Crate ghostflow_cuda

Crate ghostflow_cuda 

Source
Expand description

GhostFlow CUDA Backend - Real GPU Acceleration

This module provides real CUDA GPU acceleration when compiled with the cuda feature. Without the feature, it provides CPU fallback implementations.

Re-exports§

pub use device::CudaDevice;
pub use device::DeviceGuard;
pub use device::get_all_devices;
pub use device::select_best_device;
pub use memory::GpuMemoryPool;
pub use memory::GpuTensor;
pub use memory::get_global_gpu_pool;
pub use stream::CudaStream;
pub use stream::CudaEvent;
pub use stream::CudaTimer;
pub use tensor::CudaTensor;
pub use error::CudaError;
pub use error::CudaResult;
pub use blas::CuBlas;

Modules§

blas
cuBLAS wrapper for linear algebra operations
device
CUDA device management - Real Implementation
error
CUDA error types
ffi
CUDA FFI bindings - Real CUDA Runtime API
kernels
CUDA kernel definitions
memory
GPU memory management
ops
CUDA tensor operations
stream
CUDA streams for async execution - Real Implementation
tensor
CUDA tensor type - Real GPU tensor implementation

Functions§

cuda_version
Get CUDA version
current_device
Get current CUDA device ID
device_count
Get number of available CUDA devices
empty_cache
Empty CUDA cache (free cached memory)
init
Initialize CUDA runtime
is_available
Check if CUDA is available at runtime
memory_info
Memory info for current device
set_device
Set current CUDA device
synchronize
Synchronize all CUDA operations on current device