net/mlx5e: Add support to rx reporter diagnose

Add rx reporter, which supports diagnose call-back. Diagnostics output
include: information common to all RQs: RQ type, RQ size, RQ stride
size, CQ size and CQ stride size. In addition advertise information per
RQ and its related icosq and attached CQ.

$ devlink health diagnose pci/0000:00:0b.0 reporter rx
 Common config:
   RQ:
     type: 2 stride size: 2048 size: 8
   CQ:
     stride size: 64 size: 1024
 RQs:
   channel ix: 0 rqn: 4308 HW state: 1 SW state: 3 posted WQEs: 7 cc: 7 ICOSQ HW state: 1
   CQ:
     cqn: 1032 HW status: 0
   channel ix: 1 rqn: 4313 HW state: 1 SW state: 3 posted WQEs: 7 cc: 7 ICOSQ HW state: 1
   CQ:
     cqn: 1036 HW status: 0
   channel ix: 2 rqn: 4318 HW state: 1 SW state: 3 posted WQEs: 7 cc: 7 ICOSQ HW state: 1
   CQ:
     cqn: 1040 HW status: 0
   channel ix: 3 rqn: 4323 HW state: 1 SW state: 3 posted WQEs: 7 cc: 7 ICOSQ HW state: 1
   CQ:
     cqn: 1044 HW status: 0

$ devlink health diagnose pci/0000:00:0b.0 reporter rx -jp
{
    "Common config": {
        "RQ": {
            "type": 2,
            "stride size": 2048,
            "size": 8
        },
        "CQ": {
            "stride size": 64,
            "size": 1024
        }
    },
    "RQs": [ {
            "channel ix": 0,
            "rqn": 4308,
            "HW state": 1,
            "SW state": 3,
            "posted WQEs": 7,
            "cc": 7,
            "ICOSQ HW state": 1,
            "CQ": {
                "cqn": 1032,
                "HW status": 0
            }
        },{
            "channel ix": 1,
            "rqn": 4313,
            "HW state": 1,
            "SW state": 3,
            "posted WQEs": 7,
            "cc": 7,
            "ICOSQ HW state": 1,
            "CQ": {
                "cqn": 1036,
                "HW status": 0
            }
        },{
            "channel ix": 2,
            "rqn": 4318,
            "HW state": 1,
            "SW state": 3,
            "posted WQEs": 7,
            "cc": 7,
            "ICOSQ HW state": 1,
            "CQ": {
                "cqn": 1040,
                "HW status": 0
            }
        },{
            "channel ix": 3,
            "rqn": 4323,
            "HW state": 1,
            "SW state": 3,
            "posted WQEs": 7,
            "cc": 7,
            "ICOSQ HW state": 1,
            "CQ": {
                "cqn": 1044,
                "HW status": 0
            }
        } ]
}
Signed-off-by: Aya Levin <ayal@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

drivers/net/ethernet/mellanox/mlx5/core/Makefile

@@ -24,8 +24,8 @@ mlx5_core-y :=	main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \
 mlx5_core-$(CONFIG_MLX5_CORE_EN) += en_main.o en_common.o en_fs.o en_ethtool.o \
 		en_tx.o en_rx.o en_dim.o en_txrx.o en/xdp.o en_stats.o \
 		en_selftest.o en/port.o en/monitor_stats.o en/health.o \
-		en/reporter_tx.o en/params.o en/xsk/umem.o en/xsk/setup.o \
-		en/xsk/rx.o en/xsk/tx.o
+		en/reporter_tx.o en/reporter_rx.o en/params.o en/xsk/umem.o \
+		en/xsk/setup.o en/xsk/rx.o en/xsk/tx.o
 
 #
 # Netdev extra

drivers/net/ethernet/mellanox/mlx5/core/en.h

@@ -846,6 +846,7 @@ struct mlx5e_priv {
 	struct mlx5e_tls          *tls;
 #endif
 	struct devlink_health_reporter *tx_reporter;
+	struct devlink_health_reporter *rx_reporter;
 	struct mlx5e_xsk           xsk;
 };
 
@@ -887,6 +888,26 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget);
 int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget);
 void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq);
 
+static inline u32 mlx5e_rqwq_get_size(struct mlx5e_rq *rq)
+{
+	switch (rq->wq_type) {
+	case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ:
+		return mlx5_wq_ll_get_size(&rq->mpwqe.wq);
+	default:
+		return mlx5_wq_cyc_get_size(&rq->wqe.wq);
+	}
+}
+
+static inline u32 mlx5e_rqwq_get_cur_sz(struct mlx5e_rq *rq)
+{
+	switch (rq->wq_type) {
+	case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ:
+		return rq->mpwqe.wq.cur_sz;
+	default:
+		return rq->wqe.wq.cur_sz;
+	}
+}
+
 bool mlx5e_check_fragmented_striding_rq_cap(struct mlx5_core_dev *mdev);
 bool mlx5e_striding_rq_possible(struct mlx5_core_dev *mdev,
 				struct mlx5e_params *params);

drivers/net/ethernet/mellanox/mlx5/core/en/health.c

@@ -98,11 +98,22 @@ int mlx5e_reporter_cq_common_diagnose(struct mlx5e_cq *cq, struct devlink_fmsg *
 
 int mlx5e_health_create_reporters(struct mlx5e_priv *priv)
 {
-	return  mlx5e_reporter_tx_create(priv);
+	int err;
+
+	err = mlx5e_reporter_tx_create(priv);
+	if (err)
+		return err;
+
+	err = mlx5e_reporter_rx_create(priv);
+	if (err)
+		return err;
+
+	return 0;
 }
 
 void mlx5e_health_destroy_reporters(struct mlx5e_priv *priv)
 {
+	mlx5e_reporter_rx_destroy(priv);
 	mlx5e_reporter_tx_destroy(priv);
 }
 
@@ -111,6 +122,9 @@ void mlx5e_health_channels_update(struct mlx5e_priv *priv)
 	if (priv->tx_reporter)
 		devlink_health_reporter_state_update(priv->tx_reporter,
 						     DEVLINK_HEALTH_REPORTER_STATE_HEALTHY);
+	if (priv->rx_reporter)
+		devlink_health_reporter_state_update(priv->rx_reporter,
+						     DEVLINK_HEALTH_REPORTER_STATE_HEALTHY);
 }
 
 int mlx5e_health_sq_to_ready(struct mlx5e_channel *channel, u32 sqn)

drivers/net/ethernet/mellanox/mlx5/core/en/health.h

@@ -16,6 +16,9 @@ int mlx5e_reporter_cq_common_diagnose(struct mlx5e_cq *cq, struct devlink_fmsg *
 int mlx5e_reporter_named_obj_nest_start(struct devlink_fmsg *fmsg, char *name);
 int mlx5e_reporter_named_obj_nest_end(struct devlink_fmsg *fmsg);
 
+int mlx5e_reporter_rx_create(struct mlx5e_priv *priv);
+void mlx5e_reporter_rx_destroy(struct mlx5e_priv *priv);
+
 #define MLX5E_REPORTER_PER_Q_MAX_LEN 256
 
