* [net-next 08/13] net/mlx5: E-Switch, Support load/unload reps of specific vport types
From: Saeed Mahameed @ 2019-02-16 1:34 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Bodong Wang, Or Gerlitz, Saeed Mahameed
In-Reply-To: <20190216013452.21131-1-saeedm@mellanox.com>
From: Bodong Wang <bodong@mellanox.com>
Currently the driver loads and unloads all reps in an unbreakable
group. However, with ECPF, the reps of special vports such as uplink
and host PF should always be loaded in switchdev mode where the reps
for VFs will be loaded on-demand and unloaded on no-demand. This is
a pre-step for that change.
This patch doesn't change any functionality.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/eswitch_offloads.c | 109 ++++++++++++++----
1 file changed, 87 insertions(+), 22 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index c6c9dad69ba8..7131d41796fb 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -1291,25 +1291,47 @@ static void __esw_offloads_unload_rep(struct mlx5_eswitch *esw,
rep->rep_if[rep_type].state = REP_REGISTERED;
}
-static void esw_offloads_unload_reps_type(struct mlx5_eswitch *esw, int nvports,
- u8 rep_type)
+static void __unload_reps_special_vport(struct mlx5_eswitch *esw, u8 rep_type)
{
struct mlx5_eswitch_rep *rep;
- int vport;
-
- mlx5_esw_for_each_vf_rep_reverse(esw, vport, rep, nvports)
- __esw_offloads_unload_rep(esw, rep, rep_type);
rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_UPLINK);
__esw_offloads_unload_rep(esw, rep, rep_type);
}
-static void esw_offloads_unload_reps(struct mlx5_eswitch *esw, int nvports)
+static void __unload_reps_vf_vport(struct mlx5_eswitch *esw, int nvports,
+ u8 rep_type)
+{
+ struct mlx5_eswitch_rep *rep;
+ int i;
+
+ mlx5_esw_for_each_vf_rep_reverse(esw, i, rep, nvports)
+ __esw_offloads_unload_rep(esw, rep, rep_type);
+}
+
+static void esw_offloads_unload_vf_reps(struct mlx5_eswitch *esw, int nvports)
+{
+ u8 rep_type = NUM_REP_TYPES;
+
+ while (rep_type-- > 0)
+ __unload_reps_vf_vport(esw, nvports, rep_type);
+}
+
+static void __unload_reps_all_vport(struct mlx5_eswitch *esw, int nvports,
+ u8 rep_type)
+{
+ __unload_reps_vf_vport(esw, nvports, rep_type);
+
+ /* Special vports must be the last to unload. */
+ __unload_reps_special_vport(esw, rep_type);
+}
+
+static void esw_offloads_unload_all_reps(struct mlx5_eswitch *esw, int nvports)
{
u8 rep_type = NUM_REP_TYPES;
while (rep_type-- > 0)
- esw_offloads_unload_reps_type(esw, nvports, rep_type);
+ __unload_reps_all_vport(esw, nvports, rep_type);
}
static int __esw_offloads_load_rep(struct mlx5_eswitch *esw,
@@ -1329,39 +1351,82 @@ static int __esw_offloads_load_rep(struct mlx5_eswitch *esw,
return 0;
}
-static int esw_offloads_load_reps_type(struct mlx5_eswitch *esw, int nvports,
- u8 rep_type)
+static int __load_reps_special_vport(struct mlx5_eswitch *esw, u8 rep_type)
{
struct mlx5_eswitch_rep *rep;
- int vport;
int err;
rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_UPLINK);
err = __esw_offloads_load_rep(esw, rep, rep_type);
- if (err)
- goto out;
+ return err;
+}
- mlx5_esw_for_each_vf_rep(esw, vport, rep, nvports) {
+static int __load_reps_vf_vport(struct mlx5_eswitch *esw, int nvports,
+ u8 rep_type)
+{
+ struct mlx5_eswitch_rep *rep;
+ int err, i;
+
+ mlx5_esw_for_each_vf_rep(esw, i, rep, nvports) {
err = __esw_offloads_load_rep(esw, rep, rep_type);
if (err)
- goto err_reps;
+ goto err_vf;
}
return 0;
+err_vf:
+ __unload_reps_vf_vport(esw, --i, rep_type);
+ return err;
+}
+
+static int esw_offloads_load_vf_reps(struct mlx5_eswitch *esw, int nvports)
+{
+ u8 rep_type = 0;
+ int err;
+
+ for (rep_type = 0; rep_type < NUM_REP_TYPES; rep_type++) {
+ err = __load_reps_vf_vport(esw, nvports, rep_type);
+ if (err)
+ goto err_reps;
+ }
+
+ return err;
+
err_reps:
- esw_offloads_unload_reps_type(esw, --vport, rep_type);
-out:
+ while (rep_type-- > 0)
+ __unload_reps_vf_vport(esw, nvports, rep_type);
+ return err;
+}
+
+static int __load_reps_all_vport(struct mlx5_eswitch *esw, int nvports,
+ u8 rep_type)
+{
+ int err;
+
+ /* Special vports must be loaded first. */
+ err = __load_reps_special_vport(esw, rep_type);
+ if (err)
+ return err;
+
+ err = __load_reps_vf_vport(esw, nvports, rep_type);
+ if (err)
+ goto err_vfs;
+
+ return 0;
+
+err_vfs:
+ __unload_reps_special_vport(esw, rep_type);
return err;
}
-static int esw_offloads_load_reps(struct mlx5_eswitch *esw, int nvports)
+static int esw_offloads_load_all_reps(struct mlx5_eswitch *esw, int nvports)
{
u8 rep_type = 0;
int err;
for (rep_type = 0; rep_type < NUM_REP_TYPES; rep_type++) {
- err = esw_offloads_load_reps_type(esw, nvports, rep_type);
+ err = __load_reps_all_vport(esw, nvports, rep_type);
if (err)
goto err_reps;
}
@@ -1370,7 +1435,7 @@ static int esw_offloads_load_reps(struct mlx5_eswitch *esw, int nvports)
err_reps:
while (rep_type-- > 0)
- esw_offloads_unload_reps_type(esw, nvports, rep_type);
+ __unload_reps_all_vport(esw, nvports, rep_type);
return err;
}
@@ -1520,7 +1585,7 @@ int esw_offloads_init(struct mlx5_eswitch *esw, int vf_nvports,
if (err)
return err;
- err = esw_offloads_load_reps(esw, vf_nvports);
+ err = esw_offloads_load_all_reps(esw, vf_nvports);
if (err)
goto err_reps;
@@ -1556,7 +1621,7 @@ void esw_offloads_cleanup(struct mlx5_eswitch *esw)
u16 num_vfs = esw->dev->priv.sriov.num_vfs;
esw_offloads_devcom_cleanup(esw);
- esw_offloads_unload_reps(esw, num_vfs);
+ esw_offloads_unload_all_reps(esw, num_vfs);
esw_offloads_steering_cleanup(esw);
}
--
2.20.1
^ permalink raw reply related
* [net-next 10/13] net/mlx5: E-Switch, Assign a different position for uplink rep and vport
From: Saeed Mahameed @ 2019-02-16 1:34 UTC (permalink / raw)
To: David S. Miller
Cc: netdev, Bodong Wang, Eli Cohen, Or Gerlitz, Saeed Mahameed
In-Reply-To: <20190216013452.21131-1-saeedm@mellanox.com>
From: Bodong Wang <bodong@mellanox.com>
In offloads mode, the current implementation puts the uplink
representor at index zero of the vport reps array. It is not "natural"
to place it at index 0 since we want to put the representor for vport
0 at index 0 with the introduction of SmartNIC. A separate patch will
handle the case whether a rep is needed for vport 0 (PF vport).
