Google Supports Getting Rust Into The Linux Kernel

[s_j_newbury]

I agree. They're just an easy company to pick in a thought experiment. Facebook, Sony, Microsoft, or any other number of companies probably have more of a reason than Google.

I'm more curious in the GPL to Compatible ratio. I think it would be weird if it was actually less than 50% GPL...Like a "Huh, that's odd and not what I expected" weird.

100% is GPL, since that's the license it is distributed under. The whole point of compatible licenses, is they do not conflict with re-licensing under the GPL! Non-GPL code in the kernel is only "non-GPL" when you get it from the upstream copyright holder, or a distribution under another license, any code in distributed under "Linux" is copyright(/left) as a whole. Not that it is likely to get you into any trouble if you reuse "non-GPL" code you've copied from the kernel ostensibly under another license, since how can it be proved that's where you got it from? Although you probably want to keep that to yourself...

Mind you, IANAL

[milkylainen]

On the whole, a lot of Rust code makes my eyes hurt.
Some of it is not that bad, but damn can Rust look like somebody puked all over the place.

As per usual, I guess it's a property of whomever is writing, not the language.
I hope the readability level doesn't degrade.

[ThoreauHD]

Reminds me of this
https://twitter.com/cmuratori/status...816152064?s=20

And like all comedy, it's funny because it's true.

I don't know anything programmatically about rust. But I know a dog humping a leg when I see it.

"We have a lot of work to do regarding `alloc` and slicing `core` for things that are not needed for the kernel (floating-point, etc.).

We haven't done it just yet because it is not a trivial amount of work and we wanted to have some overall sentiment from you and the community overall before tackling everything. But it is doable and there isn't any fundamental reason that prevents it (in fact, the language supports no-allocation code)."

Seems like there are a few non-trivials to work out before Linus shoots himself in the foot.

[Volta]

It was not meant as me disagreeing. You wrote that he didn't support it, so I just wanted to add a bit more details about his attitude towards it and the concerns that were raised.

Don't worry, birdie is known troll, so he doesn't even know what he wrote.

[ddriver]

The worst part of c++ is its most powerful feature - meta programming, the api is just too awful, verbose and stiff. The language additions are rather inflexible, and they lack genericity and intuitiveness, and totally bloated the language size.

And yes, you could have perfectly usable c++ with a small subset of metaprogramming, you rarely need it all, but as one's ability to design code improve there is this urge to "design code" rather than to type the same old boilerplate over and over again like a mindless monkey.

IIRC template generation is also one of the main reasons for Linus to dislike c++, so I wonder how does rust do in the metaprogramming department.

[birdie]

It was not meant as me disagreeing. You wrote that he didn't support it, so I just wanted to add a bit more details about his attitude towards it and the concerns that were raised.

Ah, OK, I didn't catch it, sorry.

[AnAccount]

Have you actually read what I wrote? You're disagreeing with me and then you're repeating exactly what I said: Linus has very valid concerns and once they are resolved, he's OK with Rust.

It was not meant as me disagreeing. You wrote that he didn't support it, so I just wanted to add a bit more details about his attitude towards it and the concerns that were raised.

[piotrj3]

C can't go away because it is the entire computing platform.

Rust stands a good chance of replacing C++ in the future. These concerns about Rust in the Linux kernel are actually very similar to the concerns of C++ in the kernel. As of yet this hasn't happened either. Obviously the Rust guys are a little bit more noisy and the language does offer a few extra benefits above C++ in terms of (enforced) safety that might just tip the scale.

C++ has few giant disadvantages that C/Rust doesn't suffer from. Memory managment in C++ can be done in 100 diffrent ways, and that alone makes C++ code very hard to maintain without writing a book about what is allowed and what is not. Not to mention guessing what really causes bug or error in C++ is much harder (in C it is just in plain sight, Rust has much less bug errors and has very good error information, while C++ it is often invisible as well compiler error message can be longer then entire kernel code itself in templates for example).

Personally I don't have problem reading most code written in any language, it is not uncommon for me to help someone succesfully write a code in language I never touched myself, but everytime I see C++ I have vietnam flashbacks.

[danielmaciel]

Reminds me of this

https://twitter.com/cmuratori/status/1367627549816152064?s=20

[skeevy420]

This is entirely unlikely. The kernel userspace boundary makes GPL a non problem for Google especially now that they have treble in place. They only have a small fraction of the Linux kernel developers necessary to pull this off and many existing ones won't join a non GPL effort. Hiring an army of new engineers and have them understand all the nuances is not worth it when they could invest that into Fuchsia instead.

I agree. They're just an easy company to pick in a thought experiment. Facebook, Sony, Microsoft, or any other number of companies probably have more of a reason than Google.

I'm more curious in the GPL to Compatible ratio. I think it would be weird if it was actually less than 50% GPL...Like a "Huh, that's odd and not what I expected" weird.