Aman Gupta 621764b1a5 CUDA: Add mul_mat_id support for the mmf kernel (llama/15767) ...

* CUDA: Add mul_mat_id support the mmf

Add support for mul_mat_id for bs < 16

* Review: use warp_size, fix should_use_mmf condition

* Launch one block per expert, stride along n_expert_used

* templatize mul_mat_id

* Pad shmem to 16 bytes, add helper function mul_mat_f_switch_ids

* Reduce compile times by dividing mmf into f16, bf16 and f32 variants

* Divide mmf by ncols_dst

* Add missing files

* Fix MUSA/HIP builds

2025-09-20 13:42:52 +03:00

..

cmake

ggml: Skip backend library linking code when GGML_BACKEND_DL=ON (llama/15094)

2025-08-18 20:30:45 +03:00

include

cuda : fix supports_op condition for get_rows when number of blocks is too large (llama/15868)

2025-09-20 13:42:52 +03:00

src

CUDA: Add mul_mat_id support for the mmf kernel (llama/15767)

2025-09-20 13:42:52 +03:00

.gitignore

whisper : reorganize source code + improve CMake (#2256)

2024-06-26 19:34:09 +03:00

CMakeLists.txt

ggml-cpu: drop support for nnpa intrinsics (llama/15821)

2025-09-20 13:42:50 +03:00