ltqin c8397e8ef2 flashattention fwd add (80, 96) instance (#3415) ...

* add hdim (96,96) instance

* change to (80,96)

* format py

* remove 96 in optdim

* when N=6 change to llvm_amdgcn_raw_buffer_load_i32x3

[ROCm/composable_kernel commit: 92653168c2]

2025-12-17 09:16:11 -08:00

..

bfloat16.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

e8m0.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

float8.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

half.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

int8.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

integer.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

integral_constant.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

math.hpp

Fix compilation errors with latest clang22 version. (#3396)

2025-12-11 08:09:29 -08:00

mxfp_convert.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

null_type.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

numeric.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

pk_fp4.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

pk_int4.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

type_convert.hpp

chore(copyright): update copyright header for include directory (#3293)

2025-11-26 11:00:05 -07:00

vector_type.hpp

flashattention fwd add (80, 96) instance (#3415)

2025-12-17 09:16:11 -08:00