aclnnEqTensor&aclnnInplaceEqTensor

Supported Products

Product	Supported
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

Description: Compares elements of two tensors element-wise; returns True(1.) if an element in self equals the corresponding element in other, and False(0.) otherwise.
Formula: $out = (self_i == other_i) ? [True] : [False]$

Prototype

aclnnEqTensor and aclnnInplaceEqTensor implement the same function in different ways. Select a proper operator based on your requirements.
- aclnnEqTensor: An output tensor object needs to be created to store the computation result.
- aclnnInplaceEqTensor: No output tensor object needs to be created, and the computation result is stored in the memory of the input tensor.
Each operator has two-phase API calls. First, aclnnEqTensorGetWorkspaceSize or aclnnInplaceEqTensorGetWorkspaceSize is called to obtain the workspace size required for computation and the executor that contains the operator computation process. Then, aclnnEqTensor or aclnnInplaceEqTensor is called to perform computation.

aclnnStatus aclnnEqTensorGetWorkspaceSize(
  const aclTensor *self, 
  const aclTensor *other, 
  aclTensor       *out, 
  uint64_t        *workspaceSize, 
  aclOpExecutor  **executor)

aclnnStatus aclnnEqTensor(
  void          *workspace, 
  uint64_t       workspaceSize, 
  aclOpExecutor *executor, 
  aclrtStream    stream)

aclnnStatus aclnnInplaceEqTensorGetWorkspaceSize(
  const aclTensor *selfRef, 
  const aclTensor *other, 
  uint64_t        *workspaceSize, 
  aclOpExecutor  **executor)

aclnnStatus aclnnInplaceEqTensor(
  void          *workspace, 
  uint64_t       workspaceSize, 
  aclOpExecutor *executor, 
  aclrtStream    stream)

aclnnEqTensorGetWorkspaceSize

Parameters

Name	Input/Output	Description	Usage Notes	Data Type	Data Format	Dimension (Shape)	Non-contiguous Tensor
self	Input	Input tensor.	The data type must meet the data type deduction rules (see Deduction Relationship) with `other`. The shape must be broadcastable with that of `other` (see Broadcast Relationship).	DOUBLE, FLOAT16, FLOAT, BFLOAT16, INT64, INT32, INT8, UINT8, BOOL, INT16, COMPLEX64, COMPLEX128, UINT32, UINT64	ND	Must not exceed 8D.	√
other	Input	Input tensor.	The data type must meet the data type deduction rules (see Deduction Relationship) with `self`. The shape must be broadcastable with that of `self` (see Broadcast Relationship).	DOUBLE, FLOAT16, FLOAT, BFLOAT16, INT64, INT32, INT8, UINT8, BOOL, INT16, COMPLEX64, COMPLEX128, UINT32, UINT64	ND	Must not exceed 8D.	√
out	Output	Output tensor.	The data type must be convertible from BOOL (see Conversion Relationship). The shape must match the broadcast result of `self` and `other` (see Broadcast Relationship).	DOUBLE, FLOAT16, FLOAT, BFLOAT16, INT64, INT32, INT8, UINT8, BOOL, INT16, COMPLEX64, COMPLEX128, UINT64, UINT32, UINT16	ND	Must not exceed 8D.	√
workspaceSize	Output	Size of the workspace required to be allocated on the device.	-	-	-	-	-
executor	Output	Operator executor, containing the operator computation process.	-	-	-	-	-

Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products:
- The data types of selfRef and other cannot be UINT32 or UINT64.
- The data type of out cannot be UINT64, UINT32, or UINT16.
Atlas training products:
- The data type cannot be BFLOAT16.
- The data type of out cannot be BFLOAT16, UINT64, UINT32, or UINT16.

Returns

aclnnStatus: status code. For details, see aclnn Return Codes.

The first-phase API implements input parameter validation. The following error codes may be returned.

