Open
Conversation
Contributor
There was a problem hiding this comment.
Pull request overview
Upgrades the pinned spack/spack-packages version to v2026.03.0 and refreshes the associated cherry-pick list to match the new baseline.
Changes:
- Bump
SPACKPACKAGES_VERSIONfromv2025.11.0tov2026.03.0 - Remove now-unnecessary
SPACKPACKAGES_CHERRYPICKSentries (and their accompanying annotations)
wdconinc
changed the title
wdconinc
force-pushed
the
spack-packages-v2026.03.0
branch
from
62d1347 to
ff2bca7
Compare
Member
|
Contributor
Author
I'll be happy if it's only the tf env that's broken. That's also broken in #195 ... |
wdconinc
force-pushed
the
spack-packages-v2026.03.0
branch
from
498a826 to
e1ac942
Compare
wdconinc
force-pushed
the
spack-packages-v2026.03.0
branch
from
e1ac942 to
a1bd79e
Compare
Contributor
Author
|
Version differences between nighlty eic-info-2025.11.0.txt and the packages introduced here eic-info-2026.03.0.txt: |
In Spack v1.1.0, package-specific `require:` entries no longer override `all: require:` — both are applied simultaneously. The global `all: require: '%gcc'` therefore conflicts with `py-tensorflow: require: '%clang'`, causing concretization to fail with "cannot satisfy a requirement for package 'py-tensorflow'". Fix by changing the `all: require: '%gcc'` entry to `any_of: ['%gcc', '%clang']`, which permits either compiler while still constraining to the two compilers available in the image. The `all: prefer: '%gcc'` entry continues to guide most packages toward gcc. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
…'%gcc' The previous fix (any_of: ['%gcc', '%clang'] in all:require) allowed py-tensorflow to concretize with %clang but caused ~24 packages to be duplicated (once compiled with %gcc, once with %clang). Root cause analysis: py-tensorflow@2.20's own compiler constraints only require gcc@12+ and clang@16+; both gcc@13.3.0 and clang@17.0.6 are valid. The '%clang' requirement in packages.yaml was a preference, not a technical necessity. In Spack v1.1+, 'all: require: %gcc' applies to ALL packages without exception and cannot be overridden by package-specific require entries (both constraints apply simultaneously). Removing the '%clang' preference from py-tensorflow:require eliminates the compiler conflict entirely, and the all:require:'%gcc' preference guides py-tensorflow to use %gcc consistently with all other packages. Also update the misleading comment that claimed 'require for all is overwritten by package-specific entries' — this was incorrect for Spack v1.1+. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
wdconinc
force-pushed
the
spack-packages-v2026.03.0
branch
from
3509b05 to
cd3d318
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Briefly, what does this PR introduce?
This PR upgrades spack-packages to v2026.03.0:
gaudito prod/spack.yaml so it is concretized together withboost; if not then a newer boost is installed that won't work with our pinnedgaudiversion,geant4 ~timemorywas removed,protobufversioning was changed.