhe)
tbl = &arp_tbl;
#if IS_ENABLED(CONFIG_IPV6)
else if (m_neigh->family == AF_INET6)
- tbl = ipv6_stub->nd_tbl;
+ tbl = &nd_tbl;
#endif
else
return;
--
2.14.3
^ permalink raw reply related
* [net 8/8] net/mlx5e: Sync netdev vxlan ports at open
From: Saeed Mahameed @ 2018-03-23 22:05 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Shahar Klein, Saeed Mahameed
In-Reply-To: <20180323220534.19353-1-saeedm@mellanox.com>
From: Shahar Klein <shahark@mellanox.com>
When mlx5_core is loaded it is expected to sync ports
with all vxlan devices so it can support vxlan encap/decap.
This is done via udp_tunnel_get_rx_info(). Currently this
call is set in mlx5e_nic_enable() and if the netdev is not in
NETREG_REGISTERED state it will not be called.
Normally on load the netdev state is not NETREG_REGISTERED
so udp_tunnel_get_rx_info() will not be called.
Moving udp_tunnel_get_rx_info() to mlx5e_open() so
it will be called on netdev UP event and allow encap/decap.
Fixes: 610e89e05c3f ("net/mlx5e: Don't sync netdev state when not registered")
Signed-off-by: Shahar Klein <shahark@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 9 +++------
1 file changed, 3 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index e35859657217..9b4827d36e3e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -2572,6 +2572,9 @@ int mlx5e_open(struct net_device *netdev)
mlx5_set_port_admin_status(priv->mdev, MLX5_PORT_UP);
mutex_unlock(&priv->state_lock);
+ if (mlx5e_vxlan_allowed(priv->mdev))
+ udp_tunnel_get_rx_info(netdev);
+
return err;
}
@@ -4327,12 +4330,6 @@ static void mlx5e_nic_enable(struct mlx5e_priv *priv)
#ifdef CONFIG_MLX5_CORE_EN_DCB
mlx5e_dcbnl_init_app(priv);
#endif
- /* Device already registered: sync netdev system state */
- if (mlx5e_vxlan_allowed(mdev)) {
- rtnl_lock();
- udp_tunnel_get_rx_info(netdev);
- rtnl_unlock();
- }
queue_work(priv->wq, &priv->set_rx_mode_work);
--
2.14.3
^ permalink raw reply related
* Re: [PATCH v6 6/6] bnxt_en: Eliminate duplicate barriers on weakly-ordered archs
From: Michael Chan @ 2018-03-23 22:36 UTC (permalink / raw)
To: Sinan Kaya
Cc: Netdev, timur, sulrich, linux-arm-msm, linux-arm-kernel,
open list
In-Reply-To: <1521843791-21201-7-git-send-email-okaya@codeaurora.org>
On Fri, Mar 23, 2018 at 3:23 PM, Sinan Kaya <okaya@codeaurora.org> wrote:
> Code includes wmb() followed by writel(). writel() already has a barrier on
> some architectures like arm64.
>
> This ends up CPU observing two barriers back to back before executing the
> register write.
>
> Create a new wrapper function with relaxed write operator. Use the new
> wrapper when a write is following a wmb().
>
> Since code already has an explicit barrier call, changing writel() to
> writel_relaxed().
>
> Also add mmiowb() so that write code doesn't move outside of scope.
>
> Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
> ---
> drivers/net/ethernet/broadcom/bnxt/bnxt.c | 3 ++-
> drivers/net/ethernet/broadcom/bnxt/bnxt.h | 9 +++++++++
> 2 files changed, 11 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
> index 1500243..fc8ea0d 100644
> --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
> +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
> @@ -1922,7 +1922,8 @@ static int bnxt_poll_work(struct bnxt *bp, struct bnxt_napi *bnapi, int budget)
> /* Sync BD data before updating doorbell */
> wmb();
>
> - bnxt_db_write(bp, db, DB_KEY_TX | prod);
> + bnxt_db_write_relaxed(bp, db, DB_KEY_TX | prod);
> + mmiowb();
Sorry for the late review. mmiowb() is not required here because we
are in NAPI context, so only one CPU will be updating this doorbell.
Other than that, it looks good.
> }
>
> cpr->cp_raw_cons = raw_cons;
> diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
> index 1989c47..5e453b9 100644
> --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
> +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
> @@ -1401,6 +1401,15 @@ static inline u32 bnxt_tx_avail(struct bnxt *bp, struct bnxt_tx_ring_info *txr)
> ((txr->tx_prod - txr->tx_cons) & bp->tx_ring_mask);
> }
>
> +/* For TX and RX ring doorbells with no ordering guarantee*/
> +static inline void bnxt_db_write_relaxed(struct bnxt *bp, void __iomem *db,
> + u32 val)
> +{
> + writel_relaxed(val, db);
> + if (bp->flags & BNXT_FLAG_DOUBLE_DB)
> + writel_relaxed(val, db);
> +}
> +
> /* For TX and RX ring doorbells */
> static inline void bnxt_db_write(struct bnxt *bp, void __iomem *db, u32 val)
> {
> --
> 2.7.4
>
^ permalink raw reply
* [net-next 01/15] net/mlx5e: Expose PFC stall prevention counters
From: Saeed Mahameed @ 2018-03-23 22:39 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Inbar Karmy, Saeed Mahameed
In-Reply-To: <20180323223925.21678-1-saeedm@mellanox.com>
From: Inbar Karmy <inbark@mellanox.com>
Add the needed capability bit and counters to device spec description.
Expose the following two counters in ethtool:
tx_pause_storm_warning_events: when the device is stalled for a period
longer than a pre-configured watermark, the counter increase, allowing
the debug utility an insight into current device status.
tx_pause_storm_error_events: when the device is stalled for a period
longer than a pre-configured timeout, the pause transmission is disabled,
and the counter increase.
Signed-off-by: Inbar Karmy <inbark@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 19 ++++++++++++++-
drivers/net/ethernet/mellanox/mlx5/core/fw.c | 3 +++
include/linux/mlx5/device.h | 4 ++++
include/linux/mlx5/mlx5_ifc.h | 28 +++++++++++++++++++---
4 files changed, 50 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index 5f0f3493d747..2553c58dcf1c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -754,7 +754,15 @@ static const struct counter_desc pport_per_prio_pfc_stats_desc[] = {
{ "rx_%s_pause_transition", PPORT_PER_PRIO_OFF(rx_pause_transition) },
};
+static const struct counter_desc pport_pfc_stall_stats_desc[] = {
+ { "tx_pause_storm_warning_events ", PPORT_PER_PRIO_OFF(device_stall_minor_watermark_cnt) },
+ { "tx_pause_storm_error_events", PPORT_PER_PRIO_OFF(device_stall_critical_watermark_cnt) },
+};
+
#define NUM_PPORT_PER_PRIO_PFC_COUNTERS ARRAY_SIZE(pport_per_prio_pfc_stats_desc)
+#define NUM_PPORT_PFC_STALL_COUNTERS(priv) (ARRAY_SIZE(pport_pfc_stall_stats_desc) * \
+ MLX5_CAP_PCAM_FEATURE((priv)->mdev, pfcc_mask) * \
+ MLX5_CAP_DEBUG((priv)->mdev, stall_detect))
static unsigned long mlx5e_query_pfc_combined(struct mlx5e_priv *priv)
{
@@ -790,7 +798,8 @@ static int mlx5e_grp_per_prio_pfc_get_num_stats(struct mlx5e_priv *priv)
{
return (mlx5e_query_global_pause_combined(priv) +
hweight8(mlx5e_query_pfc_combined(priv))) *
- NUM_PPORT_PER_PRIO_PFC_COUNTERS;
+ NUM_PPORT_PER_PRIO_PFC_COUNTERS +
+ NUM_PPORT_PFC_STALL_COUNTERS(priv);
}
static int mlx5e_grp_per_prio_pfc_fill_strings(struct mlx5e_priv *priv,
@@ -818,6 +827,10 @@ static int mlx5e_grp_per_prio_pfc_fill_strings(struct mlx5e_priv *priv,
}
}
+ for (i = 0; i < NUM_PPORT_PFC_STALL_COUNTERS(priv); i++)
+ strcpy(data + (idx++) * ETH_GSTRING_LEN,
+ pport_pfc_stall_stats_desc[i].format);
+
return idx;
}
@@ -845,6 +858,10 @@ static int mlx5e_grp_per_prio_pfc_fill_stats(struct mlx5e_priv *priv,
}
}
+ for (i = 0; i < NUM_PPORT_PFC_STALL_COUNTERS(priv); i++)
+ data[idx++] = MLX5E_READ_CTR64_BE(&priv->stats.pport.per_prio_counters[0],
+ pport_pfc_stall_stats_desc, i);
+
return idx;
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c
index 9d11e92fb541..d7bb10ab2173 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c
@@ -183,6 +183,9 @@ int mlx5_query_hca_caps(struct mlx5_core_dev *dev)
return err;
}
+ if (MLX5_CAP_GEN(dev, debug))
+ mlx5_core_get_caps(dev, MLX5_CAP_DEBUG);
+
if (MLX5_CAP_GEN(dev, pcam_reg))
mlx5_get_pcam_reg(dev);
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index e5258ee4e38b..4b5939c78cdd 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -1013,6 +1013,7 @@ enum mlx5_cap_type {
MLX5_CAP_RESERVED,
MLX5_CAP_VECTOR_CALC,
MLX5_CAP_QOS,
+ MLX5_CAP_DEBUG,
/* NUM OF CAP Types */
MLX5_CAP_NUM
};
@@ -1140,6 +1141,9 @@ enum mlx5_qcam_feature_groups {
#define MLX5_CAP_QOS(mdev, cap)\
MLX5_GET(qos_cap, mdev->caps.hca_cur[MLX5_CAP_QOS], cap)
+#define MLX5_CAP_DEBUG(mdev, cap)\
+ MLX5_GET(debug_cap, mdev->caps.hca_cur[MLX5_CAP_DEBUG], cap)
+
#define MLX5_CAP_PCAM_FEATURE(mdev, fld) \
MLX5_GET(pcam_reg, (mdev)->caps.pcam, feature_cap_mask.enhanced_features.fld)
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 14ad84afe8ba..c7d50eccff9e 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -593,6 +593,16 @@ struct mlx5_ifc_qos_cap_bits {
u8 reserved_at_100[0x700];
};
+struct mlx5_ifc_debug_cap_bits {
+ u8 reserved_at_0[0x20];
+
+ u8 reserved_at_20[0x2];
+ u8 stall_detect[0x1];
+ u8 reserved_at_23[0x1d];
+
+ u8 reserved_at_40[0x7c0];
+};
+
struct mlx5_ifc_per_protocol_networking_offload_caps_bits {
u8 csum_cap[0x1];
u8 vlan_cap[0x1];
@@ -855,7 +865,7 @@ struct mlx5_ifc_cmd_hca_cap_bits {
u8 out_of_seq_cnt[0x1];
u8 vport_counters[0x1];
u8 retransmission_q_counters[0x1];
- u8 reserved_at_183[0x1];
+ u8 debug[0x1];
u8 modify_rq_counter_set_id[0x1];
u8 rq_delay_drop[0x1];
u8 max_qp_cnt[0xa];
@@ -1572,7 +1582,17 @@ struct mlx5_ifc_eth_per_prio_grp_data_layout_bits {
u8 rx_pause_transition_low[0x20];
- u8 reserved_at_3c0[0x400];
+ u8 reserved_at_3c0[0x40];
+
+ u8 device_stall_minor_watermark_cnt_high[0x20];
+
+ u8 device_stall_minor_watermark_cnt_low[0x20];
+
+ u8 device_stall_critical_watermark_cnt_high[0x20];
+
+ u8 device_stall_critical_watermark_cnt_low[0x20];
+
+ u8 reserved_at_480[0x340];
};
struct mlx5_ifc_eth_extended_cntrs_grp_data_layout_bits {
@@ -7874,8 +7894,10 @@ struct mlx5_ifc_peir_reg_bits {
};
struct mlx5_ifc_pcam_enhanced_features_bits {
- u8 reserved_at_0[0x7b];
+ u8 reserved_at_0[0x76];
+ u8 pfcc_mask[0x1];
+ u8 reserved_at_77[0x4];
u8 rx_buffer_fullness_counters[0x1];
u8 ptys_connector_type[0x1];
u8 reserved_at_7d[0x1];
--
2.14.3
^ permalink raw reply related
* [net-next 02/15] ethtool: Add support for configuring PFC stall prevention in ethtool
From: Saeed Mahameed @ 2018-03-23 22:39 UTC (permalink / raw)
To: David S. Miller
Cc: netdev, Inbar Karmy, Michal Kubecek, Andrew Lunn, Saeed Mahameed
In-Reply-To: <20180323223925.21678-1-saeedm@mellanox.com>
From: Inbar Karmy <inbark@mellanox.com>
In the event where the device unexpectedly becomes unresponsive
for a long period of time, flow control mechanism may propagate
pause frames which will cause congestion spreading to the entire
network.
