Linux Kernel 7.1 Released — New NTFS Driver, Intel FRED, and What’s New

Published: June 18, 2026 | Released June 14, 2026 | Covers desktop, gaming, and server impact

Linux Kernel 7.1 is here. Linus Torvalds tagged the release on June 14, 2026 — slightly ahead of schedule, fitting it in around travel plans — and it is a genuinely interesting mid-cycle release. Linux Kernel 7.1 pairs a headline feature that has been four years in the making with some hard cuts of legacy code that the kernel has been carrying far too long, plus meaningful improvements for Intel and AMD graphics, Steam Deck users, and Apple Silicon.

The merge window pulled in 12,996 non-merge changesets from 2,011 developers, 342 of whom were sending their first patch to the kernel. Here is everything that actually matters in Linux Kernel 7.1, explained for real users rather than kernel developers.

Best Linux Distro for Gaming in 2026 (AMD, NVIDIA and Beginner Picks)

---

The headline: a brand new NTFS driver, four years in the making

The biggest change in Linux Kernel 7.1 is one that will matter most to users who share storage between Linux and Windows — a completely rewritten in-kernel NTFS implementation.

The new NTFS filesystem implementation has been in the works for the last four years, featuring full write support with delayed allocation, iomap, and folio integration to improve write performance, better stability, and a new suite of userspace utilities called ntfsprogs-plus.

To understand why this matters, a little history. Linux has had NTFS read support for decades. The userspace driver NTFS-3G added write support but was always slow — running in userspace rather than in the kernel meant it could not compete with native filesystem performance. The new driver supports native in-kernel read/write operations, iomap support, folio-based memory management, removal of legacy buffer_head code, and delayed allocation for improved write efficiency.

In practice this means accessing NTFS-formatted external drives, shared partitions on dual-boot systems, and Windows network shares from Linux will be noticeably faster and more reliable. For dual-boot users especially — the exact audience currently weighing whether to switch from Windows 10 — this removes one of the most persistent friction points in the Linux experience.

One important caveat: the new driver is new kernel code, so edge cases around advanced NTFS features may take time to stabilise. For critical data on NTFS volumes, give it a few point releases before relying on it exclusively. NTFS-3G remains available as a fallback.

---

Intel FRED enabled by default

Intel users will notice that Flexible Return and Event Delivery (FRED) is now the default way the kernel handles specific system transitions. The switch was held back until publicly available hardware could be properly evaluated, and the code has since been tested thoroughly enough to flip from opt-in to opt-out.

FRED is Intel’s modern replacement for the older SYSCALL/SYSENTER mechanism for handling interrupts and exceptions. It reduces the overhead of these transitions — which happen thousands of times per second on a busy system — and is designed for Intel’s Panther Lake architecture and beyond.

For most users this is invisible. For developers and sysadmins running workloads that are interrupt-heavy — high-throughput networking, database servers, real-time processing — it is a meaningful efficiency improvement. If for any reason you need to disable it, add fred=off to your kernel boot parameters.

---

Graphics improvements — Intel Arc and AMD Radeon

Linux Kernel 7.1 brings faster Intel Arc Battlemage graphics and improvements for older AMD Radeon GPUs.

For Intel Arc users, the Battlemage improvements are particularly welcome — Arc graphics have had a complicated driver history on Linux, and each kernel release has incrementally improved the situation. Linux Kernel 7.1 continues that trend with performance and stability fixes for the Battlemage generation.

For AMD users, the improvements target older Radeon hardware — cards that the AMDGPU driver already supports but where edge cases and performance gaps remained. If you are running an older AMD card and have had intermittent issues, Linux Kernel 7.1 is worth updating to.

Apple Silicon support also sees continued improvement, with further progress on the ongoing effort to bring M-series chip support closer to feature-complete.

---

Steam Deck OLED audio fix

A bugfix for a long-standing audio issue on the Steam Deck OLED ships in Linux Kernel 7.1. This has been one of the more irritating hardware-specific bugs affecting Steam Deck OLED users on custom Linux setups — Valve’s own SteamOS had workarounds, but users running other distros on their Deck were left dealing with audio glitches. The mainline fix means all distros picking up Linux Kernel 7.1 will have this resolved.

---

Landlock security expansion

Linux Kernel 7.1 introduces a new Landlock access right for pathname UNIX domain sockets, thanks to a new LSM hook.

Landlock is Linux’s unprivileged sandboxing mechanism — it allows applications to restrict their own access to system resources without requiring root. The addition of pathname UNIX domain socket control means applications can now sandbox inter-process communication paths as well as filesystem access, closing a gap in what Landlock could previously protect.

For most desktop users this is background infrastructure. For application developers and security-conscious system administrators, it is a useful addition to the sandboxing toolkit.

---

Legacy code removal — 140,000 lines gone

Linux Kernel 7.1 finally drops i486 CPU support, removes old network driver code including ISDN-related pieces, removes bus mouse support, retires UDP-Lite kernel support, and drops some obsolete PCMCIA and PCI support — stripping over 140,000 lines of legacy code the kernel no longer needs.

The i486 removal is the most symbolic. Intel’s 486 processor launched in 1989 — maintaining special build paths for it in a 2026 kernel was pure legacy weight. Dropping it removes dead code and simplifies the build system without affecting any hardware any real user is running.

Two breaking networking changes to know about before upgrading:

UDP-Lite removal and IPv6 module mode being dropped are the most operationally disruptive items for anyone running custom kernel configs in production. If you maintain servers or custom kernel configurations that explicitly use UDP-Lite, audit your configs before upgrading. For desktop users this is irrelevant.

---

What was removed and why it matters

Removed	Reason	Impact
i486 CPU build support	35-year-old architecture, zero real users	None for anyone alive
ISDN network drivers	Technology obsolete since early 2000s	None
Bus mouse support	PS/2 predecessor, last used ~1995	None
UDP-Lite	Rare protocol, low adoption	Check configs if running custom kernels
~140,000 lines of legacy code	Maintenance overhead reduction	Smaller, cleaner kernel

---

What is next — Linux Kernel 7.2

The merge window for Linux Kernel 7.2 opens immediately following 7.1, with the first release candidate expected June 28, 2026, and a stable 7.2 targeted for late August 2026.

---

When will your distro get Linux Kernel 7.1?

The answer depends entirely on which distro you run:

Already have it or getting it imminently:

• Arch Linux — available in the main repository within days of release
• CachyOS — rolling release, picks it up immediately
• Fedora Rawhide — already in testing

Getting it in weeks:

• Fedora 44 — expect an update within 2–4 weeks
• openSUSE Tumbleweed — rolling release, fast turnaround

Getting it in months (backported):

• Ubuntu 26.04 LTS — will receive a Hardware Enablement kernel update
• Debian Testing — will pick it up during the next testing cycle
• AlmaLinux / RHEL — enterprise distros validate extensively; expect months, not weeks

Checking your current kernel version:

bash

uname -r

Manually installing on Arch:

bash

sudo pacman -Syu linux linux-headers

Manually installing on Fedora:

bash

sudo dnf update kernel

---

Should you upgrade now?

For most desktop users on rolling release distros — yes, update when it lands in your package manager. Linux Kernel 7.1 is a feature release with a solid development cycle and no last-minute scary issues flagged by Torvalds.

For servers and production systems — wait for your distro’s validated kernel package. Upgrading a production server to a manually compiled mainline kernel is not worth the risk.

For Steam Deck OLED users on custom distros — yes, the audio fix alone makes it worth updating.

For dual-boot users curious about the NTFS improvements — update and test carefully. The new driver is promising but new. Back up anything important on shared NTFS volumes before stress-testing it.

What Data Do Smart Home Devices Collect? (And What Happens to It)

What data do smart home devices collect? The honest answer is: far more than most people realise, far more than is necessary for the devices to do their jobs, and…

Matter 1.4 vs Zigbee vs Z-Wave: Best Smart Home Protocol for US Homes in 2026

By KontraNet IoT Hub | Last Updated: June 3, 2026 | Reading time: 11 min Quick Pick for US Homeowners in 2026 Use this table if you just need the answer fast: Your SituationBest…

Run AI Locally on Your PC in 2026

A Beginner’s Guide to Private AI For years, artificial intelligence has mostly lived in the cloud. Y…

Best AI Writing Tools for Bloggers in 2026 (Free & Paid)

The best AI writing tools for bloggers in 2026 aren’t just text generators anymore. The best o…