Comment by serial_dev
10 months ago
This one bit looks exactly right, though:
> As for how to move forward, (...) Either Linus takes the pull, and whatever Christoph says is irrelevant, or he doesn't, and R4L dies. Everything else is a waste of everyone's time and energy.
It does look like maintainers should have a "disagree and commit" mentality at some point, whatever decision they end up making.
I thought Rust in Linux was evaluated, discussed and agreed upon years ago. The fact that there are people still trying to sabotage it shows that they don't follow the "disagree and commit" principle.
They are more like "disagree and make the others lives a living hell until they bend to my will".
> I thought Rust in Linux was evaluated, discussed and agreed upon years ago.
I've seen Linus talk about it in one of his public chats with Dirk Hohndel as an interesting experiment that might succeed or fail, or that's the impression I got. I'm not sure everyone else got that memo.
> disagree and commit
It is not some million dollar RSUs getting vested by year end either way. A lot of them working for the love of craft and prestige. If they can just rollover on a technical disagreement then corporate office job is more suitable than open source OS kernel.
The Linux kernel is most definitely not developed for free by some college student in their bedroom for the most part. It's Intel and Red Hat and similar.
Edit: related email by Hector: https://lore.kernel.org/rust-for-linux/c5a49bcb-45cf-4295-80...
"R4L dies" is hyperbolic nonsense. There are many ways/places to integrate Rust in the kernel, even for wrappers. It's not like "I think this code should be someone else" automatically means the death of Rust.
Marcan goes full nuclear every other time someone disagrees with him. He's very much the "if it's not my preferred way then we might as well not do it at all"-type of engineer (many of us have worked with those people).
Rust is only allowed for drivers, because that sidesteps the platform support questions.
DMA is needed for an overwhelming number of useful drivers.
If you can’t use DMA from Rust, then you can’t really properly evaluate the usefulness of Rust, hence the effort is basically dead.
People can just include their own dma.rs in driver/.../mydriver/
Whether that's a good or bad approach is besides the point. It's what was suggested as an alternative, and clearly it's something that would work. "You can't use DMA from Rust" is just not true.
7 replies →
I don't know the better way to ask this question, so I'm just going to ask it:
"Congratulations, you're right, and nobody cares."
Now what? Just complain that the kernel isn't hospitable to Rust, and hope 15 years from now we're all using some Linux-compatible kernel built ground-up in Rust?
I genuinely want to know where you go from this position.
1 reply →
> Every additional bit that the another language creeps in drastically reduces the maintainability of the kernel as an integrated project. The only reason Linux managed to survive so long is by not having internal boundaries, and adding another language complely breaks this. You might not like my answer, but I will do everything I can do to stop this.
https://lwn.net/ml/all/20250131075751.GA16720@lst.de/
Maybe you can try to read what Christoph Hellwig said first.
That's a much stronger argument when you aren't at the same time yelling about the email patch system. I want R4L to succeed!