Return	Error Code	Description
ACLNN_ERR_PARAM_NULLPTR	161001	The passed `self`, `other`, or `out` is a null pointer.
ACLNN_ERR_PARAM_INVALID	161002	The passed `self`, `other`, or `out` is a null pointer.
		The number of dimensions of `self`, `other`, or `out` exceeds 8.
		Data type deduction cannot be performed for `self` and `other`.
		The shapes of `self` and `other` are not broadcastable.
		The shape of out is inconsistent with that of the broadcast result.

aclnnEqTensor

Parameters

Name	Input/Output	Description
workspace	Input	Address of the workspace to be allocated on the device.
workspaceSize	Input	Size of the workspace to be allocated on the device, which is obtained by calling the first-phase API `aclnnEqTensorGetWorkspaceSize`.
executor	Input	Operator executor, containing the operator computation process.
stream	Input	Stream for executing the task.

Returns

aclnnStatus: status code. For details, see aclnn Return Codes.

aclnnInplaceEqTensorGetWorkspaceSize

Parameters

Name	Input/Output	Description	Usage Notes	Data Type	Data Format	Dimension (Shape)	Non-contiguous Tensor
selfRef	Input/Output	Input/output tensor, `self` and `out` in the formula.	The data type must meet the data type deduction rules (see Deduction Relationship) with `other`. The shape must be broadcastable with that of `other` (see Broadcast Relationship).	DOUBLE, FLOAT16, FLOAT, BFLOAT16, INT64, INT32, INT8, UINT8, BOOL, INT16, COMPLEX64, COMPLEX128, UINT32, UINT64	ND	-	√
other	Input	`other` in the formula.	The data type must meet the data type deduction rules (see Deduction Relationship) with `selfRef`. The shape must be broadcastable with that of `selfRef` (see Broadcast Relationship).	DOUBLE, FLOAT16, FLOAT, BFLOAT16, INT64, INT32, INT8, UINT8, BOOL, INT16, COMPLEX64, COMPLEX128, UINT32, UINT64	ND	-	√
workspaceSize	Output	Size of the workspace required to be allocated on the device.	-	-	-	-	-
executor	Output	Operator executor, containing the operator computation process.	-	-	-	-	-

Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products:
- The data types of selfRef and other cannot be UINT32 or UINT64.
Atlas training products:
- The data type cannot be BFLOAT16.

Returns

aclnnStatus: status code. For details, see aclnn Return Codes.

The first-phase API implements input parameter validation. The following error codes may be returned.

Return	Error Code	Description
ACLNN_ERR_PARAM_NULLPTR	161001	The passed `selfRef` or `other` is a null pointer.
ACLNN_ERR_PARAM_INVALID	161002	The data type of `selfRef` or `other` is not supported.
		Data type deduction cannot be performed for `selfRef` and `other`.
		The shapes of `selfRef` and `other` are not broadcastable.
		The shape of `selfRef` is inconsistent with that of the broadcast result of `selfRef` and `other`.
		The number of dimensions of `selfRef` and `other` exceeds 8.
		Exactly one of `selfRef` or `other` is an empty tensor.

aclnnInplaceEqTensor

Parameters

Name	Input/Output	Description
workspace	Input	Address of the workspace to be allocated on the device.
workspaceSize	Input	Size of the workspace to be allocated on the device, which is obtained by calling the first-phase API `aclnnInplaceEqTensorGetWorkspaceSize`.
executor	Input	Operator executor, containing the operator computation process.
stream	Input	Stream for executing the task.

Returns

aclnnStatus: status code. For details, see aclnn Return Codes.

Constraints

Deterministic computation:
- aclnnEqTensor and aclnnInplaceEqTensor default to a deterministic implementation.

