Commit 051868d9 authored by farah kassabri's avatar farah kassabri Committed by Oded Gabbay

accel/habanalabs: prevent sending heartbeat before events are enabled

After the heartbeat mechanism is now expanded to be used also
for EQ health check, we shouldn't send heartbeat messages
to FW before driver allow events to be received from FW.

Because if the driver will send two heartbeats before it enables
events to be received from FW, then the EQ health check
will fail and reset the device.
Signed-off-by: default avatarfarah kassabri <fkassabri@habana.ai>
Reviewed-by: default avatarOded Gabbay <ogabbay@kernel.org>
Signed-off-by: default avatarOded Gabbay <ogabbay@kernel.org>
parent 764bfd13
......@@ -994,12 +994,7 @@ static void hl_device_eq_heartbeat(struct hl_device *hdev)
u64 event_mask = HL_NOTIFIER_EVENT_DEVICE_RESET | HL_NOTIFIER_EVENT_DEVICE_UNAVAILABLE;
struct asic_fixed_properties *prop = &hdev->asic_prop;
/*
* This feature supported in FW version 1.12.0 45.2.0 and above,
* only on those FW versions eq_health_check_supported will be set.
* Start checking eq health only after driver has enabled events from FW.
*/
if (!prop->cpucp_info.eq_health_check_supported || !hdev->init_done)
if (!prop->cpucp_info.eq_health_check_supported)
return;
if (hdev->eq_heartbeat_received)
......@@ -1015,7 +1010,8 @@ static void hl_device_heartbeat(struct work_struct *work)
struct hl_info_fw_err_info info = {0};
u64 event_mask = HL_NOTIFIER_EVENT_DEVICE_RESET | HL_NOTIFIER_EVENT_DEVICE_UNAVAILABLE;
if (!hl_device_operational(hdev, NULL))
/* Start heartbeat checks only after driver has enabled events from FW */
if (!hl_device_operational(hdev, NULL) || !hdev->init_done)
goto reschedule;
/*
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment