aclnnGeluBackwardV2

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 backward propagation of aclnnGeluV2.
Formula:

GELU Forward Computation (where x can be a scalar or a tensor):

Gelu(x)=x \cdot \Phi(x)=x/2 \cdot [1+erf(x/\sqrt{2})]

The formula for erf is:

erf(x)=\frac{2}{\sqrt \pi}\sum^{\infty}_{n=0}{\frac{(-1)^n \cdot x^{2n+1}}{n! \cdot (2n+1)}}

The relationship between gradInput and gradOutput can be expressed as:

gradInput = gradOutput \cdot (\frac{1}{2}+\frac{1}{2} \cdot erf(\frac{x}{\sqrt2})+\frac{x}{\sqrt{2\pi}} \cdot e^{-\frac{x^2}{2}})

The approximate formula for GELU is:

Gelu(x)=0.5x(1+tanh(\sqrt{2/\pi}(x+0.044715x^3)))

Function Prototype

Each operator is divided into a two-phase API. The "aclnnGeluBackwardV2GetWorkspaceSize" API must be called first to obtain the workspace size required for computation and the executor that encapsulates the operator computation flow. Then, the "aclnnGeluBackwardV2" API is called to perform the computation.

aclnnStatus aclnnGeluBackwardV2GetWorkspaceSize(
  const aclTensor *gradOutput,
  const aclTensor *self,
  char            *approximate,
  aclTensor       *gradInput,
  uint64_t        *workspaceSize,
  aclOpExecutor  **executor)

aclnnStatus aclnnGeluBackwardV2(
  void*             workspace,
  uint64_t          workspace_size,
  aclOpExecutor*    executor,
  const aclrtStream stream)

aclnnGeluBackwardV2GetWorkspaceSize

Parameters:

Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (shape)	Non-contiguous Tensor
gradOutput (aclTensor*)	Input	Weight used for gradient computation, i.e., the weight tensor by which the forward output tensor is multiplied to become a scalar, corresponding to gradOutput in the formula.	The shape must satisfy the broadcast relationship with the shape of the forward self. The dtype and the dtype of self must satisfy the data type deduction rules (see Deduction Relationship). Supports empty tensors.	FLOAT16, BFLOAT16, FLOAT32	ND	0-8	√
self (aclTensor*)	Input	Forward input value of GELU, which is x in the formula.	shape must satisfy the broadcast relationship with gradOutput's shape. dtype and gradOutput's dtype must satisfy the data type deduction rules (see deduction relationship). Empty tensors are supported.	FLOAT16, BFLOAT16, FLOAT32	ND	0-8	√
approximate (char*)	Input	Activation function mode used for computation.	Configurable to "none" or "tanh", where "none" represents the erf mode and "tanh" represents the tanh mode.	-	-	-	-
gradInput (aclTensor*)	Output	Output of the backward computation, representing the gradient of the GELU forward input parameter, i.e., the result of differentiating the input, denoted as gradInput in the formula.	The dtype is consistent with the result of data type inference performed on self and gradOutput The shape is consistent with the result of broadcasting gradOutput and self.	FLOAT16, BFLOAT16, FLOAT32	ND	0-8	√
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 inference products, Atlas training products: The supported data types are FLOAT16 and FLOAT32.

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 Code	Error Code	Description
ACLNN_ERR_PARAM_NULLPTR	161001	The passed gradOutput, self, approximate, or gradInput is a null pointer.
ACLNN_ERR_PARAM_INVALID	161002	The data types of gradOutput, self, and gradInput are not within the supported range.
		The dimensional relationship among gradOutput, self, and gradInput does not satisfy the broadcast principle.
		The data types of gradOutput, self, and gradInput do not satisfy the data type deduction rules.
		The value of approximate is not within the supported range.
		The shape is inconsistent with the shape after broadcasting gradOutput and self.

aclnnGeluBackwardV2

Parameters:

Parameter	Input/Output	Description
workspace	Input	Specifies the workspace memory address applied for on the Device side.
workspaceSize	Input	Specifies the workspace size applied for on the Device side, obtained through the first-phase API aclnnGeluBackwardV2GetWorkspaceSize.
executor	Input	Operator executor that 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:
- aclnnGeluBackwardV2 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_gelu_backward_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 shape_size = 1;
  for (auto i : shape) {
    shape_size *= i;
  }
  return shape_size;
}

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 the device-side engine.
  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 side to the device side 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 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.
  // Set deviceId based on the actual device.
  int32_t deviceId = 0;
  aclrtStream stream;
  auto ret = Init(deviceId, &stream);
  CHECK_RET(ret == 0, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);
  // 2. Construct the input and output. Customize the construction based on the API requirements.
  std::vector<int64_t> selfShape = {4, 2};
  std::vector<int64_t> gradOutputShape = {4, 2};
  std::vector<int64_t> gradInputShape = {4, 2};
  void* selfDeviceAddr = nullptr;
  void* gradOutputDeviceAddr = nullptr;
  void* gradInputDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclTensor* gradOutput = nullptr;
  aclTensor* gradInput = nullptr;
  std::vector<float> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7};
  std::vector<int> gradOutputHostData = {1, 1, 1, 1, 1, 1, 1, 1};
  std::vector<int> gradInputHostData = {0, 0, 0, 0, 0, 0, 0, 0};

  char *approximate = "tanh";
  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  ret = CreateAclTensor(gradOutputHostData, gradOutputShape, &gradOutputDeviceAddr, aclDataType::ACL_FLOAT, &gradOutput);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  ret = CreateAclTensor(gradInputHostData, gradInputShape, &gradInputDeviceAddr, aclDataType::ACL_FLOAT, &gradInput);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // 3. Call the CANN operator library API. Replace with the specific API.
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // Call the first-phase API of aclnnGeluBackwardV2.
  ret = aclnnGeluBackwardV2GetWorkspaceSize(gradOutput, self, approximate, gradInput, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnGeluBackwardV2GetWorkspaceSize 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 aclnnGeluBackwardV2.
  ret = aclnnGeluBackwardV2(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnGeluBackwardV2 failed. ERROR: %d\n", ret); return ret);
  // 4. (Boilerplate) Wait for the task execution to finish synchronously.
  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(gradInputShape);
  std::vector<float> resultData(size, 0);
  ret = aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), gradInputDeviceAddr, size * sizeof(float),
                    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(gradOutput);
  aclDestroyTensor(self);
  aclDestroyTensor(gradInput);

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