ViperEkura 34471252ab perf: pipeline fp8 gemm fragment loads and pack bf16 epilogue ...

- software-pipeline A-fragment ldmatrix: row mt+1 loads hide behind row mt MMAs
- bf16 epilogue packs two columns into one bfloat162 store (half the stores)
- fp8 vs bf16 linear: fwd 1.15x@512, 1.5x@2048, 2.8x@4096; bwd up to 2.7x, peak 35 TFLOPS

2026-08-23 21:32:20 +08:00

..

attention

refactor: reorganize CUDA kernels into per-family directories

2026-08-22 20:40:31 +08:00

common

refactor: reorganize CUDA kernels into per-family directories

2026-08-22 20:40:31 +08:00

fp8

perf: pipeline fp8 gemm fragment loads and pack bf16 epilogue

2026-08-23 21:32:20 +08:00

rotary

refactor: reorganize CUDA kernels into per-family directories

2026-08-22 20:40:31 +08:00