| .. |
|
fmha
|
Parallelize CUDA bwd along seqlen_k instead of seqlen_q
|
2022-11-05 16:26:17 -07:00 |
|
fmha_block_dgrad_fp16_kernel_loop.sm80.cu
|
Implement cross attention
|
2022-07-03 17:48:12 -07:00 |
|
fmha_block_dgrad_kernel_1xN_loop.h
|
Support all head dims that are multiples of 8, up to 128
|
2022-10-24 16:04:21 -07:00 |
|
fmha_block_fprop_fp16_kernel.sm80.cu
|
Implement cross attention
|
2022-07-03 17:48:12 -07:00 |
|
fmha_block_fprop_kernel_1xN.h
|
Support all head dims that are multiples of 8, up to 128
|
2022-10-24 16:04:21 -07:00 |
|
fmha_blockmask.h
|
Implement cross attention
|
2022-07-03 17:48:12 -07:00 |
|
fmha_bwd_hdim32.cu
|
Simplify BOOL_SWITCH macro to fix compiling error on gcc 7
|
2022-12-06 14:38:32 -08:00 |
|
fmha_bwd_hdim64.cu
|
Simplify BOOL_SWITCH macro to fix compiling error on gcc 7
|
2022-12-06 14:38:32 -08:00 |
|
fmha_bwd_hdim128.cu
|
Simplify BOOL_SWITCH macro to fix compiling error on gcc 7
|
2022-12-06 14:38:32 -08:00 |
|
fmha_bwd_launch_template.h
|
Simplify BOOL_SWITCH macro to fix compiling error on gcc 7
|
2022-12-06 14:38:32 -08:00 |
|
fmha_dgrad_kernel_1xN_loop.h
|
Fix out-of-bound memory read
|
2022-11-09 09:34:14 -08:00 |
|
fmha_fprop_kernel_1xN.h
|
Fix out-of-bound memory read
|
2022-11-09 09:34:14 -08:00 |
|
fmha_fwd_hdim32.cu
|
Simplify BOOL_SWITCH macro to fix compiling error on gcc 7
|
2022-12-06 14:38:32 -08:00 |
|
fmha_fwd_hdim64.cu
|
Simplify BOOL_SWITCH macro to fix compiling error on gcc 7
|
2022-12-06 14:38:32 -08:00 |
|
fmha_fwd_hdim128.cu
|
Simplify BOOL_SWITCH macro to fix compiling error on gcc 7
|
2022-12-06 14:38:32 -08:00 |
|
fmha_fwd_launch_template.h
|
Simplify BOOL_SWITCH macro to fix compiling error on gcc 7
|
2022-12-06 14:38:32 -08:00 |
|
fmha_kernel.h
|
Speed up compilation by splitting into separate .cu files
|
2022-11-25 16:30:18 -08:00 |
|
fmha_utils.h
|
Implement for bf16
|
2022-07-09 23:31:56 -07:00 |
|
fmha.h
|
Simplify BOOL_SWITCH macro to fix compiling error on gcc 7
|
2022-12-06 14:38:32 -08:00 |
|
philox.cuh
|
Rework dropout to decouple forward and backward
|
2022-10-21 12:04:27 -07:00 |
|
static_switch.h
|
Simplify BOOL_SWITCH macro to fix compiling error on gcc 7
|
2022-12-06 14:38:32 -08:00 |
