tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-26 19:55:04 +00:00
Code Issues Packages Projects Releases Wiki Activity

CUDA: mul_mat_q RDNA2 tunings (#2910)

Browse Source 
* CUDA: mul_mat_q RDNA2 tunings

* Update ggml-cuda.cu

Co-authored-by: Henri Vasserman <henv@hot.ee>

---------

Co-authored-by: Henri Vasserman <henv@hot.ee>
This commit is contained in:
Johannes Gäßler
2023-09-13 11:20:24 +02:00
committed by GitHub
parent 84e723653c
commit 0a5eebb45d
3 changed files with 426 additions and 45 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
CMakeLists.txt
View File

@ -388,7 +388,6 @@ if (LLAMA_HIPBLAS)
target_compile_definitions(ggml-rocm PRIVATE GGML_CUDA_DMMV_X=${LLAMA_CUDA_DMMV_X})
target_compile_definitions(ggml-rocm PRIVATE GGML_CUDA_MMV_Y=${LLAMA_CUDA_MMV_Y})
target_compile_definitions(ggml-rocm PRIVATE K_QUANTS_PER_ITERATION=${LLAMA_CUDA_KQUANTS_ITER})
target_compile_definitions(ggml-rocm PRIVATE CC_TURING=1000000000)
set_source_files_properties(ggml-cuda.cu PROPERTIES LANGUAGE CXX)
target_link_libraries(ggml-rocm PRIVATE hip::device PUBLIC hip::host roc::rocblas roc::hipblas)