Skip to content

Conversation

@benfred
Copy link
Member

@benfred benfred commented Oct 24, 2025

When converting from a DLManagedTensor to a mdspan in our c-api, we weren't checking the stride information on the dlmanaged tensor is the c-api. This caused invalid results when passing a strided matrix to functions like cuvsCagraBuild. Fix and add a unittest.

When converting from a DLManagedTensor to a mdspan in our c-api,
we weren't checking the stride information on the dlmanaged tensor
is the c-api. This caused invalid results when passing a strided
matrix to functions like cuvsCagraBuild. Fix and add a unittest.
@benfred
Copy link
Member Author

benfred commented Nov 13, 2025

/merge

@rapids-bot rapids-bot bot merged commit d3ae8fd into rapidsai:main Nov 13, 2025
240 of 244 checks passed
enp1s0 pushed a commit to enp1s0/cuvs that referenced this pull request Nov 16, 2025
When converting from a DLManagedTensor to a mdspan in our c-api, we weren't checking the stride information on the dlmanaged tensor is the c-api. This caused invalid results when passing a strided matrix to functions like cuvsCagraBuild. Fix and add a unittest.

Authors:
  - Ben Frederickson (https://github.com/benfred)
  - Corey J. Nolet (https://github.com/cjnolet)

Approvers:
  - Corey J. Nolet (https://github.com/cjnolet)

URL: rapidsai#1458
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

bug Something isn't working non-breaking Introduces a non-breaking change

Development

Successfully merging this pull request may close these issues.

[BUG] cuvsCagraBuild accepts stridden matrices as input, but the computed graph is wrong

2 participants