Microsoft AI scanner finds long-lived Rust standard library bugs
Original title: Microsoft’s AI bug hunter finds Rust flaw that went undetected for nearly 11 years
Two issues affect stable Rust, including a possible freed-memory access; Microsoft calls practical exploitation unlikely.
Microsoft FORGE Lab researchers, using the Hyenas AI vulnerability scanner, reported memory-safety bugs in Rust's standard library, including a process::exit flaw present for nearly 11 years. Calling it from a child thread can remove the main thread's alternate signal stack and turn a later stack overflow into a segmentation fault. A separate CommandExt::exec issue can leave an environment pointer referencing freed memory. Microsoft said each flaw needs unusual conditions and found no reliable exploitation path for the signal-stack bug.
Unconfirmed: Microsoft assessed practical exploitation as unlikely; the Rust project had not responded.
- Hyenas is FORGE Lab's research version of codename MDASH.
- A MapWindows::clone panic can reach uninitialized memory; Microsoft said a fix was reported.
- str::replace can emit invalid UTF-8 via a custom nightly Pattern implementation.
- A proposed execve-based fix was under review, and two reported issues remained open.
- The Rust project had not commented when CSO published.
Impact: Undefined behavior or access to freed memory under unusual conditions
| Product | Affected |
|---|---|
| Rust project Rust standard library | — |
- Vendors
- Rust project
- Products
- Rust
From the original article
4 min read at csoonline.com
Microsoft researchers have uncovered a memory-safety bug in Rust’s standard library that had gone undetected for nearly 11 years, alongside other flaws that can trigger undefined behavior even in applications written entirely in safe Rust. Two of the findings affect stable Rust, including a separate issue that could cause applications to access freed memory.
csoonline.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.