Skip to content

Conversation

@havogt
Copy link
Contributor

@havogt havogt commented Jan 22, 2026

  • Disables USE_MPI because of incompatibilities with container libraries and the libfabric injection.
  • Removes CSCS_CUDA_MPS (i.e. disable) since it doesn't seem to work on shared nodes

@edopao
Copy link
Contributor

edopao commented Feb 4, 2026

It may be that this PR is no longer needed. The AMD tests are passing on #2471.

@havogt
Copy link
Contributor Author

havogt commented Feb 5, 2026

strange... I think it still make sense to upgrade to 24.04

BASE_IMAGE: jfrog.svc.cscs.ch/dockerhub/rocm/dev-ubuntu-${UBUNTU_VERSION}:${ROCM_VERSION}-complete
EXTRA_UV_SYNC_ARGS: "--extra rocm6_0"
EXTRA_UV_ENV_VARS: "CUPY_INSTALL_USE_HIP=1 HCC_AMDGPU_TARGET=gfx942 ROCM_HOME=/opt/rocm"
UBUNTU_VERSION: '22.04'
Copy link
Contributor Author

@havogt havogt Feb 5, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

uses the default

@havogt havogt requested a review from egparedes February 5, 2026 14:00
@havogt havogt changed the title ci: Update Ubuntu version from 22.04 to 24.04 for AMD CI ci: Update Ubuntu version to 24.04 for beverin, disable MPS on santis Feb 11, 2026
@havogt havogt requested a review from edopao February 11, 2026 07:43
Copy link
Contributor

@edopao edopao left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The CSCS CI passes, we do not need to wait for the GitHub CI.

@havogt havogt merged commit 2c8fe63 into main Feb 11, 2026
31 checks passed
@havogt havogt deleted the havogt-patch-3 branch February 11, 2026 08:21
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants