* [net 1/3] net/mlx5: Free IRQs in shutdown path
2018-05-10 23:19 [pull request][net 0/3] Mellanox, mlx5 fixes 2018-05-10 Saeed Mahameed
@ 2018-05-10 23:19 ` Saeed Mahameed
2018-05-10 23:19 ` [net 2/3] net/mlx5: E-Switch, Include VF RDMA stats in vport statistics Saeed Mahameed
` (2 subsequent siblings)
3 siblings, 0 replies; 5+ messages in thread
From: Saeed Mahameed @ 2018-05-10 23:19 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Daniel Jurgens, Saeed Mahameed
From: Daniel Jurgens <danielj@mellanox.com>
Some platforms require IRQs to be free'd in the shutdown path. Otherwise
they will fail to be reallocated after a kexec.
Fixes: 8812c24d28f4 ("net/mlx5: Add fast unload support in shutdown flow")
Signed-off-by: Daniel Jurgens <danielj@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/eq.c | 28 ++++++++++++++++++++++
drivers/net/ethernet/mellanox/mlx5/core/main.c | 8 +++++++
.../net/ethernet/mellanox/mlx5/core/mlx5_core.h | 2 ++
3 files changed, 38 insertions(+)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index c1c94974e16b..1814f803bd2c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -34,6 +34,9 @@
#include <linux/module.h>
#include <linux/mlx5/driver.h>
#include <linux/mlx5/cmd.h>
+#ifdef CONFIG_RFS_ACCEL
+#include <linux/cpu_rmap.h>
+#endif
#include "mlx5_core.h"
#include "fpga/core.h"
#include "eswitch.h"
@@ -923,3 +926,28 @@ int mlx5_core_eq_query(struct mlx5_core_dev *dev, struct mlx5_eq *eq,
MLX5_SET(query_eq_in, in, eq_number, eq->eqn);
return mlx5_cmd_exec(dev, in, sizeof(in), out, outlen);
}
+
+/* This function should only be called after mlx5_cmd_force_teardown_hca */
+void mlx5_core_eq_free_irqs(struct mlx5_core_dev *dev)
+{
+ struct mlx5_eq_table *table = &dev->priv.eq_table;
+ struct mlx5_eq *eq;
+
+#ifdef CONFIG_RFS_ACCEL
+ if (dev->rmap) {
+ free_irq_cpu_rmap(dev->rmap);
+ dev->rmap = NULL;
+ }
+#endif
+ list_for_each_entry(eq, &table->comp_eqs_list, list)
+ free_irq(eq->irqn, eq);
+
+ free_irq(table->pages_eq.irqn, &table->pages_eq);
+ free_irq(table->async_eq.irqn, &table->async_eq);
+ free_irq(table->cmd_eq.irqn, &table->cmd_eq);
+#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING
+ if (MLX5_CAP_GEN(dev, pg))
+ free_irq(table->pfault_eq.irqn, &table->pfault_eq);
+#endif
+ pci_free_irq_vectors(dev->pdev);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 63a8ea31601c..e2c465b0b3f8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -1587,6 +1587,14 @@ static int mlx5_try_fast_unload(struct mlx5_core_dev *dev)
mlx5_enter_error_state(dev, true);
+ /* Some platforms requiring freeing the IRQ's in the shutdown
+ * flow. If they aren't freed they can't be allocated after
+ * kexec. There is no need to cleanup the mlx5_core software
+ * contexts.
+ */
+ mlx5_irq_clear_affinity_hints(dev);
+ mlx5_core_eq_free_irqs(dev);
+
return 0;
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index 7d001fe6e631..023882d9a22e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -128,6 +128,8 @@ int mlx5_core_eq_query(struct mlx5_core_dev *dev, struct mlx5_eq *eq,
u32 *out, int outlen);
int mlx5_start_eqs(struct mlx5_core_dev *dev);
void mlx5_stop_eqs(struct mlx5_core_dev *dev);
+/* This function should only be called after mlx5_cmd_force_teardown_hca */
+void mlx5_core_eq_free_irqs(struct mlx5_core_dev *dev);
struct mlx5_eq *mlx5_eqn2eq(struct mlx5_core_dev *dev, int eqn);
u32 mlx5_eq_poll_irq_disabled(struct mlx5_eq *eq);
void mlx5_cq_tasklet_cb(unsigned long data);
--
2.14.3
^ permalink raw reply related [flat|nested] 5+ messages in thread* [net 2/3] net/mlx5: E-Switch, Include VF RDMA stats in vport statistics
2018-05-10 23:19 [pull request][net 0/3] Mellanox, mlx5 fixes 2018-05-10 Saeed Mahameed
2018-05-10 23:19 ` [net 1/3] net/mlx5: Free IRQs in shutdown path Saeed Mahameed
@ 2018-05-10 23:19 ` Saeed Mahameed
2018-05-10 23:19 ` [net 3/3] net/mlx5e: Err if asked to offload TC match on frag being first Saeed Mahameed
2018-05-11 16:27 ` [pull request][net 0/3] Mellanox, mlx5 fixes 2018-05-10 David Miller
3 siblings, 0 replies; 5+ messages in thread
From: Saeed Mahameed @ 2018-05-10 23:19 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Adi Nissim, Saeed Mahameed
From: Adi Nissim <adin@mellanox.com>
The host side reporting of VF vport statistics didn't include the VF
RDMA traffic.
Fixes: 3b751a2a418a ("net/mlx5: E-Switch, Introduce get vf statistics")
Signed-off-by: Adi Nissim <adin@mellanox.com>
Reported-by: Ariel Almog <ariela@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 11 ++++++++++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index 332bc56306bf..1352d13eedb3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -2175,26 +2175,35 @@ int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw,
memset(vf_stats, 0, sizeof(*vf_stats));
vf_stats->rx_packets =
MLX5_GET_CTR(out, received_eth_unicast.packets) +
+ MLX5_GET_CTR(out, received_ib_unicast.packets) +
MLX5_GET_CTR(out, received_eth_multicast.packets) +
+ MLX5_GET_CTR(out, received_ib_multicast.packets) +
MLX5_GET_CTR(out, received_eth_broadcast.packets);
vf_stats->rx_bytes =
MLX5_GET_CTR(out, received_eth_unicast.octets) +
+ MLX5_GET_CTR(out, received_ib_unicast.octets) +
MLX5_GET_CTR(out, received_eth_multicast.octets) +
+ MLX5_GET_CTR(out, received_ib_multicast.octets) +
MLX5_GET_CTR(out, received_eth_broadcast.octets);
vf_stats->tx_packets =
MLX5_GET_CTR(out, transmitted_eth_unicast.packets) +
+ MLX5_GET_CTR(out, transmitted_ib_unicast.packets) +
MLX5_GET_CTR(out, transmitted_eth_multicast.packets) +
+ MLX5_GET_CTR(out, transmitted_ib_multicast.packets) +
MLX5_GET_CTR(out, transmitted_eth_broadcast.packets);
vf_stats->tx_bytes =
MLX5_GET_CTR(out, transmitted_eth_unicast.octets) +
+ MLX5_GET_CTR(out, transmitted_ib_unicast.octets) +
MLX5_GET_CTR(out, transmitted_eth_multicast.octets) +
+ MLX5_GET_CTR(out, transmitted_ib_multicast.octets) +
MLX5_GET_CTR(out, transmitted_eth_broadcast.octets);
vf_stats->multicast =
- MLX5_GET_CTR(out, received_eth_multicast.packets);
+ MLX5_GET_CTR(out, received_eth_multicast.packets) +
+ MLX5_GET_CTR(out, received_ib_multicast.packets);
vf_stats->broadcast =
MLX5_GET_CTR(out, received_eth_broadcast.packets);
--
2.14.3
^ permalink raw reply related [flat|nested] 5+ messages in thread* [net 3/3] net/mlx5e: Err if asked to offload TC match on frag being first
2018-05-10 23:19 [pull request][net 0/3] Mellanox, mlx5 fixes 2018-05-10 Saeed Mahameed
2018-05-10 23:19 ` [net 1/3] net/mlx5: Free IRQs in shutdown path Saeed Mahameed
2018-05-10 23:19 ` [net 2/3] net/mlx5: E-Switch, Include VF RDMA stats in vport statistics Saeed Mahameed
@ 2018-05-10 23:19 ` Saeed Mahameed
2018-05-11 16:27 ` [pull request][net 0/3] Mellanox, mlx5 fixes 2018-05-10 David Miller
3 siblings, 0 replies; 5+ messages in thread
From: Saeed Mahameed @ 2018-05-10 23:19 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Roi Dayan, Saeed Mahameed
From: Roi Dayan <roid@mellanox.com>
The HW doesn't support matching on frag first/later, return error if we are
asked to offload that.
Fixes: 3f7d0eb42d59 ("net/mlx5e: Offload TC matching on packets being IP fragments")
Signed-off-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_tc.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 3c534fc43400..b94276db3ce9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1261,6 +1261,10 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,
f->mask);
addr_type = key->addr_type;
+ /* the HW doesn't support frag first/later */
+ if (mask->flags & FLOW_DIS_FIRST_FRAG)
+ return -EOPNOTSUPP;
+
if (mask->flags & FLOW_DIS_IS_FRAGMENT) {
MLX5_SET(fte_match_set_lyr_2_4, headers_c, frag, 1);
MLX5_SET(fte_match_set_lyr_2_4, headers_v, frag,
--
2.14.3
^ permalink raw reply related [flat|nested] 5+ messages in thread* Re: [pull request][net 0/3] Mellanox, mlx5 fixes 2018-05-10
2018-05-10 23:19 [pull request][net 0/3] Mellanox, mlx5 fixes 2018-05-10 Saeed Mahameed
` (2 preceding siblings ...)
2018-05-10 23:19 ` [net 3/3] net/mlx5e: Err if asked to offload TC match on frag being first Saeed Mahameed
@ 2018-05-11 16:27 ` David Miller
3 siblings, 0 replies; 5+ messages in thread
From: David Miller @ 2018-05-11 16:27 UTC (permalink / raw)
To: saeedm; +Cc: netdev
From: Saeed Mahameed <saeedm@mellanox.com>
Date: Thu, 10 May 2018 16:19:12 -0700
> the following series includes some fixes for mlx5 core driver.
> Please pull and let me know if there's any problem.
Pulled.
> For -stable v4.5
> ("net/mlx5: E-Switch, Include VF RDMA stats in vport statistics")
>
> For -stable v4.10
> ("net/mlx5e: Err if asked to offload TC match on frag being first")
Queued up for -stable.
Thanks.
^ permalink raw reply [flat|nested] 5+ messages in thread