So, we want to have a different placeholder for uplink vport and
representor. It was placed at the end of vport and rep array. Since
vport number can no longer act as an index into the vport or
representors arrays, use functions to map vport numbers to indices
when accessing the vports or representors arrays, and vice versa.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Signed-off-by: Eli Cohen <eli@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/eswitch.c | 11 +++++----
.../net/ethernet/mellanox/mlx5/core/eswitch.h | 24 +++++++++++++++++++
.../mellanox/mlx5/core/eswitch_offloads.c | 11 ++-------
include/linux/mlx5/vport.h | 11 ++++++---
4 files changed, 40 insertions(+), 17 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index d1454f18c0a7..bb7f72467df9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -84,8 +84,10 @@ enum {
static struct mlx5_vport *mlx5_eswitch_get_vport(struct mlx5_eswitch *esw,
u16 vport_num)
{
+ u16 idx = mlx5_eswitch_vport_num_to_index(esw, vport_num);
+
WARN_ON(vport_num > esw->total_vports - 1);
- return &esw->vports[vport_num];
+ return &esw->vports[idx];
}
static int arm_vport_context_events_cmd(struct mlx5_core_dev *dev, u16 vport,
@@ -1756,8 +1758,7 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev)
int total_vports = MLX5_TOTAL_VPORTS(dev);
struct mlx5_eswitch *esw;
struct mlx5_vport *vport;
- int vport_num;
- int err;
+ int err, i;
if (!MLX5_VPORT_MANAGER(dev))
return 0;
@@ -1798,8 +1799,8 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev)
hash_init(esw->offloads.mod_hdr_tbl);
mutex_init(&esw->state_lock);
- mlx5_esw_for_all_vports(esw, vport_num, vport) {
- vport->vport = vport_num;
+ mlx5_esw_for_all_vports(esw, i, vport) {
+ vport->vport = mlx5_eswitch_index_to_vport_num(esw, i);
vport->info.link_state = MLX5_VPORT_ADMIN_STATE_AUTO;
vport->dev = dev;
INIT_WORK(&vport->vport_change_handler,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index fd845e6c44d5..2951c1296c3e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -374,6 +374,30 @@ static inline u16 mlx5_eswitch_manager_vport(struct mlx5_core_dev *dev)
MLX5_VPORT_ECPF : MLX5_VPORT_PF;
}
+static inline int mlx5_eswitch_uplink_idx(struct mlx5_eswitch *esw)
+{
+ /* Uplink always locate at the last element of the array.*/
+ return esw->total_vports - 1;
+}
+
+static inline int mlx5_eswitch_vport_num_to_index(struct mlx5_eswitch *esw,
+ u16 vport_num)
+{
+ if (vport_num == MLX5_VPORT_UPLINK)
+ return mlx5_eswitch_uplink_idx(esw);
+
+ return vport_num;
+}
+
+static inline int mlx5_eswitch_index_to_vport_num(struct mlx5_eswitch *esw,
+ int index)
+{
+ if (index == mlx5_eswitch_uplink_idx(esw))
+ return MLX5_VPORT_UPLINK;
+
+ return index;
+}
+
#else /* CONFIG_MLX5_ESWITCH */
/* eswitch API stubs */
static inline int mlx5_eswitch_init(struct mlx5_core_dev *dev) { return 0; }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index b702b56c457e..e787e9212174 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -85,10 +85,7 @@ enum {
static struct mlx5_eswitch_rep *mlx5_eswitch_get_rep(struct mlx5_eswitch *esw,
u16 vport_num)
{
- u16 idx = vport_num;
-
- if (vport_num == MLX5_VPORT_UPLINK)
- idx = UPLINK_REP_INDEX;
+ u16 idx = mlx5_eswitch_vport_num_to_index(esw, vport_num);
WARN_ON(idx > esw->total_vports - 1);
return &esw->offloads.vport_reps[idx];
@@ -1254,7 +1251,6 @@ int esw_offloads_init_reps(struct mlx5_eswitch *esw)
{
int total_vfs = MLX5_TOTAL_VPORTS(esw->dev);
struct mlx5_core_dev *dev = esw->dev;
- struct mlx5_esw_offload *offloads;
struct mlx5_eswitch_rep *rep;
u8 hw_id[ETH_ALEN], rep_type;
int vport;
@@ -1265,19 +1261,16 @@ int esw_offloads_init_reps(struct mlx5_eswitch *esw)
if (!esw->offloads.vport_reps)
return -ENOMEM;
- offloads = &esw->offloads;
mlx5_query_nic_vport_mac_address(dev, 0, hw_id);
mlx5_esw_for_all_reps(esw, vport, rep) {
- rep->vport = vport;
+ rep->vport = mlx5_eswitch_index_to_vport_num(esw, vport);
ether_addr_copy(rep->hw_id, hw_id);
for (rep_type = 0; rep_type < NUM_REP_TYPES; rep_type++)
rep->rep_if[rep_type].state = REP_UNREGISTERED;
}
- offloads->vport_reps[0].vport = MLX5_VPORT_UPLINK;
-
return 0;
}
diff --git a/include/linux/mlx5/vport.h b/include/linux/mlx5/vport.h
index 755aeea19e1c..134248c02786 100644
--- a/include/linux/mlx5/vport.h
+++ b/include/linux/mlx5/vport.h
@@ -36,9 +36,14 @@
#include <linux/mlx5/driver.h>
#include <linux/mlx5/device.h>
-#define MLX5_VPORT_PF_PLACEHOLDER (1u)
-#define MLX5_SPECIAL_VPORTS (MLX5_VPORT_PF_PLACEHOLDER)
-#define MLX5_TOTAL_VPORTS(mdev) (MLX5_SPECIAL_VPORTS + mlx5_core_max_vfs(mdev))
+#define MLX5_VPORT_PF_PLACEHOLDER (1u)
+#define MLX5_VPORT_UPLINK_PLACEHOLDER (1u)
+
+#define MLX5_SPECIAL_VPORTS (MLX5_VPORT_PF_PLACEHOLDER + \
+ MLX5_VPORT_UPLINK_PLACEHOLDER)
+
+#define MLX5_TOTAL_VPORTS(mdev) (MLX5_SPECIAL_VPORTS + \
+ mlx5_core_max_vfs(mdev))
#define MLX5_VPORT_MANAGER(mdev) \
(MLX5_CAP_GEN(mdev, vport_group_manager) && \
--
2.20.1
^ permalink raw reply related
* [net-next 11/13] net/mlx5: E-Switch, Consider ECPF vport depends on eswitch ownership
From: Saeed Mahameed @ 2019-02-16 1:34 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Bodong Wang, Or Gerlitz, Saeed Mahameed
In-Reply-To: <20190216013452.21131-1-saeedm@mellanox.com>
From: Bodong Wang <bodong@mellanox.com>
ECPF connects to the eswitch through vport 0xfffe. ECPF may or may
not be the eswitch manager depending on firmware configuration.
1. If ECPF is eswitch manager: ECPF will take over the eswitch manager
responsibility. A rep of the host PF shall be created at the ECPF
side for the eswitch manager to control.
