3 ms·

What I think is somewhat interesting in this case (if I've read the post correctly) half the changes were for newer versions of dependencies. If that's the case

What I think is somewhat interesting in this case (if I've read the post correctly) half the changes were for newer versions of dependencies. If that's the case, then I'm inclined to start wondering what's going on with bcachefs upstream (and the whole dependency tree), if they're not keeping up with versions.

arzig 2y ago
My read is that whatever subset of the ecosystem bcachefs-tools just hasn’t stabilized. Presumably it will, but once everything is functionally at 1.x even if they don’t reach it officially this will be become a non-issue. That said, having dabled in rust there’s a temptation to name breaking changes to improve ergonomics because the type system is so expressive. I almost wonder if this is an argument for simpler type systems to reduce churn.
SkiFire13 2y ago
Imagine that an update of bcache-fs is released and 3 dependencies (could even be transitive) were updated and the newer versions were not packaged in Debian. The maintainer now has one package to update but another 3 to package for the first time. That's additional work. AFAIK Fedora has a tool to automatize this work, but maybe Debian has additional requirements that make this harder.
- aragilar 2y ago
  Debian (and derivs) have similar tooling (and the vast majority of rust packages are tracked in a monorepo https://salsa.debian.org/rust-team/debcargo-conf https://salsa.debian.org/rust-team/debcargo-conf). My point was "some of the upstream versions were older than Debian's" which means the "Debian is behind, Debian is slow" is not true in the case. That usually only happens where an upstream is either very conservative with bumping versions, or where upstream is under-maintained, neither which at first glance seem to be true of becache-fs.
rcxdude 2y ago
It looks like the two packages that were bumped forward are ones where the changes were very minimal (it's not obvious why errno even bumped their 'major' version, and udev seem to do it on every release even when backwards compatible). Also errno 0.4 doesn't seem to exist, so I think that's a typo. (There is a somewhat frustrating trend in rust packages for them to sit on 0.x versions for a very long time, even if they're fairly heavily used in the ecosystem and pretty stable in practice) From what the author of bcachefs is saying, the forward moves were not a problem, it was moving bindgen backwards which broke the build, meaning the packaged version of bcachefs-tools stayed very out of date, causing him to get bug reports for known and fixed fairly serious bugs (ones which rendered machines unbootable).

arzig 2y ago

My read is that whatever subset of the ecosystem bcachefs-tools just hasn’t stabilized. Presumably it will, but once everything is functionally at 1.x even if they don’t reach it officially this will be become a non-issue. That said, having dabled in rust there’s a temptation to name breaking changes to improve ergonomics because the type system is so expressive. I almost wonder if this is an argument for simpler type systems to reduce churn.

SkiFire13 2y ago

Imagine that an update of bcache-fs is released and 3 dependencies (could even be transitive) were updated and the newer versions were not packaged in Debian. The maintainer now has one package to update but another 3 to package for the first time. That's additional work. AFAIK Fedora has a tool to automatize this work, but maybe Debian has additional requirements that make this harder.

aragilar 2y ago
Debian (and derivs) have similar tooling (and the vast majority of rust packages are tracked in a monorepo https://salsa.debian.org/rust-team/debcargo-conf https://salsa.debian.org/rust-team/debcargo-conf). My point was "some of the upstream versions were older than Debian's" which means the "Debian is behind, Debian is slow" is not true in the case. That usually only happens where an upstream is either very conservative with bumping versions, or where upstream is under-maintained, neither which at first glance seem to be true of becache-fs.

rcxdude 2y ago

It looks like the two packages that were bumped forward are ones where the changes were very minimal (it's not obvious why errno even bumped their 'major' version, and udev seem to do it on every release even when backwards compatible). Also errno 0.4 doesn't seem to exist, so I think that's a typo. (There is a somewhat frustrating trend in rust packages for them to sit on 0.x versions for a very long time, even if they're fairly heavily used in the ecosystem and pretty stable in practice) From what the author of bcachefs is saying, the forward moves were not a problem, it was moving bindgen backwards which broke the build, meaning the packaged version of bcachefs-tools stayed very out of date, causing him to get bug reports for known and fixed fairly serious bugs (ones which rendered machines unbootable).