Rust-Written Replacement To GNU Coreutils Progressing, Some Binaries Now Faster

[Quackdoc]

also, what the hell are these ads? it keeps pushing my comment text filed off screen as i type. i disabled my adblock on this site. how could you? Michael

try using adneaseum if you can, supposedly it just hides ads instead of blocking them, and you can even set it up to click on some ads. and it works well enough on desktop, I've not tried it on mobile yet. and if it works as it says, seems like a win win to me, (assuming preformance isn't an issue)

[re:fi.64]

I will need to see benchmarks of any language being faster than C that is not named machine code or assembly. I find it VERY hard to believe that any Rust language program can beat a pure C language program. I have some friends that are die hard Rust fans and I don't see the point in learning it when I already know C and C++. Throw in a bit of shell scripting, Perl Scripting, and good old C and you can write an entire OS!

Not sure about this particular scenario but I know:

- Rust can assume strict aliasing, helping the optimizer
- Rust's built-in data structures (e.g. their Has map) tend to be pretty well optimized compared to the average basic C implementation

[chocolate]

You can have your own opinion about licensing, but don't forget to be thankful for the engineering behind GNU that made this effort possible.

Because the GNU testsuite is excellent, we now have a proper CI using it to run the tests.

Please read the rest of the post linked by Michael in the article, it's fascinating:

https://sylvestre.ledru.info/blog/2022/01/29/an-update-on-rust-coreutils

Especially this:

Following a suggestion of Brian G, a colleague at Mozilla (he did the same for some Firefox major change), we are now collecting the history of fail/pass/error into a separate repository and generating a daily graph showing the evolution of regression.
gnu-results.png?raw=true.png

True compatibility is the most important thing and it comes before permissive vs. libre licensing.
Cheers and have a splendid weekend.

[re:fi.64]

But doesn't the Coreutils inherntly need to use unsafe code in rust (because it's directly interacting with the operating system)?
It doesn't gain the advantage of rust's memory safety, right?
Or is it still reduces the amount of unsafe code they use, and they still gain it on all the other code?
I am really asking, I don't know if my assumption is correct.

In general, all Rust projects will have some unsafe code in their dependencies at minimum, but it's by and large not the majority. In other words, the surface area for memory bugs is *much* smaller: for instance, the code to parse the contents of procfs is all safe rust, so unexpected values cannot cause memory issues there.

[skeevy420]

try using adneaseum if you can, supposedly it just hides ads instead of blocking them, and you can even set it up to click on some ads. and it works well enough on desktop, I've not tried it on mobile yet. and if it works as it says, seems like a win win to me, (assuming preformance isn't an issue)

Thanks for the suggestion. If it works like it says it works I get hassle free browsing in a manner that doesn't screw over the provider. Like you said, that's a win-win if it works as advertised. My inner anarchist likes the concept behind it -- instead of blocking ads and creating a profile that way, click everything and skew the data in the opposite manner

Not to mention that I've been meaning to do an audit of my FF plugins so installing that and dealing with its conflicts was the perfect motivation.

[Shiba]

Even better is the licensing. MIT. That means that all sorts of operating systems can pick it up and fully integrate it into their stack without worrying about violating the GPL.

OSes that run the risk of violating the GPL can gf themselves.

[bachchain]

But doesn't the Coreutils inherntly need to use unsafe code in rust (because it's directly interacting with the operating system)? It doesn't gain the advantage of rust's memory safety, right? Or is it still reduces the amount of unsafe code they use, and they still gain it on all the other code? I am really asking, I don't know if my assumption is correct.

Using unsafe doesn't invalidate rust's memory guarantees; it simply moves the burden of proof from the compiler to the programmer. And that's still an improvement over C, since it's much easier to audit and reason about a few function calls than an entire program.

[flower]

I will need to see benchmarks of any language being faster than C that is not named machine code or assembly. I find it VERY hard to believe that any Rust language program can beat a pure C language program. I have some friends that are die hard Rust fans and I don't see the point in learning it when I already know C and C++. Throw in a bit of shell scripting, Perl Scripting, and good old C and you can write an entire OS!

usually it is because more people are working on a library which oftens results in it beeing better optimized than a one-person developed single use solution.

so it could be that rust coreutils just rely more on an optimized library than a c implementation.

that said: c is faster. but it is easy to develop slow c code (it is easy to develop slow code in any language).

[sdack]

... and preformance gains, hardly a "waste of time" but if you read the article you would know that

But it is. It is a waste of time when they could have spent their time on fixing GNU Coreutils. Instead, have they created yet another clone, made another claim of RUST being safer, and without providing actual evidence.

One could have spent the time to improve GNU Coreutils performance if there is an opportunity to do so. So instead of giving back to the GNU Coreutils, which people have been using for decades, are the GNU Coreutils instead used to make a show.

Frankly, not giving back, but instead making it into some sort of cancel culture is pathetic.

[Quackdoc]

But it is. It is a waste of time when they could have spent their time on fixing GNU Coreutils. Instead, have they created yet another clone, made another claim of RUST being safer, and without providing actual evidence.

One could have spent the time to improve GNU Coreutils performance if there is an opportunity to do so. So instead of giving back to the GNU Coreutils, which people have been using for decades, are the GNU Coreutils instead used to make a show.

Frankly, not giving back, but instead making it into some sort of cancel culture is pathetic.

what the hell does this have to do with cancel culture?

I don't know if you ever programmed in C, but it is leagues easier to work with rust and optimize code. there is plenty of evidence to rust being safer to work with than C, that doesn't mean that C cannot be safe, just that it is far easier to do. no one made a show of this. they changed it to a more permissive licensing, maybe because it suits their needs better, or maybe they just like the license better.

re-write in rust, as much of a meme, IMO is a great thing, because rust is just simply IMO far easier to work with and get similar or better preformance than other languages.

it's the GNU cultists that constantly make a stink about things like this, don't get me wrong, I love gplv2, I think it is a great license. but I see no proof this has any relation to "cancel culture"