diff options
| author | farah kassabri <fkassabri@habana.ai> | 2023-08-08 12:56:47 +0300 |
|---|---|---|
| committer | Oded Gabbay <ogabbay@kernel.org> | 2023-10-09 12:37:21 +0300 |
| commit | 7c4130e6ddd709be2033a6635c91d445cb2baea5 (patch) | |
| tree | bdf924dfe7763fad06a826e80e115c6b8e9cf390 /drivers/accel/habanalabs/gaudi2 | |
| parent | 72bff371b2e26a0088d4cb1b44aee20c77e423ba (diff) | |
| download | lwn-7c4130e6ddd709be2033a6635c91d445cb2baea5.tar.gz lwn-7c4130e6ddd709be2033a6635c91d445cb2baea5.zip | |
accel/habanalabs/gaudi2: handle eq health heartbeat check
Add mechanism for fw eq health check. this will be done using two flows:
using the heartbeat mechanism and raising a dedicated interrupt to
indicate an eq failure like EQ full.
This patch will add implementation for the eq heartbeat for gaudi2 asic.
More info about the heartbeat mechanism:
Expand the heartbeat mechanism to monitor a new event that
will be sent from FW upon receiving heartbeat message.
that way driver can know that the eq is working or not.
Signed-off-by: farah kassabri <fkassabri@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Diffstat (limited to 'drivers/accel/habanalabs/gaudi2')
| -rw-r--r-- | drivers/accel/habanalabs/gaudi2/gaudi2.c | 10 |
1 files changed, 10 insertions, 0 deletions
diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c index 677900e18519..e507847bf460 100644 --- a/drivers/accel/habanalabs/gaudi2/gaudi2.c +++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c @@ -7804,6 +7804,7 @@ static inline bool is_info_event(u32 event) * an indication to an error. */ case GAUDI2_EVENT_CPU0_STATUS_NIC0_ENG0 ... GAUDI2_EVENT_CPU11_STATUS_NIC11_ENG1: + case GAUDI2_EVENT_ARC_EQ_HEARTBEAT: return true; default: return false; @@ -9765,6 +9766,11 @@ static u16 event_id_to_engine_id(struct hl_device *hdev, u16 event_type) return U16_MAX; } +static void hl_eq_heartbeat_event_handle(struct hl_device *hdev) +{ + hdev->eq_heartbeat_received = true; +} + static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_entry) { struct gaudi2_device *gaudi2 = hdev->asic_specific; @@ -10190,6 +10196,10 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent gaudi2_irq_map_table[event_type].name); break; + case GAUDI2_EVENT_ARC_EQ_HEARTBEAT: + hl_eq_heartbeat_event_handle(hdev); + error_count = GAUDI2_NA_EVENT_CAUSE; + break; default: if (gaudi2_irq_map_table[event_type].valid) { dev_err_ratelimited(hdev->dev, "Cannot find handler for event %d\n", |