aclnnScatterNd

📄 View Source Code

Applicable Products

Function

Copies data from data to out, and updates out with updates at the positions specified by indices.

Function Prototype

Each operator is divided into a two-phase API. The "aclnnScatterNdGetWorkspaceSize" API must be called first to obtain the workspace size required for computation and the executor that encapsulates the operator computation flow. Then, the "aclnnScatterNd" API is called to perform the computation.

aclnnStatus aclnnScatterNdGetWorkspaceSize(
  const aclTensor* data,
  const aclTensor* indices,
  const aclTensor* updates,
  aclTensor*       out,
  uint64_t*        workspaceSize,
  aclOpExecutor**  executor)

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

aclnnScatterNdGetWorkspaceSize

• Parameters:

  Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (Shape)	Non-Contiguous Tensor
  data (aclTensor*)	Input	Data to be copied.	The data type is the same as that of updates and out.	FLOAT16, FLOAT, BOOL, BFLOAT16	ND	1-8	√
  indices (aclTensor*)	Input	Index.	Only non-negative indices are supported. Index data in indices cannot be out of bounds.	INT32, INT64	ND	1-8	√
  updates (aclTensor*)	Input	Data to be updated.	The data type is the same as that of data and out.	FLOAT16, FLOAT, BOOL, BFLOAT16	ND	1-8	√
  out (aclTensor*)	Output	Computation result.	Data type that is the same as that of data and out. Shape that is the same as that of data.	FLOAT16, FLOAT, BOOL, BFLOAT16	ND	-	√
  workspaceSize (uint64_t*)	Output	Returns the workspace size to be applied for on the Device side.	-	-	-	-	-
  executor (aclOpExecutor**)	Output	Returns the operator executor, which contains the operator computation flow.	-	-	-	-	-

• Return value

  aclnnStatus: return code. For details, see aclnn Return Code.

  The first-phase API performs input parameter validation and returns an error in the following scenarios:

  Return Value	Return Code	Description
  ACLNN_ERR_PARAM_NULLPTR	161001	A null pointer exists in data, indices, updates, or out.
  ACLNN_ERR_PARAM_INVALID	161002	The data type is not within the supported range.
  		The shape does not meet the requirements: 1 <= rank(data) <= 8, 1 <= rank(indices) <= 8, rank(updates) = rank(data) + rank(indices) - indices.shape[-1] - 1.
  		The shape does not meet the requirements: 1 <= rank(indices) <= 8, indices.shape[-1] <= rank(data).
  		The shape does not meet the requirements: 1 <= rank(updates) <= 8, updates.shape == indices.shape[:-1] + data.shape[indices.shape[-1] :].
  		The shape does not meet the requirements: data.shape == out.shape.

aclnnScatterNd

• Parameters:

  Parameter	Input/Output	Description
  workspace	Input	Workspace memory address applied for on the Device side.
  workspaceSize	Input	Workspace size applied for on the Device side, obtained from the first-phase API aclnnScatterNdGetWorkspaceSize.
  executor	Input	Operator executor that contains the operator computation process.
  stream	Input	Specifies the stream for task execution.

• Return value

  aclnnStatus: return code. For details, see aclnn Return Code.

Constraints

None.

Example

The following provides example code for reference only. For details about compilation and running, see Compile and Run Samples.

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_scatter_nd.h"
#include "aclnn/aclnn_base.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: resource initialization.

  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 apply for device-side 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-side data to device-side 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);

  // Calculate the strides of a 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 the aclCreateTensor API 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/stream. For details, see the acl external API list.
  // Enter the deviceId based on the actual device.
  int32_t deviceId = 0;
  aclrtStream stream;
  auto ret = Init(deviceId, &stream);
  // Handle the check based on your requirements.
  CHECK_RET(ret == 0, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);
  // 2. Construct the input and output. You need to customize the construction based on the API.
  std::vector<int64_t> dataShape = {8};
  std::vector<int64_t> indicesShape = {4, 1};
  std::vector<int64_t> updatesShape = {4};
  std::vector<int64_t> outShape = {8};

  void* dataDeviceAddr = nullptr;
  void* indicesDeviceAddr = nullptr;
  void* updatesDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* data = nullptr;
  aclTensor* indices = nullptr;
  aclTensor* updates = nullptr;
  aclTensor* out = nullptr;

  std::vector<float> selfHostData = {1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0};
  std::vector<int32_t> indicesData = {4,3,1,7};
  std::vector<float> updatesData = {9.0, 10.0, 11.0, 12.0};
  std::vector<float> outData = {0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0};

  ret = CreateAclTensor(selfHostData, dataShape, &dataDeviceAddr, aclDataType::ACL_FLOAT, &data);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  ret = CreateAclTensor(indicesData, indicesShape, &indicesDeviceAddr, aclDataType::ACL_INT32, &indices);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  ret = CreateAclTensor(updatesData, updatesShape, &updatesDeviceAddr, aclDataType::ACL_FLOAT, &updates);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  ret = CreateAclTensor(outData, outShape, &outDeviceAddr, aclDataType::ACL_FLOAT, &out);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // Create the output aclTensor.
  // ret = CreateAclTensor(outData, outShape, &outDeviceAddr, aclDataType::ACL_INT32, &out);
  // CHECK_RET(ret == ACL_SUCCESS, return ret);

  // 3. Call the CANN operator library API. Replace with the specific API.
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // Call the first-phase API of aclnnAdd.
  ret = aclnnScatterNdGetWorkspaceSize(data, indices, updates, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnScatterNdGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
  // Apply for device memory based on the workspaceSize calculated 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;);
  }

  ret = aclnnScatterNd(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnScatterNd failed. ERROR: %d\n", ret); return ret);
  // 4. (Boilerplate) Synchronize and wait for task execution to 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 side memory to the host side. Modify based on the specific API definition.
  auto size = GetShapeSize(outShape);
  std::vector<float> resultData(size, 0);
  ret = aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), outDeviceAddr, size * 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 < size; i++) {
    LOG_PRINT("result[%ld] is: %f\n", i, resultData[i]);
  }

  // 6. Release aclTensor and aclScalar. Modify based on the specific API definition.
  aclDestroyTensor(data);
  aclDestroyTensor(indices);
  aclDestroyTensor(updates);
  aclDestroyTensor(out);

  // 7. Release device resources. Modify based on the specific API definition.

  aclrtFree(dataDeviceAddr);
  aclrtFree(indicesDeviceAddr);
  aclrtFree(updatesDeviceAddr);
  aclrtFree(outDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();

  return 0;
}