Skip to content

sycl: unified semantics of block offset calculation #14814

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

Open
wants to merge 1 commit into
base: master
Choose a base branch
from

Conversation

Alcpz
Copy link
Collaborator

@Alcpz Alcpz commented Jul 22, 2025

The original intent of having these block structs was to avoid calculating block indexes within the mmvq kernels to get the offset of the weights and its scales.

This PR refactors the block index calculation in the reordered mmvq kernels for q4_K and q6_K to have the same behavior as q4_0.

Also, it changes traits:: internally in the struct to match the style of Q4_K and Q6_K, which was agreed to be cleaner.

@Alcpz Alcpz requested review from s-Nick and Rbiessy July 22, 2025 11:44
@github-actions github-actions bot added ggml changes relating to the ggml tensor library for machine learning SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language labels Jul 22, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
ggml changes relating to the ggml tensor library for machine learning SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant