llama.cpp
ci: ROCm gfx1151 fixes (draft, stacked on #26461)
#26544

Merged

ci: ROCm gfx1151 fixes (draft, stacked on #26461) #26544

CISC merged 12 commits into ggml-org:master from AMD-Ecosystem:jimwu.rocm-ci-unified-memory

github-actions added testing

github-actions added devops

github-actions added ggml

github-actions added CUDA

ci: prepare for amd rocm ci

2482364b

ci: fix editorconfig-checker

0e1ab8a2

ci: fix device not recognised

811a9a5b

ci: rename gpu-amd to gpu-hip

9de5339d

ci: gpu-hip to gpu-rocm

cd4b985b

CUDA: allow integrated-GPU host output buffer in debug assert

73b81777

ci: enable unified memory for ROCm gfx1151 job

7cd0eed2

test-llama-archs: skip jamba on HIP backend

68975900

ci: use HIP_LAUNCH_BLOCKING for ROCm gfx1151 job

8b9d95c0

test-backend-sampler: skip top-k subtests on HIP backend

59dd1ced

jimw567 force pushed to 59dd1ced 101 days ago

jimw567 marked this pull request as ready for review 100 days ago

jimw567 requested a review from

JohannesGaessler 100 days ago

jimw567 requested a review from

ggerganov 100 days ago

jimw567 requested a review 100 days ago

taronaeo approved these changes on 2026-08-06

Update tests/test-backend-sampler.cpp

05659ea2

ggerganov commented on 2026-08-06

Update tests/test-backend-sampler.cpp

b7d5494a

CISC approved these changes on 2026-08-06

CISC merged a1f96d4f into master 100 days ago

Reviewers

CISC

taronaeo

ggerganov

JohannesGaessler

Assignees

No one assigned

Labels

testing devops ggml CUDA

Milestone

No milestone

llama.cpp ci: ROCm gfx1151 fixes (draft, stacked on #26461) #26544 Merged

ci: ROCm gfx1151 fixes (draft, stacked on #26461) #26544

llama.cpp
ci: ROCm gfx1151 fixes (draft, stacked on #26461)
#26544

Merged