Unverified Commit d903abfc authored by Yih-Dar's avatar Yih-Dar Committed by GitHub
Browse files

Fix AMD CI not showing GPU (#27555)



fix
Co-authored-by: default avatarydshieh <ydshieh@users.noreply.github.com>
parent fe3ce061
......@@ -38,14 +38,16 @@ jobs:
runs-on: [self-hosted, docker-gpu, amd-gpu, '${{ matrix.machine_type }}', '${{ inputs.gpu_flavor }}']
container:
image: huggingface/transformers-pytorch-amd-gpu-push-ci # <--- We test only for PyTorch for now
options: --device /dev/kfd --device /dev/dri --env HIP_VISIBLE_DEVICES --env ROCR_VISIBLE_DEVICES --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
options: --device /dev/kfd --device /dev/dri --env ROCR_VISIBLE_DEVICES --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
steps:
- name: ROCM-SMI
run: |
rocm-smi
- name: ROCM-INFO
run: |
rocminfo | grep "Agent" -A 14
- name: Show HIP environment
- name: Show ROCR environment
run: |
echo "HIP: $HIP_VISIBLE_DEVICES"
echo "ROCR: $ROCR_VISIBLE_DEVICES"
setup_gpu:
......@@ -57,7 +59,7 @@ jobs:
runs-on: [self-hosted, docker-gpu, amd-gpu, '${{ matrix.machine_type }}', '${{ inputs.gpu_flavor }}']
container:
image: huggingface/transformers-pytorch-amd-gpu-push-ci # <--- We test only for PyTorch for now
options: --device /dev/kfd --device /dev/dri --env HIP_VISIBLE_DEVICES --env ROCR_VISIBLE_DEVICES --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
options: --device /dev/kfd --device /dev/dri --env ROCR_VISIBLE_DEVICES --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
outputs:
matrix: ${{ steps.set-matrix.outputs.matrix }}
test_map: ${{ steps.set-matrix.outputs.test_map }}
......@@ -155,7 +157,7 @@ jobs:
runs-on: [self-hosted, docker-gpu, amd-gpu, '${{ matrix.machine_type }}', '${{ inputs.gpu_flavor }}']
container:
image: huggingface/transformers-pytorch-amd-gpu-push-ci # <--- We test only for PyTorch for now
options: --device /dev/kfd --device /dev/dri --env HIP_VISIBLE_DEVICES --env ROCR_VISIBLE_DEVICES --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
options: --device /dev/kfd --device /dev/dri --env ROCR_VISIBLE_DEVICES --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
steps:
# Necessary to get the correct branch name and commit SHA for `workflow_run` event
# We also take into account the `push` event (we might want to test some changes in a branch)
......@@ -206,11 +208,13 @@ jobs:
echo "matrix_folders=$matrix_folders" >> $GITHUB_ENV
- name: ROCM-SMI
run: |
rocm-smi
- name: ROCM-INFO
run: |
rocminfo | grep "Agent" -A 14
- name: Show HIP environment
- name: Show ROCR environment
run: |
echo "HIP: $HIP_VISIBLE_DEVICES"
echo "ROCR: $ROCR_VISIBLE_DEVICES"
- name: Environment
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment