Skip to content

CUDA: Pass more link libraries to device linking

Previously we dropped non-target items from the device link line because nvcc rejects paths to shared library files, and only with target items do we know the kind of library. However, this also prevents projects from linking to system-provided libraries like cublas_device that contain device code.

Fix this by passing more link items to device linking. Items that are not file paths, such as -lfoo, can simply be passed unconditionally. Items that are targets known to be shared libraries can still be skipped. Items that are paths to library files can be passed directly if they end in .a. Otherwise, pass them using -Xnvlink to bypass nvcc's front-end. The nvlink tool knows to ignore shared library files.

Also pass host linker search directories to the device linker because some environments put libraries containing device code in them.

Issue: #16317 (closed)

Edited by Brad King

Merge request reports