Skip to content

Conversation

@albestro
Copy link
Collaborator

@albestro albestro commented Nov 6, 2025

This PR aims at updating the existing recipe for paraview 5.13 to build the new version 6.
I took the chance to update also to the new spack, and consequentially to also use the new stackinator.

Unfortunately it is not clear what are the main changes, but I'll try to point out the main things.

  • switched to stackinator recipe version 2
  • packages.yaml: introduced to add requirements (and not just use preferences)
  • dependencies:
    • be more explicit about them, so they are root and end up in the view (it should ease the work when using the uenv to do a custom build)
    • bump version
  • post-install small changes and improvements
  • env-to-package.py: mainly, updated to deal with the new spack package structure

TODO

@albestro
Copy link
Collaborator Author

albestro commented Nov 6, 2025

cscs-ci run alps;system=eiger;uarch=zen2;uenv=paraview:6.0.1

@albestro
Copy link
Collaborator Author

albestro commented Nov 6, 2025

cscs-ci run alps;system=eiger;uarch=zen2;uenv=paraview:6.0.1

forgot to add the custom packages

@albestro
Copy link
Collaborator Author

cscs-ci run alps;system=eiger;uarch=zen2;uenv=paraview:6.0.1

@albestro
Copy link
Collaborator Author

albestro commented Dec 9, 2025

cscs-ci run alps;system=eiger;uarch=zen2;uenv=paraview:6.0.1

just checking if it still works after alignment with gh200 recipe

@albestro albestro force-pushed the alby/paraview-6-zen2 branch from f2bcc6d to 8a23b84 Compare December 9, 2025 10:10
bcumming pushed a commit that referenced this pull request Dec 15, 2025
This is sister PR of #267 

Excluding trivial differences (mainly EGL+CUDA vs OSMESA and related),
these PRs share a lot, so I'll try to keep them as closest as possible.
simonpintarelli pushed a commit that referenced this pull request Dec 19, 2025
This is sister PR of #267 

Excluding trivial differences (mainly EGL+CUDA vs OSMESA and related),
these PRs share a lot, so I'll try to keep them as closest as possible.
@albestro albestro force-pushed the alby/paraview-6-zen2 branch from 3f01cfd to 73bb8ea Compare January 9, 2026 08:17
@albestro
Copy link
Collaborator Author

albestro commented Jan 9, 2026

cscs-ci run alps;system=eiger;uarch=zen2;uenv=paraview:6.0.1

get back working on this, let's see what's the status and what's missing

@albestro
Copy link
Collaborator Author

cscs-ci run alps;system=eiger;uarch=zen2;uenv=paraview:6.0.1

there might be a problem with python packages due to spack/spack#51743
let's verify that's the problem, and in case I will dig more and work on a solution

it was relying on PYTHONPATH, that after this spack commit has been removed
@albestro albestro force-pushed the alby/paraview-6-zen2 branch from d4267c2 to f21e77b Compare January 15, 2026 16:31
@albestro
Copy link
Collaborator Author

cscs-ci run alps;system=eiger;uarch=zen2;uenv=paraview:6.0.1

@albestro
Copy link
Collaborator Author

cscs-ci run alps;system=eiger;uarch=zen2;uenv=paraview:6.0.1

try again

@albestro
Copy link
Collaborator Author

I have tested manually f21e77b and at least basic functionalities seem to work correctly.

image

The main difference is llvmpipe, but in my little knowledge it should be just an improvement wrt softpipe.

(paraview)[eiger][ialberto@eiger-ln003 paraview]$ srun -n4 -N1 -pdebug pvbatch -- src/coloredSphere.py
(  11.714s) [pvbatch.0       ]         vtkSMProxy.cxx:888   WARN| vtkSMSourceProxy (0x2833560): Proxy (filters, GenerateProcessIds)  has been deprecated in ParaView 5.13 and will be removed by ParaView 6.1.
        This has been replaced by 'ProcessIds'. Please consider using that instead.

ParaView Version  (6, 0, 1)
rank= 0 / 4
Vendor:   Mesa/X.org
Version:  4.5 (Compatibility Profile) Mesa 22.3.5
Renderer: llvmpipe (LLVM 15.0.7, 256 bits)
writing  coloredSphere_v6.0.1.EGL.png

@albestro
Copy link
Collaborator Author

cscs-ci run alps;system=eiger;uarch=zen2;uenv=paraview:6.0.1

@albestro albestro force-pushed the alby/paraview-6-zen2 branch from eab2e62 to ad493c3 Compare January 20, 2026 17:22
@albestro
Copy link
Collaborator Author

cscs-ci run alps;system=eiger;uarch=zen2;uenv=paraview:6.0.1

@albestro
Copy link
Collaborator Author

cscs-ci run alps;system=daint;uarch=gh200;uenv=paraview:6.0.1

unfortunately, the upstream pacakge cannot be used yet due to
conduit detection problem by vtk
RMeli pushed a commit to RMeli/alps-uenv that referenced this pull request Jan 22, 2026
This is sister PR of eth-cscs#267 

Excluding trivial differences (mainly EGL+CUDA vs OSMESA and related),
these PRs share a lot, so I'll try to keep them as closest as possible.
@albestro
Copy link
Collaborator Author

cscs-ci run alps;system=eiger;uarch=zen2;uenv=paraview:6.0.1

@albestro
Copy link
Collaborator Author

cscs-ci run alps;system=daint;uarch=gh200;uenv=paraview:6.0.1

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.

1 participant