News for hackers. Stuff that matters.
Re:Rust for Filesystems by _flux on Monday July 15, 2024 @09:32AM (#40967751)
I'm not sure what you are saying. Maybe the first proposal isn't the best one, so some exploration is warranted.

Is Kent an unwilling pawn in the game of getting Rust implemented in the Linux kernel? Or is he not serious enough about getting it in that his support should not be considered?

From what I've seen it reads to me Kent would be very happy to implement bcachefs in Rust in the Linux kernel. Bcachefs-tools does make use of it.

Re:Rust for Filesystems by pimeys on Monday July 15, 2024 @10:30AM (#40968195)
Kent has been really vocal about writing bcachefs in Rust in the IRC channel. They even started some work already, but decided to wait until the common abstractions are ready and merged.
Re:Rust for Filesystems by tucnak on Monday July 15, 2024 @09:53AM (#40967921)
> Is Kent an unwilling pawn in the game of getting Rust implemented in the Linux kernel? Or is he not serious enough about getting it in that his support should not be considered?

Neither. In fact, Kent is the only person in this story that has any agency with respect to Linux whatsoever; he's the reason we're having this conversation in the first place. And indeed, this is also the reason Rust people wish to co-opt him so hard. This pattern of behaviour is not new; in fact, pretty much all "revolutionary" movements never pass on the chance to mistake good attitude (towards them) for unequivocal alliance in "the fight."

I like Rust in theory as much as the next guy, and the code is great, for one I think the kernel developers' concerns are slightly misplaced, but in practice I wouldn't want Rust people anywhere near my machines. Linux is a pretty old, and subpar operating system; it takes a great deal of effort to execute on it & only a handful of distributions (Debian) manage to get it right in the first place! Why don't they go for some other OS? There's Plan 9, there's Fuchsia, it's not like Linux is the only path forward. But they don't want to be writing file systems; instead, the only goal that matters to them is getting to the heart of the most popular OS, so as to secure some longterm success.

It all stems from insecurity, if you ask me.

Re:Rust for Filesystems by pjc50 on Monday July 15, 2024 @10:01AM (#40967982)
> But they don't want to be writing file systems; instead, the only goal that matters to them is getting to the heart of the most popular OS, so as to secure some longterm success.

Wrong final word: security. In the war against remote exploits, harden the most widely used target first.

Rust only exists because getting people to reliably and consistently write secure C++ proved impossible.

(yes I know C++ is a much larger and more complicated language than C)

Re:Rust for Filesystems by tucnak on Monday July 15, 2024 @10:23AM (#40968140)
IBM had full safety in terms of trusted program translation and memory tagging capability since the 90s, and yet the wider industry, Linux cheapskates couldn't care less about (1) ECC memory, (2) memory tagging. Now we have Arm MTE which is a step in the right direction, and CHERI capability-programming, which is really making progress in RISC-V community.

If the Linux community was serious about, and not just cheap + ignorant, we wouldn't be having this conversation in the first place. Pandering programming language-equivalent of condom ads to this very community is not a sign of smarts, or great aspiration, but cluelessness and naivety.

Re:Rust for Filesystems by pjc50 on Monday July 15, 2024 @10:34AM (#40968228)
Man, an omni-hater, not just hating people trying to bring Rust to Linux but also hating on Linux. Must be a very rewarding sense of superiority.

> ECC memory

This is entirely Intel's fault.