[PATCH 01/26] habanalabs/gaudi2: fix BMON 3rd address range

[PATCH 01/26] habanalabs/gaudi2: fix BMON 3rd address range

[Oded Gabbay]

From: tal albo <talbo@habana.ai>

Fix programming incorrect value of address range

Signed-off-by: tal albo <talbo@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/gaudi2/gaudi2_coresight.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/misc/habanalabs/gaudi2/gaudi2_coresight.c b/drivers/misc/habanalabs/gaudi2/gaudi2_coresight.c
index 56c6ab692482..1df7a59e4101 100644
--- a/drivers/misc/habanalabs/gaudi2/gaudi2_coresight.c
+++ b/drivers/misc/habanalabs/gaudi2/gaudi2_coresight.c
@@ -2376,10 +2376,10 @@ static int gaudi2_config_bmon(struct hl_device *hdev, struct hl_debug_params *pa
 		WREG32(base_reg + mmBMON_ADDRH_S2_OFFSET, upper_32_bits(input->start_addr2));
 		WREG32(base_reg + mmBMON_ADDRL_E2_OFFSET, lower_32_bits(input->end_addr2));
 		WREG32(base_reg + mmBMON_ADDRH_E2_OFFSET, upper_32_bits(input->end_addr2));
-		WREG32(base_reg + mmBMON_ADDRL_S3_OFFSET, lower_32_bits(input->start_addr2));
-		WREG32(base_reg + mmBMON_ADDRH_S3_OFFSET, upper_32_bits(input->start_addr2));
-		WREG32(base_reg + mmBMON_ADDRL_E3_OFFSET, lower_32_bits(input->end_addr2));
-		WREG32(base_reg + mmBMON_ADDRH_E3_OFFSET, upper_32_bits(input->end_addr2));
+		WREG32(base_reg + mmBMON_ADDRL_S3_OFFSET, lower_32_bits(input->start_addr3));
+		WREG32(base_reg + mmBMON_ADDRH_S3_OFFSET, upper_32_bits(input->start_addr3));
+		WREG32(base_reg + mmBMON_ADDRL_E3_OFFSET, lower_32_bits(input->end_addr3));
+		WREG32(base_reg + mmBMON_ADDRH_E3_OFFSET, upper_32_bits(input->end_addr3));
 
 		WREG32(base_reg + mmBMON_IDL_OFFSET, 0x0);
 		WREG32(base_reg + mmBMON_IDH_OFFSET, 0x0);
-- 
2.25.1

[PATCH 02/26] habanalabs: read binning info from preboot

[Oded Gabbay]

From: farah kassabri <fkassabri@habana.ai>

Sometimes we need the binning info at a very early state of the
driver initialization. Therefore, support was added in preboot to
provide the binning info as part of the f/w descriptor and the driver
can now use that.

Signed-off-by: farah kassabri <fkassabri@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/firmware_if.c  | 24 +++++++++++++-
 drivers/misc/habanalabs/common/habanalabs.h   |  5 +++
 .../habanalabs/include/common/hl_boot_if.h    | 31 +++++++++++++------
 3 files changed, 50 insertions(+), 10 deletions(-)

diff --git a/drivers/misc/habanalabs/common/firmware_if.c b/drivers/misc/habanalabs/common/firmware_if.c
index 228b92278e48..4f364c3085fe 100644
--- a/drivers/misc/habanalabs/common/firmware_if.c
+++ b/drivers/misc/habanalabs/common/firmware_if.c
@@ -2560,13 +2560,35 @@ static int hl_fw_dynamic_init_cpu(struct hl_device *hdev,
 	}
 
 	if (!(hdev->fw_components & FW_TYPE_BOOT_CPU)) {
+		struct lkd_fw_binning_info *binning_info;
+
 		rc = hl_fw_dynamic_request_descriptor(hdev, fw_loader, 0);
 		if (rc)
 			goto protocol_err;
 
 		/* read preboot version */
-		return hl_fw_dynamic_read_device_fw_version(hdev, FW_COMP_PREBOOT,
+		rc = hl_fw_dynamic_read_device_fw_version(hdev, FW_COMP_PREBOOT,
 				fw_loader->dynamic_loader.comm_desc.cur_fw_ver);
+
+		if (rc)
+			goto out;
+
+		/* read binning info from preboot */
+		if (hdev->support_preboot_binning) {
+			binning_info = &fw_loader->dynamic_loader.comm_desc.binning_info;
+			hdev->tpc_binning = le64_to_cpu(binning_info->tpc_mask_l);
+			hdev->dram_binning = le32_to_cpu(binning_info->dram_mask);
+			hdev->edma_binning = le32_to_cpu(binning_info->edma_mask);
+			hdev->decoder_binning = le32_to_cpu(binning_info->dec_mask);
+			hdev->rotator_binning = le32_to_cpu(binning_info->rot_mask);
+
+			dev_dbg(hdev->dev,
+				"Read binning masks: tpc: 0x%llx, dram: 0x%llx, edma: 0x%x, dec: 0x%x, rot:0x%x\n",
+				hdev->tpc_binning, hdev->dram_binning, hdev->edma_binning,
+				hdev->decoder_binning, hdev->rotator_binning);
+		}
+out:
+		return rc;
 	}
 
 	/* load boot fit to FW */
diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index e2527d976ee0..9e42d0e9ce33 100644
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -3157,6 +3157,8 @@ struct hl_reset_info {
  * @edma_binning: contains mask of edma engines that is received from the f/w which
  *                   indicates which edma engines are binned-out
  * @device_release_watchdog_timeout_sec: device release watchdog timeout value in seconds.
+ * @rotator_binning: contains mask of rotators engines that is received from the f/w
+ *			which indicates which rotator engines are binned-out(Gaudi3 and above).
  * @id: device minor.
  * @id_control: minor of the control device.
  * @cdev_idx: char device index. Used for setting its name.
@@ -3214,6 +3216,7 @@ struct hl_reset_info {
  * @heartbeat: Controls if we want to enable the heartbeat mechanism vs. the f/w, which verifies
  *             that the f/w is always alive. Used only for testing.
  * @supports_ctx_switch: true if a ctx switch is required upon first submission.
+ * @support_preboot_binning: true if we support read binning info from preboot.
  */
 struct hl_device {
 	struct pci_dev			*pdev;
@@ -3322,6 +3325,7 @@ struct hl_device {
 	u32				decoder_binning;
 	u32				edma_binning;
 	u32				device_release_watchdog_timeout_sec;
+	u32				rotator_binning;
 	u16				id;
 	u16				id_control;
 	u16				cdev_idx;
@@ -3355,6 +3359,7 @@ struct hl_device {
 	u8				supports_mmu_prefetch;
 	u8				reset_upon_device_release;
 	u8				supports_ctx_switch;
+	u8				support_preboot_binning;
 
 	/* Parameters for bring-up */
 	u64				nic_ports_mask;
diff --git a/drivers/misc/habanalabs/include/common/hl_boot_if.h b/drivers/misc/habanalabs/include/common/hl_boot_if.h
index e0ea51cc7475..fe034111360e 100644
--- a/drivers/misc/habanalabs/include/common/hl_boot_if.h
+++ b/drivers/misc/habanalabs/include/common/hl_boot_if.h
@@ -439,7 +439,7 @@ struct cpu_dyn_regs {
 /* TODO: remove the desc magic after the code is updated to use message */
 /* HCDM - Habana Communications Descriptor Magic */
 #define HL_COMMS_DESC_MAGIC	0x4843444D
-#define HL_COMMS_DESC_VER	1
+#define HL_COMMS_DESC_VER	3
 
 /* HCMv - Habana Communications Message + header version */
 #define HL_COMMS_MSG_MAGIC_VALUE	0x48434D00
@@ -450,8 +450,10 @@ struct cpu_dyn_regs {
 					((ver) & HL_COMMS_MSG_MAGIC_VER_MASK))
 #define HL_COMMS_MSG_MAGIC_V0		HL_COMMS_DESC_MAGIC
 #define HL_COMMS_MSG_MAGIC_V1		HL_COMMS_MSG_MAGIC_VER(1)
+#define HL_COMMS_MSG_MAGIC_V2		HL_COMMS_MSG_MAGIC_VER(2)
+#define HL_COMMS_MSG_MAGIC_V3		HL_COMMS_MSG_MAGIC_VER(3)
 
-#define HL_COMMS_MSG_MAGIC		HL_COMMS_MSG_MAGIC_V1
+#define HL_COMMS_MSG_MAGIC		HL_COMMS_MSG_MAGIC_V3
 
 #define HL_COMMS_MSG_MAGIC_VALIDATE_MAGIC(magic)			\
 		(((magic) & HL_COMMS_MSG_MAGIC_MASK) ==			\
@@ -474,22 +476,31 @@ enum comms_msg_type {
 
 /*
  * Binning information shared between LKD and FW
- * @tpc_mask - TPC binning information
+ * @tpc_mask_l - TPC binning information lower 64 bit
  * @dec_mask - Decoder binning information
- * @hbm_mask - HBM binning information
+ * @dram_mask - DRAM binning information
  * @edma_mask - EDMA binning information
  * @mme_mask_l - MME binning information lower 32
  * @mme_mask_h - MME binning information upper 32
- * @reserved - reserved field for 64 bit alignment
+ * @rot_mask - Rotator binning information
+ * @xbar_mask - xBAR binning information
+ * @reserved - reserved field for future binning info w/o ABI change
+ * @tpc_mask_h - TPC binning information upper 64 bit
+ * @nic_mask - NIC binning information
  */
 struct lkd_fw_binning_info {
-	__le64 tpc_mask;
+	__le64 tpc_mask_l;
 	__le32 dec_mask;
-	__le32 hbm_mask;
+	__le32 dram_mask;
 	__le32 edma_mask;
 	__le32 mme_mask_l;
 	__le32 mme_mask_h;
-	__le32 reserved;
+	__le32 rot_mask;
+	__le32 xbar_mask;
+	__le32 reserved0;
+	__le64 tpc_mask_h;
+	__le64 nic_mask;
+	__le32 reserved1[8];
 };
 
 /* TODO: remove this struct after the code is updated to use message */
@@ -521,6 +532,7 @@ struct lkd_fw_comms_desc {
 	/* can be used for 1 more version w/o ABI change */
 	char reserved0[VERSION_MAX_LEN];
 	__le64 img_addr;	/* address for next FW component load */
+	struct lkd_fw_binning_info binning_info;
 };
 
 enum comms_reset_cause {
@@ -545,6 +557,7 @@ struct lkd_fw_comms_msg {
 			char reserved0[VERSION_MAX_LEN];
 			/* address for next FW component load */
 			__le64 img_addr;
+			struct lkd_fw_binning_info binning_info;
 		};
 		struct {
 			__u8 reset_cause;
@@ -552,7 +565,7 @@ struct lkd_fw_comms_msg {
 		struct {
 			__u8 fw_cfg_skip; /* 1 - skip, 0 - don't skip */
 		};
-		struct lkd_fw_binning_info binning_info;
+		struct lkd_fw_binning_info binning_conf;
 	};
 };
 
-- 
2.25.1

[PATCH 03/26] habanalabs: remove releasing of user threads from device release

[Oded Gabbay]

From: Tomer Tayar <ttayar@habana.ai>

The device file is not in use when hl_device_release() is called,
and there aren't any user threads that use IOCTLs to wait for
interrupts. Therefore there is no need to release them at this point.

Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/device.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/misc/habanalabs/common/device.c b/drivers/misc/habanalabs/common/device.c
index 87ab329e65d4..1453f2ec72d9 100644
--- a/drivers/misc/habanalabs/common/device.c
+++ b/drivers/misc/habanalabs/common/device.c
@@ -511,11 +511,6 @@ static int hl_device_release(struct inode *inode, struct file *filp)
 		return 0;
 	}
 
-	/* Each pending user interrupt holds the user's context, hence we
-	 * must release them all before calling hl_ctx_mgr_fini().
-	 */
-	hl_release_pending_user_interrupts(hpriv->hdev);
-
 	hl_ctx_mgr_fini(hdev, &hpriv->ctx_mgr);
 	hl_mem_mgr_fini(&hpriv->mem_mgr);
 
-- 
2.25.1

[PATCH 04/26] habanalabs: abort waiting user threads upon error

[Oded Gabbay]

From: Tomer Tayar <ttayar@habana.ai>

User should close the FD when being notified about an error, after
which a device reset takes place.

However, if the user has pending threads that wait for completions,
the device release won't be called and eventually the watchdog timeout
will expire, leading to hard reset and killing the user process.

To avoid it, abort such waiting threads right after the error
notification, and block following waiting operations.

Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 .../habanalabs/common/command_submission.c    | 28 +++++++++++++++++--
 drivers/misc/habanalabs/common/device.c       |  2 ++
 drivers/misc/habanalabs/common/habanalabs.h   |  1 +
 3 files changed, 28 insertions(+), 3 deletions(-)

diff --git a/drivers/misc/habanalabs/common/command_submission.c b/drivers/misc/habanalabs/common/command_submission.c
index ea0e5101c10e..cf3b82efc65c 100644
--- a/drivers/misc/habanalabs/common/command_submission.c
+++ b/drivers/misc/habanalabs/common/command_submission.c
@@ -1117,6 +1117,27 @@ void hl_release_pending_user_interrupts(struct hl_device *hdev)
 	wake_pending_user_interrupt_threads(interrupt);
 }
 
+static void force_complete_cs(struct hl_device *hdev)
+{
+	struct hl_cs *cs;
+
+	spin_lock(&hdev->cs_mirror_lock);
+
+	list_for_each_entry(cs, &hdev->cs_mirror_list, mirror_node) {
+		cs->fence->error = -EIO;
+		complete_all(&cs->fence->completion);
+	}
+
+	spin_unlock(&hdev->cs_mirror_lock);
+}
+
+void hl_abort_waitings_for_completion(struct hl_device *hdev)
+{
+	force_complete_cs(hdev);
+	force_complete_multi_cs(hdev);
+	hl_release_pending_user_interrupts(hdev);
+}
+
 static void job_wq_completion(struct work_struct *work)
 {
 	struct hl_cs_job *job = container_of(work, struct hl_cs_job,
@@ -3489,14 +3510,15 @@ static int hl_interrupt_wait_ioctl(struct hl_fpriv *hpriv, void *data)
 
 int hl_wait_ioctl(struct hl_fpriv *hpriv, void *data)
 {
+	struct hl_device *hdev = hpriv->hdev;
 	union hl_wait_cs_args *args = data;
 	u32 flags = args->in.flags;
 	int rc;
 
-	/* If the device is not operational, no point in waiting for any command submission or
-	 * user interrupt
+	/* If the device is not operational, or if an error has happened and user should release the
+	 * device, there is no point in waiting for any command submission or user interrupt.
 	 */
-	if (!hl_device_operational(hpriv->hdev, NULL))
+	if (!hl_device_operational(hpriv->hdev, NULL) || hdev->reset_info.watchdog_active)
 		return -EBUSY;
 
 	if (flags & HL_WAIT_CS_FLAGS_INTERRUPT)
diff --git a/drivers/misc/habanalabs/common/device.c b/drivers/misc/habanalabs/common/device.c
index 1453f2ec72d9..92721111b652 100644
--- a/drivers/misc/habanalabs/common/device.c
+++ b/drivers/misc/habanalabs/common/device.c
@@ -1865,6 +1865,8 @@ int hl_device_cond_reset(struct hl_device *hdev, u32 flags, u64 event_mask)
 
 	hl_ctx_put(ctx);
 
+	hl_abort_waitings_for_completion(hdev);
+
 	return 0;
 
 device_reset:
diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index 9e42d0e9ce33..7fb45610ad0c 100644
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -3791,6 +3791,7 @@ void hl_dec_fini(struct hl_device *hdev);
 void hl_dec_ctx_fini(struct hl_ctx *ctx);
 
 void hl_release_pending_user_interrupts(struct hl_device *hdev);
+void hl_abort_waitings_for_completion(struct hl_device *hdev);
 int hl_cs_signal_sob_wraparound_handler(struct hl_device *hdev, u32 q_idx,
 			struct hl_hw_sob **hw_sob, u32 count, bool encaps_sig);
 
-- 
2.25.1

[PATCH 05/26] habanalabs: don't notify user about clk throttling due to power

[Oded Gabbay]

From: Ofir Bitton <obitton@habana.ai>

As clock throttling due to high power consumption can happen very
frequently and there is no real reason to notify the user about it,
we skip this notification in all asics.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/gaudi/gaudi.c   | 7 ++++---
 drivers/misc/habanalabs/gaudi2/gaudi2.c | 7 ++++---
 2 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/misc/habanalabs/gaudi/gaudi.c b/drivers/misc/habanalabs/gaudi/gaudi.c
index 9f5e208701ba..ae78f838f987 100644
--- a/drivers/misc/habanalabs/gaudi/gaudi.c
+++ b/drivers/misc/habanalabs/gaudi/gaudi.c
@@ -7584,7 +7584,7 @@ static int tpc_krn_event_to_tpc_id(u16 tpc_dec_event_type)
 	return (tpc_dec_event_type - GAUDI_EVENT_TPC0_KRN_ERR) / 6;
 }
 
-static void gaudi_print_clk_change_info(struct hl_device *hdev, u16 event_type)
+static void gaudi_print_clk_change_info(struct hl_device *hdev, u16 event_type, u64 *event_mask)
 {
 	ktime_t zero_time = ktime_set(0, 0);
 
@@ -7612,6 +7612,7 @@ static void gaudi_print_clk_change_info(struct hl_device *hdev, u16 event_type)
 		hdev->clk_throttling.aggregated_reason |= HL_CLK_THROTTLE_THERMAL;
 		hdev->clk_throttling.timestamp[HL_CLK_THROTTLE_TYPE_THERMAL].start = ktime_get();
 		hdev->clk_throttling.timestamp[HL_CLK_THROTTLE_TYPE_THERMAL].end = zero_time;
+		*event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		dev_info_ratelimited(hdev->dev,
 			"Clock throttling due to overheating\n");
 		break;
@@ -7619,6 +7620,7 @@ static void gaudi_print_clk_change_info(struct hl_device *hdev, u16 event_type)
 	case GAUDI_EVENT_FIX_THERMAL_ENV_E:
 		hdev->clk_throttling.current_reason &= ~HL_CLK_THROTTLE_THERMAL;
 		hdev->clk_throttling.timestamp[HL_CLK_THROTTLE_TYPE_THERMAL].end = ktime_get();
+		*event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		dev_info_ratelimited(hdev->dev,
 			"Thermal envelop is safe, back to optimal clock\n");
 		break;
@@ -7887,8 +7889,7 @@ static void gaudi_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_entr
 		break;
 
 	case GAUDI_EVENT_FIX_POWER_ENV_S ... GAUDI_EVENT_FIX_THERMAL_ENV_E:
-		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
-		gaudi_print_clk_change_info(hdev, event_type);
+		gaudi_print_clk_change_info(hdev, event_type, &event_mask);
 		hl_fw_unmask_irq(hdev, event_type);
 		break;
 
diff --git a/drivers/misc/habanalabs/gaudi2/gaudi2.c b/drivers/misc/habanalabs/gaudi2/gaudi2.c
index e793fb2bdcbe..c14b3bb16f96 100644
--- a/drivers/misc/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/misc/habanalabs/gaudi2/gaudi2.c
@@ -8603,7 +8603,7 @@ static void gaudi2_handle_hbm_mc_spi(struct hl_device *hdev, u64 intr_cause_data
 				hbm_mc_spi[i].cause);
 }
 
-static void gaudi2_print_clk_change_info(struct hl_device *hdev, u16 event_type)
+static void gaudi2_print_clk_change_info(struct hl_device *hdev, u16 event_type, u64 *event_mask)
 {
 	ktime_t zero_time = ktime_set(0, 0);
 
@@ -8629,12 +8629,14 @@ static void gaudi2_print_clk_change_info(struct hl_device *hdev, u16 event_type)
 		hdev->clk_throttling.aggregated_reason |= HL_CLK_THROTTLE_THERMAL;
 		hdev->clk_throttling.timestamp[HL_CLK_THROTTLE_TYPE_THERMAL].start = ktime_get();
 		hdev->clk_throttling.timestamp[HL_CLK_THROTTLE_TYPE_THERMAL].end = zero_time;
+		*event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		dev_info_ratelimited(hdev->dev, "Clock throttling due to overheating\n");
 		break;
 
 	case GAUDI2_EVENT_CPU_FIX_THERMAL_ENV_E:
 		hdev->clk_throttling.current_reason &= ~HL_CLK_THROTTLE_THERMAL;
 		hdev->clk_throttling.timestamp[HL_CLK_THROTTLE_TYPE_THERMAL].end = ktime_get();
+		*event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		dev_info_ratelimited(hdev->dev, "Thermal envelop is safe, back to optimal clock\n");
 		break;
 
@@ -9085,8 +9087,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 	case GAUDI2_EVENT_CPU_FIX_POWER_ENV_E:
 	case GAUDI2_EVENT_CPU_FIX_THERMAL_ENV_S:
 	case GAUDI2_EVENT_CPU_FIX_THERMAL_ENV_E:
-		gaudi2_print_clk_change_info(hdev, event_type);
-		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
+		gaudi2_print_clk_change_info(hdev, event_type, &event_mask);
 		break;
 
 	case GAUDI2_EVENT_CPU_PKT_QUEUE_OUT_SYNC:
-- 
2.25.1

[PATCH 06/26] habanalabs: don't allow user to destroy CB handle more than once

[Oded Gabbay]

From: Tomer Tayar <ttayar@habana.ai>

The refcount of a CB buffer is initialized when user allocates a CB,
and is decreased when he destroys the CB handle.

If this refcount is increased also from kernel and user sends more than
one destroy requests for the handle, the buffer will be released/freed
and later be accessed when the refcount is put from kernel side.

To avoid it, prevent user from destroying the handle more than once.

Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 .../misc/habanalabs/common/command_buffer.c   | 22 +++++++++++++++++++
 drivers/misc/habanalabs/common/device.c       |  2 +-
 drivers/misc/habanalabs/common/habanalabs.h   |  6 ++++-
 .../misc/habanalabs/common/habanalabs_drv.c   |  2 +-
 drivers/misc/habanalabs/common/memory_mgr.c   |  4 +++-
 5 files changed, 32 insertions(+), 4 deletions(-)

diff --git a/drivers/misc/habanalabs/common/command_buffer.c b/drivers/misc/habanalabs/common/command_buffer.c
index 2b332991ac6a..24100501f8ca 100644
--- a/drivers/misc/habanalabs/common/command_buffer.c
+++ b/drivers/misc/habanalabs/common/command_buffer.c
@@ -298,9 +298,31 @@ int hl_cb_create(struct hl_device *hdev, struct hl_mem_mgr *mmg,
 
 int hl_cb_destroy(struct hl_mem_mgr *mmg, u64 cb_handle)
 {
+	struct hl_cb *cb;
 	int rc;
 
+	/* Make sure that a CB handle isn't destroyed by user more than once */
+	if (!mmg->is_kernel_mem_mgr) {
+		cb = hl_cb_get(mmg, cb_handle);
+		if (!cb) {
+			dev_dbg(mmg->dev, "CB destroy failed, no CB was found for handle %#llx\n",
+				cb_handle);
+			rc = -EINVAL;
+			goto out;
+		}
+
+		rc = atomic_cmpxchg(&cb->is_handle_destroyed, 0, 1);
+		hl_cb_put(cb);
+		if (rc) {
+			dev_dbg(mmg->dev, "CB destroy failed, handle %#llx was already destroyed\n",
+				cb_handle);
+			rc = -EINVAL;
+			goto out;
+		}
+	}
+
 	rc = hl_mmap_mem_buf_put_handle(mmg, cb_handle);
+out:
 	if (rc < 0)
 		return rc; /* Invalid handle */
 
diff --git a/drivers/misc/habanalabs/common/device.c b/drivers/misc/habanalabs/common/device.c
index 92721111b652..afd9d4d46574 100644
--- a/drivers/misc/habanalabs/common/device.c
+++ b/drivers/misc/habanalabs/common/device.c
@@ -853,7 +853,7 @@ static int device_early_init(struct hl_device *hdev)
 	if (rc)
 		goto free_chip_info;
 
-	hl_mem_mgr_init(hdev->dev, &hdev->kernel_mem_mgr);
+	hl_mem_mgr_init(hdev->dev, &hdev->kernel_mem_mgr, 1);
 
 	hdev->reset_wq = create_singlethread_workqueue("hl_device_reset");
 	if (!hdev->reset_wq) {
diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index 7fb45610ad0c..ecf7e5da8f1d 100644
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -872,11 +872,13 @@ struct hl_mmap_mem_buf;
  * @dev: back pointer to the owning device
  * @lock: protects handles
  * @handles: an idr holding all active handles to the memory buffers in the system.
+ * @is_kernel_mem_mgr: indicate whether the memory manager is the per-device kernel memory manager
  */
 struct hl_mem_mgr {
 	struct device *dev;
 	spinlock_t lock;
 	struct idr handles;
+	u8 is_kernel_mem_mgr;
 };
 
 /**
@@ -935,6 +937,7 @@ struct hl_mmap_mem_buf {
  * @size: holds the CB's size.
  * @roundup_size: holds the cb size after roundup to page size.
  * @cs_cnt: holds number of CS that this CB participates in.
+ * @is_handle_destroyed: atomic boolean indicating whether or not the CB handle was destroyed.
  * @is_pool: true if CB was acquired from the pool, false otherwise.
  * @is_internal: internally allocated
  * @is_mmu_mapped: true if the CB is mapped to the device's MMU.
@@ -951,6 +954,7 @@ struct hl_cb {
 	u32			size;
 	u32			roundup_size;
 	atomic_t		cs_cnt;
+	atomic_t		is_handle_destroyed;
 	u8			is_pool;
 	u8			is_internal;
 	u8			is_mmu_mapped;
@@ -3805,7 +3809,7 @@ __printf(4, 5) int hl_snprintf_resize(char **buf, size_t *size, size_t *offset,
 char *hl_format_as_binary(char *buf, size_t buf_len, u32 n);
 const char *hl_sync_engine_to_string(enum hl_sync_engine_type engine_type);
 
-void hl_mem_mgr_init(struct device *dev, struct hl_mem_mgr *mmg);
+void hl_mem_mgr_init(struct device *dev, struct hl_mem_mgr *mmg, u8 is_kernel_mem_mgr);
 void hl_mem_mgr_fini(struct hl_mem_mgr *mmg);
 int hl_mem_mgr_mmap(struct hl_mem_mgr *mmg, struct vm_area_struct *vma,
 		    void *args);
diff --git a/drivers/misc/habanalabs/common/habanalabs_drv.c b/drivers/misc/habanalabs/common/habanalabs_drv.c
index 7815c60df54e..a2983913d7c0 100644
--- a/drivers/misc/habanalabs/common/habanalabs_drv.c
+++ b/drivers/misc/habanalabs/common/habanalabs_drv.c
@@ -164,7 +164,7 @@ int hl_device_open(struct inode *inode, struct file *filp)
 	nonseekable_open(inode, filp);
 
 	hl_ctx_mgr_init(&hpriv->ctx_mgr);
-	hl_mem_mgr_init(hpriv->hdev->dev, &hpriv->mem_mgr);
+	hl_mem_mgr_init(hpriv->hdev->dev, &hpriv->mem_mgr, 0);
 
 	hpriv->taskpid = get_task_pid(current, PIDTYPE_PID);
 
diff --git a/drivers/misc/habanalabs/common/memory_mgr.c b/drivers/misc/habanalabs/common/memory_mgr.c
index 1936d653699e..e652db601f0e 100644
--- a/drivers/misc/habanalabs/common/memory_mgr.c
+++ b/drivers/misc/habanalabs/common/memory_mgr.c
@@ -309,14 +309,16 @@ int hl_mem_mgr_mmap(struct hl_mem_mgr *mmg, struct vm_area_struct *vma,
  *
  * @dev: owner device pointer
  * @mmg: structure to initialize
+ * @is_kernel_mem_mgr: indicate whether the memory manager is the per-device kernel memory manager
  *
  * Initialize an instance of unified memory manager
  */
-void hl_mem_mgr_init(struct device *dev, struct hl_mem_mgr *mmg)
+void hl_mem_mgr_init(struct device *dev, struct hl_mem_mgr *mmg, u8 is_kernel_mem_mgr)
 {
 	mmg->dev = dev;
 	spin_lock_init(&mmg->lock);
 	idr_init(&mmg->handles);
+	mmg->is_kernel_mem_mgr = is_kernel_mem_mgr;
 }
 
 /**
-- 
2.25.1

[PATCH 07/26] habanalabs: use dev_dbg() when hl_mmap_mem_buf_get() fails

[Oded Gabbay]

From: Tomer Tayar <ttayar@habana.ai>

As hl_mmap_mem_buf_get() is called also from IOCTLs which can have a
bad handle from user, modify the print for "no match to handle" to use
dev_dbg().
Calls to this function which are not dependent on user, already have an
error print when the function fails.

Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/memory_mgr.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/misc/habanalabs/common/memory_mgr.c b/drivers/misc/habanalabs/common/memory_mgr.c
index e652db601f0e..92d20ed465b4 100644
--- a/drivers/misc/habanalabs/common/memory_mgr.c
+++ b/drivers/misc/habanalabs/common/memory_mgr.c
@@ -25,8 +25,7 @@ struct hl_mmap_mem_buf *hl_mmap_mem_buf_get(struct hl_mem_mgr *mmg, u64 handle)
 	buf = idr_find(&mmg->handles, lower_32_bits(handle >> PAGE_SHIFT));
 	if (!buf) {
 		spin_unlock(&mmg->lock);
-		dev_warn(mmg->dev,
-			 "Buff get failed, no match to handle %#llx\n", handle);
+		dev_dbg(mmg->dev, "Buff get failed, no match to handle %#llx\n", handle);
 		return NULL;
 	}
 	kref_get(&buf->refcount);
-- 
2.25.1

[PATCH 08/26] habanalabs: make set_dram_properties an ASIC function

[Oded Gabbay]

From: Ohad Sharabi <osharabi@habana.ai>

As ASICs are evolving, we will need to update the DRAM properties at
various points because we may get different information from the f/w
at different points of the initialization.

This ASIC function is a foundation for this capability.

Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/habanalabs.h | 1 +
 drivers/misc/habanalabs/gaudi/gaudi.c       | 6 ++++++
 drivers/misc/habanalabs/gaudi2/gaudi2.c     | 3 ++-
 drivers/misc/habanalabs/goya/goya.c         | 6 ++++++
 4 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index ecf7e5da8f1d..893ebcba170b 100644
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -1683,6 +1683,7 @@ struct hl_asic_funcs {
 	int (*set_engine_cores)(struct hl_device *hdev, u32 *core_ids,
 					u32 num_cores, u32 core_command);
 	int (*send_device_activity)(struct hl_device *hdev, bool open);
+	int (*set_dram_properties)(struct hl_device *hdev);
 };
 
 
diff --git a/drivers/misc/habanalabs/gaudi/gaudi.c b/drivers/misc/habanalabs/gaudi/gaudi.c
index ae78f838f987..1b701a87c6fe 100644
--- a/drivers/misc/habanalabs/gaudi/gaudi.c
+++ b/drivers/misc/habanalabs/gaudi/gaudi.c
@@ -9134,6 +9134,11 @@ static u32 *gaudi_get_stream_master_qid_arr(void)
 	return gaudi_stream_master;
 }
 
+static int gaudi_set_dram_properties(struct hl_device *hdev)
+{
+	return 0;
+}
+
 static void gaudi_check_if_razwi_happened(struct hl_device *hdev)
 {
 }
@@ -9260,6 +9265,7 @@ static const struct hl_asic_funcs gaudi_funcs = {
 	.access_dev_mem = hl_access_dev_mem,
 	.set_dram_bar_base = gaudi_set_hbm_bar_base,
 	.send_device_activity = gaudi_send_device_activity,
+	.set_dram_properties = gaudi_set_dram_properties,
 };
 
 /**
diff --git a/drivers/misc/habanalabs/gaudi2/gaudi2.c b/drivers/misc/habanalabs/gaudi2/gaudi2.c
index c14b3bb16f96..10c017b8ddfa 100644
--- a/drivers/misc/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/misc/habanalabs/gaudi2/gaudi2.c
@@ -2485,7 +2485,7 @@ static int gaudi2_cpucp_info_get(struct hl_device *hdev)
 	 * at this point the DRAM parameters need to be updated according to data obtained
 	 * from the FW
 	 */
-	rc = gaudi2_set_dram_properties(hdev);
+	rc = hdev->asic_funcs->set_dram_properties(hdev);
 	if (rc)
 		return rc;
 
@@ -10467,6 +10467,7 @@ static const struct hl_asic_funcs gaudi2_funcs = {
 	.set_dram_bar_base = gaudi2_set_hbm_bar_base,
 	.set_engine_cores = gaudi2_set_engine_cores,
 	.send_device_activity = gaudi2_send_device_activity,
+	.set_dram_properties = gaudi2_set_dram_properties,
 };
 
 void gaudi2_set_asic_funcs(struct hl_device *hdev)
diff --git a/drivers/misc/habanalabs/goya/goya.c b/drivers/misc/habanalabs/goya/goya.c
index 0f083fcf81a6..ee0c7db16270 100644
--- a/drivers/misc/habanalabs/goya/goya.c
+++ b/drivers/misc/habanalabs/goya/goya.c
@@ -5420,6 +5420,11 @@ static int goya_scrub_device_dram(struct hl_device *hdev, u64 val)
 	return -EOPNOTSUPP;
 }
 
+static int goya_set_dram_properties(struct hl_device *hdev)
+{
+	return 0;
+}
+
 static int goya_send_device_activity(struct hl_device *hdev, bool open)
 {
 	return 0;
@@ -5518,6 +5523,7 @@ static const struct hl_asic_funcs goya_funcs = {
 	.access_dev_mem = hl_access_dev_mem,
 	.set_dram_bar_base = goya_set_ddr_bar_base,
 	.send_device_activity = goya_send_device_activity,
+	.set_dram_properties = goya_set_dram_properties,
 };
 
 /*
-- 
2.25.1

[PATCH 09/26] habanalabs: fix double assignment in MMU V1

[Oded Gabbay]

From: Marco Pagani <marpagan@redhat.com>

Removing double assignment of the hop2_pte_addr
variable in dram_default_mapping_fini().

Dead store reported by clang-analyzer.

Signed-off-by: Marco Pagani <marpagan@redhat.com>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/mmu/mmu_v1.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/misc/habanalabs/common/mmu/mmu_v1.c b/drivers/misc/habanalabs/common/mmu/mmu_v1.c
index 8a40de4a4761..d925dc4dd097 100644
--- a/drivers/misc/habanalabs/common/mmu/mmu_v1.c
+++ b/drivers/misc/habanalabs/common/mmu/mmu_v1.c
@@ -344,7 +344,6 @@ static void dram_default_mapping_fini(struct hl_ctx *ctx)
 		}
 	}
 
-	hop2_pte_addr = hop2_addr;
 	hop2_pte_addr = hop2_addr;
 	for (i = 0 ; i < num_of_hop3 ; i++) {
 		clear_pte(ctx, hop2_pte_addr);
-- 
2.25.1

[PATCH 10/26] habanalabs: update DRAM props according to preboot data

[Oded Gabbay]

From: Ohad Sharabi <osharabi@habana.ai>

If the f/w reports the binning masks at the preboot stage, the driver
must align its DRAM properties according to the new information.

Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/firmware_if.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/misc/habanalabs/common/firmware_if.c b/drivers/misc/habanalabs/common/firmware_if.c
index 4f364c3085fe..046866c673e2 100644
--- a/drivers/misc/habanalabs/common/firmware_if.c
+++ b/drivers/misc/habanalabs/common/firmware_if.c
@@ -2582,6 +2582,10 @@ static int hl_fw_dynamic_init_cpu(struct hl_device *hdev,
 			hdev->decoder_binning = le32_to_cpu(binning_info->dec_mask);
 			hdev->rotator_binning = le32_to_cpu(binning_info->rot_mask);
 
+			rc = hdev->asic_funcs->set_dram_properties(hdev);
+			if (rc)
+				goto out;
+
 			dev_dbg(hdev->dev,
 				"Read binning masks: tpc: 0x%llx, dram: 0x%llx, edma: 0x%x, dec: 0x%x, rot:0x%x\n",
 				hdev->tpc_binning, hdev->dram_binning, hdev->edma_binning,
-- 
2.25.1

[PATCH 11/26] habanalabs/gaudi2: count interrupt causes

[Oded Gabbay]

From: Ofir Bitton <obitton@habana.ai>

During event handling we extract interrupt cause and count it.
In case we could not find any cause we should add proper error.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/gaudi2/gaudi2.c | 361 +++++++++++++++++-------
 1 file changed, 252 insertions(+), 109 deletions(-)

diff --git a/drivers/misc/habanalabs/gaudi2/gaudi2.c b/drivers/misc/habanalabs/gaudi2/gaudi2.c
index 10c017b8ddfa..b8da2aa024ca 100644
--- a/drivers/misc/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/misc/habanalabs/gaudi2/gaudi2.c
@@ -53,6 +53,7 @@
 #define GAUDI2_HIF_HMMU_FULL_MASK		0xFFFF
 #define GAUDI2_DECODER_FULL_MASK		0x3FF
 
+#define GAUDI2_NA_EVENT_CAUSE			0xFF
 #define GAUDI2_NUM_OF_QM_ERR_CAUSE		18
 #define GAUDI2_NUM_OF_QM_LCP_ERR_CAUSE		25
 #define GAUDI2_NUM_OF_QM_ARB_ERR_CAUSE		3
@@ -6987,10 +6988,10 @@ static void print_qman_data_on_err(struct hl_device *hdev, u32 qid_base, u32 str
 		gaudi2_print_last_pqes_on_err(hdev, qid_base, i, qman_base, false);
 }
 
-static void gaudi2_handle_qman_err_generic(struct hl_device *hdev, const char *qm_name,
-						u64 qman_base, u32 qid_base)
+static int gaudi2_handle_qman_err_generic(struct hl_device *hdev, const char *qm_name,
+							u64 qman_base, u32 qid_base)
 {
-	u32 i, j, glbl_sts_val, arb_err_val, num_error_causes;
+	u32 i, j, glbl_sts_val, arb_err_val, num_error_causes, error_count = 0;
 	u64 glbl_sts_addr, arb_err_addr;
 	char reg_desc[32];
 
@@ -7013,12 +7014,14 @@ static void gaudi2_handle_qman_err_generic(struct hl_device *hdev, const char *q
 		}
 
 		for (j = 0 ; j < num_error_causes ; j++)
-			if (glbl_sts_val & BIT(j))
+			if (glbl_sts_val & BIT(j)) {
 				dev_err_ratelimited(hdev->dev, "%s %s. err cause: %s\n",
 						qm_name, reg_desc,
 						i == QMAN_STREAMS ?
 						gaudi2_qman_lower_cp_error_cause[j] :
 						gaudi2_qman_error_cause[j]);
+				error_count++;
+			}
 
 		print_qman_data_on_err(hdev, qid_base, i, qman_base);
 	}
@@ -7026,13 +7029,18 @@ static void gaudi2_handle_qman_err_generic(struct hl_device *hdev, const char *q
 	arb_err_val = RREG32(arb_err_addr);
 
 	if (!arb_err_val)
-		return;
+		goto out;
 
 	for (j = 0 ; j < GAUDI2_NUM_OF_QM_ARB_ERR_CAUSE ; j++) {
-		if (arb_err_val & BIT(j))
+		if (arb_err_val & BIT(j)) {
 			dev_err_ratelimited(hdev->dev, "%s ARB_ERR. err cause: %s\n",
 						qm_name, gaudi2_qman_arb_error_cause[j]);
+			error_count++;
+		}
 	}
+
+out:
+	return error_count;
 }
 
 static void gaudi2_razwi_rr_hbw_shared_printf_info(struct hl_device *hdev,
@@ -7675,17 +7683,17 @@ static void gaudi2_razwi_unmapped_addr_lbw_printf_info(struct hl_device *hdev, u
 }
 
 /* PSOC RAZWI interrupt occurs only when trying to access a bad address */
-static void gaudi2_ack_psoc_razwi_event_handler(struct hl_device *hdev, u64 *event_mask)
+static int gaudi2_ack_psoc_razwi_event_handler(struct hl_device *hdev, u64 *event_mask)
 {
 	u32 hbw_aw_set, hbw_ar_set, lbw_aw_set, lbw_ar_set, rtr_id, dcore_id, dcore_rtr_id, xy,
-								razwi_mask_info, razwi_intr = 0;
+						razwi_mask_info, razwi_intr = 0, error_count = 0;
 	int rtr_map_arr_len = NUM_OF_RTR_PER_DCORE * NUM_OF_DCORES;
 	u64 rtr_ctrl_base_addr;
 
 	if (hdev->pldm || !(hdev->fw_components & FW_TYPE_LINUX)) {
 		razwi_intr = RREG32(mmPSOC_GLOBAL_CONF_RAZWI_INTERRUPT);
 		if (!razwi_intr)
-			return;
+			return 0;
 	}
 
 	razwi_mask_info = RREG32(mmPSOC_GLOBAL_CONF_RAZWI_MASK_INFO);
@@ -7743,15 +7751,19 @@ static void gaudi2_ack_psoc_razwi_event_handler(struct hl_device *hdev, u64 *eve
 		gaudi2_razwi_unmapped_addr_lbw_printf_info(hdev, rtr_id,
 						rtr_ctrl_base_addr, false, event_mask);
 
+	error_count++;
+
 clear:
 	/* Clear Interrupts only on pldm or if f/w doesn't handle interrupts */
 	if (hdev->pldm || !(hdev->fw_components & FW_TYPE_LINUX))
 		WREG32(mmPSOC_GLOBAL_CONF_RAZWI_INTERRUPT, razwi_intr);
+
+	return error_count;
 }
 
-static void _gaudi2_handle_qm_sei_err(struct hl_device *hdev, u64 qman_base)
+static int _gaudi2_handle_qm_sei_err(struct hl_device *hdev, u64 qman_base)
 {
-	u32 i, sts_val, sts_clr_val = 0;
+	u32 i, sts_val, sts_clr_val = 0, error_count = 0;
 
 	sts_val = RREG32(qman_base + QM_SEI_STATUS_OFFSET);
 
@@ -7760,16 +7772,20 @@ static void _gaudi2_handle_qm_sei_err(struct hl_device *hdev, u64 qman_base)
 			dev_err_ratelimited(hdev->dev, "QM SEI. err cause: %s\n",
 						gaudi2_qm_sei_error_cause[i]);
 			sts_clr_val |= BIT(i);
+			error_count++;
 		}
 	}
 
 	WREG32(qman_base + QM_SEI_STATUS_OFFSET, sts_clr_val);
+
+	return error_count;
 }
 
-static void gaudi2_handle_qm_sei_err(struct hl_device *hdev, u16 event_type,
+static int gaudi2_handle_qm_sei_err(struct hl_device *hdev, u16 event_type,
 					struct hl_eq_razwi_info *razwi_info, u64 *event_mask)
 {
 	enum razwi_event_sources module;
+	u32 error_count = 0;
 	u64 qman_base;
 	u8 index;
 
@@ -7808,24 +7824,27 @@ static void gaudi2_handle_qm_sei_err(struct hl_device *hdev, u16 event_type,
 		module = RAZWI_ROT;
 		break;
 	default:
-		return;
+		return 0;
 	}
 
-	_gaudi2_handle_qm_sei_err(hdev, qman_base);
+	error_count = _gaudi2_handle_qm_sei_err(hdev, qman_base);
 
 	/* There is a single event per NIC macro, so should check its both QMAN blocks */
 	if (event_type >= GAUDI2_EVENT_NIC0_AXI_ERROR_RESPONSE &&
 			event_type <= GAUDI2_EVENT_NIC11_AXI_ERROR_RESPONSE)
-		_gaudi2_handle_qm_sei_err(hdev, qman_base + NIC_QM_OFFSET);
+		error_count += _gaudi2_handle_qm_sei_err(hdev,
+					qman_base + NIC_QM_OFFSET);
 
 	/* check if RAZWI happened */
 	if (razwi_info)
 		gaudi2_ack_module_razwi_event_handler(hdev, module, 0, 0, razwi_info, event_mask);
+
+	return error_count;
 }
 
-static void gaudi2_handle_qman_err(struct hl_device *hdev, u16 event_type)
+static int gaudi2_handle_qman_err(struct hl_device *hdev, u16 event_type)
 {
-	u32 qid_base;
+	u32 qid_base, error_count = 0;
 	u64 qman_base;
 	char desc[32];
 	u8 index;
@@ -7941,19 +7960,21 @@ static void gaudi2_handle_qman_err(struct hl_device *hdev, u16 event_type)
 		snprintf(desc, ARRAY_SIZE(desc), "ROTATOR1_QM");
 		break;
 	default:
-		return;
+		return 0;
 	}
 
-	gaudi2_handle_qman_err_generic(hdev, desc, qman_base, qid_base);
+	error_count = gaudi2_handle_qman_err_generic(hdev, desc, qman_base, qid_base);
 
 	/* Handle EDMA QM SEI here because there is no AXI error response event for EDMA */
 	if (event_type >= GAUDI2_EVENT_HDMA2_QM && event_type <= GAUDI2_EVENT_HDMA5_QM)
-		_gaudi2_handle_qm_sei_err(hdev, qman_base);
+		error_count += _gaudi2_handle_qm_sei_err(hdev, qman_base);
+
+	return error_count;
 }
 
-static void gaudi2_handle_arc_farm_sei_err(struct hl_device *hdev)
+static int gaudi2_handle_arc_farm_sei_err(struct hl_device *hdev)
 {
-	u32 i, sts_val, sts_clr_val = 0;
+	u32 i, sts_val, sts_clr_val = 0, error_count = 0;
 
 	sts_val = RREG32(mmARC_FARM_ARC0_AUX_ARC_SEI_INTR_STS);
 
@@ -7962,15 +7983,18 @@ static void gaudi2_handle_arc_farm_sei_err(struct hl_device *hdev)
 			dev_err_ratelimited(hdev->dev, "ARC SEI. err cause: %s\n",
 						gaudi2_arc_sei_error_cause[i]);
 			sts_clr_val |= BIT(i);
+			error_count++;
 		}
 	}
 
 	WREG32(mmARC_FARM_ARC0_AUX_ARC_SEI_INTR_CLR, sts_clr_val);
+
+	return error_count;
 }
 
-static void gaudi2_handle_cpu_sei_err(struct hl_device *hdev)
+static int gaudi2_handle_cpu_sei_err(struct hl_device *hdev)
 {
-	u32 i, sts_val, sts_clr_val = 0;
+	u32 i, sts_val, sts_clr_val = 0, error_count = 0;
 
 	sts_val = RREG32(mmCPU_IF_CPU_SEI_INTR_STS);
 
@@ -7979,50 +8003,63 @@ static void gaudi2_handle_cpu_sei_err(struct hl_device *hdev)
 			dev_err_ratelimited(hdev->dev, "CPU SEI. err cause: %s\n",
 						gaudi2_cpu_sei_error_cause[i]);
 			sts_clr_val |= BIT(i);
+			error_count++;
 		}
 	}
 
 	WREG32(mmCPU_IF_CPU_SEI_INTR_CLR, sts_clr_val);
+
+	return error_count;
 }
 
-static void gaudi2_handle_rot_err(struct hl_device *hdev, u8 rot_index,
+static int gaudi2_handle_rot_err(struct hl_device *hdev, u8 rot_index,
 					struct hl_eq_razwi_with_intr_cause *razwi_with_intr_cause,
 					u64 *event_mask)
 {
 	u64 intr_cause_data = le64_to_cpu(razwi_with_intr_cause->intr_cause.intr_cause_data);
+	u32 error_count = 0;
 	int i;
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_ROT_ERR_CAUSE ; i++)
-		if (intr_cause_data & BIT(i))
+		if (intr_cause_data & BIT(i)) {
 			dev_err_ratelimited(hdev->dev, "ROT%u. err cause: %s\n",
 						rot_index, guadi2_rot_error_cause[i]);
+			error_count++;
+		}
 
 	/* check if RAZWI happened */
 	gaudi2_ack_module_razwi_event_handler(hdev, RAZWI_ROT, rot_index, 0,
 						&razwi_with_intr_cause->razwi_info, event_mask);
+
+	return error_count;
 }
 
-static void gaudi2_tpc_ack_interrupts(struct hl_device *hdev, u8 tpc_index, char *interrupt_name,
+static int gaudi2_tpc_ack_interrupts(struct hl_device *hdev,  u8 tpc_index, char *interrupt_name,
 					struct hl_eq_razwi_with_intr_cause *razwi_with_intr_cause,
 					u64 *event_mask)
 {
 	u64 intr_cause_data = le64_to_cpu(razwi_with_intr_cause->intr_cause.intr_cause_data);
+	u32 error_count = 0;
 	int i;
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_TPC_INTR_CAUSE ; i++)
-		if (intr_cause_data & BIT(i))
+		if (intr_cause_data & BIT(i)) {
 			dev_err_ratelimited(hdev->dev, "TPC%d_%s interrupt cause: %s\n",
 					tpc_index, interrupt_name, gaudi2_tpc_interrupts_cause[i]);
+			error_count++;
+		}
 
 	/* check if RAZWI happened */
 	gaudi2_ack_module_razwi_event_handler(hdev, RAZWI_TPC, tpc_index, 0,
 						&razwi_with_intr_cause->razwi_info, event_mask);
+
+	return error_count;
 }
 
-static void gaudi2_handle_dec_err(struct hl_device *hdev, u8 dec_index, const char *interrupt_name,
+static int gaudi2_handle_dec_err(struct hl_device *hdev, u8 dec_index, const char *interrupt_name,
 				struct hl_eq_razwi_info *razwi_info, u64 *event_mask)
 {
-	u32 sts_addr, sts_val, sts_clr_val = 0;
+	u32 sts_addr, sts_val, sts_clr_val = 0, error_count = 0;
 	int i;
 
 	if (dec_index < NUM_OF_VDEC_PER_DCORE * NUM_OF_DCORES)
@@ -8042,6 +8079,7 @@ static void gaudi2_handle_dec_err(struct hl_device *hdev, u8 dec_index, const ch
 			dev_err_ratelimited(hdev->dev, "DEC%u_%s err cause: %s\n",
 					dec_index, interrupt_name, gaudi2_dec_error_cause[i]);
 			sts_clr_val |= BIT(i);
+			error_count++;
 		}
 	}
 
@@ -8051,12 +8089,14 @@ static void gaudi2_handle_dec_err(struct hl_device *hdev, u8 dec_index, const ch
 
 	/* Write 1 clear errors */
 	WREG32(sts_addr, sts_clr_val);
+
+	return error_count;
 }
 
-static void gaudi2_handle_mme_err(struct hl_device *hdev, u8 mme_index, const char *interrupt_name,
+static int gaudi2_handle_mme_err(struct hl_device *hdev, u8 mme_index, const char *interrupt_name,
 				struct hl_eq_razwi_info *razwi_info, u64 *event_mask)
 {
-	u32 sts_addr, sts_val, sts_clr_addr, sts_clr_val = 0;
+	u32 sts_addr, sts_val, sts_clr_addr, sts_clr_val = 0, error_count = 0;
 	int i;
 
 	sts_addr = mmDCORE0_MME_CTRL_LO_INTR_CAUSE + DCORE_OFFSET * mme_index;
@@ -8069,6 +8109,7 @@ static void gaudi2_handle_mme_err(struct hl_device *hdev, u8 mme_index, const ch
 			dev_err_ratelimited(hdev->dev, "MME%u_%s err cause: %s\n",
 					mme_index, interrupt_name, guadi2_mme_error_cause[i]);
 			sts_clr_val |= BIT(i);
+			error_count++;
 		}
 	}
 
