Torvalds Gets Tough on Kernel Contributors

ChocLinux writes "Linus Torvalds is cracking down on developers that add last-minute changes to the kernel during the two-week merge window. He says: 'If people miss the merge window or start abusing it with hurried last-minute things that just cause problems for -rc1, I'll just refuse to merge, and laugh in their faces derisively when they whine plaintively at me, and tell them there's going to be a new opening soon enough.'"

Comments

Good

[bblazer]

While I think that his language is a bit inflammatory, I think that it is something that needs to be done. All the last minute changes must make things very difficult to manage.

Re:Good

[slavemowgli]

It sure makes you wonder what the two-week window is actually good for, though. I mean... the whole thing was done in order to make sure that there'd be a time for submitting new stuff, and a time for shaking out bugs, and so that people would be able to tell the two apart.

So, why the fuss about last-minute merges now? If they're still in the two-week window, they should be fine; if they're not, well, then they're too late and won't get merged, but that was already clear anyway. And stuff that's not up to par quality-wise yet will (should) not be merged at all, anyway - it's not as if the code quality requirements were lowered for the two-week merge window.

What Linus seems to be doing is to effectively reduce the two-week window to a "something-less-than-two-weeks" window where noone knows exactly how much the difference is, but it does not get rid of the underlying problem: there still is a deadline, and people will still submit lots of stuff just before the deadline's there. It doesn't matter whether it's two weeks or 13 days or whatever.

The whole *point* of the "merge for two weeks, then stop merging and focus on bug fixes" was to be able live with this, so to speak. If you can't fight them, make them join you; if you can't prevent people from submitting stuff in the last minute, make sure that there's enough time *after* the last minute so that last-minute merges won't hurt you. If Linus finds it necessary to crack down on last-minute merges now - which, as I said above, is not really possible in practice (the only way to do it would be to not merge anything at all anymore, but that's obviously not a practical solution) - because there are too many, that just shows that patch pressure is too high already; further increasing it won't help. Rather, you have to look at *why* patch pressure is so high, and do something about that. For example, why not extend the two-week window to three or four weeks? It might mean that new kernel versions appear less often, but in these days of git and distributed development where tree changes are so easy to push/pull and where every distributor uses their own, heavily-patched version of the kernel, anyway, why does it matter so much? Linus has always taken a stance that quality is more important than meeting arbitrary deadlines, I think.

Or maybe I'm misunderstanding what he actually wanted to say - I did RTFA, but zdnet is not exactly what I'd call a high-quality source for kernel development news. Caveat lector...