Hansung Kim 221d5f75c2 flash: Optimize smem alloc for tcore for 8banks ...

Divide into first half & last half for warpgroup 0 & 1, and
allocate Q/K and P/V in different banks for parallel acccess.

2024-09-19 21:31:39 -07:00

..

.gitignore

flash: gitignore

2024-08-15 21:04:59 -07:00

common.h

flash: Change kernel arg to contain qkv; strip stimulus gen from host code

2024-08-15 21:03:02 -07:00

flash_impl.hpp

flash: Optimize smem alloc for tcore for 8banks

2024-09-19 21:31:39 -07:00

half.hpp

Add flash attention kernel skeleton

2024-08-14 20:46:09 -07:00

kernel.cpp

flash: Optimize smem alloc for tcore for 8banks

2024-09-19 21:31:39 -07:00

kernel.gemmini.cpp

flash: Revert to gemmini config, remove DEBUG and unnecessary checks

2024-09-12 14:25:33 -07:00

main.cpp

flash: Fix load addr for V tile; test with seqlen=128

2024-08-20 14:34:09 -07:00

Makefile

flash: Fix online softmax for DMA layout

2024-09-07 23:21:28 -07:00