net/mlx5: E-Switch, Load/unload VF reps according to event from host PF

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>

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);