Linux jumps to 7.1 — and starts deleting code just to dodge AI bug reports

A major version number that means almost nothing

Linux 7.1 landed and the first thing the thread did was deflate the headline. The jump from 6 to 7 carries no special meaning — Torvalds rolls the major number when the minor number gets uncomfortably large, nothing more.

"Is there anything particularly interesting about this? The first number of the version changes when the second number gets too big, not for any other reason" — globular-toast

So the story isn't the 7. It's what's inside 7.1. The marquee feature is a rewritten in-kernel NTFS driver with full write support — delayed allocation, iomap and folio integration for real write performance, plus a new ntfsprogs-plus userspace suite — replacing the long-troubled lineage of NTFS support. There's an Apple SMC power driver that finally reports battery metrics on Apple Silicon MacBooks running the mainline kernel, early Intel Nova Lake P graphics, new AMD graphics enablement, and a modernized swap subsystem. On the other side of the ledger, the release deletes over 140,000 lines: ISDN, a pile of legacy networking drivers and protocols, PCMCIA host controllers, and several i486-era x86 subarchitectures (M486, M486SX, ELAN).

The reason for the deletions is the actually-new thing

The NTFS work is the headline most outlets ran. But the thread fixed on a single line from the changelog that reframes the removals as something genuinely 2026.

"Moving really old and unused code out of the kernel just to get less AI-assisted bug reports is IMO one of the best consequence ever of AI. I love it. We should start trimming the fat out of everything" — TacticalCoder, a long-tenured HN contributor

The quoted changelog rationale is the load-bearing detail: ISDN and other rarely-touched driver code was pulled specifically "to avoid that influx of bug reporting" generated by AI tooling scanning code paths almost nobody exercises. The maintainers didn't fix the reports — they removed the surface the reports were aimed at. It's the first time "automated bug-report spam" appears as a stated reason to delete working code from the kernel.

What practitioners are actually waiting for

For the people who'll run it, the interest isn't abstract. The NTFS rewrite has a specific predecessor to beat.

"Looking forward for a new NTFS driver to try. I hope this new iteration will be better than ntfs3 from Paragon" — smetannik

That's the real test: ntfs3, contributed by Paragon, has been the standard for years and has its own reliability reputation. A folio-and-iomap-based rewrite promising delayed allocation is a direct bid to replace it on performance. Distros are already lining up — one reader noted Arch was still defaulting to 7.0.10 and waiting on the 7.1 rollout.

The quiet shift in 7.1 isn't the NTFS speed or the version digit. It's that the kernel now treats obsolete-hardware drivers not just as maintenance burden but as an attack surface for machine-generated noise — and is willing to amputate to make the noise stop. "How many new lines came from AI?" one reader asked. For this release, the more telling number is how many got deleted because of it.