This Month in Rust OSDev: September 2026
Welcome to a new issue of "This Month in Rust OSDev". In these posts, we give a regular overview of notable changes in the Rust operating system development ecosystem.
This series is openly developed on GitHub. Feel free to open pull requests there with content you would like to see in the next issue. If you find some issues on this page, please report them by creating an issue or using our comment form at the bottom of this page.
Please submit interesting posts and projects for the next issue by commenting on the draft pull request or via a PR on GitHub.
Disclaimer: Automated scripts and AI assistance were used for collecting and categorizing links. Everything was proofread and checked manually, with many manual tweaks.Announcements, News, and Blog Posts
Here we collect news, blog posts, etc. related to OS development in Rust.
- OSTEP & Redox: Introduction & The Process
- Blog series that maps the concepts of the Operating Systems: Three Easy Pieces textbook to Redox.
- Follow-up posts: How does scheduling work on Redox? and Memory Management in Redox - Pt. 1 Page Tables
- This Month in Redox - August 2026
- Rust in the kernel? What about Rust without the kernel!
- The OpenVMM Project
- Building a DMA based driver for the RP2350 I2C (safety not included)
- Reverse engineering my e-scooter and rewriting the firmware in rust
- The Embedded Rustacean Issue #79, Issue #80, and Issue #81
Linux-related
- Kangrejos 2026
- The Rust for Linux workshop took place on September 16–17 in Las Palmas. Slides for many sessions (pin-init, field projection, Tyr, Rust UFS driver, Rust SPDM, …) are linked on the page.
- Video: Kernel Recipes 2026 - Rust for Linux
- Talk by Miguel Ojeda.
- Compiling the kernel with gccrs
- The corresponding RustConf talk is available too: Arthur Cohen & Pierre-Emmanuel Patry: "Compiling the Linux Kernel with gccrs" | RustConf 2026
- Google's "Painful To Maintain" Binder C Linux Driver Being Removed In Favor Of Rust
- The "rnull" Rust block driver
- Guide to building kernel modules in Rust
Infrastructure and Tooling
In this section, we collect recent updates to rustc, cargo, and other tooling that are relevant to Rust OS development.
- alloc: stabilise
Allocator- After years behind the
allocator_apifeature gate, an MVP of theAllocatortrait is stable, together withBox<T, A>,Vec<T, A>,Global, and the*_with_allocatormethods. This makes custom per-collection allocators possible on stable Rust. The remaining parts of the old feature move toallocator_ext.
- After years behind the
- libcore: expose volatile atomic operations
- Adds unstable
load_volatile/store_volatilemethods on the atomic types (atomic_volatilefeature), for memory shared with devices or other address spaces, such as DMA buffers.
- Adds unstable
- x86: on targets that requires SSE, use those registers for ABI
- On SSE-requiring x86 targets, the Rust ABI now passes values in SSE registers and disabling SSE is a hard error. Kernels that disable SSE need a soft-float target.
- feat(builtin-deps): Add builtin dependencies manifest syntax
- make target feature ABI check a hard error on ARM
- Code that requests a hard-float ABI on ARM without the matching float target features is now rejected. This was a future-compat warning since Rust 1.86.
- Stabilize
unsafe_cell_access- Stabilizes
UnsafeCell::as_ref_unchecked,as_mut_unchecked, andreplace.
- Stabilizes
- support
#[target_feature(enable = ...)]on#[naked]functions - Add Tier 3 targets for Hyperlight guests
- Adds
x86_64-unknown-hyperlightandaarch64-unknown-hyperlighttargets for bare-metal guests running in Hyperlight micro-VMs.
- Adds
rust-osdev Projects
In this section, we give an overview of notable changes to the projects hosted under the rust-osdev organization.
uefi-rs
Maintained by @nicholasbishop and @phip1611
uefi makes it easy to develop Rust software that leverages safe, convenient,
and performant abstractions for UEFI functionality.
We continued last month's soundness and specification compliance work. The
common theme: the crates no longer blindly trust what the firmware reports. For
example, boot::memory_map trusted the map size reported by the firmware, so
sorting a map larger than its buffer read out of bounds, and several boot and
protocol functions turned null pointers returned by the firmware into handles or
references. In uefi-raw, we audited the pointer mutability of the protocol
definitions against the UEFI and PI specifications. This is breaking for the raw
bindings, but it lets the high-level uefi crate hand out &/&mut without
lying to the compiler.
Not everything was about fixing things. uefi-raw gained HII Internal Forms
Representation (IFR) types and uefi gained the AbsolutePointer protocol,
CString16::extend, and allocation-free iteration over the components of a
Path.
All of this was released as uefi v0.41.0, uefi-raw v0.17.0, and
uefi-macros v0.20.0.
As mentioned last month, Anthropic sponsors @phip1611 with a Max plan as part of their open source program. Much of the auditing above happened under that sponsorship, and we are happy that it helps us improve the security, robustness, and reliability of the ecosystem.
Thanks to @crawfxrd, @cwize1 and @the-shank for their contributions!
We merged the following PRs this month:
- release: uefi-macros-0.20.0, uefi-raw-0.17.0, uefi-0.41.0
- Various Fixes of Potential Undefined Behavior (LLM Assisted)
- Various Fixes of Potential Undefined Behavior (LLM Assisted) (2/N)
- Various Fixes of Potential Undefined Behavior (LLM Assisted) (3/N)
- Various Fixes of Potential Undefined Behavior (LLM Assisted) (4/N)
- Various Fixes of Potential Undefined Behavior (LLM Assisted) (5/N)
- Various Fixes of Potential Undefined Behavior and Panics (LLM Assisted) (6/N)
- Various Fixes of Potential Undefined Behavior and Panics (LLM Assisted) (8/N)
- uefi: fix aliasing UB in PciRootBridgeIo I/O accessors
- uefi: mem: fix the allocation handling of make_boxed
- uefi: media: validate the file info returned by firmware
- uefi-raw: spec violation fixes: outputs behind read-only pointers (1/5)
- uefi-raw: spec violation fixes: read-only inputs (2/5)
- uefi-raw: spec compliance: event notification context (3/5)
- uefi-raw: spec compliance: caller-owned outputs (4/5)
- uefi-raw: spec compliance: mode pointers and docs (5/5)
- uefi-raw: a few more spec adjustments regarding api_guidelines.md
- A few small Spec Fixes/Adjustments
- uefi-raw: Add HII IFR bindings
- uefi: Add AbsolutePointerProtocol
- Add EdidDiscovered protocol
- uefi: some string and path improvements
- uefi: Improve the debug format of
CStr8,CStr16, andCString16. - uefi-raw: Use efiapi for C variadic functions
- pxe: use PxeBaseCodeBootType in PxeBaseCodeSrvlist
- data_types: export FromSliceUntilNulError
- uefi-raw: improve documentation how to model UEFI types
- Various Small Release Preparation Fixes
- Some Small Fixes
acpi
Maintained by @IsaacWoods
The acpi repository contains crates for parsing the ACPI tables – data structures that the firmware of modern computers uses to relay information about the hardware to the OS.
We merged the following changes this month:
- Breaking: Improve pm1 enable registers control
- Add functions to get and clear pending events for Pm1EventRegisterBlock
- Add I2C and GPIO resource descriptors
- Add
set_interrupt_model_usedmethod for calling\_PIC - Create
pci_routing::Pin::from_pci_interrupt_pinconvenience fn. - Make RegionHandlers be Send + Sync by default
- feat: Don't take explicit references to
Handler - Use updated PhysicalMapping interface
- feat(aml): make
spinning_top,byteorder, andsmallvecoptional - Keep a copy of block stream instead of raw pointer
- Ensure that Locals passed to Methods are preserved
- Correctly unwind stack when handling
Continue - Prevent some infinite recursion and loops
- Fix a couple issues with end tags
- fix: Fix compilation with only the
allocfeature - Add basic test for ConcatenateRestTemplate
Thanks to @dewyatt, @martin-hughes, @mkroening, and @ChocolateLoverRaj for their contributions!
virtio-spec-rs
Maintained by @mkroening
The virtio-spec crate provides definitions from the Virtual I/O Device (VIRTIO) specification.