2. If ECPF is not eswitch manager: host PF will be the eswitch manager,
ECPF acts similar as a VF to the host PF. Host PF will be aware
of the ECPF vport presence and control it's rep.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/eswitch.c | 8 +-
.../net/ethernet/mellanox/mlx5/core/eswitch.h | 15 ++++
.../mellanox/mlx5/core/eswitch_offloads.c | 79 ++++++++++++++++++-
include/linux/mlx5/driver.h | 5 ++
include/linux/mlx5/mlx5_ifc.h | 3 +-
include/linux/mlx5/vport.h | 8 +-
6 files changed, 110 insertions(+), 8 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index bb7f72467df9..d2ab1ee19b2a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -1667,7 +1667,7 @@ int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs, int mode)
mlx5_reload_interface(esw->dev, MLX5_INTERFACE_PROTOCOL_ETH);
mlx5_reload_interface(esw->dev, MLX5_INTERFACE_PROTOCOL_IB);
err = esw_offloads_init(esw, nvfs,
- nvfs + MLX5_SPECIAL_VPORTS);
+ nvfs + MLX5_SPECIAL_VPORTS(esw->dev));
}
if (err)
@@ -1687,6 +1687,12 @@ int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs, int mode)
vport = mlx5_eswitch_get_vport(esw, MLX5_VPORT_PF);
esw_enable_vport(esw, vport, enabled_events);
+ /* Enable ECPF vports */
+ if (mlx5_ecpf_vport_exists(esw->dev)) {
+ vport = mlx5_eswitch_get_vport(esw, MLX5_VPORT_ECPF);
+ esw_enable_vport(esw, vport, enabled_events);
+ }
+
/* Enable VF vports */
mlx5_esw_for_each_vf_vport(esw, i, vport, nvfs)
esw_enable_vport(esw, vport, enabled_events);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 2951c1296c3e..2baa0d71380c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -380,9 +380,20 @@ static inline int mlx5_eswitch_uplink_idx(struct mlx5_eswitch *esw)
return esw->total_vports - 1;
}
+static inline int mlx5_eswitch_ecpf_idx(struct mlx5_eswitch *esw)
+{
+ return esw->total_vports - 2;
+}
+
static inline int mlx5_eswitch_vport_num_to_index(struct mlx5_eswitch *esw,
u16 vport_num)
{
+ if (vport_num == MLX5_VPORT_ECPF) {
+ if (!mlx5_ecpf_vport_exists(esw->dev))
+ esw_warn(esw->dev, "ECPF vport doesn't exist!\n");
+ return mlx5_eswitch_ecpf_idx(esw);
+ }
+
if (vport_num == MLX5_VPORT_UPLINK)
return mlx5_eswitch_uplink_idx(esw);
@@ -392,6 +403,10 @@ static inline int mlx5_eswitch_vport_num_to_index(struct mlx5_eswitch *esw,
static inline int mlx5_eswitch_index_to_vport_num(struct mlx5_eswitch *esw,
int index)
{
+ if (index == mlx5_eswitch_ecpf_idx(esw) &&
+ mlx5_ecpf_vport_exists(esw->dev))
+ return MLX5_VPORT_ECPF;
+
if (index == mlx5_eswitch_uplink_idx(esw))
return MLX5_VPORT_UPLINK;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index e787e9212174..84a33f8e3350 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -639,14 +639,35 @@ static int esw_add_fdb_peer_miss_rules(struct mlx5_eswitch *esw,
misc = MLX5_ADDR_OF(fte_match_param, spec->match_value,
misc_parameters);
+ if (mlx5_core_is_ecpf_esw_manager(esw->dev)) {
+ MLX5_SET(fte_match_set_misc, misc, source_port, MLX5_VPORT_PF);
+ flow = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb,
+ spec, &flow_act, &dest, 1);
+ if (IS_ERR(flow)) {
+ err = PTR_ERR(flow);
+ goto add_pf_flow_err;
+ }
+ flows[MLX5_VPORT_PF] = flow;
+ }
+
+ if (mlx5_ecpf_vport_exists(esw->dev)) {
+ MLX5_SET(fte_match_set_misc, misc, source_port, MLX5_VPORT_ECPF);
+ flow = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb,
+ spec, &flow_act, &dest, 1);
+ if (IS_ERR(flow)) {
+ err = PTR_ERR(flow);
+ goto add_ecpf_flow_err;
+ }
+ flows[mlx5_eswitch_ecpf_idx(esw)] = flow;
+ }
+
mlx5_esw_for_each_vf_vport(esw, i, mlx5_core_max_vfs(esw->dev)) {
MLX5_SET(fte_match_set_misc, misc, source_port, i);
flow = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb,
spec, &flow_act, &dest, 1);
if (IS_ERR(flow)) {
err = PTR_ERR(flow);
- esw_warn(esw->dev, "FDB: Failed to add peer miss flow rule err %d\n", err);
- goto add_flow_err;
+ goto add_vf_flow_err;
}
flows[i] = flow;
}
@@ -656,10 +677,18 @@ static int esw_add_fdb_peer_miss_rules(struct mlx5_eswitch *esw,
kvfree(spec);
return 0;
-add_flow_err:
+add_vf_flow_err:
nvports = --i;
mlx5_esw_for_each_vf_vport_reverse(esw, i, nvports)
mlx5_del_flow_rules(flows[i]);
+
+ if (mlx5_ecpf_vport_exists(esw->dev))
+ mlx5_del_flow_rules(flows[mlx5_eswitch_ecpf_idx(esw)]);
+add_ecpf_flow_err:
+ if (mlx5_core_is_ecpf_esw_manager(esw->dev))
+ mlx5_del_flow_rules(flows[MLX5_VPORT_PF]);
+add_pf_flow_err:
+ esw_warn(esw->dev, "FDB: Failed to add peer miss flow rule err %d\n", err);
kvfree(flows);
alloc_flows_err:
kvfree(spec);
@@ -676,6 +705,12 @@ static void esw_del_fdb_peer_miss_rules(struct mlx5_eswitch *esw)
mlx5_esw_for_each_vf_vport_reverse(esw, i, mlx5_core_max_vfs(esw->dev))
mlx5_del_flow_rules(flows[i]);
+ if (mlx5_ecpf_vport_exists(esw->dev))
+ mlx5_del_flow_rules(flows[mlx5_eswitch_ecpf_idx(esw)]);
+
+ if (mlx5_core_is_ecpf_esw_manager(esw->dev))
+ mlx5_del_flow_rules(flows[MLX5_VPORT_PF]);
+
kvfree(flows);
}
@@ -1288,6 +1323,16 @@ static void __unload_reps_special_vport(struct mlx5_eswitch *esw, u8 rep_type)
{
struct mlx5_eswitch_rep *rep;
+ if (mlx5_ecpf_vport_exists(esw->dev)) {
+ rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_ECPF);
+ __esw_offloads_unload_rep(esw, rep, rep_type);
+ }
+
+ if (mlx5_core_is_ecpf_esw_manager(esw->dev)) {
+ rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_PF);
+ __esw_offloads_unload_rep(esw, rep, rep_type);
+ }
+
rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_UPLINK);
__esw_offloads_unload_rep(esw, rep, rep_type);
}
@@ -1351,6 +1396,34 @@ static int __load_reps_special_vport(struct mlx5_eswitch *esw, u8 rep_type)
rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_UPLINK);
err = __esw_offloads_load_rep(esw, rep, rep_type);
+ if (err)
+ return err;
+
+ if (mlx5_core_is_ecpf_esw_manager(esw->dev)) {
+ rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_PF);
+ err = __esw_offloads_load_rep(esw, rep, rep_type);
+ if (err)
+ goto err_pf;
+ }
+
+ if (mlx5_ecpf_vport_exists(esw->dev)) {
+ rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_ECPF);
+ err = __esw_offloads_load_rep(esw, rep, rep_type);
+ if (err)
+ goto err_ecpf;
+ }
+
+ return 0;
+
+err_ecpf:
+ if (mlx5_core_is_ecpf_esw_manager(esw->dev)) {
+ rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_PF);
+ __esw_offloads_unload_rep(esw, rep, rep_type);
+ }
+
+err_pf:
+ rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_UPLINK);
+ __esw_offloads_unload_rep(esw, rep, rep_type);
return err;
}
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index c5454f985e1d..c2de50f02b33 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -1088,6 +1088,11 @@ static inline bool mlx5_core_is_ecpf_esw_manager(struct mlx5_core_dev *dev)
return dev->caps.embedded_cpu && MLX5_CAP_GEN(dev, eswitch_manager);
}
+static inline bool mlx5_ecpf_vport_exists(struct mlx5_core_dev *dev)
+{
+ return mlx5_core_is_pf(dev) && MLX5_CAP_ESW(dev, ecpf_vport_exists);
+}
+
#define MLX5_HOST_PF_MAX_VFS (127u)
static inline u16 mlx5_core_max_vfs(struct mlx5_core_dev *dev)
{
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 5decffe565fb..b7bb774b57b0 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -631,7 +631,8 @@ struct mlx5_ifc_e_switch_cap_bits {
u8 vport_svlan_insert[0x1];
u8 vport_cvlan_insert_if_not_exist[0x1];
u8 vport_cvlan_insert_overwrite[0x1];
- u8 reserved_at_5[0x17];
+ u8 reserved_at_5[0x16];
+ u8 ecpf_vport_exists[0x1];
u8 counter_eswitch_affinity[0x1];
u8 merged_eswitch[0x1];
u8 nic_vport_node_guid_modify[0x1];
diff --git a/include/linux/mlx5/vport.h b/include/linux/mlx5/vport.h
index 134248c02786..0eef548b9946 100644
--- a/include/linux/mlx5/vport.h
+++ b/include/linux/mlx5/vport.h
@@ -38,11 +38,13 @@
#define MLX5_VPORT_PF_PLACEHOLDER (1u)
#define MLX5_VPORT_UPLINK_PLACEHOLDER (1u)
+#define MLX5_VPORT_ECPF_PLACEHOLDER(mdev) (mlx5_ecpf_vport_exists(mdev))
-#define MLX5_SPECIAL_VPORTS (MLX5_VPORT_PF_PLACEHOLDER + \
- MLX5_VPORT_UPLINK_PLACEHOLDER)
+#define MLX5_SPECIAL_VPORTS(mdev) (MLX5_VPORT_PF_PLACEHOLDER + \
+ MLX5_VPORT_UPLINK_PLACEHOLDER + \
+ MLX5_VPORT_ECPF_PLACEHOLDER(mdev))
-#define MLX5_TOTAL_VPORTS(mdev) (MLX5_SPECIAL_VPORTS + \
+#define MLX5_TOTAL_VPORTS(mdev) (MLX5_SPECIAL_VPORTS(mdev) + \
mlx5_core_max_vfs(mdev))
#define MLX5_VPORT_MANAGER(mdev) \
--
2.20.1
^ permalink raw reply related
* [net-next 13/13] net/mlx5: E-Switch, Allow transition to offloads mode for ECPF
From: Saeed Mahameed @ 2019-02-16 1:34 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Bodong Wang, Or Gerlitz, Saeed Mahameed
In-Reply-To: <20190216013452.21131-1-saeedm@mellanox.com>
From: Bodong Wang <bodong@mellanox.com>
Currently, the e-switch driver requires going to legacy mode before
changing to the offloads mode. This makes sense for regular case as
the legacy mode is done by creating VFs.
However, it's problematic when ECPF is the eswitch manager. In such
case, ECPF will control the vports on peer host including the peer
PF and VFs. But ECPF doesn't need and shall not create VFs as the
VFs are created in the peer PF host.
Grant ECPF the ability to change from none to the offloads mode. Note
that currently the only way to go back to none mode is by unloading
the ECPF driver.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 6 ++++--
1 file changed, 4 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 91c4095ac79e..f2260391be5b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -1250,7 +1250,8 @@ static int esw_offloads_start(struct mlx5_eswitch *esw,
{
int err, err1, num_vfs = esw->dev->priv.sriov.num_vfs;
- if (esw->mode != SRIOV_LEGACY) {
+ if (esw->mode != SRIOV_LEGACY &&
+ !mlx5_core_is_ecpf_esw_manager(esw->dev)) {
NL_SET_ERR_MSG_MOD(extack,
"Can't set offloads mode, SRIOV legacy not enabled");
return -EINVAL;
@@ -1846,7 +1847,8 @@ static int mlx5_devlink_eswitch_check(struct devlink *devlink)
if(!MLX5_ESWITCH_MANAGER(dev))
return -EPERM;
- if (dev->priv.eswitch->mode == SRIOV_NONE)
+ if (dev->priv.eswitch->mode == SRIOV_NONE &&
+ !mlx5_core_is_ecpf_esw_manager(dev))
return -EOPNOTSUPP;
return 0;
--
2.20.1
^ permalink raw reply related
* [net-next 12/13] net/mlx5: E-Switch, Load/unload VF reps according to event from host PF
From: Saeed Mahameed @ 2019-02-16 1:34 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Bodong Wang, Or Gerlitz, Saeed Mahameed
In-Reply-To: <20190216013452.21131-1-saeedm@mellanox.com>
From: Bodong Wang <bodong@mellanox.com>
When host PF changes the number of VFs, the ECPF esw driver will get
a FW event. It should query the number of VFs enabled by host PF and
update the VF reps accordingly. Note that host PF can't change the
number of VFs dynamically, it has to reset the number of VFs to 0
before changing to a new positive number.
The host event is registered when driver is moving to switchdev mode,
and it's the last step to do in esw_offloads_init. It's unregistered
and the work queue is flushed when driver quits from switchdev mode.
In this way, the host event and devlink command are serialized.
When driver is enabling switchdev mode, pay attention to the following
two facts:
1. Host PF must not have VF initialized as the flow table in ECPF has
ENCAP enabled as default. Such flow table can't be created with
existing initialized VFs.
2. ECPF doesn't know how many VFs the host PF will enable, ECPF
offloads flow steering shall create the flow table/groups based on
the max number of VFs possibly supported by host PF.
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/eswitch.c | 17 ++++-
.../net/ethernet/mellanox/mlx5/core/eswitch.h | 11 +++
.../mellanox/mlx5/core/eswitch_offloads.c | 71 ++++++++++++++++++-
3 files changed, 96 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index d2ab1ee19b2a..e18af31336e6 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -39,6 +39,7 @@
#include "lib/eq.h"
#include "eswitch.h"
#include "fs_core.h"
+#include "ecpf.h"
enum {
MLX5_ACTION_NONE = 0,
@@ -1639,6 +1640,7 @@ static int eswitch_vport_event(struct notifier_block *nb,
int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs, int mode)
{
+ int vf_nvports = 0, total_nvports = 0;
struct mlx5_vport *vport;
int err;
int i, enabled_events;
@@ -1657,6 +1659,18 @@ int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs, int mode)
esw_info(esw->dev, "E-Switch enable SRIOV: nvfs(%d) mode (%d)\n", nvfs, mode);
+ if (mode == SRIOV_OFFLOADS) {
+ if (mlx5_core_is_ecpf_esw_manager(esw->dev)) {
+ err = mlx5_query_host_params_num_vfs(esw->dev, &vf_nvports);
+ if (err)
+ return err;
+ total_nvports = esw->total_vports;
+ } else {
+ vf_nvports = nvfs;
+ total_nvports = nvfs + MLX5_SPECIAL_VPORTS(esw->dev);
+ }
+ }
+
esw->mode = mode;
mlx5_lag_update(esw->dev);
@@ -1666,8 +1680,7 @@ int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs, int mode)
} else {
mlx5_reload_interface(esw->dev, MLX5_INTERFACE_PROTOCOL_ETH);
mlx5_reload_interface(esw->dev, MLX5_INTERFACE_PROTOCOL_IB);
- err = esw_offloads_init(esw, nvfs,
- nvfs + MLX5_SPECIAL_VPORTS(esw->dev));
+ err = esw_offloads_init(esw, vf_nvports, total_nvports);
}
if (err)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 2baa0d71380c..af5581a57e56 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -182,6 +182,16 @@ struct esw_mc_addr { /* SRIOV only */
u32 refcnt;
};
+struct mlx5_host_work {
+ struct work_struct work;
+ struct mlx5_eswitch *esw;
+};
+
+struct mlx5_host_info {
+ struct mlx5_nb nb;
+ u16 num_vfs;
+};
+
struct mlx5_eswitch {
struct mlx5_core_dev *dev;
struct mlx5_nb nb;
@@ -206,6 +216,7 @@ struct mlx5_eswitch {
int mode;
int nvports;
u16 manager_vport;
+ struct mlx5_host_info host_info;
};
void esw_offloads_cleanup(struct mlx5_eswitch *esw);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 84a33f8e3350..91c4095ac79e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -40,6 +40,8 @@
#include "en.h"
#include "fs_core.h"
#include "lib/devcom.h"
+#include "ecpf.h"
+#include "lib/eq.h"
enum {
FDB_FAST_PATH = 0,
@@ -1640,6 +1642,57 @@ static void esw_offloads_steering_cleanup(struct mlx5_eswitch *esw)
esw_destroy_offloads_fdb_tables(esw);
}
+static void esw_host_params_event_handler(struct work_struct *work)
+{
+ struct mlx5_host_work *host_work;
+ struct mlx5_eswitch *esw;
+ int err, num_vf = 0;
+
+ host_work = container_of(work, struct mlx5_host_work, work);
+ esw = host_work->esw;
+
+ err = mlx5_query_host_params_num_vfs(esw->dev, &num_vf);
+ if (err || num_vf == esw->host_info.num_vfs)
+ goto out;
+
+ /* Number of VFs can only change from "0 to x" or "x to 0". */
+ if (esw->host_info.num_vfs > 0) {
+ esw_offloads_unload_vf_reps(esw, esw->host_info.num_vfs);
+ } else {
+ err = esw_offloads_load_vf_reps(esw, num_vf);
+
+ if (err)
+ goto out;
+ }
+
