aclnnCummax
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：
- 第一段接口：aclnnStatus aclnnCummaxGetWorkspaceSize(const aclTensor *self, int64_t dim, aclTensor *valuesOut, aclTensor* indicesOut, uint64_t *workspaceSize, aclOpExecutor **executor)
- 第二段接口：aclnnStatus aclnnCummax(void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, aclrtStream stream)
功能描述
aclnnCummaxGetWorkspaceSize
- 接口定义：
aclnnStatus aclnnCummaxGetWorkspaceSize(const aclTensor *self, int64_t dim, aclTensor *valuesOut, aclTensor* indicesOut, uint64_t *workspaceSize, aclOpExecutor **executor)
- 参数说明：
- self：Device侧的aclTensor，输入张量，数据类型支持FLOAT、FLOAT16、DOUBLE、INT32、INT64、INT16、INT8、UINT8、BOOL，支持非连续的Tensor，数据格式支持ND，且shape需要与valuesOut、indicesOut一致。
- dim：Host侧INT64类型，指定要进行最大值计算的维度，取值范围[-self.dim(), self.dim()）。
- valuesOut：Device侧的aclTensor，输出的最大值，数据类型支持FLOAT、FLOAT16、DOUBLE、INT32、INT64、INT16、INT8、UINT8、BOOL，支持非连续的Tensor，数据格式支持ND，且shape需要与self、indicesOut一致。
- indicesOut：Device侧的aclTensor，输出的最大值索引，数据类型支持INT32、INT64，支持非连续的Tensor，数据格式支持ND，且shape需要与self、valuesOut一致。
- workspaceSize：返回用户需要在Device侧申请的workspace大小。
- executor：返回op执行器，包含了算子计算流程。
- 返回值：
返回aclnnStatus状态码，具体参见aclnn返回码。
第一段接口完成入参校验，出现以下场景时报错：
- 返回161001（ACLNN_ERR_PARAM_NULLPTR）：传入的self、valuesOut、indicesOut是空指针。
- 返回161002（ACLNN_ERR_PARAM_INVALID）：
- self、valuesOut、indicesOut的数据类型不在支持的范围之内。
- self、valuesOut、indicesOut的shape不在支持的范围之内。
- 当self为0维时，不支持传入dim。
- 输入的dim值不合法。
aclnnCummax
- 接口定义：
aclnnStatus aclnnCummax(void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, aclrtStream stream)
- 参数说明：
- workspace：在Device侧申请的workspace内存起址。
- workspaceSize：在Device侧申请的workspace大小，由第一段接口aclnnCummaxGetWorkspaceSize获取。
- executor：op执行器，包含了算子计算流程。
- stream：指定执行任务的AscendCL stream流。
- 返回值：
返回aclnnStatus状态码，具体参见aclnn返回码。
调用示例
#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_cummax.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, aclrtContext* context, aclrtStream* stream) {
// 固定写法，AscendCL初始化
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 = 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);
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);
// 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;
}
int main() {
// 1. （固定写法）device/context/stream初始化，参考AscendCL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_RET(ret == ACL_SUCCESS, 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> selfShape = {8};
std::vector<int64_t> valuesOutShape = {8};
std::vector<int64_t> indicesOutShape = {8};
void* selfDeviceAddr = nullptr;
void* valuesOutDeviceAddr = nullptr;
void* indicesOutDeviceAddr = nullptr;
aclTensor* self = nullptr;
aclTensor* valuesOut = nullptr;
aclTensor* indicesOut = nullptr;
std::vector<float> selfHostData = {3.0, 3.0, 2.0, 1.0, 3.0, 2.0, 6.0, 7.0};
std::vector<float> valuesOutHostData(8, 0.0);
std::vector<int64_t> indicesOutHostData(8, 0);
int64_t dim = 0;
// 创建self aclTensor
ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
CHECK_RET(ret == ACL_SUCCESS, return ret);
// 创建valuesOut aclTensor
ret = CreateAclTensor(valuesOutHostData, valuesOutShape, &valuesOutDeviceAddr, aclDataType::ACL_FLOAT, &valuesOut);
CHECK_RET(ret == ACL_SUCCESS, return ret);
// 创建indicesOut aclTensor
ret = CreateAclTensor(indicesOutHostData, indicesOutShape, &indicesOutDeviceAddr, aclDataType::ACL_INT64, &indicesOut);
CHECK_RET(ret == ACL_SUCCESS, return ret);
// 3. 调用CANN算子库API，需要修改为具体的API名称
uint64_t workspaceSize = 0;
aclOpExecutor* executor;
// 调用aclnnCummax第一段接口
ret = aclnnCummaxGetWorkspaceSize(self, dim, valuesOut, indicesOut, &workspaceSize, &executor);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnCummaxGetWorkspaceSize 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aclnnCummax第二段接口
ret = aclnnCummax(workspaceAddr, workspaceSize, executor, stream);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnCummax 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
// 获取valuesOut
auto size = GetShapeSize(valuesOutShape);
std::vector<float> resultData(size, 0);
ret = aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), valuesOutDeviceAddr,
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]);
}
// 获取indicesOut
auto indicesSize = GetShapeSize(indicesOutShape);
std::vector<int64_t> indicesResultData(indicesSize, 0);
ret = aclrtMemcpy(indicesResultData.data(), indicesResultData.size() * sizeof(indicesResultData[0]), indicesOutDeviceAddr,
indicesSize * sizeof(indicesResultData[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 < indicesSize; i++) {
LOG_PRINT("result[%ld] is: %ld\n", i, indicesResultData[i]);
}
// 6. 释放aclTensor和aclScalar，需要根据具体API的接口定义修改
aclDestroyTensor(self);
aclDestroyTensor(valuesOut);
aclDestroyTensor(indicesOut);
// 7. 释放divice 资源，需要根据具体API的接口定义修改
aclrtFree(selfDeviceAddr);
aclrtFree(valuesOutDeviceAddr);
aclrtFree(indicesOutDeviceAddr);
if (workspaceSize > 0) {
aclrtFree(workspaceAddr);
}
aclrtDestroyStream(stream);
aclrtDestroyContext(context);
aclrtResetDevice(deviceId);
aclFinalize();
return 0;
}
父主题： NN类算子接口