Skip to content

Commit a19b480

Browse files
Kalesh APdavem330
authored andcommitted
bnxt_en: Event handler for Thermal event
Newer FW will send a new async event when it detects that the chip's temperature has crossed the configured threshold value. The driver will now notify hwmon and will log a warning message. Link: https://lore.kernel.org/netdev/[email protected]/ Cc: Jean Delvare <[email protected]> Cc: Guenter Roeck <[email protected]> Cc: [email protected] Signed-off-by: Kalesh AP <[email protected]> Signed-off-by: Michael Chan <[email protected]> Acked-by: Guenter Roeck <[email protected]> Signed-off-by: David S. Miller <[email protected]>
1 parent 3d9cf96 commit a19b480

File tree

3 files changed

+82
-0
lines changed

3 files changed

+82
-0
lines changed

drivers/net/ethernet/broadcom/bnxt/bnxt.c

Lines changed: 52 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -2129,6 +2129,24 @@ static u16 bnxt_agg_ring_id_to_grp_idx(struct bnxt *bp, u16 ring_id)
21292129
return INVALID_HW_RING_ID;
21302130
}
21312131

2132+
#define BNXT_EVENT_THERMAL_CURRENT_TEMP(data2) \
2133+
((data2) & \
2134+
ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA2_CURRENT_TEMP_MASK)
2135+
2136+
#define BNXT_EVENT_THERMAL_THRESHOLD_TEMP(data2) \
2137+
(((data2) & \
2138+
ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA2_THRESHOLD_TEMP_MASK) >>\
2139+
ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA2_THRESHOLD_TEMP_SFT)
2140+
2141+
#define EVENT_DATA1_THERMAL_THRESHOLD_TYPE(data1) \
2142+
((data1) & \
2143+
ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_MASK)
2144+
2145+
#define EVENT_DATA1_THERMAL_THRESHOLD_DIR_INCREASING(data1) \
2146+
(((data1) & \
2147+
ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_TRANSITION_DIR) ==\
2148+
ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_TRANSITION_DIR_INCREASING)
2149+
21322150
static void bnxt_event_error_report(struct bnxt *bp, u32 data1, u32 data2)
21332151
{
21342152
u32 err_type = BNXT_EVENT_ERROR_REPORT_TYPE(data1);
@@ -2144,6 +2162,40 @@ static void bnxt_event_error_report(struct bnxt *bp, u32 data1, u32 data2)
21442162
case ASYNC_EVENT_CMPL_ERROR_REPORT_BASE_EVENT_DATA1_ERROR_TYPE_DOORBELL_DROP_THRESHOLD:
21452163
netdev_warn(bp->dev, "One or more MMIO doorbells dropped by the device!\n");
21462164
break;
2165+
case ASYNC_EVENT_CMPL_ERROR_REPORT_BASE_EVENT_DATA1_ERROR_TYPE_THERMAL_THRESHOLD: {
2166+
u32 type = EVENT_DATA1_THERMAL_THRESHOLD_TYPE(data1);
2167+
char *threshold_type;
2168+
char *dir_str;
2169+
2170+
switch (type) {
2171+
case ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_WARN:
2172+
threshold_type = "warning";
2173+
break;
2174+
case ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_CRITICAL:
2175+
threshold_type = "critical";
2176+
break;
2177+
case ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_FATAL:
2178+
threshold_type = "fatal";
2179+
break;
2180+
case ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_SHUTDOWN:
2181+
threshold_type = "shutdown";
2182+
break;
2183+
default:
2184+
netdev_err(bp->dev, "Unknown Thermal threshold type event\n");
2185+
return;
2186+
}
2187+
if (EVENT_DATA1_THERMAL_THRESHOLD_DIR_INCREASING(data1))
2188+
dir_str = "above";
2189+
else
2190+
dir_str = "below";
2191+
netdev_warn(bp->dev, "Chip temperature has gone %s the %s thermal threshold!\n",
2192+
dir_str, threshold_type);
2193+
netdev_warn(bp->dev, "Temperature (In Celsius), Current: %lu, threshold: %lu\n",
2194+
BNXT_EVENT_THERMAL_CURRENT_TEMP(data2),
2195+
BNXT_EVENT_THERMAL_THRESHOLD_TEMP(data2));
2196+
bnxt_hwmon_notify_event(bp, type);
2197+
break;
2198+
}
21472199
default:
21482200
netdev_err(bp->dev, "FW reported unknown error type %u\n",
21492201
err_type);

drivers/net/ethernet/broadcom/bnxt/bnxt_hwmon.c

Lines changed: 25 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -18,6 +18,31 @@
1818
#include "bnxt_hwrm.h"
1919
#include "bnxt_hwmon.h"
2020

21+
void bnxt_hwmon_notify_event(struct bnxt *bp, u32 type)
22+
{
23+
u32 attr;
24+
25+
if (!bp->hwmon_dev)
26+
return;
27+
28+
switch (type) {
29+
case ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_WARN:
30+
attr = hwmon_temp_max_alarm;
31+
break;
32+
case ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_CRITICAL:
33+
attr = hwmon_temp_crit_alarm;
34+
break;
35+
case ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_FATAL:
36+
case ASYNC_EVENT_CMPL_ERROR_REPORT_THERMAL_EVENT_DATA1_THRESHOLD_TYPE_SHUTDOWN:
37+
attr = hwmon_temp_emergency_alarm;
38+
break;
39+
default:
40+
return;
41+
}
42+
43+
hwmon_notify_event(&bp->pdev->dev, hwmon_temp, attr, 0);
44+
}
45+
2146
static int bnxt_hwrm_temp_query(struct bnxt *bp, u8 *temp)
2247
{
2348
struct hwrm_temp_monitor_query_output *resp;

drivers/net/ethernet/broadcom/bnxt/bnxt_hwmon.h

Lines changed: 5 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -11,9 +11,14 @@
1111
#define BNXT_HWMON_H
1212

1313
#ifdef CONFIG_BNXT_HWMON
14+
void bnxt_hwmon_notify_event(struct bnxt *bp, u32 type);
1415
void bnxt_hwmon_uninit(struct bnxt *bp);
1516
void bnxt_hwmon_init(struct bnxt *bp);
1617
#else
18+
static inline void bnxt_hwmon_notify_event(struct bnxt *bp, u32 type)
19+
{
20+
}
21+
1722
static inline void bnxt_hwmon_uninit(struct bnxt *bp)
1823
{
1924
}

0 commit comments

Comments
 (0)