Undefined symbol: ncclCommRegister

Hi, I've encountered a msccl issue using the latest nccl/nccl-test/msccl repo for allreduce test.

// msccl install step
git clone https://github.com/microsoft/msccl.git
cd msccl/
make -j src.build
cd ../

// nccl install step
git clone https://github.com/nvidia/nccl-tests.git
cd nccl-tests/
make MPI=1 NCCL_HOME=../msccl/build/ -j
cd ../

// msccl-tools install step
git clone https://github.com/microsoft/msccl-tools.git
cd msccl-tools/
pip install .
cd ../
python msccl-tools/examples/mscclang/allreduce_a100_allpairs.py --protocol=LL 8 2 > test.xml
cd ../

// allreduce test
mpirun -np 8 -x LD_LIBRARY_PATH=msccl/build/lib/:$LD_LIBRARY_PATH -x NCCL_DEBUG=INFO -x NCCL_DEBUG_SUBSYS=INIT,ENV -x MSCCL_XML_FILES=test.xml -x NCCL_ALGO=MSCCL,RING,TREE nccl-tests/build/all_reduce_perf -b 128 -e 32MB -f 2 -g 1 -c 1 -n 100 -w 100 -G 100 -z 0

// Error
nccl-tests/build/all_reduce_perf: symbol lookup error: nccl-tests/build/all_reduce_perf: undefined symbol: ncclCommRegister
nccl-tests/build/all_reduce_perf: symbol lookup error: nccl-tests/build/all_reduce_perf: undefined symbol: ncclCommRegister
nccl-tests/build/all_reduce_perf: symbol lookup error: nccl-tests/build/all_reduce_perf: undefined symbol: ncclCommRegister
nccl-tests/build/all_reduce_perf: symbol lookup error: nccl-tests/build/all_reduce_perf: undefined symbol: ncclCommRegister
nccl-tests/build/all_reduce_perf: symbol lookup error: nccl-tests/build/all_reduce_perf: undefined symbol: ncclCommRegister
nccl-tests/build/all_reduce_perf: symbol lookup error: nccl-tests/build/all_reduce_perf: undefined symbol: ncclCommRegister
nccl-tests/build/all_reduce_perf: symbol lookup error: nccl-tests/build/all_reduce_perf: undefined symbol: ncclCommRegister

Primary job terminated normally, but 1 process returned
a non-zero exit code. Per user-direction, the job has been aborted.

nccl-tests/build/all_reduce_perf: symbol lookup error: nccl-tests/build/all_reduce_perf: undefined symbol: ncclCommRegister

mpirun detected that one or more processes exited with non-zero status, thus causing
the job to be terminated. The first process to do so was:

Process name: [[14987,1],0]
Exit code: 127

Can you help me figure out this issue?

Reproduce the reported setup using the msccl, nccl-tests, and msccl-tools build commands, then run nccl-tests/build/all_reduce_perf with the shown mpirun arguments. Start by examining the all_reduce_perf failure and the linked library environment around LD_LIBRARY_PATH. Done means the allreduce test runs without the undefined symbol error.