Linux kernel mirror (for testing)
git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel
os
linux
Merge branch 'mlx5-add-sensor-name-in-temperature-message'
Tariq Toukan says:
====================
mlx5: Add sensor name in temperature message
This small series from Shahar adds the sensors names to the temperature
event messages, in addition to the existing bitmap indicators.
This improves human readability.
Series starts with simple refactoring and modifications. The top patch
adds the sensors names.
====================
Link: https://patch.msgid.link/20250213094641.226501-1-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
+39
-3
+33
-3
drivers/net/ethernet/mellanox/mlx5/core/events.c
+33
-3
drivers/net/ethernet/mellanox/mlx5/core/events.c
···
6
6
#include "mlx5_core.h"
7
7
#include "lib/eq.h"
8
8
#include "lib/events.h"
9
+
#include "hwmon.h"
9
10
10
11
struct mlx5_event_nb {
11
12
struct mlx5_nb nb;
···
154
153
return NOTIFY_OK;
155
154
}
156
155
156
+
#if IS_ENABLED(CONFIG_HWMON)
157
+
static void print_sensor_names_in_bit_set(struct mlx5_core_dev *dev, struct mlx5_hwmon *hwmon,
158
+
u64 bit_set, int bit_set_offset)
159
+
{
160
+
unsigned long *bit_set_ptr = (unsigned long *)&bit_set;
161
+
int num_bits = sizeof(bit_set) * BITS_PER_BYTE;
162
+
int i;
163
+
164
+
for_each_set_bit(i, bit_set_ptr, num_bits) {
165
+
const char *sensor_name = hwmon_get_sensor_name(hwmon, i + bit_set_offset);
166
+
167
+
mlx5_core_warn(dev, "Sensor name[%d]: %s\n", i + bit_set_offset, sensor_name);
168
+
}
169
+
}
170
+
#endif /* CONFIG_HWMON */
171
+
157
172
/* type == MLX5_EVENT_TYPE_TEMP_WARN_EVENT */
158
173
static int temp_warn(struct notifier_block *nb, unsigned long type, void *data)
159
174
{
160
175
struct mlx5_event_nb *event_nb = mlx5_nb_cof(nb, struct mlx5_event_nb, nb);
161
176
struct mlx5_events *events = event_nb->ctx;
177
+
struct mlx5_core_dev *dev = events->dev;
162
178
struct mlx5_eqe *eqe = data;
163
179
u64 value_lsb;
164
180
u64 value_msb;
165
181
166
182
value_lsb = be64_to_cpu(eqe->data.temp_warning.sensor_warning_lsb);
183
+
/* bit 1-63 are not supported for NICs,
184
+
* hence read only bit 0 (asic) from lsb.
185
+
*/
186
+
value_lsb &= 0x1;
167
187
value_msb = be64_to_cpu(eqe->data.temp_warning.sensor_warning_msb);
168
188
169
-
mlx5_core_warn(events->dev,
170
-
"High temperature on sensors with bit set %llx %llx",
171
-
value_msb, value_lsb);
189
+
if (net_ratelimit()) {
190
+
mlx5_core_warn(dev, "High temperature on sensors with bit set %#llx %#llx.\n",
191
+
value_msb, value_lsb);
192
+
#if IS_ENABLED(CONFIG_HWMON)
193
+
if (dev->hwmon) {
194
+
print_sensor_names_in_bit_set(dev, dev->hwmon, value_lsb, 0);
195
+
print_sensor_names_in_bit_set(dev, dev->hwmon, value_msb,
196
+
sizeof(value_lsb) * BITS_PER_BYTE);
197
+
}
198
+
#endif
199
+
}
172
200
173
201
return NOTIFY_OK;
174
202
}
+5
drivers/net/ethernet/mellanox/mlx5/core/hwmon.c
+5
drivers/net/ethernet/mellanox/mlx5/core/hwmon.c
+1
drivers/net/ethernet/mellanox/mlx5/core/hwmon.h
+1
drivers/net/ethernet/mellanox/mlx5/core/hwmon.h
···
10
10
11
11
int mlx5_hwmon_dev_register(struct mlx5_core_dev *mdev);
12
12
void mlx5_hwmon_dev_unregister(struct mlx5_core_dev *mdev);
13
+
const char *hwmon_get_sensor_name(struct mlx5_hwmon *hwmon, int channel);
13
14
14
15
#else
15
16
static inline int mlx5_hwmon_dev_register(struct mlx5_core_dev *mdev)