Recent searches
Search options
The next instalment of Linus vs. Kent – and Linus seems to be on the edge:
"[…] I'm contemplating just removing #bcachefs entirely from the [#Linux #kernel] mainline tree. Because you show again and again that you have no interest in trying to make mainline work. […]"
For the full discussion, start here: https://lore.kernel.org/all/cphtxla2se4gavql3re5xju7mqxld4rp6q4wbqephb6by5ibfa@5myddcaxerpb/t/#m631c24cd07f5820a4cbff8f25dff1d1a0c3cf2e7 (the quote is from one of the later mails from Linus). #LinuxKernel
@kernellogger I really hope that Linus drops bcachefs. The FS and its concept are interesting but not worth dealing with the negativity of Kent.
When rightfully criticizing his behaviour, the first thing he does is shitting on btrfs (and its devs), even if the discussion is not technical.
yup; most people can live with a sneaky remark now and then, but it's way more then that what others have to ensure from Kent…
He also snuck in an mm change that got NACKed through his fucking tree (!) then got abusive when it was reverted.
I genuinely don't understand how on any level this is being tolerated. It sets such a bad example for shit people in the kernel 'hey I'm not as bad as Kent and he gets away with it' is a thing, and I don't like it.
mistakes happen, are pointed out, and then people should learn from it; if it turns out that's not the case, something has to be and hopefully will be done, but that take a day or two.
I think we already are in the "takes a day or two" stage here (I'm ignoring here that this was someone likely to happen given the history, as that is a different story…).
hail himThat wasn't a mistake it was an intentional attempt to get an mm change in around mm developers, and it was fucking outrageous.
He did the same with fs too... :) the mm one was far, far worse though as in actually a really bad change. AND he got a CoC violation on the thread for reverting it!
The guy to me is just a bully.
@brot @kernellogger Yeah, that shitting on btrfs is just pure toxicity. Every single time. (He should ask Meta how "broken" it is...)
I'll never go back to more "basic" filesystems, so of course I'm interested in bcachefs. But its maintainer seems insufferable.
@hambier
His comms reminds me very much of the way lkml was 20 years ago, which I thought we'd pretty much all agreed to try to move away from :(
btrfs definitely has issues in both design and implementation, but those are well known and don't need to be rubbed in every five minutes. bcachefs does seem a lot cleaner and much more promising, and I genuinely think Kent has done something amazing with few resources .. but by getting merged he's "won" - time to chill out a bit!
@brot @kernellogger
@brot
Honestly from what I'd seen I was on Kent's side until fairly recently - there were certain people on lkml who would just automatically NAK any suggestion of his without engaging constructively and I'm not surprised that tipped him over the edge occasionally.
Now bcachefs has been merged though his attitude seems to have changed to "I can do what I want because my process is better" and that's... really not going to work :(
@kernellogger
Dealing with Kent afaics for a long time already was, ehh, was not easy for many people -- and the reason why some actors were not that much willing to engage constructively.
Which is…
> automatically NAK any suggestion of his without engaging constructively
…why it's so dangerous to take sides in complicated topics/discussions, as you never known if there is a non-obvious backstory that explains some behaviours.
@vbabka @kernellogger @brot ah, I don't do detail these days, so my impressions could very much be wrong. I was almost certainly thinking back well over a year ago anyway - so I likely don't mean any recent proposals.
I dip in and out of lkml periodically but have been doing so for years, and I've been trying to keep tabs on the bcachefs stuff because I have personal reasons for wanting a 'better btrfs' (mainly for container pools etc.)
@kernellogger @brot true, but at in theory most of the debate should be taking place on the public mailing lists, so the discussion is visible. I guess we don't know how much backchannel stuff happens as well though.
most of the debate *did* take place on the public mailing lists – but who in these hectic times is willing to wade through earlier discussions from the past 10+ years that lead to stances like "I'm a friendly and helpful person normally, but working with that particular person is really not worth spending my time, so I will keep my interactions and helpfulness from now on to absolut minimum required"