Linux 7.3

SMB improvements (https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/lin...), I've moved to borg, but depending on how this goes may give time machine another try:

   - Add opt-in Apple Time Machine support

     Implement the AAPL negotiation and related Finder, stream,
     COPYCHUNK, sparse-file, CHANGE_NOTIFY, and RPC compatibility
     required for Time Machine shares, allowing macOS backupd to use
     ksmbd for backups

Still sucks that it has to be reverse-engineered

> This release is a record-breaker on all fronts; it brought in 18,765 non-merge commits from 3,021 developers, 807 of whom were first-time kernel developers.

I'm reading that as many new AI assisted code patches. Not everyone does an obviously bad job at submitting AI assisted code.

Not a criticism, just an observation.

I'm excited because of https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/lin..., which is the first kernel driver to use the new Wi-Fi HaLow (s1g/900MHz Wi-Fi) support.

(Disclosure: I work for Morse)

>ARM

> Device Tree Updates

> Apple platform gains support for M3 Pro, Max and Ultra SoC

Incredible how fast support for Apple silicon is progressing.

> a long list of SMB server improvements

This is how I learned there's a mainline kernel-mode SMB server!

https://docs.kernel.org/filesystems/smb/ksmbd.html

Number of CVEs fixed at an all-time high:

https://pages.skybrian.com/2026/kernel-cve-fixes/

> This release is a record-breaker on all fronts; it brought in 18,765 non-merge commits from 3,021 developers, 807 of whom were first-time kernel developers.

hopefully without kernel-slops :P good job everyone btw!

It’d be interesting to see if rust pickup is improving or not.

First time hearing about failfs. Does it have a niche use case or is it just for testing?

Development statistics for 7.3 dropped: https://news.ycombinator.com/item?id=50155075

Genuine question(s):

1) why does it matter “my” patch was submitted? I am simply a bystander but can’t help notice a tone of being admitted to a tribe? Am I wrong ? (It’s ok if it’s a tribe just validating my observation is correct or incorrect)

2) is the Linux kernel on its way to obsolescence? Our field of software engineering and how and what it’s applied to is changing at a drastic rate. The os was built for programs to run which were built by programmers. The programmers thought about what the users of their programs needed to get stuff done and use the programs. This paradigm is changing - fast - an in 10 years it’s going to be very diff place where the users simply build programs or their agents etc. thier expectations of the OS and what “services” and interfaces the os exposes is going to change. And at some point the current os architecture is just not relevant. To build an AI first OS it’s prolly easier to build it from the ground up than keep changing Linux. It’s like we can keep trying to slap on jets to the car to make it go faster but it’s prolly just better to build a plane. We still will need to travel just what we use is going to be diff and as such older mechanisms of traveling will just be irrelevant. Why am I wrong ? What am I missing ?

It looks like this release will likely unlock the ability to develop relocatable Nix packages on Linux, thanks to some great work by Farid Zakaria and Christian Brauner! I'm hopeful that it also eventually improves the story for mixing Nix and non-Nix software.

Linux has an in-kernel SMB server like Solaris? When the hell did this happen?

Does anyone know if there has been any guidance or discussion on when a 7.x release might become a longterm supported release?

Is anyone already running this seeing /dev/input/events appear as usual? Last rc didn't work for me, events was gone :( same config 7.2.10 was fine

Can someone explain to me what failfs is relevent for?

Congrats to the kernel team! The pace of releases over the last year has been impressive. Looking forward to testing this on my machines this weekend.

I wonder how many people nowadays see something like this, clicking on it and expecting a "Download and Install" button.

How big change is this?

more rust?

nice

[flagged]

Beware - this was Torvalds' fever dream of suddenly discovering vibe coding and unleashing it on the kernel like a five year old with candy.

Expect catastrophic problems. There is likely a lot of slop in the code now + testers already found race conditions.

Torvalds' behavior in the past few weeks has been alarming.