Torvalds confirms there will be no Linux kernel 3.0

newsguy 0 Tallied Votes 680 Views Share

I stumbled across an interesting interview with Linus Torvalds over at apcmag.com in which the man himself would appear to be confirming that there will be no Linux kernel 3.0

The first hints are given when he responds to a question concerning why the 2.6 kernel has been around for so long, and explains his reasoning as to why a multi-year development cycle doesn't work concluding that the 2.6 base kernel is in such good shape that there is no pressing reason to go back to the old 'everything changes' development model. Torvalds adds "that means that we'll keep with the 2.6.x codebase, and just incrementally improve on it."

When questioned further, specifically about when we can expect to see the version 3.0 kernel, Torvalds states quite categorically that "we really don't expect to need to go to a 3.0.x version at all." This isn't the first time that Torvalds has questioned the need for a version 3.0, an email from September 2002 published at Kerneltrap says pretty much the same thing. In a reply to Jeff Garzik questioning him about a confirmation of the 2.6 numbering, Torvalds responds "I see no real reason to call it 3.0" and adds "but hey, it's just a number." But it would seem to be a final nail in the coffin of the kernel 3.0 dream, at least for the time being. After all, point increments can only go on for so long and when we hit 2.9. there may well be a renewed call for a rethink…

Dani AI

Generated

rightly picked up Linus' reluctance to bump the major number back in the 2000s, and 's version-number jokes were spot on for the era. That reluctance, however, changed: Linus officially moved the kernel from 2.6.39 to 3.0 in 2011 — but it was deliberately a numbering change, not a sweeping rewrite. (lkml.org)

Why the bump? Linus wanted to stop the “bignum” minor-version problem and thought a clean 3.x series made sense as the project entered its third decade; he emphasized there were no special incompatibilities tied to the new major number. In short: 3.0 was symbolic (and pragmatic about versioning), not a compatibility reset. (linuxfoundation.org)

What actually arrived in 3.0: modest but useful changes — btrfs gained scrubbing and auto-defragmentation, there were several new drivers and architecture tweaks, and features such as BPF JIT and some new syscalls/optimizations landed. If the thread feels like it predicted both the “no 3.0” stance and the eventual bump, that’s because Torvalds’ public position evolved from “no need” to “let’s renumber for sanity and anniversary.” (btrfs.readthedocs.io)

Practical note for anyone reading this later and wondering what it meant for users or distributions: 3.0 did not introduce wholesale breakage — treat it as a normal kernel release. To see the kernel you’re running, use:

uname -r

For kernel development context, the 3.0 release opened the usual merge window for 3.1 (stable fixes continued to use the third digit). That explains the mix of jokes, surprises and follow-ups from , and in this thread. (lkml.org)

jwenting 1,905 duckman Team Colleague

well, after 2.9 there's always 2.10 and you can take that a LOOOOOONG way, especially with the Linux tendency to use ridiculously long strings of subversions.

2.100000999.432325.432321545.35beta143523alpha542323.5234323 anyone?

happygeek 2,411 Most Valuable Poster Team Colleague Featured Poster

>2.100000999.432325.432321545.35beta143523alpha542323.5234323

ROFL!

Billll 0 Newbie Poster

3.0 is here today... it looks like Linus wants to sell something!

Tellalca 19 Posting Whiz in Training

So what's that?

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.