Examples

The following examples are for reference only. For details, see Compilation and Running Sample.

aclnnEqTensor Sample Code

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_eq_tensor.h"

#define CHECK_RET(cond, return_expr) \
  do {                               \
    if (!(cond)) {                   \
      return_expr;                   \
    }                                \
  } while (0)

#define LOG_PRINT(message, ...)     \
  do {                              \
    printf(message, ##__VA_ARGS__); \
  } while (0)

int64_t GetShapeSize(const std::vector<int64_t>& shape)
{
  int64_t shape_size = 1;
  for (auto i : shape) {
    shape_size *= i;
  }
  return shape_size;
}

int Init(int32_t deviceId, aclrtStream* stream)
{
  // (Boilerplate) Initialize resources.
  auto ret = aclInit(nullptr);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclInit failed. ERROR: %d\n", ret); return ret);
  ret = aclrtSetDevice(deviceId);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSetDevice failed. ERROR: %d\n", ret); return ret);
  ret = aclrtCreateStream(stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtCreateStream failed. ERROR: %d\n", ret); return ret);
  return 0;
}

template <typename T>
int CreateAclTensor(
    const std::vector<T>& hostData, const std::vector<int64_t>& shape, void** deviceAddr, aclDataType dataType,
    aclTensor** tensor)
{
  auto size = GetShapeSize(shape) * sizeof(T);
  // Call aclrtMalloc to allocate device memory.
  auto ret = aclrtMalloc(deviceAddr, size, ACL_MEM_MALLOC_HUGE_FIRST);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtMalloc failed. ERROR: %d\n", ret); return ret);

  // Call aclrtMemcpy to copy host data to the device memory.
  ret = aclrtMemcpy(*deviceAddr, size, hostData.data(), size, ACL_MEMCPY_HOST_TO_DEVICE);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtMemcpy failed. ERROR: %d\n", ret); return ret);

  // Compute strides of the contiguous tensor.
  std::vector<int64_t> strides(shape.size(), 1);
  for (int64_t i = shape.size() - 2; i >= 0; i--) {
    strides[i] = shape[i + 1] * strides[i + 1];
  }

  // Call aclCreateTensor to create an aclTensor.
  *tensor = aclCreateTensor(
      shape.data(), shape.size(), dataType, strides.data(), 0, aclFormat::ACL_FORMAT_ND, shape.data(), shape.size(),
      *deviceAddr);
  return 0;
}

aclError InitAcl(int32_t deviceId, aclrtStream* stream)
{
  auto ret = Init(deviceId, stream);
  CHECK_RET(ret == 0, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);
  return ACL_SUCCESS;
}

aclError CreateInputs(
    std::vector<int64_t>& selfShape, std::vector<int64_t>& otherShape, std::vector<int64_t>& outShape,
    void** selfDeviceAddr, void** otherDeviceAddr, void** outDeviceAddr, aclTensor** self, aclTensor** other,
    aclTensor** out)
{
  std::vector<double> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7};
  std::vector<double> otherHostData = {0, 1, 9, 9, 9, 9, 9, 9};
  std::vector<char> outHostData = {0, 0, 0, 0, 0, 0, 0, 0};

  auto ret = CreateAclTensor(selfHostData, selfShape, selfDeviceAddr, aclDataType::ACL_DOUBLE, self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  ret = CreateAclTensor(otherHostData, otherShape, otherDeviceAddr, aclDataType::ACL_DOUBLE, other);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  ret = CreateAclTensor(outHostData, outShape, outDeviceAddr, aclDataType::ACL_BOOL, out);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  return ACL_SUCCESS;
}

aclError ExecOpApi(
    aclTensor* self, aclTensor* other, aclTensor* out, void** workspaceAddrOut, uint64_t& workspaceSize,
    void* outDeviceAddr, std::vector<int64_t>& outShape, aclrtStream stream)
{
  aclOpExecutor* executor;

  auto ret = aclnnEqTensorGetWorkspaceSize(self, other, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnEqTensorGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);

  void* workspaceAddr = nullptr;
  if (workspaceSize > 0) {
    ret = aclrtMalloc(&workspaceAddr, workspaceSize, ACL_MEM_MALLOC_HUGE_FIRST);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("allocate workspace failed. ERROR: %d\n", ret); return ret);
  }
  *workspaceAddrOut = workspaceAddr;

  ret = aclnnEqTensor(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnEqTensor failed. ERROR: %d\n", ret); return ret);

  ret = aclrtSynchronizeStream(stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);

  auto size = GetShapeSize(outShape);
  std::vector<char> resultData(size, 0);

  ret = aclrtMemcpy(
      resultData.data(), resultData.size() * sizeof(resultData[0]), outDeviceAddr, size * sizeof(char),
      ACL_MEMCPY_DEVICE_TO_HOST);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("copy result from device to host failed. ERROR: %d\n", ret); return ret);

  for (int64_t i = 0; i < size; i++) {
    LOG_PRINT("result[%ld] is: %d\n", i, resultData[i]);
  }

  return ACL_SUCCESS;
}

int main()
{
  int32_t deviceId = 0;
  aclrtStream stream;

  auto ret = InitAcl(deviceId, &stream);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  std::vector<int64_t> selfShape = {4, 2};
  std::vector<int64_t> otherShape = {4, 2};
  std::vector<int64_t> outShape = {4, 2};

  void* selfDeviceAddr = nullptr;
  void* otherDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;

  aclTensor* self = nullptr;
  aclTensor* other = nullptr;
  aclTensor* out = nullptr;

  ret = CreateInputs(
      selfShape, otherShape, outShape, &selfDeviceAddr, &otherDeviceAddr, &outDeviceAddr, &self, &other, &out);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  uint64_t workspaceSize = 0;
  void* workspaceAddr = nullptr;

  ret = ExecOpApi(self, other, out, &workspaceAddr, workspaceSize, outDeviceAddr, outShape, stream);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // Release resources.
  aclDestroyTensor(self);
  aclDestroyTensor(other);
  aclDestroyTensor(out);

  aclrtFree(selfDeviceAddr);
  aclrtFree(otherDeviceAddr);
  aclrtFree(outDeviceAddr);

  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }

  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();
  return 0;
}

aclnnInplaceEqTensor Sample Code

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_eq_tensor.h"

#define CHECK_RET(cond, return_expr) \
  do {                               \
    if (!(cond)) {                   \
      return_expr;                   \
    }                                \
  } while (0)

#define LOG_PRINT(message, ...)     \
  do {                              \
    printf(message, ##__VA_ARGS__); \
  } while (0)

int64_t GetShapeSize(const std::vector<int64_t>& shape)
{
  int64_t shapeSize = 1;
  for (auto i : shape) {
    shapeSize *= i;
  }
  return shapeSize;
}

int Init(int32_t deviceId, aclrtStream* stream)
{
  // (Boilerplate) Initialize resources.
  auto ret = aclInit(nullptr);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclInit failed. ERROR: %d\n", ret); return ret);
  ret = aclrtSetDevice(deviceId);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSetDevice failed. ERROR: %d\n", ret); return ret);
  ret = aclrtCreateStream(stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtCreateStream failed. ERROR: %d\n", ret); return ret);
  return 0;
}

template <typename T>
int CreateAclTensor(
    const std::vector<T>& hostData, const std::vector<int64_t>& shape, void** deviceAddr, aclDataType dataType,
    aclTensor** tensor)
{
  auto size = GetShapeSize(shape) * sizeof(T);
  // Call aclrtMalloc to allocate device memory.
  auto ret = aclrtMalloc(deviceAddr, size, ACL_MEM_MALLOC_HUGE_FIRST);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtMalloc failed. ERROR: %d\n", ret); return ret);
  // Call aclrtMemcpy to copy host data to the device memory.
  ret = aclrtMemcpy(*deviceAddr, size, hostData.data(), size, ACL_MEMCPY_HOST_TO_DEVICE);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtMemcpy failed. ERROR: %d\n", ret); return ret);

