Bcachefs Fixes Pull Once Again Frustrates Linus Torvalds
phoronix.com
phoronix.com
If it's removed, I'm somewhat skeptical it can return once it's more mature. Kent seems to think that the problem is code maturity while Linus is making it clear that the problem is that Kent won't follow process and work with others. I don't think Kent's behavior will change as long as Kent is in charge of bcachefs no matter how mature the codebase gets.
(a) play better with others
(b) take your toy and go home
I've been following from the sidelines, keen to see how bcachefs pans out. When it got included in the mainline kernel I was hoping that was a sign it was soon time to give it a proper whirl.
Based on the recent kerfuffle though, it really seems like the better move is to just remove it from mainline until it's a lot more mature.
The Code or Kent?
I'd pick xfs anytime.
I don't know if it was already fixed, but ~1y ago had a really miserable experience using ZFS on my NVMe drives, zfs is like 100x slower in random read. And hey, I have fancy SSDs for a reason!