Comparing modern C++ and Rust in terms of safety and performance

C++ and Rust are two popular programming languages that offer a balance between performance and safety. While C++ has been around for decades and is widely used in many applications, Rust is a relatively new language that has gained popularity for its safety features. In this post, we will compare the safety and performance of modern C++ and Rust.

Safety

Null Pointers:

One of the most common errors in C++ is null pointer dereferencing. A null pointer is a pointer that does not point to a valid memory location. Dereferencing a null pointer can result in a segmentation fault, which can crash the program. In C++, it is the responsibility of the programmer to ensure that pointers are not null before dereferencing them.

Rust, on the other hand, has a built-in option type that ensures that a value is either present or absent. This eliminates the need for null pointers. For example, consider the following C++ code:

int* ptr = nullptr;
*ptr = 42;

This code will crash with a segmentation fault, because the pointer ptr is null. In Rust, we would write this code as follows:

let mut opt = None; 
opt = Some(42);

In this case, opt is an option type that is either Some with a value or None. If we try to access opt when it is None, Rust will throw a runtime error. This eliminates the possibility of null pointer dereferencing.

Data Races:

Another common error in concurrent programming is data races. A data race occurs when two or more threads access the same memory location simultaneously, and at least one of the accesses is a write. This can result in undefined behavior, including incorrect results or crashes.

C++ has a number of synchronization primitives that can be used to prevent data races, such as mutexes and condition variables. However, it is the responsibility of the programmer to ensure that these primitives are used correctly. In Rust, the ownership and borrowing system ensures that only one thread can modify a value at a time, preventing data races. For example, consider the following C++ code:

#include <thread>
#include <iostream>

int x = 0;
void increment() 
{
  for (int i = 0; i < 1000000; i++) 
  { x++; } 
} 

int main() 
{
  std::thread t1(increment);
  std::thread t2(increment);
  t1.join();
  t2.join();
  std::cout << x << std::endl; 
}

In this code, we have two threads that increment the variable x simultaneously. This can result in a data race, because both threads are modifying the same memory location. In Rust, we can use the ownership and borrowing system to prevent this kind of error:

use std::sync::Mutex;

let x = Mutex::new(0);

let t1 = std::thread::spawn(|| {
    for _ in 0..1000000 {
        let mut x = x.lock().unwrap();
        *x += 1;
    }
});

let t2 = std::thread::spawn(|| {
    for _ in 0..1000000 {
        let mut x = x.lock().unwrap();
        *x += 1;
    }
});

t1.join().unwrap();
t2.join().unwrap();

println!("{}", *x.lock().unwrap());

In this code, we use a mutex to ensure that only one thread can modify x at a time. This prevents data races and ensures that the output is correct.

Performance:

Compilation Time: One area where C++ has traditionally been faster than Rust is in compilation time. C++ compilers have been optimized for decades to compile large code bases quickly. However, Rust has been improving its compilation times with recent releases.

For example, consider compiling the following C++ code:

#include <iostream>

int main() {
    std::cout << "Hello, world!" << std::endl;
    return 0;
}

Using the Clang compiler on a MacBook Pro with a 2.6 GHz Intel Core i7 processor, this code takes about 0.5 seconds to compile. In comparison, compiling the equivalent Rust code:

fn main() {
    println!("Hello, world!");
}

Using the Rust compiler takes about 0.8 seconds. While C++ is faster in this example, Rust is improving its compilation times and is expected to become more competitive in the future.

Execution Time:

When it comes to execution time, C++ and Rust are both high-performance languages that can be used to write fast, low-level code. However, the specific performance characteristics of each language can vary depending on the task at hand.

For example, consider the following C++ code that calculates the sum of all integers from 1 to 1 billion:

#include <iostream>

int main() {
    long long sum = 0;
    for (int i = 1; i <= 1000000000; i++) {
        sum += i;
    }
    std::cout << sum << std::endl;
    return 0;
}

Using the Clang compiler on a MacBook Pro with a 2.6 GHz Intel Core i7 processor, this code takes about 5 seconds to execute. In comparison, the equivalent Rust code:

fn main() {
    let mut sum = 0;
    for i in 1..=1000000000 {
        sum += i;
    }
    println!("{}", sum);
}

Using the Rust compiler takes about 6 seconds to execute on the same machine. In this case, C++ is slightly faster than Rust, but the difference is not significant.

Memory Usage:

Another important factor to consider when comparing performance is memory usage. C++ and Rust both provide low-level control over memory, which can be beneficial for performance. However, this also means that the programmer is responsible for managing memory and avoiding memory leaks.

For example, consider the following C++ code that creates a large array of integers:

#include <iostream>
#include <vector>

int main() {
    std::vector<int> arr(100000000);
    std::cout << arr.size() << std::endl;
    return 0;
}

This code creates a vector of 100 million integers, which takes up about 400 MB of memory. In comparison, the equivalent Rust code:

fn main() {
    let arr = vec![0; 100000000];
    println!("{}", arr.len());
}

Using the Rust compiler takes up about 400 MB of memory. In this case, C++ and Rust have similar memory usage.

Conclusion

Both C++ and Rust offer a balance between performance and safety, but Rust’s ownership and borrowing system gives it an edge in terms of safety. Rust’s memory safety features make it less prone to common programming errors like null pointer dereferencing, data races, and memory leaks.

In terms of performance, both C++ and Rust are high-performance languages, but Rust’s memory safety features and zero-cost abstractions give it an advantage over C++. Rust’s efficient concurrency also makes it well-suited for high-performance and parallel computing.

Ultimately, the choice between C++ and Rust depends on the specific needs of your project. If you need low-level memory management and direct hardware access, C++ may be the better choice. If you prioritize safety and memory efficiency, Rust may be the way to go.