Skip to content

opencl: handle OOB write in noshuffle GEMV kernels (odd ne01)#25640

Draft
wanghqc wants to merge 1 commit into
ggml-org:masterfrom
qualcomm:hq/gemv-noshuffle-odd-ne01-guards
Draft

opencl: handle OOB write in noshuffle GEMV kernels (odd ne01)#25640
wanghqc wants to merge 1 commit into
ggml-org:masterfrom
qualcomm:hq/gemv-noshuffle-odd-ne01-guards

Conversation

@wanghqc

@wanghqc wanghqc commented Jul 14, 2026

Copy link
Copy Markdown
Contributor

Overview

This PR is to fix a critical out-of-bounds write bug: in 9 of 11 noshuffle decode GEMV kernels — the kernels pad their x-grid (to 128 for the 2-row writers, to the wave size for the 1-row writers) and then vstore2/dst[gid] with no bounds check, so on an ne01 that is not a multiple of the tiling the tail work-items write past the end of dst. Some models fail to generate coherent results due to the bug.

Additional information

None

Requirements

@github-actions github-actions Bot added ggml changes relating to the ggml tensor library for machine learning OpenCL Issues specific to the OpenCL backend labels Jul 14, 2026
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 OpenCL Issues specific to the OpenCL backend

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant