Skip to content

Conversation

madsbk
Copy link
Member

@madsbk madsbk commented Sep 29, 2025

I’m not sure this is something we need to worry about in cudf, but it came up while I was debugging with compute-sanitizer --tool initcheck.

In my opinion, it’s worthwhile to keep cudf free of valgrind and compute-sanitizer issues, when the performance overhead of doing so is very small.

@madsbk madsbk requested a review from a team as a code owner September 29, 2025 06:41
@github-actions github-actions bot added the libcudf Affects libcudf (C++/CUDA) code. label Sep 29, 2025
@madsbk madsbk added improvement Improvement / enhancement to an existing function non-breaking Non-breaking change labels Sep 29, 2025
@davidwendt
Copy link
Contributor

How difficult would it be to avoid the copy instead of adding this?
Also, it may be more efficient to use thrust::fill_uninitialized instead of cudaMemset
Could you show benchmarks for this change?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
improvement Improvement / enhancement to an existing function libcudf Affects libcudf (C++/CUDA) code. non-breaking Non-breaking change
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants