Skip to content

Conversation

@bernhardmgruber
Copy link

@bernhardmgruber bernhardmgruber commented Jan 20, 2026

  • Based on r6 from Bryce (fixes typos)
  • Added intro slide and team from last/for next "How to Build a CUDA Application in C++, From Start to Finish" course
  • Added DLI platform intro slides
  • Change namespaces from dli:: to ach::
  • Drop Final Review and Final Assignment 4 (should not be public)

The idea is to hide/remove one of the Bryce/Bernhard team intros and brev/DLI intros, depending on who teaches the lecture

I furthermore applied the following fixes discovered during the course given on 2026-01-22:

- slide 34: "Thrust execution policies are inspired them!"
- slide 51: "The median is a number",  "To compute THE median"
- slide 103: the mdspan should be reallocated to 10min
- slide 120: the Exercise Segmented Mean is fine in 6min instead of 10
- slide 128: the Exercise Use Explicit Memory Spaces is fine in 10min instead of 12
- slide 171: "work is ordered on THE GPU"
- slide 172: "on THE GPU", "THE default one"
- slide 182: thrust::copy is not done on compute_stream, as stated in the text. The text is wrong, removed it.
- slide 182: wrong variable names, lacking `_`
- slide 195: wrong variable names, lacking `_`
- slide 195: `cudaMemcpy` -> `cudaMemcpyAsync`
- slide 220: "We need the number number of thread"
- slide 240: "check if THE cell index"
- slide 256: "Atomics memory operations", remove `s`
- slide 256: `count[0]` -> `histogram[0]`
- slide 228 and others: replace `row_symmetry_check_kernel` by `symmetry_check_kernel` for consistency with other slides and exercise

@github-actions
Copy link
Contributor

👋 Thank you for your contribution!

This pull request is from a forked repository so GitHub Actions will not be able to run CI. A maintainer will review your changes shortly and manually trigger the CI.

@maintainers Please review this PR when you have a chance and follow the instructions in the CONTRIBUTING.md file to trigger the CI.

@bernhardmgruber bernhardmgruber changed the title Update CUDA C++ tutorial Update CUDA C++ tutorial to r6+ Jan 20, 2026
@bernhardmgruber
Copy link
Author

Since @brycelelbach asked me whether the files are stored in LFS, I believe so:

git lfs ls-files | select-string cuda_cpp_tutorial

8ad48309f3 * tutorials/cuda-cpp/slides/cuda_cpp_tutorial__r6.pdf
16fa224cad * tutorials/cuda-cpp/slides/cuda_cpp_tutorial__r6.pptx

* Based on r6 from Bryce (fixes typos)
* Added intro slide and team from last/for next "How to Build a CUDA Application in C++, From Start to Finish" course
* Added DLI platform intro slides
* Change namespaces from `dli::` to `ach::`
* Drop Final Review and Final Assignment 4 (should not be public)
* Include the following small fixes:
    - slide 34: "Thrust execution policies are inspired them!"
    - slide 51: "The median is a number",  "To compute THE median"
    - slide 103: the mdspan should be reallocated to 10min
    - slide 120: the Exercise Segmented Mean is fine in 6min instead of 10
    - slide 128: the Exercise Use Explicit Memory Spaces is fine in 10min instead of 12
    - slide 171: "work is ordered on THE GPU"
    - slide 172: "on THE GPU", "THE default one"
    - slide 182: thrust::copy is not done on compute_stream, as stated in the text. The text is wrong, removed it.
    - slide 182: wrong variable names, lacking `_`
    - slide 195: wrong variable names, lacking `_`
    - slide 195: `cudaMemcpy` -> `cudaMemcpyAsync`
    - slide 220: "We need the number number of thread"
    - slide 240: "check if THE cell index"
    - slide 256: "Atomics memory operations", remove `s`
    - slide 256: `count[0]` -> `histogram[0]`
    - slide 228 and others: replace `row_symmetry_check_kernel` by `symmetry_check_kernel` for consistency with other slides and exercise
@bernhardmgruber bernhardmgruber changed the title Update CUDA C++ tutorial to r6+ Update CUDA C++ tutorial to r6.1+ Jan 22, 2026
@brycelelbach
Copy link
Collaborator

Can we just call this "r6" nor "r6.1"? I don't want a super complex versioning scheme here, and extra dots in filenames throw off some OSes/tools.

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.

2 participants