Linux 7.1 Ships Ground-Up NTFS Rewrite, Intel FRED Support, and Arc Graphics Gains

Linux kernel 7.1 has landed with three features worth tracking closely: a from-scratch in-kernel NTFS driver with full read/write capability, Intel FRED support for Panther Lake silicon, and a round of performance improvements for Arc Graphics.

A New NTFS Driver in the Kernel Proper

The NTFS story is the one most directly felt by anyone running dual-boot systems or accessing Windows volumes from a Linux host. The new driver is a ground-up rewrite — not a patch on the aging ntfs.ko that shipped for years, and not a wrapper around the out-of-tree NTFS3 driver that Paragon contributed to the staging tree in 5.15. This implementation lives in the kernel proper and handles both reads and writes directly, without userspace relay through FUSE (Tom's Hardware).

The practical gap that closes here is meaningful. The original in-kernel NTFS driver was effectively read-only in any safe configuration; write support existed on paper but was widely flagged as unreliable. NTFS3 addressed that for users willing to load it explicitly, but its staging-tree lineage meant distribution packaging was uneven. A rewritten, mainline driver changes the integration calculus for distro maintainers — it becomes a default they can ship with confidence rather than an optional module they have to justify.

Worth noting: a ground-up rewrite of a filesystem driver is not a conservative choice. NTFS is a complex, proprietary-origin format with decades of on-disk variants, and correctness under edge cases — sparse files, reparse points, compression attributes — is genuinely hard to get right. The kernel community will be watching the bug backlog on this one closely over the next few cycles.

Intel FRED and Panther Lake

On the processor side, 7.1 adds support for Intel's Flexible Return and Event Delivery architecture — FRED — targeting Panther Lake, Intel's next-generation client processor family (Phoronix). FRED is Intel's replacement for the older SYSENTER/SYSEXIT and interrupt-delivery mechanisms, designed to reduce the overhead and complexity of privilege-level transitions. Where the legacy model required the OS to manage significant bookkeeping around stack switching and segment state, FRED moves more of that coordination into hardware, with the goal of both lowering transition latency and tightening the security model around event delivery.

Getting FRED support into the mainline kernel ahead of Panther Lake's broad availability is standard practice for Intel and the kernel community — it avoids the scramble that plagued earlier architecture features when hardware shipped before OS support was solid. For kernel engineers working on low-level interrupt handling or security primitives, this is the change in 7.1 with the longest tail.

Arc Graphics Performance Work

The third thread is less architecturally dramatic but operationally relevant for anyone running Intel Arc discrete GPUs on Linux. Kernel 7.1 carries performance improvements for Arc Graphics, continuing an incremental cadence of driver-level work that has been closing the gap between the Linux and Windows Arc experiences since the GPU line launched (Phoronix). The specific nature of those improvements — whether they are register-level tuning, command-submission path optimizations, or memory management changes — is not detailed in available release notes, but the direction of travel is consistent with prior cycles.

Arc on Linux has been a long rehabilitation project. Early driver quality was the platform's weakest point, and much of the credibility work has happened in-kernel rather than in userspace. Each release that chips away at that gap matters to the segment of developers and power users who want Arc's price-to-performance ratio without booting into Windows to get acceptable throughput.

Taken together, the three changes in 7.1 reflect a kernel release that is doing what mature kernel releases are supposed to do: improving hardware coverage, modernizing low-level interfaces, and reducing friction at the boundaries between Linux and the broader ecosystem it increasingly has to coexist with. The NTFS rewrite is the most user-visible of the three. The FRED work is the one with the deepest long-term implications for platform security and performance. The Arc improvements are the quietest — and for the people waiting on them, the most immediately welcome.