tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-26 11:45:21 +00:00
Code Issues Packages Projects Releases Wiki Activity

CUDA: refactor mmq, dmmv, mmvq (#7716)

Browse Source 
* CUDA: refactor mmq, dmmv, mmvq

* fix out-of-bounds write

* struct for qk, qr, qi

* fix cmake build

* mmq_type_traits
This commit is contained in:
Johannes Gäßler
2024-06-05 16:53:00 +02:00
committed by GitHub
parent 2b3389677a
commit 7d1a378b8f
112 changed files with 1783 additions and 1767 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
CMakeLists.txt
View File

@ -416,6 +416,8 @@ if (LLAMA_CUDA)
list(APPEND GGML_SOURCES_CUDA "ggml-cuda.cu")
file(GLOB SRCS "ggml-cuda/template-instances/fattn-wmma*.cu")
list(APPEND GGML_SOURCES_CUDA ${SRCS})
file(GLOB SRCS "ggml-cuda/template-instances/mmq*.cu")
list(APPEND GGML_SOURCES_CUDA ${SRCS})
add_compile_definitions(GGML_USE_CUDA)
add_compile_definitions(GGML_CUDA_USE_GRAPHS)
@ -588,6 +590,8 @@ if (LLAMA_HIPBLAS)
list(APPEND GGML_SOURCES_ROCM "ggml-cuda.cu")
file(GLOB SRCS "ggml-cuda/template-instances/fattn-wmma*.cu")
list(APPEND GGML_SOURCES_ROCM ${SRCS})
file(GLOB SRCS "ggml-cuda/template-instances/mmq*.cu")
list(APPEND GGML_SOURCES_ROCM ${SRCS})
add_compile_definitions(GGML_USE_HIPBLAS GGML_USE_CUDA)