Pat Pannuto

> There might be pushback against additional files in the repository root, but I don't see a way around this -- a many tools require that. Yeah, it's already frustrating...

> I need an ID that outlives a process, and is allocated even when the process is not. Okay, I understand the need there. I think my new question is...

If I understand the implications of #4370 and the subsequent CHERI work—which I have not yet had a chance to read in depth—I think the concern is that there will...

@bradjc @lschuermann It looks like we can't actually move away from building qemu for a little while yet — upstream ubuntu packages are too old :(

Oh dear, the flux specs seem to break `rustfmt`, which throws internal errors. I'd really hoped for this to be a clean PR, but I can't fix this right this...

> How do I know it did anything? ` Finished `flux` profile [optimized + debuginfo] target(s) in 0.22s` is the relevant line that indicates that flux succeeded in validating all...

> > How do I know it did anything? > > ` Finished`flux` profile [optimized + debuginfo] target(s) in 0.22s` is the relevant line that indicates that flux succeeded in...

> Who is responsible for fixing any issues that come up? Like anything, mostly the code author, but I think it is safe to rely on @enjhnsn2 [as a Tock...

Going to have to rethink this: It looks like we can't actually exclude this from `Cargo.lock` (it just gets added back whenever you do any kind of build). From what...

I think I've addressed all the concerns here. "`cargo test`-like" output is [on the flux team's roadmap](https://github.com/flux-rs/flux/issues/1321), and we can update `flux` when that becomes available.