To prevent this scenario, when the device is stalled for a period
longer than a pre-configured timeout, flow control mechanisms are
automatically disabled.
This patch adds support for the ETHTOOL_PFC_STALL_PREVENTION
as a tunable.
This API provides support for configuring flow control storm prevention
timeout (msec).
Signed-off-by: Inbar Karmy <inbark@mellanox.com>
Cc: Michal Kubecek <mkubecek@suse.cz>
Cc: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
include/uapi/linux/ethtool.h | 4 ++++
net/core/ethtool.c | 6 ++++++
2 files changed, 10 insertions(+)
diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index 20da156aaf64..9dc63a14a747 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -217,10 +217,14 @@ struct ethtool_value {
__u32 data;
};
+#define PFC_STORM_PREVENTION_AUTO 0xffff
+#define PFC_STORM_PREVENTION_DISABLE 0
+
enum tunable_id {
ETHTOOL_ID_UNSPEC,
ETHTOOL_RX_COPYBREAK,
ETHTOOL_TX_COPYBREAK,
+ ETHTOOL_PFC_PREVENTION_TOUT,
/*
* Add your fresh new tubale attribute above and remember to update
* tunable_strings[] in net/core/ethtool.c
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 157cd9efa4be..bb6e498c6e3d 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -121,6 +121,7 @@ tunable_strings[__ETHTOOL_TUNABLE_COUNT][ETH_GSTRING_LEN] = {
[ETHTOOL_ID_UNSPEC] = "Unspec",
[ETHTOOL_RX_COPYBREAK] = "rx-copybreak",
[ETHTOOL_TX_COPYBREAK] = "tx-copybreak",
+ [ETHTOOL_PFC_PREVENTION_TOUT] = "pfc-prevention-tout",
};
static const char
@@ -2311,6 +2312,11 @@ static int ethtool_tunable_valid(const struct ethtool_tunable *tuna)
tuna->type_id != ETHTOOL_TUNABLE_U32)
return -EINVAL;
break;
+ case ETHTOOL_PFC_PREVENTION_TOUT:
+ if (tuna->len != sizeof(u16) ||
+ tuna->type_id != ETHTOOL_TUNABLE_U16)
+ return -EINVAL;
+ break;
default:
return -EINVAL;
}
--
2.14.3
^ permalink raw reply related
* [pull request][net-next 00/15] Mellanox, mlx5 misc updates 2018-03-22
From: Saeed Mahameed @ 2018-03-23 22:39 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Saeed Mahameed
Hi Dave,
This series includes some misc updates to mlx5 core and netdev driver,
please note that there is a small change to net/core/ethtool.c and
include/uapi/linux/ethtool.h that adds new tunable for PFC stall
prevention on/off support, which was already reviewed as RFC [1].
For more information please review and see the tag log below.
Please pull and let me know if there's any problem.
P.S.: This series doesn't introduce any conflict with the ongoing
mlx5 fixes series, mlx5-fixes-2018-03-23.
[1] https://patchwork.ozlabs.org/cover/838314/
Thanks,
Saeed.
---
The following changes since commit 6686c459e1449a3ee5f3fd313b0a559ace7a700e:
Merge branch 'hns3-VF-reset' (2018-03-22 15:29:05 -0400)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git tags/mlx5-updates-2018-03-22
for you to fetch changes up to 50f0e3bc5e4b5079b1e9b829db1a2246757965d3:
net/mlx5e: Add VLAN offload features to hw_enc_features (2018-03-22 16:28:31 -0700)
----------------------------------------------------------------
mlx5-updates-2018-03-22 (Misc updates)
This series includes misc updates for mlx5 core and netdev dirver,
Highlights:
>From Inbar, three patches to add support for PFC stall prevention
statistics and enable/disable through new ethtool tunable, as requested
from previous submission.
>From Moshe, four patches, added more drop counters:
- drop counter for netdev steering miss
- drop counter for when VF logical link is down
- drop counter for when netdev logical link is down.
>From Or, three patches to support vlan push/pop offload via tc HW action,
for newer HW (Connectx-5 and onward) via HW steering flow actions rather
than the emulated path for the older HW brands.
And five more misc small trivial patches.
----------------------------------------------------------------
Aviv Heller (1):
net/mlx5e: Add VLAN offload features to hw_enc_features
Gal Pressman (3):
net/mlx5e: Remove redundant check in get ethtool stats
net/mlx5e: Make choose LRO timeout function static
net/mlx5e: Add a helper macro in set features ndo
Inbar Karmy (3):
net/mlx5e: Expose PFC stall prevention counters
ethtool: Add support for configuring PFC stall prevention in ethtool
net/mlx5e: PFC stall prevention support
Leon Romanovsky (1):
net/mlx5: Protect from command bit overflow
Moshe Shemesh (4):
net/mlx5: Add support for QUERY_VNIC_ENV command
net/mlx5e: Add vnic steering drop statistics
net/mlx5: Add packet dropped while vport down statistics
net/mlx5e: Add interface down dropped packets statistics
Or Gerlitz (3):
net/mlx5: E-Switch, Use same source for offloaded actions check
net/mlx5: Add core support for vlan push/pop steering action
net/mlx5e: Offload tc vlan push/pop using HW action
drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 4 +-
.../mellanox/mlx5/core/diag/fs_tracepoint.h | 2 +
drivers/net/ethernet/mellanox/mlx5/core/en.h | 4 +-
.../net/ethernet/mellanox/mlx5/core/en_ethtool.c | 60 +++++++++-
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 77 +++++++------
drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 126 +++++++++++++++++++--
drivers/net/ethernet/mellanox/mlx5/core/en_stats.h | 6 +
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 15 ++-
drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 31 ++++-
drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 13 ++-
.../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 30 +++--
drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c | 10 +-
drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 4 +-
drivers/net/ethernet/mellanox/mlx5/core/fw.c | 3 +
drivers/net/ethernet/mellanox/mlx5/core/port.c | 64 +++++++++--
drivers/net/ethernet/mellanox/mlx5/core/vport.c | 26 +++++
include/linux/mlx5/device.h | 4 +
include/linux/mlx5/fs.h | 7 ++
include/linux/mlx5/mlx5_ifc.h | 116 +++++++++++++++++--
include/linux/mlx5/port.h | 6 +
include/linux/mlx5/vport.h | 3 +
include/uapi/linux/ethtool.h | 4 +
net/core/ethtool.c | 6 +
23 files changed, 531 insertions(+), 90 deletions(-)
^ permalink raw reply
* [net-next 04/15] net/mlx5: Add support for QUERY_VNIC_ENV command
From: Saeed Mahameed @ 2018-03-23 22:39 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Moshe Shemesh, Saeed Mahameed
In-Reply-To: <20180323223925.21678-1-saeedm@mellanox.com>
From: Moshe Shemesh <moshe@mellanox.com>
Add support for new FW command QUERY_VNIC_ENV.
The command is used by the driver to query vnic diagnostic statistics
from FW.
Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 2 ++
include/linux/mlx5/mlx5_ifc.h | 50 ++++++++++++++++++++++++++-
2 files changed, 51 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
index e9a1fbcc4adf..fe5428667ad1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
@@ -359,6 +359,7 @@ static int mlx5_internal_err_ret_value(struct mlx5_core_dev *dev, u16 op,
case MLX5_CMD_OP_MODIFY_HCA_VPORT_CONTEXT:
case MLX5_CMD_OP_QUERY_HCA_VPORT_GID:
case MLX5_CMD_OP_QUERY_HCA_VPORT_PKEY:
+ case MLX5_CMD_OP_QUERY_VNIC_ENV:
case MLX5_CMD_OP_QUERY_VPORT_COUNTER:
case MLX5_CMD_OP_ALLOC_Q_COUNTER:
case MLX5_CMD_OP_QUERY_Q_COUNTER:
@@ -501,6 +502,7 @@ const char *mlx5_command_str(int command)
MLX5_COMMAND_STR_CASE(MODIFY_HCA_VPORT_CONTEXT);
MLX5_COMMAND_STR_CASE(QUERY_HCA_VPORT_GID);
MLX5_COMMAND_STR_CASE(QUERY_HCA_VPORT_PKEY);
+ MLX5_COMMAND_STR_CASE(QUERY_VNIC_ENV);
MLX5_COMMAND_STR_CASE(QUERY_VPORT_COUNTER);
MLX5_COMMAND_STR_CASE(ALLOC_Q_COUNTER);
MLX5_COMMAND_STR_CASE(DEALLOC_Q_COUNTER);
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index f3200a9696d6..52e373dd2679 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -143,6 +143,7 @@ enum {
MLX5_CMD_OP_MODIFY_HCA_VPORT_CONTEXT = 0x763,
MLX5_CMD_OP_QUERY_HCA_VPORT_GID = 0x764,
MLX5_CMD_OP_QUERY_HCA_VPORT_PKEY = 0x765,
+ MLX5_CMD_OP_QUERY_VNIC_ENV = 0x76f,
MLX5_CMD_OP_QUERY_VPORT_COUNTER = 0x770,
MLX5_CMD_OP_ALLOC_Q_COUNTER = 0x771,
MLX5_CMD_OP_DEALLOC_Q_COUNTER = 0x772,
@@ -875,7 +876,7 @@ struct mlx5_ifc_cmd_hca_cap_bits {
u8 vhca_group_manager[0x1];
u8 ib_virt[0x1];
u8 eth_virt[0x1];
- u8 reserved_at_1a4[0x1];
+ u8 vnic_env_queue_counters[0x1];
u8 ets[0x1];
u8 nic_flow_table[0x1];
u8 eswitch_flow_table[0x1];
@@ -2386,6 +2387,24 @@ struct mlx5_ifc_xrc_srqc_bits {
u8 reserved_at_180[0x80];
};
+struct mlx5_ifc_vnic_diagnostic_statistics_bits {
+ u8 counter_error_queues[0x20];
+
+ u8 total_error_queues[0x20];
+
+ u8 send_queue_priority_update_flow[0x20];
+
+ u8 reserved_at_60[0x20];
+
+ u8 nic_receive_steering_discard[0x40];
+
+ u8 receive_discard_vport_down[0x40];
+
+ u8 transmit_discard_vport_down[0x40];
+
+ u8 reserved_at_140[0xec0];
+};
+
struct mlx5_ifc_traffic_counter_bits {
u8 packets[0x40];
@@ -3661,6 +3680,35 @@ struct mlx5_ifc_query_vport_state_in_bits {
u8 reserved_at_60[0x20];
};
+struct mlx5_ifc_query_vnic_env_out_bits {
+ u8 status[0x8];
+ u8 reserved_at_8[0x18];
+
+ u8 syndrome[0x20];
+
+ u8 reserved_at_40[0x40];
+
+ struct mlx5_ifc_vnic_diagnostic_statistics_bits vport_env;
+};
+
+enum {
+ MLX5_QUERY_VNIC_ENV_IN_OP_MOD_VPORT_DIAG_STATISTICS = 0x0,
+};
+
+struct mlx5_ifc_query_vnic_env_in_bits {
+ u8 opcode[0x10];
+ u8 reserved_at_10[0x10];
+
+ u8 reserved_at_20[0x10];
+ u8 op_mod[0x10];
+
+ u8 other_vport[0x1];
+ u8 reserved_at_41[0xf];
+ u8 vport_number[0x10];
+
+ u8 reserved_at_60[0x20];
+};
+
struct mlx5_ifc_query_vport_counter_out_bits {
u8 status[0x8];
u8 reserved_at_8[0x18];
--
2.14.3
^ permalink raw reply related
* [net-next 08/15] net/mlx5: E-Switch, Use same source for offloaded actions check
From: Saeed Mahameed @ 2018-03-23 22:39 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Or Gerlitz, Saeed Mahameed
In-Reply-To: <20180323223925.21678-1-saeedm@mellanox.com>
From: Or Gerlitz <ogerlitz@mellanox.com>
Align the checks for modify header and encap actions with the
rest of the code.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 0a8303c1b52f..21ebe3e80e6e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -88,10 +88,10 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw,
if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_DECAP)
spec->match_criteria_enable |= MLX5_MATCH_INNER_HEADERS;
- if (attr->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR)
+ if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR)
flow_act.modify_id = attr->mod_hdr_id;
- if (attr->action & MLX5_FLOW_CONTEXT_ACTION_ENCAP)
+ if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_ENCAP)
flow_act.encap_id = attr->encap_id;
rule = mlx5_add_flow_rules((struct mlx5_flow_table *)esw->fdb_table.fdb,
--
2.14.3
^ permalink raw reply related
* [net-next 07/15] net/mlx5e: Add interface down dropped packets statistics
From: Saeed Mahameed @ 2018-03-23 22:39 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Moshe Shemesh, Saeed Mahameed
In-Reply-To: <20180323223925.21678-1-saeedm@mellanox.com>
From: Moshe Shemesh <moshe@mellanox.com>
Added the following packets drop counter:
Rx interface down dropped packets - counts packets which were received
while the ETH interface was down.
This counter will be shown on ethtool as a new counter called
rx_if_down_packets.
The implementation allocates a q_counter for drop rq which gets all the
received traffic while the interface is down.
Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en.h | 3 +-
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 46 +++++++++++++--------
drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 48 ++++++++++++++++------
drivers/net/ethernet/mellanox/mlx5/core/en_stats.h | 1 +
4 files changed, 69 insertions(+), 29 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 4c9360b25532..48e0b2a747d9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -781,7 +781,8 @@ struct mlx5e_priv {
struct net_device *netdev;
struct mlx5e_stats stats;
struct hwtstamp_config tstamp;
- u16 q_counter;
+ u16 q_counter;
+ u16 drop_rq_q_counter;
#ifdef CONFIG_MLX5_CORE_EN_DCB
struct mlx5e_dcbx dcbx;
#endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index da94c8cba5ee..f8bc3bcdf046 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -615,8 +615,7 @@ static int mlx5e_create_rq(struct mlx5e_rq *rq,
static int mlx5e_modify_rq_state(struct mlx5e_rq *rq, int curr_state,
int next_state)
{
- struct mlx5e_channel *c = rq->channel;
- struct mlx5_core_dev *mdev = c->mdev;
+ struct mlx5_core_dev *mdev = rq->mdev;
void *in;
void *rqc;
@@ -1768,14 +1767,16 @@ static void mlx5e_build_rq_param(struct mlx5e_priv *priv,
param->wq.linear = 1;
}
-static void mlx5e_build_drop_rq_param(struct mlx5_core_dev *mdev,
+static void mlx5e_build_drop_rq_param(struct mlx5e_priv *priv,
struct mlx5e_rq_param *param)
{
+ struct mlx5_core_dev *mdev = priv->mdev;
void *rqc = param->rqc;
void *wq = MLX5_ADDR_OF(rqc, rqc, wq);
MLX5_SET(wq, wq, wq_type, MLX5_WQ_TYPE_LINKED_LIST);
MLX5_SET(wq, wq, log_wq_stride, ilog2(sizeof(struct mlx5e_rx_wqe)));
+ MLX5_SET(rqc, rqc, counter_set_id, priv->drop_rq_q_counter);
param->wq.buf_numa_node = dev_to_node(&mdev->pdev->dev);
}
@@ -2643,15 +2644,16 @@ static int mlx5e_alloc_drop_cq(struct mlx5_core_dev *mdev,
return mlx5e_alloc_cq_common(mdev, param, cq);
}
-static int mlx5e_open_drop_rq(struct mlx5_core_dev *mdev,
+static int mlx5e_open_drop_rq(struct mlx5e_priv *priv,
struct mlx5e_rq *drop_rq)
{
+ struct mlx5_core_dev *mdev = priv->mdev;
struct mlx5e_cq_param cq_param = {};
struct mlx5e_rq_param rq_param = {};
struct mlx5e_cq *cq = &drop_rq->cq;
int err;
- mlx5e_build_drop_rq_param(mdev, &rq_param);
+ mlx5e_build_drop_rq_param(priv, &rq_param);
err = mlx5e_alloc_drop_cq(mdev, cq, &cq_param);
if (err)
@@ -2669,6 +2671,10 @@ static int mlx5e_open_drop_rq(struct mlx5_core_dev *mdev,
if (err)
goto err_free_rq;
+ err = mlx5e_modify_rq_state(drop_rq, MLX5_RQC_STATE_RST, MLX5_RQC_STATE_RDY);
+ if (err)
+ mlx5_core_warn(priv->mdev, "modify_rq_state failed, rx_if_down_packets won't be counted %d\n", err);
+
return 0;
err_free_rq:
@@ -4183,7 +4189,7 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
mlx5e_ipsec_build_netdev(priv);
}
-static void mlx5e_create_q_counter(struct mlx5e_priv *priv)
+static void mlx5e_create_q_counters(struct mlx5e_priv *priv)
{
struct mlx5_core_dev *mdev = priv->mdev;
int err;
@@ -4193,14 +4199,21 @@ static void mlx5e_create_q_counter(struct mlx5e_priv *priv)
mlx5_core_warn(mdev, "alloc queue counter failed, %d\n", err);
priv->q_counter = 0;
}
+
+ err = mlx5_core_alloc_q_counter(mdev, &priv->drop_rq_q_counter);
+ if (err) {
+ mlx5_core_warn(mdev, "alloc drop RQ counter failed, %d\n", err);
+ priv->drop_rq_q_counter = 0;
+ }
}
-static void mlx5e_destroy_q_counter(struct mlx5e_priv *priv)
+static void mlx5e_destroy_q_counters(struct mlx5e_priv *priv)
{
- if (!priv->q_counter)
- return;
+ if (priv->q_counter)
+ mlx5_core_dealloc_q_counter(priv->mdev, priv->q_counter);
- mlx5_core_dealloc_q_counter(priv->mdev, priv->q_counter);
+ if (priv->drop_rq_q_counter)
+ mlx5_core_dealloc_q_counter(priv->mdev, priv->drop_rq_q_counter);
}
static void mlx5e_nic_init(struct mlx5_core_dev *mdev,
@@ -4439,18 +4452,18 @@ int mlx5e_attach_netdev(struct mlx5e_priv *priv)
if (err)
goto out;
- err = mlx5e_open_drop_rq(mdev, &priv->drop_rq);
+ mlx5e_create_q_counters(priv);
+
+ err = mlx5e_open_drop_rq(priv, &priv->drop_rq);
if (err) {
mlx5_core_err(mdev, "open drop rq failed, %d\n", err);
- goto err_cleanup_tx;
+ goto err_destroy_q_counters;
}
err = profile->init_rx(priv);
if (err)
goto err_close_drop_rq;
- mlx5e_create_q_counter(priv);
-
if (profile->enable)
profile->enable(priv);
@@ -4459,7 +4472,8 @@ int mlx5e_attach_netdev(struct mlx5e_priv *priv)
err_close_drop_rq:
mlx5e_close_drop_rq(&priv->drop_rq);
-err_cleanup_tx:
+err_destroy_q_counters:
+ mlx5e_destroy_q_counters(priv);
profile->cleanup_tx(priv);
out:
@@ -4476,9 +4490,9 @@ void mlx5e_detach_netdev(struct mlx5e_priv *priv)
profile->disable(priv);
flush_workqueue(priv->wq);
- mlx5e_destroy_q_counter(priv);
profile->cleanup_rx(priv);
mlx5e_close_drop_rq(&priv->drop_rq);
+ mlx5e_destroy_q_counters(priv);
profile->cleanup_tx(priv);
cancel_delayed_work_sync(&priv->update_stats_work);
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index 552510c03ef2..c0dab9a8969e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -170,11 +170,24 @@ static const struct counter_desc q_stats_desc[] = {
{ MLX5E_DECLARE_STAT(struct mlx5e_qcounter_stats, rx_out_of_buffer) },
};
+static const struct counter_desc drop_rq_stats_desc[] = {
+ { MLX5E_DECLARE_STAT(struct mlx5e_qcounter_stats, rx_if_down_packets) },
+};
+
#define NUM_Q_COUNTERS ARRAY_SIZE(q_stats_desc)
+#define NUM_DROP_RQ_COUNTERS ARRAY_SIZE(drop_rq_stats_desc)
static int mlx5e_grp_q_get_num_stats(struct mlx5e_priv *priv)
{
- return priv->q_counter ? NUM_Q_COUNTERS : 0;
+ int num_stats = 0;
+
+ if (priv->q_counter)
+ num_stats += NUM_Q_COUNTERS;
+
+ if (priv->drop_rq_q_counter)
+ num_stats += NUM_DROP_RQ_COUNTERS;
+
+ return num_stats;
}
static int mlx5e_grp_q_fill_strings(struct mlx5e_priv *priv, u8 *data, int idx)
@@ -182,7 +195,13 @@ static int mlx5e_grp_q_fill_strings(struct mlx5e_priv *priv, u8 *data, int idx)
int i;
for (i = 0; i < NUM_Q_COUNTERS && priv->q_counter; i++)
- strcpy(data + (idx++) * ETH_GSTRING_LEN, q_stats_desc[i].format);
+ strcpy(data + (idx++) * ETH_GSTRING_LEN,
+ q_stats_desc[i].format);
+
+ for (i = 0; i < NUM_DROP_RQ_COUNTERS && priv->drop_rq_q_counter; i++)
+ strcpy(data + (idx++) * ETH_GSTRING_LEN,
+ drop_rq_stats_desc[i].format);
+
return idx;
}
@@ -191,7 +210,11 @@ static int mlx5e_grp_q_fill_stats(struct mlx5e_priv *priv, u64 *data, int idx)
int i;
for (i = 0; i < NUM_Q_COUNTERS && priv->q_counter; i++)
- data[idx++] = MLX5E_READ_CTR32_CPU(&priv->stats.qcnt, q_stats_desc, i);
+ data[idx++] = MLX5E_READ_CTR32_CPU(&priv->stats.qcnt,
+ q_stats_desc, i);
+ for (i = 0; i < NUM_DROP_RQ_COUNTERS && priv->drop_rq_q_counter; i++)
+ data[idx++] = MLX5E_READ_CTR32_CPU(&priv->stats.qcnt,
+ drop_rq_stats_desc, i);
return idx;
}
@@ -199,16 +222,17 @@ static void mlx5e_grp_q_update_stats(struct mlx5e_priv *priv)
{
struct mlx5e_qcounter_stats *qcnt = &priv->stats.qcnt;
u32 out[MLX5_ST_SZ_DW(query_q_counter_out)];
- int err;
-
- if (!priv->q_counter)
- return;
-
- err = mlx5_core_query_q_counter(priv->mdev, priv->q_counter, 0, out, sizeof(out));
- if (err)
- return;
- qcnt->rx_out_of_buffer = MLX5_GET(query_q_counter_out, out, out_of_buffer);
+ if (priv->q_counter &&
+ !mlx5_core_query_q_counter(priv->mdev, priv->q_counter, 0, out,
+ sizeof(out)))
+ qcnt->rx_out_of_buffer = MLX5_GET(query_q_counter_out,
+ out, out_of_buffer);
+ if (priv->drop_rq_q_counter &&
+ !mlx5_core_query_q_counter(priv->mdev, priv->drop_rq_q_counter, 0,
+ out, sizeof(out)))
+ qcnt->rx_if_down_packets = MLX5_GET(query_q_counter_out, out,
+ out_of_buffer);
}
#define VNIC_ENV_OFF(c) MLX5_BYTE_OFF(query_vnic_env_out, c)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index 847388ff8ca8..43a72efa28c0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -97,6 +97,7 @@ struct mlx5e_sw_stats {
struct mlx5e_qcounter_stats {
u32 rx_out_of_buffer;
+ u32 rx_if_down_packets;
};
struct mlx5e_vnic_env_stats {
--
2.14.3
^ permalink raw reply related
* [net-next 13/15] net/mlx5e: Make choose LRO timeout function static
From: Saeed Mahameed @ 2018-03-23 22:39 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Gal Pressman, Saeed Mahameed
In-Reply-To: <20180323223925.21678-1-saeedm@mellanox.com>
From: Gal Pressman <galp@mellanox.com>
The function is used in en_main.c only, we can make it static and remove
its declaration from en.h
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en.h | 1 -
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 2 +-
2 files changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 48e0b2a747d9..294bc9f175a5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -1062,7 +1062,6 @@ void mlx5e_cleanup_nic_tx(struct mlx5e_priv *priv);
int mlx5e_close(struct net_device *netdev);
int mlx5e_open(struct net_device *netdev);
void mlx5e_update_stats_work(struct work_struct *work);
-u32 mlx5e_choose_lro_timeout(struct mlx5_core_dev *mdev, u32 wanted_timeout);
int mlx5e_bits_invert(unsigned long a, int size);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index f8bc3bcdf046..eb38a35564c7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -3967,7 +3967,7 @@ void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode)
MLX5_CQ_PERIOD_MODE_START_FROM_CQE);
}
-u32 mlx5e_choose_lro_timeout(struct mlx5_core_dev *mdev, u32 wanted_timeout)
+static u32 mlx5e_choose_lro_timeout(struct mlx5_core_dev *mdev, u32 wanted_timeout)
{
int i;
--
2.14.3
^ permalink raw reply related
* [net-next 03/15] net/mlx5e: PFC stall prevention support
From: Saeed Mahameed @ 2018-03-23 22:39 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Inbar Karmy, Saeed Mahameed
In-Reply-To: <20180323223925.21678-1-saeedm@mellanox.com>
From: Inbar Karmy <inbark@mellanox.com>
Implement set/get functions to configure PFC stall prevention
timeout by tunables api through ethtool.
By default the stall prevention timeout is configured to 8 sec.
Timeout range is: 80-8000 msec.
Enabling stall prevention without a specific timeout will set
the timeout to 100 msec.
Signed-off-by: Inbar Karmy <inbark@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/en_ethtool.c | 57 +++++++++++++++++++
drivers/net/ethernet/mellanox/mlx5/core/port.c | 64 +++++++++++++++++++---
include/linux/mlx5/mlx5_ifc.h | 17 ++++--
include/linux/mlx5/port.h | 6 ++
4 files changed, 132 insertions(+), 12 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index cc8048f68f11..62061fd23143 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -1066,6 +1066,57 @@ static int mlx5e_get_rxnfc(struct net_device *netdev,
return err;
}
+#define MLX5E_PFC_PREVEN_AUTO_TOUT_MSEC 100
+#define MLX5E_PFC_PREVEN_TOUT_MAX_MSEC 8000
+#define MLX5E_PFC_PREVEN_MINOR_PRECENT 85
+#define MLX5E_PFC_PREVEN_TOUT_MIN_MSEC 80
+#define MLX5E_DEVICE_STALL_MINOR_WATERMARK(critical_tout) \
+ max_t(u16, MLX5E_PFC_PREVEN_TOUT_MIN_MSEC, \
+ (critical_tout * MLX5E_PFC_PREVEN_MINOR_PRECENT) / 100)
+
+static int mlx5e_get_pfc_prevention_tout(struct net_device *netdev,
+ u16 *pfc_prevention_tout)
+{
+ struct mlx5e_priv *priv = netdev_priv(netdev);
+ struct mlx5_core_dev *mdev = priv->mdev;
+
+ if (!MLX5_CAP_PCAM_FEATURE((priv)->mdev, pfcc_mask) ||
+ !MLX5_CAP_DEBUG((priv)->mdev, stall_detect))
+ return -EOPNOTSUPP;
+
+ return mlx5_query_port_stall_watermark(mdev, pfc_prevention_tout, NULL);
+}
+
+static int mlx5e_set_pfc_prevention_tout(struct net_device *netdev,
+ u16 pfc_preven)
+{
+ struct mlx5e_priv *priv = netdev_priv(netdev);
+ struct mlx5_core_dev *mdev = priv->mdev;
+ u16 critical_tout;
+ u16 minor;
+
+ if (!MLX5_CAP_PCAM_FEATURE((priv)->mdev, pfcc_mask) ||
+ !MLX5_CAP_DEBUG((priv)->mdev, stall_detect))
+ return -EOPNOTSUPP;
+
+ critical_tout = (pfc_preven == PFC_STORM_PREVENTION_AUTO) ?
+ MLX5E_PFC_PREVEN_AUTO_TOUT_MSEC :
+ pfc_preven;
+
+ if (critical_tout != PFC_STORM_PREVENTION_DISABLE &&
+ (critical_tout > MLX5E_PFC_PREVEN_TOUT_MAX_MSEC ||
+ critical_tout < MLX5E_PFC_PREVEN_TOUT_MIN_MSEC)) {
+ netdev_info(netdev, "%s: pfc prevention tout not in range (%d-%d)\n",
+ __func__, MLX5E_PFC_PREVEN_TOUT_MIN_MSEC,
+ MLX5E_PFC_PREVEN_TOUT_MAX_MSEC);
+ return -EINVAL;
+ }
+
+ minor = MLX5E_DEVICE_STALL_MINOR_WATERMARK(critical_tout);
+ return mlx5_set_port_stall_watermark(mdev, critical_tout,
+ minor);
+}
+
static int mlx5e_get_tunable(struct net_device *dev,
const struct ethtool_tunable *tuna,
void *data)
@@ -1077,6 +1128,9 @@ static int mlx5e_get_tunable(struct net_device *dev,
case ETHTOOL_TX_COPYBREAK:
*(u32 *)data = priv->channels.params.tx_max_inline;
break;
+ case ETHTOOL_PFC_PREVENTION_TOUT:
+ err = mlx5e_get_pfc_prevention_tout(dev, data);
+ break;
default:
err = -EINVAL;
break;
@@ -1118,6 +1172,9 @@ static int mlx5e_set_tunable(struct net_device *dev,
break;
mlx5e_switch_priv_channels(priv, &new_channels, NULL);
+ break;
+ case ETHTOOL_PFC_PREVENTION_TOUT:
+ err = mlx5e_set_pfc_prevention_tout(dev, *(u16 *)data);
break;
default:
err = -EINVAL;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c
index c37d00cd472a..fa9d0760dd36 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/port.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c
@@ -483,6 +483,17 @@ int mlx5_core_query_ib_ppcnt(struct mlx5_core_dev *dev,
}
EXPORT_SYMBOL_GPL(mlx5_core_query_ib_ppcnt);
+static int mlx5_query_pfcc_reg(struct mlx5_core_dev *dev, u32 *out,
+ u32 out_size)
+{
+ u32 in[MLX5_ST_SZ_DW(pfcc_reg)] = {0};
+
+ MLX5_SET(pfcc_reg, in, local_port, 1);
+
+ return mlx5_core_access_reg(dev, in, sizeof(in), out,
+ out_size, MLX5_REG_PFCC, 0, 0);
+}
+
int mlx5_set_port_pause(struct mlx5_core_dev *dev, u32 rx_pause, u32 tx_pause)
{
u32 in[MLX5_ST_SZ_DW(pfcc_reg)] = {0};
@@ -500,13 +511,10 @@ EXPORT_SYMBOL_GPL(mlx5_set_port_pause);
int mlx5_query_port_pause(struct mlx5_core_dev *dev,
u32 *rx_pause, u32 *tx_pause)
{
- u32 in[MLX5_ST_SZ_DW(pfcc_reg)] = {0};
u32 out[MLX5_ST_SZ_DW(pfcc_reg)];
int err;
- MLX5_SET(pfcc_reg, in, local_port, 1);
- err = mlx5_core_access_reg(dev, in, sizeof(in), out,
- sizeof(out), MLX5_REG_PFCC, 0, 0);
+ err = mlx5_query_pfcc_reg(dev, out, sizeof(out));
if (err)
return err;
@@ -520,6 +528,49 @@ int mlx5_query_port_pause(struct mlx5_core_dev *dev,
}
EXPORT_SYMBOL_GPL(mlx5_query_port_pause);
+int mlx5_set_port_stall_watermark(struct mlx5_core_dev *dev,
+ u16 stall_critical_watermark,
+ u16 stall_minor_watermark)
+{
+ u32 in[MLX5_ST_SZ_DW(pfcc_reg)] = {0};
+ u32 out[MLX5_ST_SZ_DW(pfcc_reg)];
+
+ MLX5_SET(pfcc_reg, in, local_port, 1);
+ MLX5_SET(pfcc_reg, in, pptx_mask_n, 1);
+ MLX5_SET(pfcc_reg, in, pprx_mask_n, 1);
+ MLX5_SET(pfcc_reg, in, ppan_mask_n, 1);
+ MLX5_SET(pfcc_reg, in, critical_stall_mask, 1);
+ MLX5_SET(pfcc_reg, in, minor_stall_mask, 1);
+ MLX5_SET(pfcc_reg, in, device_stall_critical_watermark,
+ stall_critical_watermark);
+ MLX5_SET(pfcc_reg, in, device_stall_minor_watermark, stall_minor_watermark);
+
+ return mlx5_core_access_reg(dev, in, sizeof(in), out,
+ sizeof(out), MLX5_REG_PFCC, 0, 1);
+}
+
+int mlx5_query_port_stall_watermark(struct mlx5_core_dev *dev,
+ u16 *stall_critical_watermark,
+ u16 *stall_minor_watermark)
+{
+ u32 out[MLX5_ST_SZ_DW(pfcc_reg)];
+ int err;
+
+ err = mlx5_query_pfcc_reg(dev, out, sizeof(out));
+ if (err)
+ return err;
+
+ if (stall_critical_watermark)
+ *stall_critical_watermark = MLX5_GET(pfcc_reg, out,
+ device_stall_critical_watermark);
+
+ if (stall_minor_watermark)
+ *stall_minor_watermark = MLX5_GET(pfcc_reg, out,
+ device_stall_minor_watermark);
+
+ return 0;
+}
+
int mlx5_set_port_pfc(struct mlx5_core_dev *dev, u8 pfc_en_tx, u8 pfc_en_rx)
{
u32 in[MLX5_ST_SZ_DW(pfcc_reg)] = {0};
@@ -538,13 +589,10 @@ EXPORT_SYMBOL_GPL(mlx5_set_port_pfc);
int mlx5_query_port_pfc(struct mlx5_core_dev *dev, u8 *pfc_en_tx, u8 *pfc_en_rx)
{
- u32 in[MLX5_ST_SZ_DW(pfcc_reg)] = {0};
u32 out[MLX5_ST_SZ_DW(pfcc_reg)];
int err;
- MLX5_SET(pfcc_reg, in, local_port, 1);
- err = mlx5_core_access_reg(dev, in, sizeof(in), out,
- sizeof(out), MLX5_REG_PFCC, 0, 0);
+ err = mlx5_query_pfcc_reg(dev, out, sizeof(out));
if (err)
return err;
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index c7d50eccff9e..f3200a9696d6 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -7833,7 +7833,11 @@ struct mlx5_ifc_pifr_reg_bits {
struct mlx5_ifc_pfcc_reg_bits {
u8 reserved_at_0[0x8];
u8 local_port[0x8];
- u8 reserved_at_10[0x10];
+ u8 reserved_at_10[0xb];
+ u8 ppan_mask_n[0x1];
+ u8 minor_stall_mask[0x1];
+ u8 critical_stall_mask[0x1];
+ u8 reserved_at_1e[0x2];
u8 ppan[0x4];
u8 reserved_at_24[0x4];
@@ -7843,17 +7847,22 @@ struct mlx5_ifc_pfcc_reg_bits {
u8 pptx[0x1];
u8 aptx[0x1];
- u8 reserved_at_42[0x6];
+ u8 pptx_mask_n[0x1];
+ u8 reserved_at_43[0x5];
u8 pfctx[0x8];
u8 reserved_at_50[0x10];
u8 pprx[0x1];
u8 aprx[0x1];
- u8 reserved_at_62[0x6];
+ u8 pprx_mask_n[0x1];
+ u8 reserved_at_63[0x5];
u8 pfcrx[0x8];
u8 reserved_at_70[0x10];
- u8 reserved_at_80[0x80];
+ u8 device_stall_minor_watermark[0x10];
+ u8 device_stall_critical_watermark[0x10];
+
+ u8 reserved_at_a0[0x60];
};
struct mlx5_ifc_pelc_reg_bits {
diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h
index 035f0d4dc9fe..34aed6032f86 100644
--- a/include/linux/mlx5/port.h
+++ b/include/linux/mlx5/port.h
@@ -151,6 +151,12 @@ int mlx5_set_port_pfc(struct mlx5_core_dev *dev, u8 pfc_en_tx, u8 pfc_en_rx);
int mlx5_query_port_pfc(struct mlx5_core_dev *dev, u8 *pfc_en_tx,
u8 *pfc_en_rx);
+int mlx5_set_port_stall_watermark(struct mlx5_core_dev *dev,
+ u16 stall_critical_watermark,
+ u16 stall_minor_watermark);
+int mlx5_query_port_stall_watermark(struct mlx5_core_dev *dev,
+ u16 *stall_critical_watermark, u16 *stall_minor_watermark);
+
int mlx5_max_tc(struct mlx5_core_dev *mdev);
int mlx5_set_port_prio_tc(struct mlx5_core_dev *mdev, u8 *prio_tc);
--
2.14.3
^ permalink raw reply related
* [net-next 06/15] net/mlx5: Add packet dropped while vport down statistics
From: Saeed Mahameed @ 2018-03-23 22:39 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Moshe Shemesh, Saeed Mahameed
In-Reply-To: <20180323223925.21678-1-saeedm@mellanox.com>
From: Moshe Shemesh <moshe@mellanox.com>
Added the following packets dropped while vport down statistics:
Rx dropped while vport down - counts packets which were steered by
e-switch to a vport, but dropped since the vport was down. This counter
will be shown on ip link tool as part of the vport rx_dropped counter.
Tx dropped while vport down - counts packets which were transmitted by
a vport, but dropped due to vport logical link down. This counter
will be shown on ip link tool as part of the vport tx_dropped counter.
The counters are read from FW by command QUERY_VNIC_ENV.
Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 31 +++++++++++++++++++----
drivers/net/ethernet/mellanox/mlx5/core/vport.c | 26 +++++++++++++++++++
include/linux/mlx5/mlx5_ifc.h | 4 ++-
include/linux/mlx5/vport.h | 3 +++
4 files changed, 58 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index 77b7272eaaa8..332bc56306bf 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -2096,17 +2096,19 @@ int mlx5_eswitch_set_vport_rate(struct mlx5_eswitch *esw, int vport,
return err;
}
-static void mlx5_eswitch_query_vport_drop_stats(struct mlx5_core_dev *dev,
- int vport_idx,
- struct mlx5_vport_drop_stats *stats)
+static int mlx5_eswitch_query_vport_drop_stats(struct mlx5_core_dev *dev,
+ int vport_idx,
+ struct mlx5_vport_drop_stats *stats)
{
struct mlx5_eswitch *esw = dev->priv.eswitch;
struct mlx5_vport *vport = &esw->vports[vport_idx];
+ u64 rx_discard_vport_down, tx_discard_vport_down;
u64 bytes = 0;
u16 idx = 0;
+ int err = 0;
if (!vport->enabled || esw->mode != SRIOV_LEGACY)
- return;
+ return 0;
if (vport->egress.drop_counter) {
idx = vport->egress.drop_counter->id;
@@ -2117,6 +2119,23 @@ static void mlx5_eswitch_query_vport_drop_stats(struct mlx5_core_dev *dev,
idx = vport->ingress.drop_counter->id;
mlx5_fc_query(dev, idx, &stats->tx_dropped, &bytes);
}
+
+ if (!MLX5_CAP_GEN(dev, receive_discard_vport_down) &&
+ !MLX5_CAP_GEN(dev, transmit_discard_vport_down))
+ return 0;
+
+ err = mlx5_query_vport_down_stats(dev, vport_idx,
+ &rx_discard_vport_down,
+ &tx_discard_vport_down);
+ if (err)
+ return err;
+
+ if (MLX5_CAP_GEN(dev, receive_discard_vport_down))
+ stats->rx_dropped += rx_discard_vport_down;
+ if (MLX5_CAP_GEN(dev, transmit_discard_vport_down))
+ stats->tx_dropped += tx_discard_vport_down;
+
+ return 0;
}
int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw,
@@ -2180,7 +2199,9 @@ int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw,
vf_stats->broadcast =
MLX5_GET_CTR(out, received_eth_broadcast.packets);
- mlx5_eswitch_query_vport_drop_stats(esw->dev, vport, &stats);
+ err = mlx5_eswitch_query_vport_drop_stats(esw->dev, vport, &stats);
+ if (err)
+ goto free_out;
vf_stats->rx_dropped = stats.rx_dropped;
vf_stats->tx_dropped = stats.tx_dropped;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c
index dfe36cf6fbea..177e076b8d17 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c
@@ -1070,6 +1070,32 @@ int mlx5_core_query_vport_counter(struct mlx5_core_dev *dev, u8 other_vport,
}
EXPORT_SYMBOL_GPL(mlx5_core_query_vport_counter);
+int mlx5_query_vport_down_stats(struct mlx5_core_dev *mdev, u16 vport,
+ u64 *rx_discard_vport_down,
+ u64 *tx_discard_vport_down)
+{
+ u32 out[MLX5_ST_SZ_DW(query_vnic_env_out)] = {0};
+ u32 in[MLX5_ST_SZ_DW(query_vnic_env_in)] = {0};
+ int err;
+
+ MLX5_SET(query_vnic_env_in, in, opcode,
+ MLX5_CMD_OP_QUERY_VNIC_ENV);
+ MLX5_SET(query_vnic_env_in, in, op_mod, 0);
+ MLX5_SET(query_vnic_env_in, in, vport_number, vport);
+ if (vport)
+ MLX5_SET(query_vnic_env_in, in, other_vport, 1);
+
+ err = mlx5_cmd_exec(mdev, in, sizeof(in), out, sizeof(out));
+ if (err)
+ return err;
+
+ *rx_discard_vport_down = MLX5_GET64(query_vnic_env_out, out,
+ vport_env.receive_discard_vport_down);
+ *tx_discard_vport_down = MLX5_GET64(query_vnic_env_out, out,
+ vport_env.transmit_discard_vport_down);
+ return 0;
+}
+
int mlx5_core_modify_hca_vport_context(struct mlx5_core_dev *dev,
u8 other_vport, u8 port_num,
int vf,
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 9202113f552c..1f3483d40055 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -1009,7 +1009,9 @@ struct mlx5_ifc_cmd_hca_cap_bits {
u8 log_max_xrcd[0x5];
u8 nic_receive_steering_discard[0x1];
- u8 reserved_at_341[0x7];
+ u8 receive_discard_vport_down[0x1];
+ u8 transmit_discard_vport_down[0x1];
+ u8 reserved_at_343[0x5];
u8 log_max_flow_counter_bulk[0x8];
u8 max_flow_counter_15_0[0x10];
diff --git a/include/linux/mlx5/vport.h b/include/linux/mlx5/vport.h
index 64e193e87394..9208cb8809ac 100644
--- a/include/linux/mlx5/vport.h
+++ b/include/linux/mlx5/vport.h
@@ -107,6 +107,9 @@ int mlx5_modify_nic_vport_vlans(struct mlx5_core_dev *dev,
int mlx5_nic_vport_enable_roce(struct mlx5_core_dev *mdev);
int mlx5_nic_vport_disable_roce(struct mlx5_core_dev *mdev);
+int mlx5_query_vport_down_stats(struct mlx5_core_dev *mdev, u16 vport,
+ u64 *rx_discard_vport_down,
+ u64 *tx_discard_vport_down);
int mlx5_core_query_vport_counter(struct mlx5_core_dev *dev, u8 other_vport,
int vf, u8 port_num, void *out,
size_t out_sz);
--
2.14.3
^ permalink raw reply related
* [net-next 11/15] net/mlx5: Protect from command bit overflow
From: Saeed Mahameed @ 2018-03-23 22:39 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Leon Romanovsky, Saeed Mahameed
In-Reply-To: <20180323223925.21678-1-saeedm@mellanox.com>
From: Leon Romanovsky <leonro@mellanox.com>
The system with CONFIG_UBSAN enabled on produces the following error
during driver initialization. The reason to it that max_reg_cmds can be
larger enough to cause to "1 << max_reg_cmds" overflow the unsigned long.
================================================================================
UBSAN: Undefined behaviour in drivers/net/ethernet/mellanox/mlx5/core/cmd.c:1805:42
signed integer overflow:
-2147483648 - 1 cannot be represented in type 'int'
CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.15.0-rc2-00032-g06cda2358d9b-dirty #724
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.7.5-0-ge51488c-20140602_164612-nilsson.home.kraxel.org 04/01/2014
Call Trace:
dump_stack+0xe9/0x18f
? dma_virt_alloc+0x81/0x81
ubsan_epilogue+0xe/0x4e
handle_overflow+0x187/0x20c
mlx5_cmd_init+0x73a/0x12b0
mlx5_load_one+0x1c3d/0x1d30
init_one+0xd02/0xf10
pci_device_probe+0x26c/0x3b0
driver_probe_device+0x622/0xb40
__driver_attach+0x175/0x1b0
bus_for_each_dev+0xef/0x190
bus_add_driver+0x2db/0x490
driver_register+0x16b/0x1e0
__pci_register_driver+0x177/0x1b0
init+0x6d/0x92
do_one_initcall+0x15b/0x270
kernel_init_freeable+0x2d8/0x3d0
kernel_init+0x14/0x190
ret_from_fork+0x24/0x30
================================================================================
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
index fe5428667ad1..21cd1703a862 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
@@ -1804,7 +1804,7 @@ int mlx5_cmd_init(struct mlx5_core_dev *dev)
cmd->checksum_disabled = 1;
cmd->max_reg_cmds = (1 << cmd->log_sz) - 1;
- cmd->bitmask = (1 << cmd->max_reg_cmds) - 1;
+ cmd->bitmask = (1UL << cmd->max_reg_cmds) - 1;
cmd->cmdif_rev = ioread32be(&dev->iseg->cmdif_rev_fw_sub) >> 16;
if (cmd->cmdif_rev > CMD_IF_REV) {
--
2.14.3
^ permalink raw reply related
* [net-next 12/15] net/mlx5e: Remove redundant check in get ethtool stats
From: Saeed Mahameed @ 2018-03-23 22:39 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Gal Pressman, Saeed Mahameed
In-Reply-To: <20180323223925.21678-1-saeedm@mellanox.com>
From: Gal Pressman <galp@mellanox.com>
ethtool core code makes sure data isn't NULL before calling
get_ethtool_stats, testing it again in the driver is redundant.
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index 62061fd23143..d415e67b557b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -203,9 +203,6 @@ void mlx5e_ethtool_get_ethtool_stats(struct mlx5e_priv *priv,
{
int i, idx = 0;
- if (!data)
- return;
-
mutex_lock(&priv->state_lock);
mlx5e_update_stats(priv);
mutex_unlock(&priv->state_lock);
--
2.14.3
^ permalink raw reply related
* [net-next 05/15] net/mlx5e: Add vnic steering drop statistics
From: Saeed Mahameed @ 2018-03-23 22:39 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Moshe Shemesh, Saeed Mahameed
In-Reply-To: <20180323223925.21678-1-saeedm@mellanox.com>
From: Moshe Shemesh <moshe@mellanox.com>
Added the following packets drop counter:
Rx steering missed dropped packets - counts packets which were dropped
due to miss on NIC rx steering rules.
This counter will be shown on ethtool as a new counter called
rx_steer_missed_packets.
Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 65 ++++++++++++++++++++++
drivers/net/ethernet/mellanox/mlx5/core/en_stats.h | 5 ++
include/linux/mlx5/mlx5_ifc.h | 3 +-
3 files changed, 72 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index 2553c58dcf1c..552510c03ef2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -211,6 +211,65 @@ static void mlx5e_grp_q_update_stats(struct mlx5e_priv *priv)
qcnt->rx_out_of_buffer = MLX5_GET(query_q_counter_out, out, out_of_buffer);
}
+#define VNIC_ENV_OFF(c) MLX5_BYTE_OFF(query_vnic_env_out, c)
+static const struct counter_desc vnic_env_stats_desc[] = {
+ { "rx_steer_missed_packets",
+ VNIC_ENV_OFF(vport_env.nic_receive_steering_discard) },
+};
+
+#define NUM_VNIC_ENV_COUNTERS ARRAY_SIZE(vnic_env_stats_desc)
+
+static int mlx5e_grp_vnic_env_get_num_stats(struct mlx5e_priv *priv)
+{
+ return MLX5_CAP_GEN(priv->mdev, nic_receive_steering_discard) ?
+ NUM_VNIC_ENV_COUNTERS : 0;
+}
+
+static int mlx5e_grp_vnic_env_fill_strings(struct mlx5e_priv *priv, u8 *data,
+ int idx)
+{
+ int i;
+
+ if (!MLX5_CAP_GEN(priv->mdev, nic_receive_steering_discard))
+ return idx;
+
+ for (i = 0; i < NUM_VNIC_ENV_COUNTERS; i++)
+ strcpy(data + (idx++) * ETH_GSTRING_LEN,
+ vnic_env_stats_desc[i].format);
+ return idx;
+}
+
+static int mlx5e_grp_vnic_env_fill_stats(struct mlx5e_priv *priv, u64 *data,
+ int idx)
+{
+ int i;
+
+ if (!MLX5_CAP_GEN(priv->mdev, nic_receive_steering_discard))
+ return idx;
+
+ for (i = 0; i < NUM_VNIC_ENV_COUNTERS; i++)
+ data[idx++] = MLX5E_READ_CTR64_BE(priv->stats.vnic.query_vnic_env_out,
+ vnic_env_stats_desc, i);
+ return idx;
+}
+
+static void mlx5e_grp_vnic_env_update_stats(struct mlx5e_priv *priv)
+{
+ u32 *out = (u32 *)priv->stats.vnic.query_vnic_env_out;
+ int outlen = MLX5_ST_SZ_BYTES(query_vnic_env_out);
+ u32 in[MLX5_ST_SZ_DW(query_vnic_env_in)] = {0};
+ struct mlx5_core_dev *mdev = priv->mdev;
+
+ if (!MLX5_CAP_GEN(priv->mdev, nic_receive_steering_discard))
+ return;
+
+ MLX5_SET(query_vnic_env_in, in, opcode,
+ MLX5_CMD_OP_QUERY_VNIC_ENV);
+ MLX5_SET(query_vnic_env_in, in, op_mod, 0);
+ MLX5_SET(query_vnic_env_in, in, other_vport, 0);
+ mlx5_cmd_exec(mdev, in, sizeof(in), out, outlen);
+}
+
#define VPORT_COUNTER_OFF(c) MLX5_BYTE_OFF(query_vport_counter_out, c)
static const struct counter_desc vport_stats_desc[] = {
{ "rx_vport_unicast_packets",
@@ -1111,6 +1170,12 @@ const struct mlx5e_stats_grp mlx5e_stats_grps[] = {
.update_stats_mask = MLX5E_NDO_UPDATE_STATS,
.update_stats = mlx5e_grp_q_update_stats,
},
+ {
+ .get_num_stats = mlx5e_grp_vnic_env_get_num_stats,
+ .fill_strings = mlx5e_grp_vnic_env_fill_strings,
+ .fill_stats = mlx5e_grp_vnic_env_fill_stats,
+ .update_stats = mlx5e_grp_vnic_env_update_stats,
+ },
{
.get_num_stats = mlx5e_grp_vport_get_num_stats,
.fill_strings = mlx5e_grp_vport_fill_strings,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index 0b3320a2b072..847388ff8ca8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -99,6 +99,10 @@ struct mlx5e_qcounter_stats {
u32 rx_out_of_buffer;
};
+struct mlx5e_vnic_env_stats {
+ __be64 query_vnic_env_out[MLX5_ST_SZ_QW(query_vnic_env_out)];
+};
+
#define VPORT_COUNTER_GET(vstats, c) MLX5_GET64(query_vport_counter_out, \
vstats->query_vport_out, c)
@@ -201,6 +205,7 @@ struct mlx5e_ch_stats {
struct mlx5e_stats {
struct mlx5e_sw_stats sw;
struct mlx5e_qcounter_stats qcnt;
+ struct mlx5e_vnic_env_stats vnic;
struct mlx5e_vport_stats vport;
struct mlx5e_pport_stats pport;
struct rtnl_link_stats64 vf_vport;
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 52e373dd2679..9202113f552c 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -1008,7 +1008,8 @@ struct mlx5_ifc_cmd_hca_cap_bits {
u8 reserved_at_330[0xb];
u8 log_max_xrcd[0x5];
- u8 reserved_at_340[0x8];
+ u8 nic_receive_steering_discard[0x1];
+ u8 reserved_at_341[0x7];
u8 log_max_flow_counter_bulk[0x8];
u8 max_flow_counter_15_0[0x10];
--
2.14.3
^ permalink raw reply related
* [net-next 14/15] net/mlx5e: Add a helper macro in set features ndo
From: Saeed Mahameed @ 2018-03-23 22:39 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Gal Pressman, Saeed Mahameed
In-Reply-To: <20180323223925.21678-1-saeedm@mellanox.com>
From: Gal Pressman <galp@mellanox.com>
Add a new macro to prevent copy-pasting the same code for each new
feature.
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 26 ++++++++++-------------
1 file changed, 11 insertions(+), 15 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index eb38a35564c7..fb936a978f20 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -3242,24 +3242,20 @@ static int mlx5e_set_features(struct net_device *netdev,
netdev_features_t features)
{
netdev_features_t oper_features = netdev->features;
- int err;
+ int err = 0;
+
+#define MLX5E_HANDLE_FEATURE(feature, handler) \
+ mlx5e_handle_feature(netdev, &oper_features, features, feature, handler)
- err = mlx5e_handle_feature(netdev, &oper_features, features,
- NETIF_F_LRO, set_feature_lro);
- err |= mlx5e_handle_feature(netdev, &oper_features, features,
- NETIF_F_HW_VLAN_CTAG_FILTER,
+ err |= MLX5E_HANDLE_FEATURE(NETIF_F_LRO, set_feature_lro);
+ err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_VLAN_CTAG_FILTER,
set_feature_cvlan_filter);
- err |= mlx5e_handle_feature(netdev, &oper_features, features,
- NETIF_F_HW_TC, set_feature_tc_num_filters);
- err |= mlx5e_handle_feature(netdev, &oper_features, features,
- NETIF_F_RXALL, set_feature_rx_all);
- err |= mlx5e_handle_feature(netdev, &oper_features, features,
- NETIF_F_RXFCS, set_feature_rx_fcs);
- err |= mlx5e_handle_feature(netdev, &oper_features, features,
- NETIF_F_HW_VLAN_CTAG_RX, set_feature_rx_vlan);
+ err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_TC, set_feature_tc_num_filters);
+ err |= MLX5E_HANDLE_FEATURE(NETIF_F_RXALL, set_feature_rx_all);
+ err |= MLX5E_HANDLE_FEATURE(NETIF_F_RXFCS, set_feature_rx_fcs);
+ err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_VLAN_CTAG_RX, set_feature_rx_vlan);
#ifdef CONFIG_RFS_ACCEL
- err |= mlx5e_handle_feature(netdev, &oper_features, features,
- NETIF_F_NTUPLE, set_feature_arfs);
+ err |= MLX5E_HANDLE_FEATURE(NETIF_F_NTUPLE, set_feature_arfs);
#endif
if (err) {
--
2.14.3
^ permalink raw reply related
* [net-next 10/15] net/mlx5e: Offload tc vlan push/pop using HW action
From: Saeed Mahameed @ 2018-03-23 22:39 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Or Gerlitz, Saeed Mahameed
In-Reply-To: <20180323223925.21678-1-saeedm@mellanox.com>
From: Or Gerlitz <ogerlitz@mellanox.com>
Currently, we are emulating the offload of vlan push/pop actions using
global setup as done by commit f5f82476090f ("net/mlx5: E-Switch, Support
VLAN actions in the offloads mode"). With newer NICs, we can apply a flow
action for that matter, do that while keeping the emulated path for the
older HW brands.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 15 ++++++++-----
drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 10 ++++++++-
.../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 26 +++++++++++++++++-----
3 files changed, 40 insertions(+), 11 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 7c33df2034f0..3e4a7e81b67f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -2530,12 +2530,17 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, struct tcf_exts *exts,
if (tcf_vlan_action(a) == TCA_VLAN_ACT_POP) {
attr->action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_POP;
} else if (tcf_vlan_action(a) == TCA_VLAN_ACT_PUSH) {
- if (tcf_vlan_push_proto(a) != htons(ETH_P_8021Q) ||
- tcf_vlan_push_prio(a))
- return -EOPNOTSUPP;
-
attr->action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH;
- attr->vlan = tcf_vlan_push_vid(a);
+ attr->vlan_vid = tcf_vlan_push_vid(a);
+ if (mlx5_eswitch_vlan_actions_supported(priv->mdev)) {
+ attr->vlan_prio = tcf_vlan_push_prio(a);
+ attr->vlan_proto = tcf_vlan_push_proto(a);
+ if (!attr->vlan_proto)
+ attr->vlan_proto = htons(ETH_P_8021Q);
+ } else if (tcf_vlan_push_proto(a) != htons(ETH_P_8021Q) ||
+ tcf_vlan_push_prio(a)) {
+ return -EOPNOTSUPP;
+ }
} else { /* action is TCA_VLAN_ACT_MODIFY */
return -EOPNOTSUPP;
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index a435eb7971c6..4cd773fa55e3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -232,7 +232,9 @@ struct mlx5_esw_flow_attr {
struct mlx5_eswitch_rep *out_rep;
int action;
- u16 vlan;
+ __be16 vlan_proto;
+ u16 vlan_vid;
+ u8 vlan_prio;
bool vlan_handled;
u32 encap_id;
u32 mod_hdr_id;
@@ -255,6 +257,12 @@ int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw,
int __mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw,
int vport, u16 vlan, u8 qos, u8 set_flags);
+static inline bool mlx5_eswitch_vlan_actions_supported(struct mlx5_core_dev *dev)
+{
+ return MLX5_CAP_ESW_FLOWTABLE_FDB(dev, pop_vlan) &&
+ MLX5_CAP_ESW_FLOWTABLE_FDB(dev, push_vlan);
+}
+
#define MLX5_DEBUG_ESWITCH_MASK BIT(3)
#define esw_info(dev, format, ...) \
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 21ebe3e80e6e..35e256eb2f6e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -58,8 +58,16 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw,
if (esw->mode != SRIOV_OFFLOADS)
return ERR_PTR(-EOPNOTSUPP);
- /* per flow vlan pop/push is emulated, don't set that into the firmware */
- flow_act.action = attr->action & ~(MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH | MLX5_FLOW_CONTEXT_ACTION_VLAN_POP);
+ flow_act.action = attr->action;
