Should all consumers be rebuilt? Similar to major USE_GCC upgrades.
Is MFH to 2018Q4 desirable? Less versions to support and may help getting aarch64 back again.
Details
Details
- poudriere bulk -t is green on 10.4 i386/amd64, 11.2 i386/amd64, 12.0 i386/amd64
- poudriere bulk -t is green for DEFAULT_VERSIONS+=ssl=libressl-devel
- poudriere bulk -t is green for all consumers on 11.2 amd64
Diff Detail
Diff Detail
- Repository
- rP FreeBSD ports repository
- Lint
Lint Not Applicable - Unit
Tests Not Applicable
Event Timeline
Comment Actions
Rust 1.30 has been released.
I'd appreciate if you can review before Saturday 01:00 UTC i.e., the next package build. Users would be able to grab binary package during weekend rather than wait until Tuesday if not later. In the case approval cannot be granted, please, state the rationale soon, so I can schedule landing/further work accordingly.
Comment Actions
Should all consumers be rebuilt? Similar to major USE_GCC upgrades.
I think we should. Poudriere doesn't rebuild packages when build dependencies versions change, so we might catch problems in ports only later.
Is MFH to 2018Q4 desirable? Less versions to support and may help getting aarch64 back again.
Is there any reason not to MFH it?
Comment Actions
Minor complexity:
- consumers of different versions from /head may require additional testing
- MFH script cannot deal with PORTREVISION bumps in non-yet-existing ports but pre-commit hook forbids incomplete mergeinfo
- rP481972 mismerged PORTREVISION bump in devel/rust-cbindgen