@@ -8078,23 +8119,29 @@ static void gaudi2_handle_mme_err(struct hl_device *hdev, u8 mme_index, const ch
 							event_mask);
 
 	WREG32(sts_clr_addr, sts_clr_val);
+
+	return error_count;
 }
 
-static void gaudi2_handle_mme_sbte_err(struct hl_device *hdev, u8 mme_index, u8 sbte_index,
+static int gaudi2_handle_mme_sbte_err(struct hl_device *hdev, u8 mme_index, u8 sbte_index,
 					u64 intr_cause_data)
 {
-	int i;
+	int i, error_count = 0;
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_MME_SBTE_ERR_CAUSE ; i++)
-		if (intr_cause_data & BIT(i))
+		if (intr_cause_data & BIT(i)) {
 			dev_err_ratelimited(hdev->dev, "MME%uSBTE%u_AXI_ERR_RSP err cause: %s\n",
 					mme_index, sbte_index, guadi2_mme_sbte_error_cause[i]);
+			error_count++;
+		}
+
+	return error_count;
 }
 
-static void gaudi2_handle_mme_wap_err(struct hl_device *hdev, u8 mme_index,
+static int gaudi2_handle_mme_wap_err(struct hl_device *hdev, u8 mme_index,
 					struct hl_eq_razwi_info *razwi_info, u64 *event_mask)
 {
-	u32 sts_addr, sts_val, sts_clr_addr, sts_clr_val = 0;
+	u32 sts_addr, sts_val, sts_clr_addr, sts_clr_val = 0, error_count = 0;
 	int i;
 
 	sts_addr = mmDCORE0_MME_ACC_INTR_CAUSE + DCORE_OFFSET * mme_index;
@@ -8108,6 +8155,7 @@ static void gaudi2_handle_mme_wap_err(struct hl_device *hdev, u8 mme_index,
 					"MME%u_WAP_SOURCE_RESULT_INVALID err cause: %s\n",
 					mme_index, guadi2_mme_wap_error_cause[i]);
 			sts_clr_val |= BIT(i);
+			error_count++;
 		}
 	}
 
@@ -8118,10 +8166,13 @@ static void gaudi2_handle_mme_wap_err(struct hl_device *hdev, u8 mme_index,
 						event_mask);
 
 	WREG32(sts_clr_addr, sts_clr_val);
+
+	return error_count;
 }
 
-static void gaudi2_handle_kdma_core_event(struct hl_device *hdev, u64 intr_cause_data)
+static int gaudi2_handle_kdma_core_event(struct hl_device *hdev, u64 intr_cause_data)
 {
+	u32 error_count = 0;
 	int i;
 
 	/* If an AXI read or write error is received, an error is reported and
@@ -8130,19 +8181,28 @@ static void gaudi2_handle_kdma_core_event(struct hl_device *hdev, u64 intr_cause
 	 * the actual error caused by a LBW KDMA transaction.
 	 */
 	for (i = 0 ; i < GAUDI2_NUM_OF_DMA_CORE_INTR_CAUSE ; i++)
-		if (intr_cause_data & BIT(i))
+		if (intr_cause_data & BIT(i)) {
 			dev_err_ratelimited(hdev->dev, "kdma core err cause: %s\n",
 						gaudi2_kdma_core_interrupts_cause[i]);
+			error_count++;
+		}
+
+	return error_count;
 }
 
-static void gaudi2_handle_dma_core_event(struct hl_device *hdev, u64 intr_cause_data)
+static int gaudi2_handle_dma_core_event(struct hl_device *hdev, u64 intr_cause_data)
 {
+	u32 error_count = 0;
 	int i;
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_DMA_CORE_INTR_CAUSE ; i++)
-		if (intr_cause_data & BIT(i))
+		if (intr_cause_data & BIT(i)) {
 			dev_err_ratelimited(hdev->dev, "dma core err cause: %s\n",
 						gaudi2_dma_core_interrupts_cause[i]);
+			error_count++;
+		}
+
+	return error_count;
 }
 
 static void gaudi2_print_pcie_mstr_rr_mstr_if_razwi_info(struct hl_device *hdev, u64 *event_mask)
@@ -8178,9 +8238,10 @@ static void gaudi2_print_pcie_mstr_rr_mstr_if_razwi_info(struct hl_device *hdev,
 	}
 }
 
-static void gaudi2_print_pcie_addr_dec_info(struct hl_device *hdev, u64 intr_cause_data,
+static int gaudi2_print_pcie_addr_dec_info(struct hl_device *hdev, u64 intr_cause_data,
 						u64 *event_mask)
 {
+	u32 error_count = 0;
 	int i;
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_PCIE_ADDR_DEC_ERR_CAUSE ; i++) {
@@ -8189,6 +8250,7 @@ static void gaudi2_print_pcie_addr_dec_info(struct hl_device *hdev, u64 intr_cau
 
 		dev_err_ratelimited(hdev->dev, "PCIE ADDR DEC Error: %s\n",
 					gaudi2_pcie_addr_dec_error_cause[i]);
+		error_count++;
 
 		switch (intr_cause_data & BIT_ULL(i)) {
 		case PCIE_WRAP_PCIE_IC_SEI_INTR_IND_AXI_LBW_ERR_INTR_MASK:
@@ -8198,33 +8260,44 @@ static void gaudi2_print_pcie_addr_dec_info(struct hl_device *hdev, u64 intr_cau
 			break;
 		}
 	}
+
+	return error_count;
 }
 
-static void gaudi2_handle_pif_fatal(struct hl_device *hdev, u64 intr_cause_data)
+static int gaudi2_handle_pif_fatal(struct hl_device *hdev, u64 intr_cause_data)
 
 {
+	u32 error_count = 0;
 	int i;
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_PMMU_FATAL_ERR_CAUSE ; i++) {
-		if (intr_cause_data & BIT_ULL(i))
+		if (intr_cause_data & BIT_ULL(i)) {
 			dev_err_ratelimited(hdev->dev, "PMMU PIF err cause: %s\n",
 					gaudi2_pmmu_fatal_interrupts_cause[i]);
+			error_count++;
+		}
 	}
+
+	return error_count;
 }
 
-static void gaudi2_handle_hif_fatal(struct hl_device *hdev, u16 event_type, u64 intr_cause_data)
+static int gaudi2_handle_hif_fatal(struct hl_device *hdev, u16 event_type, u64 intr_cause_data)
 {
-	u32 dcore_id, hif_id;
+	u32 dcore_id, hif_id, error_count = 0;
 	int i;
 
 	dcore_id = (event_type - GAUDI2_EVENT_HIF0_FATAL) / 4;
 	hif_id = (event_type - GAUDI2_EVENT_HIF0_FATAL) % 4;
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_HIF_FATAL_ERR_CAUSE ; i++) {
-		if (intr_cause_data & BIT_ULL(i))
+		if (intr_cause_data & BIT_ULL(i)) {
 			dev_err_ratelimited(hdev->dev, "DCORE%u_HIF%u: %s\n", dcore_id, hif_id,
 					gaudi2_hif_fatal_interrupts_cause[i]);
+			error_count++;
+		}
 	}
+
+	return error_count;
 }
 
 static void gaudi2_handle_page_error(struct hl_device *hdev, u64 mmu_base, bool is_pmmu,
@@ -8270,10 +8343,10 @@ static void gaudi2_handle_access_error(struct hl_device *hdev, u64 mmu_base, boo
 	WREG32(mmu_base + MMU_OFFSET(mmDCORE0_HMMU0_MMU_ACCESS_ERROR_CAPTURE), 0);
 }
 
-static void gaudi2_handle_mmu_spi_sei_generic(struct hl_device *hdev, const char *mmu_name,
+static int gaudi2_handle_mmu_spi_sei_generic(struct hl_device *hdev, const char *mmu_name,
 						u64 mmu_base, bool is_pmmu, u64 *event_mask)
 {
-	u32 spi_sei_cause, interrupt_clr = 0x0;
+	u32 spi_sei_cause, interrupt_clr = 0x0, error_count = 0;
 	int i;
 
 	spi_sei_cause = RREG32(mmu_base + MMU_SPI_SEI_CAUSE_OFFSET);
@@ -8290,6 +8363,8 @@ static void gaudi2_handle_mmu_spi_sei_generic(struct hl_device *hdev, const char
 
 			if (gaudi2_mmu_spi_sei[i].clear_bit >= 0)
 				interrupt_clr |= BIT(gaudi2_mmu_spi_sei[i].clear_bit);
+
+			error_count++;
 		}
 	}
 
@@ -8298,12 +8373,14 @@ static void gaudi2_handle_mmu_spi_sei_generic(struct hl_device *hdev, const char
 
 	/* Clear interrupt */
 	WREG32(mmu_base + MMU_INTERRUPT_CLR_OFFSET, interrupt_clr);
+
+	return error_count;
 }
 
-static void gaudi2_handle_sm_err(struct hl_device *hdev, u8 sm_index)
+static int gaudi2_handle_sm_err(struct hl_device *hdev, u8 sm_index)
 {
-	u32 sei_cause_addr, sei_cause_val, sei_cause_cause, sei_cause_log;
-	u32 cq_intr_addr, cq_intr_val, cq_intr_queue_index;
+	u32 sei_cause_addr, sei_cause_val, sei_cause_cause, sei_cause_log,
+		cq_intr_addr, cq_intr_val, cq_intr_queue_index, error_count = 0;
 	int i;
 
 	sei_cause_addr = mmDCORE0_SYNC_MNGR_GLBL_SM_SEI_CAUSE + DCORE_OFFSET * sm_index;
@@ -8328,6 +8405,7 @@ static void gaudi2_handle_sm_err(struct hl_device *hdev, u8 sm_index)
 					gaudi2_sm_sei_cause[i].cause_name,
 					gaudi2_sm_sei_cause[i].log_name,
 					sei_cause_log & gaudi2_sm_sei_cause[i].log_mask);
+			error_count++;
 			break;
 		}
 
@@ -8343,16 +8421,20 @@ static void gaudi2_handle_sm_err(struct hl_device *hdev, u8 sm_index)
 
 		dev_err_ratelimited(hdev->dev, "SM%u err. err cause: CQ_INTR. queue index: %u\n",
 				sm_index, cq_intr_queue_index);
+		error_count++;
 
 		/* Clear CQ_INTR */
 		WREG32(cq_intr_addr, 0);
 	}
+
+	return error_count;
 }
 
-static void gaudi2_handle_mmu_spi_sei_err(struct hl_device *hdev, u16 event_type, u64 *event_mask)
+static int gaudi2_handle_mmu_spi_sei_err(struct hl_device *hdev, u16 event_type, u64 *event_mask)
 {
 	bool is_pmmu = false;
 	char desc[32];
+	u32 error_count = 0;
 	u64 mmu_base;
 	u8 index;
 
@@ -8404,10 +8486,12 @@ static void gaudi2_handle_mmu_spi_sei_err(struct hl_device *hdev, u16 event_type
 		snprintf(desc, ARRAY_SIZE(desc), "PMMU");
 		break;
 	default:
-		return;
+		return 0;
 	}
 
-	gaudi2_handle_mmu_spi_sei_generic(hdev, desc, mmu_base, is_pmmu, event_mask);
+	error_count = gaudi2_handle_mmu_spi_sei_generic(hdev, desc, mmu_base, is_pmmu, event_mask);
+
+	return error_count;
 }
 
 
@@ -8586,21 +8670,30 @@ static bool gaudi2_handle_hbm_mc_sei_err(struct hl_device *hdev, u16 event_type,
 	return require_hard_reset;
 }
 
-static void gaudi2_handle_hbm_cattrip(struct hl_device *hdev, u64 intr_cause_data)
+static int gaudi2_handle_hbm_cattrip(struct hl_device *hdev, u64 intr_cause_data)
 {
-	dev_err(hdev->dev,
-		"HBM catastrophic temperature error (CATTRIP) cause %#llx\n",
-		intr_cause_data);
+	if (intr_cause_data) {
+		dev_err(hdev->dev,
+			"HBM catastrophic temperature error (CATTRIP) cause %#llx\n",
+			intr_cause_data);
+		return 1;
+	}
+
+	return 0;
 }
 
-static void gaudi2_handle_hbm_mc_spi(struct hl_device *hdev, u64 intr_cause_data)
+static int gaudi2_handle_hbm_mc_spi(struct hl_device *hdev, u64 intr_cause_data)
 {
-	u32 i;
+	u32 i, error_count = 0;
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_HBM_MC_SPI_CAUSE ; i++)
-		if (intr_cause_data & hbm_mc_spi[i].mask)
+		if (intr_cause_data & hbm_mc_spi[i].mask) {
 			dev_dbg(hdev->dev, "HBM spi event: notification cause(%s)\n",
 				hbm_mc_spi[i].cause);
+			error_count++;
+		}
+
+	return error_count;
 }
 
 static void gaudi2_print_clk_change_info(struct hl_device *hdev, u16 event_type, u64 *event_mask)
@@ -8657,9 +8750,9 @@ static void gaudi2_print_out_of_sync_info(struct hl_device *hdev,
 		le32_to_cpu(sync_err->pi), le32_to_cpu(sync_err->ci), q->pi, atomic_read(&q->ci));
 }
 
-static void gaudi2_handle_pcie_p2p_msix(struct hl_device *hdev)
+static int gaudi2_handle_pcie_p2p_msix(struct hl_device *hdev)
 {
-	u32 p2p_intr, msix_gw_intr;
+	u32 p2p_intr, msix_gw_intr, error_count = 0;
 
 	p2p_intr = RREG32(mmPCIE_WRAP_P2P_INTR);
 	msix_gw_intr = RREG32(mmPCIE_WRAP_MSIX_GW_INTR);
@@ -8670,6 +8763,7 @@ static void gaudi2_handle_pcie_p2p_msix(struct hl_device *hdev)
 			RREG32(mmPCIE_WRAP_P2P_REQ_ID));
 
 		WREG32(mmPCIE_WRAP_P2P_INTR, 0x1);
+		error_count++;
 	}
 
 	if (msix_gw_intr) {
@@ -8678,13 +8772,16 @@ static void gaudi2_handle_pcie_p2p_msix(struct hl_device *hdev)
 			RREG32(mmPCIE_WRAP_MSIX_GW_VEC));
 
 		WREG32(mmPCIE_WRAP_MSIX_GW_INTR, 0x1);
+		error_count++;
 	}
+
+	return error_count;
 }
 
-static void gaudi2_handle_pcie_drain(struct hl_device *hdev,
+static int gaudi2_handle_pcie_drain(struct hl_device *hdev,
 			struct hl_eq_pcie_drain_ind_data *drain_data)
 {
-	u64 lbw_rd, lbw_wr, hbw_rd, hbw_wr, cause;
+	u64 lbw_rd, lbw_wr, hbw_rd, hbw_wr, cause, error_count = 0;
 
 	cause = le64_to_cpu(drain_data->intr_cause.intr_cause_data);
 	lbw_rd = le64_to_cpu(drain_data->drain_rd_addr_lbw);
@@ -8692,26 +8789,37 @@ static void gaudi2_handle_pcie_drain(struct hl_device *hdev,
 	hbw_rd = le64_to_cpu(drain_data->drain_rd_addr_hbw);
 	hbw_wr = le64_to_cpu(drain_data->drain_wr_addr_hbw);
 
-	if (cause & BIT_ULL(0))
+	if (cause & BIT_ULL(0)) {
 		dev_err_ratelimited(hdev->dev,
 			"PCIE AXI drain LBW completed, read_err %u, write_err %u\n",
 			!!lbw_rd, !!lbw_wr);
+		error_count++;
+	}
 
-	if (cause & BIT_ULL(1))
+	if (cause & BIT_ULL(1)) {
 		dev_err_ratelimited(hdev->dev,
 			"PCIE AXI drain HBW completed, raddr %#llx, waddr %#llx\n",
 			hbw_rd, hbw_wr);
+		error_count++;
+	}
+
+	return error_count;
 }
 
-static void gaudi2_handle_psoc_drain(struct hl_device *hdev, u64 intr_cause_data)
+static int gaudi2_handle_psoc_drain(struct hl_device *hdev, u64 intr_cause_data)
 {
+	u32 error_count = 0;
 	int i;
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_AXI_DRAIN_ERR_CAUSE ; i++) {
-		if (intr_cause_data & BIT_ULL(i))
+		if (intr_cause_data & BIT_ULL(i)) {
 			dev_err_ratelimited(hdev->dev, "PSOC %s completed\n",
 				gaudi2_psoc_axi_drain_interrupts_cause[i]);
+			error_count++;
+		}
 	}
+
+	return error_count;
 }
 
 static void gaudi2_print_cpu_pkt_failure_info(struct hl_device *hdev,
@@ -8724,8 +8832,7 @@ static void gaudi2_print_cpu_pkt_failure_info(struct hl_device *hdev,
 		le32_to_cpu(sync_err->pi), le32_to_cpu(sync_err->ci), q->pi, atomic_read(&q->ci));
 }
 
-static void hl_arc_event_handle(struct hl_device *hdev,
-					struct hl_eq_engine_arc_intr_data *data)
+static int hl_arc_event_handle(struct hl_device *hdev, struct hl_eq_engine_arc_intr_data *data)
 {
 	struct hl_engine_arc_dccm_queue_full_irq *q;
 	u32 intr_type, engine_id;
@@ -8742,9 +8849,10 @@ static void hl_arc_event_handle(struct hl_device *hdev,
 		dev_err_ratelimited(hdev->dev,
 				"ARC DCCM Full event: EngId: %u, Intr_type: %u, Qidx: %u\n",
 				engine_id, intr_type, q->queue_index);
-		break;
+		return 1;
 	default:
 		dev_err_ratelimited(hdev->dev, "Unknown ARC event type\n");
+		return 0;
 	}
 }
 
@@ -8752,7 +8860,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 {
 	struct gaudi2_device *gaudi2 = hdev->asic_specific;
 	bool reset_required = false, is_critical = false;
-	u32 ctl, reset_flags = HL_DRV_RESET_HARD;
+	u32 ctl, reset_flags = HL_DRV_RESET_HARD, error_count = 0;
 	int index, sbte_index;
 	u64 event_mask = 0;
 	u16 event_type;
@@ -8779,6 +8887,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		reset_required = gaudi2_handle_ecc_event(hdev, event_type, &eq_entry->ecc_data);
 		is_critical = eq_entry->ecc_data.is_critical;
+		error_count++;
 		break;
 
 	case GAUDI2_EVENT_TPC0_QM ... GAUDI2_EVENT_PDMA1_QM:
@@ -8786,48 +8895,50 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 	case GAUDI2_EVENT_ROTATOR0_ROT0_QM ... GAUDI2_EVENT_ROTATOR1_ROT1_QM:
 		fallthrough;
 	case GAUDI2_EVENT_NIC0_QM0 ... GAUDI2_EVENT_NIC11_QM1:
-		gaudi2_handle_qman_err(hdev, event_type);
+		error_count = gaudi2_handle_qman_err(hdev, event_type);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
 	case GAUDI2_EVENT_ARC_AXI_ERROR_RESPONSE_0:
 		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
-		gaudi2_handle_arc_farm_sei_err(hdev);
+		error_count = gaudi2_handle_arc_farm_sei_err(hdev);
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_CPU_AXI_ERR_RSP:
-		gaudi2_handle_cpu_sei_err(hdev);
+		error_count = gaudi2_handle_cpu_sei_err(hdev);
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_PDMA_CH0_AXI_ERR_RSP:
 	case GAUDI2_EVENT_PDMA_CH1_AXI_ERR_RSP:
 		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
-		gaudi2_handle_qm_sei_err(hdev, event_type, &eq_entry->razwi_info, &event_mask);
+		error_count = gaudi2_handle_qm_sei_err(hdev, event_type,
+					&eq_entry->razwi_info, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
 	case GAUDI2_EVENT_ROTATOR0_AXI_ERROR_RESPONSE:
 	case GAUDI2_EVENT_ROTATOR1_AXI_ERROR_RESPONSE:
 		index = event_type - GAUDI2_EVENT_ROTATOR0_AXI_ERROR_RESPONSE;
-		gaudi2_handle_rot_err(hdev, index, &eq_entry->razwi_with_intr_cause, &event_mask);
-		gaudi2_handle_qm_sei_err(hdev, event_type, NULL, &event_mask);
+		error_count = gaudi2_handle_rot_err(hdev, index,
+					&eq_entry->razwi_with_intr_cause, &event_mask);
+		error_count += gaudi2_handle_qm_sei_err(hdev, event_type, NULL, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
 	case GAUDI2_EVENT_TPC0_AXI_ERR_RSP ... GAUDI2_EVENT_TPC24_AXI_ERR_RSP:
 		index = event_type - GAUDI2_EVENT_TPC0_AXI_ERR_RSP;
-		gaudi2_tpc_ack_interrupts(hdev, index, "AXI_ERR_RSP",
+		error_count = gaudi2_tpc_ack_interrupts(hdev, index, "AXI_ERR_RSP",
 						&eq_entry->razwi_with_intr_cause, &event_mask);
-		gaudi2_handle_qm_sei_err(hdev, event_type, NULL, &event_mask);
+		error_count += gaudi2_handle_qm_sei_err(hdev, event_type, NULL, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
 	case GAUDI2_EVENT_DEC0_AXI_ERR_RSPONSE ... GAUDI2_EVENT_DEC9_AXI_ERR_RSPONSE:
 		index = event_type - GAUDI2_EVENT_DEC0_AXI_ERR_RSPONSE;
-		gaudi2_handle_dec_err(hdev, index, "AXI_ERR_RESPONSE", &eq_entry->razwi_info,
-					&event_mask);
+		error_count = gaudi2_handle_dec_err(hdev, index, "AXI_ERR_RESPONSE",
+					&eq_entry->razwi_info, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
@@ -8858,8 +8969,8 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 	case GAUDI2_EVENT_TPC24_KERNEL_ERR:
 		index = (event_type - GAUDI2_EVENT_TPC0_KERNEL_ERR) /
 			(GAUDI2_EVENT_TPC1_KERNEL_ERR - GAUDI2_EVENT_TPC0_KERNEL_ERR);
-		gaudi2_tpc_ack_interrupts(hdev, index, "KRN_ERR", &eq_entry->razwi_with_intr_cause,
-						&event_mask);
+		error_count = gaudi2_tpc_ack_interrupts(hdev, index, "KRN_ERR",
+				&eq_entry->razwi_with_intr_cause, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
@@ -8875,7 +8986,8 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 	case GAUDI2_EVENT_DEC9_SPI:
 		index = (event_type - GAUDI2_EVENT_DEC0_SPI) /
 				(GAUDI2_EVENT_DEC1_SPI - GAUDI2_EVENT_DEC0_SPI);
-		gaudi2_handle_dec_err(hdev, index, "SPI", &eq_entry->razwi_info, &event_mask);
+		error_count = gaudi2_handle_dec_err(hdev, index, "SPI",
+					&eq_entry->razwi_info, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
@@ -8886,9 +8998,9 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		index = (event_type - GAUDI2_EVENT_MME0_CTRL_AXI_ERROR_RESPONSE) /
 				(GAUDI2_EVENT_MME1_CTRL_AXI_ERROR_RESPONSE -
 						GAUDI2_EVENT_MME0_CTRL_AXI_ERROR_RESPONSE);
-		gaudi2_handle_mme_err(hdev, index,
+		error_count = gaudi2_handle_mme_err(hdev, index,
 				"CTRL_AXI_ERROR_RESPONSE", &eq_entry->razwi_info, &event_mask);
-		gaudi2_handle_qm_sei_err(hdev, event_type, NULL, &event_mask);
+		error_count += gaudi2_handle_qm_sei_err(hdev, event_type, NULL, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
@@ -8899,8 +9011,8 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		index = (event_type - GAUDI2_EVENT_MME0_QMAN_SW_ERROR) /
 				(GAUDI2_EVENT_MME1_QMAN_SW_ERROR -
 					GAUDI2_EVENT_MME0_QMAN_SW_ERROR);
-		gaudi2_handle_mme_err(hdev, index, "QMAN_SW_ERROR", &eq_entry->razwi_info,
-					&event_mask);
+		error_count = gaudi2_handle_mme_err(hdev, index, "QMAN_SW_ERROR",
+					&eq_entry->razwi_info, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
@@ -8911,25 +9023,26 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		index = (event_type - GAUDI2_EVENT_MME0_WAP_SOURCE_RESULT_INVALID) /
 				(GAUDI2_EVENT_MME1_WAP_SOURCE_RESULT_INVALID -
 					GAUDI2_EVENT_MME0_WAP_SOURCE_RESULT_INVALID);
-		gaudi2_handle_mme_wap_err(hdev, index, &eq_entry->razwi_info, &event_mask);
+		error_count = gaudi2_handle_mme_wap_err(hdev, index,
+					&eq_entry->razwi_info, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
 	case GAUDI2_EVENT_KDMA_CH0_AXI_ERR_RSP:
 	case GAUDI2_EVENT_KDMA0_CORE:
-		gaudi2_handle_kdma_core_event(hdev,
+		error_count = gaudi2_handle_kdma_core_event(hdev,
 					le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_HDMA2_CORE ... GAUDI2_EVENT_PDMA1_CORE:
-		gaudi2_handle_dma_core_event(hdev,
+		error_count = gaudi2_handle_dma_core_event(hdev,
 					le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
 	case GAUDI2_EVENT_PCIE_ADDR_DEC_ERR:
-		gaudi2_print_pcie_addr_dec_info(hdev,
+		error_count = gaudi2_print_pcie_addr_dec_info(hdev,
 				le64_to_cpu(eq_entry->intr_cause.intr_cause_data), &event_mask);
 		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
@@ -8939,27 +9052,27 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 	case GAUDI2_EVENT_HMMU_0_AXI_ERR_RSP ... GAUDI2_EVENT_HMMU_12_AXI_ERR_RSP:
 	case GAUDI2_EVENT_PMMU0_PAGE_FAULT_WR_PERM ... GAUDI2_EVENT_PMMU0_SECURITY_ERROR:
 	case GAUDI2_EVENT_PMMU_AXI_ERR_RSP_0:
-		gaudi2_handle_mmu_spi_sei_err(hdev, event_type, &event_mask);
+		error_count = gaudi2_handle_mmu_spi_sei_err(hdev, event_type, &event_mask);
 		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
 	case GAUDI2_EVENT_HIF0_FATAL ... GAUDI2_EVENT_HIF12_FATAL:
-		gaudi2_handle_hif_fatal(hdev, event_type,
+		error_count = gaudi2_handle_hif_fatal(hdev, event_type,
 				le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
 		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_PMMU_FATAL_0:
-		gaudi2_handle_pif_fatal(hdev,
+		error_count = gaudi2_handle_pif_fatal(hdev,
 				le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
 		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_PSOC63_RAZWI_OR_PID_MIN_MAX_INTERRUPT:
-		gaudi2_ack_psoc_razwi_event_handler(hdev, &event_mask);
+		error_count = gaudi2_ack_psoc_razwi_event_handler(hdev, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
@@ -8969,33 +9082,39 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 			reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 			reset_required = true;
 		}
+		error_count++;
 		break;
 
 	case GAUDI2_EVENT_HBM_CATTRIP_0 ... GAUDI2_EVENT_HBM_CATTRIP_5:
-		gaudi2_handle_hbm_cattrip(hdev, le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
+		error_count = gaudi2_handle_hbm_cattrip(hdev,
+				le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_HBM0_MC0_SPI ... GAUDI2_EVENT_HBM5_MC1_SPI:
-		gaudi2_handle_hbm_mc_spi(hdev, le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
+		error_count = gaudi2_handle_hbm_mc_spi(hdev,
+				le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_PCIE_DRAIN_COMPLETE:
-		gaudi2_handle_pcie_drain(hdev, &eq_entry->pcie_drain_ind_data);
+		error_count = gaudi2_handle_pcie_drain(hdev, &eq_entry->pcie_drain_ind_data);
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_PSOC59_RPM_ERROR_OR_DRAIN:
-		gaudi2_handle_psoc_drain(hdev, le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
+		error_count = gaudi2_handle_psoc_drain(hdev,
+				le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_CPU_AXI_ECC:
+		error_count = GAUDI2_NA_EVENT_CAUSE;
 		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 	case GAUDI2_EVENT_CPU_L2_RAM_ECC:
+		error_count = GAUDI2_NA_EVENT_CAUSE;
 		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
@@ -9009,25 +9128,30 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		sbte_index = (event_type - GAUDI2_EVENT_MME0_SBTE0_AXI_ERR_RSP) %
 				(GAUDI2_EVENT_MME1_SBTE0_AXI_ERR_RSP -
 					GAUDI2_EVENT_MME0_SBTE0_AXI_ERR_RSP);
-		gaudi2_handle_mme_sbte_err(hdev, index, sbte_index,
-						le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
+		error_count = gaudi2_handle_mme_sbte_err(hdev, index, sbte_index,
+					le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 	case GAUDI2_EVENT_VM0_ALARM_A ... GAUDI2_EVENT_VM3_ALARM_B:
+		error_count = GAUDI2_NA_EVENT_CAUSE;
 		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 	case GAUDI2_EVENT_PSOC_AXI_ERR_RSP:
+		error_count = GAUDI2_NA_EVENT_CAUSE;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 	case GAUDI2_EVENT_PSOC_PRSTN_FALL:
+		error_count = GAUDI2_NA_EVENT_CAUSE;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 	case GAUDI2_EVENT_PCIE_APB_TIMEOUT:
+		error_count = GAUDI2_NA_EVENT_CAUSE;
 		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 	case GAUDI2_EVENT_PCIE_FATAL_ERR:
+		error_count = GAUDI2_NA_EVENT_CAUSE;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 	case GAUDI2_EVENT_TPC0_BMON_SPMU:
@@ -9080,6 +9204,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 	case GAUDI2_EVENT_DEC8_BMON_SPMU:
 	case GAUDI2_EVENT_DEC9_BMON_SPMU:
 	case GAUDI2_EVENT_ROTATOR0_BMON_SPMU ... GAUDI2_EVENT_SM3_BMON_SPMU:
+		error_count = GAUDI2_NA_EVENT_CAUSE;
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
@@ -9088,65 +9213,83 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 	case GAUDI2_EVENT_CPU_FIX_THERMAL_ENV_S:
 	case GAUDI2_EVENT_CPU_FIX_THERMAL_ENV_E:
 		gaudi2_print_clk_change_info(hdev, event_type, &event_mask);
+		error_count = GAUDI2_NA_EVENT_CAUSE;
 		break;
 
 	case GAUDI2_EVENT_CPU_PKT_QUEUE_OUT_SYNC:
 		gaudi2_print_out_of_sync_info(hdev, &eq_entry->pkt_sync_err);
+		error_count = GAUDI2_NA_EVENT_CAUSE;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_PCIE_FLR_REQUESTED:
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
+		error_count = GAUDI2_NA_EVENT_CAUSE;
 		/* Do nothing- FW will handle it */
 		break;
 
 	case GAUDI2_EVENT_PCIE_P2P_MSIX:
-		gaudi2_handle_pcie_p2p_msix(hdev);
+		error_count = gaudi2_handle_pcie_p2p_msix(hdev);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
 	case GAUDI2_EVENT_SM0_AXI_ERROR_RESPONSE ... GAUDI2_EVENT_SM3_AXI_ERROR_RESPONSE:
 		index = event_type - GAUDI2_EVENT_SM0_AXI_ERROR_RESPONSE;
-		gaudi2_handle_sm_err(hdev, index);
+		error_count = gaudi2_handle_sm_err(hdev, index);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
 	case GAUDI2_EVENT_PSOC_MME_PLL_LOCK_ERR ... GAUDI2_EVENT_DCORE2_HBM_PLL_LOCK_ERR:
+		error_count = GAUDI2_NA_EVENT_CAUSE;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_CPU_CPLD_SHUTDOWN_CAUSE:
 		dev_info(hdev->dev, "CPLD shutdown cause, reset reason: 0x%llx\n",
 						le64_to_cpu(eq_entry->data[0]));
+		error_count = GAUDI2_NA_EVENT_CAUSE;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 	case GAUDI2_EVENT_CPU_CPLD_SHUTDOWN_EVENT:
 		dev_err(hdev->dev, "CPLD shutdown event, reset reason: 0x%llx\n",
 						le64_to_cpu(eq_entry->data[0]));
+		error_count = GAUDI2_NA_EVENT_CAUSE;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_CPU_PKT_SANITY_FAILED:
 		gaudi2_print_cpu_pkt_failure_info(hdev, &eq_entry->pkt_sync_err);
+		error_count = GAUDI2_NA_EVENT_CAUSE;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_ARC_DCCM_FULL:
-		hl_arc_event_handle(hdev, &eq_entry->arc_data);
+		error_count = hl_arc_event_handle(hdev, &eq_entry->arc_data);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
 	case GAUDI2_EVENT_CPU_FP32_NOT_SUPPORTED:
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
+		error_count = GAUDI2_NA_EVENT_CAUSE;
 		is_critical = true;
 		break;
 
 	default:
-		if (gaudi2_irq_map_table[event_type].valid)
+		if (gaudi2_irq_map_table[event_type].valid) {
 			dev_err_ratelimited(hdev->dev, "Cannot find handler for event %d\n",
 						event_type);
+			error_count = GAUDI2_NA_EVENT_CAUSE;
+		}
 	}
 
+	/* Make sure to dump an error in case no error cause was printed so far.
+	 * Note that although we have counted the errors, we use this number as
+	 * a boolean.
+	 */
+	if (error_count == 0 && !is_info_event(event_type))
+		dev_err_ratelimited(hdev->dev,
+			"No Error cause for H/W event %d\n", event_type);
+
 	if ((gaudi2_irq_map_table[event_type].reset || reset_required) &&
 				(hdev->hard_reset_on_fw_events ||
 				(hdev->asic_prop.fw_security_enabled && is_critical)))
-- 
2.25.1

[PATCH 12/26] habanalabs/gaudi2: remove duplicated event prints

[Oded Gabbay]

From: Ofir Bitton <obitton@habana.ai>

In order to reduce error log, we try to minimize the dumped rows
while keeping all relevant error info. In addition we completely
remove clock throttling debug logs.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/gaudi2/gaudi2.c | 339 +++++++++++-------------
 1 file changed, 149 insertions(+), 190 deletions(-)

diff --git a/drivers/misc/habanalabs/gaudi2/gaudi2.c b/drivers/misc/habanalabs/gaudi2/gaudi2.c
index b8da2aa024ca..8373239ad1bc 100644
--- a/drivers/misc/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/misc/habanalabs/gaudi2/gaudi2.c
@@ -6804,38 +6804,37 @@ static inline bool is_info_event(u32 event)
 	switch (event) {
 	case GAUDI2_EVENT_CPU_CPLD_SHUTDOWN_CAUSE:
 	case GAUDI2_EVENT_CPU_FIX_POWER_ENV_S ... GAUDI2_EVENT_CPU_FIX_THERMAL_ENV_E:
+
+	/* return in case of NIC status event - these events are received periodically and not as
+	 * an indication to an error.
+	 */
+	case GAUDI2_EVENT_CPU0_STATUS_NIC0_ENG0 ... GAUDI2_EVENT_CPU11_STATUS_NIC11_ENG1:
 		return true;
 	default:
 		return false;
 	}
 }
 
-static void gaudi2_print_irq_info(struct hl_device *hdev, u16 event_type)
+void gaudi2_print_event(struct hl_device *hdev, u16 event_type,
+			bool ratelimited, const char *fmt, ...)
 {
-	char desc[64] = "";
-	bool event_valid = false;
-
-	/* return in case of NIC status event - these events are received periodically and not as
-	 * an indication to an error, thus not printed.
-	 */
-	if (event_type >= GAUDI2_EVENT_CPU0_STATUS_NIC0_ENG0 &&
-		event_type <= GAUDI2_EVENT_CPU11_STATUS_NIC11_ENG1)
-		return;
+	struct va_format vaf;
+	va_list args;
 
-	if (gaudi2_irq_map_table[event_type].valid) {
-		snprintf(desc, sizeof(desc), gaudi2_irq_map_table[event_type].name);
-		event_valid = true;
-	}
-
-	if (!event_valid)
-		snprintf(desc, sizeof(desc), "N/A");
+	va_start(args, fmt);
+	vaf.fmt = fmt;
+	vaf.va = &args;
 
-	if (is_info_event(event_type))
-		dev_info_ratelimited(hdev->dev, "Received H/W interrupt %d [\"%s\"]\n",
-								event_type, desc);
+	if (ratelimited)
+		dev_err_ratelimited(hdev->dev, "%s: %pV\n",
+			gaudi2_irq_map_table[event_type].valid ?
+			gaudi2_irq_map_table[event_type].name : "N/A Event", &vaf);
 	else
-		dev_err_ratelimited(hdev->dev, "Received H/W interrupt %d [\"%s\"]\n",
-								event_type, desc);
+		dev_err(hdev->dev, "%s: %pV\n",
+			gaudi2_irq_map_table[event_type].valid ?
+			gaudi2_irq_map_table[event_type].name : "N/A Event", &vaf);
+
+	va_end(args);
 }
 
 static bool gaudi2_handle_ecc_event(struct hl_device *hdev, u16 event_type,
@@ -6848,7 +6847,7 @@ static bool gaudi2_handle_ecc_event(struct hl_device *hdev, u16 event_type,
 	ecc_syndrom = le64_to_cpu(ecc_data->ecc_syndrom);
 	memory_wrapper_idx = ecc_data->memory_wrapper_idx;
 
-	dev_err(hdev->dev,
+	gaudi2_print_event(hdev, event_type, !ecc_data->is_critical,
 		"ECC error detected. address: %#llx. Syndrom: %#llx. block id %u. critical %u.\n",
 		ecc_address, ecc_syndrom, memory_wrapper_idx, ecc_data->is_critical);
 
@@ -6988,7 +6987,7 @@ static void print_qman_data_on_err(struct hl_device *hdev, u32 qid_base, u32 str
 		gaudi2_print_last_pqes_on_err(hdev, qid_base, i, qman_base, false);
 }
 
-static int gaudi2_handle_qman_err_generic(struct hl_device *hdev, const char *qm_name,
+static int gaudi2_handle_qman_err_generic(struct hl_device *hdev, u16 event_type,
 							u64 qman_base, u32 qid_base)
 {
 	u32 i, j, glbl_sts_val, arb_err_val, num_error_causes, error_count = 0;
@@ -7015,11 +7014,11 @@ static int gaudi2_handle_qman_err_generic(struct hl_device *hdev, const char *qm
 
 		for (j = 0 ; j < num_error_causes ; j++)
 			if (glbl_sts_val & BIT(j)) {
-				dev_err_ratelimited(hdev->dev, "%s %s. err cause: %s\n",
-						qm_name, reg_desc,
-						i == QMAN_STREAMS ?
-						gaudi2_qman_lower_cp_error_cause[j] :
-						gaudi2_qman_error_cause[j]);
+				gaudi2_print_event(hdev, event_type, true,
+					"%s. err cause: %s", reg_desc,
+					i == QMAN_STREAMS ?
+					gaudi2_qman_lower_cp_error_cause[j] :
+					gaudi2_qman_error_cause[j]);
 				error_count++;
 			}
 
@@ -7033,8 +7032,9 @@ static int gaudi2_handle_qman_err_generic(struct hl_device *hdev, const char *qm
 
 	for (j = 0 ; j < GAUDI2_NUM_OF_QM_ARB_ERR_CAUSE ; j++) {
 		if (arb_err_val & BIT(j)) {
-			dev_err_ratelimited(hdev->dev, "%s ARB_ERR. err cause: %s\n",
-						qm_name, gaudi2_qman_arb_error_cause[j]);
+			gaudi2_print_event(hdev, event_type, true,
+				"ARB_ERR. err cause: %s",
+				gaudi2_qman_arb_error_cause[j]);
 			error_count++;
 		}
 	}
@@ -7761,7 +7761,7 @@ static int gaudi2_ack_psoc_razwi_event_handler(struct hl_device *hdev, u64 *even
 	return error_count;
 }
 
-static int _gaudi2_handle_qm_sei_err(struct hl_device *hdev, u64 qman_base)
+static int _gaudi2_handle_qm_sei_err(struct hl_device *hdev, u64 qman_base, u16 event_type)
 {
 	u32 i, sts_val, sts_clr_val = 0, error_count = 0;
 
@@ -7769,8 +7769,8 @@ static int _gaudi2_handle_qm_sei_err(struct hl_device *hdev, u64 qman_base)
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_QM_SEI_ERR_CAUSE ; i++) {
 		if (sts_val & BIT(i)) {
-			dev_err_ratelimited(hdev->dev, "QM SEI. err cause: %s\n",
-						gaudi2_qm_sei_error_cause[i]);
+			gaudi2_print_event(hdev, event_type, true,
+				"err cause: %s", gaudi2_qm_sei_error_cause[i]);
 			sts_clr_val |= BIT(i);
 			error_count++;
 		}
@@ -7827,13 +7827,13 @@ static int gaudi2_handle_qm_sei_err(struct hl_device *hdev, u16 event_type,
 		return 0;
 	}
 
-	error_count = _gaudi2_handle_qm_sei_err(hdev, qman_base);
+	error_count = _gaudi2_handle_qm_sei_err(hdev, qman_base, event_type);
 
 	/* There is a single event per NIC macro, so should check its both QMAN blocks */
 	if (event_type >= GAUDI2_EVENT_NIC0_AXI_ERROR_RESPONSE &&
 			event_type <= GAUDI2_EVENT_NIC11_AXI_ERROR_RESPONSE)
 		error_count += _gaudi2_handle_qm_sei_err(hdev,
-					qman_base + NIC_QM_OFFSET);
+					qman_base + NIC_QM_OFFSET, event_type);
 
 	/* check if RAZWI happened */
 	if (razwi_info)
@@ -7846,7 +7846,6 @@ static int gaudi2_handle_qman_err(struct hl_device *hdev, u16 event_type)
 {
 	u32 qid_base, error_count = 0;
 	u64 qman_base;
-	char desc[32];
 	u8 index;
 
 	switch (event_type) {
@@ -7854,125 +7853,104 @@ static int gaudi2_handle_qman_err(struct hl_device *hdev, u16 event_type)
 		index = event_type - GAUDI2_EVENT_TPC0_QM;
 		qid_base = GAUDI2_QUEUE_ID_DCORE0_TPC_0_0 + index * QMAN_STREAMS;
 		qman_base = mmDCORE0_TPC0_QM_BASE + index * DCORE_TPC_OFFSET;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE0_TPC%d_QM", index);
 		break;
 	case GAUDI2_EVENT_TPC6_QM ... GAUDI2_EVENT_TPC11_QM:
 		index = event_type - GAUDI2_EVENT_TPC6_QM;
 		qid_base = GAUDI2_QUEUE_ID_DCORE1_TPC_0_0 + index * QMAN_STREAMS;
 		qman_base = mmDCORE1_TPC0_QM_BASE + index * DCORE_TPC_OFFSET;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE1_TPC%d_QM", index);
 		break;
 	case GAUDI2_EVENT_TPC12_QM ... GAUDI2_EVENT_TPC17_QM:
 		index = event_type - GAUDI2_EVENT_TPC12_QM;
 		qid_base = GAUDI2_QUEUE_ID_DCORE2_TPC_0_0 + index * QMAN_STREAMS;
 		qman_base = mmDCORE2_TPC0_QM_BASE + index * DCORE_TPC_OFFSET;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE2_TPC%d_QM", index);
 		break;
 	case GAUDI2_EVENT_TPC18_QM ... GAUDI2_EVENT_TPC23_QM:
 		index = event_type - GAUDI2_EVENT_TPC18_QM;
 		qid_base = GAUDI2_QUEUE_ID_DCORE3_TPC_0_0 + index * QMAN_STREAMS;
 		qman_base = mmDCORE3_TPC0_QM_BASE + index * DCORE_TPC_OFFSET;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE3_TPC%d_QM", index);
 		break;
 	case GAUDI2_EVENT_TPC24_QM:
 		qid_base = GAUDI2_QUEUE_ID_DCORE0_TPC_6_0;
 		qman_base = mmDCORE0_TPC6_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE0_TPC6_QM");
 		break;
 	case GAUDI2_EVENT_MME0_QM:
 		qid_base = GAUDI2_QUEUE_ID_DCORE0_MME_0_0;
 		qman_base = mmDCORE0_MME_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE0_MME_QM");
 		break;
 	case GAUDI2_EVENT_MME1_QM:
 		qid_base = GAUDI2_QUEUE_ID_DCORE1_MME_0_0;
 		qman_base = mmDCORE1_MME_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE1_MME_QM");
 		break;
 	case GAUDI2_EVENT_MME2_QM:
 		qid_base = GAUDI2_QUEUE_ID_DCORE2_MME_0_0;
 		qman_base = mmDCORE2_MME_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE2_MME_QM");
 		break;
 	case GAUDI2_EVENT_MME3_QM:
 		qid_base = GAUDI2_QUEUE_ID_DCORE3_MME_0_0;
 		qman_base = mmDCORE3_MME_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE3_MME_QM");
 		break;
 	case GAUDI2_EVENT_HDMA0_QM:
 		qid_base = GAUDI2_QUEUE_ID_DCORE0_EDMA_0_0;
 		qman_base = mmDCORE0_EDMA0_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE0_EDMA0_QM");
 		break;
 	case GAUDI2_EVENT_HDMA1_QM:
 		qid_base = GAUDI2_QUEUE_ID_DCORE0_EDMA_1_0;
 		qman_base = mmDCORE0_EDMA1_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE0_EDMA1_QM");
 		break;
 	case GAUDI2_EVENT_HDMA2_QM:
 		qid_base = GAUDI2_QUEUE_ID_DCORE1_EDMA_0_0;
 		qman_base = mmDCORE1_EDMA0_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE1_EDMA0_QM");
 		break;
 	case GAUDI2_EVENT_HDMA3_QM:
 		qid_base = GAUDI2_QUEUE_ID_DCORE1_EDMA_1_0;
 		qman_base = mmDCORE1_EDMA1_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE1_EDMA1_QM");
 		break;
 	case GAUDI2_EVENT_HDMA4_QM:
 		qid_base = GAUDI2_QUEUE_ID_DCORE2_EDMA_0_0;
 		qman_base = mmDCORE2_EDMA0_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE2_EDMA0_QM");
 		break;
 	case GAUDI2_EVENT_HDMA5_QM:
 		qid_base = GAUDI2_QUEUE_ID_DCORE2_EDMA_1_0;
 		qman_base = mmDCORE2_EDMA1_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE2_EDMA1_QM");
 		break;
 	case GAUDI2_EVENT_HDMA6_QM:
 		qid_base = GAUDI2_QUEUE_ID_DCORE3_EDMA_0_0;
 		qman_base = mmDCORE3_EDMA0_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE3_EDMA0_QM");
 		break;
 	case GAUDI2_EVENT_HDMA7_QM:
 		qid_base = GAUDI2_QUEUE_ID_DCORE3_EDMA_1_0;
 		qman_base = mmDCORE3_EDMA1_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE3_EDMA1_QM");
 		break;
 	case GAUDI2_EVENT_PDMA0_QM:
 		qid_base = GAUDI2_QUEUE_ID_PDMA_0_0;
 		qman_base = mmPDMA0_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "PDMA0_QM");
 		break;
 	case GAUDI2_EVENT_PDMA1_QM:
 		qid_base = GAUDI2_QUEUE_ID_PDMA_1_0;
 		qman_base = mmPDMA1_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "PDMA1_QM");
 		break;
 	case GAUDI2_EVENT_ROTATOR0_ROT0_QM:
 		qid_base = GAUDI2_QUEUE_ID_ROT_0_0;
 		qman_base = mmROT0_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "ROTATOR0_QM");
 		break;
 	case GAUDI2_EVENT_ROTATOR1_ROT1_QM:
 		qid_base = GAUDI2_QUEUE_ID_ROT_1_0;
 		qman_base = mmROT1_QM_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "ROTATOR1_QM");
 		break;
 	default:
 		return 0;
 	}
 
-	error_count = gaudi2_handle_qman_err_generic(hdev, desc, qman_base, qid_base);
+	error_count = gaudi2_handle_qman_err_generic(hdev, event_type, qman_base, qid_base);
 
 	/* Handle EDMA QM SEI here because there is no AXI error response event for EDMA */
 	if (event_type >= GAUDI2_EVENT_HDMA2_QM && event_type <= GAUDI2_EVENT_HDMA5_QM)
-		error_count += _gaudi2_handle_qm_sei_err(hdev, qman_base);
+		error_count += _gaudi2_handle_qm_sei_err(hdev, qman_base, event_type);
 
 	return error_count;
 }
 
-static int gaudi2_handle_arc_farm_sei_err(struct hl_device *hdev)
+static int gaudi2_handle_arc_farm_sei_err(struct hl_device *hdev, u16 event_type)
 {
 	u32 i, sts_val, sts_clr_val = 0, error_count = 0;
 
@@ -7980,8 +7958,8 @@ static int gaudi2_handle_arc_farm_sei_err(struct hl_device *hdev)
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_ARC_SEI_ERR_CAUSE ; i++) {
 		if (sts_val & BIT(i)) {
-			dev_err_ratelimited(hdev->dev, "ARC SEI. err cause: %s\n",
-						gaudi2_arc_sei_error_cause[i]);
+			gaudi2_print_event(hdev, event_type, true,
+				"err cause: %s", gaudi2_arc_sei_error_cause[i]);
 			sts_clr_val |= BIT(i);
 			error_count++;
 		}
@@ -7992,7 +7970,7 @@ static int gaudi2_handle_arc_farm_sei_err(struct hl_device *hdev)
 	return error_count;
 }
 
-static int gaudi2_handle_cpu_sei_err(struct hl_device *hdev)
+static int gaudi2_handle_cpu_sei_err(struct hl_device *hdev, u16 event_type)
 {
 	u32 i, sts_val, sts_clr_val = 0, error_count = 0;
 
@@ -8000,8 +7978,8 @@ static int gaudi2_handle_cpu_sei_err(struct hl_device *hdev)
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_CPU_SEI_ERR_CAUSE ; i++) {
 		if (sts_val & BIT(i)) {
-			dev_err_ratelimited(hdev->dev, "CPU SEI. err cause: %s\n",
-						gaudi2_cpu_sei_error_cause[i]);
+			gaudi2_print_event(hdev, event_type, true,
+				"err cause: %s", gaudi2_cpu_sei_error_cause[i]);
 			sts_clr_val |= BIT(i);
 			error_count++;
 		}
@@ -8012,7 +7990,7 @@ static int gaudi2_handle_cpu_sei_err(struct hl_device *hdev)
 	return error_count;
 }
 
-static int gaudi2_handle_rot_err(struct hl_device *hdev, u8 rot_index,
+static int gaudi2_handle_rot_err(struct hl_device *hdev, u8 rot_index, u16 event_type,
 					struct hl_eq_razwi_with_intr_cause *razwi_with_intr_cause,
 					u64 *event_mask)
 {
@@ -8022,8 +8000,8 @@ static int gaudi2_handle_rot_err(struct hl_device *hdev, u8 rot_index,
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_ROT_ERR_CAUSE ; i++)
 		if (intr_cause_data & BIT(i)) {
-			dev_err_ratelimited(hdev->dev, "ROT%u. err cause: %s\n",
-						rot_index, guadi2_rot_error_cause[i]);
+			gaudi2_print_event(hdev, event_type, true,
+				"err cause: %s", guadi2_rot_error_cause[i]);
 			error_count++;
 		}
 
@@ -8034,7 +8012,7 @@ static int gaudi2_handle_rot_err(struct hl_device *hdev, u8 rot_index,
 	return error_count;
 }
 
-static int gaudi2_tpc_ack_interrupts(struct hl_device *hdev,  u8 tpc_index, char *interrupt_name,
+static int gaudi2_tpc_ack_interrupts(struct hl_device *hdev,  u8 tpc_index, u16 event_type,
 					struct hl_eq_razwi_with_intr_cause *razwi_with_intr_cause,
 					u64 *event_mask)
 {
@@ -8044,8 +8022,8 @@ static int gaudi2_tpc_ack_interrupts(struct hl_device *hdev,  u8 tpc_index, char
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_TPC_INTR_CAUSE ; i++)
 		if (intr_cause_data & BIT(i)) {
-			dev_err_ratelimited(hdev->dev, "TPC%d_%s interrupt cause: %s\n",
-					tpc_index, interrupt_name, gaudi2_tpc_interrupts_cause[i]);
+			gaudi2_print_event(hdev, event_type, true,
+				"interrupt cause: %s",  gaudi2_tpc_interrupts_cause[i]);
 			error_count++;
 		}
 
@@ -8056,7 +8034,7 @@ static int gaudi2_tpc_ack_interrupts(struct hl_device *hdev,  u8 tpc_index, char
 	return error_count;
 }
 
-static int gaudi2_handle_dec_err(struct hl_device *hdev, u8 dec_index, const char *interrupt_name,
+static int gaudi2_handle_dec_err(struct hl_device *hdev, u8 dec_index, u16 event_type,
 				struct hl_eq_razwi_info *razwi_info, u64 *event_mask)
 {
 	u32 sts_addr, sts_val, sts_clr_val = 0, error_count = 0;
@@ -8076,8 +8054,8 @@ static int gaudi2_handle_dec_err(struct hl_device *hdev, u8 dec_index, const cha
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_DEC_ERR_CAUSE ; i++) {
 		if (sts_val & BIT(i)) {
-			dev_err_ratelimited(hdev->dev, "DEC%u_%s err cause: %s\n",
-					dec_index, interrupt_name, gaudi2_dec_error_cause[i]);
+			gaudi2_print_event(hdev, event_type, true,
+				"err cause: %s", gaudi2_dec_error_cause[i]);
 			sts_clr_val |= BIT(i);
 			error_count++;
 		}
@@ -8093,7 +8071,7 @@ static int gaudi2_handle_dec_err(struct hl_device *hdev, u8 dec_index, const cha
 	return error_count;
 }
 
-static int gaudi2_handle_mme_err(struct hl_device *hdev, u8 mme_index, const char *interrupt_name,
+static int gaudi2_handle_mme_err(struct hl_device *hdev, u8 mme_index, u16 event_type,
 				struct hl_eq_razwi_info *razwi_info, u64 *event_mask)
 {
 	u32 sts_addr, sts_val, sts_clr_addr, sts_clr_val = 0, error_count = 0;
@@ -8106,8 +8084,8 @@ static int gaudi2_handle_mme_err(struct hl_device *hdev, u8 mme_index, const cha
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_MME_ERR_CAUSE ; i++) {
 		if (sts_val & BIT(i)) {
-			dev_err_ratelimited(hdev->dev, "MME%u_%s err cause: %s\n",
-					mme_index, interrupt_name, guadi2_mme_error_cause[i]);
+			gaudi2_print_event(hdev, event_type, true,
+				"err cause: %s", guadi2_mme_error_cause[i]);
 			sts_clr_val |= BIT(i);
 			error_count++;
 		}
@@ -8123,22 +8101,22 @@ static int gaudi2_handle_mme_err(struct hl_device *hdev, u8 mme_index, const cha
 	return error_count;
 }
 
-static int gaudi2_handle_mme_sbte_err(struct hl_device *hdev, u8 mme_index, u8 sbte_index,
+static int gaudi2_handle_mme_sbte_err(struct hl_device *hdev, u16 event_type,
 					u64 intr_cause_data)
 {
 	int i, error_count = 0;
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_MME_SBTE_ERR_CAUSE ; i++)
 		if (intr_cause_data & BIT(i)) {
-			dev_err_ratelimited(hdev->dev, "MME%uSBTE%u_AXI_ERR_RSP err cause: %s\n",
-					mme_index, sbte_index, guadi2_mme_sbte_error_cause[i]);
+			gaudi2_print_event(hdev, event_type, true,
+				"err cause: %s", guadi2_mme_sbte_error_cause[i]);
 			error_count++;
 		}
 
 	return error_count;
 }
 
-static int gaudi2_handle_mme_wap_err(struct hl_device *hdev, u8 mme_index,
+static int gaudi2_handle_mme_wap_err(struct hl_device *hdev, u8 mme_index, u16 event_type,
 					struct hl_eq_razwi_info *razwi_info, u64 *event_mask)
 {
 	u32 sts_addr, sts_val, sts_clr_addr, sts_clr_val = 0, error_count = 0;
@@ -8151,9 +8129,8 @@ static int gaudi2_handle_mme_wap_err(struct hl_device *hdev, u8 mme_index,
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_MME_WAP_ERR_CAUSE ; i++) {
 		if (sts_val & BIT(i)) {
-			dev_err_ratelimited(hdev->dev,
-					"MME%u_WAP_SOURCE_RESULT_INVALID err cause: %s\n",
-					mme_index, guadi2_mme_wap_error_cause[i]);
+			gaudi2_print_event(hdev, event_type, true,
+				"err cause: %s", guadi2_mme_wap_error_cause[i]);
 			sts_clr_val |= BIT(i);
 			error_count++;
 		}
@@ -8170,7 +8147,8 @@ static int gaudi2_handle_mme_wap_err(struct hl_device *hdev, u8 mme_index,
 	return error_count;
 }
 
-static int gaudi2_handle_kdma_core_event(struct hl_device *hdev, u64 intr_cause_data)
+static int gaudi2_handle_kdma_core_event(struct hl_device *hdev, u16 event_type,
+					u64 intr_cause_data)
 {
 	u32 error_count = 0;
 	int i;
@@ -8182,23 +8160,24 @@ static int gaudi2_handle_kdma_core_event(struct hl_device *hdev, u64 intr_cause_
 	 */
 	for (i = 0 ; i < GAUDI2_NUM_OF_DMA_CORE_INTR_CAUSE ; i++)
 		if (intr_cause_data & BIT(i)) {
-			dev_err_ratelimited(hdev->dev, "kdma core err cause: %s\n",
-						gaudi2_kdma_core_interrupts_cause[i]);
+			gaudi2_print_event(hdev, event_type, true,
+				"err cause: %s", gaudi2_kdma_core_interrupts_cause[i]);
 			error_count++;
 		}
 
 	return error_count;
 }
 
-static int gaudi2_handle_dma_core_event(struct hl_device *hdev, u64 intr_cause_data)
+static int gaudi2_handle_dma_core_event(struct hl_device *hdev, u16 event_type,
+					u64 intr_cause_data)
 {
 	u32 error_count = 0;
 	int i;
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_DMA_CORE_INTR_CAUSE ; i++)
 		if (intr_cause_data & BIT(i)) {
-			dev_err_ratelimited(hdev->dev, "dma core err cause: %s\n",
-						gaudi2_dma_core_interrupts_cause[i]);
+			gaudi2_print_event(hdev, event_type, true,
+				"err cause: %s", gaudi2_dma_core_interrupts_cause[i]);
 			error_count++;
 		}
 
@@ -8238,8 +8217,8 @@ static void gaudi2_print_pcie_mstr_rr_mstr_if_razwi_info(struct hl_device *hdev,
 	}
 }
 
-static int gaudi2_print_pcie_addr_dec_info(struct hl_device *hdev, u64 intr_cause_data,
-						u64 *event_mask)
+static int gaudi2_print_pcie_addr_dec_info(struct hl_device *hdev, u16 event_type,
+					u64 intr_cause_data, u64 *event_mask)
 {
 	u32 error_count = 0;
 	int i;
@@ -8248,8 +8227,8 @@ static int gaudi2_print_pcie_addr_dec_info(struct hl_device *hdev, u64 intr_caus
 		if (!(intr_cause_data & BIT_ULL(i)))
 			continue;
 
-		dev_err_ratelimited(hdev->dev, "PCIE ADDR DEC Error: %s\n",
-					gaudi2_pcie_addr_dec_error_cause[i]);
+		gaudi2_print_event(hdev, event_type, true,
+			"err cause: %s", gaudi2_pcie_addr_dec_error_cause[i]);
 		error_count++;
 
 		switch (intr_cause_data & BIT_ULL(i)) {
@@ -8264,7 +8243,8 @@ static int gaudi2_print_pcie_addr_dec_info(struct hl_device *hdev, u64 intr_caus
 	return error_count;
 }
 
-static int gaudi2_handle_pif_fatal(struct hl_device *hdev, u64 intr_cause_data)
+static int gaudi2_handle_pif_fatal(struct hl_device *hdev, u16 event_type,
+				u64 intr_cause_data)
 
 {
 	u32 error_count = 0;
@@ -8272,8 +8252,8 @@ static int gaudi2_handle_pif_fatal(struct hl_device *hdev, u64 intr_cause_data)
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_PMMU_FATAL_ERR_CAUSE ; i++) {
 		if (intr_cause_data & BIT_ULL(i)) {
-			dev_err_ratelimited(hdev->dev, "PMMU PIF err cause: %s\n",
-					gaudi2_pmmu_fatal_interrupts_cause[i]);
+			gaudi2_print_event(hdev, event_type, true,
+				"err cause: %s", gaudi2_pmmu_fatal_interrupts_cause[i]);
 			error_count++;
 		}
 	}
@@ -8283,16 +8263,13 @@ static int gaudi2_handle_pif_fatal(struct hl_device *hdev, u64 intr_cause_data)
 
 static int gaudi2_handle_hif_fatal(struct hl_device *hdev, u16 event_type, u64 intr_cause_data)
 {
-	u32 dcore_id, hif_id, error_count = 0;
+	u32 error_count = 0;
 	int i;
 
-	dcore_id = (event_type - GAUDI2_EVENT_HIF0_FATAL) / 4;
-	hif_id = (event_type - GAUDI2_EVENT_HIF0_FATAL) % 4;
-
 	for (i = 0 ; i < GAUDI2_NUM_OF_HIF_FATAL_ERR_CAUSE ; i++) {
 		if (intr_cause_data & BIT_ULL(i)) {
-			dev_err_ratelimited(hdev->dev, "DCORE%u_HIF%u: %s\n", dcore_id, hif_id,
-					gaudi2_hif_fatal_interrupts_cause[i]);
+			gaudi2_print_event(hdev, event_type, true,
+				"err cause: %s", gaudi2_hif_fatal_interrupts_cause[i]);
 			error_count++;
 		}
 	}
@@ -8343,7 +8320,7 @@ static void gaudi2_handle_access_error(struct hl_device *hdev, u64 mmu_base, boo
 	WREG32(mmu_base + MMU_OFFSET(mmDCORE0_HMMU0_MMU_ACCESS_ERROR_CAPTURE), 0);
 }
 
-static int gaudi2_handle_mmu_spi_sei_generic(struct hl_device *hdev, const char *mmu_name,
+static int gaudi2_handle_mmu_spi_sei_generic(struct hl_device *hdev, u16 event_type,
 						u64 mmu_base, bool is_pmmu, u64 *event_mask)
 {
 	u32 spi_sei_cause, interrupt_clr = 0x0, error_count = 0;
@@ -8353,8 +8330,8 @@ static int gaudi2_handle_mmu_spi_sei_generic(struct hl_device *hdev, const char
 
 	for (i = 0 ; i < GAUDI2_NUM_OF_MMU_SPI_SEI_CAUSE ; i++) {
 		if (spi_sei_cause & BIT(i)) {
-			dev_err_ratelimited(hdev->dev, "%s SPI_SEI ERR. err cause: %s\n",
-						mmu_name, gaudi2_mmu_spi_sei[i].cause);
+			gaudi2_print_event(hdev, event_type, true,
+				"err cause: %s", gaudi2_mmu_spi_sei[i].cause);
 
 			if (i == 0)
 				gaudi2_handle_page_error(hdev, mmu_base, is_pmmu, event_mask);
@@ -8377,7 +8354,7 @@ static int gaudi2_handle_mmu_spi_sei_generic(struct hl_device *hdev, const char
 	return error_count;
 }
 
-static int gaudi2_handle_sm_err(struct hl_device *hdev, u8 sm_index)
+static int gaudi2_handle_sm_err(struct hl_device *hdev, u16 event_type, u8 sm_index)
 {
 	u32 sei_cause_addr, sei_cause_val, sei_cause_cause, sei_cause_log,
 		cq_intr_addr, cq_intr_val, cq_intr_queue_index, error_count = 0;
@@ -8400,11 +8377,11 @@ static int gaudi2_handle_sm_err(struct hl_device *hdev, u8 sm_index)
 			if (!(sei_cause_cause & BIT(i)))
 				continue;
 
-			dev_err_ratelimited(hdev->dev, "SM%u SEI ERR. err cause: %s. %s: 0x%X\n",
-					sm_index,
-					gaudi2_sm_sei_cause[i].cause_name,
-					gaudi2_sm_sei_cause[i].log_name,
-					sei_cause_log & gaudi2_sm_sei_cause[i].log_mask);
+			gaudi2_print_event(hdev, event_type, true,
+				"err cause: %s. %s: 0x%X\n",
+				gaudi2_sm_sei_cause[i].cause_name,
+				gaudi2_sm_sei_cause[i].log_name,
+				sei_cause_log & gaudi2_sm_sei_cause[i].log_mask);
 			error_count++;
 			break;
 		}
@@ -8433,7 +8410,6 @@ static int gaudi2_handle_sm_err(struct hl_device *hdev, u8 sm_index)
 static int gaudi2_handle_mmu_spi_sei_err(struct hl_device *hdev, u16 event_type, u64 *event_mask)
 {
 	bool is_pmmu = false;
-	char desc[32];
 	u32 error_count = 0;
 	u64 mmu_base;
 	u8 index;
@@ -8442,54 +8418,46 @@ static int gaudi2_handle_mmu_spi_sei_err(struct hl_device *hdev, u16 event_type,
 	case GAUDI2_EVENT_HMMU0_PAGE_FAULT_OR_WR_PERM ... GAUDI2_EVENT_HMMU3_SECURITY_ERROR:
 		index = (event_type - GAUDI2_EVENT_HMMU0_PAGE_FAULT_OR_WR_PERM) / 3;
 		mmu_base = mmDCORE0_HMMU0_MMU_BASE + index * DCORE_HMMU_OFFSET;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE0_HMMU%d", index);
 		break;
 	case GAUDI2_EVENT_HMMU_0_AXI_ERR_RSP ... GAUDI2_EVENT_HMMU_3_AXI_ERR_RSP:
 		index = (event_type - GAUDI2_EVENT_HMMU_0_AXI_ERR_RSP);
 		mmu_base = mmDCORE0_HMMU0_MMU_BASE + index * DCORE_HMMU_OFFSET;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE0_HMMU%d", index);
 		break;
 	case GAUDI2_EVENT_HMMU8_PAGE_FAULT_WR_PERM ... GAUDI2_EVENT_HMMU11_SECURITY_ERROR:
 		index = (event_type - GAUDI2_EVENT_HMMU8_PAGE_FAULT_WR_PERM) / 3;
 		mmu_base = mmDCORE1_HMMU0_MMU_BASE + index * DCORE_HMMU_OFFSET;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE1_HMMU%d", index);
 		break;
 	case GAUDI2_EVENT_HMMU_8_AXI_ERR_RSP ... GAUDI2_EVENT_HMMU_11_AXI_ERR_RSP:
 		index = (event_type - GAUDI2_EVENT_HMMU_8_AXI_ERR_RSP);
 		mmu_base = mmDCORE1_HMMU0_MMU_BASE + index * DCORE_HMMU_OFFSET;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE1_HMMU%d", index);
 		break;
 	case GAUDI2_EVENT_HMMU7_PAGE_FAULT_WR_PERM ... GAUDI2_EVENT_HMMU4_SECURITY_ERROR:
 		index = (event_type - GAUDI2_EVENT_HMMU7_PAGE_FAULT_WR_PERM) / 3;
 		mmu_base = mmDCORE2_HMMU0_MMU_BASE + index * DCORE_HMMU_OFFSET;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE2_HMMU%d", index);
 		break;
 	case GAUDI2_EVENT_HMMU_7_AXI_ERR_RSP ... GAUDI2_EVENT_HMMU_4_AXI_ERR_RSP:
 		index = (event_type - GAUDI2_EVENT_HMMU_7_AXI_ERR_RSP);
 		mmu_base = mmDCORE2_HMMU0_MMU_BASE + index * DCORE_HMMU_OFFSET;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE2_HMMU%d", index);
 		break;
 	case GAUDI2_EVENT_HMMU15_PAGE_FAULT_WR_PERM ... GAUDI2_EVENT_HMMU12_SECURITY_ERROR:
 		index = (event_type - GAUDI2_EVENT_HMMU15_PAGE_FAULT_WR_PERM) / 3;
 		mmu_base = mmDCORE3_HMMU0_MMU_BASE + index * DCORE_HMMU_OFFSET;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE3_HMMU%d", index);
 		break;
 	case GAUDI2_EVENT_HMMU_15_AXI_ERR_RSP ... GAUDI2_EVENT_HMMU_12_AXI_ERR_RSP:
 		index = (event_type - GAUDI2_EVENT_HMMU_15_AXI_ERR_RSP);
 		mmu_base = mmDCORE3_HMMU0_MMU_BASE + index * DCORE_HMMU_OFFSET;
-		snprintf(desc, ARRAY_SIZE(desc), "DCORE3_HMMU%d", index);
 		break;
 	case GAUDI2_EVENT_PMMU0_PAGE_FAULT_WR_PERM ... GAUDI2_EVENT_PMMU0_SECURITY_ERROR:
 	case GAUDI2_EVENT_PMMU_AXI_ERR_RSP_0:
 		is_pmmu = true;
 		mmu_base = mmPMMU_HBW_MMU_BASE;
-		snprintf(desc, ARRAY_SIZE(desc), "PMMU");
 		break;
 	default:
 		return 0;
 	}
 
-	error_count = gaudi2_handle_mmu_spi_sei_generic(hdev, desc, mmu_base, is_pmmu, event_mask);
+	error_count = gaudi2_handle_mmu_spi_sei_generic(hdev, event_type, mmu_base,
+							is_pmmu, event_mask);
 
 	return error_count;
 }
@@ -8611,22 +8579,17 @@ static bool gaudi2_handle_hbm_mc_sei_err(struct hl_device *hdev, u16 event_type,
 
 	cause_idx = sei_data->hdr.sei_cause;
 	if (cause_idx > GAUDI2_NUM_OF_HBM_SEI_CAUSE - 1) {
-		dev_err_ratelimited(hdev->dev, "Invalid HBM SEI event cause (%d) provided by FW\n",
-					cause_idx);
+		gaudi2_print_event(hdev, event_type, true,
+			"err cause: %s",
+			"Invalid HBM SEI event cause (%d) provided by FW\n", cause_idx);
 		return true;
 	}
 
-	if (sei_data->hdr.is_critical)
-		dev_err(hdev->dev,
-			"System Critical Error Interrupt - HBM(%u) MC(%u) MC_CH(%u) MC_PC(%u). Error cause: %s\n",
-			hbm_id, mc_id, sei_data->hdr.mc_channel, sei_data->hdr.mc_pseudo_channel,
-			hbm_mc_sei_cause[cause_idx]);
-
-	else
-		dev_err_ratelimited(hdev->dev,
-			"System Non-Critical Error Interrupt - HBM(%u) MC(%u) MC_CH(%u) MC_PC(%u). Error cause: %s\n",
-			hbm_id, mc_id, sei_data->hdr.mc_channel, sei_data->hdr.mc_pseudo_channel,
-			hbm_mc_sei_cause[cause_idx]);
+	gaudi2_print_event(hdev, event_type, !sei_data->hdr.is_critical,
+		"System %s Error Interrupt - HBM(%u) MC(%u) MC_CH(%u) MC_PC(%u). Error cause: %s\n",
+		sei_data->hdr.is_critical ? "Critical" : "Non-critical",
+		hbm_id, mc_id, sei_data->hdr.mc_channel, sei_data->hdr.mc_pseudo_channel,
+		hbm_mc_sei_cause[cause_idx]);
 
 	/* Print error-specific info */
 	switch (cause_idx) {
@@ -8670,12 +8633,12 @@ static bool gaudi2_handle_hbm_mc_sei_err(struct hl_device *hdev, u16 event_type,
 	return require_hard_reset;
 }
 
-static int gaudi2_handle_hbm_cattrip(struct hl_device *hdev, u64 intr_cause_data)
+static int gaudi2_handle_hbm_cattrip(struct hl_device *hdev, u16 event_type,
+				u64 intr_cause_data)
 {
 	if (intr_cause_data) {
-		dev_err(hdev->dev,
-			"HBM catastrophic temperature error (CATTRIP) cause %#llx\n",
-			intr_cause_data);
+		gaudi2_print_event(hdev, event_type, true,
+			"temperature error cause: %#llx", intr_cause_data);
 		return 1;
 	}
 
@@ -8708,13 +8671,13 @@ static void gaudi2_print_clk_change_info(struct hl_device *hdev, u16 event_type,
 		hdev->clk_throttling.aggregated_reason |= HL_CLK_THROTTLE_POWER;
 		hdev->clk_throttling.timestamp[HL_CLK_THROTTLE_TYPE_POWER].start = ktime_get();
 		hdev->clk_throttling.timestamp[HL_CLK_THROTTLE_TYPE_POWER].end = zero_time;
-		dev_info_ratelimited(hdev->dev, "Clock throttling due to power consumption\n");
+		dev_dbg_ratelimited(hdev->dev, "Clock throttling due to power consumption\n");
 		break;
 
 	case GAUDI2_EVENT_CPU_FIX_POWER_ENV_E:
 		hdev->clk_throttling.current_reason &= ~HL_CLK_THROTTLE_POWER;
 		hdev->clk_throttling.timestamp[HL_CLK_THROTTLE_TYPE_POWER].end = ktime_get();
-		dev_info_ratelimited(hdev->dev, "Power envelop is safe, back to optimal clock\n");
+		dev_dbg_ratelimited(hdev->dev, "Power envelop is safe, back to optimal clock\n");
 		break;
 
 	case GAUDI2_EVENT_CPU_FIX_THERMAL_ENV_S:
@@ -8741,16 +8704,18 @@ static void gaudi2_print_clk_change_info(struct hl_device *hdev, u16 event_type,
 	mutex_unlock(&hdev->clk_throttling.lock);
 }
 
-static void gaudi2_print_out_of_sync_info(struct hl_device *hdev,
+static void gaudi2_print_out_of_sync_info(struct hl_device *hdev, u16 event_type,
 					struct cpucp_pkt_sync_err *sync_err)
 {
 	struct hl_hw_queue *q = &hdev->kernel_queues[GAUDI2_QUEUE_ID_CPU_PQ];
 
-	dev_err(hdev->dev, "Out of sync with FW, FW: pi=%u, ci=%u, LKD: pi=%u, ci=%d\n",
-		le32_to_cpu(sync_err->pi), le32_to_cpu(sync_err->ci), q->pi, atomic_read(&q->ci));
+	gaudi2_print_event(hdev, event_type, false,
+		"FW: pi=%u, ci=%u, LKD: pi=%u, ci=%d\n",
+		le32_to_cpu(sync_err->pi), le32_to_cpu(sync_err->ci),
+		q->pi, atomic_read(&q->ci));
 }
 
-static int gaudi2_handle_pcie_p2p_msix(struct hl_device *hdev)
+static int gaudi2_handle_pcie_p2p_msix(struct hl_device *hdev, u16 event_type)
 {
 	u32 p2p_intr, msix_gw_intr, error_count = 0;
 
@@ -8758,7 +8723,7 @@ static int gaudi2_handle_pcie_p2p_msix(struct hl_device *hdev)
 	msix_gw_intr = RREG32(mmPCIE_WRAP_MSIX_GW_INTR);
 
 	if (p2p_intr) {
-		dev_err_ratelimited(hdev->dev,
+		gaudi2_print_event(hdev, event_type, true,
 			"pcie p2p transaction terminated due to security, req_id(0x%x)\n",
 			RREG32(mmPCIE_WRAP_P2P_REQ_ID));
 
@@ -8767,7 +8732,7 @@ static int gaudi2_handle_pcie_p2p_msix(struct hl_device *hdev)
 	}
 
 	if (msix_gw_intr) {
-		dev_err_ratelimited(hdev->dev,
+		gaudi2_print_event(hdev, event_type, true,
 			"pcie msi-x gen denied due to vector num check failure, vec(0x%X)\n",
 			RREG32(mmPCIE_WRAP_MSIX_GW_VEC));
 
@@ -8822,17 +8787,18 @@ static int gaudi2_handle_psoc_drain(struct hl_device *hdev, u64 intr_cause_data)
 	return error_count;
 }
 
-static void gaudi2_print_cpu_pkt_failure_info(struct hl_device *hdev,
+static void gaudi2_print_cpu_pkt_failure_info(struct hl_device *hdev, u16 event_type,
 					struct cpucp_pkt_sync_err *sync_err)
 {
 	struct hl_hw_queue *q = &hdev->kernel_queues[GAUDI2_QUEUE_ID_CPU_PQ];
 
-	dev_warn(hdev->dev,
+	gaudi2_print_event(hdev, event_type, false,
 		"FW reported sanity check failure, FW: pi=%u, ci=%u, LKD: pi=%u, ci=%d\n",
 		le32_to_cpu(sync_err->pi), le32_to_cpu(sync_err->ci), q->pi, atomic_read(&q->ci));
 }
 
-static int hl_arc_event_handle(struct hl_device *hdev, struct hl_eq_engine_arc_intr_data *data)
+static int hl_arc_event_handle(struct hl_device *hdev, u16 event_type,
+					struct hl_eq_engine_arc_intr_data *data)
 {
 	struct hl_engine_arc_dccm_queue_full_irq *q;
 	u32 intr_type, engine_id;
@@ -8846,12 +8812,12 @@ static int hl_arc_event_handle(struct hl_device *hdev, struct hl_eq_engine_arc_i
 	case ENGINE_ARC_DCCM_QUEUE_FULL_IRQ:
 		q = (struct hl_engine_arc_dccm_queue_full_irq *) &payload;
 
-		dev_err_ratelimited(hdev->dev,
+		gaudi2_print_event(hdev, event_type, true,
 				"ARC DCCM Full event: EngId: %u, Intr_type: %u, Qidx: %u\n",
 				engine_id, intr_type, q->queue_index);
 		return 1;
 	default:
-		dev_err_ratelimited(hdev->dev, "Unknown ARC event type\n");
+		gaudi2_print_event(hdev, event_type, true, "Unknown ARC event type\n");
 		return 0;
 	}
 }
@@ -8860,8 +8826,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 {
 	struct gaudi2_device *gaudi2 = hdev->asic_specific;
 	bool reset_required = false, is_critical = false;
-	u32 ctl, reset_flags = HL_DRV_RESET_HARD, error_count = 0;
-	int index, sbte_index;
+	u32 index, ctl, reset_flags = HL_DRV_RESET_HARD, error_count = 0;
 	u64 event_mask = 0;
 	u16 event_type;
 
@@ -8877,8 +8842,6 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 	gaudi2->events_stat[event_type]++;
 	gaudi2->events_stat_aggregate[event_type]++;
 
-	gaudi2_print_irq_info(hdev, event_type);
-
 	switch (event_type) {
 	case GAUDI2_EVENT_PCIE_CORE_SERR ... GAUDI2_EVENT_ARC0_ECC_DERR:
 		fallthrough;
@@ -8901,12 +8864,12 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 
 	case GAUDI2_EVENT_ARC_AXI_ERROR_RESPONSE_0:
 		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
-		error_count = gaudi2_handle_arc_farm_sei_err(hdev);
+		error_count = gaudi2_handle_arc_farm_sei_err(hdev, event_type);
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_CPU_AXI_ERR_RSP:
-		error_count = gaudi2_handle_cpu_sei_err(hdev);
+		error_count = gaudi2_handle_cpu_sei_err(hdev, event_type);
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
@@ -8921,7 +8884,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 	case GAUDI2_EVENT_ROTATOR0_AXI_ERROR_RESPONSE:
 	case GAUDI2_EVENT_ROTATOR1_AXI_ERROR_RESPONSE:
 		index = event_type - GAUDI2_EVENT_ROTATOR0_AXI_ERROR_RESPONSE;
-		error_count = gaudi2_handle_rot_err(hdev, index,
+		error_count = gaudi2_handle_rot_err(hdev, index, event_type,
 					&eq_entry->razwi_with_intr_cause, &event_mask);
 		error_count += gaudi2_handle_qm_sei_err(hdev, event_type, NULL, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
@@ -8929,7 +8892,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 
 	case GAUDI2_EVENT_TPC0_AXI_ERR_RSP ... GAUDI2_EVENT_TPC24_AXI_ERR_RSP:
 		index = event_type - GAUDI2_EVENT_TPC0_AXI_ERR_RSP;
-		error_count = gaudi2_tpc_ack_interrupts(hdev, index, "AXI_ERR_RSP",
+		error_count = gaudi2_tpc_ack_interrupts(hdev, index, event_type,
 						&eq_entry->razwi_with_intr_cause, &event_mask);
 		error_count += gaudi2_handle_qm_sei_err(hdev, event_type, NULL, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
@@ -8937,8 +8900,8 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 
 	case GAUDI2_EVENT_DEC0_AXI_ERR_RSPONSE ... GAUDI2_EVENT_DEC9_AXI_ERR_RSPONSE:
 		index = event_type - GAUDI2_EVENT_DEC0_AXI_ERR_RSPONSE;
-		error_count = gaudi2_handle_dec_err(hdev, index, "AXI_ERR_RESPONSE",
-					&eq_entry->razwi_info, &event_mask);
+		error_count = gaudi2_handle_dec_err(hdev, index, event_type,
+						&eq_entry->razwi_info, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
@@ -8969,8 +8932,8 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 	case GAUDI2_EVENT_TPC24_KERNEL_ERR:
 		index = (event_type - GAUDI2_EVENT_TPC0_KERNEL_ERR) /
 			(GAUDI2_EVENT_TPC1_KERNEL_ERR - GAUDI2_EVENT_TPC0_KERNEL_ERR);
-		error_count = gaudi2_tpc_ack_interrupts(hdev, index, "KRN_ERR",
-				&eq_entry->razwi_with_intr_cause, &event_mask);
+		error_count = gaudi2_tpc_ack_interrupts(hdev, index, event_type,
+					&eq_entry->razwi_with_intr_cause, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
@@ -8986,7 +8949,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 	case GAUDI2_EVENT_DEC9_SPI:
 		index = (event_type - GAUDI2_EVENT_DEC0_SPI) /
 				(GAUDI2_EVENT_DEC1_SPI - GAUDI2_EVENT_DEC0_SPI);
-		error_count = gaudi2_handle_dec_err(hdev, index, "SPI",
+		error_count = gaudi2_handle_dec_err(hdev, index, event_type,
 					&eq_entry->razwi_info, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
@@ -8998,8 +8961,8 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		index = (event_type - GAUDI2_EVENT_MME0_CTRL_AXI_ERROR_RESPONSE) /
 				(GAUDI2_EVENT_MME1_CTRL_AXI_ERROR_RESPONSE -
 						GAUDI2_EVENT_MME0_CTRL_AXI_ERROR_RESPONSE);
-		error_count = gaudi2_handle_mme_err(hdev, index,
-				"CTRL_AXI_ERROR_RESPONSE", &eq_entry->razwi_info, &event_mask);
+		error_count = gaudi2_handle_mme_err(hdev, index, event_type,
+				&eq_entry->razwi_info, &event_mask);
 		error_count += gaudi2_handle_qm_sei_err(hdev, event_type, NULL, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
@@ -9011,7 +8974,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		index = (event_type - GAUDI2_EVENT_MME0_QMAN_SW_ERROR) /
 				(GAUDI2_EVENT_MME1_QMAN_SW_ERROR -
 					GAUDI2_EVENT_MME0_QMAN_SW_ERROR);
-		error_count = gaudi2_handle_mme_err(hdev, index, "QMAN_SW_ERROR",
+		error_count = gaudi2_handle_mme_err(hdev, index, event_type,
 					&eq_entry->razwi_info, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
@@ -9023,26 +8986,26 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		index = (event_type - GAUDI2_EVENT_MME0_WAP_SOURCE_RESULT_INVALID) /
 				(GAUDI2_EVENT_MME1_WAP_SOURCE_RESULT_INVALID -
 					GAUDI2_EVENT_MME0_WAP_SOURCE_RESULT_INVALID);
-		error_count = gaudi2_handle_mme_wap_err(hdev, index,
+		error_count = gaudi2_handle_mme_wap_err(hdev, index, event_type,
 					&eq_entry->razwi_info, &event_mask);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
 	case GAUDI2_EVENT_KDMA_CH0_AXI_ERR_RSP:
 	case GAUDI2_EVENT_KDMA0_CORE:
-		error_count = gaudi2_handle_kdma_core_event(hdev,
+		error_count = gaudi2_handle_kdma_core_event(hdev, event_type,
 					le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_HDMA2_CORE ... GAUDI2_EVENT_PDMA1_CORE:
-		error_count = gaudi2_handle_dma_core_event(hdev,
+		error_count = gaudi2_handle_dma_core_event(hdev, event_type,
 					le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
 	case GAUDI2_EVENT_PCIE_ADDR_DEC_ERR:
-		error_count = gaudi2_print_pcie_addr_dec_info(hdev,
+		error_count = gaudi2_print_pcie_addr_dec_info(hdev, event_type,
 				le64_to_cpu(eq_entry->intr_cause.intr_cause_data), &event_mask);
 		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
@@ -9065,7 +9028,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		break;
 
 	case GAUDI2_EVENT_PMMU_FATAL_0:
-		error_count = gaudi2_handle_pif_fatal(hdev,
+		error_count = gaudi2_handle_pif_fatal(hdev, event_type,
 				le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
 		reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
@@ -9086,7 +9049,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		break;
 
 	case GAUDI2_EVENT_HBM_CATTRIP_0 ... GAUDI2_EVENT_HBM_CATTRIP_5:
-		error_count = gaudi2_handle_hbm_cattrip(hdev,
+		error_count = gaudi2_handle_hbm_cattrip(hdev, event_type,
 				le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
@@ -9122,14 +9085,8 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 	case GAUDI2_EVENT_MME1_SBTE0_AXI_ERR_RSP ... GAUDI2_EVENT_MME1_SBTE4_AXI_ERR_RSP:
 	case GAUDI2_EVENT_MME2_SBTE0_AXI_ERR_RSP ... GAUDI2_EVENT_MME2_SBTE4_AXI_ERR_RSP:
 	case GAUDI2_EVENT_MME3_SBTE0_AXI_ERR_RSP ... GAUDI2_EVENT_MME3_SBTE4_AXI_ERR_RSP:
-		index = (event_type - GAUDI2_EVENT_MME0_SBTE0_AXI_ERR_RSP) /
-				(GAUDI2_EVENT_MME1_SBTE0_AXI_ERR_RSP -
-					GAUDI2_EVENT_MME0_SBTE0_AXI_ERR_RSP);
-		sbte_index = (event_type - GAUDI2_EVENT_MME0_SBTE0_AXI_ERR_RSP) %
-				(GAUDI2_EVENT_MME1_SBTE0_AXI_ERR_RSP -
-					GAUDI2_EVENT_MME0_SBTE0_AXI_ERR_RSP);
-		error_count = gaudi2_handle_mme_sbte_err(hdev, index, sbte_index,
-					le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
+		error_count = gaudi2_handle_mme_sbte_err(hdev, event_type,
+						le64_to_cpu(eq_entry->intr_cause.intr_cause_data));
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 	case GAUDI2_EVENT_VM0_ALARM_A ... GAUDI2_EVENT_VM3_ALARM_B:
@@ -9217,7 +9174,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		break;
 
 	case GAUDI2_EVENT_CPU_PKT_QUEUE_OUT_SYNC:
-		gaudi2_print_out_of_sync_info(hdev, &eq_entry->pkt_sync_err);
+		gaudi2_print_out_of_sync_info(hdev, event_type, &eq_entry->pkt_sync_err);
 		error_count = GAUDI2_NA_EVENT_CAUSE;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
@@ -9229,13 +9186,13 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		break;
 
 	case GAUDI2_EVENT_PCIE_P2P_MSIX:
-		error_count = gaudi2_handle_pcie_p2p_msix(hdev);
+		error_count = gaudi2_handle_pcie_p2p_msix(hdev, event_type);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
 	case GAUDI2_EVENT_SM0_AXI_ERROR_RESPONSE ... GAUDI2_EVENT_SM3_AXI_ERROR_RESPONSE:
 		index = event_type - GAUDI2_EVENT_SM0_AXI_ERROR_RESPONSE;
-		error_count = gaudi2_handle_sm_err(hdev, index);
+		error_count = gaudi2_handle_sm_err(hdev, event_type, index);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
@@ -9258,13 +9215,13 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		break;
 
 	case GAUDI2_EVENT_CPU_PKT_SANITY_FAILED:
-		gaudi2_print_cpu_pkt_failure_info(hdev, &eq_entry->pkt_sync_err);
+		gaudi2_print_cpu_pkt_failure_info(hdev, event_type, &eq_entry->pkt_sync_err);
 		error_count = GAUDI2_NA_EVENT_CAUSE;
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		break;
 
 	case GAUDI2_EVENT_ARC_DCCM_FULL:
-		error_count = hl_arc_event_handle(hdev, &eq_entry->arc_data);
+		error_count = hl_arc_event_handle(hdev, event_type, &eq_entry->arc_data);
 		event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
 		break;
 
@@ -9286,7 +9243,9 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 	 * Note that although we have counted the errors, we use this number as
 	 * a boolean.
 	 */
-	if (error_count == 0 && !is_info_event(event_type))
+	if (error_count == GAUDI2_NA_EVENT_CAUSE && !is_info_event(event_type))
+		gaudi2_print_event(hdev, event_type, true, "%d", event_type);
+	else if (error_count == 0)
 		dev_err_ratelimited(hdev->dev,
 			"No Error cause for H/W event %d\n", event_type);
 
-- 
2.25.1

[PATCH 13/26] habanalabs: adjacent timestamps should be more accurate

[Oded Gabbay]

From: Tamir Gilad-Raz <tgiladraz@habana.ai>

timestamp events that expire on the same interrupt will get the same
timestamp value

Signed-off-by: Tamir Gilad-Raz <tgiladraz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/irq.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/misc/habanalabs/common/irq.c b/drivers/misc/habanalabs/common/irq.c
index 94d537fd4fde..8bbcc223df91 100644
--- a/drivers/misc/habanalabs/common/irq.c
+++ b/drivers/misc/habanalabs/common/irq.c
@@ -228,7 +228,7 @@ static void hl_ts_free_objects(struct work_struct *work)
  * list to a dedicated workqueue to do the actual put.
  */
 static int handle_registration_node(struct hl_device *hdev, struct hl_user_pending_interrupt *pend,
-						struct list_head **free_list)
+						struct list_head **free_list, ktime_t now)
 {
 	struct timestamp_reg_free_node *free_node;
 	u64 timestamp;
@@ -246,7 +246,7 @@ static int handle_registration_node(struct hl_device *hdev, struct hl_user_pendi
 	if (!free_node)
 		return -ENOMEM;
 
-	timestamp = ktime_get_ns();
+	timestamp = ktime_to_ns(now);
 
 	*pend->ts_reg_info.timestamp_kernel_addr = timestamp;
 
@@ -298,7 +298,7 @@ static void handle_user_interrupt(struct hl_device *hdev, struct hl_user_interru
 			if (pend->ts_reg_info.buf) {
 				if (!reg_node_handle_fail) {
 					rc = handle_registration_node(hdev, pend,
-									&ts_reg_free_list_head);
+								&ts_reg_free_list_head, now);
 					if (rc)
 						reg_node_handle_fail = true;
 				}
-- 
2.25.1

[PATCH 14/26] habanalabs: skip device idle check in hpriv_release if in reset

[Oded Gabbay]

From: Tomer Tayar <ttayar@habana.ai>

When user context is released and hpriv_release() is called, there is a
device idle status check, to understand if user has left the device not
idle and then a reset is required.

However, if the user process is killed because of device hard reset,
the device at this point would always be not idle, because the device
engines were already forcefully halted.

Modify hpriv_release() to skip the idle check if reset is in progress.

Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/device.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/misc/habanalabs/common/device.c b/drivers/misc/habanalabs/common/device.c
index afd9d4d46574..71f958a2e91b 100644
--- a/drivers/misc/habanalabs/common/device.c
+++ b/drivers/misc/habanalabs/common/device.c
@@ -428,8 +428,10 @@ static void hpriv_release(struct kref *ref)
 	 */
 	reset_device = hdev->reset_upon_device_release || hdev->reset_info.watchdog_active;
 
-	/* Unless device is reset in any case, check idle status and reset if device is not idle */
-	if (!reset_device && hdev->pdev && !hdev->pldm)
+	/* Check the device idle status and reset if not idle.
+	 * Skip it if already in reset, or if device is going to be reset in any case.
+	 */
+	if (!hdev->reset_info.in_reset && !reset_device && hdev->pdev && !hdev->pldm)
 		device_is_idle = hdev->asic_funcs->is_device_idle(hdev, idle_mask,
 							HL_BUSY_ENGINES_MASK_EXT_SIZE, NULL);
 	if (!device_is_idle) {
-- 
2.25.1

[PATCH 15/26] habanalabs/gaudi2: support abrupt device reset event

[Oded Gabbay]

From: Ofir Bitton <obitton@habana.ai>

In certain scenarios, firmware might encounter a fatal event for
which a device reset is required. Hence, a proper notification
is needed for driver to be aware and initiate a reset sequence.

In secured environments the reset will be performed by firmware
without an explicit request from the driver.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/gaudi2/gaudi2.c                         | 1 +
 drivers/misc/habanalabs/include/gaudi2/gaudi2_async_events.h    | 1 +
 .../habanalabs/include/gaudi2/gaudi2_async_ids_map_extended.h   | 2 ++
 3 files changed, 4 insertions(+)

diff --git a/drivers/misc/habanalabs/gaudi2/gaudi2.c b/drivers/misc/habanalabs/gaudi2/gaudi2.c
index 8373239ad1bc..ba3b0ae76ebf 100644
--- a/drivers/misc/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/misc/habanalabs/gaudi2/gaudi2.c
@@ -9226,6 +9226,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		break;
 
 	case GAUDI2_EVENT_CPU_FP32_NOT_SUPPORTED:
+	case GAUDI2_EVENT_DEV_RESET_REQ:
 		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
 		error_count = GAUDI2_NA_EVENT_CAUSE;
 		is_critical = true;
diff --git a/drivers/misc/habanalabs/include/gaudi2/gaudi2_async_events.h b/drivers/misc/habanalabs/include/gaudi2/gaudi2_async_events.h
index 305b576222e6..50852cc80373 100644
--- a/drivers/misc/habanalabs/include/gaudi2/gaudi2_async_events.h
+++ b/drivers/misc/habanalabs/include/gaudi2/gaudi2_async_events.h
@@ -958,6 +958,7 @@ enum gaudi2_async_event_id {
 	GAUDI2_EVENT_CPU11_STATUS_NIC11_ENG1 = 1318,
 	GAUDI2_EVENT_ARC_DCCM_FULL = 1319,
 	GAUDI2_EVENT_CPU_FP32_NOT_SUPPORTED = 1320,
+	GAUDI2_EVENT_DEV_RESET_REQ = 1321,
 	GAUDI2_EVENT_SIZE,
 };
 
diff --git a/drivers/misc/habanalabs/include/gaudi2/gaudi2_async_ids_map_extended.h b/drivers/misc/habanalabs/include/gaudi2/gaudi2_async_ids_map_extended.h
index d510cb10c883..82be01bea98e 100644
--- a/drivers/misc/habanalabs/include/gaudi2/gaudi2_async_ids_map_extended.h
+++ b/drivers/misc/habanalabs/include/gaudi2/gaudi2_async_ids_map_extended.h
@@ -2665,6 +2665,8 @@ static struct gaudi2_async_events_ids_map gaudi2_irq_map_table[] = {
 		.msg = 1, .reset = 0, .name = "ARC_DCCM_FULL" },
 	{ .fc_id = 1320, .cpu_id = 626, .valid = 1,
 		.msg = 1, .reset = 1, .name = "FP32_NOT_SUPPORTED" },
+	{ .fc_id = 1321, .cpu_id = 627, .valid = 1,
+		.msg = 1, .reset = 1, .name = "DEV_RESET_REQ" },
 };
 
 #endif /* __GAUDI2_ASYNC_IDS_MAP_EVENTS_EXT_H_ */
-- 
2.25.1

[PATCH 16/26] habanalabs: define traces for COMMS protocol

[Oded Gabbay]

From: Ohad Sharabi <osharabi@habana.ai>

As the COMMS protocol is being used more widely in our driver,
an available debug tool for the handshake will be handy.

This commit defines tracepoints to various key points of the COMMS
protocol.

Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 include/trace/events/habanalabs.h | 36 +++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/include/trace/events/habanalabs.h b/include/trace/events/habanalabs.h
index f05c5fa668a2..10233e13cee4 100644
--- a/include/trace/events/habanalabs.h
+++ b/include/trace/events/habanalabs.h
@@ -87,6 +87,42 @@ DEFINE_EVENT(habanalabs_dma_alloc_template, habanalabs_dma_free,
 	TP_PROTO(struct device *dev, u64 cpu_addr, u64 dma_addr, size_t size, const char *caller),
 	TP_ARGS(dev, cpu_addr, dma_addr, size, caller));
 
+DECLARE_EVENT_CLASS(habanalabs_comms_template,
+	TP_PROTO(struct device *dev, char *op_str),
+
+	TP_ARGS(dev, op_str),
+
+	TP_STRUCT__entry(
+		__string(dname, dev_name(dev))
+		__field(char *, op_str)
+	),
+
+	TP_fast_assign(
+		__assign_str(dname, dev_name(dev));
+		__entry->op_str = op_str;
+	),
+
+	TP_printk("%s: cms: %s",
+		__get_str(dname),
+		__entry->op_str)
+);
+
+DEFINE_EVENT(habanalabs_comms_template, habanalabs_comms_protocol_cmd,
+	TP_PROTO(struct device *dev, char *op_str),
+	TP_ARGS(dev, op_str));
+
+DEFINE_EVENT(habanalabs_comms_template, habanalabs_comms_send_cmd,
+	TP_PROTO(struct device *dev, char *op_str),
+	TP_ARGS(dev, op_str));
+
+DEFINE_EVENT(habanalabs_comms_template, habanalabs_comms_wait_status,
+	TP_PROTO(struct device *dev, char *op_str),
+	TP_ARGS(dev, op_str));
+
+DEFINE_EVENT(habanalabs_comms_template, habanalabs_comms_wait_status_done,
+	TP_PROTO(struct device *dev, char *op_str),
+	TP_ARGS(dev, op_str));
+
 #endif /* if !defined(_TRACE_HABANALABS_H) || defined(TRACE_HEADER_MULTI_READ) */
 
 /* This part must be outside protection */
-- 
2.25.1

[PATCH 17/26] habanalabs: trace COMMS protocol

[Oded Gabbay]

From: Ohad Sharabi <osharabi@habana.ai>

Call COMMS tracepoints from within the dynamic CPU FW load.
This can help debug failures or delays in the dynamic FW load flow.

Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/firmware_if.c | 31 ++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/drivers/misc/habanalabs/common/firmware_if.c b/drivers/misc/habanalabs/common/firmware_if.c
index 046866c673e2..ebb81cf89f02 100644
--- a/drivers/misc/habanalabs/common/firmware_if.c
+++ b/drivers/misc/habanalabs/common/firmware_if.c
@@ -14,8 +14,32 @@
 #include <linux/ctype.h>
 #include <linux/vmalloc.h>
 
+#include <trace/events/habanalabs.h>
+
 #define FW_FILE_MAX_SIZE		0x1400000 /* maximum size of 20MB */
 
+static char *comms_cmd_str_arr[COMMS_INVLD_LAST] = {
+	[COMMS_NOOP] = __stringify(COMMS_NOOP),
+	[COMMS_CLR_STS] = __stringify(COMMS_CLR_STS),
+	[COMMS_RST_STATE] = __stringify(COMMS_RST_STATE),
+	[COMMS_PREP_DESC] = __stringify(COMMS_PREP_DESC),
+	[COMMS_DATA_RDY] = __stringify(COMMS_DATA_RDY),
+	[COMMS_EXEC] = __stringify(COMMS_EXEC),
+	[COMMS_RST_DEV] = __stringify(COMMS_RST_DEV),
+	[COMMS_GOTO_WFE] = __stringify(COMMS_GOTO_WFE),
+	[COMMS_SKIP_BMC] = __stringify(COMMS_SKIP_BMC),
+	[COMMS_PREP_DESC_ELBI] = __stringify(COMMS_PREP_DESC_ELBI),
+};
+
+static char *comms_sts_str_arr[COMMS_STS_INVLD_LAST] = {
+	[COMMS_STS_NOOP] = __stringify(COMMS_STS_NOOP),
+	[COMMS_STS_ACK] = __stringify(COMMS_STS_ACK),
+	[COMMS_STS_OK] = __stringify(COMMS_STS_OK),
+	[COMMS_STS_ERR] = __stringify(COMMS_STS_ERR),
+	[COMMS_STS_VALID_ERR] = __stringify(COMMS_STS_VALID_ERR),
+	[COMMS_STS_TIMEOUT_ERR] = __stringify(COMMS_STS_TIMEOUT_ERR),
+};
+
 static char *extract_fw_ver_from_str(const char *fw_str)
 {
 	char *str, *fw_ver, *whitespace;
@@ -1634,6 +1658,7 @@ static void hl_fw_dynamic_send_cmd(struct hl_device *hdev,
 	val = FIELD_PREP(COMMS_COMMAND_CMD_MASK, cmd);
 	val |= FIELD_PREP(COMMS_COMMAND_SIZE_MASK, size);
 
+	trace_habanalabs_comms_send_cmd(hdev->dev, comms_cmd_str_arr[cmd]);
 	WREG32(le32_to_cpu(dyn_regs->kmd_msg_to_cpu), val);
 }
 
@@ -1691,6 +1716,8 @@ static int hl_fw_dynamic_wait_for_status(struct hl_device *hdev,
 
 	dyn_regs = &fw_loader->dynamic_loader.comm_desc.cpu_dyn_regs;
 
+	trace_habanalabs_comms_wait_status(hdev->dev, comms_sts_str_arr[expected_status]);
+
 	/* Wait for expected status */
 	rc = hl_poll_timeout(
 		hdev,
@@ -1706,6 +1733,8 @@ static int hl_fw_dynamic_wait_for_status(struct hl_device *hdev,
 		return -EIO;
 	}
 
+	trace_habanalabs_comms_wait_status_done(hdev->dev, comms_sts_str_arr[expected_status]);
+
 	/*
 	 * skip storing FW response for NOOP to preserve the actual desired
 	 * FW status
@@ -1778,6 +1807,8 @@ int hl_fw_dynamic_send_protocol_cmd(struct hl_device *hdev,
 {
 	int rc;
 
+	trace_habanalabs_comms_protocol_cmd(hdev->dev, comms_cmd_str_arr[cmd]);
+
 	/* first send clear command to clean former commands */
 	rc = hl_fw_dynamic_send_clear_cmd(hdev, fw_loader);
 	if (rc)
-- 
2.25.1

[PATCH 18/26] habanalabs: set log level for descriptor validation to debug

[Oded Gabbay]

From: farah kassabri <fkassabri@habana.ai>

This warning doesn't have real consequences, and therefore can be
printed in debug level.

Signed-off-by: farah kassabri <fkassabri@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/firmware_if.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/misc/habanalabs/common/firmware_if.c b/drivers/misc/habanalabs/common/firmware_if.c
index ebb81cf89f02..537b1ae3fcb7 100644
--- a/drivers/misc/habanalabs/common/firmware_if.c
+++ b/drivers/misc/habanalabs/common/firmware_if.c
@@ -1932,11 +1932,11 @@ static int hl_fw_dynamic_validate_descriptor(struct hl_device *hdev,
 	int rc;
 
 	if (le32_to_cpu(fw_desc->header.magic) != HL_COMMS_DESC_MAGIC)
-		dev_warn(hdev->dev, "Invalid magic for dynamic FW descriptor (%x)\n",
+		dev_dbg(hdev->dev, "Invalid magic for dynamic FW descriptor (%x)\n",
 				fw_desc->header.magic);
 
 	if (fw_desc->header.version != HL_COMMS_DESC_VER)
-		dev_warn(hdev->dev, "Invalid version for dynamic FW descriptor (%x)\n",
+		dev_dbg(hdev->dev, "Invalid version for dynamic FW descriptor (%x)\n",
 				fw_desc->header.version);
 
 	/*
-- 
2.25.1

[PATCH 19/26] habanalabs: remove support to export dmabuf from handle

[Oded Gabbay]

From: Ohad Sharabi <osharabi@habana.ai>

The API to the user which allows exporting DMA buffer from handle is
deprecated here. It was never used as it is relevant only for Gaudi2,
and the user stack has yet to add support for dmabuf in Gaudi2.

Looking forward, a modified API to export DMA buffer for ASICs that
supports virtual memory will be added.

Until the new API will be ready- exporting DMA buffer will not be
supported for ASICs with virtual memory.

Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/habanalabs.h |   5 -
 drivers/misc/habanalabs/common/memory.c     | 143 ++------------------
 2 files changed, 9 insertions(+), 139 deletions(-)

diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index 893ebcba170b..e68928b59c1e 100644
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -1744,8 +1744,6 @@ struct hl_cs_counters_atomic {
  * struct hl_dmabuf_priv - a dma-buf private object.
  * @dmabuf: pointer to dma-buf object.
  * @ctx: pointer to the dma-buf owner's context.
- * @phys_pg_pack: pointer to physical page pack if the dma-buf was exported for
- *                memory allocation handle.
  * @device_address: physical address of the device's memory. Relevant only
  *                  if phys_pg_pack is NULL (dma-buf was exported from address).
  *                  The total size can be taken from the dmabuf object.
@@ -1753,7 +1751,6 @@ struct hl_cs_counters_atomic {
 struct hl_dmabuf_priv {
 	struct dma_buf			*dmabuf;
 	struct hl_ctx			*ctx;
-	struct hl_vm_phys_pg_pack	*phys_pg_pack;
 	uint64_t			device_address;
 };
 
@@ -2117,7 +2114,6 @@ struct hl_vm_hw_block_list_node {
  * @node: used to attach to deletion list that is used when all the allocations are cleared
  *        at the teardown of the context.
  * @mapping_cnt: number of shared mappings.
- * @exporting_cnt: number of dma-buf exporting.
  * @asid: the context related to this list.
  * @page_size: size of each page in the pack.
  * @flags: HL_MEM_* flags related to this list.
@@ -2133,7 +2129,6 @@ struct hl_vm_phys_pg_pack {
 	u64			total_size;
 	struct list_head	node;
 	atomic_t		mapping_cnt;
-	u32			exporting_cnt;
 	u32			asid;
 	u32			page_size;
 	u32			flags;
diff --git a/drivers/misc/habanalabs/common/memory.c b/drivers/misc/habanalabs/common/memory.c
index 7c5c18be294a..864a8a1c6067 100644
--- a/drivers/misc/habanalabs/common/memory.c
+++ b/drivers/misc/habanalabs/common/memory.c
@@ -371,12 +371,6 @@ static int free_device_memory(struct hl_ctx *ctx, struct hl_mem_in *args)
 		return -EINVAL;
 	}
 
-	if (phys_pg_pack->exporting_cnt) {
-		spin_unlock(&vm->idr_lock);
-		dev_dbg(hdev->dev, "handle %u is exported, cannot free\n", handle);
-		return -EINVAL;
-	}
-
 	/* must remove from idr before the freeing of the physical pages as the refcount of the pool
 	 * is also the trigger of the idr destroy
 	 */
@@ -1700,29 +1694,19 @@ static struct sg_table *hl_map_dmabuf(struct dma_buf_attachment *attachment,
 					enum dma_data_direction dir)
 {
 	struct dma_buf *dma_buf = attachment->dmabuf;
-	struct hl_vm_phys_pg_pack *phys_pg_pack;
 	struct hl_dmabuf_priv *hl_dmabuf;
 	struct hl_device *hdev;
 	struct sg_table *sgt;
 
 	hl_dmabuf = dma_buf->priv;
 	hdev = hl_dmabuf->ctx->hdev;
-	phys_pg_pack = hl_dmabuf->phys_pg_pack;
 
 	if (!attachment->peer2peer) {
 		dev_dbg(hdev->dev, "Failed to map dmabuf because p2p is disabled\n");
 		return ERR_PTR(-EPERM);
 	}
 
-	if (phys_pg_pack)
-		sgt = alloc_sgt_from_device_pages(hdev,
-						phys_pg_pack->pages,
-						phys_pg_pack->npages,
-						phys_pg_pack->page_size,
-						attachment->dev,
-						dir);
-	else
-		sgt = alloc_sgt_from_device_pages(hdev,
+	sgt = alloc_sgt_from_device_pages(hdev,
 						&hl_dmabuf->device_address,
 						1,
 						hl_dmabuf->dmabuf->size,
@@ -1763,18 +1747,8 @@ static void hl_unmap_dmabuf(struct dma_buf_attachment *attachment,
 static void hl_release_dmabuf(struct dma_buf *dmabuf)
 {
 	struct hl_dmabuf_priv *hl_dmabuf = dmabuf->priv;
-	struct hl_ctx *ctx = hl_dmabuf->ctx;
-	struct hl_device *hdev = ctx->hdev;
-	struct hl_vm *vm = &hdev->vm;
-
-	if (hl_dmabuf->phys_pg_pack) {
-		spin_lock(&vm->idr_lock);
-		hl_dmabuf->phys_pg_pack->exporting_cnt--;
-		spin_unlock(&vm->idr_lock);
-	}
 
 	hl_ctx_put(hl_dmabuf->ctx);
-
 	kfree(hl_dmabuf);
 }
 
@@ -1785,7 +1759,7 @@ static const struct dma_buf_ops habanalabs_dmabuf_ops = {
 	.release = hl_release_dmabuf,
 };
 
-static int export_dmabuf_common(struct hl_ctx *ctx,
+static int export_dmabuf(struct hl_ctx *ctx,
 				struct hl_dmabuf_priv *hl_dmabuf,
 				u64 total_size, int flags, int *dmabuf_fd)
 {
@@ -1849,6 +1823,11 @@ static int export_dmabuf_from_addr(struct hl_ctx *ctx, u64 device_addr,
 
 	prop = &hdev->asic_prop;
 
+	if (prop->dram_supports_virtual_memory) {
+		dev_dbg(hdev->dev, "Export not supported for devices with virtual memory\n");
+		return -EOPNOTSUPP;
+	}
+
 	if (!IS_ALIGNED(device_addr, PAGE_SIZE)) {
 		dev_dbg(hdev->dev,
 			"exported device memory address 0x%llx should be aligned to 0x%lx\n",
@@ -1890,99 +1869,7 @@ static int export_dmabuf_from_addr(struct hl_ctx *ctx, u64 device_addr,
 
 	hl_dmabuf->device_address = device_addr;
 
-	rc = export_dmabuf_common(ctx, hl_dmabuf, size, flags, dmabuf_fd);
-	if (rc)
-		goto err_free_dmabuf_wrapper;
-
-	return 0;
-
-err_free_dmabuf_wrapper:
-	kfree(hl_dmabuf);
-	return rc;
-}
-
-/**
- * export_dmabuf_from_handle() - export a dma-buf object for the given memory
- *                               handle.
- * @ctx: pointer to the context structure.
- * @handle: device memory allocation handle.
- * @flags: DMA-BUF file/FD flags.
- * @dmabuf_fd: pointer to result FD that represents the dma-buf object.
- *
- * Create and export a dma-buf object for an existing memory allocation inside
- * the device memory, and return a FD which is associated with the dma-buf
- * object.
- *
- * Return: 0 on success, non-zero for failure.
- */
-static int export_dmabuf_from_handle(struct hl_ctx *ctx, u64 handle, int flags,
-					int *dmabuf_fd)
-{
-	struct hl_vm_phys_pg_pack *phys_pg_pack;
-	struct hl_dmabuf_priv *hl_dmabuf;
-	struct hl_device *hdev = ctx->hdev;
-	struct asic_fixed_properties *prop;
-	struct hl_vm *vm = &hdev->vm;
-	u64 bar_address;
-	int rc, i;
-
-	prop = &hdev->asic_prop;
-
-	if (upper_32_bits(handle)) {
-		dev_dbg(hdev->dev, "no match for handle 0x%llx\n", handle);
-		return -EINVAL;
-	}
-
-	spin_lock(&vm->idr_lock);
-
-	phys_pg_pack = idr_find(&vm->phys_pg_pack_handles, (u32) handle);
-	if (!phys_pg_pack) {
-		spin_unlock(&vm->idr_lock);
-		dev_dbg(hdev->dev, "no match for handle 0x%x\n", (u32) handle);
-		return -EINVAL;
-	}
-
-	/* increment now to avoid freeing device memory while exporting */
-	phys_pg_pack->exporting_cnt++;
-
-	spin_unlock(&vm->idr_lock);
-
-	if (phys_pg_pack->vm_type != VM_TYPE_PHYS_PACK) {
-		dev_dbg(hdev->dev, "handle 0x%llx does not represent DRAM memory\n", handle);
-		rc = -EINVAL;
-		goto err_dec_exporting_cnt;
-	}
-
-	for (i = 0 ; i < phys_pg_pack->npages ; i++) {
-
-		bar_address = hdev->dram_pci_bar_start +
-						(phys_pg_pack->pages[i] -
-						prop->dram_base_address);
-
-		if (bar_address + phys_pg_pack->page_size >
-			hdev->dram_pci_bar_start + prop->dram_pci_bar_size ||
-			bar_address + phys_pg_pack->page_size < bar_address) {
-
-			dev_dbg(hdev->dev,
-				"DRAM memory range 0x%llx (+0x%x) is outside of PCI BAR boundaries\n",
-				phys_pg_pack->pages[i],
-				phys_pg_pack->page_size);
-
-			rc = -EINVAL;
-			goto err_dec_exporting_cnt;
-		}
-	}
-
-	hl_dmabuf = kzalloc(sizeof(*hl_dmabuf), GFP_KERNEL);
-	if (!hl_dmabuf) {
-		rc = -ENOMEM;
-		goto err_dec_exporting_cnt;
-	}
-
-	hl_dmabuf->phys_pg_pack = phys_pg_pack;
-
-	rc = export_dmabuf_common(ctx, hl_dmabuf, phys_pg_pack->total_size,
-				flags, dmabuf_fd);
+	rc = export_dmabuf(ctx, hl_dmabuf, size, flags, dmabuf_fd);
 	if (rc)
 		goto err_free_dmabuf_wrapper;
 
@@ -1990,12 +1877,6 @@ static int export_dmabuf_from_handle(struct hl_ctx *ctx, u64 handle, int flags,
 
 err_free_dmabuf_wrapper:
 	kfree(hl_dmabuf);
-
-err_dec_exporting_cnt:
-	spin_lock(&vm->idr_lock);
-	phys_pg_pack->exporting_cnt--;
-	spin_unlock(&vm->idr_lock);
-
 	return rc;
 }
 
@@ -2269,13 +2150,7 @@ int hl_mem_ioctl(struct hl_fpriv *hpriv, void *data)
 		break;
 
 	case HL_MEM_OP_EXPORT_DMABUF_FD:
-		if (hdev->asic_prop.dram_supports_virtual_memory)
-			rc = export_dmabuf_from_handle(ctx,
-					args->in.export_dmabuf_fd.handle,
-					args->in.flags,
-					&dmabuf_fd);
-		else
-			rc = export_dmabuf_from_addr(ctx,
+		rc = export_dmabuf_from_addr(ctx,
 					args->in.export_dmabuf_fd.handle,
 					args->in.export_dmabuf_fd.mem_size,
 					args->in.flags,
-- 
2.25.1

[PATCH 20/26] habanalabs: helper function to validate export params

[Oded Gabbay]

From: Ohad Sharabi <osharabi@habana.ai>

Validate export parameters in a dedicated function instead of in the
main export flow.
This will be useful later when support to export dmabuf for devices
with virtual memory will be added.

Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/memory.c | 79 ++++++++++++++-----------
 1 file changed, 44 insertions(+), 35 deletions(-)

diff --git a/drivers/misc/habanalabs/common/memory.c b/drivers/misc/habanalabs/common/memory.c
index 864a8a1c6067..e3b2e882b037 100644
--- a/drivers/misc/habanalabs/common/memory.c
+++ b/drivers/misc/habanalabs/common/memory.c
@@ -1797,36 +1797,10 @@ static int export_dmabuf(struct hl_ctx *ctx,
 	return rc;
 }
 
-/**
- * export_dmabuf_from_addr() - export a dma-buf object for the given memory
- *                             address and size.
- * @ctx: pointer to the context structure.
- * @device_addr:  device memory physical address.
- * @size: size of device memory.
- * @flags: DMA-BUF file/FD flags.
- * @dmabuf_fd: pointer to result FD that represents the dma-buf object.
- *
- * Create and export a dma-buf object for an existing memory allocation inside
- * the device memory, and return a FD which is associated with the dma-buf
- * object.
- *
- * Return: 0 on success, non-zero for failure.
- */
-static int export_dmabuf_from_addr(struct hl_ctx *ctx, u64 device_addr,
-					u64 size, int flags, int *dmabuf_fd)
+static int validate_export_params(struct hl_device *hdev, u64 device_addr, u64 size)
 {
-	struct hl_dmabuf_priv *hl_dmabuf;
-	struct hl_device *hdev = ctx->hdev;
-	struct asic_fixed_properties *prop;
+	struct asic_fixed_properties *prop = &hdev->asic_prop;
 	u64 bar_address;
-	int rc;
-
-	prop = &hdev->asic_prop;
-
-	if (prop->dram_supports_virtual_memory) {
-		dev_dbg(hdev->dev, "Export not supported for devices with virtual memory\n");
-		return -EOPNOTSUPP;
-	}
 
 	if (!IS_ALIGNED(device_addr, PAGE_SIZE)) {
 		dev_dbg(hdev->dev,
@@ -1843,26 +1817,61 @@ static int export_dmabuf_from_addr(struct hl_ctx *ctx, u64 device_addr,
 	}
 
 	if (device_addr < prop->dram_user_base_address ||
-				device_addr + size > prop->dram_end_address ||
-				device_addr + size < device_addr) {
+				(device_addr + size) > prop->dram_end_address ||
+				(device_addr + size) < device_addr) {
 		dev_dbg(hdev->dev,
 			"DRAM memory range 0x%llx (+0x%llx) is outside of DRAM boundaries\n",
 			device_addr, size);
 		return -EINVAL;
 	}
 
-	bar_address = hdev->dram_pci_bar_start +
-			(device_addr - prop->dram_base_address);
+	bar_address = hdev->dram_pci_bar_start + (device_addr - prop->dram_base_address);
 
-	if (bar_address + size >
-			hdev->dram_pci_bar_start + prop->dram_pci_bar_size ||
-			bar_address + size < bar_address) {
+	if ((bar_address + size) > (hdev->dram_pci_bar_start + prop->dram_pci_bar_size) ||
+			(bar_address + size) < bar_address) {
 		dev_dbg(hdev->dev,
 			"DRAM memory range 0x%llx (+0x%llx) is outside of PCI BAR boundaries\n",
 			device_addr, size);
 		return -EINVAL;
 	}
 
+	return 0;
+}
+
+/**
+ * export_dmabuf_from_addr() - export a dma-buf object for the given memory
+ *                             address and size.
+ * @ctx: pointer to the context structure.
+ * @device_addr:  device memory physical address.
+ * @size: size of device memory.
+ * @flags: DMA-BUF file/FD flags.
+ * @dmabuf_fd: pointer to result FD that represents the dma-buf object.
+ *
+ * Create and export a dma-buf object for an existing memory allocation inside
+ * the device memory, and return a FD which is associated with the dma-buf
+ * object.
+ *
+ * Return: 0 on success, non-zero for failure.
+ */
+static int export_dmabuf_from_addr(struct hl_ctx *ctx, u64 device_addr,
+					u64 size, int flags, int *dmabuf_fd)
+{
+	struct hl_dmabuf_priv *hl_dmabuf;
+	struct hl_device *hdev = ctx->hdev;
+	struct asic_fixed_properties *prop;
+	int rc;
+
+	prop = &hdev->asic_prop;
+
+	if (prop->dram_supports_virtual_memory) {
+		dev_dbg(hdev->dev, "Export not supported for devices with virtual memory\n");
+		return -EOPNOTSUPP;
+	}
+
+	rc = validate_export_params(hdev, device_addr, size);
+	if (rc)
+		return rc;
+
 	hl_dmabuf = kzalloc(sizeof(*hl_dmabuf), GFP_KERNEL);
 	if (!hl_dmabuf)
 		return -ENOMEM;
-- 
2.25.1

[PATCH 21/26] habanalabs: modify export dmabuf API

[Oded Gabbay]

From: Ohad Sharabi <osharabi@habana.ai>

A previous commit deprecated the option to export from handle, leaving
the code with no support for devices with virtual memory.

This commit modifies the export API in a way that unifies the uAPI to
user address for both cases (i.e. with and without MMU support) and add
the actual support for devices with virtual memory.

Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/habanalabs.h |   9 +
 drivers/misc/habanalabs/common/memory.c     | 219 +++++++++++++++++---
 include/uapi/misc/habanalabs.h              |  21 +-
 3 files changed, 218 insertions(+), 31 deletions(-)

diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index e68928b59c1e..ef5a765f3313 100644
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -1744,6 +1744,9 @@ struct hl_cs_counters_atomic {
  * struct hl_dmabuf_priv - a dma-buf private object.
  * @dmabuf: pointer to dma-buf object.
  * @ctx: pointer to the dma-buf owner's context.
+ * @phys_pg_pack: pointer to physical page pack if the dma-buf was exported
+ *                where virtual memory is supported.
+ * @memhash_hnode: pointer to the memhash node. this object holds the export count.
  * @device_address: physical address of the device's memory. Relevant only
  *                  if phys_pg_pack is NULL (dma-buf was exported from address).
  *                  The total size can be taken from the dmabuf object.
@@ -1751,6 +1754,8 @@ struct hl_cs_counters_atomic {
 struct hl_dmabuf_priv {
 	struct dma_buf			*dmabuf;
 	struct hl_ctx			*ctx;
+	struct hl_vm_phys_pg_pack	*phys_pg_pack;
+	struct hl_vm_hash_node		*memhash_hnode;
 	uint64_t			device_address;
 };
 
@@ -2078,12 +2083,16 @@ struct hl_cs_parser {
  *				hl_userptr).
  * @node: node to hang on the hash table in context object.
  * @vaddr: key virtual address.
+ * @handle: memory handle for device memory allocation.
  * @ptr: value pointer (hl_vm_phys_pg_list or hl_userptr).
+ * @export_cnt: number of exports from within the VA block.
  */
 struct hl_vm_hash_node {
 	struct hlist_node	node;
 	u64			vaddr;
+	u64			handle;
 	void			*ptr;
+	int			export_cnt;
 };
 
 /**
diff --git a/drivers/misc/habanalabs/common/memory.c b/drivers/misc/habanalabs/common/memory.c
index e3b2e882b037..c7c27ffa6309 100644
--- a/drivers/misc/habanalabs/common/memory.c
+++ b/drivers/misc/habanalabs/common/memory.c
@@ -19,7 +19,9 @@ MODULE_IMPORT_NS(DMA_BUF);
 #define HL_MMU_DEBUG	0
 
 /* use small pages for supporting non-pow2 (32M/40M/48M) DRAM phys page sizes */
-#define DRAM_POOL_PAGE_SIZE SZ_8M
+#define DRAM_POOL_PAGE_SIZE	SZ_8M
+
+#define MEM_HANDLE_INVALID	ULONG_MAX
 
 static int allocate_timestamps_buffers(struct hl_fpriv *hpriv,
 			struct hl_mem_in *args, u64 *handle);
@@ -1234,6 +1236,7 @@ static int map_device_va(struct hl_ctx *ctx, struct hl_mem_in *args, u64 *device
 
 	hnode->ptr = vm_type;
 	hnode->vaddr = ret_vaddr;
+	hnode->handle = is_userptr ? MEM_HANDLE_INVALID : handle;
 
 	mutex_lock(&ctx->mem_hash_lock);
 	hash_add(ctx->mem_hash, &hnode->node, ret_vaddr);
@@ -1307,6 +1310,12 @@ static int unmap_device_va(struct hl_ctx *ctx, struct hl_mem_in *args,
 		return -EINVAL;
 	}
 
+	if (hnode->export_cnt) {
+		mutex_unlock(&ctx->mem_hash_lock);
+		dev_err(hdev->dev, "failed to unmap %#llx, memory is exported\n", vaddr);
+		return -EINVAL;
+	}
+
 	hash_del(&hnode->node);
 	mutex_unlock(&ctx->mem_hash_lock);
 
@@ -1694,19 +1703,29 @@ static struct sg_table *hl_map_dmabuf(struct dma_buf_attachment *attachment,
 					enum dma_data_direction dir)
 {
 	struct dma_buf *dma_buf = attachment->dmabuf;
+	struct hl_vm_phys_pg_pack *phys_pg_pack;
 	struct hl_dmabuf_priv *hl_dmabuf;
 	struct hl_device *hdev;
 	struct sg_table *sgt;
 
 	hl_dmabuf = dma_buf->priv;
 	hdev = hl_dmabuf->ctx->hdev;
+	phys_pg_pack = hl_dmabuf->phys_pg_pack;
 
 	if (!attachment->peer2peer) {
 		dev_dbg(hdev->dev, "Failed to map dmabuf because p2p is disabled\n");
 		return ERR_PTR(-EPERM);
 	}
 
-	sgt = alloc_sgt_from_device_pages(hdev,
+	if (phys_pg_pack)
+		sgt = alloc_sgt_from_device_pages(hdev,
+						phys_pg_pack->pages,
+						phys_pg_pack->npages,
+						phys_pg_pack->page_size,
+						attachment->dev,
+						dir);
+	else
+		sgt = alloc_sgt_from_device_pages(hdev,
 						&hl_dmabuf->device_address,
 						1,
 						hl_dmabuf->dmabuf->size,
@@ -1747,8 +1766,15 @@ static void hl_unmap_dmabuf(struct dma_buf_attachment *attachment,
 static void hl_release_dmabuf(struct dma_buf *dmabuf)
 {
 	struct hl_dmabuf_priv *hl_dmabuf = dmabuf->priv;
+	struct hl_ctx *ctx = hl_dmabuf->ctx;
+
+	if (hl_dmabuf->memhash_hnode) {
+		mutex_lock(&ctx->mem_hash_lock);
+		hl_dmabuf->memhash_hnode->export_cnt--;
+		mutex_unlock(&ctx->mem_hash_lock);
+	}
 
-	hl_ctx_put(hl_dmabuf->ctx);
+	hl_ctx_put(ctx);
 	kfree(hl_dmabuf);
 }
 
@@ -1797,11 +1823,8 @@ static int export_dmabuf(struct hl_ctx *ctx,
 	return rc;
 }
 
-static int validate_export_params(struct hl_device *hdev, u64 device_addr, u64 size)
+static int validate_export_params_common(struct hl_device *hdev, u64 device_addr, u64 size)
 {
-	struct asic_fixed_properties *prop = &hdev->asic_prop;
-	u64 bar_address;
-
 	if (!IS_ALIGNED(device_addr, PAGE_SIZE)) {
 		dev_dbg(hdev->dev,
 			"exported device memory address 0x%llx should be aligned to 0x%lx\n",
@@ -1816,6 +1839,19 @@ static int validate_export_params(struct hl_device *hdev, u64 device_addr, u64 s
 		return -EINVAL;
 	}
 
+	return 0;
+}
+
+static int validate_export_params_no_mmu(struct hl_device *hdev, u64 device_addr, u64 size)
+{
+	struct asic_fixed_properties *prop = &hdev->asic_prop;
+	u64 bar_address;
+	int rc;
+
+	rc = validate_export_params_common(hdev, device_addr, size);
+	if (rc)
+		return rc;
+
 	if (device_addr < prop->dram_user_base_address ||
 				(device_addr + size) > prop->dram_end_address ||
 				(device_addr + size) < device_addr) {
@@ -1838,12 +1874,115 @@ static int validate_export_params(struct hl_device *hdev, u64 device_addr, u64 s
 	return 0;
 }
 
