|
Uni20
Asynchronous tensor and runtime library.
|
CUDA-specific awaiters for device and runtime resource scheduling. More...
#include <uni20/async/cuda_task.hpp>#include <uni20/backend/cuda/resource_pool.hpp>#include <uni20/backend/cuda/runtime.hpp>#include <uni20/config.hpp>#include "cuda_error.hpp"#include <cuda_runtime_api.h>#include <atomic>Go to the source code of this file.
Classes | |
| class | uni20::async::CudaDeviceAwaiter |
| Awaiter that establishes or changes a CUDA task's device affinity. More... | |
| class | uni20::async::CudaResourceAcquireAwaiter< Resource > |
| CUDA-task awaiter for one provider resource from a fixed-capacity pool. More... | |
| class | uni20::async::CudaStreamAcquireAwaiter |
| CUDA-task awaiter for an actually-idle stream-pool lease. More... | |
Namespaces | |
| namespace | uni20 |
| namespace | uni20::async |
| This represents the reverse-mode adjoint for functions of complex variables, assuming a real-valued scalar loss function ( L \colon \mathbb{C} \to \mathbb{R} ). | |
| namespace | uni20::cuda |
CUDA-specific awaiters for device and runtime resource scheduling.