aclnnFloorDivides&aclnnInplaceFloorDivides

📄 View source code

Supported Products

Function

• Description: Performs division and rounds down the remainder.
• Formula:$$out_i = floor(\frac{input_i}{other})$$

Prototype

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

  • aclnnFloorDivides: An output tensor object needs to be created to store the computation result.
  • aclnnInplaceFloorDivides: 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, aclnnFloorDividesGetWorkspaceSize or aclnnInplaceFloorDividesGetWorkspaceSize is called to obtain input parameters and calculate the required workspace size based on the computation process. Then, aclnnFloorDivides or aclnnInplaceFloorDivides is called to perform computation.

• aclnnStatus aclnnFloorDividesGetWorkspaceSize(const aclTensor *self, const aclScalar *other, aclTensor *out, uint64_t *workspaceSize, aclOpExecutor **executor)

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

• aclnnStatus aclnnInplaceFloorDividesGetWorkspaceSize(aclTensor *selfRef, const aclScalar *other, uint64_t *workspaceSize, aclOpExecutor **executor)

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

aclnnFloorDividesGetWorkspaceSize

• Parameters:

  • self (aclTensor*, compute input): aclTensor on the device. The shape must meet the broadcast relationship with other. Non-contiguous tensors are supported. The data format can be ND.
    • 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, INT8, UINT8, INT16, BOOL, or BFLOAT16, and must meet the data type deduction rules (see Deduction Relationship) with other.
    • Atlas training products: The data type can be FLOAT, FLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, or BOOL, and must meet the data type deduction rules (see Deduction Relationship) with other.
  • other (aclScalar*, compute input): aclScalar on the host.
    • 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, INT8, UINT8, INT16, BOOL, or BFLOAT16, and must meet the data type deduction rules (see Deduction Relationship) with self.
    • Atlas training products: The data type can be FLOAT, FLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, or BOOL, and must meet the data type deduction rules (see Deduction Relationship) with self.
  • out (aclTensor*, compute output): aclTensor on the device. The data type must be convertible from the deduced data type of self and other. The shape must be the broadcasted shape of self and other. Non-contiguous tensors are supported. The data format can be ND.
    • 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, INT8, UINT8, INT16, BOOL, or BFLOAT16.
    • Atlas training products: The data type can be FLOAT, FLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, or BOOL.
  • workspaceSize (uint64_t*, output): size of the workspace to be allocated on the device.
  • executor (aclOpExecutor**, output): operator executor, containing the operator computation process.

• Returns:

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

The first-phase API performs input parameter validation. The following errors may be returned:
  161001 (ACLNN_ERR_PARAM_NULLPTR): 1. The passed self, other, or out is a null pointer.
  161002 (ACLNN_ERR_PARAM_INVALID): 1. The data type or format of self or other is not supported.
                                    2. self or other do not meet the data type deduction rules.
                                    3. The deduced data type cannot be converted to the data type of out.
                                    4. The shape of self exceeds 8D.

aclnnFloorDivides

• Parameters:

  • workspace (void*, input): memory 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 aclnnFloorDividesGetWorkspaceSize.
  • 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.

aclnnInplaceFloorDividesGetWorkspaceSize

• Parameters:

  • selfRef (aclTensor*, compute input): aclTensor on the device. The shape must meet the broadcast relationship with other. Non-contiguous tensors are supported. The data format can be ND.
    • 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, INT8, UINT8, INT16, BOOL, or BFLOAT16, and must meet the data type deduction rules (see Deduction Relationship) with other.
    • Atlas training products: The data type can be FLOAT, FLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, or BOOL, and must meet the data type deduction rules (see Deduction Relationship) with other.
  • other (aclScalar*, compute input): aclScalar on the host.
    • 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, INT8, UINT8, INT16, BOOL, or BFLOAT16, and must meet the data type deduction rules (see Deduction Relationship) with selfRef.
    • Atlas training products: The data type can be FLOAT, FLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, or BOOL, and must meet the data type deduction rules (see Deduction Relationship) with selfRef.
  • workspaceSize (uint64_t*, output): size of the workspace to be allocated on the device.
  • executor (aclOpExecutor**, output): operator executor, containing the operator computation process.

• Returns:

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

The first-phase API performs input parameter validation. The following errors may be returned:
  161001 (ACLNN_ERR_PARAM_NULLPTR): 1. The passed selfRef or other is a null pointer.
  161002 (ACLNN_ERR_PARAM_INVALID): 1. The data type or format of selfRef or other is not supported.
                                    2. selfRef or other do not meet the data type deduction rules.
                                    3. The shape of selfRef or other exceeds 8D.

aclnnInplaceFloorDivides

• Parameters:

  • workspace (void*, input): memory 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 aclnnInplaceFloorDividesGetWorkspaceSize.
  • 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 computation:
  • aclnnFloorDivides and aclnnInplaceFloorDivides default to a deterministic implementation.

• Due to the limited precision of the FLOAT16 and BFLOAT16, not all decimals can be represented, which may lead to errors in floor computations. It is recommended to use higher-precision data types, such as FLOAT32.

Calling Example

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

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_floor_divide.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 the 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;
}

int main() {
  // 1. (Boilerplate) Initialize the device and stream. For details, see the ACL API manual.
  // Set the device ID 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> otherShape = {4, 2};
  std::vector<int64_t> outShape = {4, 2};
  void* selfDeviceAddr = nullptr;
  void* otherDeviceAddr = 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};
  std::vector<float> outHostData(8, 0);
  float otherValue = 2.0f;
  // 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 aclnnFloorDivides.
  ret = aclnnFloorDividesGetWorkspaceSize(self, other, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnFloorDividesGetWorkspaceSize 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 aclnnFloorDivides.
  ret = aclnnFloorDivides(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnFloorDivides failed. ERROR: %d\n", ret); return ret);

  uint64_t inplaceWorkspaceSize = 0;
  aclOpExecutor* inplaceExecutor;
  // Call the first-phase API of aclnnInplaceFloorDivides.
  ret = aclnnInplaceFloorDividesGetWorkspaceSize(self, other, &inplaceWorkspaceSize, &inplaceExecutor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceFloorDividesGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
  // Allocate device memory based on workspaceSize computed by the first-phase API.
  void* inplaceWorkspaceAddr = nullptr;
  if (inplaceWorkspaceSize > 0) {
    ret = aclrtMalloc(&inplaceWorkspaceAddr, inplaceWorkspaceSize, 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 aclnnInplaceFloorDivides.
  ret = aclnnInplaceFloorDivides(inplaceWorkspaceAddr, inplaceWorkspaceSize, inplaceExecutor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceFloorDivides failed. ERROR: %d\n", ret); return ret);

  // 4. (Boilerplate) 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 memory 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]);
  }

  auto inplaceSize = GetShapeSize(selfShape);
  std::vector<float> inplaceResultData(inplaceSize, 0);
  ret = aclrtMemcpy(inplaceResultData.data(), inplaceResultData.size() * sizeof(inplaceResultData[0]), selfDeviceAddr,
                    inplaceSize * sizeof(float), 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 < inplaceSize; i++) {
    LOG_PRINT("inplaceResult[%ld] is: %f\n", i, inplaceResultData[i]);
  }

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

  // 7. Free device resources. Modify the code based on the API definition.
  aclrtFree(selfDeviceAddr);
  aclrtFree(otherDeviceAddr);
  aclrtFree(outDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();

  return 0;
}