Iamax
asdBlasIsamax
asdBlasIsamax算子调用示例：
#include <iostream>
#include <vector>
#include "asdsip.h"
#include "acl/acl.h"
#include "acl_meta.h"
using namespace AsdSip;
#define ASD_STATUS_CHECK(err) \
do { \
AsdSip::AspbStatus err_ = (err); \
if (err_ != AsdSip::NO_ERROR) { \
std::cout << "Execute failed." << std::endl; \
exit(-1); \
} \
} while (0)
#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)
{
// 固定写法，acl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 = 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侧数据复制到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(int argc, char **argv)
{
int 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);
int64_t n = 6;
int64_t incx = 1;
int64_t xSize = n;
int64_t ySize = 1;
std::vector<float> tensorInXData;
tensorInXData.reserve(xSize);
for (int64_t i = 0; i < xSize; i++) {
if (i < xSize / 2) {
tensorInXData[i] = 1.0 * i;
} else {
tensorInXData[i] = xSize - 1.0 * i;
}
}
std::vector<int32_t> tensorOutYData;
tensorOutYData.reserve(ySize);
std::cout << "------- input X -------" << std::endl;
for (int64_t i = 0; i < xSize; i++) {
std::cout << tensorInXData[i] << " ";
}
std::cout << std::endl;
std::vector<int64_t> xShape = {xSize};
std::vector<int64_t> yShape = {ySize};
aclTensor *inputX = nullptr;
aclTensor *outputY = nullptr;
void *inputXDeviceAddr = nullptr;
void *outputYDeviceAddr = nullptr;
ret = CreateAclTensor(tensorInXData, xShape, &inputXDeviceAddr, aclDataType::ACL_FLOAT, &inputX);
CHECK_RET(ret == ACL_SUCCESS, return ret);
ret = CreateAclTensor(tensorOutYData, yShape, &outputYDeviceAddr, aclDataType::ACL_INT32, &outputY);
CHECK_RET(ret == ACL_SUCCESS, return ret);
asdBlasHandle handle;
asdBlasCreate(handle);
size_t lwork = 0;
void *buffer = nullptr;
asdBlasMakeIamaxPlan(handle);
asdBlasGetWorkspaceSize(handle, lwork);
std::cout << "lwork = " << lwork << std::endl;
if (lwork > 0) {
ret = aclrtMalloc(&buffer, static_cast<int64_t>(lwork), ACL_MEM_MALLOC_HUGE_FIRST);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("allocate workspace failed. ERROR: %d\n", ret); return ret);
}
asdBlasSetWorkspace(handle, buffer);
asdBlasSetStream(handle, stream);
ASD_STATUS_CHECK(asdBlasIsamax(handle, n, inputX, incx, outputY));
asdBlasSynchronize(handle);
asdBlasDestroy(handle);
ret = aclrtMemcpy(tensorOutYData.data(),
ySize * sizeof(int32_t),
outputYDeviceAddr,
ySize * sizeof(int32_t),
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);
std::cout << "------- result -------" << std::endl;
std::cout << tensorOutYData[0] << std::endl;
std::cout << "Execute successfully." << std::endl;
aclDestroyTensor(inputX);
aclDestroyTensor(outputY);
aclrtFree(inputXDeviceAddr);
aclrtFree(outputYDeviceAddr);
aclrtDestroyStream(stream);
aclrtResetDevice(deviceId);
aclFinalize();
return 0;
}
asdBlasIcamax
asdBlasIcamax算子调用示例：
#include <iostream>
#include <vector>
#include "asdsip.h"
#include "acl/acl.h"
#include "acl_meta.h"
using namespace AsdSip;
#define ASD_STATUS_CHECK(err) \
do { \
AsdSip::AspbStatus err_ = (err); \
if (err_ != AsdSip::NO_ERROR) { \
std::cout << "Execute failed." << std::endl; \
exit(-1); \
} \
} while (0)
#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)
{
// 固定写法，acl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 = 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侧数据复制到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(int argc, char **argv)
{
int 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);
int64_t n = 6;
int64_t incx = 1;
int64_t xSize = 6;
int64_t ySize = 1;
std::vector<std::complex<float>> tensorInXData;
tensorInXData.reserve(xSize);
for (int64_t i = 0; i < xSize; i++) {
if (i < xSize / 2) {
tensorInXData[i] = {(float)(1.0 * i), (float)(1.0 * i)};
} else {
tensorInXData[i] = {(float)(xSize - 1.0 * i), (float)(xSize - 1.0 * i)};
}
}
std::vector<int32_t> tensorOutYData;
tensorOutYData.reserve(ySize);
std::cout << "n: " << n << std::endl;
std::cout << "------- input x -------" << std::endl;
for (int64_t i = 0; i < xSize; i++) {
std::cout << tensorInXData[i] << " ";
}
std::cout << std::endl;
std::vector<int64_t> xShape = {xSize};
std::vector<int64_t> yShape = {ySize};
aclTensor *inputX = nullptr;
aclTensor *outputY = nullptr;
void *inputXDeviceAddr = nullptr;
void *outputYDeviceAddr = nullptr;
ret = CreateAclTensor(tensorInXData, xShape, &inputXDeviceAddr, aclDataType::ACL_COMPLEX64, &inputX);
CHECK_RET(ret == ACL_SUCCESS, return ret);
ret = CreateAclTensor(tensorOutYData, yShape, &outputYDeviceAddr, aclDataType::ACL_INT32, &outputY);
CHECK_RET(ret == ACL_SUCCESS, return ret);
asdBlasHandle handle;
asdBlasCreate(handle);
size_t lwork = 0;
void *buffer = nullptr;
asdBlasMakeIamaxPlan(handle);
asdBlasGetWorkspaceSize(handle, lwork);
std::cout << "lwork = " << lwork << std::endl;
if (lwork > 0) {
ret = aclrtMalloc(&buffer, static_cast<int64_t>(lwork), ACL_MEM_MALLOC_HUGE_FIRST);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("allocate workspace failed. ERROR: %d\n", ret); return ret);
}
asdBlasSetWorkspace(handle, buffer);
asdBlasSetStream(handle, stream);
ASD_STATUS_CHECK(asdBlasIcamax(handle, n, inputX, incx, outputY));
asdBlasSynchronize(handle);
asdBlasDestroy(handle);
ret = aclrtMemcpy(tensorOutYData.data(),
ySize * sizeof(int32_t),
outputYDeviceAddr,
ySize * sizeof(int32_t),
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);
std::cout << "------- result -------" << std::endl;
std::cout << tensorOutYData[0] << std::endl;
std::cout << "Execute successfully." << std::endl;
aclDestroyTensor(inputX);
aclDestroyTensor(outputY);
aclrtFree(inputXDeviceAddr);
aclrtFree(outputYDeviceAddr);
aclrtDestroyStream(stream);
aclrtResetDevice(deviceId);
aclFinalize();
return 0;
}
父主题： BLAS