 struct mlx5e_err_ctx {

drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c

+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2019 Mellanox Technologies.
+
+#include "health.h"
+#include "params.h"
+
+static int mlx5e_query_rq_state(struct mlx5_core_dev *dev, u32 rqn, u8 *state)
+{
+	int outlen = MLX5_ST_SZ_BYTES(query_rq_out);
+	void *out;
+	void *rqc;
+	int err;
+
+	out = kvzalloc(outlen, GFP_KERNEL);
+	if (!out)
+		return -ENOMEM;
+
+	err = mlx5_core_query_rq(dev, rqn, out);
+	if (err)
+		goto out;
+
+	rqc = MLX5_ADDR_OF(query_rq_out, out, rq_context);
+	*state = MLX5_GET(rqc, rqc, state);
+
+out:
+	kvfree(out);
+	return err;
+}
+
+static int mlx5e_rx_reporter_build_diagnose_output(struct mlx5e_rq *rq,
+						   struct devlink_fmsg *fmsg)
+{
+	struct mlx5e_priv *priv = rq->channel->priv;
+	struct mlx5e_params *params;
+	struct mlx5e_icosq *icosq;
+	u8 icosq_hw_state;
+	int wqes_sz;
+	u8 hw_state;
+	u16 wq_head;
+	int err;
+
+	params = &priv->channels.params;
+	icosq = &rq->channel->icosq;
+	err = mlx5e_query_rq_state(priv->mdev, rq->rqn, &hw_state);
+	if (err)
+		return err;
+
+	err = mlx5_core_query_sq_state(priv->mdev, icosq->sqn, &icosq_hw_state);
+	if (err)
+		return err;
+
+	wqes_sz = mlx5e_rqwq_get_cur_sz(rq);
+	wq_head = params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ ?
+		  rq->mpwqe.wq.head : mlx5_wq_cyc_get_head(&rq->wqe.wq);
+
+	err = devlink_fmsg_obj_nest_start(fmsg);
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u32_pair_put(fmsg, "channel ix", rq->channel->ix);
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u32_pair_put(fmsg, "rqn", rq->rqn);
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u8_pair_put(fmsg, "HW state", hw_state);
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u8_pair_put(fmsg, "SW state", rq->state);
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u32_pair_put(fmsg, "posted WQEs", wqes_sz);
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u32_pair_put(fmsg, "cc", wq_head);
+	if (err)
+		return err;
+
+	err = devlink_fmsg_u8_pair_put(fmsg, "ICOSQ HW state", icosq_hw_state);
+	if (err)
+		return err;
+
+	err = mlx5e_reporter_cq_diagnose(&rq->cq, fmsg);
+	if (err)
+		return err;
+
+	err = devlink_fmsg_obj_nest_end(fmsg);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static int mlx5e_rx_reporter_diagnose(struct devlink_health_reporter *reporter,
+				      struct devlink_fmsg *fmsg)
+{
+	struct mlx5e_priv *priv = devlink_health_reporter_priv(reporter);
+	struct mlx5e_params *params = &priv->channels.params;
+	struct mlx5e_rq *generic_rq;
+	u32 rq_stride, rq_sz;
+	int i, err = 0;
+
+	mutex_lock(&priv->state_lock);
+
+	if (!test_bit(MLX5E_STATE_OPENED, &priv->state))
+		goto unlock;
+
+	generic_rq = &priv->channels.c[0]->rq;
+	rq_sz = mlx5e_rqwq_get_size(generic_rq);
+	rq_stride = BIT(mlx5e_mpwqe_get_log_stride_size(priv->mdev, params, NULL));
+
+	err = mlx5e_reporter_named_obj_nest_start(fmsg, "Common config");
+	if (err)
+		goto unlock;
+
+	err = mlx5e_reporter_named_obj_nest_start(fmsg, "RQ");
+	if (err)
+		goto unlock;
+
+	err = devlink_fmsg_u8_pair_put(fmsg, "type", params->rq_wq_type);
+	if (err)
+		goto unlock;
+
+	err = devlink_fmsg_u64_pair_put(fmsg, "stride size", rq_stride);
+	if (err)
+		goto unlock;
+
+	err = devlink_fmsg_u32_pair_put(fmsg, "size", rq_sz);
+	if (err)
+		goto unlock;
+
+	err = mlx5e_reporter_named_obj_nest_end(fmsg);
+	if (err)
+		goto unlock;
+
+	err = mlx5e_reporter_cq_common_diagnose(&generic_rq->cq, fmsg);
+	if (err)
+		goto unlock;
+
+	err = mlx5e_reporter_named_obj_nest_end(fmsg);
+	if (err)
+		goto unlock;
+
+	err = devlink_fmsg_arr_pair_nest_start(fmsg, "RQs");
+	if (err)
+		goto unlock;
+
+	for (i = 0; i < priv->channels.num; i++) {
+		struct mlx5e_rq *rq = &priv->channels.c[i]->rq;
+
+		err = mlx5e_rx_reporter_build_diagnose_output(rq, fmsg);
+		if (err)
+			goto unlock;
+	}
+	err = devlink_fmsg_arr_pair_nest_end(fmsg);
+	if (err)
+		goto unlock;
+unlock:
+	mutex_unlock(&priv->state_lock);
+	return err;
+}
+
+static const struct devlink_health_reporter_ops mlx5_rx_reporter_ops = {
+	.name = "rx",
+	.diagnose = mlx5e_rx_reporter_diagnose,
+};
+
+int mlx5e_reporter_rx_create(struct mlx5e_priv *priv)
+{
+	struct devlink *devlink = priv_to_devlink(priv->mdev);
+	struct devlink_health_reporter *reporter;
+
+	reporter = devlink_health_reporter_create(devlink,
+						  &mlx5_rx_reporter_ops,
+						  0, false, priv);
+	if (IS_ERR(reporter)) {
+		netdev_warn(priv->netdev, "Failed to create rx reporter, err = %ld\n",
+			    PTR_ERR(reporter));
+		return PTR_ERR(reporter);
+	}
+	priv->rx_reporter = reporter;
+	return 0;
+}
+
+void mlx5e_reporter_rx_destroy(struct mlx5e_priv *priv)
+{
+	if (!priv->rx_reporter)
+		return;
+
+	devlink_health_reporter_destroy(priv->rx_reporter);
+}

drivers/net/ethernet/mellanox/mlx5/core/en_main.c

@@ -247,26 +247,6 @@ static inline void mlx5e_build_umr_wqe(struct mlx5e_rq *rq,
 	ucseg->mkey_mask     = cpu_to_be64(MLX5_MKEY_MASK_FREE);
 }
 
-static u32 mlx5e_rqwq_get_size(struct mlx5e_rq *rq)
-{
-	switch (rq->wq_type) {
-	case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ:
-		return mlx5_wq_ll_get_size(&rq->mpwqe.wq);
-	default:
-		return mlx5_wq_cyc_get_size(&rq->wqe.wq);
-	}
-}
-
-static u32 mlx5e_rqwq_get_cur_sz(struct mlx5e_rq *rq)
-{
-	switch (rq->wq_type) {
-	case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ:
-		return rq->mpwqe.wq.cur_sz;
-	default:
-		return rq->wqe.wq.cur_sz;
-	}
-}
-
 static int mlx5e_rq_alloc_mpwqe_info(struct mlx5e_rq *rq,
 				     struct mlx5e_channel *c)
 {