aclnnSliceV2

Applicable Products

Function

Description: Extracts a tensor $out$ from the input tensor $self$ based on the given axes $axes$, ranges $[starts, ends]$, and strides $steps$. $starts[i]$ and $ends[i]$ can take values outside $[0, self.shape[axes[i]]]$. After taking the values, they are converted to valid values according to the following formulas, assuming $self.shape[axes[i]] = N$:

$out.shape$ differs from $self.shape$ on the axes specified in axes, and remains the same on other axes:

Function Prototype

Each operator is divided into a two-phase API. The "aclnnSliceV2GetWorkspaceSize" API must be called first to obtain the workspace size required for computation and the executor that encapsulates the operator computation flow. Then, the "aclnnSliceV2" API is called to perform the computation.

aclnnStatus aclnnSliceV2GetWorkspaceSize(
    const aclTensor*   self,
    const aclIntArray* starts,
    const aclIntArray* ends,
    const aclIntArray* axes,
    const aclIntArray* steps,
    aclTensor*         out,
    uint64_t*          workspaceSize,
    aclOpExecutor**    executor)

aclnnStatus aclnnSliceV2(
    void*          workspace,
    uint64_t       workspaceSize,
    aclOpExecutor* executor,
    aclrtStream    stream)

aclnnSliceV2GetWorkspaceSize

• Parameters

  Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Shape	Non-contiguous Tensor
  self (aclTensor*)	Input	self in the formula.	-	FLOAT、FLOAT16、INT32、INT64、INT8、UINT8、BOOL、BFLOAT16	ND	1-8	√
  starts (aclIntArray*)	Input	starts in the formula.	Starting index of the slice position. The number of elements in starts, ends, axes, and steps must be the same.	-	-	-	-
  ends (aclIntArray*)	Input	ends in the formula.	End index of the slice position. starts, ends, axes, and steps have the same number of elements.	-	-	-	-
  axes (aclIntArray*)	Input	Axes in the formula, specifying the axes to slice.	starts, ends, axes, and steps have the same number of elements. The elements of axes must be within the range [-self.dim(), self.dim() - 1], and the specified axes must not be duplicated.	-	-	-	-
  steps (aclIntArray*)	Input	The steps in the formula.	starts, ends, axes, and steps have the same number of elements, and all elements of steps are positive integers.	-	-	-	-
  out (aclTensor*)	Output	out in the formula.	The shape satisfies the deduction rules in the formula.	FLOAT, FLOAT16, INT32, INT64, INT8, UINT8, BOOL, BFLOAT16	ND	0-8	√
  workspaceSize (uint64_t*)	Output	Returns the workspace size to be requested on the Device side.	-	-	-	-	-
  executor（aclOpExecutor**）	Output	Returns the op executor, which contains the operator computation flow.	-	-	-	-	-

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

• Return Value

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

  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	Indicates that one or more of the input parameters self, starts, ends, axes, steps, or out are null pointers.
  ACLNN_ERR_PARAM_INVALID	161002	The data types of self and out are not within the supported range.
  		The shape dimension of self is greater than 8.
  		A value in axes is out of bounds, not within [-self.dim(), self.dim() - 1], or the axes represented are duplicated.
  		A value in steps is less than or equal to 0.
  		The data types of self and out are inconsistent.
  		The shape of out does not satisfy the deduction rules in Function.
  		The number of elements in starts, ends, axes, and steps is different.

aclnnSliceV2

• Parameters

  Parameter	Input/Output	Description
  workspace	Input	Address of the workspace memory applied for on the Device side.
  workspaceSize	Input	Size of the workspace applied for on the Device side, obtained through the first-phase API aclnnSliceV2GetWorkspaceSize.
  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 Codes.

Constraints

• Deterministic computation:

  • aclnnSliceV2 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_slice_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: 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);

  // Compute 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. Custom construction is required based on the API.
  std::vector<int64_t> selfShape = {4, 2};
  std::vector<int64_t> outShape = {2, 1};
  void* selfDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclTensor* out = nullptr;
  aclIntArray *dim = nullptr;
  aclIntArray *start = nullptr;
  aclIntArray *end = nullptr;
  aclIntArray *step = nullptr;
  std::vector<float> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7};
  std::vector<float> outHostData = {0, 0};
  int64_t dimValue[] = {0,1};
  int64_t startValue[] = {1,1};
  int64_t endValue[] = {3,2};
  int64_t stepValue[] = {1,1};
  // Create the self aclTensor.
  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  dim = aclCreateIntArray(dimValue, 2);
  CHECK_RET(dim != nullptr, return ret);
  start = aclCreateIntArray(startValue, 2);
  CHECK_RET(start != nullptr, return ret);
  end = aclCreateIntArray(endValue, 2);
  CHECK_RET(end != nullptr, return ret);
  step = aclCreateIntArray(stepValue, 2);
  CHECK_RET(step != nullptr, 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 aclnnSliceV2.
  ret = aclnnSliceV2GetWorkspaceSize(self, start, end, dim, step, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnSliceV2GetWorkspaceSize 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 aclnnSliceV2.
  ret = aclnnSliceV2(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnSliceV2 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 memory to the host memory. 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(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. Modify based on the specific API definition.
  aclDestroyTensor(self);
  aclDestroyIntArray(dim);
  aclDestroyIntArray(start);
  aclDestroyIntArray(end);
  aclDestroyIntArray(step);
  aclDestroyTensor(out);

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