0 0 0 C++

CUDA: optimize MMQ int8 tensor core performance (#8062)

* CUDA: optimize MMQ int8 tensor core performance

* only a single get_mma_tile_x_k function

* simplify code, make functions constexpr

Johannes Gäßler committed 2y ago

9a590c82262dd518137f85406e65e452fdf2aca3

Parent: 52fc870

Committed by GitHub <noreply@github.com> on 6/24/2024, 10:41:23 AM