+ esw->host_info.num_vfs = num_vf;
+
+out:
+ kfree(host_work);
+}
+
+static int esw_host_params_event(struct notifier_block *nb,
+ unsigned long type, void *data)
+{
+ struct mlx5_host_work *host_work;
+ struct mlx5_host_info *host_info;
+ struct mlx5_eswitch *esw;
+
+ host_work = kzalloc(sizeof(*host_work), GFP_ATOMIC);
+ if (!host_work)
+ return NOTIFY_DONE;
+
+ host_info = mlx5_nb_cof(nb, struct mlx5_host_info, nb);
+ esw = container_of(host_info, struct mlx5_eswitch, host_info);
+
+ host_work->esw = esw;
+
+ INIT_WORK(&host_work->work, esw_host_params_event_handler);
+ queue_work(esw->work_queue, &host_work->work);
+
+ return NOTIFY_OK;
+}
+
int esw_offloads_init(struct mlx5_eswitch *esw, int vf_nvports,
int total_nvports)
{
@@ -1656,6 +1709,14 @@ int esw_offloads_init(struct mlx5_eswitch *esw, int vf_nvports,
goto err_reps;
esw_offloads_devcom_init(esw);
+
+ if (mlx5_core_is_ecpf_esw_manager(esw->dev)) {
+ MLX5_NB_INIT(&esw->host_info.nb, esw_host_params_event,
+ HOST_PARAMS_CHANGE);
+ mlx5_eq_notifier_register(esw->dev, &esw->host_info.nb);
+ esw->host_info.num_vfs = vf_nvports;
+ }
+
return 0;
err_reps:
@@ -1684,7 +1745,15 @@ static int esw_offloads_stop(struct mlx5_eswitch *esw,
void esw_offloads_cleanup(struct mlx5_eswitch *esw)
{
- u16 num_vfs = esw->dev->priv.sriov.num_vfs;
+ u16 num_vfs;
+
+ if (mlx5_core_is_ecpf_esw_manager(esw->dev)) {
+ mlx5_eq_notifier_unregister(esw->dev, &esw->host_info.nb);
+ flush_workqueue(esw->work_queue);
+ num_vfs = esw->host_info.num_vfs;
+ } else {
+ num_vfs = esw->dev->priv.sriov.num_vfs;
+ }
esw_offloads_devcom_cleanup(esw);
esw_offloads_unload_all_reps(esw, num_vfs);
--
2.20.1
^ permalink raw reply related
* [PATCH net-next] igc: Make function igc_write_rss_indir_tbl() static
From: Wei Yongjun @ 2019-02-16 2:04 UTC (permalink / raw)
To: Jeff Kirsher, Sasha Neftin
Cc: Wei Yongjun, intel-wired-lan, netdev, kernel-janitors
Fixes the following sparse warning:
drivers/net/ethernet/intel/igc/igc_ethtool.c:646:6: warning:
symbol 'igc_write_rss_indir_tbl' was not declared. Should it be static?
Fixes: 8c5ad0dae93c ("igc: Add ethtool support")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
---
drivers/net/ethernet/intel/igc/igc_ethtool.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/igc/igc_ethtool.c b/drivers/net/ethernet/intel/igc/igc_ethtool.c
index eff37a6c0afa..544239422577 100644
--- a/drivers/net/ethernet/intel/igc/igc_ethtool.c
+++ b/drivers/net/ethernet/intel/igc/igc_ethtool.c
@@ -643,7 +643,7 @@ static int igc_set_coalesce(struct net_device *netdev,
return 0;
}
-void igc_write_rss_indir_tbl(struct igc_adapter *adapter)
+static void igc_write_rss_indir_tbl(struct igc_adapter *adapter)
{
struct igc_hw *hw = &adapter->hw;
u32 reg = IGC_RETA(0);
^ permalink raw reply related
* Re: [PATCH net 2/2] tcp: tcp_v4_err() should be more careful
From: Neal Cardwell @ 2019-02-16 2:21 UTC (permalink / raw)
To: Eric Dumazet
Cc: David S . Miller, netdev, Eric Dumazet, Yuchung Cheng,
soukjin bae
In-Reply-To: <20190215213621.183537-3-edumazet@google.com>
On Fri, Feb 15, 2019 at 4:36 PM Eric Dumazet <edumazet@google.com> wrote:
>
> ICMP handlers are not very often stressed, we should
> make them more resilient to bugs that might surface in
> the future.
>
> If there is no packet in retransmit queue, we should
> avoid a NULL deref.
>
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Reported-by: soukjin bae <soukjin.bae@samsung.com>
> ---
> net/ipv4/tcp_ipv4.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
Acked-by: Neal Cardwell <ncardwell@google.com>
Thanks!
neal
^ permalink raw reply
* Re: [PATCH net 1/2] tcp: clear icsk_backoff in tcp_write_queue_purge()
From: Neal Cardwell @ 2019-02-16 2:21 UTC (permalink / raw)
To: Eric Dumazet
Cc: David S . Miller, netdev, Eric Dumazet, Yuchung Cheng,
soukjin bae
In-Reply-To: <20190215213621.183537-2-edumazet@google.com>
On Fri, Feb 15, 2019 at 4:36 PM Eric Dumazet <edumazet@google.com> wrote:
>
> soukjin bae reported a crash in tcp_v4_err() handling
> ICMP_DEST_UNREACH after tcp_write_queue_head(sk)
> returned a NULL pointer.
>
> Current logic should have prevented this :
>
> if (seq != tp->snd_una || !icsk->icsk_retransmits ||
> !icsk->icsk_backoff || fastopen)
> break;
>
> Problem is the write queue might have been purged
> and icsk_backoff has not been cleared.
>
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Reported-by: soukjin bae <soukjin.bae@samsung.com>
> ---
> net/ipv4/tcp.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
Acked-by: Neal Cardwell <ncardwell@google.com>
Thanks!
neal
^ permalink raw reply
* Re: [PATCH net 2/2] tcp: tcp_v4_err() should be more careful
From: Soheil Hassas Yeganeh @ 2019-02-16 2:31 UTC (permalink / raw)
To: Neal Cardwell
Cc: Eric Dumazet, David S . Miller, netdev, Eric Dumazet,
Yuchung Cheng, soukjin bae
In-Reply-To: <CADVnQymt=dtyQDhMNFz4Wmbeqbies_nwvU7fxawyizOji5H7eg@mail.gmail.com>
On Fri, Feb 15, 2019 at 9:21 PM Neal Cardwell <ncardwell@google.com> wrote:
>
> On Fri, Feb 15, 2019 at 4:36 PM Eric Dumazet <edumazet@google.com> wrote:
> >
> > ICMP handlers are not very often stressed, we should
> > make them more resilient to bugs that might surface in
> > the future.
> >
> > If there is no packet in retransmit queue, we should
> > avoid a NULL deref.
> >
> > Signed-off-by: Eric Dumazet <edumazet@google.com>
> > Reported-by: soukjin bae <soukjin.bae@samsung.com>
> > ---
> > net/ipv4/tcp_ipv4.c | 5 ++++-
> > 1 file changed, 4 insertions(+), 1 deletion(-)
>
> Acked-by: Neal Cardwell <ncardwell@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Thanks you for the fix, Eric!
>
> Thanks!
>
> neal
^ permalink raw reply
* [PATCH net-next] ptr_ring: remove duplicated include from ptr_ring.h
From: YueHaibing @ 2019-02-16 2:37 UTC (permalink / raw)
To: davem, mst, xiyou.wangcong, parri.andrea; +Cc: linux-kernel, netdev, YueHaibing
Remove duplicated include.
