Skip to content

fix: incorrect kernel launching parameter in parallel_for_gpu_soa - #537

Open
XDflight wants to merge 1 commit into
NVlabs:masterfrom
XDflight:master
Open

fix: incorrect kernel launching parameter in parallel_for_gpu_soa#537
XDflight wants to merge 1 commit into
NVlabs:masterfrom
XDflight:master

Conversation

@XDflight

Copy link
Copy Markdown

Looks like a typo to me. I found this issue while debugging a program that uses parallel_for_gpu_soa.

Copilot AI lite review requested due to automatic review settings August 11, 2026 08:29

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

Pull request overview

Fixes an incorrect CUDA kernel launch configuration in parallel_for_gpu_soa by using the intended 2D grid (x = n_blocks_linear(n_elements), y = n_dims) so the kernel can correctly index both elements and dimensions via blockIdx.x and blockIdx.y.

Changes:

  • Use the precomputed dim3 blocks = { n_blocks_linear(n_elements), n_dims, 1 } when launching parallel_for_soa_kernel.
  • Ensure the launch grid matches the kernel’s indexing scheme (blockIdx.y for dimension selection).

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

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