NVRTC Batch 5 #1180
Merged
NVRTC Batch 5 #1180
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
|
This is really, really great! Thanks!! |
|
Same here, on behalf of the Python community, thank you very much, this whole undertaking will be an important step for reusable special functions throughout the ecosystem! |
|
Only CI failure is a bad connect downloading packages. Should be safe to merge. |
|
Excellent! I'll start looking into getting this working in CuPy. |
Let us know how it goes and feel free to ping us with issues you find. |
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Outdated suggestions cannot be applied.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.
Finally, we have enough foundation on NVRTC that we can now add distributions.
Adds NVRTC support to: arcsine dist, bernoulli dist, cauchy dist, exponential dist, extreme value dist, holtsmark dist, landau dist, laplace dist, logistic dist, mapairy dist, and saspoint5 dist. This is the same support that currently exists for NVCC
Replaces the use of
thurst::pairwithcuda::std::pairso that it works with NVRTC. Implement our own versions ofcuda::std::forwardandcuda::std::make_tuplesince they are missing from libcu++.NVRTC support is almost in line with what is supported in NVCC and SYCL minus a few outstanding special functions that are not in the libcu++ library. Those will be in the next batch so that development moving forward will apply to all platforms at once rather than piecemeal.
Completed CI runs on device can be found at: cppalliance/cuda-math#13
CC: @steppi, @izaid, @dschmitz89