Support building ArrayFire with the device symbols flag #2957
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Outdated suggestions cannot be applied.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.
ArrayFire was failing to build with NVCC's -G(device symbols) flag.
Description
Support compiling of ArrayFire with the -G flag in NVCC. This flag embeds device symbols in the binary
so that you could debug CUDA kernels using cuda-gdb. Compiling with device debug symbols was failing
because we were calling host functions from functions that supported both device and host code.
This PR also addresses an issue where we were using the default stream when calling CUB functions
in reduce by key.
Tests for reduce by keys don't pass in the current master with debug symbols because of the newer
version of CUB used. The older version of CUB passes the tests but it is not compatible with CUDA 11.
This is still being investigated and I have NOT included the fix as part of this PR.
Changes to Users
N/A
Checklist
[ ] Functions added to unified API[ ] Functions documented