+static int validate_export_params(struct hl_device *hdev, u64 device_addr, u64 size, u64 offset,
+					struct hl_vm_phys_pg_pack *phys_pg_pack)
+{
+	struct asic_fixed_properties *prop = &hdev->asic_prop;
+	u64 bar_address;
+	int i, rc;
+
+	rc = validate_export_params_common(hdev, device_addr, size);
+	if (rc)
+		return rc;
+
+	if ((offset + size) > phys_pg_pack->total_size) {
+		dev_dbg(hdev->dev, "offset %#llx and size %#llx exceed total map size %#llx\n",
+				offset, size, phys_pg_pack->total_size);
+		return -EINVAL;
+	}
+
+	for (i = 0 ; i < phys_pg_pack->npages ; i++) {
+
+		bar_address = hdev->dram_pci_bar_start +
+					(phys_pg_pack->pages[i] - prop->dram_base_address);
+
+		if ((bar_address + phys_pg_pack->page_size) >
+				(hdev->dram_pci_bar_start + prop->dram_pci_bar_size) ||
+				(bar_address + phys_pg_pack->page_size) < bar_address) {
+			dev_dbg(hdev->dev,
+				"DRAM memory range 0x%llx (+0x%x) is outside of PCI BAR boundaries\n",
+					phys_pg_pack->pages[i],
+					phys_pg_pack->page_size);
+
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static struct hl_vm_hash_node *memhash_node_export_get(struct hl_ctx *ctx, u64 addr)
+{
+	struct hl_device *hdev = ctx->hdev;
+	struct hl_vm_hash_node *hnode;
+
+	/* get the memory handle */
+	mutex_lock(&ctx->mem_hash_lock);
+	hash_for_each_possible(ctx->mem_hash, hnode, node, (unsigned long)addr)
+		if (addr == hnode->vaddr)
+			break;
+
+	if (!hnode) {
+		mutex_unlock(&ctx->mem_hash_lock);
+		dev_dbg(hdev->dev, "map address %#llx not found\n", addr);
+		return ERR_PTR(-EINVAL);
+	}
+
+	if (upper_32_bits(hnode->handle)) {
+		mutex_unlock(&ctx->mem_hash_lock);
+		dev_dbg(hdev->dev, "invalid handle %#llx for map address %#llx\n",
+				hnode->handle, addr);
+		return ERR_PTR(-EINVAL);
+	}
+
+	/*
+	 * node found, increase export count so this memory cannot be unmapped
+	 * and the hash node cannot be deleted.
+	 */
+	hnode->export_cnt++;
+	mutex_unlock(&ctx->mem_hash_lock);
+
+	return hnode;
+}
+
+static void memhash_node_export_put(struct hl_ctx *ctx, struct hl_vm_hash_node *hnode)
+{
+	mutex_lock(&ctx->mem_hash_lock);
+	hnode->export_cnt--;
+	mutex_unlock(&ctx->mem_hash_lock);
+}
+
+static struct hl_vm_phys_pg_pack *get_phys_pg_pack_from_hash_node(struct hl_device *hdev,
+							struct hl_vm_hash_node *hnode)
+{
+	struct hl_vm_phys_pg_pack *phys_pg_pack;
+	struct hl_vm *vm = &hdev->vm;
+
+	spin_lock(&vm->idr_lock);
+	phys_pg_pack = idr_find(&vm->phys_pg_pack_handles, (u32) hnode->handle);
+	if (!phys_pg_pack) {
+		spin_unlock(&vm->idr_lock);
+		dev_dbg(hdev->dev, "no match for handle 0x%x\n", (u32) hnode->handle);
+		return ERR_PTR(-EINVAL);
+	}
+
+	spin_unlock(&vm->idr_lock);
+
+	if (phys_pg_pack->vm_type != VM_TYPE_PHYS_PACK) {
+		dev_dbg(hdev->dev, "handle 0x%llx does not represent DRAM memory\n", hnode->handle);
+		return ERR_PTR(-EINVAL);
+	}
+
+	return phys_pg_pack;
+}
+
 /**
  * export_dmabuf_from_addr() - export a dma-buf object for the given memory
  *                             address and size.
  * @ctx: pointer to the context structure.
- * @device_addr:  device memory physical address.
- * @size: size of device memory.
+ * @addr: device address.
+ * @size: size of device memory to export.
+ * @offset: the offset into the buffer from which to start exporting
  * @flags: DMA-BUF file/FD flags.
  * @dmabuf_fd: pointer to result FD that represents the dma-buf object.
  *
@@ -1853,37 +1992,66 @@ static int validate_export_params(struct hl_device *hdev, u64 device_addr, u64 s
  *
  * Return: 0 on success, non-zero for failure.
  */
-static int export_dmabuf_from_addr(struct hl_ctx *ctx, u64 device_addr,
-					u64 size, int flags, int *dmabuf_fd)
+static int export_dmabuf_from_addr(struct hl_ctx *ctx, u64 addr, u64 size, u64 offset,
+					int flags, int *dmabuf_fd)
 {
-	struct hl_dmabuf_priv *hl_dmabuf;
-	struct hl_device *hdev = ctx->hdev;
+	struct hl_vm_phys_pg_pack *phys_pg_pack = NULL;
+	struct hl_vm_hash_node *hnode = NULL;
 	struct asic_fixed_properties *prop;
+	struct hl_dmabuf_priv *hl_dmabuf;
+	struct hl_device *hdev;
+	u64 export_addr;
 	int rc;
 
+	hdev = ctx->hdev;
 	prop = &hdev->asic_prop;
 
-	if (prop->dram_supports_virtual_memory) {
-		dev_dbg(hdev->dev, "Export not supported for devices with virtual memory\n");
-		return -EOPNOTSUPP;
+	/* offset must be 0 in devices without virtual memory support */
+	if (!prop->dram_supports_virtual_memory && offset) {
+		dev_dbg(hdev->dev, "offset is not allowed in device without virtual memory\n");
+		return -EINVAL;
 	}
 
-	rc = validate_export_params(hdev, device_addr, size);
-	if (rc)
-		return rc;
+	export_addr = addr + offset;
 
 	hl_dmabuf = kzalloc(sizeof(*hl_dmabuf), GFP_KERNEL);
 	if (!hl_dmabuf)
 		return -ENOMEM;
 
-	hl_dmabuf->device_address = device_addr;
+	if (prop->dram_supports_virtual_memory) {
+		hnode = memhash_node_export_get(ctx, addr);
+		if (IS_ERR(hnode)) {
+			rc = PTR_ERR(hnode);
+			goto err_free_dmabuf_wrapper;
+		}
+		phys_pg_pack = get_phys_pg_pack_from_hash_node(hdev, hnode);
+		if (IS_ERR(phys_pg_pack)) {
+			rc = PTR_ERR(phys_pg_pack);
+			goto dec_memhash_export_cnt;
+		}
+		rc = validate_export_params(hdev, export_addr, size, offset, phys_pg_pack);
+		if (rc)
+			goto dec_memhash_export_cnt;
+
+		hl_dmabuf->phys_pg_pack = phys_pg_pack;
+		hl_dmabuf->memhash_hnode = hnode;
+	} else {
+		rc = validate_export_params_no_mmu(hdev, export_addr, size);
+		if (rc)
+			goto err_free_dmabuf_wrapper;
+	}
+
+	hl_dmabuf->device_address = export_addr;
 
 	rc = export_dmabuf(ctx, hl_dmabuf, size, flags, dmabuf_fd);
 	if (rc)
-		goto err_free_dmabuf_wrapper;
+		goto dec_memhash_export_cnt;
 
 	return 0;
 
+dec_memhash_export_cnt:
+	if (prop->dram_supports_virtual_memory)
+		memhash_node_export_put(ctx, hnode);
 err_free_dmabuf_wrapper:
 	kfree(hl_dmabuf);
 	return rc;
@@ -2160,10 +2328,11 @@ int hl_mem_ioctl(struct hl_fpriv *hpriv, void *data)
 
 	case HL_MEM_OP_EXPORT_DMABUF_FD:
 		rc = export_dmabuf_from_addr(ctx,
-					args->in.export_dmabuf_fd.handle,
-					args->in.export_dmabuf_fd.mem_size,
-					args->in.flags,
-					&dmabuf_fd);
+				args->in.export_dmabuf_fd.addr,
+				args->in.export_dmabuf_fd.mem_size,
+				args->in.export_dmabuf_fd.offset,
+				args->in.flags,
+				&dmabuf_fd);
 		memset(args, 0, sizeof(*args));
 		args->out.fd = dmabuf_fd;
 		break;
diff --git a/include/uapi/misc/habanalabs.h b/include/uapi/misc/habanalabs.h
index 3b995e841eb8..c67d18901c1d 100644
--- a/include/uapi/misc/habanalabs.h
+++ b/include/uapi/misc/habanalabs.h
@@ -1851,15 +1851,24 @@ struct hl_mem_in {
 		/**
 		 * structure for exporting DMABUF object (used with
 		 * the HL_MEM_OP_EXPORT_DMABUF_FD op)
-		 * @handle: handle returned from HL_MEM_OP_ALLOC.
-		 *          in Gaudi, where we don't have MMU for the device memory, the
-		 *          driver expects a physical address (instead of a handle) in the
-		 *          device memory space.
-		 * @mem_size: size of memory allocation. Relevant only for GAUDI
+		 * @addr: for Gaudi1, the driver expects a physical address
+		 *        inside the device's DRAM. this is because in Gaudi1
+		 *        we don't have MMU that covers the device's DRAM.
+		 *        for all other ASICs, the driver expects a device
+		 *        virtual address that represents the start address of
+		 *        a mapped DRAM memory area inside the device.
+		 *        the address must be the same as was received from the
+		 *        driver during a previous HL_MEM_OP_MAP operation.
+		 * @mem_size: size of memory to export.
+		 * @offset: for Gaudi1, this value must be 0. For all other ASICs,
+		 *          the driver expects an offset inside of the memory area
+		 *          describe by addr. the offset represents the start
+		 *          address of that the exported dma-buf object describes.
 		 */
 		struct {
-			__u64 handle;
+			__u64 addr;
 			__u64 mem_size;
+			__u64 offset;
 		} export_dmabuf_fd;
 	};
 
-- 
2.25.1

[PATCH 22/26] habanalabs: fix dmabuf to export only required size

[Oded Gabbay]

From: Ohad Sharabi <osharabi@habana.ai>

This patch fixes a bug that was found in the dmabuf flow.
Bug description as found on Gaudi2 device:
1. User allocates 4MB of device memory
    - Note that although the allocation size was 4MB the HMMU allocated
      a full page of 768MB to back the request.
    - The user gets a memory handle that points to a single page (768MB)
    - Mapping the handle, the user gets virtual address to the start of
      the page.
2. User exports the buffer
3. User registers the exported buffer in the importer. This flow has
   a callback to the exporter which in turn converts the phys_page_pack
   to an SG list for the importer. This SG list is of single entry of
   size 768MB. However, the size that was passed to the importer was
   only 4MB.

The solution for this is to make sure the importer gets exposure only
to the exported size.

This will be done by fixing the SG created by the exporter to be of
the total size of the actual exported memory requested by the user.

Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/habanalabs.h |  2 ++
 drivers/misc/habanalabs/common/memory.c     | 35 +++++++++++++++------
 2 files changed, 28 insertions(+), 9 deletions(-)

diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index ef5a765f3313..de715c91a87e 100644
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -2120,6 +2120,7 @@ struct hl_vm_hw_block_list_node {
  * @pages: the physical page array.
  * @npages: num physical pages in the pack.
  * @total_size: total size of all the pages in this list.
+ * @exported_size: buffer exported size.
  * @node: used to attach to deletion list that is used when all the allocations are cleared
  *        at the teardown of the context.
  * @mapping_cnt: number of shared mappings.
@@ -2136,6 +2137,7 @@ struct hl_vm_phys_pg_pack {
 	u64			*pages;
 	u64			npages;
 	u64			total_size;
+	u64			exported_size;
 	struct list_head	node;
 	atomic_t		mapping_cnt;
 	u32			asid;
diff --git a/drivers/misc/habanalabs/common/memory.c b/drivers/misc/habanalabs/common/memory.c
index c7c27ffa6309..3f05bb398c70 100644
--- a/drivers/misc/habanalabs/common/memory.c
+++ b/drivers/misc/habanalabs/common/memory.c
@@ -1548,10 +1548,10 @@ static int set_dma_sg(struct scatterlist *sg, u64 bar_address, u64 chunk_size,
 }
 
 static struct sg_table *alloc_sgt_from_device_pages(struct hl_device *hdev, u64 *pages, u64 npages,
-						u64 page_size, struct device *dev,
-						enum dma_data_direction dir)
+						u64 page_size, u64 exported_size,
+						struct device *dev, enum dma_data_direction dir)
 {
-	u64 chunk_size, bar_address, dma_max_seg_size;
+	u64 chunk_size, bar_address, dma_max_seg_size, cur_size_to_export, cur_npages;
 	struct asic_fixed_properties *prop;
 	int rc, i, j, nents, cur_page;
 	struct scatterlist *sg;
@@ -1577,16 +1577,23 @@ static struct sg_table *alloc_sgt_from_device_pages(struct hl_device *hdev, u64
 	if (!sgt)
 		return ERR_PTR(-ENOMEM);
 
+	/* remove export size restrictions in case not explicitly defined */
+	cur_size_to_export = exported_size ? exported_size : (npages * page_size);
+
 	/* If the size of each page is larger than the dma max segment size,
 	 * then we can't combine pages and the number of entries in the SGL
 	 * will just be the
 	 * <number of pages> * <chunks of max segment size in each page>
 	 */
-	if (page_size > dma_max_seg_size)
-		nents = npages * DIV_ROUND_UP_ULL(page_size, dma_max_seg_size);
-	else
+	if (page_size > dma_max_seg_size) {
+		/* we should limit number of pages according to the exported size */
+		cur_npages = DIV_ROUND_UP_SECTOR_T(cur_size_to_export, page_size);
+		nents = cur_npages * DIV_ROUND_UP_SECTOR_T(page_size, dma_max_seg_size);
+	} else {
+		cur_npages = npages;
+
 		/* Get number of non-contiguous chunks */
-		for (i = 1, nents = 1, chunk_size = page_size ; i < npages ; i++) {
+		for (i = 1, nents = 1, chunk_size = page_size ; i < cur_npages ; i++) {
 			if (pages[i - 1] + page_size != pages[i] ||
 					chunk_size + page_size > dma_max_seg_size) {
 				nents++;
@@ -1596,6 +1603,7 @@ static struct sg_table *alloc_sgt_from_device_pages(struct hl_device *hdev, u64
 
 			chunk_size += page_size;
 		}
+	}
 
 	rc = sg_alloc_table(sgt, nents, GFP_KERNEL | __GFP_ZERO);
 	if (rc)
@@ -1618,7 +1626,8 @@ static struct sg_table *alloc_sgt_from_device_pages(struct hl_device *hdev, u64
 			else
 				cur_device_address += dma_max_seg_size;
 
-			chunk_size = min(size_left, dma_max_seg_size);
+			/* make sure not to export over exported size */
+			chunk_size = min3(size_left, dma_max_seg_size, cur_size_to_export);
 
 			bar_address = hdev->dram_pci_bar_start + cur_device_address;
 
@@ -1626,6 +1635,8 @@ static struct sg_table *alloc_sgt_from_device_pages(struct hl_device *hdev, u64
 			if (rc)
 				goto error_unmap;
 
+			cur_size_to_export -= chunk_size;
+
 			if (size_left > dma_max_seg_size) {
 				size_left -= dma_max_seg_size;
 			} else {
@@ -1637,7 +1648,7 @@ static struct sg_table *alloc_sgt_from_device_pages(struct hl_device *hdev, u64
 		/* Merge pages and put them into the scatterlist */
 		for_each_sgtable_dma_sg(sgt, sg, i) {
 			chunk_size = page_size;
-			for (j = cur_page + 1 ; j < npages ; j++) {
+			for (j = cur_page + 1 ; j < cur_npages ; j++) {
 				if (pages[j - 1] + page_size != pages[j] ||
 						chunk_size + page_size > dma_max_seg_size)
 					break;
@@ -1648,10 +1659,13 @@ static struct sg_table *alloc_sgt_from_device_pages(struct hl_device *hdev, u64
 			bar_address = hdev->dram_pci_bar_start +
 					(pages[cur_page] - prop->dram_base_address);
 
+			/* make sure not to export over exported size */
+			chunk_size = min(chunk_size, cur_size_to_export);
 			rc = set_dma_sg(sg, bar_address, chunk_size, dev, dir);
 			if (rc)
 				goto error_unmap;
 
+			cur_size_to_export -= chunk_size;
 			cur_page = j;
 		}
 	}
@@ -1722,6 +1736,7 @@ static struct sg_table *hl_map_dmabuf(struct dma_buf_attachment *attachment,
 						phys_pg_pack->pages,
 						phys_pg_pack->npages,
 						phys_pg_pack->page_size,
+						phys_pg_pack->exported_size,
 						attachment->dev,
 						dir);
 	else
@@ -1729,6 +1744,7 @@ static struct sg_table *hl_map_dmabuf(struct dma_buf_attachment *attachment,
 						&hl_dmabuf->device_address,
 						1,
 						hl_dmabuf->dmabuf->size,
+						0,
 						attachment->dev,
 						dir);
 
@@ -2033,6 +2049,7 @@ static int export_dmabuf_from_addr(struct hl_ctx *ctx, u64 addr, u64 size, u64 o
 		if (rc)
 			goto dec_memhash_export_cnt;
 
+		phys_pg_pack->exported_size = size;
 		hl_dmabuf->phys_pg_pack = phys_pg_pack;
 		hl_dmabuf->memhash_hnode = hnode;
 	} else {
-- 
2.25.1

[PATCH 23/26] habanalabs: fix handling of wait CS for interrupting signals

[Oded Gabbay]

From: Tomer Tayar <ttayar@habana.ai>

The -ERESTARTSYS return value is not handled correctly when a signal is
received while waiting for CS completion.
This can lead to bad output values to user when waiting for a single CS
completion, and more severe, it can cause a non-stopping loop when
waiting to multi-CS completion and until a CS timeout.

Fix the handling and exit the waiting if this return value is received.

Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 .../misc/habanalabs/common/command_submission.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/drivers/misc/habanalabs/common/command_submission.c b/drivers/misc/habanalabs/common/command_submission.c
index cf3b82efc65c..0ec8cdcbb1f5 100644
--- a/drivers/misc/habanalabs/common/command_submission.c
+++ b/drivers/misc/habanalabs/common/command_submission.c
@@ -2590,7 +2590,9 @@ static int hl_wait_for_fence(struct hl_ctx *ctx, u64 seq, struct hl_fence *fence
 		*status = CS_WAIT_STATUS_BUSY;
 	}
 
-	if (error == -ETIMEDOUT || error == -EIO)
+	if (completion_rc == -ERESTARTSYS)
+		rc = completion_rc;
+	else if (error == -ETIMEDOUT || error == -EIO)
 		rc = error;
 
 	return rc;
@@ -2849,6 +2851,9 @@ static int hl_wait_multi_cs_completion(struct multi_cs_data *mcs_data,
 	if (completion_rc > 0)
 		mcs_data->timestamp = mcs_compl->timestamp;
 
+	if (completion_rc == -ERESTARTSYS)
+		return completion_rc;
+
 	mcs_data->wait_status = completion_rc;
 
 	return 0;
@@ -2994,15 +2999,15 @@ static int hl_multi_cs_wait_ioctl(struct hl_fpriv *hpriv, void *data)
 free_seq_arr:
 	kfree(cs_seq_arr);
 
-	if (rc)
-		return rc;
-
-	if (mcs_data.wait_status == -ERESTARTSYS) {
+	if (rc == -ERESTARTSYS) {
 		dev_err_ratelimited(hdev->dev,
 				"user process got signal while waiting for Multi-CS\n");
-		return -EINTR;
+		rc = -EINTR;
 	}
 
+	if (rc)
+		return rc;
+
 	/* update output args */
 	memset(args, 0, sizeof(*args));
 
-- 
2.25.1

[PATCH 24/26] habanalabs: put fences in case of unexpected wait status

[Oded Gabbay]

From: Tomer Tayar <ttayar@habana.ai>

Need to put fences even if an unexpected status value is received while
waiting for a fence.

Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/command_submission.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/misc/habanalabs/common/command_submission.c b/drivers/misc/habanalabs/common/command_submission.c
index 0ec8cdcbb1f5..1543ef993f8e 100644
--- a/drivers/misc/habanalabs/common/command_submission.c
+++ b/drivers/misc/habanalabs/common/command_submission.c
@@ -2722,7 +2722,8 @@ static int hl_cs_poll_fences(struct multi_cs_data *mcs_data, struct multi_cs_com
 			break;
 		default:
 			dev_err(hdev->dev, "Invalid fence status\n");
-			return -EINVAL;
+			rc = -EINVAL;
+			break;
 		}
 
 	}
-- 
2.25.1

[PATCH 25/26] habanalabs/gaudi2: wait for preboot ready if HW state is dirty

[Oded Gabbay]

From: Ohad Sharabi <osharabi@habana.ai>

Instead of waiting for BTM indication we should wait for preboot ready.
Consider the below scenario:
    1. FW update is being triggered
           - setting the dirty bit
    2. hard reset will be triggered due to the dirty bit
    3. FW initiates the reset:
           - dirty bit cleared
           - BTM indication cleared
           - preboot ready indication cleared
    4. during hard reset:
           - BTM indication will be set
           - BIST test performed and another reset triggered
    5. only after this reset the preboot will set the preboot ready

When polling on BTM indication alone we can lose sync with FW while
trying to communicate with FW that is during reset.
To overcome this we will always wait to preboot ready indication.

Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/firmware_if.c |  2 +-
 drivers/misc/habanalabs/common/habanalabs.h  |  1 +
 drivers/misc/habanalabs/gaudi2/gaudi2.c      | 26 +++++++++++++++++++-
 3 files changed, 27 insertions(+), 2 deletions(-)

diff --git a/drivers/misc/habanalabs/common/firmware_if.c b/drivers/misc/habanalabs/common/firmware_if.c
index 537b1ae3fcb7..cda0bf3dbf1b 100644
--- a/drivers/misc/habanalabs/common/firmware_if.c
+++ b/drivers/misc/habanalabs/common/firmware_if.c
@@ -1352,7 +1352,7 @@ static void detect_cpu_boot_status(struct hl_device *hdev, u32 status)
 	}
 }
 
-static int hl_fw_wait_preboot_ready(struct hl_device *hdev)
+int hl_fw_wait_preboot_ready(struct hl_device *hdev)
 {
 	struct pre_fw_load_props *pre_fw_load = &hdev->fw_loader.pre_fw_load;
 	u32 status;
diff --git a/drivers/misc/habanalabs/common/habanalabs.h b/drivers/misc/habanalabs/common/habanalabs.h
index de715c91a87e..e5443bf7fe12 100644
--- a/drivers/misc/habanalabs/common/habanalabs.h
+++ b/drivers/misc/habanalabs/common/habanalabs.h
@@ -3745,6 +3745,7 @@ int hl_fw_cpucp_power_get(struct hl_device *hdev, u64 *power);
 void hl_fw_ask_hard_reset_without_linux(struct hl_device *hdev);
 void hl_fw_ask_halt_machine_without_linux(struct hl_device *hdev);
 int hl_fw_init_cpu(struct hl_device *hdev);
+int hl_fw_wait_preboot_ready(struct hl_device *hdev);
 int hl_fw_read_preboot_status(struct hl_device *hdev);
 int hl_fw_dynamic_send_protocol_cmd(struct hl_device *hdev,
 				struct fw_load_mgr *fw_loader,
diff --git a/drivers/misc/habanalabs/gaudi2/gaudi2.c b/drivers/misc/habanalabs/gaudi2/gaudi2.c
index ba3b0ae76ebf..5242b6f6bf95 100644
--- a/drivers/misc/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/misc/habanalabs/gaudi2/gaudi2.c
@@ -5484,7 +5484,31 @@ static void gaudi2_hw_fini(struct hl_device *hdev, bool hard_reset, bool fw_rese
 
 skip_reset:
 	if (driver_performs_reset || hard_reset)
-		gaudi2_poll_btm_indication(hdev, reset_sleep_ms, poll_timeout_us);
+		/*
+		 * Instead of waiting for BTM indication we should wait for preboot ready:
+		 * Consider the below scenario:
+		 * 1. FW update is being triggered
+		 *        - setting the dirty bit
+		 * 2. hard reset will be triggered due to the dirty bit
+		 * 3. FW initiates the reset:
+		 *        - dirty bit cleared
+		 *        - BTM indication cleared
+		 *        - preboot ready indication cleared
+		 * 4. during hard reset:
+		 *        - BTM indication will be set
+		 *        - BIST test performed and another reset triggered
+		 * 5. only after this reset the preboot will set the preboot ready
+		 *
+		 * when polling on BTM indication alone we can lose sync with FW while trying to
+		 * communicate with FW that is during reset.
+		 * to overcome this we will always wait to preboot ready indication
+		 */
+		if ((hdev->fw_components & FW_TYPE_PREBOOT_CPU)) {
+			msleep(reset_sleep_ms);
+			hl_fw_wait_preboot_ready(hdev);
+		} else {
+			gaudi2_poll_btm_indication(hdev, reset_sleep_ms, poll_timeout_us);
+		}
 	else
 		gaudi2_get_soft_rst_done_indication(hdev, poll_timeout_us);
 
-- 
2.25.1

[PATCH 26/26] habanalabs: fix wrong variable type used for vzalloc

[Oded Gabbay]

From: farah kassabri <fkassabri@habana.ai>

vzalloc expects void* and not void __iomem*.

Signed-off-by: farah kassabri <fkassabri@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/firmware_if.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/misc/habanalabs/common/firmware_if.c b/drivers/misc/habanalabs/common/firmware_if.c
index cda0bf3dbf1b..ee4d1c5ca527 100644
--- a/drivers/misc/habanalabs/common/firmware_if.c
+++ b/drivers/misc/habanalabs/common/firmware_if.c
@@ -2019,9 +2019,10 @@ static int hl_fw_dynamic_read_and_validate_descriptor(struct hl_device *hdev,
 						struct fw_load_mgr *fw_loader)
 {
 	struct lkd_fw_comms_desc *fw_desc;
-	void __iomem *src, *temp_fw_desc;
 	struct pci_mem_region *region;
 	struct fw_response *response;
+	void *temp_fw_desc;
+	void __iomem *src;
 	u16 fw_data_size;
 	enum pci_region region_id;
 	int rc;
-- 
2.25.1