GCC 13 & LLVM 17

[h-vetinari]

We've had GCC 13 and Clang 17 live for a while, and Clang 18 was released in March, and GCC 14 is imminent. Should be good enough to start discussing the next update cycle.

To-Do list based on docs (was conda-forge/conda-forge.github.io#1950)

• Update all respective feedstocks (listing only those for activation; they wouldn't pass if the implementation wasn't complete)
  • GCC 13
    • GCC 14.1, 13.3 & 12.4 ctng-compilers-feedstock#133
    • GCC 14.1, 13.3, 12.4 ctng-compiler-activation-feedstock#121
    • gfortran 14.2, 13.3, 12.4 gfortran_impl_osx-64-feedstock#82
    • GCC 13.2; bump min_runtime_version ctng-compiler-activation-feedstock#97
  • Clang 17
    • LLVM 17.0.6 clang-compiler-activation-feedstock#113 (osx)
    • Add clang{,xx} compilers ctng-compiler-activation-feedstock#102 (linux; not part of global pinning)
    • clang 17 + msvc 14.38 clang-win-activation-feedstock#32 (win; no part of the global pinning)
• Tested that compilers work
  • WIP: Test new compilers scipy-feedstock#250 (includes fortran)
• Checked nvcc-bounds (see here & here)
  • Cannot update because we're already at the limit for CUDA 11.2 & 11.8 (GCC 11) resp. CUDA 12.0 (GCC 12)
• Checked release notes for ABI-relevant changes
  • GCC release notes & porting guide
    • Nothing too scary looking. Some transitive header deps in libstdcxx were removed (solved by regular IWYU).
  • Clang potentially breaking changes
    • Nothing relevant
  • Clang ABI changes
    • Minor changes (IMO) of the kind we had ignored in 16, 15 & 14

Previously: #4991 #4215 #3731 #3290 #3261 #2802 #2010

[conda-forge-webservices]

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe) and found it was in an excellent condition.

[isuruf]

See conda-forge/ctng-compiler-activation-feedstock#96

[hmaarrfk]

A breaking change seems to have been introduced in 13.2 ....
conda-forge/ctng-compilers-feedstock#113

[hmaarrfk]

shall we try?

[h-vetinari]

I'm fine with it obviously, but I'm not in a rush on this one. Wanted to bring it up in a core call perhaps, but those have had more high-urgency topics recently.

[h-vetinari]

Note: I'm mainly waiting for the release of GCC 14.1 (which is usually followed by a point release for the previous major version from all the bug fixes accumulated during release preparation); once that is in place I think we can run the migration.

[h-vetinari]

@conda-forge/core, I'd like to follow through on this compiler update. This has been a long time coming and was also discussed in the last core call.

The one issue is that we currently don't have gfortran 13.3 on OSX (see conda-forge/gfortran_impl_osx-64-feedstock#82 which has been stuck for 2 months), but we do have 13.2 (confirmed working e.g. in conda-forge/scipy-feedstock#250), so that should be good enough IMO to move forward.

[h-vetinari]

In case this got lost over the weekend, PTAL @conda-forge/core. If there are no further comments, I'm planning to merge in 24h.