Skip to content

cpp.md

Latest commit

 

History

History
82 lines (62 loc) · 3.2 KB

cpp.md

File metadata and controls

82 lines (62 loc) · 3.2 KB
Raw

C++ and CMake

Include apxchol.h and link the apxchol::core CMake target:

#include "apxchol.h"

apxchol::solver solver(A);            // A: apxchol::csc_view of CSC arrays
auto result = solver.solve(b, 1e-8);  // b: std::vector<double>; check result.converged

The basic example shows one-shot, reusable and batched solves. Using the solver covers the solver itself, and the C++ reference lists every option.

Add the library to a CMake project in one of three ways, described below: build it from a source checkout, embed the repository root, or use an installed SDK such as a prebuilt SDK archive.

Use from CMake

For a source checkout, call the helper in your consuming project's top-level CMakeLists.txt, after its project(...) declaration:

include(path/to/apxchol/cmake/apxchol_core.cmake)
apxchol_add_core()
target_link_libraries(my_program PRIVATE apxchol::core)

This builds just the library and enables any GPU language in the consuming project's scope. Set APXCHOL_DEVICE when configuring the consuming project. This builds portable code by default; CUDA consumers must also choose CMAKE_CUDA_ARCHITECTURES for distribution instead of its native default. The core keeps its -march flag private. A -march or -mcpu in CMAKE_CXX_FLAGS (or CMAKE_CXX_FLAGS_<CONFIG>) replaces apxchol's own.

Alternatively, embed the repository root directly:

add_subdirectory(path/to/apxchol)
target_link_libraries(my_program PRIVATE apxchol::core)

Root embedding defaults to the portable library only. Set APXCHOL_BUILD_CLI, APXCHOL_BUILD_TESTS or APXCHOL_BUILD_EXAMPLES before add_subdirectory to opt into those targets. CLI dependencies are fetched only when the CLI is enabled; tests with the CLI disabled still include the operator-input unit tests.

For CUDA or Metal, enable CUDA or OBJCXX in the consuming parent before add_subdirectory, for example with project(my_project LANGUAGES CXX CUDA). The top-level helper above enables the selected language for you.

For a separately installed SDK, use:

find_package(apxchol CONFIG REQUIRED)
target_link_libraries(my_program PRIVATE apxchol::core)

Build and install that SDK with:

cmake -S cmake/core -B build-sdk -DCMAKE_INSTALL_PREFIX=/path/to/prefix
cmake --build build-sdk -j6
cmake --install build-sdk

Configure the consuming project with -DCMAKE_PREFIX_PATH=/path/to/prefix. From 1.0, a version request such as find_package(apxchol 1.0 CONFIG REQUIRED) accepts any installed 1.x release from 1.0 on; a 0.x request, such as 0.3, accepts only that minor release. The shared library's soname carries the major and minor version (currently libapxchol_core.so.0.3), because the binary ABI may change between minor releases; rebuild consumers when you upgrade. CUDA SDK consumers need the CUDA Toolkit. Use compatible compiler, standard-library and OpenMP runtime settings.

ExternalProject_Add builds in a separate project. Use it to build/install cmake/core, then configure the consuming project against that install with find_package. It does not create an immediately linkable apxchol::core target in the parent configuration.