Merge branch 'Devlink-health-updates'

Eran Ben Elisha says:

====================
Devlink health updates

This patchset includes a fix [patch 01] to the devlink health state update, in
case recover was aborted.

In addition, it includes a small enhancement to the infrastructure in order to
allow direct state update in run-time, and use it from mlx5e tx reporter.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

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

@@ -2302,6 +2302,10 @@ int mlx5e_open_channels(struct mlx5e_priv *priv,
 			goto err_close_channels;
 	}
 
+	if (!IS_ERR_OR_NULL(priv->tx_reporter))
+		devlink_health_reporter_state_update(priv->tx_reporter,
+						     DEVLINK_HEALTH_REPORTER_STATE_HEALTHY);
+
 	kvfree(cparam);
 	return 0;
 

include/net/devlink.h

@@ -447,6 +447,11 @@ typedef void devlink_snapshot_data_dest_t(const void *data);
 struct devlink_fmsg;
 struct devlink_health_reporter;
 
+enum devlink_health_reporter_state {
+	DEVLINK_HEALTH_REPORTER_STATE_HEALTHY,
+	DEVLINK_HEALTH_REPORTER_STATE_ERROR,
+};
+
 /**
  * struct devlink_health_reporter_ops - Reporter operations
  * @name: reporter name
@@ -715,6 +720,9 @@ void *
 devlink_health_reporter_priv(struct devlink_health_reporter *reporter);
 int devlink_health_report(struct devlink_health_reporter *reporter,
 			  const char *msg, void *priv_ctx);
+void
+devlink_health_reporter_state_update(struct devlink_health_reporter *reporter,
+				     enum devlink_health_reporter_state state);
 
 void devlink_compat_running_version(struct net_device *dev,
 				    char *buf, size_t len);
@@ -1204,6 +1212,12 @@ devlink_health_report(struct devlink_health_reporter *reporter,
 	return 0;
 }
 
+static inline void
+devlink_health_reporter_state_update(struct devlink_health_reporter *reporter,
+				     enum devlink_health_reporter_state state)
+{
+}
+
 static inline void
 devlink_compat_running_version(struct net_device *dev, char *buf, size_t len)
 {

include/trace/events/devlink.h

@@ -140,6 +140,37 @@ TRACE_EVENT(devlink_health_recover_aborted,
 		  __entry->time_since_last_recover)
 );
 
+/*
+ * Tracepoint for devlink health reporter state update:
+ */
+TRACE_EVENT(devlink_health_reporter_state_update,
+	TP_PROTO(const struct devlink *devlink, const char *reporter_name,
+		 bool new_state),
+
+	TP_ARGS(devlink, reporter_name, new_state),
+
+	TP_STRUCT__entry(
+		__string(bus_name, devlink->dev->bus->name)
+		__string(dev_name, dev_name(devlink->dev))
+		__string(driver_name, devlink->dev->driver->name)
+		__string(reporter_name, reporter_name)
+		__field(u8, new_state)
+	),
+
+	TP_fast_assign(
+		__assign_str(bus_name, devlink->dev->bus->name);
+		__assign_str(dev_name, dev_name(devlink->dev));
+		__assign_str(driver_name, devlink->dev->driver->name);
+		__assign_str(reporter_name, reporter_name);
+		__entry->new_state = new_state;
+	),
+
+	TP_printk("bus_name=%s dev_name=%s driver_name=%s reporter_name=%s: new_state=%d",
+		  __get_str(bus_name), __get_str(dev_name),
+		  __get_str(driver_name), __get_str(reporter_name),
+		  __entry->new_state)
+);
+
 #endif /* _TRACE_DEVLINK_H */
 
 /* This part must be outside protection */

net/core/devlink.c

@@ -4409,11 +4409,6 @@ struct devlink_health_reporter {
 	u64 last_recovery_ts;
 };
 
-enum devlink_health_reporter_state {
-	DEVLINK_HEALTH_REPORTER_STATE_HEALTHY,
-	DEVLINK_HEALTH_REPORTER_STATE_ERROR,
-};
-
 void *
 devlink_health_reporter_priv(struct devlink_health_reporter *reporter)
 {
@@ -4498,6 +4493,23 @@ devlink_health_reporter_destroy(struct devlink_health_reporter *reporter)
 }
 EXPORT_SYMBOL_GPL(devlink_health_reporter_destroy);
 
+void
+devlink_health_reporter_state_update(struct devlink_health_reporter *reporter,
+				     enum devlink_health_reporter_state state)
+{
+	if (WARN_ON(state != DEVLINK_HEALTH_REPORTER_STATE_HEALTHY &&
+		    state != DEVLINK_HEALTH_REPORTER_STATE_ERROR))
+		return;
+
+	if (reporter->health_state == state)
+		return;
+
+	reporter->health_state = state;
+	trace_devlink_health_reporter_state_update(reporter->devlink,
+						   reporter->ops->name, state);
+}
+EXPORT_SYMBOL_GPL(devlink_health_reporter_state_update);
+
 static int
 devlink_health_reporter_recover(struct devlink_health_reporter *reporter,
 				void *priv_ctx)
@@ -4569,16 +4581,19 @@ static int devlink_health_do_dump(struct devlink_health_reporter *reporter,
 int devlink_health_report(struct devlink_health_reporter *reporter,
 			  const char *msg, void *priv_ctx)
 {
+	enum devlink_health_reporter_state prev_health_state;
 	struct devlink *devlink = reporter->devlink;
 
 	/* write a log message of the current error */
 	WARN_ON(!msg);
 	trace_devlink_health_report(devlink, reporter->ops->name, msg);
 	reporter->error_count++;
+	prev_health_state = reporter->health_state;
+	reporter->health_state = DEVLINK_HEALTH_REPORTER_STATE_ERROR;
 
 	/* abort if the previous error wasn't recovered */
 	if (reporter->auto_recover &&
-	    (reporter->health_state != DEVLINK_HEALTH_REPORTER_STATE_HEALTHY ||
+	    (prev_health_state != DEVLINK_HEALTH_REPORTER_STATE_HEALTHY ||
 	     jiffies - reporter->last_recovery_ts <
 	     msecs_to_jiffies(reporter->graceful_period))) {
 		trace_devlink_health_recover_aborted(devlink,