Andrey Talman

HI @sberryman Thank you I could reproduce this with our 1.11 docker image . This issue can be mitigated by following these steps: https://developer.nvidia.com/blog/updating-the-cuda-linux-gpg-repository-key I am looking forward on fixing...

This is our build file for magma in windows: ```https://github.com/pytorch/builder/blob/main/windows/internal/build_magma.bat``` We also have this runbook on building magma for windows: https://github.com/pytorch/builder/blob/main/CUDA_UPGRADE_GUIDE.MD#5-modify-code-to-install-the-new-cuda-for-windows-and-update-magma-for-windows Here is a log of recent magma_cuda build: https://github.com/pytorch/builder/runs/7220211738?check_suite_focus=true

@IvanYashchuk after this is merged we will have to trigger magma rebuild on linux and windows

> Rebuild for Linux will be triggered automatically after the merge, right? https://github.com/pytorch/builder/blob/main/.github/workflows/build-magma-linux.yml Does not look like it will, could you please add ``` magma/package_files/* ``` To path: in both...

@pytorchmergebot rebase

We should proceed with the following schedule on dropping support for 10.2 : * Release 1.12 - announce * Release 1.13 - actual drop

One solution of dropping CUDA 10.2 that was discussed with @ptrblck is to link with CUDA dynamically rather then statically as done now. Then we download CUDA runtime on the...

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](https://github.com/pytorch/builder/files/8558618/Runner_20220425-222940-utc.log)

@pytorchmergebot -f "Already run through all tests"

@pytorchbot --help