+ /* if per flow vlan pop/push is emulated, don't set that into the firmware */
+ if (!mlx5_eswitch_vlan_actions_supported(esw->dev))
+ flow_act.action &= ~(MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH |
+ MLX5_FLOW_CONTEXT_ACTION_VLAN_POP);
+ else if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH) {
+ flow_act.vlan.ethtype = ntohs(attr->vlan_proto);
+ flow_act.vlan.vid = attr->vlan_vid;
+ flow_act.vlan.prio = attr->vlan_prio;
+ }
if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) {
dest[i].type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
@@ -185,7 +193,7 @@ static int esw_add_vlan_action_check(struct mlx5_esw_flow_attr *attr,
/* protects against (1) setting rules with different vlans to push and
* (2) setting rules w.o vlans (attr->vlan = 0) && w. vlans to push (!= 0)
*/
- if (push && in_rep->vlan_refcount && (in_rep->vlan != attr->vlan))
+ if (push && in_rep->vlan_refcount && (in_rep->vlan != attr->vlan_vid))
goto out_notsupp;
return 0;
@@ -202,6 +210,10 @@ int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw,
bool push, pop, fwd;
int err = 0;
+ /* nop if we're on the vlan push/pop non emulation mode */
+ if (mlx5_eswitch_vlan_actions_supported(esw->dev))
+ return 0;
+
push = !!(attr->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH);
pop = !!(attr->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP);
fwd = !!(attr->action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST);
@@ -239,11 +251,11 @@ int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw,
if (vport->vlan_refcount)
goto skip_set_push;
- err = __mlx5_eswitch_set_vport_vlan(esw, vport->vport, attr->vlan, 0,
+ err = __mlx5_eswitch_set_vport_vlan(esw, vport->vport, attr->vlan_vid, 0,
SET_VLAN_INSERT | SET_VLAN_STRIP);
if (err)
goto out;
- vport->vlan = attr->vlan;
+ vport->vlan = attr->vlan_vid;
skip_set_push:
vport->vlan_refcount++;
}
@@ -261,6 +273,10 @@ int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw,
bool push, pop, fwd;
int err = 0;
+ /* nop if we're on the vlan push/pop non emulation mode */
+ if (mlx5_eswitch_vlan_actions_supported(esw->dev))
+ return 0;
+
if (!attr->vlan_handled)
return 0;
--
2.14.3
^ permalink raw reply related
* [net-next 09/15] net/mlx5: Add core support for vlan push/pop steering action
From: Saeed Mahameed @ 2018-03-23 22:39 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Or Gerlitz, Saeed Mahameed
In-Reply-To: <20180323223925.21678-1-saeedm@mellanox.com>
From: Or Gerlitz <ogerlitz@mellanox.com>
Newer NICs (ConnectX-5 and onward) can apply vlan pop or push as an
action taking place during flow steering. Add the core bits for that.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.h | 2 ++
drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 3 ---
drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c | 10 +++++++++-
drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 4 +++-
include/linux/mlx5/fs.h | 7 +++++++
include/linux/mlx5/mlx5_ifc.h | 16 ++++++++++++++--
6 files changed, 35 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.h
index a6ba57fbb414..09f178a3fcab 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.h
@@ -136,6 +136,8 @@ TRACE_EVENT(mlx5_fs_del_fg,
{MLX5_FLOW_CONTEXT_ACTION_ENCAP, "ENCAP"},\
{MLX5_FLOW_CONTEXT_ACTION_DECAP, "DECAP"},\
{MLX5_FLOW_CONTEXT_ACTION_MOD_HDR, "MOD_HDR"},\
+ {MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH, "VLAN_PUSH"},\
+ {MLX5_FLOW_CONTEXT_ACTION_VLAN_POP, "VLAN_POP"},\
{MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_PRIO, "NEXT_PRIO"}
TRACE_EVENT(mlx5_fs_set_fte,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 98d2177d0806..a435eb7971c6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -227,9 +227,6 @@ enum {
SET_VLAN_INSERT = BIT(1)
};
-#define MLX5_FLOW_CONTEXT_ACTION_VLAN_POP 0x4000
-#define MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH 0x8000
-
struct mlx5_esw_flow_attr {
struct mlx5_eswitch_rep *in_rep;
struct mlx5_eswitch_rep *out_rep;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
index 645f83cac34d..ef5afd7c9325 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c
@@ -317,7 +317,7 @@ static int mlx5_cmd_set_fte(struct mlx5_core_dev *dev,
fte->dests_size * MLX5_ST_SZ_BYTES(dest_format_struct);
u32 out[MLX5_ST_SZ_DW(set_fte_out)] = {0};
struct mlx5_flow_rule *dst;
- void *in_flow_context;
+ void *in_flow_context, *vlan;
void *in_match_value;
void *in_dests;
u32 *in;
@@ -340,11 +340,19 @@ static int mlx5_cmd_set_fte(struct mlx5_core_dev *dev,
in_flow_context = MLX5_ADDR_OF(set_fte_in, in, flow_context);
MLX5_SET(flow_context, in_flow_context, group_id, group_id);
+
MLX5_SET(flow_context, in_flow_context, flow_tag, fte->action.flow_tag);
MLX5_SET(flow_context, in_flow_context, action, fte->action.action);
MLX5_SET(flow_context, in_flow_context, encap_id, fte->action.encap_id);
MLX5_SET(flow_context, in_flow_context, modify_header_id,
fte->action.modify_id);
+
+ vlan = MLX5_ADDR_OF(flow_context, in_flow_context, push_vlan);
+
+ MLX5_SET(vlan, vlan, ethtype, fte->action.vlan.ethtype);
+ MLX5_SET(vlan, vlan, vid, fte->action.vlan.vid);
+ MLX5_SET(vlan, vlan, prio, fte->action.vlan.prio);
+
in_match_value = MLX5_ADDR_OF(flow_context, in_flow_context,
match_value);
memcpy(in_match_value, &fte->val, sizeof(fte->val));
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index 3ba07c7096ef..de51e7c39bc8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -1439,7 +1439,9 @@ static bool check_conflicting_actions(u32 action1, u32 action2)
if (xored_actions & (MLX5_FLOW_CONTEXT_ACTION_DROP |
MLX5_FLOW_CONTEXT_ACTION_ENCAP |
MLX5_FLOW_CONTEXT_ACTION_DECAP |
- MLX5_FLOW_CONTEXT_ACTION_MOD_HDR))
+ MLX5_FLOW_CONTEXT_ACTION_MOD_HDR |
+ MLX5_FLOW_CONTEXT_ACTION_VLAN_POP |
+ MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH))
return true;
return false;
diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h
index b957e52434f8..47aecc4fa8c2 100644
--- a/include/linux/mlx5/fs.h
+++ b/include/linux/mlx5/fs.h
@@ -142,6 +142,12 @@ struct mlx5_flow_group *
mlx5_create_flow_group(struct mlx5_flow_table *ft, u32 *in);
void mlx5_destroy_flow_group(struct mlx5_flow_group *fg);
+struct mlx5_fs_vlan {
+ u16 ethtype;
+ u16 vid;
+ u8 prio;
+};
+
struct mlx5_flow_act {
u32 action;
bool has_flow_tag;
@@ -149,6 +155,7 @@ struct mlx5_flow_act {
u32 encap_id;
u32 modify_id;
uintptr_t esp_id;
+ struct mlx5_fs_vlan vlan;
};
#define MLX5_DECLARE_FLOW_ACT(name) \
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 1f3483d40055..c19e611d2782 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -314,7 +314,10 @@ struct mlx5_ifc_flow_table_prop_layout_bits {
u8 flow_table_modify[0x1];
u8 encap[0x1];
u8 decap[0x1];
- u8 reserved_at_9[0x17];
+ u8 reserved_at_9[0x1];
+ u8 pop_vlan[0x1];
+ u8 push_vlan[0x1];
+ u8 reserved_at_c[0x14];
u8 reserved_at_20[0x2];
u8 log_max_ft_size[0x6];
@@ -2311,10 +2314,19 @@ enum {
MLX5_FLOW_CONTEXT_ACTION_ENCAP = 0x10,
MLX5_FLOW_CONTEXT_ACTION_DECAP = 0x20,
MLX5_FLOW_CONTEXT_ACTION_MOD_HDR = 0x40,
+ MLX5_FLOW_CONTEXT_ACTION_VLAN_POP = 0x80,
+ MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH = 0x100,
+};
+
+struct mlx5_ifc_vlan_bits {
+ u8 ethtype[0x10];
+ u8 prio[0x3];
+ u8 cfi[0x1];
+ u8 vid[0xc];
};
struct mlx5_ifc_flow_context_bits {
- u8 reserved_at_0[0x20];
+ struct mlx5_ifc_vlan_bits push_vlan;
u8 group_id[0x20];
--
2.14.3
^ permalink raw reply related
* [net-next 15/15] net/mlx5e: Add VLAN offload features to hw_enc_features
From: Saeed Mahameed @ 2018-03-23 22:39 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Aviv Heller, Saeed Mahameed
In-Reply-To: <20180323223925.21678-1-saeedm@mellanox.com>
From: Aviv Heller <avivh@mellanox.com>
We support outer VLAN offload in driver and HW regardless of whether
an encapsulation is present in the next headers.
Exposing this in hw_enc_features will allow us to offload outer VLANs
in cases where encapsulation protocols like VXLAN and IPsec are used.
Signed-off-by: Aviv Heller <avivh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index fb936a978f20..1d36d7569f44 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -4106,6 +4106,9 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
netdev->vlan_features |= NETIF_F_RXCSUM;
netdev->vlan_features |= NETIF_F_RXHASH;
+ netdev->hw_enc_features |= NETIF_F_HW_VLAN_CTAG_TX;
+ netdev->hw_enc_features |= NETIF_F_HW_VLAN_CTAG_RX;
+
if (!!MLX5_CAP_ETH(mdev, lro_cap))
netdev->vlan_features |= NETIF_F_LRO;
--
2.14.3
^ 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