Problems for building cuda_voxelizer in Linux

[halehOssadat]

GLM is mostly headers and it doens't have a library to be linked so there is no need to include it in TARGET_LINK_LIBRARIES and because Cmake can not find it, it is better that its include directory be set by the user the same as trimesh2.

[Forceflow]

GLM can be found in distro repositories. I don't think your suggested solution is the best way to go.

[mvrozanti]

I'm actually having trouble compiling because of that. Took me a while to notice that while I had GLM installed, the target glm::glm was not found and that that was to be expected. @halehOssadat can you please share what you did?

[mvrozanti]

I ended up adding this to CMakeLists.txt I found on this repository which also uses GLM and it worked then:

FIND_PACKAGE(glm QUIET)
IF(glm_FOUND)
  IF(TARGET glm::glm)
    MESSAGE(STATUS "Found glm")
  ELSE()
    # fallback for old glm version in EmuELEC and Nix
    GET_TARGET_PROPERTY(GLM_INCLUDE_DIR glm INTERFACE_INCLUDE_DIRECTORIES)
    MESSAGE(STATUS "Found glm: ${GLM_INCLUDE_DIR}")
    ADD_LIBRARY(glm::glm INTERFACE IMPORTED)
    SET_TARGET_PROPERTIES(glm::glm PROPERTIES
      INTERFACE_INCLUDE_DIRECTORIES ${GLM_INCLUDE_DIR})
  ENDIF()
ELSE()
  # fallback for old glm version in UBPorts
  FIND_PATH(GLM_INCLUDE_DIR
    NAMES glm/glm.hpp
    PATHS ${GLM_ROOT_DIR}/include)
  IF(NOT GLM_INCLUDE_DIR)
    MESSAGE(FATAL_ERROR "glm library missing")
  ENDIF()

  MESSAGE(STATUS "Found glm: ${GLM_INCLUDE_DIR}")
  ADD_LIBRARY(glm::glm INTERFACE IMPORTED)
  SET_TARGET_PROPERTIES(glm::glm PROPERTIES
    INTERFACE_INCLUDE_DIRECTORIES ${GLM_INCLUDE_DIR})
ENDIF()

[Forceflow]

Thank you for commenting.

What distro are you running? It seems to work fine on the build system that Github uses...