This is my workflow yml (full file):
jobs:
cuda-build:
strategy:
fail-fast: false
matrix:
cuda: [ '11.0.1', '11.5.0', '12.0.0', '12.5.0' ]
name: GPU build (CUDA ${{ matrix.cuda }})
runs-on: ubuntu-latest
steps:
- uses: jimver/cuda-toolkit@v0.2.19
id: cuda-toolkit
with:
cuda: ${{ matrix.cuda }}
method: 'network'
sub-packages: '[ "nvcc" ]'
Among 4 versions of cuda: 11.0.1, 11.5.0, 12.0.0, 12.5.0, only 12.5.0 works, the rest fail executing command /usr/bin/sudo apt-get -y install cuda-nvcc-12-0:
/usr/bin/sudo apt-get -y install cuda-nvcc-12-0
Reading package lists...
Building dependency tree...
Reading state information...
Package cuda-nvcc-12-0 is not available, but is referred to by another package.
This may mean that the package is missing, has been obsoleted, or
is only available from another source
E: Package 'cuda-nvcc-12-0' has no installation candidate
Error: Error: The process '/usr/bin/sudo' failed with exit code 100
You may check the action here.
This is my workflow yml (full file):
Among 4 versions of cuda: 11.0.1, 11.5.0, 12.0.0, 12.5.0, only 12.5.0 works, the rest fail executing command
/usr/bin/sudo apt-get -y install cuda-nvcc-12-0:You may check the action here.