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Outdated suggestions cannot be applied.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.
CUDA_ADD_LIBRARY
is deprecated after 3.10 (reference), and it limits the CUDA parallel compilation. Because many architectures are compiled serially,warp-ctc
requires almost 2 minutes to be compiled.In this PR,
add_library
+ set project language with CUDA instead ofcuda_add_library
. Then user can pass-t0
through-DCMAKE_CUDA_FLAGS
to enable CUDA parallel compilation. The compile time can be decreased from 2min to tens of second (~10x).CMAKE_CXX_STANDARD
andCMAKE_CUDA_STANDARD
to replace manually adding-std=c++11