Linux kernel mirror (for testing)
git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel
os
linux
Merge branch 'mlx5-health-syndrome'
Tariq Toukan says:
====================
mlx5: Trust lockdown health syndrome
This series introduces a new error type in the health syndrome,
specifically for trust lock-down. Additionally, it exposes the CRR bit
in the health buffer, which, when set, indicates that the error cannot
be recovered without a process involving a cold reset. We add The CRR
bit value to the health buffer info log and update it to be logged on
any syndrome.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
+14
-1
+14
-1
drivers/net/ethernet/mellanox/mlx5/core/health.c
+14
-1
drivers/net/ethernet/mellanox/mlx5/core/health.c
···
96
96
return rfr_severity >> MLX5_RFR_BIT_OFFSET;
97
97
}
98
98
99
+
static int mlx5_health_get_crr(u8 rfr_severity)
100
+
{
101
+
return (rfr_severity >> MLX5_CRR_BIT_OFFSET) & 0x01;
102
+
}
103
+
99
104
static bool sensor_fw_synd_rfr(struct mlx5_core_dev *dev)
100
105
{
101
106
struct mlx5_core_health *health = &dev->priv.health;
···
380
375
return "High temperature";
381
376
case MLX5_INITIAL_SEG_HEALTH_SYNDROME_ICM_PCI_POISONED_ERR:
382
377
return "ICM fetch PCI data poisoned error";
378
+
case MLX5_INITIAL_SEG_HEALTH_SYNDROME_TRUST_LOCKDOWN_ERR:
379
+
return "Trust lockdown error";
383
380
default:
384
381
return "unrecognized error";
385
382
}
···
449
442
mlx5_log(dev, severity, "time %u\n", ioread32be(&h->time));
450
443
mlx5_log(dev, severity, "hw_id 0x%08x\n", ioread32be(&h->hw_id));
451
444
mlx5_log(dev, severity, "rfr %d\n", mlx5_health_get_rfr(rfr_severity));
445
+
mlx5_log(dev, severity, "crr %d\n", mlx5_health_get_crr(rfr_severity));
452
446
mlx5_log(dev, severity, "severity %d (%s)\n", severity, mlx5_loglevel_str(severity));
453
447
mlx5_log(dev, severity, "irisc_index %d\n", ioread8(&h->irisc_index));
454
448
mlx5_log(dev, severity, "synd 0x%x: %s\n", ioread8(&h->synd),
455
449
hsynd_str(ioread8(&h->synd)));
456
450
mlx5_log(dev, severity, "ext_synd 0x%04x\n", ioread16be(&h->ext_synd));
457
451
mlx5_log(dev, severity, "raw fw_ver 0x%08x\n", ioread32be(&h->fw_ver));
452
+
if (mlx5_health_get_crr(rfr_severity))
453
+
mlx5_core_warn(dev, "Cold reset is required\n");
458
454
}
459
455
460
456
static int
···
809
799
health->prev = count;
810
800
if (health->miss_counter == MAX_MISSES) {
811
801
mlx5_core_err(dev, "device's health compromised - reached miss count\n");
802
+
health->synd = ioread8(&h->synd);
812
803
print_health_info(dev);
813
804
queue_work(health->wq, &health->report_work);
814
805
}
815
806
816
807
prev_synd = health->synd;
817
808
health->synd = ioread8(&h->synd);
818
-
if (health->synd && health->synd != prev_synd)
809
+
if (health->synd && health->synd != prev_synd) {
810
+
print_health_info(dev);
819
811
queue_work(health->wq, &health->report_work);
812
+
}
820
813
821
814
out:
822
815
mod_timer(&health->timer, get_next_poll_jiffies(dev));