Re: [PATCH iwl-next v1] ice: fw and port health status

From
Tony Nguyen <anthony.l.nguyen@intel.com>
Date
2024-11-22 19:33:17
Also in
intel-wired-lan
Reply to
Konrad Knitter, 2024-11-18

On 11/18/2024 2:48 AM, Konrad Knitter wrote:
quoted hunk ↗ jump to hunk
Firmware generates events for global events or port specific events.

Driver shall subscribe for health status events from firmware on supported
FW versions >= 1.7.6.
Driver shall expose those under specific health reporter, two new
reporters are introduced:
- FW health reporter shall represent global events (problems with the
image, recovery mode);
- Port health reporter shall represent port-specific events (module
failure).

Firmware only reports problems when those are detected, it does not store
active fault list.
Driver will hold only last global and last port-specific event.
Driver will report all events via devlink health report,
so in case of multiple events of the same source they can be reviewed
using devlink autodump feature.

$ devlink health

pci/0000:b1:00.3:
   reporter fw
     state healthy error 0 recover 0 auto_dump true
   reporter port
     state error error 1 recover 0 last_dump_date 2024-03-17
	last_dump_time 09:29:29 auto_dump true

$ devlink health diagnose pci/0000:b1:00.3 reporter port

   Syndrome: 262
   Description: Module is not present.
   Possible Solution: Check that the module is inserted correctly.
   Port Number: 0

Tested on Intel Corporation Ethernet Controller E810-C for SFP

Co-developed-by: Sharon Haroni <redacted>
Signed-off-by: Sharon Haroni <redacted>
Co-developed-by: Nicholas Nunley <redacted>
Signed-off-by: Nicholas Nunley <redacted>
Co-developed-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Brett Creeley <brett.creeley@amd.com>
Signed-off-by: Konrad Knitter <redacted>
---
  .../net/ethernet/intel/ice/devlink/health.c   | 290 +++++++++++++++++-
  .../net/ethernet/intel/ice/devlink/health.h   |  12 +
  .../net/ethernet/intel/ice/ice_adminq_cmd.h   |  87 ++++++
  drivers/net/ethernet/intel/ice/ice_common.c   |  37 +++
  drivers/net/ethernet/intel/ice/ice_common.h   |   2 +
  drivers/net/ethernet/intel/ice/ice_main.c     |   3 +
  drivers/net/ethernet/intel/ice/ice_type.h     |   5 +
  7 files changed, 429 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/intel/ice/devlink/health.c b/drivers/net/ethernet/intel/ice/devlink/health.c
index c7a8b8c9e1ca..4e6c6891e207 100644
--- a/drivers/net/ethernet/intel/ice/devlink/health.c
+++ b/drivers/net/ethernet/intel/ice/devlink/health.c
@@ -1,13 +1,272 @@
  // SPDX-License-Identifier: GPL-2.0
  /* Copyright (c) 2024, Intel Corporation. */
  
-#include "health.h"
  #include "ice.h"
+#include "ice_adminq_cmd.h" /* for enum ice_aqc_health_status_elem */
+#include "health.h"
Is there a reason you're re-ordering health.h?
quoted hunk ↗ jump to hunk
  #include "ice_ethtool_common.h"
  
  #define ICE_DEVLINK_FMSG_PUT_FIELD(fmsg, obj, name) \
  	devlink_fmsg_put(fmsg, #name, (obj)->name)
  
+#define ICE_HEALTH_STATUS_DATA_SIZE 2
+
+struct ice_health_status {
+	enum ice_aqc_health_status code;
+	const char *description;
+	const char *solution;
+	const char *data_label[ICE_HEALTH_STATUS_DATA_SIZE];
+};
+
+/**
Wrong style, should be '/*'

