speice.io/_drafts/understanding-allocations-in-rust.md

layout	title	description	category	tags
post	Allocations in Rust	An introduction to the memory model		rust

There's an alchemy of distilling complex technical topics into articles and videos that change the way programmers see the tools they interact with on a regular basis. I knew what a linker was, but there's a staggering amount of complexity in between main() and your executable. Rust programmers use the Box type all the time, but there's a rich history of the Rust language itself wrapped up in how special it is.

In a similar vein, I want you to look at code and understand how memory is used; the complex choreography of operating system, compiler, and program that frees you to focus on functionality far-flung from frivolous book-keeping. The Rust compiler relieves a great deal of the cognitive burden associated with memory management, but we're going to step into its world for a while.

Let's learn a bit about memory in Rust.

Table of Contents

This post is intended as both guide and reference material; we'll work to establish an understanding of the different memory types Rust makes use of, then summarize each section for easy citation in the future. To that end, a table of contents is provided to assist in easy navigation:

• Foreword
• Stacking Up: Non-Heap Memory Types
• Piling On: Rust and the Heap
• Compiler Optimizations Make Everything Complicated
• Summary: When Does Rust Allocate?
• Appendix and Further Reading

Foreword

There's a simple checklist to see if you can skip over reading this article. You must:

1. Only write #![no_std] crates
2. Never use unsafe
3. Never use #![feature(alloc)]

For some uses of Rust, typically embedded devices, these constraints make sense. They're working with very limited memory, and the program binary size itself may significantly affect what's available! There's no operating system able to manage this "virtual memory" junk, but that's not an issue because there's only one running application. The embedonomicon is ever in mind, and interacting with the "real world" through extra peripherals is accomplished by reading and writing to exact memory addresses.

Most Rust programs find these requirements overly burdensome though. C++ developers would struggle without access to std::vector (except those hardcore no-STL guys), and Rust developers would struggle without std::vec. But in this scenario, std::vec is actually part of the alloc crate, and thus off-limits (because the alloc crate requires #![feature(alloc)]). Also, Box is right out for the same reason.

Whether writing code for embedded devices or not, the important thing in both situations is how much you know before your application starts about what its memory usage will look like. In embedded devices, there's a small, fixed amount of memory to use. In a browser, you have no idea how large google.com's home page is until you start trying to download it. The compiler uses this information (or lack thereof) to optimize how memory is used; put simply, your code runs faster when the compiler can guarantee exactly how much memory your program needs while it's running. This post is all about understanding the optimization tricks the compiler uses, and how you can help the compiler and make your programs more efficient.

Now let's address some conditions and caveats before going much further:

• We'll focus on "safe" Rust only; unsafe lets you use platform-specific allocation API's (think the [libc] and [winapi] implementations of [malloc]) that we'll ignore.
• We'll assume a "debug" build of Rust code (what you get with cargo run and cargo test) and address (pun intended) "release" mode at the end (cargo run --release and cargo test --release).
• All content will be run using Rust 1.31, as that's the highest currently supported in the Compiler Exporer. As such, we'll avoid talking about things like compile-time evaluation of static that are available in nightly.
• Because of the nature of the content, some (very simple) assembly-level code is involved. We'll keep this to a minimum, but I needed a refresher on the push and pop instructions while writing this post.

And finally, I'll do what I can to flag potential future changes, but the Rust docs have a notice worth repeating:

Rust does not currently have a rigorously and formally defined memory model.

-- the docs

Stacking Up: Non-Heap Memory Types

We'll start with the "happy path": what happens when Rust is able to figure out at compile time how much memory will be used in your program.

This is important because of the extra optimizations Rust uses when it can predict how much memory is needed! Let's go over a quick example:

const MICROS_PER_MILLI: u32 = 1000;
const NANOS_PER_MICRO: u32 = 1000;

pub fn millis_to_nanos(millis: u32) -> u32 {
    let micros = millis * MICROS_PER_MILLI;
    let nanos = micros * NANOS_PER_MICRO;

    return nanos;
}

-- Compiler Explorer

Forgive the overly simple code, but this shows off what the compiler can figure out about your program:

1. There's one u32 passed to the function, and two u32's used in the function body. Each one is 4 bytes, for a total of 12 bytes. We can temporarily reserve space for all variables because we know exactly how much space is needed.
   • If you're looking at the assembly: millis is stored in edi, micros is stored in eax, and nanos is stored in ecx. The eax register is re-used to store the final result.
2. Because MICROS_PER_MILLI and NANOS_PER_MICRO are constants, the compiler never allocates memory, and just burns the constants into the final program.
   • Look for the instructions mov edi, 1000 and mov ecx, 1000.

Given this information, the compiler can efficiently lay out your memory usage so that the program never needs to ask the kernel/allocator for memory! This example was a bit silly though, so let's talk about the more interesting details.

const and static: Program Allocations

The first memory type we'll look at is pretty special: when Rust can prove that a value is fixed for the life of a program, and when a reference is valid for the duration of the program (static, not specifically 'static). Understanding the distinction between value and reference is important for reasons we'll go into below. The full specification for these two memory types is available, but we'll take a hands-on approach to the topic.

const

The quick summary is this: const declares a read-only block of memory that is loaded as part of your program binary (during the call to exec(3)). Any const value resulting from calling a const fn is guaranteed to be materialized at compile-time (meaning that access at runtime will not invoke the const fn), even though the const fn functions are available at run-time as well. The compiler can choose to copy the constant value wherever it is deemed practical. Getting the address of a const value is legal, but not guaranteed to be the same even when referring to the same named identifier.

The first point is a bit strange - "read-only memory". The Rust book mentions in a couple places that using mut with constants is illegal, but it's also important to demonstrate just how immutable they are. Typically in Rust you can use "inner mutability" to modify things that aren't declared mut. RefCell provides an API to guarantee at runtime that some consistency rules are enforced:

use std::cell::RefCell;

fn my_mutator(cell: &RefCell<u8>) {
    // Even though we're given an immutable reference,
    // the `replace` method allows us to modify the inner value.
    cell.replace(14);
}

fn main() {
    let cell = RefCell::new(25);
    // Prints out 25
    println!("Cell: {:?}", cell);
    my_mutator(&cell);
    // Prints out 14
    println!("Cell: {:?}", cell);
}

-- Rust Playground

When const is involved though, modifications are silently ignored:

use std::cell::RefCell;

const CELL: RefCell<u8> = RefCell::new(25);

fn my_mutator(cell: &RefCell<u8>) {
    cell.replace(14);
}

fn main() {
    // First line prints 25 as expected
    println!("Cell: {:?}", &CELL);
    my_mutator(&CELL);
    // Second line *still* prints 25
    println!("Cell: {:?}", &CELL);
}

-- Rust Playground

And a second example using Once:

use std::sync::Once;

const SURPRISE: Once = Once::new();

fn main() {
    // This is how `Once` is supposed to be used
    SURPRISE.call_once(|| println!("Initializing..."));
    // Because `Once` is a `const` value, we never record it
    // having been initialized the first time, and this closure
    // will also execute.
    SURPRISE.call_once(|| println!("Initializing again???"));
}

-- Rust Playground

When the const specification refers to "rvalues", this is what they mean. Clippy will treat this as an error, but it's still something to be aware of.

The next thing to mention is that const values are loaded into memory as part of your program binary. Because of this, any const values declared in your program will be "realized" at compile-time; accessing them may trigger a main-memory lookup (with a fixed address, so your CPU may be able to prefetch the value), but that's it.

use std::cell::RefCell;

const CELL: RefCell<u32> = RefCell::new(24);

pub fn multiply(value: u32) -> u32 {
    value * (*CELL.get_mut())
}

-- Compiler Explorer

The compiler only creates one RefCell, and uses it everywhere. However, that value is fully realized at compile time, and is fully stored in the .L__unnamed_1 section.

If it's helpful though, the compiler can choose to copy const values.

const FACTOR: u32 = 1000;

pub fn multiply(value: u32) -> u32 {
    value * FACTOR
}

pub fn multiply_twice(value: u32) -> u32 {
    value * FACTOR * FACTOR
}

-- Compiler Explorer

In this example, the FACTOR value is turned into the mov edi, 1000 instruction in both the multiply and multiply_twice functions; the "1000" value is never "stored" anywhere, as it's small enough to inline into the assembly instructions.

Finally, getting the address of a const value is possible but not guaranteed to be unique (given that the compiler can choose to copy values). In my testing I was never able to get the compiler to copy a const value and get differing pointers, but the specifications are clear enough: don't rely on pointers to const values being consistent. To be frank, caring about locations for const values is almost certainly a code smell.

static

Static variables are related to const variables, but take a slightly different approach. When the compiler can guarantee that a reference is fixed for the life of a program, you end up with a static variable (as opposed to values that are fixed for the duration a program is running). Because of this reference/value distinction, static variables behave much more like what people expect from "global" variables. We'll look at regular static variables first, and then address the lazy_static!() and thread_local!() macros later.

More generally, static variables are globally unique locations in memory, the contents of which are loaded as part of your program being read into main memory. They allow initialization with both raw values and const fn calls, and the initial value is loaded along with the program/library binary. All static variables must be of a type that implements the Sync marker trait. And while static mut variables are allowed, mutating a static is considered an unsafe operation.

The single biggest difference between const and static is the guarantees provided about uniqueness. Where const variables may or may not be copied in code, static variables are guarantee to be unique. If we take a previous const example and change it to static, the difference should be clear:

static FACTOR: u32 = 1000;

pub fn multiply(value: u32) -> u32 {
    value * FACTOR
}

pub fn multiply_twice(value: u32) -> u32 {
    value * FACTOR * FACTOR
}

-- Compiler Explorer

Where previously there were plenty of references to multiplying by 1000, the new assembly refers to FACTOR as a named memory location instead. No initialization work needs to be done, but the compiler can no longer prove the value never changes during execution.

Next, let's talk about initialization. The simplest case is initializing static variables with either scalar or struct notation:

#[derive(Debug)]
struct MyStruct {
    x: u32
}

static MY_STRUCT: MyStruct = MyStruct {
    // You can even reference other statics
    // declared later
    x: MY_VAL
};

static MY_VAL: u32 = 24;

fn main() {
    println!("Static MyStruct: {:?}", MY_STRUCT);
}

-- Rust Playground

Things get a bit weirder when using const fn. In most cases, things just work:

#[derive(Debug)]
struct MyStruct {
    x: u32
}

impl MyStruct {
    const fn new() -> MyStruct {
        MyStruct { x: 24 }
    }
}

static MY_STRUCT: MyStruct = MyStruct::new();

fn main() {
    println!("const fn Static MyStruct: {:?}", MY_STRUCT);
}

-- Rust Playground

However, there's a caveat: you're currently not allowed to use const fn to initialize static variables of types that aren't marked Sync. As an example, even though RefCell::new() is const fn, because RefCell isn't Sync, you'll get an error at compile time:

use std::cell::RefCell;

// error[E0277]: `std::cell::RefCell<u8>` cannot be shared between threads safely
static MY_LOCK: RefCell<u8> = RefCell::new(0);

-- Rust Playground

It's likely that this will change in the future though, so be on the lookout.

Which leads well to the next point: static variable types must implement the Sync marker. Because they're globally unique, it must be safe for you to access static variables from any thread at any time. Most struct definitions automatically implement the Sync trait because they contain only elements which themselves implement Sync. This is why earlier examples could get away with initializing statics, even though we never included an impl Sync for MyStruct in the code. For more on the Sync trait, the Nomicon has a much more thorough treatment. But as an example, Rust refuses to compile our earlier example if we add a non-Sync element to the struct definition:

use std::cell::RefCell;

struct MyStruct {
    x: u32,
    y: RefCell<u8>,
}

// error[E0277]: `std::cell::RefCell<u8>` cannot be shared between threads safely
static MY_STRUCT: MyStruct = MyStruct {
    x: 8,
    y: RefCell::new(8)
};

-- Rust Playground

Finally, while static mut variables are allowed, mutating them is an unsafe operation. Unlike const however, interior mutability is acceptable. To demonstrate:

use std::sync::Once;

// This example adapted from https://doc.rust-lang.org/std/sync/struct.Once.html#method.call_once
static INIT: Once = Once::new();

fn main() {
    // Note that while `INIT` is declared immutable, we're still allowed
    // to mutate its interior
    INIT.call_once(|| println!("Initializing..."));
    // This code won't panic, as the interior of INIT was modified
    // as part of the previous `call_once`
    INIT.call_once(|| panic!("INIT was called twice!"));
}

-- Rust Playground

push and pop: Stack Allocations

const and static are perfectly fine, but it's very rare that we know at compile-time about either references or values that will be the same for the entire time our program is running. Put another way, it's not often the case that either you or your compiler know how much memory your entire program will need.

However, there are still some optimizations the compiler can do if it knows how much memory individual functions will need. Specifically, the compiler can make use of "stack" memory (as opposed to "heap" memory) which can be managed far faster in both the short- and long-term. When requesting memory, the push instruction can typically complete in 1 or 2 cycles (<1 nanosecond on modern CPUs). Heap memory instead requires using an allocator (specialized software to track what memory is in use) to reserve space. And when you're finished with memory, the pop instruction likewise runs in 1-3 cycles, as opposed to an allocator needing to worry about memory fragmentation and other issues. All sorts of incredibly sophisticated techniques have been used to design allocators:

• Garbage Collection strategies like Tracing (used in Java) and Reference counting (used in Python)
• Thread-local structures to prevent locking the allocator in tcmalloc
• Arena structures used in jemalloc, which until recently was the primary allocator for Rust programs!

