Hacker News Reader

Comment by geofft | Hacker News Reader

No, most #![no_std] libraries do not link libc in the way you describe (having a direct FFI dependency on malloc) - they link liballoc, which has a pluggable allocator interface. On some platforms and some configurations (including most normal userspace platforms), the default allocator used by liballoc uses malloc.

It's actually hard to go out of your way and call malloc directly, because FFI calls are unsafe. It's a lot easier to use Box/Vec/String/etc., all of which are defined in liballoc and use the pluggable allocator interface.

I know this because I've successfully used #![no_std] libraries in places where libc doesn't exist and no function named malloc exists, and they do work. If you're having a linker issue it's almost certainly because you haven't changed the default allocator - if you have an example of this I'd be happy to take a look at debugging it.

View on HN

> and they do work.

Maybe you are just using different no_std libraries that I am using, but pretty much all of the no_std libraries that I use have `libc` as a direct dependency.

Not only for malloc, many of them just needs to write something to stdout or a file, generate random numbers, ..., and that's hard to do without libc. Why they advertise themselves as no_std escapes my comprehension.

Huh, the libc crate itself claims #![no_std] support. I agree that's confusing and counterintuitive... I see what it means in terms of semantics but I don't understand why that's useful.

The libc crate does correctly claim #![no_std] support, because #![no_std] means "Does not require the Rust standard library", and libc does not require it.

The two main issues I see with #![no_std] are:

* its a flag for "doesn't link the standard library" but the standard library is often too high-level for bare metal apps that want to be in precise control about everything that gets linked

* it isn't a contract of any kind, so you can't rely on it for anything. In fact, this code is correct and works as intended:

    #![no_std]
    extern crate std;

This is problematic, because many #![no_std] libraries end up linking libstd "by accident", even though that's probably not their intent.

So I agree with you that #![no_std] isn't a silver bullet. I think it is still useful in that it lets you avoid linking the standard library by default, which is necessary condition for embedded development. It is not a sufficient condition, in that in practice you actually want to forbid the standard library and other libraries from being linked, and not just "don't link them by default, but do so by accident".

`#![no_std]` really means that sysroot isn't available for your platform and you gotta roll your own.

Project I'm working on right now is `#[no_std]`, but my own sysroot covers 90% of real `std`. Which is why I even export it as `std`, so I can use pretty much all crates I would usually use. Because if your platform has an allocator, doesn't matter where it came from, then you can add `liballoc`. `libstd` = `libcore` + `liballoc` + locks + allocator + threads + a few other things that usually depend liballoc. Which means that you have nearly entire libstd available to you if your target platform has an allocator.

Most `#[no_std]` crates that depend on allocator clearly state that they depend on `liballoc` and that you gotta provide one. What I'm trying to say `#[no_std]` doesn't mean embed, it just means sysroot isnt' available for target.

Reply on news.ycombinator.com