llama.cpp
fix: CUDA device PCI bus ID de-dupe OOMing (ignoring other 3 gpus entirely)
#22533

Merged

fix: CUDA device PCI bus ID de-dupe OOMing (ignoring other 3 gpus entirely) #22533

JohannesGaessler merged 2 commits into ggml-org:master from lucyknada:patch-2

fix: CUDA device PCI bus ID detection for multi-GPU de-dupe

e7e466df

lucyknada requested a review 107 days ago

github-actions added Nvidia GPU

github-actions added ggml

gaugarg-nv commented on 2026-04-30

HIP, MUSA macros

422f9ea6

JohannesGaessler requested a review from

IMbackK 105 days ago

gaugarg-nv approved these changes on 2026-05-02

JohannesGaessler approved these changes on 2026-05-02

JohannesGaessler merged d05fe1d7 into master 104 days ago

Reviewers

JohannesGaessler

gaugarg-nv

IMbackK

Assignees

No one assigned

Labels

Nvidia GPU ggml

Milestone

No milestone