Centralize cuda_arch capability definition

[Copilot]

Wait for:

• feat: build arrow with cuda for eic_cuda #199

cuda_arch=75 was duplicated across 5 package specs in cuda/spack.yaml and tf/spack.yaml, making it hard to retarget GPU architecture.

Changes

• spack-environment/packages.yaml: Added cuda_arch=75 as a conditional requirement in packages: all: require with when: '+cuda', so it applies to all CUDA-enabled packages without affecting CPU-only packages.
• cuda/spack.yaml and tf/spack.yaml: Drop inline cuda_arch=75 from acts, arrow, celeritas, py-torch, and py-tensorflow specs.

# spack-environment/packages.yaml
packages:
  all:
    require:
    - when: '+cuda'
      any_of: [cuda_arch=75, '@:']

To target a different GPU architecture (currently Compute Capability 7.5 / Turing: T4, RTX 2xxx, Quadro RTX), change cuda_arch=75 in this single location.

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.

---

📍 Connect Copilot coding agent with Jira, Azure Boards or Linear to delegate work to Copilot in one click without leaving your project management tool.

[wdconinc]

Looks sensible. Maybe there's a risk this will enable cuda_arch and therefore +cuda for packages where we currently don't explicitly enable +cuda, but would that be a bad thing?

[Copilot]

Pull request overview

This PR centralizes the CUDA compute capability setting for the Spack CUDA-related environments by introducing a shared cuda_arch.yaml include and removing per-spec cuda_arch=75 pins.

Changes:

• Add a new spack-environment/cuda_arch.yaml and include it from CUDA/TensorFlow environments.
• Remove explicit cuda_arch=75 from CUDA-enabled specs in spack-environment/cuda/spack.yaml.
• Remove explicit cuda_arch=75 from the TensorFlow CUDA spec in spack-environment/tf/spack.yaml.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.

File	Description
spack-environment/tf/spack.yaml	Includes centralized CUDA arch config; drops per-spec cuda_arch=75 on TensorFlow.
spack-environment/cuda/spack.yaml	Includes centralized CUDA arch config; drops per-spec cuda_arch=75 on multiple CUDA specs.
spack-environment/cuda_arch.yaml	New shared config intended to define CUDA architecture in one place.

---

You can also share your feedback on Copilot code review. Take the survey.

[wdconinc]

Appears to propagate the cuda_arch correctly now, e.g. https://github.com/eic/containers/actions/runs/23980557567/job/69944240377?pr=202#step:15:581.

[Copilot]

Pull request overview

Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.

[Copilot]

The PR description says cuda_arch=75 is added as a conditional requirement (and the file header notes the preferred require: any_of: [...] pattern for optional variants), but the change adds a conditional spec: cuda_arch=75 entry under packages:all:prefer and does not mirror it in packages:all:require. This may not have the intended effect and diverges from the established pattern used for other global constraints (e.g., build_system / build_type use require: any_of: [...]). Consider moving/duplicating this into require using the existing any_of: [cuda_arch=75, '@:'] style so it applies consistently while remaining a no-op for packages that don't support cuda_arch.

[Copilot]

This environment still hard-codes cuda_arch=75 for g4adept later in the same specs: list (g4adept +cuda cuda_arch=75). If the goal is to centralize the CUDA architecture, consider removing the remaining inline cuda_arch from g4adept as well (or documenting why it needs to stay explicitly set).