aclnnLinalgCross

Applicable Products

Product	Supported
Ascend 950PR/Ascend 950DT	√
Atlas A3 training products/Atlas A3 inference products	√
Atlas A2 training products/Atlas A2 inference products	√
Atlas 200I/500 A2 inference products	√
Atlas inference products	√
Atlas training products	√

Function

Description: Performs the linear_cross operation on the input tensors.
Formula:

out = self\times other = \begin{vmatrix}i&j&k\\x_1&y_1&z_1\\x_2&y_2&z_2\end{vmatrix} = (y_1z_2-y_2z_1)i-(x_1z_2-x_2z_1)j+(x_1y_2-x_2y_1)k

Function Prototype

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

aclnnStatus aclnnLinalgCrossGetWorkspaceSize(
  const aclTensor*          self,
  const aclTensor*          other,
  int64_t                   dim,
  aclTensor*                out,
  uint64_t*                 workspaceSize,
  aclOpExecutor**           executor)

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

aclnnLinalgCrossGetWorkspaceSize

Parameters:

Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension	Non-contiguous Tensor
self (aclTensor*)	Input	self in the formula.	The data type is consistent with other and out. It must satisfy the broadcast relationship with other, and the shape value after broadcasting along the axis specified by dim must be 3.	INT8, INT16, INT32, INT64, UINT8, FLOAT16, BFLOAT16, FLOAT, FLOAT64, COMPLEX64, COMPLEX128	ND	0-8	√
other (aclTensor*)	Input	other in the formula.	Data type: Consistent with self and out. Must satisfy the broadcast relationship with self.	INT8, INT16, INT32, INT64, UINT8, FLOAT16, BFLOAT16, FLOAT, FLOAT64, COMPLEX64, COMPLEX128	ND	0-8	√
dim (int64_t)	Input	Specifies the axis along which to compute the linalg cross product for self.	Defaults to -1 if not specified, with a valid range of [-self dimension count, self dimension count - 1].	-	-	-	-
out (aclTensor*)	Output	The out in the formula.	The data type is consistent with self and other. The shape must be consistent with the broadcast shape of self and other.	INT8, INT16, INT32, INT64, UINT8, FLOAT16, BFLOAT16, FLOAT, FLOAT64, COMPLEX64, COMPLEX128	ND	-	√
workspaceSize (uint64_t*)	Output	Returns the workspace size to be allocated on the Device side.	-	-	-	-	-
executor (aclOpExecutor**)	Output	Returns the operator executor, which contains the operator computation flow.	-	-	-	-	-

Atlas training products BFLOAT16 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 self or out is a null pointer.
ACLNN_ERR_PARAM_INVALID	161002	The data types of self, other, and out are inconsistent, or the data format is not within the supported range.
		The dimension of self or other is greater than 8.
		self and other do not satisfy the broadcast relationship.
		The shape of self and other after broadcast is inconsistent with out.
		The shape of self in the corresponding dim dimension is not 3.
		The value of dim is not within the range of [-number of dimensions of self, number of dimensions of self - 1].

aclnnLinalgCross

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 aclnnLinalgCrossGetWorkspaceSize.
executor	Input	The op executor, which contains the operator computation flow.
stream	Input	Specifies the stream on which the task is executed.

Return value

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

Constraints

Deterministic computation:
- aclnnLinalgCross 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_linalg_cross.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 allocate memory on the device side.
  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 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 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;
}

aclError InitAcl(int32_t deviceId, aclrtStream* stream)
{
  auto ret = Init(deviceId, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);
  return ACL_SUCCESS;
}

aclError CreateInputs(
    std::vector<int64_t>& selfShape, std::vector<int64_t>& otherShape, std::vector<int64_t>& outShape,
    void** selfDeviceAddr, void** otherDeviceAddr, void** outDeviceAddr, aclTensor** self, aclTensor** other,
    aclTensor** out)
{
  std::vector<double> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7, 8};
  std::vector<double> otherHostData = {1, 1, 1, 2, 2, 2, 3, 3, 3};
  std::vector<double> outHostData = {0, 0, 0, 0, 0, 0, 0, 0, 0};

  // Create the self aclTensor.
  auto ret = CreateAclTensor(selfHostData, selfShape, selfDeviceAddr, aclDataType::ACL_DOUBLE, self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // Create the other aclTensor.
  ret = CreateAclTensor(otherHostData, otherShape, otherDeviceAddr, aclDataType::ACL_DOUBLE, other);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // Create the out aclTensor.
  ret = CreateAclTensor(outHostData, outShape, outDeviceAddr, aclDataType::ACL_DOUBLE, out);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  return ACL_SUCCESS;
}

aclError ExecOpApi(
    aclTensor* self, aclTensor* other, aclTensor* out, int64_t dim, void** workspaceAddrOut, uint64_t& workspaceSize,
    void* outDeviceAddr, std::vector<int64_t>& outShape, aclrtStream stream)
{
  aclOpExecutor* executor;

  // Call the first-phase API of aclnnLinalgCross.
  auto ret = aclnnLinalgCrossGetWorkspaceSize(self, other, dim, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnLinalgCrossGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);

  // Apply for device memory based on workspaceSize.
  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);
  }
  *workspaceAddrOut = workspaceAddr;

  // Call the second-phase API of aclnnLinalgCross.
  ret = aclnnLinalgCross(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnLinalgCross failed. ERROR: %d\n", ret); return ret);

  // Synchronize.
  ret = aclrtSynchronizeStream(stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);

  // Copy the result from device to host.
  auto size = GetShapeSize(outShape);
  std::vector<double> 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: %lf\n", i, resultData[i]);
  }

  return ACL_SUCCESS;
}

int main()
{
  // Initialize the device/stream.
  int32_t deviceId = 0;
  aclrtStream stream;

  CHECK_RET(InitAcl(deviceId, &stream) == ACL_SUCCESS, return -1);

  // Construct the input and output.
  std::vector<int64_t> selfShape = {3, 3};
  std::vector<int64_t> otherShape = {3, 3};
  std::vector<int64_t> outShape = {3, 3};

  void* selfDeviceAddr = nullptr;
  void* otherDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;

  aclTensor* self = nullptr;
  aclTensor* other = nullptr;
  aclTensor* out = nullptr;

  aclError ret = CreateInputs(
      selfShape, otherShape, outShape, &selfDeviceAddr, &otherDeviceAddr, &outDeviceAddr, &self, &other, &out);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // Call the CANN operator API.
  int64_t dim = 1;
  uint64_t workspaceSize = 0;
  void* workspaceAddr = nullptr;

  ret = ExecOpApi(self, other, out, dim, &workspaceAddr, workspaceSize, outDeviceAddr, outShape, stream);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // Release aclTensor.
  aclDestroyTensor(self);
  aclDestroyTensor(other);
  aclDestroyTensor(out);

  // Release device resources.
  aclrtFree(selfDeviceAddr);
  aclrtFree(otherDeviceAddr);
  aclrtFree(outDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();
  return 0;
}