You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
-[x] (HV) Bump OSX compilers to [clang 18](https://github.com/conda-forge/conda-forge-pinning-feedstock/pull/6571)
47
+
-[x] (HV) Basic infrastructure for [Alma 9](https://github.com/conda-forge/conda-forge.github.io/issues/2342)
48
+
- Thanks to a bunch of consolidation and automation for [sysroot](https://github.com/conda-forge/linux-sysroot-feedstock/pull/79), [CDTs](https://github.com/conda-forge/cdt-builds/blob/main/cdt_slugs.yaml) & images ([pending](https://github.com/conda-forge/docker-images/pull/287)), this became very easy to add now.
49
+
- Note: CDTs do not contain distro or glibc version anymore, just `conda`. This works because constraints on `sysroot_linux-<arch>` and `__glibc` ensure separation.
- Last blocker is around usage of `yum_requirements.txt`, I did a [survey](https://github.com/conda-forge/conda-forge-pinning-feedstock/issues/6283#issuecomment-2440281086) and it looks like we can update to alma8 with minimal issues ([PR](https://github.com/conda-forge/conda-forge-pinning-feedstock/pull/6626))
52
+
- AFAICT, we should be able to get rid of `DEFAULT_LINUX_VERSION` completely
53
+
- (MRB) I think we can deprecate the `DEFAULT_LINUX_VERSION` option in smithy too if we remove it from the pinnings
54
+
- CONCLUSION: reenable it per issue here: https://github.com/conda-forge/conda-forge-pinning-feedstock/issues/6629
55
+
-[x] (HV) Lints/warnings for [`yum_requirements.txt`](https://github.com/search?q=org%3Aconda-forge+path%3Arecipe%2Fyum_requirements.txt&type=code) / CDTs
56
+
- Based on cruft found in above survey, we should IMO warn/error on incorrect entries in `yum_requirements.txt`
57
+
- May even want to go further and warn if `yum_requirements` don't show up as CDTs in meta.yaml (though this would need some metadata [fixes](https://github.com/conda-forge/cdt-builds/issues/72) in our CDTs though)
58
+
-[x] (HV) Flang (=windows fortran) [migration](https://conda-forge.org/status/migration/?name=flang19) running, propose to update:
59
+
- write a piggyback migrator (to be written) that switches away from `m2w64_*` compilers (otherwise migration PRs are no-ops) -- [example](https://github.com/conda-forge/lapack-feedstock/pull/74/commits/f472d1aed2b4fedb0a55639d4e0786592da4485e) of the kind of rewrite that would be necessary
60
+
-[exclude](https://github.com/conda-forge/conda-forge-pinning-feedstock/pull/6618) r-feedstocks, as those are consistently on [`m2w64_*` stack](https://github.com/search?type=code&q=org%3Aconda-forge+%2Fcompiler%5C%28%5B%27%22%5Dm2w64_fortran%5B%27%22%5D%5C%29%2F)
61
+
-[x] (MRB) Fixed some bugs in the rerender infra. Had to add `--force` to `git add` for the transferred patches.
62
+
-[x] (HV) OpenSSL releases reach `host` environments as soon as they are available with a `>=x.y``run_exports`.
- IF: Package the latest patch version of the last minor when a new minor comes, and then wait a bit til other distros catch up to see what happens with the fresh release.
65
+
- HV: Patch releases usually take longer than a couple weeks, unless there are serious security bugs.
0 commit comments