Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

14 files changed, 291 insertions, 203 deletions

diff --git a/drivers/misc/habanalabs/common/command_submission.c b/drivers/misc/habanalabs/common/command_submission.c
index beb482310a58..b2b3d2b0f808 100644
--- a/drivers/misc/habanalabs/common/command_submission.c
+++ b/drivers/misc/habanalabs/common/command_submission.c
@@ -472,8 +472,11 @@ static int allocate_cs(struct hl_device *hdev, struct hl_ctx *ctx,
 	cntr = &hdev->aggregated_cs_counters;
 
 	cs = kzalloc(sizeof(*cs), GFP_ATOMIC);
-	if (!cs)
+	if (!cs) {
+		atomic64_inc(&ctx->cs_counters.out_of_mem_drop_cnt);
+		atomic64_inc(&cntr->out_of_mem_drop_cnt);
 		return -ENOMEM;
+	}
 
 	cs->ctx = ctx;
 	cs->submitted = false;
@@ -486,6 +489,8 @@ static int allocate_cs(struct hl_device *hdev, struct hl_ctx *ctx,
 
 	cs_cmpl = kmalloc(sizeof(*cs_cmpl), GFP_ATOMIC);
 	if (!cs_cmpl) {
+		atomic64_inc(&ctx->cs_counters.out_of_mem_drop_cnt);
+		atomic64_inc(&cntr->out_of_mem_drop_cnt);
 		rc = -ENOMEM;
 		goto free_cs;
 	}
@@ -513,6 +518,8 @@ static int allocate_cs(struct hl_device *hdev, struct hl_ctx *ctx,
 	cs->jobs_in_queue_cnt = kcalloc(hdev->asic_prop.max_queues,
 			sizeof(*cs->jobs_in_queue_cnt), GFP_ATOMIC);
 	if (!cs->jobs_in_queue_cnt) {
+		atomic64_inc(&ctx->cs_counters.out_of_mem_drop_cnt);
+		atomic64_inc(&cntr->out_of_mem_drop_cnt);
 		rc = -ENOMEM;
 		goto free_fence;
 	}
@@ -562,7 +569,7 @@ void hl_cs_rollback_all(struct hl_device *hdev)
 	for (i = 0 ; i < hdev->asic_prop.completion_queues_count ; i++)
 		flush_workqueue(hdev->cq_wq[i]);
 
-	/* Make sure we don't have leftovers in the H/W queues mirror list */
+	/* Make sure we don't have leftovers in the CS mirror list */
 	list_for_each_entry_safe(cs, tmp, &hdev->cs_mirror_list, mirror_node) {
 		cs_get(cs);
 		cs->aborted = true;
@@ -764,11 +771,14 @@ static int hl_cs_sanity_checks(struct hl_fpriv *hpriv, union hl_cs_args *args)
 
 static int hl_cs_copy_chunk_array(struct hl_device *hdev,
 					struct hl_cs_chunk **cs_chunk_array,
-					void __user *chunks, u32 num_chunks)
+					void __user *chunks, u32 num_chunks,
+					struct hl_ctx *ctx)
 {
 	u32 size_to_copy;
 
 	if (num_chunks > HL_MAX_JOBS_PER_CS) {
+		atomic64_inc(&ctx->cs_counters.validation_drop_cnt);
+		atomic64_inc(&hdev->aggregated_cs_counters.validation_drop_cnt);
 		dev_err(hdev->dev,
 			"Number of chunks can NOT be larger than %d\n",
 			HL_MAX_JOBS_PER_CS);
@@ -777,11 +787,16 @@ static int hl_cs_copy_chunk_array(struct hl_device *hdev,
 
 	*cs_chunk_array = kmalloc_array(num_chunks, sizeof(**cs_chunk_array),
 					GFP_ATOMIC);
-	if (!*cs_chunk_array)
+	if (!*cs_chunk_array) {
+		atomic64_inc(&ctx->cs_counters.out_of_mem_drop_cnt);
+		atomic64_inc(&hdev->aggregated_cs_counters.out_of_mem_drop_cnt);
 		return -ENOMEM;
+	}
 
 	size_to_copy = num_chunks * sizeof(struct hl_cs_chunk);
 	if (copy_from_user(*cs_chunk_array, chunks, size_to_copy)) {
+		atomic64_inc(&ctx->cs_counters.validation_drop_cnt);
+		atomic64_inc(&hdev->aggregated_cs_counters.validation_drop_cnt);
 		dev_err(hdev->dev, "Failed to copy cs chunk array from user\n");
 		kfree(*cs_chunk_array);
 		return -EFAULT;
@@ -797,6 +812,7 @@ static int cs_ioctl_default(struct hl_fpriv *hpriv, void __user *chunks,
 	struct hl_device *hdev = hpriv->hdev;
 	struct hl_cs_chunk *cs_chunk_array;
 	struct hl_cs_counters_atomic *cntr;
+	struct hl_ctx *ctx = hpriv->ctx;
 	struct hl_cs_job *job;
 	struct hl_cs *cs;
 	struct hl_cb *cb;
@@ -805,7 +821,8 @@ static int cs_ioctl_default(struct hl_fpriv *hpriv, void __user *chunks,
 	cntr = &hdev->aggregated_cs_counters;
 	*cs_seq = ULLONG_MAX;
 
-	rc = hl_cs_copy_chunk_array(hdev, &cs_chunk_array, chunks, num_chunks);
+	rc = hl_cs_copy_chunk_array(hdev, &cs_chunk_array, chunks, num_chunks,
+			hpriv->ctx);
 	if (rc)
 		goto out;
 
@@ -832,8 +849,8 @@ static int cs_ioctl_default(struct hl_fpriv *hpriv, void __user *chunks,
 		rc = validate_queue_index(hdev, chunk, &queue_type,
 						&is_kernel_allocated_cb);
 		if (rc) {
-			atomic64_inc(&hpriv->ctx->cs_counters.parsing_drop_cnt);
-			atomic64_inc(&cntr->parsing_drop_cnt);
+			atomic64_inc(&ctx->cs_counters.validation_drop_cnt);
+			atomic64_inc(&cntr->validation_drop_cnt);
 			goto free_cs_object;
 		}
 
@@ -841,8 +858,8 @@ static int cs_ioctl_default(struct hl_fpriv *hpriv, void __user *chunks,
 			cb = get_cb_from_cs_chunk(hdev, &hpriv->cb_mgr, chunk);
 			if (!cb) {
 				atomic64_inc(
-				&hpriv->ctx->cs_counters.parsing_drop_cnt);
-				atomic64_inc(&cntr->parsing_drop_cnt);
+					&ctx->cs_counters.validation_drop_cnt);
+				atomic64_inc(&cntr->validation_drop_cnt);
 				rc = -EINVAL;
 				goto free_cs_object;
 			}
@@ -856,8 +873,7 @@ static int cs_ioctl_default(struct hl_fpriv *hpriv, void __user *chunks,
 		job = hl_cs_allocate_job(hdev, queue_type,
 						is_kernel_allocated_cb);
 		if (!job) {
-			atomic64_inc(
-			&hpriv->ctx->cs_counters.out_of_mem_drop_cnt);
+			atomic64_inc(&ctx->cs_counters.out_of_mem_drop_cnt);
 			atomic64_inc(&cntr->out_of_mem_drop_cnt);
 			dev_err(hdev->dev, "Failed to allocate a new job\n");
 			rc = -ENOMEM;
@@ -891,7 +907,7 @@ static int cs_ioctl_default(struct hl_fpriv *hpriv, void __user *chunks,
 
