Linus today on LKML[1]: """Apparently nobody actually runs #linux-next. I knew it didn't get a lot of testing, but apparently it's more like "no testing at all" than "not a lot"."""

Wondering if the problem he and Alexei ran into only triggers in some environments – then the situation might not be as bad as Linus makes it sound.

But at the same time he has a point. One of the reasons why I stay away from using -next regularly: it includes the mm-unstable/mm-nonmm-unstable branches. I often wonder if that is actually violating the rules for -next inclusion, but I'm not sure[2].

[1] https://lore.kernel.org/all/CAHk-=wipM[email protected]/

[2] can't remember where they are written down, but I think it was round about "only push code to -next that you'd send to Linus for mainline inclusion tomorrow it he'd suddenly open the merge window" -- but I might be mistaken…

#kernel #LinuxKernel