pytorch
M1 workers are available however not getting assigned to workflow
When trying to use macos m1 workers, I see that they are not picking up:
Following Runners are available but idling:
i-0c79eb93109020f30 Organization self-hosted macOS ARM64 macos-m1-11 Runner group: Default Idle i-094da233f1820a8db Organization self-hosted macOS ARM64 macos-m1-11 Runner group: Default Idle i-0cfedc4a23459f38f Organization self-hosted macOS ARM64 macos-m1-11 Runner group: Default Idle i-0f7db7b1f2e1563d2 Organization self-hosted macOS ARM64 macos-m1-11 Runner group: Default Idle i-037f422dc3efbbfb5 Organization self-hosted macOS ARM64 macos-m1-11 Runner group: Default Idle
Trying to run the job (also used macos-m1-11 as label and same result):
Started 19h 4m 27s ago Requested labels: self-hosted, macOS, ARM64 Job defined at: pytorch/builder/.github/workflows/build-macos-arm64-domains-wheels.yml@refs/pull/1001/merge Waiting for a runner to pick up this job...
Please investigate and fix.
cc @seemethere @malfet
continue investigating this issue I see that daemon is running on the host and it does communicate with github. Log attached: Runner_20220425-222940-utc.log
I were able to compile and install simple workflow using my personal M1, see https://github.com/malfet/deleteme/runs/6295939723?check_suite_focus=true#step:3:7 Will try to replicate this behavior on our cloud instances
And I can successfully build torchvision natively on M1 (but can't upload it yet), see WIP https://github.com/pytorch/vision/pull/5948 / https://github.com/pytorch/vision/runs/6299765936?check_suite_focus=true
Would these runners be available for libtorch-macos m1 ? I have a WIP to cross-compile for M1 https://github.com/pytorch/builder/pull/1078 but it would be unnecessary if we have native M1 runners.
