]> git.djapps.eu Git - pkg/ggml/sources/ggml/commit
vulkan: fix diag_mask_inf (llama/11323)
authorJeff Bolz <redacted>
Thu, 23 Jan 2025 07:01:17 +0000 (01:01 -0600)
committerGeorgi Gerganov <redacted>
Wed, 29 Jan 2025 10:57:00 +0000 (12:57 +0200)
commitc3debeca1586428ac2f4428094027fc5a8ed08e4
tree4bcd99394e88c684f50386372e684c516cab981e
parent417f0100f957da05b0eb78a13e275a78a319744c
vulkan: fix diag_mask_inf (llama/11323)

With robustbufferaccess disabled, this shader was showing OOB stores. There
is a bounds check in the code, but the workgrouop dimensions were reversed vs
CUDA and it was running the wrong number of threads. So fix the workgroup
dimensions and disable robustness for this pipeline.
src/ggml-vulkan/ggml-vulkan.cpp
src/ggml-vulkan/vulkan-shaders/diag_mask_inf.comp