❯ podman run --user 0 --rm -it registry.redhat.io/rhai/base-image-cuda-rhel9:3.2.0-1765367347
Trying to pull registry.redhat.io/rhai/base-image-cuda-rhel9:3.2.0-1765367347...
Getting image source signatures
Checking if image destination supports signatures
Copying blob sha256:11f81ffe9d0743a9f909d61688991d71cff5592d2111bde11e472d4393ba9860
Copying blob sha256:27e7cfc996422af11fa0e0744f1b18263b72fc31cfa467cbcb9486174a364503
ldd /lib64/ucx/libuct_cuda.so.0.0.0
Copying config sha256:ba1a31a3777d93ae6845bd6d6e163c1e350a79f6d663e7f2862dd6d84bfa4e6b
Writing manifest to image destination
Storing signatures
ldd /lib64/ucx/libuct_cuda.so.0.0.0
(app-root) /opt/app-root$ ldd /lib64/ucx/libuct_cuda.so.0.0.0
	linux-vdso.so.1 (0x0000ffff914dc000)
	/usr/lib64/libjemalloc.so.2 (0x0000ffff9135b000)
	libuct.so.0 => /lib64/libuct.so.0 (0x0000ffff91307000)
	libucs.so.0 => /lib64/libucs.so.0 (0x0000ffff9128f000)
	libm.so.6 => /lib64/libm.so.6 (0x0000ffff911ee000)
	libucm.so.0 => /lib64/libucm.so.0 (0x0000ffff911bd000)
	libcuda.so.1 => /usr/local/cuda/compat/libcuda.so.1 (0x0000ffff8b91a000)
	libnvidia-ml.so.1 => not found
	libc.so.6 => /lib64/libc.so.6 (0x0000ffff8b76c000)
	/lib/ld-linux-aarch64.so.1 (0x0000ffff914a0000)
	libdl.so.2 => /lib64/libdl.so.2 (0x0000ffff8b74b000)
	librt.so.1 => /lib64/librt.so.1 (0x0000ffff8b72a000)
	libpthread.so.0 => /lib64/libpthread.so.0 (0x0000ffff8b709000)
(app-root) /opt/app-root$ find / -name "libnvidia-ml.so*"
(app-root) /opt/app-root$ find / -name "libnvidia-ml*"
(app-root) /opt/app-root$ 

SUBFAILED[dlib='/lib64/ucx/libuct_cuda.so.0.0.0'] tests/containers/base_image_test.py::TestBaseImage::test_elf_files_can_link_runtime_libs[ghcr.io/red-hat-data-services/notebooks/workbench-images:cuda-jupyter-tensorflow-ubi9-python-3.12-_2c70a7ff3733116f48d692ab9dd8ac657ace03bd] - Failed: dlib='/lib64/ucx/libuct_cuda.so.0.0.0' has unsatisfied dependencies deps='libnvidia-ml.so.1 => not found'

SUBFAILED[dlib='/lib64/ucx/libuct_cuda_gdrcopy.so.0.0.0'] tests/containers/base_image_test.py::TestBaseImage::test_elf_files_can_link_runtime_libs[ghcr.io/red-hat-data-services/notebooks/workbench-images:cuda-jupyter-tensorflow-ubi9-python-3.12-_2c70a7ff3733116f48d692ab9dd8ac657ace03bd] - Failed: dlib='/lib64/ucx/libuct_cuda_gdrcopy.so.0.0.0' has unsatisfied dependencies deps='libnvidia-ml.so.1 => not found'

SUBFAILED[dlib='/lib64/ucx/libuct_cuda_gdrcopy.so.0.0.0'] tests/containers/base_image_test.py::TestBaseImage::test_elf_files_can_link_runtime_libs[ghcr.io/red-hat-data-services/notebooks/workbench-images:cuda-jupyter-tensorflow-ubi9-python-3.12-_2c70a7ff3733116f48d692ab9dd8ac657ace03bd] - Failed: dlib='/lib64/ucx/libuct_cuda_gdrcopy.so.0.0.0' has unsatisfied dependencies deps='libnvidia-ml.so.1 => not found'