aclrtGetErrorVerbose

Note: This API is for trial use and may be changed in later versions. It is not available in production environments.

Applicable Products

Product	Supported
Ascend 950PR/Ascend 950DT	√
Atlas A3 training products/Atlas A3 inference products	√
Atlas A2 training products/Atlas A2 inference products	√
Atlas 200I/500 A2 inference products	☓
Atlas inference products	☓
Atlas training products	☓

Function

Obtains the detailed error information after a device fault occurs. This API must be called after a fault event is obtained and before task submission is terminated.

Prototype

aclError aclrtGetErrorVerbose(int32_t deviceId, aclrtErrorInfo *errorInfo);

Parameters

Parameter	Input/Output	Description
deviceId	Input	Device ID. The value must be the same as the device ID in aclrtSetDevice.
errorInfo	Output	Error message. typedef enum { ACL_RT_NO_ERROR = 0, // No error ACL_RT_ERROR_MEMORY = 1, // Memory error ACL_RT_ERROR_L2 = 2, // L2 buffer error ACL_RT_ERROR_AICORE = 3, // AI Core error ACL_RT_ERROR_LINK = 4, // Network error ACL_RT_ERROR_L3_PORT = 5, // L3 port error ACL_RT_ERROR_OTHERS = 0xFFFF, // Other errors } aclrtErrorType; typedef enum aclrtAicoreErrorType { ACL_RT_AICORE_ERROR_UNKNOWN, // Unknown error ACL_RT_AICORE_ERROR_SW, // Software check recommended ACL_RT_AICORE_ERROR_HW_LOCAL, // Device hardware check recommended } aclrtAicoreErrorType; #define ACL_RT_MEM_UCE_INFO_MAX_NUM 20 typedef struct { size_t arraySize; // Size of the memUceInfoArray array aclrtMemUceInfo memUceInfoArray[ACL_RT_MEM_UCE_INFO_MAX_NUM]; // Array of virtual addresses where a memory UCE occurs } aclrtMemUceInfoArray; typedef union aclrtErrorInfoDetail { aclrtMemUceInfoArray uceInfo; // Memory UCE information aclrtAicoreErrorType aicoreErrType; // AI Core error } aclrtErrorInfoDetail; typedef struct aclrtErrorInfo { uint8_t tryRepair; // Whether repairing is required. 0: no; 1: yes uint8_t hasDetail; // Whether error details are available. 0: no; 1: yes uint8_t reserved[2]; // Reserved parameter aclrtErrorType errorType; // Error type aclrtErrorInfoDetail detail; // Error details } aclrtErrorInfo;

Parameter

Input/Output

Description

deviceId

Input

Device ID.

The value must be the same as the device ID in aclrtSetDevice.

errorInfo

Output

Error message.

typedef enum { 
    ACL_RT_NO_ERROR = 0,          // No error
    ACL_RT_ERROR_MEMORY = 1,       // Memory error
    ACL_RT_ERROR_L2 = 2,          // L2 buffer error
    ACL_RT_ERROR_AICORE = 3,      // AI Core error
    ACL_RT_ERROR_LINK = 4,        // Network error
    ACL_RT_ERROR_L3_PORT = 5,     // L3 port error
    ACL_RT_ERROR_OTHERS = 0xFFFF, // Other errors
} aclrtErrorType;

typedef enum aclrtAicoreErrorType { 
    ACL_RT_AICORE_ERROR_UNKNOWN,   // Unknown error
    ACL_RT_AICORE_ERROR_SW,       // Software check recommended
    ACL_RT_AICORE_ERROR_HW_LOCAL, // Device hardware check recommended
} aclrtAicoreErrorType;

#define ACL_RT_MEM_UCE_INFO_MAX_NUM 20
typedef struct {
    size_t arraySize; // Size of the memUceInfoArray array
    aclrtMemUceInfo memUceInfoArray[ACL_RT_MEM_UCE_INFO_MAX_NUM];  // Array of virtual addresses where a memory UCE occurs
} aclrtMemUceInfoArray;

typedef union aclrtErrorInfoDetail { 
aclrtMemUceInfoArray uceInfo;        //  Memory UCE information
    aclrtAicoreErrorType aicoreErrType; // AI Core error
} aclrtErrorInfoDetail;

typedef struct aclrtErrorInfo { 
    uint8_t tryRepair;           // Whether repairing is required. 0: no; 1: yes
    uint8_t hasDetail;           // Whether error details are available. 0: no; 1: yes
    uint8_t reserved[2];         // Reserved parameter
    aclrtErrorType errorType;    // Error type
    aclrtErrorInfoDetail detail; // Error details
} aclrtErrorInfo;

Returns

0 on success; otherwise, failure. For details, see aclError.

Restrictions

If the minimum timeout interval queried by calling aclrtGetOpTimeoutInterval is less than 100 ms, the detailed fault information returned by this API may be inaccurate.
For the Atlas A3 training products/Atlas A3 inference products and Atlas A2 training products/Atlas A2 inference products, only ACL_RT_NO_ERROR (no error), ACL_RT_ERROR_MEMORY (memory error), and ACL_RT_ERROR_OTHERS (other errors) can be obtained.

Parent topic: Exception Handling