Re: ostree oxidation: internal components written in Rust

[Alex Kiernan <alex kiernan gmail com>]

On Wed, Dec 19, 2018 at 2:03 PM Colin Walters <walters verbum org> wrote:
> On Tue, Dec 18, 2018, at 1:53 PM, Anton Gerasimov via ostree-list wrote:
> > Hi,
> >
> >
> > > We are using ostree in embedded Linux, built with minimal dependencies (in
> > > terms of packages and cross-build tools) building it from Yocto.
> > >
> > > Our interest is in having the ostree core available in C.
> >
> > Same here. It is not impossible to build Rust code in Yocto, but it
> > increases maintenance burden and build time significantly.
>
> Hmm; is there any discussion of this elsewhere - have any handy
> links to e.g. mailing list archives?  Can you elaborate a bit
> on the maintenance overhead?  Is it around dealing with crates.io sync?
> On that topic I found
> https://lists.yoctoproject.org/pipermail/yocto/2017-March/035028.html
>
> For build time, is that building rustc for the host?  (including LLVM?) Or just general Rust
> compilation times even on target?
>
> Bigger picture...I really don't want to lose OSTree's existing user base.
> It's really great (and humbling) to see what you all are doing with it!
> That conservatism is what has held me back from pushing the Rust
> requirement more.  (Also the fact that it's been very useful for us to
> gain experience with it in the rpm-ostree context)
>
> I think last time this came up on the list we floated the idea of having a Rust requirement
> only for *build time* things, such as the static delta building code (which
> would hugely benefit from oxidation).   It's been almost 2 years since
> then, and Rust has rather notably increased in traction.
> (And Felix posted new bindings!)
>
> We could try to actually move some server-side stuff out into a separate
> repository (which...I guess really gets straight into https://github.com/alexlarsson/repo-manager )
>
> But...there's a whole lot of stuff I'd love to do in the libostree core that
> become a lot nicer with Rust.  For example I'd like to do more parallelization
> (particularly of commits and checkouts).  Another good example is
> tackling the monster that is ostree-repo-pull.c.
>
> So: we may need to try to negotiate some sort of timeframe here.

For us we're actively looking at Rust as a language for use on an
embedded Arm device (again with Yocto as the build environment) even
though we have some concerns about the support (both in terms of
what's tested and support of meta-rust).

A lot of this seems to me to be a problem of critical mass... Rust
looks like an extremely attractive language in the embedded space but
right now it looks like a lot of work for little return, but until we
have more things we want to use that are in Rust, that problem will
persist.

-- 
Alex Kiernan