mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-11 08:50:17 +00:00

Files

Jeff Huang e3556fed04 Optimize batch prefill kernel performance for VECTORIZED_LAYOUT KV cache (#3657 )

- Add multi-dimensional page index support (YsGatherDims) in tile_scatter_gather
- Add is_gather_dim() and get_gather_index() for multi-dim page lookup
- Override MakeVDramTileDistribution() for VECTORIZED_LAYOUT to match
  GEMM's BWarpDstrEncoding (K decomposition: {K2, K0, K1})
- Add GetGemmKDecomposition() to retrieve kABKLane and kKPerThread
- Add static_assert for RowMajor VLayout requirement in batch prefill

Co-authored-by: Po Yen Chen <PoYen.Chen@amd.com>

2026-01-29 07:18:41 +08:00

algorithm

Shuffle fix for gfx950 (#3491 )

2026-01-13 09:21:29 -08:00

arch

[CK_TILE] ABQuant New Preshuffle (#3638 )

2026-01-27 23:46:49 -08:00

container

Shuffle fix for gfx950 (#3491 )

2026-01-13 09:21:29 -08:00

numeric

fix some syntax errors (#3658 )

2026-01-27 09:59:39 -08:00

tensor

Optimize batch prefill kernel performance for VECTORIZED_LAYOUT KV cache (#3657 )

2026-01-29 07:18:41 +08:00

utility

Revert "Revert " Fp8 block scale quantization for fmha fwd (#3330 )" (#3633 )" (#3635 )

2026-01-23 09:03:22 -08:00

config.hpp

Add support to gfx1153 and fix gfx115X WMMA config (#3496 )

2026-01-05 10:03:30 -08:00

README.md

introducing ck_tile! (#1216 )

2024-04-15 19:27:12 -05:00

README.md

ck_tile/core

ck_tile/core contains every basic functions and structures to create a GPU kernel using ck_tile. User should only include ck_tile/core.hpp this single header to use all the functionality. Everything is under ck_tile namespace. The coding style under this folder should be similar to std (snake_case for structure/function, Camel for template types...)

algorithm/
    coordinate transform and some other reusable algorithm
arch/
    contains some basic device building block like mma, buffer addressing, etc...
container/
    contains basic container data structure, array/sequence/tuple/...
numeric/
    data type, and data type related math
tensor/
    tensor descriptors and tile level API
utility/
    other utility function for both host/device