NVIDIA
activate_custom_mpi.sh: do not link against mlir
See https://github.com/NVIDIA/cuda-quantum/issues/2892
It might be worth it to further trim down what the ELF file links against, but the other files seem benign for now.
cudaq@2570547-lcedt:/opt/nvidia/cudaq/distributed_interfaces$ readelf -d libcudaq_distributed_interface_mpi.so | grep cudaq
0x000000000000001d (RUNPATH) Library runpath: [/usr/local/llvm/lib:/opt/nvidia/cudaq/lib:/opt/nvidia/cudaq/lib/plugins:/opt/nvidia/cudaq/distributed_interfaces:/usr/local/openmpi/lib64:/usr/local/openmpi/lib]
0x0000000000000001 (NEEDED) Shared library: [libcudaq.so]
0x0000000000000001 (NEEDED) Shared library: [libcudaq-common.so]
0x0000000000000001 (NEEDED) Shared library: [libcudaq-ensmallen.so]
0x0000000000000001 (NEEDED) Shared library: [libcudaq-nlopt.so]
0x0000000000000001 (NEEDED) Shared library: [libcudaq-spin.so]
0x0000000000000001 (NEEDED) Shared library: [libcudaq-operator.so]
0x0000000000000001 (NEEDED) Shared library: [libcudaq-comm-plugin.so]
0x0000000000000001 (NEEDED) Shared library: [libcudaq-pyscf.so]
0x0000000000000001 (NEEDED) Shared library: [libcudaq-em-default.so]
0x0000000000000001 (NEEDED) Shared library: [libcudaq-platform-default.so]
This pull request requires additional validation before any workflows can run on NVIDIA's runners.
Pull request vetters can view their responsibilities here.
Contributors can view more details about this message here.
![copy-pr-bot[bot] avatar](https://avatars.githubusercontent.com/in/354886?v=4 "Published by a pub.dev verified publisher"){kind=small}
![github-actions[bot] avatar](https://avatars.githubusercontent.com/in/15368?v=4 "Published by a pub.dev verified publisher"){kind=small}
I will modify this PR shortly to instead of just dropping the mlir links, add a new flag to nvq++.in. Something like --disable-cudaq-links. This is ultimately what I'd like to do anyways, so let's just do it right the first time rather than having the --disable-mlir-links as a bandaid.