891202d458
* Update yolov5.cpp implement upsample with IResizeLayer * Update yolov5.cpp repair mistake and test it * add FasterRcnn(R50C4) maskrcnn is working * add FasterRcnn(R50C4) adjust indent * update README adjust coding style
41 lines
1.1 KiB
C++
41 lines
1.1 KiB
C++
#pragma once
|
|
|
|
#include <cuda_runtime_api.h>
|
|
#include <stdexcept>
|
|
#include <cstdint>
|
|
|
|
#define CUDA_ALIGN 256
|
|
|
|
template <typename T>
|
|
inline size_t get_size_aligned(size_t num_elem) {
|
|
size_t size = num_elem * sizeof(T);
|
|
size_t extra_align = 0;
|
|
if (size % CUDA_ALIGN != 0) {
|
|
extra_align = CUDA_ALIGN - size % CUDA_ALIGN;
|
|
}
|
|
return size + extra_align;
|
|
}
|
|
|
|
template <typename T>
|
|
inline T *get_next_ptr(size_t num_elem, void *&workspace, size_t &workspace_size) {
|
|
size_t size = get_size_aligned<T>(num_elem);
|
|
if (size > workspace_size) {
|
|
throw std::runtime_error("Workspace is too small!");
|
|
}
|
|
workspace_size -= size;
|
|
T *ptr = reinterpret_cast<T *>(workspace);
|
|
workspace = reinterpret_cast<void *>(reinterpret_cast<uintptr_t>(workspace) + size);
|
|
return ptr;
|
|
}
|
|
|
|
#ifndef CUDA_CHECK
|
|
#define CUDA_CHECK(callstr)\
|
|
{\
|
|
cudaError_t error_code = callstr;\
|
|
if (error_code != cudaSuccess) {\
|
|
std::cerr << "CUDA error " << error_code << " at " << __FILE__ << ":" << __LINE__;\
|
|
assert(0);\
|
|
}\
|
|
}
|
|
#endif // CUDA_CHECK
|