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
---
include/linux/ptr_ring.h | 1 -
1 file changed, 1 deletion(-)
diff --git a/include/linux/ptr_ring.h b/include/linux/ptr_ring.h
index 186cd8e..8da46ac 100644
--- a/include/linux/ptr_ring.h
+++ b/include/linux/ptr_ring.h
@@ -26,7 +26,6 @@
#include <linux/cache.h>
#include <linux/types.h>
#include <linux/compiler.h>
-#include <linux/cache.h>
#include <linux/slab.h>
#include <asm/errno.h>
#endif
--
2.7.4
^ permalink raw reply related
* Re: [PATCH net-next] mdio_bus: Fix PTR_ERR() usage after initialization to constant
From: YueHaibing @ 2019-02-16 2:40 UTC (permalink / raw)
To: Al Viro; +Cc: Andrew Lunn, davem, f.fainelli, hkallweit1, linux-kernel, netdev
In-Reply-To: <20190201042411.GM2217@ZenIV.linux.org.uk>
On 2019/2/1 12:24, Al Viro wrote:
> On Tue, Jan 29, 2019 at 11:30:27AM +0800, YueHaibing wrote:
>>>> gpiod = fwnode_get_named_gpiod(&mdiodev->dev.of_node->fwnode,
>>>> "reset-gpios", 0, GPIOD_OUT_LOW,
>>>> "PHY reset");
>>>> - if (PTR_ERR(gpiod) == -ENOENT ||
>>>> - PTR_ERR(gpiod) == -ENOSYS)
>>>> - gpiod = NULL;
>>>> - else if (IS_ERR(gpiod))
>>>> - return PTR_ERR(gpiod);
>>>> + if (IS_ERR(gpiod)) {
>>>> + ret = PTR_ERR(gpiod);
>>>> + if (ret == -ENOENT || ret == -ENOSYS)
>>>> + gpiod = NULL;
>>>> + else
>>>> + return ret;
>>>> + }
>
> Rule of the thumb: PTR_ERR(p) == -E... is almost always better off
> as p == ERR_PTR(-E...)
Ok, will fix it.
>
> .
>
^ permalink raw reply
* Re: [net-next 5/5] net: sock: remove the definition of SOCK_DEBUG()
From: Yafang Shao @ 2019-02-16 2:50 UTC (permalink / raw)
To: Joe Perches
Cc: Eric Dumazet, Cong Wang, David Miller, Daniel Borkmann, netdev,
shaoyafang
In-Reply-To: <620a93625ac3709bf6df5e31bcdcdd319db140ef.camel@perches.com>
On Sat, Feb 16, 2019 at 2:51 AM Joe Perches <joe@perches.com> wrote:
>
> On Fri, 2019-02-15 at 10:22 -0800, Eric Dumazet wrote:
> > On Fri, Feb 15, 2019 at 10:13 AM Cong Wang <xiyou.wangcong@gmail.com> wrote:
> > > On Fri, Feb 15, 2019 at 8:26 AM Eric Dumazet <edumazet@google.com> wrote:
> > > > On Fri, Feb 15, 2019 at 6:50 AM Yafang Shao <laoar.shao@gmail.com> wrote:
> > > > > As SOCK_DEBUG() isn't used any more, we can get ride of it now.
> > > > >
> > > >
> > > > No, we are still using this infrastructure from time to time.
> > > >
> > > > I told you I agreed to remove the current (obsolete) TCP call sites,
> > > > I never suggested to remove SOCK_DEBUG() completely.
> > >
> > > Since when do we upstream care about any out-of-tree users?
> > >
> > > You can always carry a patch to keep it downstream if you want,
> > > no one can stop you doing it.
> >
> > Somehow the patch series seems to present things in this way :
> >
> > Eric Dumazet suggested to remove completely the SOCK_DEBUG() interface.
>
> Well, you kinda did.
> It's certainly reasonable to interpret what you wrote as such.
>
> On Tue, 2019-02-12 at 18:15 -0800, Eric Dumazet wrote:
> > Just remove all SOCK_DEBUG() calls, there are leftovers of very ancient times.
>
> My suggestion would be to undefine SOCK_DEBUGGING.
>
This seems like a reasonable trade-off decision.
I will change it like this.
> Something like:
> ---
> include/net/sock.h | 13 ++++++++-----
> 1 file changed, 8 insertions(+), 5 deletions(-)
>
> diff --git a/include/net/sock.h b/include/net/sock.h
> index 328cb7cb7b0b..7e39bdfa342a 100644
> --- a/include/net/sock.h
> +++ b/include/net/sock.h
> @@ -81,14 +81,17 @@
> */
>
> /* Define this to get the SOCK_DBG debugging facility. */
> -#define SOCK_DEBUGGING
> +/* #define SOCK_DEBUGGING */
> #ifdef SOCK_DEBUGGING
> -#define SOCK_DEBUG(sk, msg...) do { if ((sk) && sock_flag((sk), SOCK_DBG)) \
> - printk(KERN_DEBUG msg); } while (0)
> +#define SOCK_DEBUG(sk, fmt, ...) \
> +do { \
> + if ((sk) && sock_flag((sk), SOCK_DBG)) \
> + printk(KERN_DEBUG fmt, ##__VA_ARGS__); \
> +} while (0)
> #else
> /* Validate arguments and do nothing */
> -static inline __printf(2, 3)
> -void SOCK_DEBUG(const struct sock *sk, const char *msg, ...)
> +__printf(2, 3)
> +static inline void SOCK_DEBUG(const struct sock *sk, const char *fmt, ...)
> {
> }
> #endif
>
>
Thanks
Yafang
^ permalink raw reply
* RE: [PATCH 1/2] i40e: move i40e_xsk_umem function
From: Brown, Aaron F @ 2019-02-16 2:54 UTC (permalink / raw)
To: Björn Töpel, intel-wired-lan@lists.osuosl.org
Cc: Topel, Bjorn, Karlsson, Magnus, magnus.karlsson@gmail.com,
netdev@vger.kernel.org, Sokolowski, Jan
In-Reply-To: <20190212085205.7848-2-bjorn.topel@gmail.com>
> From: netdev-owner@vger.kernel.org [mailto:netdev-
> owner@vger.kernel.org] On Behalf Of Björn Töpel
> Sent: Tuesday, February 12, 2019 12:52 AM
> To: intel-wired-lan@lists.osuosl.org
> Cc: Topel, Bjorn <bjorn.topel@intel.com>; Karlsson, Magnus
> <magnus.karlsson@intel.com>; magnus.karlsson@gmail.com;
> netdev@vger.kernel.org; Sokolowski, Jan <jan.sokolowski@intel.com>
> Subject: [PATCH 1/2] i40e: move i40e_xsk_umem function
>
> From: Björn Töpel <bjorn.topel@intel.com>
>
> The i40e_xsk_umem function was explicitly inlined in i40e.h. There is
> no reason for that, so move it to i40e_main.c instead.
>
> Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
> ---
> drivers/net/ethernet/intel/i40e/i40e.h | 14 --------------
> drivers/net/ethernet/intel/i40e/i40e_main.c | 20 ++++++++++++++++++++
> 2 files changed, 20 insertions(+), 14 deletions(-)
>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
^ permalink raw reply
* RE: [PATCH 2/2] i40e: add tracking of AF_XDP ZC state for each queue pair
From: Brown, Aaron F @ 2019-02-16 2:54 UTC (permalink / raw)
To: Björn Töpel, intel-wired-lan@lists.osuosl.org
Cc: Topel, Bjorn, Karlsson, Magnus, magnus.karlsson@gmail.com,
netdev@vger.kernel.org, Sokolowski, Jan
In-Reply-To: <20190212085205.7848-3-bjorn.topel@gmail.com>
> From: netdev-owner@vger.kernel.org [mailto:netdev-
> owner@vger.kernel.org] On Behalf Of Björn Töpel
> Sent: Tuesday, February 12, 2019 12:52 AM
> To: intel-wired-lan@lists.osuosl.org
> Cc: Topel, Bjorn <bjorn.topel@intel.com>; Karlsson, Magnus
> <magnus.karlsson@intel.com>; magnus.karlsson@gmail.com;
> netdev@vger.kernel.org; Sokolowski, Jan <jan.sokolowski@intel.com>
> Subject: [PATCH 2/2] i40e: add tracking of AF_XDP ZC state for each queue
> pair
>
> From: Björn Töpel <bjorn.topel@intel.com>
>
> In commit f3fef2b6e1cc ("i40e: Remove umem from VSI") a regression was
> introduced; When the VSI was reset, the setup code would try to enable
> AF_XDP ZC unconditionally (as long as there was a umem placed in the
> netdev._rx struct). Here, we add a bitmap to the VSI that tracks if a
> certain queue pair has been "zero-copy enabled" via the ndo_bpf. The
> bitmap is used in i40e_xsk_umem, and enables zero-copy if and only if
> XDP is enabled, the corresponding qid in the bitmap is set and the
> umem is non-NULL.
