llama.cpp
cuda : extend GGML_OP_PAD to work with non-cont src0
#19429

Merged

cuda : extend GGML_OP_PAD to work with non-cont src0 #19429

ggerganov merged 2 commits into master from gg/cuda-pad-non-cont

ggerganov requested a review from

JohannesGaessler 3 days ago

github-actions added Nvidia GPU

github-actions added ggml

ggerganov force pushed from 23aa6675 to b3507fa6 3 days ago

cuda : extend GGML_OP_PAD to work with non-cont src0

b7c7c950

tests : add permuted pad

084b3d8d

ggerganov force pushed from b3507fa6 to 084b3d8d 2 days ago

JohannesGaessler approved these changes on 2026-02-09

github-actions added testing

ggerganov merged a0d58553 into master 2 days ago

ggerganov deleted the gg/cuda-pad-non-cont branch 2 days ago

Reviewers

JohannesGaessler

Assignees

No one assigned

Labels

testing Nvidia GPU ggml

Milestone

No milestone