But no matter how sophisticated your allocator is, the principle remains: the fastest allocator is the one you never use. As such, we're not going to go in detail on how exactly the push and pop instructions work, and we'll focus instead on the conditions that enable the Rust compiler to use stack-based allocation for variables.

Now, one question I hope you're asking is "how do we distinguish stack- and heap-based allocations in Rust code?" There are three strategies I'm going to use for this:

1. Any time the push or pop instructions are used, or the rsp register is modified, this is a stack allocation:

   pub fn stack_alloc(x: u32) -> u32 {
       // Space for `y` is allocated by subtracting from `rsp`,
       // and then populated
       let y = [1u8, 2, 3, 4];
       // Space for `y` is deallocated by adding back to `rsp`
       x
   }
   

   -- Compiler Explorer
2. Any time call core::ptr::drop_in_place occurs, a heap allocation has occurred sometime in the past and it is now time for us to de-allocate the memory:

   pub fn heap_alloc(x: usize) -> usize {
       // Space for elements in a vector has to be allocated
       // on the heap, and is then de-allocated once the
       // vector goes out of scope
       let y: Vec<u8> = Vec::with_capacity(x);
       x
   }
   

   -- Compiler Explorer (drop_in_place happens on line 1321)
3. Using a special GlobalAlloc implementation to track when heap allocations occur. For this post, I'll be using qadapt to trigger a panic if heap allocations occur; code that doesn't panic doesn't use heap allocations, and by necessity uses stack allocation instead.

With all that in mind, let's get into the details. The unfortunate thing about stack allocations in Rust is that there's not a good way to glance at code and figure out where allocations on the heap happen. Looking at other languages, Java mostly cares about new MyObject() (yes, I'm conveniently ignoring autoboxing). C makes things clear with calls to malloc(3). C++ has the new keyword, std::make_unique(), and std::make_shared() (though things are admittedly more complex with RAII). All languages exist on a memory management spectrum, from Zig forcing you to provide an allocator, to Python/Ruby/JavaScript assuming you generally never worry about those details.

So what can be done to make sure your program is using stack allocations? A couple of guidelines are in order:

For code you control:

• Never using types in the alloc crate is sufficient. While you should always review its contents, the most notable members are Box, refcount types (Rc, Arc)
• Dynamically resizable types need to be treated with care; we'll go into detail later, but pay attention to String, Vec, and HashMap
• Enums and other wrapper types will not trigger heap allocations unless the underlying type also needs heap allocation. You can use Option, Result with reckless abandon.
• Arrays are guaranteed to be stack-allocated in all circumstances.

For code outside your control:

• Review the code to make sure it abides by the guidelines above
• Use a custom allocator like qadapt as an automated check to make sure that stack allocations are used in code you care about.

Example: Why doesn't Vec::new() go to the allocator?

Questions:

1. What is the "Push" instruction? Why do we like the stack?
2. How does Rust allocate arguments to the function?
3. How does Rust allocate variables created in the function but never returned?
4. How does Rust allocate variables created in the function and returned?
5. How do Option<> or Result<> affect structs?
6. How are arrays allocated?
7. Legal to pass an array as an argument?
8. Can you force a heap allocation with arrays that are larger than stack size?
   • Check ulimit -s
9. Can you force heap allocation by returning something that escapes the stack?
   • Will #[inline(always)] move this back to a stack allocation?

Piling On - Rust and the Heap

Example: How to trigger a heap allocation

Questions:

1. Where do collection types allocate memory?
2. Does a Box<> always allocate heap?
   • Yes, with exception of compiler optimizations
3. Passing Box vs. genericizing/monomorphization
   • If it uses dyn Trait, it's on the heap?
   • What if the trait implements Sized?
4. Other pointer types? Do Rc<>/Arc<> force heap allocation?
   • Maybe? Part of the alloc crate, but should use qadapt to check
5. How many allocations happen before main() is called?

Compiler Optimizations Make Everything Complicated

1. Box<> getting inlined into stack allocations
2. Vec::push() === Vec::with_capacity() for fixed/predictable capacities
3. Inlining statics that don't change value

Appendix and Further Reading

[libc]: CRATES.IO LINK [winapi]: CRATES.IO LINK [malloc]: MANPAGE LINK