This project aims to be unopinionated regarding actual VIRTIO drivers that are implemented on top of this crate.
This month, the crate was upgraded to version 1.4 of the VIRTIO specification and gained virtio-blk definitions, released as v0.4.0:
- chore: Release version 0.4.0
- feat: add virtio-blk definitions
- feat: update
virtio::Idto spec 1.4 - feat: add
DeviceStatus::SUSPENDfrom spec 1.4 - feat: update features to spec 1.4
- feat(driver_notifications): upgrade to spec 1.4
- feat(net): upgrade to spec 1.4
- feat(pci): upgrade to spec 1.4
- feat(mmio): upgrade to spec 1.4
- feat: make enums ABI-compatible
x86_64
Maintained by @phil-opp, @josephlr, and @Freax13
The x86_64 crate provides various abstractions for x86_64 systems, including wrappers for CPU instructions, access to processor-specific registers, and abstraction types for architecture-specific structures such as page tables and descriptor tables.
We published a first release candidate for v0.16.0 this month. We merged the following changes:
- release 0.16.0-rc.0
- increase the Minimum Supported Rust Version to 1.98
- make memory encryption bit an upper limit for physical address bits
- feat(paging): Implement
IntoIteratorandCopyfor ranges - fix: Add
#[track_caller]to many address, page, and frame methods - fix(paging): Fix panic on displaying a page table with the last physical frame
- clean up features
Thanks to @mkroening for their contributions!
multiboot2
Maintained by @phip1611
Convenient and safe parsing of Multiboot2 Boot Information (MBI) structures and the contained information tags. Usable in no_std environments, such as a kernel. An optional builder feature also allows the construction of the corresponding structures.
The raw_type! macro we announced last month landed. For users, this means that
values unknown to the specification - an unknown header tag type, architecture,
or memory area type - no longer produce undefined behavior. They now arrive as a
Custom variant that can simply be ignored.
On the builder side, structures built on the stack could leak uninitialized padding bytes into the output. Built structures are now byte-wise identical to before, except that the padding between tags is guaranteed to be zeroed - so what a bootloader hands to a kernel no longer depends on whatever was on the stack.
Released as multiboot2 v0.27.0 and v0.28.0, multiboot2-header v0.11.0, and
multiboot2-common v0.6.0 and v0.7.0. These come with a few breaking changes,
most notably that Header and MaybeDynSized are now unsafe traits, which
affects users implementing custom tags.
We merged the following PRs this month:
uart_16550
Maintained by @phip1611
Simple yet highly configurable low-level driver for 16550 UART devices, typically known and used as serial ports or COM ports.
v0.8.1 adds the public method Uart16550::check_present(), which probes for a
device through the scratch register. init() now delegates its existing
presence check to it, so users can run the same probe on their own before
touching the device.
The repository also gained a real-hw-test crate member that builds a bootable
EFI image. This makes it much easier to verify the driver on real hardware
rather than only in virtual machines - which is exactly what this crate was
rewritten for.
We merged the following PRs this month:
Other Projects
In this section, we describe updates to Rust OS projects that are not directly related to the rust-osdev organization. Feel free to create a pull request with the updates of your OS project for the next post.
No project updates were submitted this month.
Join Us?
Are you interested in Rust-based operating system development? Our rust-osdev organization is always open to new members and new projects. Just let us know if you want to join! A good way to get in touch is our Zulip chat.