Linux 7.0 Released (linuxiac.com) 28
"The new Linux kernel was released and it's kind of a big deal," writes longtime Slashdot reader rexx mainframe. "Here is what you can expect." Linuxiac reports: A key update in Linux 7.0 is the removal of the experimental label from Rust support. That (of course) does not make Rust a dominant language in kernel development, but it is still an important step in its gradual integration into the project. Another notable security-related change is the addition of ML-DSA post-quantum signatures for kernel module authentication, while support for SHA-1-based module-signing schemes has been removed.
The kernel now includes BPF-based filtering for io_uring operations, providing administrators with improved control in restricted environments. Additionally, BTF type lookups are now faster due to binary search. At the same time, this release continues ongoing cleanup in the kernel's lower layers. The removal of linuxrc initrd code advances the transition to initramfs as the sole early-userspace boot mechanism.
Linux 7.0 also introduces NULLFS, an immutable and empty root filesystem designed for systems that mount the real root later. Plus, preemption handling is now simpler on most architectures, with further improvements to restartable sequences, workqueues, RCU internals, slab allocation, and type-based hardening. Filesystems and storage receive several updates as well. Non-blocking timestamp updates now function correctly, and filesystems must explicitly opt in to leases rather than receiving them by default. Phoronix has compiled a list of the many exciting changes.
Linus Torvalds himself announced the release, which can be downloaded directly from his git tree or from the kernel.org website.
Linux 7.0 has a major new version number but it's "largely a numbering reset [...], not a sign of some unusually disruptive release," notes Linuxiac.
The kernel now includes BPF-based filtering for io_uring operations, providing administrators with improved control in restricted environments. Additionally, BTF type lookups are now faster due to binary search. At the same time, this release continues ongoing cleanup in the kernel's lower layers. The removal of linuxrc initrd code advances the transition to initramfs as the sole early-userspace boot mechanism.
Linux 7.0 also introduces NULLFS, an immutable and empty root filesystem designed for systems that mount the real root later. Plus, preemption handling is now simpler on most architectures, with further improvements to restartable sequences, workqueues, RCU internals, slab allocation, and type-based hardening. Filesystems and storage receive several updates as well. Non-blocking timestamp updates now function correctly, and filesystems must explicitly opt in to leases rather than receiving them by default. Phoronix has compiled a list of the many exciting changes.
Linus Torvalds himself announced the release, which can be downloaded directly from his git tree or from the kernel.org website.
Linux 7.0 has a major new version number but it's "largely a numbering reset [...], not a sign of some unusually disruptive release," notes Linuxiac.
A number reset (Score:2)
Yes, number reset now a major feature!!
Re: (Score:2)
Yes, number reset now a major feature!!
A sign of OS maturity. :-)
Re: (Score:3)
Here you go, enjoy reading
https://linux.slashdot.org/sto... [slashdot.org]
https://news.slashdot.org/stor... [slashdot.org]
https://tech.slashdot.org/stor... [slashdot.org]
https://linux.slashdot.org/sto... [slashdot.org]
Re: (Score:2, Insightful)
Does that mean we'll get a new Slashdot announcement for future Linux releases
Hopefully. Thats the kind of stuff this particular site was doing before all the virtue-soys showed up and made it a climate news and EV fanboi site. Hell, minor releases of Enlightenment used to be news, not that you'd know what that is.
Re: (Score:2, Insightful)
Hopefully. Thats the kind of stuff this particular site was doing before all the virtue-soys showed up and made it a climate news and EV fanboi site.
You do know that nobody but the editors decide which content you're going to see, right? It's called maximizing engagement and they give absolutely zero fucks about anything else.
Re: (Score:2)
It's called maximizing engagement
Is that the motivating prerogative? If so then why is it a minor eddy in the torrent of tech news that people are astonished to learn is still up?
Re: (Score:2)
Re: (Score:2)
"Slashdot. News for Nerds. Stuff that Matters"
Re: (Score:2)
God, Enlightenment. That's a name I haven't heard, in a long, long time.
Probably not something you should upgrade to yet (Score:5, Informative)
If you or some dependency of something you run uses PostgreSQL, be aware that Linux 7.0 has changes that causes a 50% performance hit on the former. The Linux people are adamant that the PGSQL people should change their code, despite the fact it's not due to a bug or anything similar.
Until you can migrate to a newer PGSQL with the changes that the Linux people are demanding, with time taken to test and make sure these work (it's not a trivial fix, the PGSQL people literally have to rewrite a critical part of the code), you should probably pin an earlier kernel, or use one patched to support PREEMPT_NONE.
Here's a non-AI article that explains the issue: https://www.phoronix.com/news/... [phoronix.com]
If I were a distro maker, I'd not touch Linux 7.x until the PostgreSQL people have had a chance to release changes and the code is mature enough to use, though alas that could be years given bugs and security issues with anything the PGSQL people do could take years to surface.
Re: (Score:1)
If people that are concerned with database performance are doing day zero upgrades of a X.0 kernel release, they're idiots and they deserve what they get.
Anyone that has been around the block a few times knows you wait and let the idiots shake out all the problems like this, and you upgrade on the X.1 that fixes all the problems they could have never seen coming due to mountains of edge cases that cannot effectively be tested for.
Re:Probably not something you should upgrade to ye (Score:5, Insightful)
upgrades of a X.0 kernel release
Except the location of the Xes is arbitrary. I was modded down by the people who didn't like living in a world where we had kernels with numbers 2.4.27 (disappointingly a couple with low UIDs as well) where major, minor, and points releases meant something, but the difference between kernel 7.0 and kernel 6.19 is that Linus thought it was a good day to change a number.
And before anyone mods me troll, to quote the man himself https://lkml.org/lkml/2026/2/2... [lkml.org]:
We have a new major number purely because I'm easily confused and not
good with big numbers.
Re: (Score:2)
> that causes a 50% performance hit
That's approaching "breaking userspace".
I'll read more about this but unless ancient pgsql code was doing something stupid it sounds like a third option needs to be considered.
Re: (Score:2)
There are really enough long-term support kernels to make this not an issue.
Re: (Score:2)
Largely nonsense.
They are defaulting to PREEMPT_NONE but nothing to stop you changing that.
Newer versions of PostgreSQL already avoid certain spinlocks that hit this performance issue which is, as stated, doing something dumb in PostgreSQL which kernels can't have insight into, and which the new kernel defaults now penalise.
And nobody, with a brain, would just be running anything critical on the very latest kernel without testing it first.
By the time 7.0.1 comes out, PostgreSQL will have sorted it and nobod
Can be avoided with config (Score:2)
The problem doesn't occur if you have huge pages enabled, which is a good idea for a database machine anyway, as running without huge pages has almost as much of an impact on Postgres performance as this regression does. So no need to way for postfix to ship the spinlock bug fix.
Why NULLFS: (Score:5, Informative)
I was curious so I looked up the details about NULLFS.
Apparently, there is an issue with swapping the root filesystem which is done using the syscall pivot_root()... but not with initramfs,
per the man page... [man7.org]
The rootfs (initial ramfs) cannot be pivot_root()ed. The recommended method of changing the root filesystem in this case is to delete everything in rootfs, overmount rootfs with the new root, attach stdin/stdout/stderr to the new /dev/console, and exec the new init(1). Helper programs for this process exist; see switch_root(8).
So basically, this fixes a long-standing hack that well... is not safe in some cases, most notably with with containers (CVE-2020-15257). The proper solution was to make a simple null filesystem that could use pivot_root and swap out the rootfs without hacks.
More details here: https://lwn.net/Articles/10621... [lwn.net]
And here: https://www.linkedin.com/pulse... [linkedin.com]
So will it be another decade... (Score:2)
...before Linux 7 takes over Linux XP in number of users?
kind of a big deal (Score:2)
Y'know, the guy who's name makes the Li in Linux
I guess that quote is included, but it kinda defeats the whole article.
Weird and stupid clickbait but it's nice to see people excited about Linux.
Re: (Score:2)
"kind of a big deal" the guy specifically said that it isn't. Y'know, the guy who's name makes the Li in Linux I guess that quote is included, but it kinda defeats the whole article. Weird and stupid clickbait but it's nice to see people excited about Linux.
Increasing UDP throughput simply by inlining a function is a big deal!
Corporate Release Cycle Linux (Score:1)