aclnnMv

Applicable Products

Function

• Description: Computes the product of the matrix input and the vector vec.

• Formula:

  $$out = input * vec$$

  Where $input$ is an n*m two-dimensional tensor, $vec$ is a one-dimensional tensor of length m, and $out$ is a one-dimensional tensor of length n.

• Example:

  a = tensor([[1, 2, 3],
              [4, 5, 6]],dtype=float)
  b = tensor([1, 1, 1], dtype=float)
  c = mv(a, b)
  c: tensor([6, 15],dtype=float)

Function Prototype

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

aclnnStatus aclnnMvGetWorkspaceSize(
  const aclTensor *self,
  const aclTensor *vec,
  aclTensor       *out,
  int8_t          cubeMathType,
  uint64_t        *workspaceSize,
  aclOpExecutor   **executor)

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

aclnnMvGetWorkspaceSize

• Input Parameters:

  Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Shape	Non-contiguous Tensor
  self	Input	The input in the formula.	Other formats are read in ND format unless otherwise specified, but unexpected behavior may occur during continued execution.	BFLOAT16, FLOAT, FLOAT16	ND	2	√
  vec	Input	In the formula, the data type of vec is the same as that of self.	Other formats are read in ND format unless otherwise specified, but unexpected behavior may occur during subsequent execution.	BFLOAT16, FLOAT, FLOAT16	ND	1	√
  out	Output	The output "out" in the formula.	Other formats are read in ND format unless otherwise specified, but unexpected behavior may occur during continued execution.	BFLOAT16, FLOAT, FLOAT16	ND	1	√
  cubeMathType	Input	Specifies the computation logic of the Cube unit.	If the input data types have a deduction relationship, this parameter processes the deduced data type by default. Supported enum values are as follows: 0: KEEP_DTYPE, keeps the input data type for computation. 1: ALLOW_FP32_DOWN_PRECISION, allows down-precision computation of the input data. 2: USE_FP16, allows down-precision of the input to FLOAT16 for computation. 3: USE_HF32, allows down-precision of the input to HFLOAT32 for computation.	INT8	-	-	-
  workspaceSize	Output	Returns the workspace size to be applied for on the Device side.	-	-	-	-	-
  executor	Output	Returns the operator executor, which contains the operator computation flow.	-	-	-	-	-

  • Atlas training products, Atlas inference products:

    • BFLOAT16 data type is not supported.

    • When the input data type is FLOAT32, cubeMathType=0 is not supported.

    • For cubeMathType=1, when the input data type is FLOAT32, it is converted to FLOAT16 for computation. When the input is of other data types, no processing is performed.

    • cubeMathType=3 is not supported.

  • Atlas A2 training products/Atlas A2 inference products, Atlas A3 training products/Atlas A3 inference products, Ascend 950PR/Ascend 950DT:

    • cubeMathType=1: When the input data type is FLOAT32, it is converted to HFLOAT32 for computation. When the input is of other data types, no processing is performed.

    • cubeMathType=2: This option is not supported when the input data type is BFLOAT16.

    • cubeMathType=3: When the input data type is FLOAT32, it will be converted to HFLOAT32 for computation. This option is not supported when the input is of other data types.

• 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 input self, vec, or out is a null pointer.
  ACLNN_ERR_PARAM_INVALID	161002	The data types of self, vec, and out are not within the supported range.
  		The data types of self, vec, and out are different.
  		The shapes of self, vec, and out do not meet the constraints.
  		The passed cubeMathType does not meet the constraints.

aclnnMv

• 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 through the first-phase API aclnnMvGetWorkspaceSize.
  executor	Input	Operator executor, which contains the operator computation flow.
  stream	Input	Specifies the stream for executing the task.

• Return value

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

Constraints

• Deterministic description:

  • Atlas training products and Atlas inference products: aclnnMv defaults to a non-deterministic implementation. Deterministic computation can be enabled by calling aclrtCtxSetSysParamOpt.

  • Ascend 950PR/Ascend 950DT: aclnnMv 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_mv.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 code for 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);
  // Copy host data to the device memory by calling 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];
  }

  // Create an aclTensor by calling aclCreateTensor.
  *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.
  // Set deviceId based on the 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.
  std::vector<int64_t> selfShape = {2, 3};
  std::vector<int64_t> vecShape = {3};
  std::vector<int64_t> outShape = {2};
  void* selfDeviceAddr = nullptr;
  void* vecDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclTensor* vec = nullptr;
  aclTensor* out = nullptr;
  std::vector<float> selfHostData = {1, 2, 3, 4, 5, 6};
  std::vector<float> vecHostData = {7, 8, 9};
  std::vector<float> outHostData = {0, 0};
  int8_t cubeMathType = 1;
  // Create the self aclTensor.
  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the vec aclTensor.
  ret = CreateAclTensor(vecHostData, vecShape, &vecDeviceAddr, aclDataType::ACL_FLOAT, &vec);
  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 it with the specific API name.
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // Call the first-phase API of aclnnMv.
  ret = aclnnMvGetWorkspaceSize(self, vec, out, cubeMathType, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnMvGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
  // Allocate device memory based on the workspaceSize computed 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 aclnnMv.
  ret = aclnnMv(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnMv failed. ERROR: %d\n", ret); return ret);

  // 4. (Boilerplate) Wait synchronously for the 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 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. Modify based on the specific API definition.
  aclDestroyTensor(self);
  aclDestroyTensor(vec);
  aclDestroyTensor(out);
  // 7. Release device resources. Parameters need to be defined based on the specific API.
  aclrtFree(selfDeviceAddr);
  aclrtFree(vecDeviceAddr);
  aclrtFree(outDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();
  return 0;
}