habanalabs: expose undefined opcode status via info ioctl

The info ioctl retrieves information on the last undefined opcode
occurred.
Signed-off-by: Tal Cohen <talcohen@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>

drivers/misc/habanalabs/common/habanalabs_ioctl.c

@@ -610,6 +610,28 @@ static int razwi_info(struct hl_fpriv *hpriv, struct hl_info_args *args)
 	return copy_to_user(out, &info, min_t(size_t, max_size, sizeof(info))) ? -EFAULT : 0;
 }
 
+static int undefined_opcode_info(struct hl_fpriv *hpriv, struct hl_info_args *args)
+{
+	struct hl_device *hdev = hpriv->hdev;
+	u32 max_size = args->return_size;
+	struct hl_info_undefined_opcode_event info = {0};
+	void __user *out = (void __user *) (uintptr_t) args->return_pointer;
+
+	if ((!max_size) || (!out))
+		return -EINVAL;
+
+	info.timestamp = ktime_to_ns(hdev->last_error.undef_opcode.timestamp);
+	info.engine_id = hdev->last_error.undef_opcode.engine_id;
+	info.cq_addr = hdev->last_error.undef_opcode.cq_addr;
+	info.cq_size = hdev->last_error.undef_opcode.cq_size;
+	info.stream_id = hdev->last_error.undef_opcode.stream_id;
+	info.cb_addr_streams_len = hdev->last_error.undef_opcode.cb_addr_streams_len;
+	memcpy(info.cb_addr_streams, hdev->last_error.undef_opcode.cb_addr_streams,
+			sizeof(info.cb_addr_streams));
+
+	return copy_to_user(out, &info, min_t(size_t, max_size, sizeof(info))) ? -EFAULT : 0;
+}
+
 static int dev_mem_alloc_page_sizes_info(struct hl_fpriv *hpriv, struct hl_info_args *args)
 {
 	void __user *out = (void __user *) (uintptr_t) args->return_pointer;
@@ -718,6 +740,9 @@ static int _hl_info_ioctl(struct hl_fpriv *hpriv, void *data,
 	case HL_INFO_RAZWI_EVENT:
 		return razwi_info(hpriv, args);
 
+	case HL_INFO_UNDEFINED_OPCODE_EVENT:
+		return undefined_opcode_info(hpriv, args);
+
 	case HL_INFO_DEV_MEM_ALLOC_PAGE_SIZES:
 		return dev_mem_alloc_page_sizes_info(hpriv, args);
 

include/uapi/misc/habanalabs.h

@@ -352,6 +352,7 @@ enum hl_server_type {
  * HL_INFO_REGISTER_EVENTFD   - Register eventfd for event notifications.
  * HL_INFO_UNREGISTER_EVENTFD - Unregister eventfd
  * HL_INFO_GET_EVENTS         - Retrieve the last occurred events
+ * HL_INFO_UNDEFINED_OPCODE_EVENT - Retrieve last undefined opcode error information.
  */
 #define HL_INFO_HW_IP_INFO			0
 #define HL_INFO_HW_EVENTS			1
@@ -380,6 +381,7 @@ enum hl_server_type {
 #define HL_INFO_REGISTER_EVENTFD		28
 #define HL_INFO_UNREGISTER_EVENTFD		29
 #define HL_INFO_GET_EVENTS			30
+#define HL_INFO_UNDEFINED_OPCODE_EVENT		31
 
 #define HL_INFO_VERSION_MAX_LEN			128
 #define HL_INFO_CARD_NAME_MAX_LEN		16
@@ -656,6 +658,34 @@ struct hl_info_razwi_event {
 	__u8 pad[2];
 };
 
+#define MAX_QMAN_STREAMS_INFO		4
+#define OPCODE_INFO_MAX_ADDR_SIZE	8
+/**
+ * struct hl_info_undefined_opcode_event - info about last undefined opcode error
+ * @timestamp: timestamp of the undefined opcode error
+ * @cb_addr_streams: CB addresses (per stream) that are currently exists in the PQ
+ *                   entiers. In case all streams array entries are
+ *                   filled with values, it means the execution was in Lower-CP.
+ * @cq_addr: the address of the current handled command buffer
+ * @cq_size: the size of the current handled command buffer
+ * @cb_addr_streams_len: num of streams - actual len of cb_addr_streams array.
+ *                       should be equal to 1 incase of undefined opcode
+ *                       in Upper-CP (specific stream) and equal to 4 incase
+ *                       of undefined opcode in Lower-CP.
+ * @engine_id: engine-id that the error occurred on
+ * @stream_id: the stream id the error occurred on. In case the stream equals to
+ *             MAX_QMAN_STREAMS_INFO it means the error occurred on a Lower-CP.
+ */
+struct hl_info_undefined_opcode_event {
+	__s64 timestamp;
+	__u64 cb_addr_streams[MAX_QMAN_STREAMS_INFO][OPCODE_INFO_MAX_ADDR_SIZE];
+	__u64 cq_addr;
+	__u32 cq_size;
+	__u32 cb_addr_streams_len;
+	__u32 engine_id;
+	__u32 stream_id;
+};
+
 /**
  * struct hl_info_dev_memalloc_page_sizes - valid page sizes in device mem alloc information.
  * @page_order_bitmask: bitmap in which a set bit represents the order of the supported page size