Rust for embedded

[legacy]

engines

they are usually written in C/C++
sometimes directly in assembly

eLua can pass arguments, do foreign-calls, and get results
it works like a wrapper, but it adds a ton of scripting capabilities

[Marco]

And with following good programming practices, it is not that hard to write good programs in C.

So, how many exploitable overflow bugs do you think are hidden in your internet facing code after using those good programming practices?

[ataradov]

So, how many exploitable overflow bugs do you think are hidden in your internet facing code after using those good programming practices?

I don't know, obviously. But there is no proof that Rust actually helps anything here. Rust only helps if you never-ever use unsafe keyword, which never happens in a real program.

If anything, unsafe sections simply point out sections where you should look for exploits first. Convenient language for both sides, I guess

[Marco]

It will inevitably happen in hardware interfacing code, it need never happen in anything else as long as you can swallow the performance overhead for runtime bounds checking.

[legacy]

as long as you can swallow the performance overhead for runtime bounds checking.

that's the same problem with ADA, and the reason why I was suggesting big-iron processors like PowerPC rather then small machines

[tggzzz]

It will inevitably happen in hardware interfacing code, it need never happen in anything else as long as you can swallow the performance overhead for runtime bounds checking.

But who cares about correctness[1] so long as it is fast? No, that's not a joke, unfortunately.

[1] or at least lack of easily and mechanically detectable errors

[Feynman]

We were and still are definitely lacking a decent programming language for embedded systems.

But since the C-beast is pretty much tamed (MISRA, Lint and stuff), I think there is no need anymore for another "high level" language for deeply embedded systems.

[SreejitS]

We were and still are definitely lacking a decent programming language for embedded systems.

Agreed 

[legacy]

We were and still are definitely lacking a decent programming language for embedded systems.

Agreed 

I agree, too 

[autobot]

I just wish to lower the entry barrier for embedded systems....

The arduino was very sucsessful in that but had big limitations. The mbed(and mbed  OS) could be the answer - combining global collaboration and code sharing, under a well designed structure(with a lot of attention to tools/security/power/cost etc) and somewhat higher barrier than the arduino in order to create higher quality in contributed code.