aclnnGatherV2

📄 View Source Code

Applicable Products

Function

• Description: Extracts elements from the input tensor along the specified dimension dim based on the index numbers in index, and saves them to the out tensor.

• Example:

  For example, given an input tensor $x=\begin{bmatrix}1 & 2 & 3 \\ 4 & 5 & 6 \\ 7 & 8 & 9\end{bmatrix}$ and an index tensor idx=[1, 0].

  • Result when dim=0: $y=\begin{bmatrix}4 & 5 & 6 \\ 1 & 2 & 3\end{bmatrix}$

  • Result when dim=1: $y=\begin{bmatrix}2 & 1\\ 5 & 4\\ 8 & 7\end{bmatrix}$

  The specific computation process is as follows: Taking a 3D tensor as an example, a tensor x with shape (3,2,2) = $\begin{bmatrix}[[1,&2],&[3,&4]], \\ [[5,&6],&[7,&8]], \\ [[9,&10],&[11,&12]]\end{bmatrix}$, idx = [1, 0], the subscripts corresponding to dim=0, 1, 2 of tensor x are $l, m, n$ respectively, and idx is one-dimensional (the zero-dimensional case is treated as one-dimensional with size 1).

  • When dim is 0: I = index[i];    y$[i][m][n]$ = x$[I][m][n]$

  • When dim is 1: J = index[j];     y$[l][j][n]$ = x$[l][J][n]$

  • When dim is 2: K = index[k];   y$[l][m][k]$ = x$[l][m][K]$

Function Prototype

Each operator is divided into a two-phase API. The "aclnnGatherV2GetWorkspaceSize" API must be called first to obtain the workspace size required for computation and the executor that encapsulates the operator computation flow. Then, the "aclnnGatherV2" API is called to perform the computation.

aclnnStatus aclnnGatherV2GetWorkspaceSize(
 const aclTensor *self,
 int64_t          dim,
 const aclTensor *index,
 aclTensor       *out,
 uint64_t        *workspaceSize,
 aclOpExecutor  **executor)

aclnnStatus aclnnGatherV2(
 void          *workspace,
 uint64_t       workspaceSize,
 aclOpExecutor *executor,
 aclrtStream    stream)

aclnnGatherV2GetWorkspaceSize

• Parameters

  Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Shape	Non-contiguous Tensor
  self	Input	Input aclTensor.	-	FLOAT、FLOAT16、INT64、INT32、INT16、INT8、UINT8、BOOL、DOUBLE、COMPLEX64、COMPLEX128、BFLOAT16	ND	1-8	√
  dim	Input	Inputs an integer.	The value range is [-self.dim(), self.dim()-1].	INT64	-	-	-
  index	Input	Input aclTensor.	Supports data dimensions from 1 to 8. The value range is within [0, self.shape[dim]) (including 0, excluding self.shape[dim]).	INT64, INT32	ND	1-8	√
  out	Output	Output aclTensor.	The data type must be the same as that of self. The data dimension supports 1 to 8 dimensions. The number of dimensions equals the number of self dimensions plus the number of index dimensions minus one. Except that the dim dimension is expanded to match the shape of index, the lengths of other dimensions are consistent with the corresponding dimensions of self.	Same as self	ND	1-8	-
  workspaceSize	Output	Returns the workspace size to be requested on the Device side.	-	-	-	-	-
  executor	Output	Returns the operator executor, which contains the operator computation process.	-	-	-	-	-

  • Atlas inference products, Atlas training products: The BFLOAT16 data type is not supported.

• 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	Error Code	Description
  ACLNN_ERR_PARAM_NULLPTR	161001	The self, index, or out parameter is a null pointer.
  ACLNN_ERR_PARAM_INVALID	161002	The data types of parameters self and index are not within the supported range.
  		When self.dim()=0, the value range of parameter dim is not within [-1, 0]; when self.dim()>0, the value range of parameter dim is not within [-self.dim(), self.dim()-1].
  		The number of dimensions of self and index exceeds 8.
  		The data types of self and out are inconsistent.
  		The shape of out does not satisfy the requirement that, except for the dim dimension which is expanded to match the shape of index, the lengths of all other dimensions are consistent with the corresponding dimensions of self.

aclnnGatherV2

• 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 aclnnGatherV2GetWorkspaceSize.
  executor	Input	Operator executor, which 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:

  • aclnnGatherV2 defaults to a deterministic implementation.

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_gather_v2.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-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);
  // Copy data from the host to the device using aclrtMemcpy.
  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 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/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. Customize the construction based on the API interface.
  std::vector<int64_t> selfShape = {2, 2};
  std::vector<int64_t> indexShape = {2, 2};
  std::vector<int64_t> outShape = {2, 2, 2};
  void* selfDeviceAddr = nullptr;
  void* indexDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclTensor* index = nullptr;
  aclTensor* out = nullptr;
  std::vector<float> selfHostData = {0, 1, 2, 3};
  std::vector<int64_t> indexHostData = {-2, -1, 0, 1};
  std::vector<float> outHostData = {0, 0, 0, 0, 0, 0, 0, 0};
  int64_t dim = 0;
  // Create the self aclTensor.
  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the index aclTensor.
  ret = CreateAclTensor(indexHostData, indexShape, &indexDeviceAddr, aclDataType::ACL_INT64, &index);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the 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. Replace with the specific API name.
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // Call the first-phase API of aclnnGatherV2.
  ret = aclnnGatherV2GetWorkspaceSize(self, dim, index, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnGatherV2GetWorkspaceSize 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 aclnnGatherV2.
  ret = aclnnGatherV2(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnGatherV2 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 device memory to host memory. Modify this 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(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(self);
  aclDestroyTensor(index);
  aclDestroyTensor(out);

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