>
> Fixes: f3fef2b6e1cc ("i40e: Remove umem from VSI")
> Signed-off-by: Björn Töpel <bjorn.topel@intel.com>
> ---
> drivers/net/ethernet/intel/i40e/i40e.h | 2 ++
> drivers/net/ethernet/intel/i40e/i40e_main.c | 10 +++++++++-
> drivers/net/ethernet/intel/i40e/i40e_xsk.c | 3 +++
> 3 files changed, 14 insertions(+), 1 deletion(-)
>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
^ permalink raw reply
* [PATCH v2 net-next] mdio_bus: Fix PTR_ERR() usage after initialization to constant
From: YueHaibing @ 2019-02-16 2:59 UTC (permalink / raw)
To: davem, andrew, f.fainelli, hkallweit1; +Cc: linux-kernel, netdev, YueHaibing
Fix coccinelle warning:
./drivers/net/phy/mdio_bus.c:51:5-12: ERROR: PTR_ERR applied after initialization to constant on line 44
./drivers/net/phy/mdio_bus.c:52:5-12: ERROR: PTR_ERR applied after initialization to constant on line 44
fix this by using IS_ERR before PTR_ERR
Fixes: bafbdd527d56 ("phylib: Add device reset GPIO support")
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
---
v2: remove variable 'ret'
---
drivers/net/phy/mdio_bus.c | 11 ++++++-----
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c
index 3d31358..debc74c 100644
--- a/drivers/net/phy/mdio_bus.c
+++ b/drivers/net/phy/mdio_bus.c
@@ -48,11 +48,12 @@ static int mdiobus_register_gpiod(struct mdio_device *mdiodev)
gpiod = fwnode_get_named_gpiod(&mdiodev->dev.of_node->fwnode,
"reset-gpios", 0, GPIOD_OUT_LOW,
"PHY reset");
- if (PTR_ERR(gpiod) == -ENOENT ||
- PTR_ERR(gpiod) == -ENOSYS)
- gpiod = NULL;
- else if (IS_ERR(gpiod))
- return PTR_ERR(gpiod);
+ if (IS_ERR(gpiod)) {
+ if (PTR_ERR(gpiod) == -ENOENT || PTR_ERR(gpiod) == -ENOSYS)
+ gpiod = NULL;
+ else
+ return PTR_ERR(gpiod);
+ }
mdiodev->reset = gpiod;
--
2.7.4
^ permalink raw reply related
* Re: pull-request: mac80211 2019-02-15
From: David Miller @ 2019-02-16 3:42 UTC (permalink / raw)
To: johannes; +Cc: netdev, linux-wireless
In-Reply-To: <20190215125146.18856-1-johannes@sipsolutions.net>
From: Johannes Berg <johannes@sipsolutions.net>
Date: Fri, 15 Feb 2019 13:51:45 +0100
> It's clear things are winding down, this is basically just the stuff
> from Herbert that we've been discussing. I threw in a simple error
> path fix, mostly because it's simple :-)
>
> Please pull and let me know if there's any problem.
Ok, pulled, thanks Johannes.
^ permalink raw reply
* [PATCH bpf-next] tools/libbpf: support bigger BTF data sizes
From: Andrii Nakryiko @ 2019-02-16 3:52 UTC (permalink / raw)
To: andrii.nakryiko, netdev, bpf, kernel-team, yhs, ast, kafai,
daniel, acme
Cc: Andrii Nakryiko
While it's understandable why kernel limits number of BTF types to 65535
and size of string section to 64KB, in libbpf as user-space library it's
too restrictive. E.g., pahole converting DWARF to BTF type information
for Linux kernel generates more than 3 million BTF types and more than
3MB of strings, before deduplication. So to allow btf__dedup() to do its
work, we need to be able to load bigger BTF sections using btf__new().
Singed-off-by: Andrii Nakryiko <andriin@fb.com>
---
tools/lib/bpf/btf.c | 15 ++++++++-------
1 file changed, 8 insertions(+), 7 deletions(-)
diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index ade1c32fb083..68b50e9bbde1 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -16,7 +16,8 @@
#define max(a, b) ((a) > (b) ? (a) : (b))
#define min(a, b) ((a) < (b) ? (a) : (b))
-#define BTF_MAX_NR_TYPES 65535
+#define BTF_MAX_NR_TYPES 0x7fffffff
+#define BTF_MAX_STR_OFFSET 0x7fffffff
#define IS_MODIFIER(k) (((k) == BTF_KIND_TYPEDEF) || \
((k) == BTF_KIND_VOLATILE) || \
@@ -175,7 +176,7 @@ static int btf_parse_str_sec(struct btf *btf)
const char *start = btf->nohdr_data + hdr->str_off;
const char *end = start + btf->hdr->str_len;
- if (!hdr->str_len || hdr->str_len - 1 > BTF_MAX_NAME_OFFSET ||
+ if (!hdr->str_len || hdr->str_len - 1 > BTF_MAX_STR_OFFSET ||
start[0] || end[-1]) {
pr_debug("Invalid BTF string section\n");
return -EINVAL;
@@ -1882,7 +1883,7 @@ static int btf_dedup_prim_types(struct btf_dedup *d)
*/
static inline bool is_type_mapped(struct btf_dedup *d, uint32_t type_id)
{
- return d->map[type_id] <= BTF_MAX_TYPE;
+ return d->map[type_id] <= BTF_MAX_NR_TYPES;
}
/*
@@ -2033,7 +2034,7 @@ static int btf_dedup_is_equiv(struct btf_dedup *d, __u32 cand_id,
canon_id = resolve_fwd_id(d, canon_id);
hypot_type_id = d->hypot_map[canon_id];
- if (hypot_type_id <= BTF_MAX_TYPE)
+ if (hypot_type_id <= BTF_MAX_NR_TYPES)
return hypot_type_id == cand_id;
if (btf_dedup_hypot_map_add(d, canon_id, cand_id))
@@ -2252,7 +2253,7 @@ static int btf_dedup_struct_type(struct btf_dedup *d, __u32 type_id)
__u32 h;
/* already deduped or is in process of deduping (loop detected) */
- if (d->map[type_id] <= BTF_MAX_TYPE)
+ if (d->map[type_id] <= BTF_MAX_NR_TYPES)
return 0;
t = d->btf->types[type_id];
@@ -2329,7 +2330,7 @@ static int btf_dedup_ref_type(struct btf_dedup *d, __u32 type_id)
if (d->map[type_id] == BTF_IN_PROGRESS_ID)
return -ELOOP;
- if (d->map[type_id] <= BTF_MAX_TYPE)
+ if (d->map[type_id] <= BTF_MAX_NR_TYPES)
return resolve_type_id(d, type_id);
t = d->btf->types[type_id];
@@ -2509,7 +2510,7 @@ static int btf_dedup_remap_type_id(struct btf_dedup *d, __u32 type_id)
resolved_type_id = resolve_type_id(d, type_id);
new_type_id = d->hypot_map[resolved_type_id];
- if (new_type_id > BTF_MAX_TYPE)
+ if (new_type_id > BTF_MAX_NR_TYPES)
return -EINVAL;
return new_type_id;
}
--
2.17.1
^ permalink raw reply related
* Re: [PATCH net] net: ip6_gre: initialize erspan_ver just for erspan tunnels
From: David Miller @ 2019-02-16 4:15 UTC (permalink / raw)
To: lorenzo.bianconi; +Cc: netdev, petrm
In-Reply-To: <a08d2d1dbe3a498244867308adf37259bed30f37.1550239457.git.lorenzo.bianconi@redhat.com>
From: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Date: Fri, 15 Feb 2019 15:10:32 +0100
> After commit c706863bc890 ("net: ip6_gre: always reports o_key to
> userspace"), ip6gre and ip6gretap tunnels started reporting TUNNEL_KEY
> output flag even if it is not configured.
