Linux Kernel 4.1 Will Be an LTS Release 46
New submitter prisoninmate writes: The Linux Foundation's LinuxLTSI (Long-Term Support Initiative) group has confirmed on Twitter that the next LTS version of the Linux kernel will be 4.1. The information has also been confirmed by Greg Kroah-Hartman, a renowned kernel developer who is currently maintaining several kernel branches, including a few LTS ones. When Linux kernel 4.1 is released, it will become the LTS version of 2015 and the most advanced long-term support release. This is significant because the LTSI releases are (or will be) everywhere, in a "Linux is everywhere" sense. As the initiative's page puts it, "The LTSI tree is expected to be a usable base for the majority of embedded systems, as well as the base for ecosystem players (e.g., semiconductor vendors, set-vendors, software component vendors, distributors, and system/application framework providers). ... The goal is to reduce the number of private trees currently in use in the CE industry and encourage more collaboration and sharing of development resources."
The CE Industry? (Score:5, Informative)
Jargon a little harder, asswipes! That's Consumer Electronics for those of us not in manufacturing.
Re:The CE Industry? (Score:5, Funny)
ITWAE (In Technolgy We Acronym Everything)
Re: (Score:2)
Re: (Score:1)
No. They were referring to Windows CE, normally shortened to "wince". Which also accurately describes how it makes you feel when you use it.
Here's looking at you, Android (Score:3)
"Reduce the number of private trees" --> Yeah, like the ancient (by mainline standards) kernels in most releases of Android... The sooner GOOG learns how to play nice with the rest of the Linux developers and get their customizations contributed upstream, the better off we'll all be. Though, admittedly, AOSP is doing a pretty decent job of that nowadays. The more egregious sinners are the device manufacturers.
Re: (Score:1)
That quote immediately made me think of this, completely non-obligatory XKCD [xkcd.com].
Re: (Score:1)
Randall Munroe is a humorless twat.
Skynet (Score:5, Funny)
Guess it means this picture makes sense and we are all doomed soon.
https://lh5.googleusercontent.... [googleusercontent.com]
Re: (Score:1)
And in the latest instalment, a Terminator comes back and persuades NORAD to upgrade it's systems to Windows
Er... how old is this message? (Score:2, Informative)
"When Linux kernel 4.1 is released, it will become the LTS version of 2015 and the most advanced long-term support release."
But... it HAS been released. I'm using it. Right now. As I type this.
Re:Er... how old is this message? (Score:4, Informative)
Not quite yet - still on 4.1-rc6, final version not yet released.
(At least as far as kernel.org [kernel.org] is concerned).
Re: (Score:2)
Not quite yet - still on 4.1-rc6, final version not yet released.
(At least as far as kernel.org [kernel.org] is concerned).
Correct!
Release number (Score:3)
You know what would have been a good candidate for LTS? 4!
Not necessarily the release itself but the number. The numbers are arbitrary so just make them the lts releases. Rather than 4.1 or 4.13 or 5.9..
Re: (Score:2)
No. You don't LTS the new hotness. Add features, patch them, then support that.
Win 95 rev B, Win 98 SE, latest service pack.. the only difference here is an honest version bump.
Re: (Score:2)
There's no "new hotness" about 4.0. Linux moved a long time ago away from the model of using a major version number change to indicate large changes or major new features.
Re: (Score:2)
It should move back.
Re: (Score:1)
Well, that's a very well-reasoned argument that I'm sure will persuade Linus to change his mind.
Re: (Score:2)
Perhaps he should seek Kay Sievers' input.
Re: (Score:1)
You know what would have been a good candidate for LTS? 4!
Factorial four? But that's 24, and I don't think they'd want to skip all of the intervening versions.
Re: (Score:2)
Making Terminator all the more likely :-) (Score:1)
2-3 Years is NOT Long-Term (Score:5, Interesting)
When will the industry realize that for an embedded system long-term is not 2 years but more like 30 years? When these systems are built into buildings, cars, aircraft, etc. they are expected to last for a long time.
Re: (Score:3)
I don't think "buildings, cars, aircraft, etc" would ever put a vanilla version of linux in their systems and expect a third party to 100% maintain them. That's just bad business. Give and take.
Re: (Score:2)
The problem comes when features start getting added to those systems that expose them to the outside world. For example someone wants to monitor and control their building remotely (or just to control it from their smartphone when they are in the building). So you start to get interconnections between the building management network and the internet and/or office wifi.
Re: (Score:2)
The kernel team will take an LTS release 2-3 years out, but most enterprise-class distros will provide least 5 years of support. RHEL 5 was released in March 2007 and with Extended Lifecycle Support it'll last to November 2020 so that's 13.5 years. And at that point I'd just sign up for the CVE list and see if there's any critical and relevant bugs for the most recently supported kernel that needs backporting, my guess is that it'd be once a leap year from there on out. It seems rather unlikely that hackers
Re: (Score:2)
For CEs that belong in a landfill nearly the day they are built, 2 years is an excellent plan. We have some of the constraints that you find in the rest of embedded, but we tend to be in a constant race to hit certain market dates with various features and price points. And we depend on your average user to be willing to discard their old widget and buy a new widget. (and hopefully that's your widget and not your competitor's)
Old News (Score:1)