aclnnLayerNormBackward

接口原型

每个算子有两段接口，必须先调用“aclnnXxxGetWorkspaceSize”接口获取入参并根据计算流程计算所需workspace大小，再调用“aclnnXxx”接口执行计算。两段式接口如下：

功能描述

算子功能：aclnnLayerNorm的反向传播。

aclnnLayerNormBackwardGetWorkspaceSize

aclnnLayerNormBackward

约束与限制

无

调用示例

#include <vector>
#include "acl/acl.h"
#include "aclnnop/level2/aclnn_layer_norm_backward.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 GetIntShapeSize(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, aclrtContext* context, aclrtStream* stream) {
  // 固定写法，acl初始化
  auto ret = aclrtSetDevice(deviceId);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSetDevice failed. ERROR: %d\n", ret); return ret);
  ret = aclrtCreateContext(context, deviceId);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtCreateContext failed. ERROR: %d\n", ret); return ret);
  ret = aclrtSetCurrentContext(*context);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSetCurrentContext 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);

  ret = aclInit(nullptr);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclInit 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 = GetIntShapeSize(shape) * sizeof(T);
  // 调用aclrtMalloc申请device侧内存
  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);

  // 调用aclrtMemcpy将host侧数据拷贝到device侧内存上
  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);

  // 计算连续tensor的strides
  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];
  }

  // 调用aclCreateTensor接口创建aclTensor
  *tensor = aclCreateTensor(shape.data(), shape.size(), dataType, strides.data(), 0, aclFormat::ACL_FORMAT_ND,
                            shape.data(), shape.size(), *deviceAddr);
  return 0;
}

template <typename T>
int CreateAclIntArray(const std::vector<T>& hostData, void** deviceAddr, aclIntArray** intArray) {
  auto size = GetIntShapeSize(hostData) * sizeof(T);
  // 调用aclrtMalloc申请device侧内存
  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);

  // 调用aclrtMemcpy将host侧数据拷贝到device侧内存上
  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);

  // 调用aclCreateTensor接口创建aclTensor
  *intArray = aclCreateIntArray(hostData.data(), hostData.size());
  return 0;
}

int main() {
  // 1. （固定写法）device/context/stream初始化, 参考acl对外接口列表
  // 根据自己的实际device填写deviceId
  int32_t deviceId = 0;
  aclrtContext context;
  aclrtStream stream;
  auto ret = Init(deviceId, &context, &stream);
  // check根据自己的需要处理
  CHECK_RET(ret == 0, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);

  // 2. 构造输入与输出，需要根据API的接口自定义构造
  std::vector<int64_t> xShape = {2, 2};
  std::vector<int64_t> meanShape = {2, 1};
  std::vector<int64_t> normShape = {2};
  void* dyDeviceAddr = nullptr;
  void* xDeviceAddr = nullptr;
  void* normShapeAddr = nullptr;
  void* meanDeviceAddr = nullptr;
  void* rstdDeviceAddr = nullptr;
  void* weightDeviceAddr = nullptr;
  void* biasDeviceAddr = nullptr;
  void* maskAddr = nullptr;
  void* outDeviceAddr = nullptr;
  void* dwDeviceAddr = nullptr;
  void* dbDeviceAddr = nullptr;
  aclTensor* dy = nullptr;
  aclTensor* x = nullptr;
  aclIntArray* norm = nullptr; 
  aclTensor* mean = nullptr;
  aclTensor* rstd = nullptr;
  aclTensor* weight = nullptr;
  aclTensor* bias = nullptr;
  aclBoolArray* mask = nullptr;
  aclTensor* out = nullptr;
  aclTensor* dw = nullptr;
  aclTensor* db = nullptr;

  std::vector<float> dyHostData = {2,3,4,5};
  std::vector<float> xHostData = {2,3,4,5};
  std::vector<int64_t> normData = {2};
  std::vector<float> meanHostData = {2, 3};
  std::vector<float> rstdHostData = {4, 5};
  std::vector<float> weightHostData = {1, 1};
  std::vector<float> biasHostData = {0, 0};
  std::vector<bool> maskData = {true, true, true};
  std::vector<float> outHostData = {0, 0, 0, 0};
  std::vector<float> dwHostData = {0, 0};
  std::vector<float> dbHostData = {0, 0};

