Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

ggml : add GGML_CUDA_USE_GRAPHS option, restore GGML_CUDA_FORCE_CUBLAS #8140

Merged
merged 1 commit into from
Jun 26, 2024
Merged

ggml : add GGML_CUDA_USE_GRAPHS option, restore GGML_CUDA_FORCE_CUBLAS #8140

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -80,6 +80,7 @@ set(GGML_SANITIZE_UNDEFINED ${LLAMA_SANITIZE_UNDEFINED})
set(GGML_ALL_WARNINGS ${LLAMA_ALL_WARNINGS})
set(GGML_FATAL_WARNINGS ${LLAMA_FATAL_WARNINGS})
set(GGML_LLAMAFILE ON)
set(GGML_CUDA_USE_GRAPHS ON)

# transition helpers
function (llama_option_depr TYPE OLD NEW)
Expand Down
2 changes: 2 additions & 0 deletions ggml/CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -109,6 +109,7 @@ option(GGML_LLAMAFILE "ggml: use ggml SGEMM"
option(GGML_CUDA "ggml: use CUDA" OFF)
option(GGML_CUDA_FORCE_DMMV "ggml: use dmmv instead of mmvq CUDA kernels" OFF)
option(GGML_CUDA_FORCE_MMQ "ggml: use mmq kernels instead of cuBLAS" OFF)
option(GGML_CUDA_FORCE_CUBLAS "ggml: always use cuBLAS instead of mmq kernels" OFF)
set (GGML_CUDA_DMMV_X "32" CACHE STRING "ggml: x stride for dmmv CUDA kernels")
set (GGML_CUDA_MMV_Y "1" CACHE STRING "ggml: y block size for mmv CUDA kernels")
option(GGML_CUDA_F16 "ggml: use 16 bit floats for some calculations" OFF)
Expand All @@ -119,6 +120,7 @@ set (GGML_CUDA_PEER_MAX_BATCH_SIZE "128" CACHE STRING
option(GGML_CUDA_NO_PEER_COPY "ggml: do not use peer to peer copies" OFF)
option(GGML_CUDA_NO_VMM "ggml: do not try to use CUDA VMM" OFF)
option(GGML_CUDA_FA_ALL_QUANTS "ggml: compile all quants for FlashAttention" OFF)
option(GGML_CUDA_USE_GRAPHS "ggml: use CUDA graphs (llama.cpp only)" OFF)

option(GGML_CURL "ggml: use libcurl to download model from an URL" OFF)
option(GGML_HIPBLAS "ggml: use hipBLAS" OFF)
Expand Down
5 changes: 4 additions & 1 deletion ggml/src/CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -295,12 +295,15 @@ if (GGML_CUDA)

list(APPEND GGML_CDEF_PUBLIC GGML_USE_CUDA)

add_compile_definitions(GGML_CUDA_USE_GRAPHS)
add_compile_definitions(GGML_CUDA_DMMV_X=${GGML_CUDA_DMMV_X})
add_compile_definitions(GGML_CUDA_MMV_Y=${GGML_CUDA_MMV_Y})
add_compile_definitions(K_QUANTS_PER_ITERATION=${GGML_CUDA_KQUANTS_ITER})
add_compile_definitions(GGML_CUDA_PEER_MAX_BATCH_SIZE=${GGML_CUDA_PEER_MAX_BATCH_SIZE})

if (GGML_CUDA_USE_GRAPHS)
add_compile_definitions(GGML_CUDA_USE_GRAPHS)
endif()

if (GGML_CUDA_FORCE_DMMV)
add_compile_definitions(GGML_CUDA_FORCE_DMMV)
endif()
Expand Down
Loading