drivers/net/ethernet/intel/ice/devlink/health.c:22: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
quoted hunk ↗ jump to hunk
+ * In addition to the health status codes provided below, the firmware might
+ * generate Health Status Codes that are not pertinent to the end-user.
+ * For instance, Health Code 0x1002 is triggered when the command fails.
+ * Such codes should be disregarded by the end-user.
+ * The below lookup requires to be sorted by code.
+ */
+
+static const char *const ice_common_port_solutions =
+	"Check your cable connection. Change or replace the module or cable. Manually set speed and duplex.";
+static const char *const ice_port_number_label = "Port Number";
+static const char *const ice_update_nvm_solution = "Update to the latest NVM image.";
...
quoted hunk ↗ jump to hunk
+static void ice_describe_status_code(struct devlink_fmsg *fmsg,
+				     struct ice_aqc_health_status_elem *hse)
+{
+	static const char *const aux_label[] = { "Aux Data 1", "Aux Data 2" };
+	const struct ice_health_status *health_code;
+	u32 internal_data[2];
+	u16 status_code;
+
+	status_code = le16_to_cpu(hse->health_status_code);
+
+	devlink_fmsg_put(fmsg, "Syndrome", status_code);
+	if (status_code != 0) {
if (status_code) {...
quoted hunk ↗ jump to hunk
+		internal_data[0] = le32_to_cpu(hse->internal_data1);
+		internal_data[1] = le32_to_cpu(hse->internal_data2);
+
+		health_code = ice_get_health_status(status_code);
+
+		if (!health_code)
+			return;
Please don't separate the error check with a newline. Other occurrences in this patch as well, please fix those too.
quoted hunk ↗ jump to hunk
+
+		devlink_fmsg_string_pair_put(fmsg, "Description", health_code->description);
+
+		if (health_code->solution)
+			devlink_fmsg_string_pair_put(fmsg, "Possible Solution",
+						     health_code->solution);
+
+		for (int i = 0; i < ICE_HEALTH_STATUS_DATA_SIZE; i++) {
+			if (internal_data[i] != ICE_AQC_HEALTH_STATUS_UNDEFINED_DATA)
+				devlink_fmsg_u32_pair_put(fmsg,
+							  health_code->data_label[i] ?
+							  health_code->data_label[i] :
+							  aux_label[i],
+							  internal_data[i]);
+		}
+	}
+}
+
...
quoted hunk ↗ jump to hunk
+void ice_process_health_status_event(struct ice_pf *pf, struct ice_rq_event_info *event)
+{
+	const struct ice_aqc_health_status_elem *health_info;
+	const struct ice_health_status *health_code;
+	u16 status_code, count;
+
+	health_info = (struct ice_aqc_health_status_elem *)event->msg_buf;
+	count = le16_to_cpu(event->desc.params.get_health_status.health_status_count);
+
+	if (count > (event->buf_len / sizeof(*health_info))) {
+		dev_err(ice_pf_to_dev(pf), "Received a health status event with invalid element count\n");
+		return;
+	}
+
+	for (int i = 0; i < count; i++) {
+		status_code = le16_to_cpu(health_info->health_status_code);
+		health_code = ice_get_health_status(status_code);
Looks like the scope of these vars can be reduced to this loop.
quoted hunk ↗ jump to hunk
+
+		if (health_code) {
+			switch (health_info->event_source) {
+			case ICE_AQC_HEALTH_STATUS_GLOBAL:
+				pf->health_reporters.fw_status = *health_info;
+				devlink_health_report(pf->health_reporters.fw,
+						      "FW syndrome reported", NULL);
+				break;
+			case ICE_AQC_HEALTH_STATUS_PF:
+			case ICE_AQC_HEALTH_STATUS_PORT:
+				pf->health_reporters.port_status = *health_info;
+				devlink_health_report(pf->health_reporters.port,
+						      "Port syndrome reported", NULL);
+				break;
+			default:
+				dev_err(ice_pf_to_dev(pf), "Health code with unknown source\n");
+			}
+		} else {
+			u32 data1, data2;
+			u16 source;
+
+			source = le16_to_cpu(health_info->event_source);
+			data1 = le32_to_cpu(health_info->internal_data1);
+			data2 = le32_to_cpu(health_info->internal_data2);
+			dev_dbg(ice_pf_to_dev(pf),
+				"Received internal health status code 0x%08x, source: 0x%08x, data1: 0x%08x, data2: 0x%08x",
+				status_code, source, data1, data2);
+		}
+		health_info++;
+	}
+}
...
quoted hunk ↗ jump to hunk
@@ -27,15 +29,21 @@ enum ice_mdd_src {
   * struct ice_health - stores ice devlink health reporters and accompanied data
   * @tx_hang: devlink health reporter for tx_hang event
   * @mdd: devlink health reporter for MDD detection event
+ * @fw: devlink health reporter for FW Health Status events
+ * @port: devlink health reporter for Port Health Status events
These should be in the order of the struct i.e. 'mdd' should be in-between.
quoted hunk ↗ jump to hunk
   * @tx_hang_buf: pre-allocated place to put info for Tx hang reporter from
   *               non-sleeping context
   * @tx_ring: ring that the hang occured on
   * @head: descriptior head
   * @intr: interrupt register value
   * @vsi_num: VSI owning the queue that the hang occured on
+ * @fw_status: buffer for last received FW Status event
+ * @port_status: buffer for last received Port Status event
   */
  struct ice_health {
+	struct devlink_health_reporter *fw;
  	struct devlink_health_reporter *mdd;
+	struct devlink_health_reporter *port;
  	struct devlink_health_reporter *tx_hang;
  	struct_group_tagged(ice_health_tx_hang_buf, tx_hang_buf,
  		struct ice_tx_ring *tx_ring;
...
+/**
+ * ice_is_fw_health_report_supported
drivers/net/ethernet/intel/ice/ice_common.c:6052: warning: missing initial short description on line:
 * ice_is_fw_health_report_supported
quoted hunk ↗ jump to hunk
+ * @hw: pointer to the hardware structure
+ *
+ * Return true if firmware supports health status reports,
Return isn't recognized, it should be Return:

drivers/net/ethernet/intel/ice/ice_common.c:6059: warning: No description found for return value of 'ice_is_fw_health_report_supported'

quoted hunk ↗ jump to hunk
+ * false otherwise
+ */
+bool ice_is_fw_health_report_supported(struct ice_hw *hw)
+{
+	return ice_is_fw_api_min_ver(hw, ICE_FW_API_HEALTH_REPORT_MAJ,
+				     ICE_FW_API_HEALTH_REPORT_MIN,
+				     ICE_FW_API_HEALTH_REPORT_PATCH);
+}
+
+/**
+ * ice_aq_set_health_status_cfg - Configure FW health events
+ * @hw: pointer to the HW struct
+ * @event_source: type of diagnostic events to enable
+ *
+ * Configure the health status event types that the firmware will send to this
+ * PF. The supported event types are: PF-specific, all PFs, and global.
+ * Return: 0 on success, negative error code otherwise.
IMO a newline separating the Return: would be make it easier to differentiate.

Thanks,
Tony
+ */
+int ice_aq_set_health_status_cfg(struct ice_hw *hw, u8 event_source)

Thread (3 messages) 3 messages, 3 authors, last reply 2024-11-22
Keyboard shortcuts
hback out one level
jnext message in thread / next series-diff position
kprevious message in thread / previous series-diff position
ldrill in
Escclose help / fold thread tree
?toggle this help