Hacker News Reader

Comment by Aurornis | Hacker News Reader

> Rust is evolving far too fast to be used in code which needs to run for years to decades down the line.

Code doesn’t stop running on existing hardware when the language changes in a future compiler. You can still use the same old toolchain.

I’ve done a lot of embedded development in a past life. Keeping old tool chains around for each old platform was standard.

I would much rather go through the easy process of switching to an older Rust tool chain to build something than all of the games we played to keep entire VMs archived with a snapshot of a vendor tool chain that worked to build something.

View on HN

I remember a coworker having to fight with an old platform's build not working because our user/group IDs were bigger than 2^16. I can't remember which utility was causing the problem, I'd have to guess tar. This is when we learned to play the archive a VM game.

I can't imagine theres much overlap between "we will need to update this firmware for the next decade." and "Let's bet the farm on the documentation being perfect, and all the downloads still available."

I know a defence company that has a bunch of vaxes stored in low oxygen environments because they legally have to be able to provide software updates to firmware they’ve written for the next 20 or so years and it was written on a vax.

They had some great stories trying to get something or other running again where they had to fly one of the original designers over to hand solder a board back into action.

How we do that today is a bit of an interesting problem I don’t think they’ve convincingly solved; basically maintaining nightly builds forever — a couple 1U’s of kubernetes in deep storage ain’t gonna do it, you’re not gonna be able to solder a xeon back to life..

I know I’d rather be trying get a load of c99 rebuilt for some mips or other after 20 years that some random version of rust.

> I know a defence company that has a bunch of vaxes stored in low oxygen environments because they legally have to be able to provide software updates to firmware they’ve written for the next 20 or so years and it was written on a vax.

So uh, will these ever make it to an auction site you think?

I have an Acer Chromebook with Celeron N3060 CPU and it runs the SIMH VAX emulator with 64MB for the VAX at the same speed as a Vaxstation 4000/60 and likely the disk is much faster.

I like OpenVMS and am slowly learning more about it; no reason to wait until you see those hit eBay :-)

> I know I’d rather be trying get a load of c99 rebuilt for some mips or other after 20 years that some random version of rust.

Rust 1.0 is 11 years old and it's still trivial to compile Rust code from then. I doubt that will change in the next 9 years.

C is an absolute nightmare in comparison. I tried to compile some old C code I had for Nordic nRF51 chips, only a few years after the chips became available. I gave up. Broken links, missing documentation, etc. etc. I can see why other people here are saying it's standard practice to archive a VM. Not really necessary for Rust.

shrug I guess we have different experiences. Any even small sized rust application I've come across rivals nodejs for amount of deps it pulls in.

Sure, just vendor them in, simples, right?

You could do (it is simple), but all the code is stored on crates.io so there's no real need.

An exception is for crates that wrap C code which might get the code from elsewhere but those are quite rare.

And they’re immutable, forever? No one can pull a leftpad?

Yes, they are immutable. It's only possible to "yank" a specific version, which will prevent new dependencies, but it will still be available for download for existing dependencies.

https://doc.rust-lang.org/cargo/commands/cargo-yank.html

To elaborate on "prevent new dependencies", dependency resolution will never choose a yanked version. However the yanked version remains hosted.

I tried to compile and run some C code from 1991 using a modern C compiler:

https://github.com/sshine/dikumud/commits/master/

It wasn't plug-and-play.

My guess is that when Rust code gets 30 years old, the problem would more likely be that you can't find an already compiled compiler that will work for that old code, and that the compilers themselves need bootstrapping. So you'll just fast-forward the code to work on a new compiler instead.

> you can't find an already compiled compiler that will work for that old code

If it's pure Rust code then the latest Rust compiler will almost certainly work.

The issue will be if some of the crates wrap C code... but that kind of proves the point.

> Rust 1.0 is 11 years old and it's still trivial to compile Rust code from then. I doubt that will change in the next 9 years.

Maybe it's trivial to compile Rust code but it's not trivial to build a project with dependencies. I'm trying to get my feet wet with an official USB example project from Embassy on my RP2040. It doesn't work in the latest git repo for some unknown reason (might be my fault, probably is, but it's not obvious to me).

I'm assuming it worked at some point, maybe something changed and someone forgot to update something somewhere (there are lots of example projects). So I thought I'd "git bisect" until I find a working version and go from there. Well, I cannot get it to build against anything older than a year ago and that version also isn't working for me. It's dependency and Rust edition hell.

Isn't Charon VAX officially supported by HP/Compaq/DEC these days?

People work supporting past embedded codebases and developing new code intended to run on them for decades.

If the toolchain moves on, the product is suck on whatever architecture and developers are stuck running an emulator/docker of some particular vintage of Debian sid.

> You can still use the same old toolchain.

If you keep your old computer around, yes.

The good news is that C seems also contaminated with "move fast, break things " phylosophy. The modern code writer is not able to make things that last more than a couple of months.

Previous versions of the Rust compiler don't just up and disappear just because I moved to a new workstation or setup a new build server. I understand it's not optimal to rely on a download always being available, but even then, that is not at all exclusive to any single language. Why would earlier versions of Rust be susceptible to this but not something like gcc? I don't see it.

> You can still use the same old toolchain.

Unless you find out the compiler was buggy and was producing faulty binaries, but the new compiler can no longer compile the old code.

What you are describing happens all the time. Usually the toolchain provider will continue updating a list of known issues for some time after EOL. Beyond that you have third parties that do it for decades, if the platform is big enough. They collect bug reports from the industry, investigate them, then create lists that you subscribe to. Those lists include detailed examples, explanations, and usually linter rules to detect code that could trigger the bug.

The truth is: If the toolchain was good enough to ship your product, has time to go EOL, and then you do a patch that surfaces an esoteric toolchain bug, then the odds are that you'll know exactly what triggered the bug and you can work around it by writing different code.

Because even if the newer shinier compiler/toolchain had the issue fixed, most companies wouldn't upgrade to it at that point. It's almost never desirable to change your toolchain for a shipping product, you're just introducing more unknowns.

> Because even if the newer shinier compiler/toolchain had the issue fixed, most companies wouldn't upgrade to it at that point. It's almost never desirable to change your toolchain for a shipping product, you're just introducing more unknowns.

This reaction to toolchain stability is quite defensive, and was needed for C, but isn't universally needed. C toolchain updates could break your product because of how loose the C language can be; I've had code that had benign undefined behaviour, until a toolchain update brought in an optimisation that broke it.

Another outcome of a toolchain update could be "no bugs introduced, existing bugs in your codebase now found by diagnostics".

Coding around compiler bugs is pretty standard practice in usecases where you have some random vendor compiler that is based on some ancient gcc if you’re lucky or completely in-house if you’re not.

Fortunately there are more than two compiler versions in the world.

It's easy to install different rust toolchains. You could increment the toolchain version forward and find the next toolchain to include the fix, or even backport the fix to a custom toolchain if you want.

The comments acting like Rust is breaking code all the time are also pretty lost. I've been developing Rust since +/- the 1.0 days and this isn't a common occurrence. When something does need to change it's usually a tightening up of something that was incorrect in the past, and it's easy to fix.

Some of these comments act like everything is going to collapse at any moment and the old code will be unusable, which is pretty ridiculous

Reply on news.ycombinator.com