  // Compute strides of the contiguous tensor.
  std::vector<int64_t> strides(shape.size(), 1);
  for (int64_t i = shape.size() - 2; i >= 0; i--) {
    strides[i] = shape[i + 1] * strides[i + 1];
  }

  // Call aclCreateTensor to create an aclTensor.
  *tensor = aclCreateTensor(
      shape.data(), shape.size(), dataType, strides.data(), 0, aclFormat::ACL_FORMAT_ND, shape.data(), shape.size(),
      *deviceAddr);
  return 0;
}

aclError InitAcl(int32_t deviceId, aclrtStream* stream)
{
  auto ret = Init(deviceId, stream);
  CHECK_RET(ret == 0, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);
  return ACL_SUCCESS;
}

aclError CreateInputs(
    std::vector<int64_t>& selfShape, std::vector<int64_t>& otherShape, void** selfDeviceAddr, void** otherDeviceAddr,
    aclTensor** self, aclTensor** other)
{
  std::vector<double> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7};
  std::vector<double> otherHostData = {1, 1, 3, 3, 5, 5, 7, 7};

  //Create a self tensor.
  auto ret = CreateAclTensor(selfHostData, selfShape, selfDeviceAddr, aclDataType::ACL_DOUBLE, self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  //Create an other tensor.
  ret = CreateAclTensor(otherHostData, otherShape, otherDeviceAddr, aclDataType::ACL_DOUBLE, other);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  return ACL_SUCCESS;
}

aclError ExecOpApi(
    aclTensor* self, aclTensor* other, void** workspaceAddrOut, uint64_t& workspaceSize, void* selfDeviceAddr,
    std::vector<int64_t>& selfShape, aclrtStream stream)
{
  aclOpExecutor* executor;

  // Obtain the workspace size.
  auto ret = aclnnInplaceEqTensorGetWorkspaceSize(self, other, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceEqTensorGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);

  // Allocate workspace.
  void* workspaceAddr = nullptr;
  if (workspaceSize > 0) {
    ret = aclrtMalloc(&workspaceAddr, workspaceSize, ACL_MEM_MALLOC_HUGE_FIRST);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("allocate workspace failed. ERROR: %d\n", ret); return ret);
  }
  *workspaceAddrOut = workspaceAddr;

  // Execute the operator.
  ret = aclnnInplaceEqTensor(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceEqTensor failed. ERROR: %d\n", ret); return ret);

  // Synchronize.
  ret = aclrtSynchronizeStream(stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);

  // Copy the output.
  auto size = GetShapeSize(selfShape);
  std::vector<double> resultData(size);

  ret = aclrtMemcpy(
      resultData.data(), resultData.size() * sizeof(resultData[0]), selfDeviceAddr, size * sizeof(double),
      ACL_MEMCPY_DEVICE_TO_HOST);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("copy result from device to host failed. ERROR: %d\n", ret); return ret);

  for (int64_t i = 0; i < size; i++) {
    LOG_PRINT("result[%ld] is: %lf\n", i, resultData[i]);
  }

  return ACL_SUCCESS;
}

int main()
{
  int32_t deviceId = 0;
  aclrtStream stream;

  auto ret = InitAcl(deviceId, &stream);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  std::vector<int64_t> selfShape = {4, 2};
  std::vector<int64_t> otherShape = {4, 2};
  void* selfDeviceAddr = nullptr;
  void* otherDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclTensor* other = nullptr;

  ret = CreateInputs(selfShape, otherShape, &selfDeviceAddr, &otherDeviceAddr, &self, &other);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  uint64_t workspaceSize = 0;
  void* workspaceAddr = nullptr;

  ret = ExecOpApi(self, other, &workspaceAddr, workspaceSize, selfDeviceAddr, selfShape, stream);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // Destroy.
  aclDestroyTensor(self);
  aclDestroyTensor(other);

  // Release.
  aclrtFree(selfDeviceAddr);
  aclrtFree(otherDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }

  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();
  return 0;
}