Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add foundational CUDA support and basic kernel #459

Draft
wants to merge 23 commits into
base: main
Choose a base branch
from
Draft

Conversation

corybarr
Copy link
Contributor

@corybarr corybarr commented Aug 28, 2023

This PR add the foundational classes for adding CUDA kernels to tilesets. In this initial implementation, a selectable kernel is created per tile. Notes:

  • If you want to see the placeholder kernel's output to verify it's working, uncomment the print statement in CudaKernels.h::createVoxelsKernel
  • This PR runs only a placeholder kernel. The next PR will finalize handling the arguments non-trivial kernels need per type.
  • There is a test USDA file that will start to load the Melbourne point-cloud dataset
  • Note that the base branch is add-cuda-cmake, not main
  • This PR also contains code to pass the geometric error per tile down to a tile for later use in kernels

melbournePointCloudTileset.zip

@corybarr corybarr changed the base branch from main to add-cuda-cmake August 28, 2023 19:14
CudaKernelArgs args,
int elementCount)
: kernelType(cudaKernelType)
, kernelArgs(std::move(args))
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This clang convention of starting with a comma seems odd. Is this correct?

int elementCount;

private:
// omni::fabric::PrimBucketList _bucketList;
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Placeholder for incorporating the kernel args in the next PR.

void runAllRunners();
void initialize();
void runRunner(CudaRunner& runner);
void** packArgs(CudaKernelArgs cudaKernelArgs, CudaKernelType cudaKernelType);
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Placeholder for next PR. Signature might change.

@corybarr corybarr marked this pull request as ready for review August 28, 2023 19:19
@weegeekps
Copy link
Contributor

@corybarr @lilleyse Does this need to be merged prior to the release on Friday?

@corybarr corybarr marked this pull request as draft August 29, 2023 19:20
Base automatically changed from add-cuda-cmake to main September 5, 2023 19:14
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants