New Linux Kernel Drama: Torvalds Drops Bcachefs Support After Clash (itsfoss.com) 85
Bcachefs "pitches itself as a filesystem that 'doesn't eat your data'," writes the open source/Linux blog It's FOSS. Although it was last October that Bcachefs developer Kent Overstreet was restricted from participating in the Linux 6.13 kernel development cycle (after ending a mailing list post with "Get your head examined. And get the fuck out of here with this shit.")
And now with the upcoming Linux kernel 6.17 release, Linus Torvalds has decided to drop Bcachefs support, they report, "owing to growing tensions" with Overstreet: The decision follows a series of disagreements over how fixes and changes for it were submitted during the 6.16 release cycle... Kent filed a pull request to add a new feature called "journal-rewind". It was meant to improve bcachefs repair functionality, but it landed during the release candidate (RC) phase, a time usually reserved for bug fixes, not new features, as Linus pointed out. [Adding "I remain steadfastly convinced that anybody who uses bcachefs is expecting it to be experimental. They had better."]
Theodore Ts'o, a long-time kernel developer and maintainer of ext4, also chimed in, saying that Kent's approach risks introducing regressions, especially when changes affect sensitive parts of a filesystem like journaling. He reminded Kent that the rules around the merge window have been a long-standing consensus in the kernel community, and it's Linus's job to enforce them. After some more back and forth, Kent pushed back, arguing that the rules around the merge window aren't absolute and should allow for flexibility, even more so when user data is at stake. He then went ahead and resubmitted the patch, citing instances from XFS and Btrfs where similar fixes made it into the kernel during RCs. Linus merged it into his tree, but ultimately decided to drop Bcachefs entirely in the 6.17 merge window.
To which Kent responded by clarifying that he wasn't trying to shut Linus out of Bcachefs' decisions, stressing that he values Linus's input...
This of course follows the great Torvalds-Overstreet "filesystem people never learn" throwdown back in April.
And now with the upcoming Linux kernel 6.17 release, Linus Torvalds has decided to drop Bcachefs support, they report, "owing to growing tensions" with Overstreet: The decision follows a series of disagreements over how fixes and changes for it were submitted during the 6.16 release cycle... Kent filed a pull request to add a new feature called "journal-rewind". It was meant to improve bcachefs repair functionality, but it landed during the release candidate (RC) phase, a time usually reserved for bug fixes, not new features, as Linus pointed out. [Adding "I remain steadfastly convinced that anybody who uses bcachefs is expecting it to be experimental. They had better."]
Theodore Ts'o, a long-time kernel developer and maintainer of ext4, also chimed in, saying that Kent's approach risks introducing regressions, especially when changes affect sensitive parts of a filesystem like journaling. He reminded Kent that the rules around the merge window have been a long-standing consensus in the kernel community, and it's Linus's job to enforce them. After some more back and forth, Kent pushed back, arguing that the rules around the merge window aren't absolute and should allow for flexibility, even more so when user data is at stake. He then went ahead and resubmitted the patch, citing instances from XFS and Btrfs where similar fixes made it into the kernel during RCs. Linus merged it into his tree, but ultimately decided to drop Bcachefs entirely in the 6.17 merge window.
To which Kent responded by clarifying that he wasn't trying to shut Linus out of Bcachefs' decisions, stressing that he values Linus's input...
This of course follows the great Torvalds-Overstreet "filesystem people never learn" throwdown back in April.
Re: (Score:2)
Speaking of self-centered cunts, anyone have ANY fucking idea how Linux is going to survive without Linus? (Trolls, feel free to choke hard on the sheer amount of common fucking sense in that question.)
This scenario has already happened with vim
Re: (Score:2)
Re: (Score:2)
anyone have ANY fucking idea how Linux is going to survive without Linus?
This scenario has already happened with vim
Linux doesn't use vim any more?
This is about the passing of Bram Moolenaar, Benevolent dictator for life of Vim, the fact that Linus as well is a mortal, and that without someone as assertive as Linus it will be hard to keep the project steered.
Re: (Score:2)
That's a real problem Python has survived Guido's retirement quite nicely, though. But there was lots of pre-planning. I haven't attended to whether such pre-planning is happening in the kernel, but there were a few obvious candidates the last time I looked.
Don't know what the story is about and don't care (Score:2)
But why did you propagate the AC brain fart Subject?
Not feeling motivated to search for signs of intelligent life on today's Slashdot. But I'll give the Funny tab a click just to make sure it's as empty as usual. Probably could be a "feeping creaturitis" joke on the topic, but...
Re: (Score:2)
Linus build a good team of core maintainers. Linux ran fine when Linus was on his sabbatical.
Re: Kent by name, K*unt by nature (Score:2)
How will Apple continue without Steve Jobs?
Turns out, a lot of these complicated technical projects are a team effort. As there is some leadership, the project survives.
All the drama (Score:2, Insightful)
Re:All the drama (Score:5, Funny)
Is always with filesystem developers.
*reiserfs enters the chat*
Re: (Score:2)
Re: (Score:1)
It was almost comical watching people here defend that guy. Some choice arguments.
He's just weird and kind of autistic.
His mail order bride is missing and he seems pretty unconcerned.
The guy just happened to buy books about crime and forensics right after she went missing? Not suspicious to me!
Look who here hasn't removed the back seat from their car and then forgot where it went?
Re:All the drama (Score:5, Informative)
Re: (Score:2)
I don't see why not. No jokes about token ring here, just that many kinds of people murder people.
Re: (Score:2)
However, most driver programmers don't know what they are doing. That isn't going to change in the term. Rust allows them to write drivers without writing memory leaks or buffer overruns. Net win.
Re: nope, also rust people (Score:3, Insightful)
Whenever I throw this in a discussion, it usually goes in the "Good prog
Re: (Score:2)
The example that I remember was "if (a = 5) ". It compiles and runs. But it is a typo that is not flagged
It gets flagged as a warning by most compilers. In GCC, don't disable -Wno-parentheses
Re: nope, also rust people (Score:2)
Also, Yoda notation is a thing.
Re: nope, also rust people (Score:2)
Re: (Score:2)
The bad programmer apologizes and fixes it. The good one plays the blame game and even may throw in a tantrum.
Arrogant programmers throw a tantrum and blame. Skill is an orthogonal trait to arrogance.
Good programmers ask:
1) How did I make that mistake?
2) How can I avoid making the same mistake in the future?
3) Did I make a similar mistake anywhere else (go verify)?
If you do that, you can write secure code in C. If you don't, then you can't write secure code in any language (because memory errors are not the only errors).
Re: nope, also rust people (Score:2)
Re: (Score:2)
That is too binary in my opinion.
Which part of what I said is too binary?
Re: (Score:2)
I do not know when you studied. But C originally as designed to be a kind of portable macro assembler.
Basically if you want to bring that analogy/point to the extreme: it did not much more than abstracting away the names of the registers. Or invent on very small machines, like 6502, its on machine abstraction - which did not have much to do with the original processor architecture.
Basically old C compilers only looked at the syntax, if the { and } matched and there was no missing semicolon ... it just spit
Free Pascal (Score:2)
For all the attempts to correct the warts and remove limitations to make Pascal usable for large system development--Modula 2, Ada, Oberon, Component Pascal--it appears that Pascal World has coalesced around the Delphi 7 object Pascal dialect for which Free Pascal is the mullti-platform compiler?
Re: (Score:2)
Good programmers ask: [...] If you don't, then you can't write secure code in any language
All fun and games until you manage a large project and your team gets juniors apparently picked at random by HR (or the AI bot). It's not about how great a programmer you are, it's about how disastrous the new interns can be.
Re: (Score:2)
If they are interns, you have to teach them.
Wrong way around (Score:3)
"The bad programmer apologizes and fixes it. The good one plays the blame game and even may throw in a tantrum. "
ITYM the other way around because thats been my experience. Good programmers know their limits and will admit to cocking things up. Incompentents blame other people for their failings as in other areas of life.
Re: (Score:2)
Arrogant programmers get upset. Arrogance is orthogonal to skill (but eventually limits it).
Re: (Score:2)
Good programmers say, "How can I avoid making that mistake in the future? Did I make a similar mistake anywhere else?"
Indeed. You only throw out the tool if it is useless. Otherwise you _fix_ it. And you always aknowledge that _you_ screwed up or you will never advance your skills. Far too many people do not understand that one.
Arrogant programmers get upset. Arrogance is orthogonal to skill (but eventually limits it).
Arrogance reduces and limits learning. And we all need to learn in order to get better, no matter how intelligent. And thereby arrogant people without insight into what their actual skills are usually mess up badly at some point. If very arrgogant, they do it again and again and again. The story her
Re: (Score:2)
Indeed. But there are tons of bad programmers that mistakenly think they are good programmers. And that is the actual root of the problem: Big-ego-small-skills. As usual with the human race when things get screwed up. The Dunning-Kruger Effect really is the most important research result, ever.
Re: Wrong way around (Score:2)
Re: (Score:2)
Here is another one: gcc flags this as with a warning when using -Wall (which you should always do), even without -fanalyzer.
Ther is really not need to throw out the kid with the bathwater.
Re: (Score:2)
Everybody makes mistakes.
Most of the time he fixes them himself.
Most mistakes are what they are: a mistake.
In other words, they do not write such code because they are stupid, ignorant, uneducated or bad programmers, they write it: because they made a mistake. An oversight.
This particular mistake never happened to me (as far as I remember, but my memory could be faulty).
Then again, some people started to use "if (5 = a) " instead. Which does not compile. But in English and German this code sounds ugly, if y
if a assign 5 is non-zero then ... (Score:2)
I have warnings as errors turned off, so it does compile.
Re: (Score:2)
Well, ...
the code I push into Git works on my machine
Re: (Score:2)
However, most driver programmers don't know what they are doing. That isn't going to change in the term. Rust allows them to write drivers without writing memory leaks or buffer overruns. Net win.
Probably not. Probably just a waste of effort overall. Because most drivers have to use _unsafe_ Rust or they do not work. And _unsafe_ Rust is about the same as C in risks, minus the part of the coders that know what they are doing.
My take is Linus just allowed it to shut the Rust fanatics up.
Re: (Score:2)
What Rust lets you do is confine the "unsafe" parts so that you do most of the logic with full safety on and then pass the results to a much smaller "unsafe" part that interacts directly with the hardware.
Re: (Score:2)
That requires skill and insight. And hence all advantages are lost.
Re: (Score:2)
You have to use unsafe in 5 lines of code.
And the other 10,000 lines do not need it.
What is your complaint with that?
Re: (Score:2)
Not always, but they tend to me more ... project-centered ... than most.
How to Win Friends and Influence People (Score:5, Informative)
"I positively enjoy working with you - when you're not being a dick, but you can be genuinely impossible sometimes."
- Overstreet replying to Linus threatening not to merge any of his future code.
If that's what young people call an apology, no wonder Torvalds is sick of dealing with him. Yet the media will twist this as 'evil Linus'.
Re:How to Win Friends and Influence People (Score:5, Interesting)
All Overstreet needs to do is figure out how a merge window works. He doesn't need to write a perfect apology.
Re: (Score:3, Insightful)
All Overstreet needs to do is...
All Overstreet needs to is continue development as he wishes. There is no fundamental reason bcachefs must be included in Linus's mainline kernel. The kernel has loadable modules. This work can simply be a loadable module. There are tools to make this next to transparent to an end user, up to and including as a root file system. ZFS On Linux has existed this way for 15 years now. There are entire commercial empires built around it, and it has never, at any point, been in the Linux mainline code base.
Re:How to Win Friends and Influence People (Score:5, Funny)
2025 sure is weird:
- Linux is the most popular OS kernel in the world.
- Windows supports running Linux natively on the OS.
- Linus is the most level headed voice in the room.
Somehow the last one was the least likely to be on my bingo card.
Re: (Score:2)
Linus was always level-headed. He just prioritizes clarity over politeness and I fail to see how that is bad. Many people think it is though, but these people never made any great piece of technology a reality.
The one thing Linus should have probably learned is to be "icily polite" and learn how to tell people they are complete and utter wastes of oxygen without ever using any impolite expressions. Well, maybe too many people are to thick to get it if that mode is used, so maybe it would not have done much.
Re: (Score:2)
Linus is Finnish, but Linux is an international effort. Perhaps this international audience made Linus evolve his ways of expression.
Mutually Assured Insincerity (Score:2)
How being diplomatic works.
Re: (Score:2)
Okay, so the discussion did produce a good joke. Not the feeping creaturitis I was looking for, but still...
Messiah Complex (Score:2)
Sounds like this Overstreet guy suffers from a bad case of it. But then its fairly common amongst the supposed "L337" in tech IME. Yes they're smart , but they think they're a genius and can't deal very well with people which may or may not be aspergers related though plenty of people who are bad with others are neurotypical but just utter dicks.
Re:Messiah Complex (Score:5, Informative)
What Overstreet misses is that protecting releases is criticlly important for something like the Linux kernel. Makes him not smart in general: He sees his thing, but he does not understand how it fits in the genral scheme of things. And then he fails again by being aggressive instead of learning something and that is a sign of immaturity.
Re: (Score:2)
It's Overstreet who seems to behave like someone with Asperger's. Aspies can be very smart and very good at what they do, which can make them very useful, but often their social skills are like these of a spoiled 5 years old brat having a tantrum.
It's too bad it had come to this... But perhaps Linus had to choose between two evils and hopefully he chose the lesser one.
Re: (Score:2)
Ah, and before someone accuses me of being unfair and bigoted to Aspies, I'm an Aspie myself, and my own social skills are nothing to write home about... But I at least am aware of my limitations and make conscious efforts to mitigate them.
Perhaps it comes with time and experience, for I'm not young anymore.
Re: (Score:2)
Indeed. And protecting the release and minimuzing risks of post-release problems is very much not "beign a dick". It is called "being competent".
Oversteet is a bully with an outsized ego, compared to his skills. And while his skills seem to be narrow, but pretty good in that narrow area, his ego negates their worth. Obviously, he lacks the insight to understand that and that is a problem of personal maturity. Some people are so in awe of what they can do, that they stop learning and end up lacking criticall
Great loss; could have been worse (Score:2)
This is a great loss for all open source filesystem users. What could Bcachefs have become? However what would have been a truly great loss is if Overstreet had learned to work well with others and then we lost it. Bcachefs was not yet really on the path to future adoption because it had neither broad developer support nor corporate backing. It was one guy's pet project, no matter how technologically promising it was.
Re:Great loss; could have been worse (Score:5, Interesting)
Out of curiosity, what was the technological promising thing about it?
I only see the constant last-minute fixes it seems to need and that user space tooling is dropped from Debian for being unmaintainable, which tells me that - regardless what technological wonder it may be - it is certainly not a filesystem I would use.
Re:Great loss; could have been worse (Score:5, Informative)
Storage tiering, compression tiering, native encryption, caching strategies, erasure coding (with no write hole) https://derg.nz/kb/howto/2023/... [derg.nz] Also bcachefs is faster than btrfs https://www.phoronix.com/revie... [phoronix.com]
Re: (Score:3)
I only see the constant last-minute fixes it seems to need and that user space tooling is dropped from Debian for being unmaintainable, which tells me that - regardless what technological wonder it may be - it is certainly not a filesystem I would use.
Blog post from the debian maintainer https://jonathancarter.org/202... [jonathancarter.org] debian bug https://bugs.debian.org/cgi-bi... [debian.org] I can only concur with your conclusion.
Re: (Score:2)
Indeed. "High performance" does lose to "unmaintainable" every time when you need to depend on something.
Re: Great loss; could have been worse (Score:2)
Yes a very dark day for Linux. It so badly needs a NG filesystem.
Re: (Score:1)
Re: (Score:2)
This is beneath you. Shame!
Re: (Score:2)
Linus is the project head in the Linux project. An open source project of this magnitude can be successful only if a set of rules are agreed upon, that would govern the project. We have enough examples of other projects that exploded in disarray because everyone did as he pleased. Linus already proved that he can play by rules when he toned down his speech due to others' complaints of his rudeness, and this is commendable.
Who else to enforce the rules than the project lead or people appointed by them?
Re: (Score:2)
Putting aside the rest of your comment here,
I hope Linux actually has a succession plan for when The Almighty Linus dies. Somehow I donâ(TM)t think canonization is gonna help the merge schedule.
There is such a plan. Whether it can succeed is another question. This is a legitimate and worrying concern. It's a shame you chose to pose it as you did. I think this should be thought about more anyway, though.
This quote from Overstreet makes me glad he's out (Score:4, Insightful)
Overstreet: Yeah, Al just pointed me at generic_set_sb_d_ops().
I don't want AI slop in the kernel I rely on for work. Fuck that guy.
Re: (Score:3)
AI isn't the problem, it's that search-engine'ing his way through the dark
Re:This quote from Overstreet makes me glad he's o (Score:5, Informative)
Move stuff out of the kernel (Score:2)
Re: (Score:3)
Kent Overstreet about the recovery mode that triggered the feud: " we can run it in userspace today. The thing is, with it in the kernel side codebase, we can test it with an -o nochanges mount, and that way the user can verify with their own eyes that the filesystem looks the way it should and their data is there. For this particular recovery mode, that's essential." https://lwn.net/Articles/10272... [lwn.net] Also "A fuse implementation will never be performant enough to replace the kernel implementation, "
Re: (Score:2)
Linux is monolithic. Out-of-Tree Patches work, but they require constant work. For reliable out-of-tree features you want to use a micro kernel. Maybe ask Tanenbaum about it instead of Torvalds.
What drama? (Score:5, Interesting)
Is that the deragned press reporting I hear about?
What happens is that a non-collaborator tried to work with a collaborative project and could not adjust. Hence he got removed. As it should be.
This person endangered the release process. That is basically a hanging offense. He also seems to thick to understand how competent software release works. He could simply have appologized, retracted the change, make it part of the next kernel release and promised to do better. Instead he defended his destructive and risky behavior. That is not acceptable. Such behavior creates technological debt and problematic or failed releases.
On the plus side, it is reassuring to see that the Linux kernel team refuses to accept shoddy practices and to descent on an MS-like level of slapdash and incompetence.
Open source software and closed minds (Score:1)
Team: America World Police (Score:3)
Linus is a "dick" (according to Kent). Bugs are like pussies. And Kent is an asshole. Pussies don't like dicks, because pussies get fucked by dicks. But dicks also fuck assholes. Assholes think they can deal with pussies their way, because because pussies are only an inch and a half away from assholes. The problem with assholes is that sometimes they shit too much or shit when it isn't appropriate — and it takes a dick to show them that. I don't know much in this crazy, crazy world, but I do know that if you don't let this dick fuck this asshole, we're going to have our dicks and pussies all covered in shit!
The social awkwardness with CS types is very real (Score:2)
The number of CS students with seriously problematic behavior/socialization is jaw-dropping. Every semester, they deal with many cases of CS students who exhibit basically every problematic social behavior you could list. The instructors, professors, staff, and administrators are constantly dealing with it. It's not a majority, but the numbers are surpr
The confusion didn't help. (Score:2)
That there was a file system. AND A caching system with more or less the same name. Bcache and Bcachefs
The former being a way to semi speed up raid array's with a cache ssd.