Statically link the CUDA runtime into a single shared library #164
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Jason Lowe <[email protected]>
|
build |
1 similar comment
|
build |
|
build |
pxLi
approved these changes
Apr 11, 2022
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Closes #146.
This is a major update to the way this project builds and is designed to build all of the libcudf, libcudfjni, and libspark_rapids_jni code into a single shared library that is statically linked with the CUDA runtime. Because we are no longer leveraging the contents of a cudf-like jar, the project has been collapsed into a single top-level Maven module that pulls in the cudf Java sources.
During the build, CMake is invoked to build libcudf and libcudfjni as static libraries. Those static libraries are then whole-archive linked into a libcudf.so library which is statically linked with the CUDA runtime. In addition a stub libcudfjni.so library is created as well that depends on libcudf.so. This somewhat contorted build is done for the following reasons:
This depends on rapidsai/cudf#10619.