Gahow Wang
0e5c7d22e2
Route Tensor::matmul and matmul_backward through cuBLAS Sgemm instead of the hand-written tiled kernel. fp32 → same GEMM up to rounding order, so the T3 cuBLAS tolerance and downstream grad-checks are preserved. - cublas.rs: thread-local persistent handle + row-major sgemm helper with transpose flags (col-major⟺row-major as the T3 oracle does). - matmul_backward: dA/dB via cuBLAS OP_T, dropping the two transpose kernels + their allocations the T3 version ran. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
10 lines
161 B
Rust
10 lines
161 B
Rust
#[cfg(not(no_cuda))]
|
|
pub mod cublas;
|
|
pub mod device;
|
|
pub mod error;
|
|
pub mod ffi;
|
|
pub mod memory;
|
|
|
|
pub use error::{CudaError, Result};
|
|
pub use memory::GpuBuffer;
|