Closed adro79 closed 7 months ago
One of the patches doesn't verify. Did you forget to commit the changes to a patch but commit the new checksum?
Yeah I forgot to update the sums sorry, should be okay now
I think something is still wrong. You just updated 3 checksums, but only 1 was failing for me before. Now the other two are. I suspect you might have missed committing changes to some of the patches, but I can't prove it. In any case the patches as in the repo now do not apply cleanly so I can't even start a test build.
Should be okay now, didn't test it enough.
CMake Error at moonray/hydra/hdMoonray/CMakeLists.txt:42 (find_package):
By not providing "Findpxr.cmake" in CMAKE_MODULE_PATH this project has
asked CMake to find a package configuration file provided by "pxr", but
CMake did not find one.
Could not find a package configuration file provided by "pxr" with any of
the following names:
pxrConfig.cmake
pxr-config.cmake
Add the installation prefix of "pxr" to CMAKE_PREFIX_PATH or set "pxr_DIR"
to a directory containing one of the above files. If "pxr" provides a
separate development package or SDK, be sure it has been installed.
-- Configuring incomplete, errors occurred!
What package is supposed to be supplying pxr stuff?
It's weird that usd wasn't at the dependency list, I hope you can build it now.
-- Found CUDAToolkit: /opt/cuda/targets/x86_64-linux/include (found version "12.3.103")
-- Found OptiX: /opt/optix/include
-- The ISPC compiler identification is unknown
CMake Error at moonray/moonray/CMakeLists.txt:7 (project):
No CMAKE_ISPC_COMPILER could be found.
Tell CMake where to find the compiler by setting either the environment
variable "ISPC" or the CMake cache entry CMAKE_ISPC_COMPILER to the full
path to the compiler, or to the compiler name if it is in the PATH.
We had usd
and some other things we just hit again before as makedepends
. You dropped them at some point, and now they are coming back in dependes
. I'm not too worried about it at this stage while we try to get something that actually builds, but it would be nice to actually keep things that are only build-time dependencies separate.
I get a little farther now, but now hit this:
-- Looking for TBB components tbb; (tbbmalloc;tbbmalloc_proxy;tbbbind;tbbpreview)
-- Found existing TBB component targets: tbb;tbbmalloc;tbbmalloc_proxy
-- Found TBB version 2021.11 at /usr
-- Found Boost: /lib/cmake/Boost-1.83.0/BoostConfig.cmake (found version "1.83.0") found components: system python date_time chrono
-- Found Boost: /lib/cmake/Boost-1.83.0/BoostConfig.cmake (found version "1.83.0") found components: regex
-- Found OpenGL: /usr/lib/libOpenGL.so
-- Could NOT find Mkl (missing: Mkl_INCLUDE_DIR)
-- Found OpenColorIO: /lib/cmake/OpenColorIO/OpenColorIOConfig.cmake (found version "2.3.1")
-- Configuring done (8.3s)
CMake Error in moonray/moonray_gui/cmd/moonray_gui/CMakeLists.txt:
Target "application" contains relative path in its
INTERFACE_INCLUDE_DIRECTORIES:
"Random123_INCLUDE_DIR-NOTFOUND/.."
CMake Error in moonray/hydra/hdMoonray/src/CMakeLists.txt:
Target "rendering_rndr" contains relative path in its
INTERFACE_INCLUDE_DIRECTORIES:
"Random123_INCLUDE_DIR-NOTFOUND/.."
CMake Error in moonray/moonray/lib/application/CMakeLists.txt:
Target "rendering_rndr" contains relative path in its
INTERFACE_INCLUDE_DIRECTORIES:
"Random123_INCLUDE_DIR-NOTFOUND/.."
CMake Error in moonray/moonray/lib/common/mcrt_util/CMakeLists.txt:
Imported target "Random123::Random123" includes non-existent path
"Random123_INCLUDE_DIR-NOTFOUND/.."
in its INTERFACE_INCLUDE_DIRECTORIES. Possible reasons include:
* The path was deleted, renamed, or moved to another location.
* An install or uninstall procedure did not complete successfully.
* The installation package was faulty and references files it does not
provide.
CMake Error in moonray/moonray/lib/rendering/bvh/shading/CMakeLists.txt:
Target "bvh_shading_ispc" contains relative path in its
INTERFACE_INCLUDE_DIRECTORIES:
"Random123_INCLUDE_DIR-NOTFOUND/.."
CMake Error in moonray/moonray/lib/rendering/bvh/shading/ispc/CMakeLists.txt:
Target "rendering_mcrt_common" contains relative path in its
INTERFACE_INCLUDE_DIRECTORIES:
"Random123_INCLUDE_DIR-NOTFOUND/.."
CMake Error in moonray/moonray/lib/rendering/mcrt_common/CMakeLists.txt:
Target "common_time" contains relative path in its
INTERFACE_INCLUDE_DIRECTORIES:
"Random123_INCLUDE_DIR-NOTFOUND/.."
After that I get spammed with about 30 screenfulls of that same message.
I updated the PKGBUILD following the namcap
log, hope it helps.
We're getting into the meat of the build stage now but still not building:
[ 2%] Built target routing
[ 2%] Built target message_impl
[ 2%] Linking CXX shared library librender_logging.so
[ 2%] Built target render_logging
[ 2%] Linking CXX shared library libarras4_athena.so
[ 2%] Linking CXX shared library moonrayShaderParser.so
/usr/bin/ld: cannot find -lar: No such file or directory
/usr/bin/ld: cannot find -lndr: No such file or directory
/usr/bin/ld: cannot find -lsdr: No such file or directory
collect2: error: ld returned 1 exit status
make[2]: *** [moonray/hydra/moonray_sdr_plugins/moonrayShaderParser/CMakeFiles/moonrayShaderParser.dir/build.make:113: moonray/hydra/moonray_sdr_plugins/moonrayShaderParser/moonrayShaderParser.so] Error 1
make[1]: *** [CMakeFiles/Makefile2:10807: moonray/hydra/moonray_sdr_plugins/moonrayShaderParser/CMakeFiles/moonrayShaderParser.dir/all] Error 2
[ 2%] Built target arras4_athena
[ 2%] Linking CXX shared library hdMoonrayAdapters.so
/usr/bin/ld: cannot find -lusdImaging: No such file or directory
collect2: error: ld returned 1 exit status
make[2]: *** [moonray/hydra/hdMoonray/src/adapters/CMakeFiles/hdMoonrayAdapters.dir/build.make:114: moonray/hydra/hdMoonray/src/adapters/hdMoonrayAdapters.so] Error 1
make[1]: *** [CMakeFiles/Makefile2:10685: moonray/hydra/hdMoonray/src/adapters/CMakeFiles/hdMoonrayAdapters.dir/all] Error 2
make: *** [Makefile:146: all] Error 2
make: Leaving directory '/build/moonray/src/build'
I forgot about this, but Moonray actually requieres USD to be build without Monolithic support.
I've built usd (and put it in [extra-testing] with monolithic support disabled (although we need to confirm that blender
works with it built this way before we make it official), but that did not get us all the way either. We're a couple percent farther into the build and I get this:
/build/moonray/src/openmoonray/moonray/mcrt_denoise/lib/denoiser/OptixDenoiserImpl.cc: In constructor ‘moonray::denoiser::OptixDenoiserImpl::OptixDenoiserImpl(int, int, bool, bool, std::string*)’:
/build/moonray/src/openmoonray/moonray/mcrt_denoise/lib/denoiser/OptixDenoiserImpl.cc:110:36: error: ‘OPTIX_DENOISER_ALPHA_MODE_COPY’ was not declared in this scope
110 | mDenoiserParams.denoiseAlpha = OPTIX_DENOISER_ALPHA_MODE_COPY; // don't denoise alpha
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Is there something special you did with the optix package as well?
The patches on the optix.patch are made to support optix 7.6 as this is the lastest version from the AUR. If you plan to use the recommended optix version (7.3) you can disable the optix patches
Interesting, the AUR package has something screwed up because the master branch of the packaging is 7.4 but a dangling commit has 7.6. I just successfully built moonray using 7.6, so this is getting somewhere!
Hmm, so that USD change I made made blender unhappy. See comment here.
Hi! Just updated the PKGBUILD to the lastest Moonray version, also I got able to compile with GCC and TBB.