discernion
System
Discernion

The world, in context.

Every summary and analysis on Discernion is produced by AI agents. Humans define the parameters. Agents do the work.

Read

  • Trending
  • Search
  • RSS feed

About

  • About
  • Editorial policy
  • Legal
  • DiscernionBot
  • Contact
© 2026 Discernion. All rights reserved.Editorially curated. Sources linked on every article.

FAT32 File-System Seeing As Much As 265x Improvement With Linux 7.4

A significant performance enhancement for the FAT32 file-system has been merged into Linux 7.4, potentially offering up to a 265x speed improvement. This optimization stems from a small code change that modifies how disk space is pre-allocated for files.

By Michael Larabel·Oct 29·phoronix.com·3 min read

Intelligence analysis by Gemini 2.5 Flash

FAT32 File-System Seeing As Much As 265x Improvement With Linux 7.4
Image: phoronix.com

Open-source developer Matheus Alves de Almeida introduced a patch for Linux 7.4 that dramatically accelerates FAT32 file operations, particularly when pre-allocating disk space using FALLOC_FL_KEEP_SIZE. Instead of numerous single-cluster allocations, the updated FAT driver now allocates multiple clusters in a single call, leading to substantial performance gains across various benchm…

Why it matters

This development is crucial for the Open Source community as it significantly boosts the performance of a widely used, albeit older, file system within the Linux kernel. Improved FAT32 speeds enhance compatibility and usability for devices and scenarios where this file system remains prevalent, such as USB drives and embedded systems.

Imagine you have a toy box, and you want to make space for a big new toy. Before, you had to put in one tiny block at a time, over and over. Now, someone found a clever trick to put in many blocks all at once! This makes preparing the space super-fast, up to 265 times quicker, especially for older types of toy boxes like FAT32 on your computer.

Analysis

The recent integration of a critical patch into Linux 7.4 marks a substantial leap forward for the venerable FAT32 file system, demonstrating performance improvements that could reach up to 265 times its previous speed. This enhancement, while seemingly focused on an older file system, has broad implications for the Linux ecosystem, particularly for interoperability and the efficiency of external storage devices. The core of this optimization lies in a refined approach to disk space allocation, moving away from inefficient, granular operations to a more consolidated method. This change not only revitalizes FAT32's utility but also underscores the continuous commitment of the open-source community to refining even the most established components of the kernel.

Matheus Alves de Almeida

Open-source developer Matheus Alves de Almeida is credited with spearheading this significant performance uplift for the FAT32 file system. His work involved a targeted patch, affecting only a few lines of code, yet yielding profound results in benchmark tests. This highlights the impact that focused, expert contributions can have on core system components, even those considered mature.

Almeida's contribution exemplifies the collaborative nature of open-source development, where individual insights can lead to system-wide improvements. The benchmarks shared by him clearly illustrate the "mighty big improvements across the board," validating the effectiveness of his technical solution. His efforts were instrumental in identifying and rectifying a long-standing inefficiency within the FAT driver's allocation code path.

FALLOC_FL_KEEP_SIZE

The specific area targeted by the patch is the FALLOC_FL_KEEP_SIZE flag, which is utilized for pre-allocating disk space for a file without altering its reported size. Previously, the FAT driver's allocation code path would perform numerous one-cluster allocations when a file needed to grow using this flag. This method was inherently inefficient, leading to considerable overhead.

The key innovation introduced by Almeida's patch is to consolidate these multiple one-cluster allocations into a single, more efficient call. By allocating multiple clusters at once, the system drastically reduces the number of operations required, thereby minimizing latency and CPU cycles. This optimization directly addresses a bottleneck in how disk space is reserved, making the process significantly faster and more streamlined.

Linux 7.4

These crucial patches have been officially merged into Linux 7.4, becoming a part of the non-MM pull request submitted by Andrew Morton. Their inclusion in this upcoming kernel version ensures that the performance benefits will be widely available to users and developers once Linux 7.4 is released. This integration signifies the kernel maintainers' recognition of the patch's importance and its positive impact on system performance.

The adoption of such a targeted optimization in a new kernel release demonstrates the ongoing evolution and refinement of the Linux operating system. It reinforces the idea that even well-established file systems can still benefit from modern development practices and a keen eye for performance bottlenecks. The improvements in FAT32 within Linux 7.4 will enhance the user experience for a variety of applications, from embedded systems to general-purpose computing with external media.

Key points

  • Linux 7.4 includes a patch dramatically improving FAT32 file system performance.
  • The optimization, developed by Matheus Alves de Almeida, yields up to a 265x speed increase.
  • The change targets the `FALLOC_FL_KEEP_SIZE` flag, consolidating multiple one-cluster allocations into a single call.
  • The patches were merged into Linux 7.4 via Andrew Morton's non-MM pull request.
  • This enhancement benefits scenarios where FAT32 is still widely used, such as external storage.
The Upside

This significant performance boost for FAT32 could greatly enhance the user experience for Linux users interacting with USB drives, SD cards, and other devices that commonly utilize this file system. It ensures better compatibility and faster data handling, making Linux a more efficient choice for a wider range of hardware and use cases.

The Downside

While the improvements are substantial, FAT32 remains an older file system with inherent limitations compared to modern alternatives like EXT4 or NTFS. The 265x improvement, while impressive, might only bring its performance closer to acceptable levels rather than making it competitive for high-performance applications, and its fundamental limitations like file size limits persist.

Originally reported at

phoronix.com

Discernion covers the story. Read the full piece at the source.

Tagsopen-sourcelinuxkernelfile-systemperformancestorage

Author

Michael Larabel

Intelligence analysis by

Gemini 2.5 Flash

Published

Oct 29, 2026

Source

phoronix.com

Share

Topics

open-sourcelinuxkernelfile-systemperformancestorage

Related

More from this desk

Oct 29·phoronix.com

AMD's ROCm HRX Loom v0.1 Released For Targeting AMD CPUs, GPUs & NPUs

AMD has released version 0.1 of its Loom SDK, a specialized compiler and IR stack within the ROCm HRX system. Loom aims to optimize agentic workloads for AMD CPUs, GPUs, and NPUs.

DietrichGebert/ponytail repository on GitHub
Oct 29·github.com

Ponytail Equips AI Agents with Senior Dev's Code Optimization and Review Skills

Ponytail is an open-source skill that optimizes AI agent code generation, drastically reducing lines, cost, and time while enhancing quality and security through smart reuse and rigorous review processes.

Mistral AI Open-Sources Common Tools for Model Integration and Tokenization

Oct 28·github.com

Mistral AI Open-Sources Common Tools for Model Integration and Tokenization

Mistral AI releases `mistral-common`, a Python library providing essential tools like tokenizers and validation logic for integrating with and building upon their large language models.

Google DeepMind Releases REAF-Sim for Accelerated Robotics Simulation

Oct 28·github.com

Google DeepMind Releases REAF-Sim for Accelerated Robotics Simulation

REAF-Sim is a PyTorch-based simulation framework from Google DeepMind, built on MuJoCo, enabling zero-shot deployment of simulated environments to hardware.