> ip6gre_fill_info checks erspan_ver value to add TUNNEL_KEY for
> erspan tunnels, however in commit 84581bdae9587 ("erspan: set
> erspan_ver to 1 by default when adding an erspan dev")
> erspan_ver is initialized to 1 even for ip6gre or ip6gretap
> Fix the issue moving erspan_ver initialization in a dedicated routine
>
> Fixes: c706863bc890 ("net: ip6_gre: always reports o_key to userspace")
> Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Applied, thanks.
^ permalink raw reply
* Re: [PATCH][next] mlxsw: core: fix spelling mistake "temprature" -> "temperature"
From: David Miller @ 2019-02-16 4:17 UTC (permalink / raw)
To: colin.king; +Cc: jiri, idosch, netdev, kernel-janitors, linux-kernel
In-Reply-To: <20190215151153.17206-1-colin.king@canonical.com>
From: Colin King <colin.king@canonical.com>
Date: Fri, 15 Feb 2019 15:11:53 +0000
> From: Colin Ian King <colin.king@canonical.com>
>
> There is a spelling mistake in several dev_err messages, fix these.
>
> Signed-off-by: Colin Ian King <colin.king@canonical.com>
Applied, thanks Colin.
^ permalink raw reply
* Re: [PATCH] net: phy: xgmiitorgmii: Support generic PHY status read
From: David Miller @ 2019-02-16 4:21 UTC (permalink / raw)
To: paul.kocialkowski
Cc: netdev, linux-arm-kernel, linux-kernel, andrew, f.fainelli,
hkallweit1, michal.simek, thomas.petazzoni
In-Reply-To: <20190215161708.18645-1-paul.kocialkowski@bootlin.com>
From: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
Date: Fri, 15 Feb 2019 17:17:08 +0100
> Some PHY drivers like the generic one do not provide a read_status
> callback on their own but rely on genphy_read_status being called
> directly.
>
> With the current code, this results in a NULL function pointer call.
> Call genphy_read_status instead when there is no specific callback.
>
> Signed-off-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
Applied, thanks.
Unfortunately I only noticed your updated version with the Fixes tag
after pushing this version out. I'll be more careful next time :)
^ permalink raw reply
* Re: [PATCH] net: Fix for_each_netdev_feature on Big endian
From: David Miller @ 2019-02-16 4:24 UTC (permalink / raw)
To: hauke.mehrtens; +Cc: netdev, jarod, hauke, stable
In-Reply-To: <20190215165854.25842-1-hauke.mehrtens@intel.com>
From: Hauke Mehrtens <hauke.mehrtens@intel.com>
Date: Fri, 15 Feb 2019 17:58:54 +0100
> The features attribute is of type u64 and stored in the native endianes on
> the system. The for_each_set_bit() macro takes a pointer to a 32 bit array
> and goes over the bits in this area. On little Endian systems this also
> works with an u64 as the most significant bit is on the highest address,
> but on big endian the words are swapped. When we expect bit 15 here we get
> bit 47 (15 + 32).
>
> This patch converts it more or less to its own for_each_set_bit()
> implementation which works on 64 bit integers directly. This is then
> completely in host endianness and should work like expected.
>
> Fixes: fd867d51f ("net/core: generic support for disabling netdev features down stack")
> Cc: stable@vger.kernel.org
> Signed-off-by: Hauke Mehrtens <hauke.mehrtens@intel.com>
Applied and queued up for -stable.
Please do not CC: stable for networking fixes, I handle -stable submissions
myself manually and this is documented in the netdev FAQ.
^ permalink raw reply
* Re: [PATCH net-next] net/ipv6: prefer rcu_access_pointer() over rcu_dereference()
From: David Miller @ 2019-02-16 4:26 UTC (permalink / raw)
To: pabeni; +Cc: netdev, dsahern
In-Reply-To: <52d4aee68b816ce36c0d4f4398ae143a2ebbdba4.1550249827.git.pabeni@redhat.com>
From: Paolo Abeni <pabeni@redhat.com>
Date: Fri, 15 Feb 2019 18:15:37 +0100
> rt6_cache_allowed_for_pmtu() checks for rt->from presence, but
> it does not access the RCU protected pointer. We can use
> rcu_access_pointer() and clean-up the code a bit. No functional
> changes intended.
>
> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Applied, thanks.
^ permalink raw reply
* Re: [PATCH net] net: validate untrusted gso packets without csum offload
From: David Miller @ 2019-02-16 4:31 UTC (permalink / raw)
To: willemdebruijn.kernel
Cc: netdev, edumazet, jasowang, maximmi, willemb, syzkaller
In-Reply-To: <20190215171547.247018-1-willemdebruijn.kernel@gmail.com>
From: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
Date: Fri, 15 Feb 2019 12:15:47 -0500
> From: Willem de Bruijn <willemb@google.com>
>
> Syzkaller again found a path to a kernel crash through bad gso input.
> By building an excessively large packet to cause an skb field to wrap.
>
> If VIRTIO_NET_HDR_F_NEEDS_CSUM was set this would have been dropped in
> skb_partial_csum_set.
>
> GSO packets that do not set checksum offload are suspicious and rare.
> Most callers of virtio_net_hdr_to_skb already pass them to
> skb_probe_transport_header.
>
> Move that test forward, change it to detect parse failure and drop
> packets on failure as those cleary are not one of the legitimate
> VIRTIO_NET_HDR_GSO types.
>
> Fixes: bfd5f4a3d605 ("packet: Add GSO/csum offload support.")
> Fixes: f43798c27684 ("tun: Allow GSO using virtio_net_hdr")
> Reported-by: syzbot <syzkaller@googlegroups.com>
> Signed-off-by: Willem de Bruijn <willemb@google.com>
Applied and queued up for -stable, thanks Willem.
> This captures a variety of bad gso packets, but to tighten further:
>
> - drop SKB_GSO_DODGY packets with ipip/sit/.. , which cannot be legal.
> by ipip_gso_segment wrappers around inet_gso_segment
> expands on 121d57af308d ("gso: validate gso_type in GSO handlers")
>
> - limit the number of ipv6 exthdrs allowed from dodgy sources.
> not sure where to draw the line. but not at 64K ;)
>
> - validate the network and transport protocol returned in
> skb_probe_transport_header against the VIRTIO_NET_HDR_GSO type
>
> - probe all dodgy GSO packets, also those that set checksum offload.
> this will have a performance impact, discussed previously in
> http://patchwork.ozlabs.org/patch/861874/
> but it would have blocked this latest bug as well
>
> All but the last one seem pretty uncontroversial to me. If no one
> objects I plan to send those to net-next.
No objections from me.
^ permalink raw reply
* Re: [PATCH net-next] selftests: forwarding: Add some missing configuration symbols
From: David Miller @ 2019-02-16 4:32 UTC (permalink / raw)
To: f.fainelli; +Cc: netdev, shuah, idosch, jiri, linux-kselftest, linux-kernel
In-Reply-To: <20190215181457.21101-1-f.fainelli@gmail.com>
From: Florian Fainelli <f.fainelli@gmail.com>
Date: Fri, 15 Feb 2019 10:14:52 -0800
> For the forwarding selftests to work, we need network namespaces when
> using veth/vrf otherwise ping/ping6 commands like these:
>
> ip vrf exec vveth0 /bin/ping 192.0.2.2 -c 10 -i 0.1 -w 5
>
> will fail because network namespaces may not be enabled.
>
> Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Applied.
^ permalink raw reply
* Re: [PATCH net-next 0/7] s390/qeth: updates 2019-02-15
From: David Miller @ 2019-02-16 4:35 UTC (permalink / raw)
To: jwi; +Cc: netdev, linux-s390, schwidefsky, heiko.carstens, raspl, ubraun
In-Reply-To: <20190215182231.90709-1-jwi@linux.ibm.com>
From: Julian Wiedmann <jwi@linux.ibm.com>
Date: Fri, 15 Feb 2019 19:22:24 +0100
> please apply a few more qeth patches to net-next. Along with some smaller
> improvements, this revamps our code for the SW statistics that are exposed
> through ETHTOOL_GSTATS.
Applied, thanks.
^ permalink raw reply
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