cmake_minimum_required(VERSION 3.21)

project(Ollama C CXX)

include(CheckLanguage)

find_package(Threads REQUIRED)

set(CMAKE_BUILD_TYPE Release)
set(BUILD_SHARED_LIBS ON)

set(CMAKE_CXX_STANDARD 17)
set(CMAKE_CXX_STANDARD_REQUIRED ON)
set(CMAKE_CXX_EXTENSIONS OFF)

set(GGML_BUILD ON)
set(GGML_SHARED ON)
set(GGML_CCACHE ON)
set(GGML_BACKEND_DL ON)
set(GGML_BACKEND_SHARED ON)
set(GGML_SCHED_MAX_COPIES 4)

set(GGML_LLAMAFILE ON)
set(GGML_CPU_ALL_VARIANTS ON)
set(GGML_CUDA_PEER_MAX_BATCH_SIZE 128)
set(GGML_CUDA_GRAPHS ON)

set(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${CMAKE_BINARY_DIR}/lib)
set(CMAKE_LIBRARY_OUTPUT_DIRECTORY ${CMAKE_BINARY_DIR}/lib)

include_directories(${CMAKE_CURRENT_SOURCE_DIR}/ml/backend/ggml/ggml/src)
include_directories(${CMAKE_CURRENT_SOURCE_DIR}/ml/backend/ggml/ggml/src/include)
include_directories(${CMAKE_CURRENT_SOURCE_DIR}/ml/backend/ggml/ggml/src/ggml-cpu)
include_directories(${CMAKE_CURRENT_SOURCE_DIR}/ml/backend/ggml/ggml/src/ggml-cpu/amx)

set(GGML_CPU ON)
add_subdirectory(${CMAKE_CURRENT_SOURCE_DIR}/ml/backend/ggml/ggml/src)
set_property(TARGET ggml PROPERTY EXCLUDE_FROM_ALL TRUE)

check_language(CUDA)
if(CMAKE_CUDA_COMPILER)
    if(CMAKE_VERSION VERSION_GREATER_EQUAL "3.24" AND NOT CMAKE_CUDA_ARCHITECTURES)
        set(CMAKE_CUDA_ARCHITECTURES "native")
    endif()

    add_subdirectory(${CMAKE_CURRENT_SOURCE_DIR}/ml/backend/ggml/ggml/src/ggml-cuda)
endif()

check_language(HIP)
if(CMAKE_HIP_COMPILER)
    add_subdirectory(${CMAKE_CURRENT_SOURCE_DIR}/ml/backend/ggml/ggml/src/ggml-hip)
endif()