 		rc = cs_parser(hpriv, job);
 		if (rc) {
-			atomic64_inc(&hpriv->ctx->cs_counters.parsing_drop_cnt);
+			atomic64_inc(&ctx->cs_counters.parsing_drop_cnt);
 			atomic64_inc(&cntr->parsing_drop_cnt);
 			dev_err(hdev->dev,
 				"Failed to parse JOB %d.%llu.%d, err %d, rejecting the CS\n",
@@ -901,8 +917,8 @@ static int cs_ioctl_default(struct hl_fpriv *hpriv, void __user *chunks,
 	}
 
 	if (int_queues_only) {
-		atomic64_inc(&hpriv->ctx->cs_counters.parsing_drop_cnt);
-		atomic64_inc(&cntr->parsing_drop_cnt);
+		atomic64_inc(&ctx->cs_counters.validation_drop_cnt);
+		atomic64_inc(&cntr->validation_drop_cnt);
 		dev_err(hdev->dev,
 			"Reject CS %d.%llu because only internal queues jobs are present\n",
 			cs->ctx->asid, cs->sequence);
@@ -1042,7 +1058,7 @@ out:
 }
 
 static int cs_ioctl_extract_signal_seq(struct hl_device *hdev,
-		struct hl_cs_chunk *chunk, u64 *signal_seq)
+		struct hl_cs_chunk *chunk, u64 *signal_seq, struct hl_ctx *ctx)
 {
 	u64 *signal_seq_arr = NULL;
 	u32 size_to_copy, signal_seq_arr_len;
@@ -1052,6 +1068,8 @@ static int cs_ioctl_extract_signal_seq(struct hl_device *hdev,
 
 	/* currently only one signal seq is supported */
 	if (signal_seq_arr_len != 1) {
+		atomic64_inc(&ctx->cs_counters.validation_drop_cnt);
+		atomic64_inc(&hdev->aggregated_cs_counters.validation_drop_cnt);
 		dev_err(hdev->dev,
 			"Wait for signal CS supports only one signal CS seq\n");
 		return -EINVAL;
@@ -1060,13 +1078,18 @@ static int cs_ioctl_extract_signal_seq(struct hl_device *hdev,
 	signal_seq_arr = kmalloc_array(signal_seq_arr_len,
 					sizeof(*signal_seq_arr),
 					GFP_ATOMIC);
-	if (!signal_seq_arr)
+	if (!signal_seq_arr) {
+		atomic64_inc(&ctx->cs_counters.out_of_mem_drop_cnt);
+		atomic64_inc(&hdev->aggregated_cs_counters.out_of_mem_drop_cnt);
 		return -ENOMEM;
+	}
 
 	size_to_copy = chunk->num_signal_seq_arr * sizeof(*signal_seq_arr);
 	if (copy_from_user(signal_seq_arr,
 				u64_to_user_ptr(chunk->signal_seq_arr),
 				size_to_copy)) {
+		atomic64_inc(&ctx->cs_counters.validation_drop_cnt);
+		atomic64_inc(&hdev->aggregated_cs_counters.validation_drop_cnt);
 		dev_err(hdev->dev,
 			"Failed to copy signal seq array from user\n");
 		rc = -EFAULT;
@@ -1153,6 +1176,7 @@ static int cs_ioctl_signal_wait(struct hl_fpriv *hpriv, enum hl_cs_type cs_type,
 	struct hl_device *hdev = hpriv->hdev;
 	struct hl_cs_compl *sig_waitcs_cmpl;
 	u32 q_idx, collective_engine_id = 0;
+	struct hl_cs_counters_atomic *cntr;
 	struct hl_fence *sig_fence = NULL;
 	struct hl_ctx *ctx = hpriv->ctx;
 	enum hl_queue_type q_type;
@@ -1160,9 +1184,11 @@ static int cs_ioctl_signal_wait(struct hl_fpriv *hpriv, enum hl_cs_type cs_type,
 	u64 signal_seq;
 	int rc;
 
+	cntr = &hdev->aggregated_cs_counters;
 	*cs_seq = ULLONG_MAX;
 
-	rc = hl_cs_copy_chunk_array(hdev, &cs_chunk_array, chunks, num_chunks);
+	rc = hl_cs_copy_chunk_array(hdev, &cs_chunk_array, chunks, num_chunks,
+			ctx);
 	if (rc)
 		goto out;
 
@@ -1170,6 +1196,8 @@ static int cs_ioctl_signal_wait(struct hl_fpriv *hpriv, enum hl_cs_type cs_type,
 	chunk = &cs_chunk_array[0];
 
 	if (chunk->queue_index >= hdev->asic_prop.max_queues) {
+		atomic64_inc(&ctx->cs_counters.validation_drop_cnt);
+		atomic64_inc(&cntr->validation_drop_cnt);
 		dev_err(hdev->dev, "Queue index %d is invalid\n",
 			chunk->queue_index);
 		rc = -EINVAL;
@@ -1181,6 +1209,8 @@ static int cs_ioctl_signal_wait(struct hl_fpriv *hpriv, enum hl_cs_type cs_type,
 	q_type = hw_queue_prop->type;
 
 	if (!hw_queue_prop->supports_sync_stream) {
+		atomic64_inc(&ctx->cs_counters.validation_drop_cnt);
+		atomic64_inc(&cntr->validation_drop_cnt);
 		dev_err(hdev->dev,
 			"Queue index %d does not support sync stream operations\n",
 			q_idx);
@@ -1190,6 +1220,8 @@ static int cs_ioctl_signal_wait(struct hl_fpriv *hpriv, enum hl_cs_type cs_type,
 
 	if (cs_type == CS_TYPE_COLLECTIVE_WAIT) {
 		if (!(hw_queue_prop->collective_mode == HL_COLLECTIVE_MASTER)) {
+			atomic64_inc(&ctx->cs_counters.validation_drop_cnt);
+			atomic64_inc(&cntr->validation_drop_cnt);
 			dev_err(hdev->dev,
 				"Queue index %d is invalid\n", q_idx);
 			rc = -EINVAL;
@@ -1200,12 +1232,14 @@ static int cs_ioctl_signal_wait(struct hl_fpriv *hpriv, enum hl_cs_type cs_type,
 	}
 
 	if (cs_type == CS_TYPE_WAIT || cs_type == CS_TYPE_COLLECTIVE_WAIT) {
-		rc = cs_ioctl_extract_signal_seq(hdev, chunk, &signal_seq);
+		rc = cs_ioctl_extract_signal_seq(hdev, chunk, &signal_seq, ctx);
 		if (rc)
 			goto free_cs_chunk_array;
 
 		sig_fence = hl_ctx_get_fence(ctx, signal_seq);
 		if (IS_ERR(sig_fence)) {
+			atomic64_inc(&ctx->cs_counters.validation_drop_cnt);
+			atomic64_inc(&cntr->validation_drop_cnt);
 			dev_err(hdev->dev,
 				"Failed to get signal CS with seq 0x%llx\n",
 				signal_seq);
@@ -1223,6 +1257,8 @@ static int cs_ioctl_signal_wait(struct hl_fpriv *hpriv, enum hl_cs_type cs_type,
 			container_of(sig_fence, struct hl_cs_compl, base_fence);
 
 		if (sig_waitcs_cmpl->type != CS_TYPE_SIGNAL) {
+			atomic64_inc(&ctx->cs_counters.validation_drop_cnt);
+			atomic64_inc(&cntr->validation_drop_cnt);
 			dev_err(hdev->dev,
 				"CS seq 0x%llx is not of a signal CS\n",
 				signal_seq);
@@ -1270,8 +1306,11 @@ static int cs_ioctl_signal_wait(struct hl_fpriv *hpriv, enum hl_cs_type cs_type,
 	else if (cs_type == CS_TYPE_COLLECTIVE_WAIT)
 		rc = hdev->asic_funcs->collective_wait_create_jobs(hdev, ctx,
 				cs, q_idx, collective_engine_id);
-	else
+	else {
+		atomic64_inc(&ctx->cs_counters.validation_drop_cnt);
+		atomic64_inc(&cntr->validation_drop_cnt);
 		rc = -EINVAL;
+	}
 
 	if (rc)
 		goto free_cs_object;
diff --git a/drivers/misc/habanalabs/common/device.c b/drivers/misc/habanalabs/common/device.c
index 5871162a8442..1456eabf9601 100644
--- a/drivers/misc/habanalabs/common/device.c
+++ b/drivers/misc/habanalabs/common/device.c
@@ -17,12 +17,12 @@ enum hl_device_status hl_device_status(struct hl_device *hdev)
 {
 	enum hl_device_status status;
 
-	if (hdev->disabled)
-		status = HL_DEVICE_STATUS_MALFUNCTION;
-	else if (atomic_read(&hdev->in_reset))
+	if (atomic_read(&hdev->in_reset))
 		status = HL_DEVICE_STATUS_IN_RESET;
 	else if (hdev->needs_reset)
 		status = HL_DEVICE_STATUS_NEEDS_RESET;
+	else if (hdev->disabled)
+		status = HL_DEVICE_STATUS_MALFUNCTION;
 	else
 		status = HL_DEVICE_STATUS_OPERATIONAL;
 
@@ -1092,6 +1092,7 @@ kill_processes:
 						GFP_KERNEL);
 		if (!hdev->kernel_ctx) {
 			rc = -ENOMEM;
+			hl_mmu_fini(hdev);
 			goto out_err;
 		}
 
@@ -1103,6 +1104,7 @@ kill_processes:
 				"failed to init kernel ctx in hard reset\n");
 			kfree(hdev->kernel_ctx);
 			hdev->kernel_ctx = NULL;
+			hl_mmu_fini(hdev);
 			goto out_err;
 		}
 	}
diff --git a/drivers/misc/habanalabs/common/firmware_if.c b/drivers/misc/habanalabs/common/firmware_if.c
index 0e1c629e9800..20f77f58edef 100644
--- a/drivers/misc/habanalabs/common/firmware_if.c
+++ b/drivers/misc/habanalabs/common/firmware_if.c
@@ -627,25 +627,38 @@ int hl_fw_read_preboot_status(struct hl_device *hdev, u32 cpu_boot_status_reg,
 	security_status = RREG32(cpu_security_boot_status_reg);
 
 	/* We read security status multiple times during boot:
-	 * 1. preboot - we check if fw security feature is supported
-	 * 2. boot cpu - we get boot cpu security status
-	 * 3. FW application - we get FW application security status
+	 * 1. preboot - a. Check whether the security status bits are valid
+	 *              b. Check whether fw security is enabled
+	 *              c. Check whether hard reset is done by preboot
+	 * 2. boot cpu - a. Fetch boot cpu security status
+	 *               b. Check whether hard reset is done by boot cpu
+	 * 3. FW application - a. Fetch fw application security status
+	 *                     b. Check whether hard reset is done by fw app
 	 *
 	 * Preboot:
 	 * Check security status bit (CPU_BOOT_DEV_STS0_ENABLED), if it is set
 	 * check security enabled bit (CPU_BOOT_DEV_STS0_SECURITY_EN)
 	 */
 	if (security_status & CPU_BOOT_DEV_STS0_ENABLED) {
-		hdev->asic_prop.fw_security_status_valid = 1;
-		prop->fw_security_disabled =
-			!(security_status & CPU_BOOT_DEV_STS0_SECURITY_EN);
+		prop->fw_security_status_valid = 1;
+
+		if (security_status & CPU_BOOT_DEV_STS0_SECURITY_EN)
+			prop->fw_security_disabled = false;
+		else
+			prop->fw_security_disabled = true;
+
+		if (security_status & CPU_BOOT_DEV_STS0_FW_HARD_RST_EN)
+			prop->hard_reset_done_by_fw = true;
 	} else {
-		hdev->asic_prop.fw_security_status_valid = 0;
+		prop->fw_security_status_valid = 0;
 		prop->fw_security_disabled = true;
 	}
 
+	dev_dbg(hdev->dev, "Firmware preboot hard-reset is %s\n",
+			prop->hard_reset_done_by_fw ? "enabled" : "disabled");
+
 	dev_info(hdev->dev, "firmware-level security is %s\n",
-		prop->fw_security_disabled ? "disabled" : "enabled");
+			prop->fw_security_disabled ? "disabled" : "enabled");
 
 	return 0;
 }
@@ -655,6 +668,7 @@ int hl_fw_init_cpu(struct hl_device *hdev, u32 cpu_boot_status_reg,
 			u32 cpu_security_boot_status_reg, u32 boot_err0_reg,
 			bool skip_bmc, u32 cpu_timeout, u32 boot_fit_timeout)
 {
+	struct asic_fixed_properties *prop = &hdev->asic_prop;
 	u32 status;
 	int rc;
 
@@ -723,11 +737,22 @@ int hl_fw_init_cpu(struct hl_device *hdev, u32 cpu_boot_status_reg,
 	/* Read U-Boot version now in case we will later fail */
 	hdev->asic_funcs->read_device_fw_version(hdev, FW_COMP_UBOOT);
 
+	/* Clear reset status since we need to read it again from boot CPU */
+	prop->hard_reset_done_by_fw = false;
+
 	/* Read boot_cpu security bits */
-	if (hdev->asic_prop.fw_security_status_valid)
-		hdev->asic_prop.fw_boot_cpu_security_map =
+	if (prop->fw_security_status_valid) {
+		prop->fw_boot_cpu_security_map =
 				RREG32(cpu_security_boot_status_reg);
 
+		if (prop->fw_boot_cpu_security_map &
+				CPU_BOOT_DEV_STS0_FW_HARD_RST_EN)
+			prop->hard_reset_done_by_fw = true;
+	}
+
+	dev_dbg(hdev->dev, "Firmware boot CPU hard-reset is %s\n",
+			prop->hard_reset_done_by_fw ? "enabled" : "disabled");
+
 	if (rc) {
 		detect_cpu_boot_status(hdev, status);
 		rc = -EIO;
@@ -796,18 +821,21 @@ int hl_fw_init_cpu(struct hl_device *hdev, u32 cpu_boot_status_reg,
 		goto out;
 	}
 
+	/* Clear reset status since we need to read again from app */
+	prop->hard_reset_done_by_fw = false;
+
 	/* Read FW application security bits */
-	if (hdev->asic_prop.fw_security_status_valid) {
-		hdev->asic_prop.fw_app_security_map =
+	if (prop->fw_security_status_valid) {
+		prop->fw_app_security_map =
 				RREG32(cpu_security_boot_status_reg);
 
-		if (hdev->asic_prop.fw_app_security_map &
+		if (prop->fw_app_security_map &
 				CPU_BOOT_DEV_STS0_FW_HARD_RST_EN)
-			hdev->asic_prop.hard_reset_done_by_fw = true;
+			prop->hard_reset_done_by_fw = true;
 	}
 
-	dev_dbg(hdev->dev, "Firmware hard-reset is %s\n",
-		hdev->asic_prop.hard_reset_done_by_fw ? "enabled" : "disabled");
+	dev_dbg(hdev->dev, "Firmware application CPU hard-reset is %s\n",
+			prop->hard_reset_done_by_fw ? "enabled" : "disabled");
 
 	dev_info(hdev->dev, "Successfully loaded firmware to device\n");
 
diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index 571eda6ef5ab..e0d7f5fbaa5c 100644
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -944,7 +944,7 @@ struct hl_asic_funcs {
 	u32 (*get_signal_cb_size)(struct hl_device *hdev);
 	u32 (*get_wait_cb_size)(struct hl_device *hdev);
 	u32 (*gen_signal_cb)(struct hl_device *hdev, void *data, u16 sob_id,
-			u32 size);
+			u32 size, bool eb);
 	u32 (*gen_wait_cb)(struct hl_device *hdev,
 			struct hl_gen_wait_properties *prop);
 	void (*reset_sob)(struct hl_device *hdev, void *data);
@@ -1000,6 +1000,7 @@ struct hl_va_range {
  * @queue_full_drop_cnt: dropped due to queue full
  * @device_in_reset_drop_cnt: dropped due to device in reset
  * @max_cs_in_flight_drop_cnt: dropped due to maximum CS in-flight
+ * @validation_drop_cnt: dropped due to error in validation
  */
 struct hl_cs_counters_atomic {
 	atomic64_t out_of_mem_drop_cnt;
@@ -1007,6 +1008,7 @@ struct hl_cs_counters_atomic {
 	atomic64_t queue_full_drop_cnt;
 	atomic64_t device_in_reset_drop_cnt;
 	atomic64_t max_cs_in_flight_drop_cnt;
+	atomic64_t validation_drop_cnt;
 };
 
 /**
diff --git a/drivers/misc/habanalabs/common/habanalabs_drv.c b/drivers/misc/habanalabs/common/habanalabs_drv.c
index 6bbb6bca6860..032d114f01ea 100644
--- a/drivers/misc/habanalabs/common/habanalabs_drv.c
+++ b/drivers/misc/habanalabs/common/habanalabs_drv.c
@@ -544,6 +544,7 @@ static struct pci_driver hl_pci_driver = {
 	.id_table = ids,
 	.probe = hl_pci_probe,
 	.remove = hl_pci_remove,
+	.shutdown = hl_pci_remove,
 	.driver.pm = &hl_pm_ops,
 	.err_handler = &hl_pci_err_handler,
 };
diff --git a/drivers/misc/habanalabs/common/habanalabs_ioctl.c b/drivers/misc/habanalabs/common/habanalabs_ioctl.c
index 32e6af1db4e3..12efbd9d2e3a 100644
--- a/drivers/misc/habanalabs/common/habanalabs_ioctl.c
+++ b/drivers/misc/habanalabs/common/habanalabs_ioctl.c
@@ -335,6 +335,8 @@ static int cs_counters_info(struct hl_fpriv *hpriv, struct hl_info_args *args)
 			atomic64_read(&cntr->device_in_reset_drop_cnt);
 	cs_counters.total_max_cs_in_flight_drop_cnt =
 			atomic64_read(&cntr->max_cs_in_flight_drop_cnt);
+	cs_counters.total_validation_drop_cnt =
+			atomic64_read(&cntr->validation_drop_cnt);
 
 	if (hpriv->ctx) {
 		cs_counters.ctx_out_of_mem_drop_cnt =
@@ -352,6 +354,9 @@ static int cs_counters_info(struct hl_fpriv *hpriv, struct hl_info_args *args)
 		cs_counters.ctx_max_cs_in_flight_drop_cnt =
 				atomic64_read(
 			&hpriv->ctx->cs_counters.max_cs_in_flight_drop_cnt);
+		cs_counters.ctx_validation_drop_cnt =
+				atomic64_read(
+				&hpriv->ctx->cs_counters.validation_drop_cnt);
 	}
 
 	return copy_to_user(out, &cs_counters,
@@ -406,7 +411,7 @@ static int total_energy_consumption_info(struct hl_fpriv *hpriv,
 static int pll_frequency_info(struct hl_fpriv *hpriv, struct hl_info_args *args)
 {
 	struct hl_device *hdev = hpriv->hdev;
-	struct hl_pll_frequency_info freq_info = {0};
+	struct hl_pll_frequency_info freq_info = { {0} };
 	u32 max_size = args->return_size;
 	void __user *out = (void __user *) (uintptr_t) args->return_pointer;
 	int rc;
diff --git a/drivers/misc/habanalabs/common/hw_queue.c b/drivers/misc/habanalabs/common/hw_queue.c
index 7caf868d1585..76217258780a 100644
--- a/drivers/misc/habanalabs/common/hw_queue.c
+++ b/drivers/misc/habanalabs/common/hw_queue.c
@@ -418,8 +418,11 @@ static void init_signal_cs(struct hl_device *hdev,
 		"generate signal CB, sob_id: %d, sob val: 0x%x, q_idx: %d\n",
 		cs_cmpl->hw_sob->sob_id, cs_cmpl->sob_val, q_idx);
 
+	/* we set an EB since we must make sure all oeprations are done
+	 * when sending the signal
+	 */
 	hdev->asic_funcs->gen_signal_cb(hdev, job->patched_cb,
-				cs_cmpl->hw_sob->sob_id, 0);
+				cs_cmpl->hw_sob->sob_id, 0, true);
 
 	kref_get(&hw_sob->kref);
 
diff --git a/drivers/misc/habanalabs/common/pci.c b/drivers/misc/habanalabs/common/pci.c
index 923b2606e29f..b4725e6101f6 100644
--- a/drivers/misc/habanalabs/common/pci.c
+++ b/drivers/misc/habanalabs/common/pci.c
@@ -130,10 +130,8 @@ static int hl_pci_elbi_write(struct hl_device *hdev, u64 addr, u32 data)
 	if ((val & PCI_CONFIG_ELBI_STS_MASK) == PCI_CONFIG_ELBI_STS_DONE)
 		return 0;
 
-	if (val & PCI_CONFIG_ELBI_STS_ERR) {
-		dev_err(hdev->dev, "Error writing to ELBI\n");
+	if (val & PCI_CONFIG_ELBI_STS_ERR)
 		return -EIO;
-	}
 
 	if (!(val & PCI_CONFIG_ELBI_STS_MASK)) {
 		dev_err(hdev->dev, "ELBI write didn't finish in time\n");
@@ -160,8 +158,12 @@ int hl_pci_iatu_write(struct hl_device *hdev, u32 addr, u32 data)
 
 	dbi_offset = addr & 0xFFF;
 
-	rc = hl_pci_elbi_write(hdev, prop->pcie_aux_dbi_reg_addr, 0x00300000);
-	rc |= hl_pci_elbi_write(hdev, prop->pcie_dbi_base_address + dbi_offset,
+	/* Ignore result of writing to pcie_aux_dbi_reg_addr as it could fail
+	 * in case the firmware security is enabled
+	 */
+	hl_pci_elbi_write(hdev, prop->pcie_aux_dbi_reg_addr, 0x00300000);
+
+	rc = hl_pci_elbi_write(hdev, prop->pcie_dbi_base_address + dbi_offset,
 				data);
 
 	if (rc)
@@ -244,9 +246,11 @@ int hl_pci_set_inbound_region(struct hl_device *hdev, u8 region,
 
 	rc |= hl_pci_iatu_write(hdev, offset + 0x4, ctrl_reg_val);
 
-	/* Return the DBI window to the default location */
-	rc |= hl_pci_elbi_write(hdev, prop->pcie_aux_dbi_reg_addr, 0);
-	rc |= hl_pci_elbi_write(hdev, prop->pcie_aux_dbi_reg_addr + 4, 0);
+	/* Return the DBI window to the default location
+	 * Ignore result of writing to pcie_aux_dbi_reg_addr as it could fail
+	 * in case the firmware security is enabled
+	 */
+	hl_pci_elbi_write(hdev, prop->pcie_aux_dbi_reg_addr, 0);
 
 	if (rc)
 		dev_err(hdev->dev, "failed to map bar %u to 0x%08llx\n",
@@ -294,9 +298,11 @@ int hl_pci_set_outbound_region(struct hl_device *hdev,
 	/* Enable */
 	rc |= hl_pci_iatu_write(hdev, 0x004, 0x80000000);
 
-	/* Return the DBI window to the default location */
-	rc |= hl_pci_elbi_write(hdev, prop->pcie_aux_dbi_reg_addr, 0);
-	rc |= hl_pci_elbi_write(hdev, prop->pcie_aux_dbi_reg_addr + 4, 0);
+	/* Return the DBI window to the default location
+	 * Ignore result of writing to pcie_aux_dbi_reg_addr as it could fail
+	 * in case the firmware security is enabled
+	 */
+	hl_pci_elbi_write(hdev, prop->pcie_aux_dbi_reg_addr, 0);
 
 	return rc;
 }
diff --git a/drivers/misc/habanalabs/gaudi/gaudi.c b/drivers/misc/habanalabs/gaudi/gaudi.c
index 1f1926607c5e..8c09e4466af8 100644
--- a/drivers/misc/habanalabs/gaudi/gaudi.c
+++ b/drivers/misc/habanalabs/gaudi/gaudi.c
@@ -151,19 +151,6 @@ static const u16 gaudi_packet_sizes[MAX_PACKET_ID] = {
 	[PACKET_LOAD_AND_EXE]	= sizeof(struct packet_load_and_exe)
 };
 
-static const u32 gaudi_pll_base_addresses[GAUDI_PLL_MAX] = {
-	[CPU_PLL] = mmPSOC_CPU_PLL_NR,
-	[PCI_PLL] = mmPSOC_PCI_PLL_NR,
-	[SRAM_PLL] = mmSRAM_W_PLL_NR,
-	[HBM_PLL] = mmPSOC_HBM_PLL_NR,
-	[NIC_PLL] = mmNIC0_PLL_NR,
-	[DMA_PLL] = mmDMA_W_PLL_NR,
-	[MESH_PLL] = mmMESH_W_PLL_NR,
-	[MME_PLL] = mmPSOC_MME_PLL_NR,
-	[TPC_PLL] = mmPSOC_TPC_PLL_NR,
-	[IF_PLL] = mmIF_W_PLL_NR
-};
-
 static inline bool validate_packet_id(enum packet_id id)
 {
 	switch (id) {
@@ -374,7 +361,7 @@ static int gaudi_cpucp_info_get(struct hl_device *hdev);
 static void gaudi_disable_clock_gating(struct hl_device *hdev);
 static void gaudi_mmu_prepare(struct hl_device *hdev, u32 asid);
 static u32 gaudi_gen_signal_cb(struct hl_device *hdev, void *data, u16 sob_id,
-				u32 size);
+				u32 size, bool eb);
 static u32 gaudi_gen_wait_cb(struct hl_device *hdev,
 				struct hl_gen_wait_properties *prop);
 
@@ -667,12 +654,6 @@ static int gaudi_early_init(struct hl_device *hdev)
 	if (rc)
 		goto free_queue_props;
 
-	if (gaudi_get_hw_state(hdev) == HL_DEVICE_HW_STATE_DIRTY) {
-		dev_info(hdev->dev,
-			"H/W state is dirty, must reset before initializing\n");
-		hdev->asic_funcs->hw_fini(hdev, true);
-	}
-
 	/* Before continuing in the initialization, we need to read the preboot
 	 * version to determine whether we run with a security-enabled firmware
 	 */
@@ -685,6 +666,12 @@ static int gaudi_early_init(struct hl_device *hdev)
 		goto pci_fini;
 	}
 
+	if (gaudi_get_hw_state(hdev) == HL_DEVICE_HW_STATE_DIRTY) {
+		dev_info(hdev->dev,
+			"H/W state is dirty, must reset before initializing\n");
+		hdev->asic_funcs->hw_fini(hdev, true);
+	}
+
 	return 0;
 
 pci_fini:
@@ -703,93 +690,60 @@ static int gaudi_early_fini(struct hl_device *hdev)
 }
 
 /**
- * gaudi_fetch_pll_frequency - Fetch PLL frequency values
+ * gaudi_fetch_psoc_frequency - Fetch PSOC frequency values
  *
  * @hdev: pointer to hl_device structure
- * @pll_index: index of the pll to fetch frequency from
- * @pll_freq: pointer to store the pll frequency in MHz in each of the available
- *            outputs. if a certain output is not available a 0 will be set
  *
  */
-static int gaudi_fetch_pll_frequency(struct hl_device *hdev,
-				enum gaudi_pll_index pll_index,
-				u16 *pll_freq_arr)
+static int gaudi_fetch_psoc_frequency(struct hl_device *hdev)
 {
-	u32 nr = 0, nf = 0, od = 0, pll_clk = 0, div_fctr, div_sel,
-			pll_base_addr = gaudi_pll_base_addresses[pll_index];
-	u16 freq = 0;
-	int i, rc;
-
-	if (hdev->asic_prop.fw_security_status_valid &&
-			(hdev->asic_prop.fw_app_security_map &
-					CPU_BOOT_DEV_STS0_PLL_INFO_EN)) {
-		rc = hl_fw_cpucp_pll_info_get(hdev, pll_index, pll_freq_arr);
+	struct asic_fixed_properties *prop = &hdev->asic_prop;
+	u32 nr = 0, nf = 0, od = 0, div_fctr = 0, pll_clk, div_sel;
+	u16 pll_freq_arr[HL_PLL_NUM_OUTPUTS], freq;
+	int rc;
 
-		if (rc)
-			return rc;
-	} else if (hdev->asic_prop.fw_security_disabled) {
+	if (hdev->asic_prop.fw_security_disabled) {
 		/* Backward compatibility */
-		nr = RREG32(pll_base_addr + PLL_NR_OFFSET);
-		nf = RREG32(pll_base_addr + PLL_NF_OFFSET);
-		od = RREG32(pll_base_addr + PLL_OD_OFFSET);
-
-		for (i = 0; i < HL_PLL_NUM_OUTPUTS; i++) {
-			div_fctr = RREG32(pll_base_addr +
-					PLL_DIV_FACTOR_0_OFFSET + i * 4);
-			div_sel = RREG32(pll_base_addr +
-					PLL_DIV_SEL_0_OFFSET + i * 4);
+		div_fctr = RREG32(mmPSOC_CPU_PLL_DIV_FACTOR_2);
+		div_sel = RREG32(mmPSOC_CPU_PLL_DIV_SEL_2);
+		nr = RREG32(mmPSOC_CPU_PLL_NR);
+		nf = RREG32(mmPSOC_CPU_PLL_NF);
+		od = RREG32(mmPSOC_CPU_PLL_OD);
 
-			if (div_sel == DIV_SEL_REF_CLK ||
+		if (div_sel == DIV_SEL_REF_CLK ||
 				div_sel == DIV_SEL_DIVIDED_REF) {
-				if (div_sel == DIV_SEL_REF_CLK)
-					freq = PLL_REF_CLK;
-				else
-					freq = PLL_REF_CLK / (div_fctr + 1);
-			} else if (div_sel == DIV_SEL_PLL_CLK ||
-					div_sel == DIV_SEL_DIVIDED_PLL) {
-				pll_clk = PLL_REF_CLK * (nf + 1) /
-						((nr + 1) * (od + 1));
-				if (div_sel == DIV_SEL_PLL_CLK)
-					freq = pll_clk;
-				else
-					freq = pll_clk / (div_fctr + 1);
-			} else {
-				dev_warn(hdev->dev,
-					"Received invalid div select value: %d",
-					div_sel);
-			}
-
-			pll_freq_arr[i] = freq;
+			if (div_sel == DIV_SEL_REF_CLK)
+				freq = PLL_REF_CLK;
+			else
+				freq = PLL_REF_CLK / (div_fctr + 1);
+		} else if (div_sel == DIV_SEL_PLL_CLK ||
+			div_sel == DIV_SEL_DIVIDED_PLL) {
+			pll_clk = PLL_REF_CLK * (nf + 1) /
+					((nr + 1) * (od + 1));
+			if (div_sel == DIV_SEL_PLL_CLK)
+				freq = pll_clk;
+			else
+				freq = pll_clk / (div_fctr + 1);
+		} else {
+			dev_warn(hdev->dev,
+				"Received invalid div select value: %d",
+				div_sel);
+			freq = 0;
 		}
 	} else {
-		dev_err(hdev->dev, "Failed to fetch PLL frequency values\n");
-		return -EIO;
-	}
+		rc = hl_fw_cpucp_pll_info_get(hdev, CPU_PLL, pll_freq_arr);
 
-	return 0;
-}
-
-/**
- * gaudi_fetch_psoc_frequency - Fetch PSOC frequency values
- *
- * @hdev: pointer to hl_device structure
- *
- */
-static int gaudi_fetch_psoc_frequency(struct hl_device *hdev)
-{
-	struct asic_fixed_properties *prop = &hdev->asic_prop;
-	u16 pll_freq[HL_PLL_NUM_OUTPUTS];
-	int rc;
+		if (rc)
+			return rc;
 
-	rc = gaudi_fetch_pll_frequency(hdev, CPU_PLL, pll_freq);
-	if (rc)
-		return rc;
+		freq = pll_freq_arr[2];
+	}
 
-	prop->psoc_timestamp_frequency = pll_freq[2];
-	prop->psoc_pci_pll_nr = 0;
-	prop->psoc_pci_pll_nf = 0;
-	prop->psoc_pci_pll_od = 0;
-	prop->psoc_pci_pll_div_factor = 0;
+	prop->psoc_timestamp_frequency = freq;
+	prop->psoc_pci_pll_nr = nr;
+	prop->psoc_pci_pll_nf = nf;
+	prop->psoc_pci_pll_od = od;
+	prop->psoc_pci_pll_div_factor = div_fctr;
 
 	return 0;
 }
@@ -884,11 +838,17 @@ static int gaudi_init_tpc_mem(struct hl_device *hdev)
 	size_t fw_size;
 	void *cpu_addr;
 	dma_addr_t dma_handle;
-	int rc;
+	int rc, count = 5;
 
+again:
 	rc = request_firmware(&fw, GAUDI_TPC_FW_FILE, hdev->dev);
+	if (rc == -EINTR && count-- > 0) {
+		msleep(50);
+		goto again;
+	}
+
 	if (rc) {
-		dev_err(hdev->dev, "Firmware file %s is not found!\n",
+		dev_err(hdev->dev, "Failed to load firmware file %s\n",
 				GAUDI_TPC_FW_FILE);
 		goto out;
 	}
@@ -1110,7 +1070,7 @@ static void gaudi_collective_slave_init_job(struct hl_device *hdev,
 		prop->collective_sob_id, queue_id);
 
 	cb_size += gaudi_gen_signal_cb(hdev, job->user_cb,
-			prop->collective_sob_id, cb_size);
+			prop->collective_sob_id, cb_size, false);
 }
 
 static void gaudi_collective_wait_init_cs(struct hl_cs *cs)
@@ -2449,8 +2409,6 @@ static void gaudi_init_golden_registers(struct hl_device *hdev)
 	gaudi_init_e2e(hdev);
 	gaudi_init_hbm_cred(hdev);
 
-	hdev->asic_funcs->disable_clock_gating(hdev);
-
 	for (tpc_id = 0, tpc_offset = 0;
 				tpc_id < TPC_NUMBER_OF_ENGINES;
 				tpc_id++, tpc_offset += TPC_CFG_OFFSET) {
@@ -3462,6 +3420,9 @@ static void gaudi_set_clock_gating(struct hl_device *hdev)
 	if (hdev->in_debug)
 		return;
 
+	if (!hdev->asic_prop.fw_security_disabled)
+		return;
+
 	for (i = GAUDI_PCI_DMA_1, qman_offset = 0 ; i < GAUDI_HBM_DMA_1 ; i++) {
 		enable = !!(hdev->clock_gating_mask &
 				(BIT_ULL(gaudi_dma_assignment[i])));
@@ -3513,7 +3474,7 @@ static void gaudi_disable_clock_gating(struct hl_device *hdev)
 	u32 qman_offset;
 	int i;
 
-	if (!(gaudi->hw_cap_initialized & HW_CAP_CLK_GATE))
+	if (!hdev->asic_prop.fw_security_disabled)
 		return;
 
 	for (i = 0, qman_offset = 0 ; i < DMA_NUMBER_OF_CHANNELS ; i++) {
@@ -3806,7 +3767,7 @@ static int gaudi_init_cpu_queues(struct hl_device *hdev, u32 cpu_timeout)
 static void gaudi_pre_hw_init(struct hl_device *hdev)
 {
 	/* Perform read from the device to make sure device is up */
-	RREG32(mmPCIE_DBI_DEVICE_ID_VENDOR_ID_REG);
+	RREG32(mmHW_STATE);
 
 	if (hdev->asic_prop.fw_security_disabled) {
 		/* Set the access through PCI bars (Linux driver only) as
@@ -3847,6 +3808,13 @@ static int gaudi_hw_init(struct hl_device *hdev)
 		return rc;
 	}
 
+	/* In case the clock gating was enabled in preboot we need to disable
+	 * it here before touching the MME/TPC registers.
+	 * There is no need to take clk gating mutex because when this function
+	 * runs, no other relevant code can run
+	 */
+	hdev->asic_funcs->disable_clock_gating(hdev);
+
 	/* SRAM scrambler must be initialized after CPU is running from HBM */
 	gaudi_init_scrambler_sram(hdev);
 
@@ -3885,7 +3853,7 @@ static int gaudi_hw_init(struct hl_device *hdev)
 	}
 
 	/* Perform read from the device to flush all configuration */
-	RREG32(mmPCIE_DBI_DEVICE_ID_VENDOR_ID_REG);
+	RREG32(mmHW_STATE);
 
 	return 0;
 
