llama.cpp
cuda: align snake fusion matcher with the other backends
#25460

Merged

cuda: align snake fusion matcher with the other backends #25460

ggerganov merged 2 commits into ggml-org:master from ServeurpersoCom:ggml/cuda-snake-fusion-parity

cuda: fix snake fusion type predicate, a and inv_b are F32

0372ab74

cuda: reject snake fusion on non-contiguous operands

70ff329e

ServeurpersoCom requested a review 75 days ago

github-actions added ggml

github-actions added CUDA

am17an approved these changes on 2026-07-09

am17an added merge ready

ggerganov merged 2021515a into master 75 days ago

Reviewers

am17an

Assignees

No one assigned

Labels

ggml merge ready CUDA

Milestone

No milestone