aclRfft1D
支持的产品型号
- Atlas A2训练系列产品/Atlas 800I A2推理产品。
接口原型
每个算子分为两段式接口,必须先调用“aclRfft1DGetWorkspaceSize”接口获取计算所需workspace大小以及包含了算子计算流程的执行器,再调用“aclRfft1D”接口执行计算。
- aclnnStatus aclRfft1DGetWorkspaceSize(const aclTensor* self, int64_t n, int64_t dim, int64_t norm, aclTensor* out, uint64_t* workspaceSize, aclOpExecutor** executor)
- aclnnStatus aclRfft1D(void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, aclrtStream stream)
功能描述
算子功能:aclRfft1D计算张量self的RFFT
计算公式:
示例: 假设self为 {1, 2, 3, 4} 则out = {10, 0, -2, 2, -2, 0} = {10 + 0j, -2 + 2j, -2 + 0j} (自定义)
aclRfft1DGetWorkspaceSize
参数说明:
- self(aclTensor*, calculation input): 即公式中的输入。数据类型:FLOAT。数据格式:ND。不支持不连续的Tensor。.
- n(int64_t, calculation optional attribute): 表示信号长度。数据类型:INT64。如果给定,则在计算Rfft1D之前,输入将被补零或修剪到此长度。支持的最大值为262144。
- dim(int64_t, calculation optional attribute): 表示维度。数据类型:INT64。如果给定,则RFFT将应用于指定的维度。支持的值为-1。
- norm(int64_t, calculation optional attribute): 表示归一化模式。数据类型:INT64。默认值为1。1表示不归一化,2表示按1/n归一化,3表示按1/sqrt(n)归一化。
- out (aclTensor*, calculated output): 表示公式中的输出。数据类型:FLOAT。数据格式:ND。不支持不连续的张量。
- workspaceSize (uint64_t, input parameter): NPU设备上申请的workspace大小,在第一段调用aclRfft1DGetWorkspaceSize接口获取。
- executor(aclOpExecutor*, input parameter): 算子执行器,包含算子计算过程。
返回值:
aclnnStatus:返回状态码,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): 1. 输入或输出tensor为空。 返回161002(ACLNN_ERR_PARAM_INVALID): 1. self的数据类型不在支持的范围内。 返回561103(ACLNN_ERR_INNER_NULLPTR): 1. 中间结果为null。 返回561101(ACLNN_ERR_INNER_CREATE_EXECUTOR): 1. 执行者为null。
aclRfft1D
参数说明:
- workspace(void*, 入参):在Device侧申请的workspace内存地址。
- workspaceSize(uint64_t, 入参):在Device侧申请的workspace大小,由第一段接口aclRfft1DGetWorkspaceSize获取。
- executor(aclOpExecutor*, 入参):op执行器,包含了算子计算流程。
- stream(aclrtStream, 入参):指定执行任务的AscendCL Stream流。
返回值:
aclnnStatus:返回状态码,具体参见aclnn返回码。
约束与限制
- 第1种情况:支持"n"的范围[1, 4096]和2的n次幂最大262144
- 第2种情况:支持的"dim"值为-1
- 第3种情况:支持的"norm"值为1、2、3
调用示例
#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/acl_rfft1d.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) {
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);
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);
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);
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];
}
*tensor = aclCreateTensor(shape.data(), shape.size(), dataType, strides.data(), 0, aclFormat::ACL_FORMAT_ND,
shape.data(), shape.size(), *deviceAddr);
return 0;
}
int main() {
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);
std::vector<int64_t> selfShape = {1, 1, 8};
std::vector<int64_t> outShape = {1, 1, 5, 2};
void* selfDeviceAddr = nullptr;
void* outDeviceAddr = nullptr;
aclTensor* self = nullptr;
aclTensor* out = nullptr;
std::vector<float> selfHostData = {1, 2, 3, 4, 5, 6, 7, 8};
std::vector<float> outHostData = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0};
ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
CHECK_RET(ret == ACL_SUCCESS, return ret);
ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_FLOAT, &out);
CHECK_RET(ret == ACL_SUCCESS, return ret);
int n = 8;
int dim = -1;
int norm = 1; // backward - 1, forward - 2, ortho - 3
uint64_t workspaceSize = 0;
aclOpExecutor* executor;
ret = aclRfft1DGetWorkspaceSize(self, n, dim, norm, out, &workspaceSize, &executor);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclRfft1DGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
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);
}
ret = aclRfft1D(workspaceAddr, workspaceSize, executor, stream);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclRfft1D failed. ERROR: %d\n", ret); return ret);
ret = aclrtSynchronizeStream(stream);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);
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]);
}
aclDestroyTensor(self);
aclDestroyTensor(out);
aclrtFree(selfDeviceAddr);
aclrtFree(outDeviceAddr);
if (workspaceSize > 0) {
aclrtFree(workspaceAddr);
}
aclrtDestroyStream(stream);
aclrtResetDevice(deviceId);
aclFinalize();
return 0;
}