On Wed, Jan 13, 2021 at 05:30:53PM -0500, Leo Famulari wrote: > Using `guix weather`, we can check substitute availability for the > staging branch: > > ------ > master branch > aarch64: 66% > x86_64: 93% > i686: 85% > armhf: 51% > > staging branch > aarch64: 44% > x86_64: 80% > i686: 60% > armhf: 30% > ------ Updates since 9 days ago: ------ master branch aarch64: 72% x86_64: 94% i686: 86% armhf: 46% staging branch aarch64: 48% x86_64: 79% i686: 70% armhf: 30% ------ Some improvement can be seen. There are still a huge number of failures on CI that can't be reproduced locally, especially for aarch64. Since working around the failure of mesa on i686, that platform shows major improvement. The big failures on x86_64 are all rather niche — Java and OCaml mostly. armhf continues to decline because we are no longer building substitutes for it. There are some failures I think are important, though. For example, the vtk package is broken due to incompatibility with new Freetype, which breaks FreeCAD. On #guix, Marius said "I looked into VTK before the holidays; the Freetype issue is fixed in version 9, but that has other problems, such as making it impossible to unbundle the dozens of libraries that we are currently unbundling [...] it is possible to backport the VTK commits that fix Freetype compatibility, but it will be a lot of work and a huge patch (it was a major cleanup IIRC)." I'm CC-ing Ekaitz Zarraga, who has been working on FreeCAD. I'm not sure what we can do about this problem in the short term. Marius, can you give more info about the bundling problem? Full weather report attached, including failures of packages with at least 10 dependents. Can you identify any outstanding issues? Or are we ready to merge the branch into master? Although the numbers are lower for staging, the high rate of spurious failures and the successful system reconfigurations reported by Guix users makes me think that the situation is fine.