Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Linux CUDA packages don't restore

See original GitHub issue

The Linux CUDA packages failed to restore in the DiffSharp repo build, see https://github.com/DiffSharp/DiffSharp/actions/runs/958984202

         /home/runner/.nuget/packages/libtorch-cuda-11.1-linux-x64-part2-primary/1.9.0.5/buildTransitive/netstandard2.0/libtorch-cuda-11.1-linux-x64-part2-primary.targets(204,5): error MSB4018: The "FileRestitch" task failed unexpectedly. [/home/runner/work/DiffSharp/DiffSharp/bundles/DiffSharp-cuda-linux/DiffSharp-cuda-linux.fsproj]
       /home/runner/.nuget/packages/libtorch-cuda-11.1-linux-x64-part2-primary/1.9.0.5/buildTransitive/netstandard2.0/libtorch-cuda-11.1-linux-x64-part2-primary.targets(204,5): error MSB4018: System.Exception: Error downloading and reviving packages. Reconsituted file contents have incorrect SHA [/home/runner/work/DiffSharp/DiffSharp/bundles/DiffSharp-cuda-linux/DiffSharp-cuda-linux.fsproj]
       /home/runner/.nuget/packages/libtorch-cuda-11.1-linux-x64-part2-primary/1.9.0.5/buildTransitive/netstandard2.0/libtorch-cuda-11.1-linux-x64-part2-primary.targets(204,5): error MSB4018: 	Expected SHA: $94b99584debeb3c52a5d3f173d157db81289d7741923bccbcec10d8b9ff09f0e [/home/runner/work/DiffSharp/DiffSharp/bundles/DiffSharp-cuda-linux/DiffSharp-cuda-linux.fsproj]

Issue Analytics

State:
Created 2 years ago
Comments:7 (7 by maintainers)

Top GitHub Comments

1reaction

dsymecommented, Jun 22, 2021

Should be fixed by these versions

    <LibTorchNugetVersion>1.9.0.6</LibTorchNugetVersion>
    <TorchSharpVersion>0.91.52672</TorchSharpVersion>

0reactions

dsymecommented, Jun 23, 2021

OK, cool, dotnet / MSBuild can handle > 2GB native binaries as of 6.0.100-preview.5.21302.13, I’ve tested that in the DiffSharp repository. I will adjust the README to indicate that using this toolchain is necessary if consuming libtorch via the LINUX CUDA package.

So we now have both 1.9.0 and Mac support done.