@@ -3927,7 +3895,10 @@ static void gaudi_hw_fini(struct hl_device *hdev, bool hard_reset)
 	/* I don't know what is the state of the CPU so make sure it is
 	 * stopped in any means necessary
 	 */
-	WREG32(mmPSOC_GLOBAL_CONF_KMD_MSG_TO_CPU, KMD_MSG_GOTO_WFE);
+	if (hdev->asic_prop.hard_reset_done_by_fw)
+		WREG32(mmPSOC_GLOBAL_CONF_KMD_MSG_TO_CPU, KMD_MSG_RST_DEV);
+	else
+		WREG32(mmPSOC_GLOBAL_CONF_KMD_MSG_TO_CPU, KMD_MSG_GOTO_WFE);
 
 	WREG32(mmGIC_DISTRIBUTOR__5_GICD_SETSPI_NSR, GAUDI_EVENT_HALT_MACHINE);
 
@@ -3971,11 +3942,15 @@ static void gaudi_hw_fini(struct hl_device *hdev, bool hard_reset)
 
 		WREG32(mmPSOC_GLOBAL_CONF_SW_ALL_RST,
 			1 << PSOC_GLOBAL_CONF_SW_ALL_RST_IND_SHIFT);
-	}
 
-	dev_info(hdev->dev,
-		"Issued HARD reset command, going to wait %dms\n",
-		reset_timeout_ms);
+		dev_info(hdev->dev,
+			"Issued HARD reset command, going to wait %dms\n",
+			reset_timeout_ms);
+	} else {
+		dev_info(hdev->dev,
+			"Firmware performs HARD reset, going to wait %dms\n",
+			reset_timeout_ms);
+	}
 
 	/*
 	 * After hard reset, we can't poll the BTM_FSM register because the PSOC
@@ -7936,7 +7911,7 @@ static u32 gaudi_get_wait_cb_size(struct hl_device *hdev)
 }
 
 static u32 gaudi_gen_signal_cb(struct hl_device *hdev, void *data, u16 sob_id,
-				u32 size)
+				u32 size, bool eb)
 {
 	struct hl_cb *cb = (struct hl_cb *) data;
 	struct packet_msg_short *pkt;
@@ -7953,7 +7928,7 @@ static u32 gaudi_gen_signal_cb(struct hl_device *hdev, void *data, u16 sob_id,
 	ctl |= FIELD_PREP(GAUDI_PKT_SHORT_CTL_OP_MASK, 0); /* write the value */
 	ctl |= FIELD_PREP(GAUDI_PKT_SHORT_CTL_BASE_MASK, 3); /* W_S SOB base */
 	ctl |= FIELD_PREP(GAUDI_PKT_SHORT_CTL_OPCODE_MASK, PACKET_MSG_SHORT);
-	ctl |= FIELD_PREP(GAUDI_PKT_SHORT_CTL_EB_MASK, 1);
+	ctl |= FIELD_PREP(GAUDI_PKT_SHORT_CTL_EB_MASK, eb);
 	ctl |= FIELD_PREP(GAUDI_PKT_SHORT_CTL_RB_MASK, 1);
 	ctl |= FIELD_PREP(GAUDI_PKT_SHORT_CTL_MB_MASK, 1);
 
diff --git a/drivers/misc/habanalabs/gaudi/gaudiP.h b/drivers/misc/habanalabs/gaudi/gaudiP.h
index f2d91f4fcffe..a7ab2d7e57d4 100644
--- a/drivers/misc/habanalabs/gaudi/gaudiP.h
+++ b/drivers/misc/habanalabs/gaudi/gaudiP.h
@@ -105,13 +105,6 @@
 #define MME_ACC_OFFSET		(mmMME1_ACC_BASE - mmMME0_ACC_BASE)
 #define SRAM_BANK_OFFSET	(mmSRAM_Y0_X1_RTR_BASE - mmSRAM_Y0_X0_RTR_BASE)
 
-#define PLL_NR_OFFSET		0
-#define PLL_NF_OFFSET		(mmPSOC_CPU_PLL_NF - mmPSOC_CPU_PLL_NR)
-#define PLL_OD_OFFSET		(mmPSOC_CPU_PLL_OD - mmPSOC_CPU_PLL_NR)
-#define PLL_DIV_FACTOR_0_OFFSET	(mmPSOC_CPU_PLL_DIV_FACTOR_0 - \
-				mmPSOC_CPU_PLL_NR)
-#define PLL_DIV_SEL_0_OFFSET	(mmPSOC_CPU_PLL_DIV_SEL_0 - mmPSOC_CPU_PLL_NR)
-
 #define NUM_OF_SOB_IN_BLOCK		\
 	(((mmSYNC_MNGR_E_N_SYNC_MNGR_OBJS_SOB_OBJ_2047 - \
 	mmSYNC_MNGR_E_N_SYNC_MNGR_OBJS_SOB_OBJ_0) + 4) >> 2)
diff --git a/drivers/misc/habanalabs/gaudi/gaudi_coresight.c b/drivers/misc/habanalabs/gaudi/gaudi_coresight.c
index 2e3612e1ee28..88a09d42e111 100644
--- a/drivers/misc/habanalabs/gaudi/gaudi_coresight.c
+++ b/drivers/misc/habanalabs/gaudi/gaudi_coresight.c
@@ -9,6 +9,7 @@
 #include "../include/gaudi/gaudi_coresight.h"
 #include "../include/gaudi/asic_reg/gaudi_regs.h"
 #include "../include/gaudi/gaudi_masks.h"
+#include "../include/gaudi/gaudi_reg_map.h"
 
 #include <uapi/misc/habanalabs.h>
 #define SPMU_SECTION_SIZE		MME0_ACC_SPMU_MAX_OFFSET
@@ -874,7 +875,7 @@ int gaudi_debug_coresight(struct hl_device *hdev, void *data)
 	}
 
 	/* Perform read from the device to flush all configuration */
-	RREG32(mmPCIE_DBI_DEVICE_ID_VENDOR_ID_REG);
+	RREG32(mmHW_STATE);
 
 	return rc;
 }
diff --git a/drivers/misc/habanalabs/goya/goya.c b/drivers/misc/habanalabs/goya/goya.c
index 3e5eb9e3d7bd..b8b4aa636b7c 100644
--- a/drivers/misc/habanalabs/goya/goya.c
+++ b/drivers/misc/habanalabs/goya/goya.c
@@ -613,12 +613,6 @@ static int goya_early_init(struct hl_device *hdev)
 	if (rc)
 		goto free_queue_props;
 
-	if (goya_get_hw_state(hdev) == HL_DEVICE_HW_STATE_DIRTY) {
-		dev_info(hdev->dev,
-			"H/W state is dirty, must reset before initializing\n");
-		hdev->asic_funcs->hw_fini(hdev, true);
-	}
-
 	/* Before continuing in the initialization, we need to read the preboot
 	 * version to determine whether we run with a security-enabled firmware
 	 */
@@ -631,6 +625,12 @@ static int goya_early_init(struct hl_device *hdev)
 		goto pci_fini;
 	}
 
+	if (goya_get_hw_state(hdev) == HL_DEVICE_HW_STATE_DIRTY) {
+		dev_info(hdev->dev,
+			"H/W state is dirty, must reset before initializing\n");
+		hdev->asic_funcs->hw_fini(hdev, true);
+	}
+
 	if (!hdev->pldm) {
 		val = RREG32(mmPSOC_GLOBAL_CONF_BOOT_STRAP_PINS);
 		if (val & PSOC_GLOBAL_CONF_BOOT_STRAP_PINS_SRIOV_EN_MASK)
@@ -694,32 +694,47 @@ static void goya_qman0_set_security(struct hl_device *hdev, bool secure)
 static void goya_fetch_psoc_frequency(struct hl_device *hdev)
 {
 	struct asic_fixed_properties *prop = &hdev->asic_prop;
-	u32 trace_freq = 0;
-	u32 pll_clk = 0;
-	u32 div_fctr = RREG32(mmPSOC_PCI_PLL_DIV_FACTOR_1);
-	u32 div_sel = RREG32(mmPSOC_PCI_PLL_DIV_SEL_1);
-	u32 nr = RREG32(mmPSOC_PCI_PLL_NR);
-	u32 nf = RREG32(mmPSOC_PCI_PLL_NF);
-	u32 od = RREG32(mmPSOC_PCI_PLL_OD);
-
-	if (div_sel == DIV_SEL_REF_CLK || div_sel == DIV_SEL_DIVIDED_REF) {
-		if (div_sel == DIV_SEL_REF_CLK)
-			trace_freq = PLL_REF_CLK;
-		else
-			trace_freq = PLL_REF_CLK / (div_fctr + 1);
-	} else if (div_sel == DIV_SEL_PLL_CLK ||
-					div_sel == DIV_SEL_DIVIDED_PLL) {
-		pll_clk = PLL_REF_CLK * (nf + 1) / ((nr + 1) * (od + 1));
-		if (div_sel == DIV_SEL_PLL_CLK)
-			trace_freq = pll_clk;
-		else
-			trace_freq = pll_clk / (div_fctr + 1);
+	u32 nr = 0, nf = 0, od = 0, div_fctr = 0, pll_clk, div_sel;
+	u16 pll_freq_arr[HL_PLL_NUM_OUTPUTS], freq;
+	int rc;
+
+	if (hdev->asic_prop.fw_security_disabled) {
+		div_fctr = RREG32(mmPSOC_PCI_PLL_DIV_FACTOR_1);
+		div_sel = RREG32(mmPSOC_PCI_PLL_DIV_SEL_1);
+		nr = RREG32(mmPSOC_PCI_PLL_NR);
+		nf = RREG32(mmPSOC_PCI_PLL_NF);
+		od = RREG32(mmPSOC_PCI_PLL_OD);
+
+		if (div_sel == DIV_SEL_REF_CLK ||
+				div_sel == DIV_SEL_DIVIDED_REF) {
+			if (div_sel == DIV_SEL_REF_CLK)
+				freq = PLL_REF_CLK;
+			else
+				freq = PLL_REF_CLK / (div_fctr + 1);
+		} else if (div_sel == DIV_SEL_PLL_CLK ||
+				div_sel == DIV_SEL_DIVIDED_PLL) {
+			pll_clk = PLL_REF_CLK * (nf + 1) /
+					((nr + 1) * (od + 1));
+			if (div_sel == DIV_SEL_PLL_CLK)
+				freq = pll_clk;
+			else
+				freq = pll_clk / (div_fctr + 1);
+		} else {
+			dev_warn(hdev->dev,
+				"Received invalid div select value: %d",
+				div_sel);
+			freq = 0;
+		}
 	} else {
-		dev_warn(hdev->dev,
-			"Received invalid div select value: %d", div_sel);
+		rc = hl_fw_cpucp_pll_info_get(hdev, PCI_PLL, pll_freq_arr);
+
+		if (rc)
+			return;
+
+		freq = pll_freq_arr[1];
 	}
 
-	prop->psoc_timestamp_frequency = trace_freq;
+	prop->psoc_timestamp_frequency = freq;
 	prop->psoc_pci_pll_nr = nr;
 	prop->psoc_pci_pll_nf = nf;
 	prop->psoc_pci_pll_od = od;
@@ -5324,7 +5339,7 @@ static u32 goya_get_wait_cb_size(struct hl_device *hdev)
 }
 
 static u32 goya_gen_signal_cb(struct hl_device *hdev, void *data, u16 sob_id,
-		u32 size)
+				u32 size, bool eb)
 {
 	return 0;
 }
diff --git a/drivers/misc/habanalabs/include/common/hl_boot_if.h b/drivers/misc/habanalabs/include/common/hl_boot_if.h
index e5801ecf0cb2..b637dfd69f6e 100644
--- a/drivers/misc/habanalabs/include/common/hl_boot_if.h
+++ b/drivers/misc/habanalabs/include/common/hl_boot_if.h
@@ -145,11 +145,15 @@
  *					implemented. This means that FW will
  *					perform hard reset procedure on
  *					receiving the halt-machine event.
- *					Initialized in: linux
+ *					Initialized in: preboot, u-boot, linux
  *
  * CPU_BOOT_DEV_STS0_PLL_INFO_EN	FW retrieval of PLL info is enabled.
  *					Initialized in: linux
  *
+ * CPU_BOOT_DEV_STS0_CLK_GATE_EN	Clock Gating enabled.
+ *					FW initialized Clock Gating.
+ *					Initialized in: preboot
+ *
  * CPU_BOOT_DEV_STS0_ENABLED		Device status register enabled.
  *					This is a main indication that the
  *					running FW populates the device status
@@ -171,6 +175,7 @@
 #define CPU_BOOT_DEV_STS0_DRAM_SCR_EN			(1 << 9)
 #define CPU_BOOT_DEV_STS0_FW_HARD_RST_EN		(1 << 10)
 #define CPU_BOOT_DEV_STS0_PLL_INFO_EN			(1 << 11)
+#define CPU_BOOT_DEV_STS0_CLK_GATE_EN			(1 << 13)
 #define CPU_BOOT_DEV_STS0_ENABLED			(1 << 31)
 
 enum cpu_boot_status {
@@ -204,6 +209,8 @@ enum kmd_msg {
 	KMD_MSG_GOTO_WFE,
 	KMD_MSG_FIT_RDY,
 	KMD_MSG_SKIP_BMC,
+	RESERVED,
+	KMD_MSG_RST_DEV,
 };
 
 enum cpu_msg_status {
diff --git a/drivers/misc/pvpanic.c b/drivers/misc/pvpanic.c
index 951b37da5e3c..41cab297d66e 100644
--- a/drivers/misc/pvpanic.c
+++ b/drivers/misc/pvpanic.c
@@ -55,12 +55,23 @@ static int pvpanic_mmio_probe(struct platform_device *pdev)
 	struct resource *res;
 
 	res = platform_get_mem_or_io(pdev, 0);
-	if (res && resource_type(res) == IORESOURCE_IO)
+	if (!res)
+		return -EINVAL;
+
+	switch (resource_type(res)) {
+	case IORESOURCE_IO:
 		base = devm_ioport_map(dev, res->start, resource_size(res));
-	else
+		if (!base)
+			return -ENOMEM;
+		break;
+	case IORESOURCE_MEM:
 		base = devm_ioremap_resource(dev, res);
-	if (IS_ERR(base))
-		return PTR_ERR(base);
+		if (IS_ERR(base))
+			return PTR_ERR(base);
+		break;
+	default:
+		return -EINVAL;
+	}
 
 	atomic_notifier_chain_register(&panic_notifier_list,
 				       &pvpanic_panic_nb);