aclnnBatchMatMul

Applicable Products

Function

• Description: Performs matrix multiplication between tensor self and tensor mat2. Only three-dimensional tensors are supported. The first dimension is the batch dimension, and the last two dimensions perform matrix multiplication. Broadcasting is also supported when the batch axis of one input is 1. See the example.

• Formula:

  $$out = self@mat2$$

• Example:

  The shape of self is [A, M, K], the shape of mat2 is [A, K, N], and the shape of the computed output out is [A, M, N]. The first dimensions are equal, and the last two dimensions perform matrix multiplication. The shape of self is [A, M, K], the shape of mat2 is [1, K, N], and the shape of the computed output out is [A, M, N]. The first dimension of mat2 is 1, which will be broadcast to A, and the last two dimensions perform matrix multiplication. The shape of self is [1, M, K], the shape of mat2 is [B, K, N], and the shape of the computed output out is [B, M, N]. The first dimension of self is 1, which will be broadcast to B, and the last two dimensions perform matrix multiplication.

Function Prototype

Each operator is divided into a two-phase API. You must first call the "aclnnBatchMatMulGetWorkspaceSize" API to obtain the input parameters and calculate the required workspace size based on the process, and then call the "aclnnBatchMatMul" API to perform the computation.

aclnnStatus aclnnBatchMatMulGetWorkspaceSize(
  const aclTensor *self,
  const aclTensor *mat2,
  aclTensor       *out,
  int8_t           cubeMathType,
  uint64_t        *workspaceSize,
  aclOpExecutor   **executor)

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

aclnnBatchMatMulGetWorkspaceSize

• Parameters:

  Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (Shape)	Non-Contiguous Tensor
  self	Input	Represents the first matrix in matrix multiplication, i.e., self in the formula.	The data type must satisfy the data type deduction rules with mat2 (see Deduction Relationship and Constraints).	BFLOAT16, FLOAT16, FLOAT32	ND	3	√
  mat2	Input	Represents the second matrix in matrix multiplication, mat2 in the formula.	The data type needs to satisfy the data type deduction rules with self (see Deduction Relationship and Constraints). The reduce dimension of mat2 needs to be equal in size to the reduce dimension of self.	BFLOAT16, FLOAT16, FLOAT32	ND	3	√
  out	Output	Represents the output matrix of matrix multiplication, out in the formula.	The data type must be consistent with the data type derived from self and mat2 (see deduction relationship and Constraints).	BFLOAT16, FLOAT16, FLOAT32	ND	3	-
  cubeMathType	Input	Specifies the computation logic of the Cube unit.	If the input data types have a deduction relationship, this parameter processes the data type after deduction by default. Supported enumeration values are as follows: 0: KEEP_DTYPE. Keeps the input data type for computation. 1: ALLOW_FP32_DOWN_PRECISION. Supports down-precision computation of the input data. 2: USE_FP16. Supports down-precision of the input to FLOAT16 for computation. 3: USE_HF32. Supports down-precision of the input to HFLOAT32 for computation.	INT8	-	-	-
  workspaceSize	Output	Returns the workspace size to be allocated on the Device side.	-	-	-	-	-
  executor	Output	Returns the operator executor, which contains the operator computation flow.	-	-	-	-	-

  • 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 will be converted to HFLOAT32 for computation; when the input is of other data types, no processing is performed.

    • cubeMathType=2: When the input data type is FLOAT32, it will be converted to FLOAT16 for computation; when the input is of other data types, no processing is performed.

    • cubeMathType=3: When the input data type is FLOAT32, it will be converted to HFLOAT32 for computation; when the input is of other data types, no processing is performed.

  • Atlas training products, Atlas inference products:

    • BFLOAT16 data type is not supported.

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

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

    • cubeMathType=3 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 passed self, mat2, or out is a null pointer.
  ACLNN_ERR_PARAM_INVALID	161002	The data type of self, mat2, or out is not within the supported range.
  		The data format of self, mat2, or out is not within the supported range.
  		The dimensions of self and mat2 are not three-dimensional.
  		The last dimension of self is not equal to the second-to-last dimension of mat2.

aclnnBatchMatMul

• Parameters:

  Parameter	Input/Output	Description
  workspace	Input	Workspace memory address allocated on the Device side.
  workspaceSize	Input	Workspace size allocated on the Device side, obtained from the first-phase API aclnnBatchMatMulGetWorkspaceSize.
  executor	Input	Operator executor, which contains the operator computation flow.
  stream	Input	Specifies the stream that executes the task.

• Return value

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

Constraints

• Deterministic computation

  • Atlas training products, Atlas inference products: aclnnBatchMatMul defaults to a deterministic implementation.

  • Ascend 950PR/Ascend 950DT: aclnnBatchMatMul 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_batch_matmul.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 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.
  // Fill in the deviceId based on your actual device.
  int32_t deviceId = 0;
  aclrtStream stream;
  auto ret = Init(deviceId, &stream);
  // Handle the check based on your needs.
  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 = {1, 2, 3};
  std::vector<int64_t> mat2Shape = {1, 3, 4};
  std::vector<int64_t> outShape = {1, 2, 4};
  void* selfDeviceAddr = nullptr;
  void* mat2DeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclTensor* mat2 = nullptr;
  aclTensor* out = nullptr;
  std::vector<float> selfHostData = {0, 1, 2, 3, 4, 5};
  std::vector<float> mat2HostData = {1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 4};
  std::vector<float> outHostData(8, 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 mat2 aclTensor.
  ret = CreateAclTensor(mat2HostData, mat2Shape, &mat2DeviceAddr, aclDataType::ACL_FLOAT, &mat2);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create out aclTensor.
  ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_FLOAT, &out);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  uint64_t workspaceSize = 0;
  aclOpExecutor* executor = nullptr;

  // aclnnBatchMatMul API call example
  // 3. Call the CANN operator library API. Replace with the specific API name.
  // Call the first-phase API of aclnnBatchMatMul.
  ret = aclnnBatchMatMulGetWorkspaceSize(self, mat2, out, cubeMathType, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnBatchMatMulGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
  // Apply for 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 aclnnBatchMatMul.
  ret = aclnnBatchMatMul(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnBatchMatMul failed. ERROR: %d\n", ret); return ret);

  // 4. (Boilerplate) Synchronously 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-side memory to the Host side. 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 and aclScalar. Modify based on the specific API definition.
  aclDestroyTensor(self);
  aclDestroyTensor(mat2);
  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;
}