aclnnScatterNdUpdate

📄 View Source Code

Applicable Products

Function

Updates the values in the tensor varRef one by one using the values from the tensor updates at the specified indices.

Function Prototype

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

aclnnStatus aclnnScatterNdUpdateGetWorkspaceSize(
  aclTensor*       varRef,
  const aclTensor* indices,
  const aclTensor* updates,
  uint64_t*        workspaceSize,
  aclOpExecutor**  executor)

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

aclnnScatterNdUpdateGetWorkspaceSize

• Parameters:

  Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (Shape)	Non-Contiguous Tensor
  varRef (aclTensor*)	Input/Output	Specifies the target tensor to be updated.	The data type must be the same as that of updates.	FLOAT16, BFLOAT16, FLOAT32, INT64, BOOL, INT8	ND	1-8	√
  indices (aclTensor*)	Input	Specifies the index tensor.	Index data in indices does not support out-of-bounds access.	INT32, INT64	ND	-	√
  updates (aclTensor*)	Input	Updated data tensor.	The data type must be consistent with varRef.	FLOAT16, BFLOAT16, FLOAT32, INT64, BOOL, INT8	ND	-	√
  workspaceSize (uint64_t*)	Output	Returns the workspace size to be applied for on the Device side.	-	-	-	-	-
  executor (aclOpExecutor**)	Output	Returns the op 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	varRef, indices, or updates is a null pointer.
  ACLNN_ERR_PARAM_INVALID	161002/561103	The shapes of varRef, indices, and updates do not meet the requirements.
  		The data types of varRef, indices, and updates are not within the supported range.

aclnnScatterNdUpdate

• Parameters:

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

• Return value

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

Constraints

• Deterministic computation:

  • aclnnScatterNdUpdate defaults to a deterministic implementation.

• Input shape constraints:

  • indices must be at least 2-dimensional, and the size of its last dimension cannot exceed the number of dimensions of varRef.

  • Assume the size of the last dimension of indices is a, then the shape of updates equals the shape of indices excluding the last dimension plus the shape of varRef excluding the first a dimensions. For example: if the shape of varRef is (4, 5, 6) and the shape of indices is (3, 2), then the shape of updates must be (3, 6).

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_update.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 data from the host side to the 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. See the acl API manual.
  // Enter the deviceId based on your 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 and output, which requires custom construction based on the API.
  std::vector<int64_t> varRefShape = {2, 3, 7};
  std::vector<int64_t> indicesShape = {2, 2};
  std::vector<int64_t> updatesShape = {2, 7};
  void* varRefDeviceAddr = nullptr;
  void* indicesDeviceAddr = nullptr;
  void* updatesDeviceAddr = nullptr;
  aclTensor* varRef = nullptr;
  aclTensor* indices = nullptr;
  aclTensor* updates = nullptr;
  std::vector<float> varRefHostData = {0.3816, 0.3939, 0.8474, 0.1652, 0.6049, 0.3315, 0.4954,
                                     0.3284, 0.7060, 0.4359, 0.6514, 0.9476, 0.4708, 0.0656,
                                     0.9652, 0.9512, 0.6452, 0.1981, 0.4159, 0.9575, 0.1516,
                                     0.4987, 0.9107, 0.6635, 0.4119, 0.4845, 0.5558, 0.2749,
                                     0.6230, 0.1180, 0.2400, 0.9971, 0.4093, 0.5561, 0.4023,
                                     0.6612, 0.4109, 0.8470, 0.9733, 0.6947, 0.7980, 0.7957};
  std::vector<int64_t> indicesHostData = {5, 0, 1, 5};
  std::vector<float> updatesHostData = {0.7804, 0.3411, 0.6674, 0.8468, 0.6679, 0.5549, 0.9893,
                                    0.2086, 0.2473, 0.5110, 0.4549, 0.3113, 0.8490, 0.9217};
  // Create the varRef aclTensor.
  ret = CreateAclTensor(varRefHostData, varRefShape, &varRefDeviceAddr, aclDataType::ACL_FLOAT, &varRef);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the indices aclTensor.
  ret = CreateAclTensor(indicesHostData, indicesShape, &indicesDeviceAddr, aclDataType::ACL_INT64, &indices);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the updates aclTensor.
  ret = CreateAclTensor(updatesHostData, updatesShape, &updatesDeviceAddr, aclDataType::ACL_FLOAT, &updates);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // 3. Call the CANN operator library API. Replace with the specific API name.
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // Call the first-phase API of aclnnScatterNdUpdate.
  ret = aclnnScatterNdUpdateGetWorkspaceSize(varRef, indices, updates, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnScatterNdUpdateGetWorkspaceSize 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);
  }
  // Call the second-phase API of aclnnScatterNdUpdate.
  ret = aclnnScatterNdUpdate(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnScatterNdUpdate failed. ERROR: %d\n", ret); return ret);

  // 4. (Boilerplate) Synchronously wait for task execution to finish.
  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 memory. Modify based on the specific API definition.
  auto size = GetShapeSize(varRefShape);
  std::vector<float> resultData(size, 0);
  ret = aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), varRefDeviceAddr,
                    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 based on the specific API definition.
  aclDestroyTensor(varRef);
  aclDestroyTensor(indices);
  aclDestroyTensor(updates);
  // 7. Release device resources. Set parameters based on the specific API definition.
  aclrtFree(varRefDeviceAddr);
  aclrtFree(indicesDeviceAddr);
  aclrtFree(updatesDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();
  return 0;
}