  // 创建dy aclTensor
  ret = CreateAclTensor(dyHostData, xShape, &dyDeviceAddr, aclDataType::ACL_FLOAT, &dy);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建x aclTensor
  ret = CreateAclTensor(xHostData, xShape, &xDeviceAddr, aclDataType::ACL_FLOAT, &x);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建normalizedShape aclIntArray
  ret = CreateAclIntArray(normData, &normShapeAddr, &norm);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建mean aclTensor
  ret = CreateAclTensor(meanHostData, meanShape, &meanDeviceAddr, aclDataType::ACL_FLOAT, &mean);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建rstd aclTensor
  ret = CreateAclTensor(rstdHostData, meanShape, &rstdDeviceAddr, aclDataType::ACL_FLOAT, &rstd);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建weight aclTensor
  ret = CreateAclTensor(weightHostData, normShape, &weightDeviceAddr, aclDataType::ACL_FLOAT, &weight);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建bias aclTensor
  ret = CreateAclTensor(biasHostData, normShape, &biasDeviceAddr, aclDataType::ACL_FLOAT, &bias);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建outputmask aclBoolArray
  bool mask_data[3]={true, true, true};
  mask = aclCreateBoolArray(&(mask_data[0]), 3);
  // ret = CreateAclBoolArray(maskData, &maskAddr, &mask);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建out aclTensor
  ret = CreateAclTensor(outHostData, xShape, &outDeviceAddr, aclDataType::ACL_FLOAT, &out);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建dw aclTensor
  ret = CreateAclTensor(dwHostData, normShape, &dwDeviceAddr, aclDataType::ACL_FLOAT, &dw);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建db aclTensor
  ret = CreateAclTensor(dbHostData, normShape, &dbDeviceAddr, aclDataType::ACL_FLOAT, &db);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // 3. 调用CANN算子库API，需要修改为具体的HostApi
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // 调用aclnnLayerNormBackward第一段接口
  ret = aclnnLayerNormBackwardGetWorkspaceSize(dy, x, norm, mean, rstd, weight, bias, mask, out, dw, db, &workspaceSize,&executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnLayerNormBackwardGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
  // 根据第一段接口计算出的workspaceSize申请device内存
  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;);
  }
  // 调用aclnnLayerNormBackward第二段接口
  ret = aclnnLayerNormBackward(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnLayerNormBackward failed. ERROR: %d\n", ret); return ret);

  // 4. （固定写法）同步等待任务执行结束
  ret = aclrtSynchronizeStream(stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);

  // 5. 获取输出的值，将device侧内存上的结果拷贝至host侧，需要根据具体API的接口定义修改
  auto size = GetIntShapeSize(xShape);
  std::vector<float> resultData(size, 0);
  ret = aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), outDeviceAddr, 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]);
  }
  auto size1 = GetIntShapeSize(normShape);
  std::vector<float> resultData1(size1, 0);
  ret = aclrtMemcpy(resultData1.data(), resultData1.size() * sizeof(resultData1[0]), dwDeviceAddr,
                    size1 * 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 < size1; i++) {
    LOG_PRINT("result[%ld] is: %f\n", i, resultData1[i]);
  }
  auto size2 = GetIntShapeSize(normShape);
  std::vector<float> resultData2(size2, 0);
  ret = aclrtMemcpy(resultData2.data(), resultData2.size() * sizeof(resultData2[0]), dbDeviceAddr,
                    size2 * 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 < size2; i++) {
    LOG_PRINT("result[%ld] is: %f\n", i, resultData2[i]);
  }

  // 6. 释放aclTensor和aclIntArray，需要根据具体API的接口定义修改
  aclDestroyTensor(dy);
  aclDestroyTensor(x);
  aclDestroyIntArray(norm);
  aclDestroyTensor(mean);
  aclDestroyTensor(rstd);
  aclDestroyTensor(weight);
  aclDestroyTensor(bias);
  aclDestroyBoolArray(mask);
  aclDestroyTensor(out);
  aclDestroyTensor(dw);
  aclDestroyTensor(db);
  return 0;
}
父主题： NN类算子接口（aclnn）