Skip to content
NewA calmer, faster ZeroHour

Branch Target Reuse: Spectre-v2 Attacks in JIT Engines

Lobsters · securityby vusec.net via fro ()7 min read5 · 4 reads
AI summary · grok-4.7Summaries by Vani AI

VUSec's Branch Target Reuse attack reuses stale JIT branch predictions to speculatively leak memory on Intel CPUs.

VUSec researchers present Branch Target Reuse, a Spectre-v2 attack that reuses stale indirect branch predictions after JIT code is freed and reallocated. An end-to-end Linux cBPF exploit leaks arbitrary kernel memory on modern Intel CPUs, including a root password hash from the su process, at about 8 bytes per second and bypasses enabled mitigations plus constant blinding. Proofs of concept also show speculative execution into Firefox SpiderMonkey's WebAssembly constant pool and a way to skip GraalVM sandbox masking, though those are not full end-to-end exploits.

Key facts
  • Stale branch-target buffer entries survive JIT code free and reuse.
  • Linux cBPF exploit leaks about 8 bytes per second, including a root hash.
  • Researchers bypassed cBPF constant blinding, which is off by default.
  • SpiderMonkey proof of concept leaks tens of bytes per second on Intel.
  • GraalVM masking can be skipped, but GC often clears BTB entries.
Analyze this story with Vani AIImpact, who is affected and what defenders should do, in a new chat

From the original article

7 min read at vusec.net

Practical Spectre-v2 Attacks in JIT Engines via Stale Branch Prediction Entries We present Branch Target Reuse (BTR) , a new Spectre-v2 attack targeting just-in-time (JIT) compilers. BTR affects the JIT engines found in web browsers, language runtimes, and the operating system kernel, across multiple CPU vendors.

Lobsters · security

By vusec.net via fro · vusec.net

Read the full article at Lobsters · security

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.