Documentation
Download

aclnnMuls&aclnnInplaceMuls

📄 View source code

Supported Products

ProductSupported
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

  • Operator function: Performs multiplication computation between the tensor and scalar.
  • Formula:outi=selfi×otherout_i = self_i \times other

Prototype

  • aclnnMuls and aclnnInplaceMuls implement the same function in different ways. Select a proper operator based on your requirements.

    • aclnnMuls: An output tensor object needs to be created to store the computation result.
    • aclnnInplaceMuls: 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, aclnnMulsGetWorkspaceSize or aclnnInplaceMulsGetWorkspaceSize is called to obtain the workspace size required for computation and the executor that contains the operator computation process. Then, aclnnMuls or aclnnInplaceMuls is called to perform computation.

    • aclnnStatus aclnnMulsGetWorkspaceSize(const aclTensor *self, const aclScalar *other, aclTensor *out, uint64_t *workspaceSize, aclOpExecutor **executor)
    • aclnnStatus aclnnMuls(void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, aclrtStream stream)
    • aclnnStatus aclnnInplaceMulsGetWorkspaceSize(aclTensor *selfRef, const aclScalar *other, uint64_t *workspaceSize, aclOpExecutor **executor)
    • aclnnStatus aclnnInplaceMuls(void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, aclrtStream stream)

aclnnMulsGetWorkspaceSize

  • Parameters:

    • self (const aclTensor *, computation input): input self in the formula, which is an aclTensor on the device. The shape dimensions cannot be greater than 8. Non-contiguous tensors are supported. The data format can be ND.

      • Atlas inference products and Atlas training products: The data type can be FLOAT, FLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, BOOL, COMPLEX128, or COMPLEX64, and must meet the deduction relationship (see Deduction Relationship) with other.
      • Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products: The data type can be FLOAT, FLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, BOOL, COMPLEX128, COMPLEX64, or BFLOAT16, and must meet the deduction relationship (see Deduction Relationship) with other.

    • other (aclScalar *, computation input): input other in the formula, which is an aclScalar on the host.

      • Atlas inference products and Atlas training products: The data type can be FLOAT, FLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, BOOL, COMPLEX128, or COMPLEX64, and must meet the deduction relationship (see Deduction Relationship) with self.
      • Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products: The data type can be FLOAT, FLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, BOOL, COMPLEX128, COMPLEX64, or BFLOAT16, and must meet the deduction relationship (see Deduction Relationship) with self.

    • out (aclTensor *, computation output): output out in the formula, which is an aclTensor on the device. The shape is the same as that of the input self. Non-contiguous tensors are supported. The data format can be ND. The data type must be convertible from that after self and other are deduced. For details, see Conversion Relationships.

      • Atlas inference products and Atlas training products: The data type can be FLOAT, FLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, BOOL, COMPLEX128, or COMPLEX64.
      • Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products: The data type can be FLOAT, FLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, BOOL, COMPLEX128, COMPLEX64, or BFLOAT16.

    • workspaceSize (uint64_t *, output): size of the workspace to be allocated on the device.

    • executor (aclOpExecutor **, output): operator executor, covering the operator computation process.

  • Returns:

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

   The first-phase API implements input parameter verification. The following errors may be thrown.
 161001 (ACLNN_ERR_PARAM_NULLPTR): 1. self, other, or out is a null pointer.
 161002 (ACLNN_ERR_PARAM_INVALID): 1. The data types of self and out are not supported.
                                   2. Data type deduction cannot be performed for self and other.
                                   3. The deduced data type cannot be converted to the type of out.
                                   4. The shapes of self and out are inconsistent.
                                   5. The dimensions of self and out are greater than 8.

aclnnMuls

  • Parameters:

    • workspace (void *, input): address of the workspace to be allocated on the device.
    • workspaceSize (uint64_t, input): size of the workspace to be allocated on the device, which is obtained by calling the first-phase API aclnnMulsGetWorkspaceSize.
    • executor (aclOpExecutor *, input): operator executor, containing the operator computation process.
    • stream (aclrtStream, input): stream for executing the task.

  • Returns:

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

aclnnInplaceMulsGetWorkspaceSize

  • Parameters:

    • selfRef (aclTensor *, computation input | computation output): input and output tensor, that is, self and out in the formula. It is the aclTensor on the device. Non-contiguous tensors are supported. The data format can be ND.
      • Atlas inference products and Atlas training products: The data type can be FLOAT, FLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, BOOL, COMPLEX128, or COMPLEX64, and must meet the deduction relationship (see Deduction Relationship) with other.
      • Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products: The data type can be FLOAT, FLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, BOOL, COMPLEX128, COMPLEX64, or BFLOAT16, and must meet the deduction relationship (see Deduction Relationship) with other.
    • other (aclScalar *, computation input): input <idp:inline displayname="code" id="code17858186219">other in the formula, which is an aclScalar on the host.
      • Atlas inference products and Atlas training products: The data type can be FLOAT, FLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, BOOL, COMPLEX128, or COMPLEX64, and must meet the deduction relationship (see Deduction Relationship) with self.
      • Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products: The data type can be FLOAT, FLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, BOOL, COMPLEX128, COMPLEX64, or BFLOAT16, and must meet the deduction relationship (see Deduction Relationship) with self.
    • workspaceSize (uint64_t *, output): size of the workspace to be allocated on the device.
    • executor (aclOpExecutor **, output): operator executor, covering the operator computation process.

  • Returns:

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

   The first-phase API implements input parameter verification. The following errors may be thrown.
 161001 (ACLNN_ERR_PARAM_NULLPTR): 1. The input selfRef and other are null pointers.
 161002 (ACLNN_ERR_PARAM_INVALID): 1. The data type of selfRef is not supported.
                                   2. Data type deduction cannot be performed for self and other.
                                   3. The deduced data type cannot be converted to the type of self.

aclnnInplaceMuls

  • Parameters:

    • workspace (void *, input): address of the workspace to be allocated on the device.
    • workspaceSize (uint64_t, input): size of the workspace to be allocated on the device, which is obtained by calling the first-phase API aclnnInplaceMulsGetWorkspaceSize.
    • executor (aclOpExecutor *, input): operator executor, containing the operator computation process.
    • stream (aclrtStream, input): stream for executing the task.

  • Returns:

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

Constraints

  • Deterministic computing:

    • aclnnMuls&aclnnInplaceMuls defaults to a deterministic implementation.

    When other is of the float type, there is no precision loss. When other is of the int32 or int64 type, precision loss may occur if the value of other is greater than 2^24. In this case, aclnnMul is recommended.

Example

The following example is for reference only. For details, see Compilation and Running Sample.

aclnnMuls sample code:

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_mul.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) {
  // (Fixed writing) 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 memory on the device.
  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 the data on the host to the memory on the device.
  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);

  // Calculate the strides of consecutive tensors.
  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;
}

int main() {
  // 1. (Fixed writing) Initialize the device and stream. For details, see the ACL API manual.
    // Set deviceId based on the actual device.
  int32_t deviceId = 0;
  aclrtStream stream;
  auto ret = Init(deviceId, &stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);

  // 2. Construct inputs and outputs based on the API definition.
  std::vector<int64_t> selfShape = {4, 2};
  std::vector<int64_t> outShape = {4, 2};
  void* selfDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclScalar* other = nullptr;
  aclTensor* out = nullptr;
  std::vector<float> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7};
  float otherValue = 1.2f;
  std::vector<float> outHostData(8, 0);
  // Create a self aclTensor.
  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create an other aclScalar.
  other = aclCreateScalar(&otherValue, aclDataType::ACL_FLOAT);
  CHECK_RET(other != nullptr, return ret);
  // Create an out aclTensor.
  ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_FLOAT, &out);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // 3. Call the CANN operator library API, which needs to be replaced with the actual API.
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // Call the first-phase API of aclnnMuls.
  ret = aclnnMulsGetWorkspaceSize(self, other, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnMulsGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
  // Allocate device memory based on workspaceSize computed by the first-phase API.
  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);
  }
  // Call the second-phase API of aclnnMuls.
  ret = aclnnMuls(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnMuls failed. ERROR: %d\n", ret); return ret);

  // 4. (Fixed writing) Wait until the task execution is complete.
  ret = aclrtSynchronizeStream(stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);

  // 5. Obtain the output value and copy the result from the device to the host. Modify the code based on the API definition.
  auto size = GetShapeSize(outShape);
  std::vector<float> resultData(size, 0);
  ret = aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), outDeviceAddr,
                    size * sizeof(resultData[0]), 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: %f\n", i, resultData[i]);
  }

  // 6. Release aclTensor and aclScalar. Modify the configuration based on the API definition.
  aclDestroyTensor(self);
  aclDestroyScalar(other);
  aclDestroyTensor(out);

  // 7. Release device resources.
  aclrtFree(selfDeviceAddr);
  aclrtFree(outDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();

  return 0;
}

aclnnInplaceMuls sample code:

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_mul.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) {
  // (Fixed writing) 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 memory on the device.
  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 the data on the host to the memory on the device.
  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);

  // Calculate the strides of consecutive tensors.
  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;
}

int main() {
  // 1. (Fixed writing) Initialize the device and stream. For details, see the ACL API manual.
    // Set deviceId based on the actual device.
  int32_t deviceId = 0;
  aclrtStream stream;
  auto ret = Init(deviceId, &stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);

  // 2. Construct the input based on the API.
  std::vector<int64_t> selfRefShape = {4, 2};
  std::vector<int64_t> outShape = {4, 2};
  void* selfRefDeviceAddr = nullptr;
  aclTensor* selfRef = nullptr;
  aclScalar* other = nullptr;
  std::vector<float> selfRefHostData = {0, 1, 2, 3, 4, 5, 6, 7};
  float otherValue = 1.2f;
  // Create a selfRef aclTensor.
  ret = CreateAclTensor(selfRefHostData, selfRefShape, &selfRefDeviceAddr, aclDataType::ACL_FLOAT, &selfRef);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create an other aclScalar.
  other = aclCreateScalar(&otherValue, aclDataType::ACL_FLOAT);
  CHECK_RET(other != nullptr, return ret);

  // 3. Call the CANN operator library API, which needs to be replaced with the actual API.
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // Call the first-phase API of aclnnInplaceMuls.
  ret = aclnnInplaceMulsGetWorkspaceSize(selfRef, other, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceMulsGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
  // Allocate device memory based on workspaceSize computed by the first-phase API.
  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);
  }
  // Call the second-phase API of aclnnInplaceMuls.
  ret = aclnnInplaceMuls(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceMuls failed. ERROR: %d\n", ret); return ret);

  // 4. (Fixed writing) Wait until the task execution is complete.
  ret = aclrtSynchronizeStream(stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);

  // 5. Obtain the output value and copy the result from the device to the host. Modify the code based on the API definition.
  auto size = GetShapeSize(selfRefShape);
  std::vector<float> resultData(size, 0);
  ret = aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), selfRefDeviceAddr,
                    size * sizeof(resultData[0]), 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: %f\n", i, resultData[i]);
  }

  // 6. Release aclTensor and aclScalar. Modify the configuration based on the API definition.
  aclDestroyTensor(selfRef);
  aclDestroyScalar(other);

  // 7. Release device resources. Modify the configuration based on the API definition.
  aclrtFree(selfRefDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();
  return 0;
}
This site uses cookies to improve your browsing experience. By continuing to browse this site, you agree to our use of cookies.View Details