Skip to content
KitploitKITPLOIT
ToolsExploitsBlog
Log in
Submit
ToolsExploitsBlog
Submit

Hacking, PenTest, and Cybersecurity Tools for Your Security Arsenal!

Kitploit is a directory of hacking, cybersecurity, and pentesting tools. Discover the latest project updates to find vulnerabilities, analyze systems, automate testing, and strengthen your security.

FeedsContactPrivacy© 2026 Kitploit

Tool Directory

Categories

View all categories
Loading categories
chronomaly — Android kernel exploit for CVE-2025-38352, previously exploited in-the-wild. Targets vulnerable x86_64 Linux kernels v5.10.x. | Kitploit
Tools/GitHubGitHub/farazsth98/chronomaly
Android SecurityVulnerability AnalysisExploitationPapers & ResearchLearning & EducationBinary Exploitation
GitHubfarazsth98/chronomaly

chronomaly

Android kernel exploit for CVE-2025-38352, previously exploited in-the-wild. Targets vulnerable x86_64 Linux kernels v5.10.x.

View Repository
310484110 months agoReviewed by Kitploit
Share

Chronomaly

Chronomaly is a kernel exploit for the Android / Linux kernel using CVE-2025-38352. The exploit was written specifically for Linux kernel v5.10.157, but should work against all vulnerable v5.10.x kernels, as it does not require any specific kernel text offsets to work.

I covered the vulnerability in detail in a three-part blog post series, all the way from PoC to exploit:

  • Part 1 - In-the-wild Android Kernel Vulnerability Analysis + PoC
  • Part 2 - Extending The Race Window Without a Kernel Patch
  • Part 3 - Uncovering Chronomaly

demo

Build Setup

This exploit has only been tested against an x86_64 Linux kernel v5.10.157 running in QEMU. I asked a friend to send me their Pixel 6a kernel config to base my kernel config off of, and these are the config options important for this exploit (I started from the kernelCTF config as a base):

  • CONFIG_POSIX_CPU_TIMERS_TASK_WORK=n
  • CONFIG_PREEMPT=y (Full Preemption, no RT)
  • CONFIG_SLAB_MERGE_DEFAULT=n
  • DEBUG_LIST=n
  • BUG_ON_DATA_CORRUPTION=n
  • LIST_HARDENED=n

To disable CONFIG_POSIX_CPU_TIMERS_TASK_WORK, you can follow the steps laid out in my first blog post here.

Refer to the qemu.sh file for my QEMU run script. I used 4 cores and 3 GB RAM for testing.

Exploit parameters you'll need to change

Since the exploit depends on CPU timers, there are two parameters you may need to change to adapt it to your environment.

CPU_USAGE_THRESHOLD

This parameter is used when consuming CPU time to fire the timers inside the race_func(). It must be set such that:

  • The timers don't fire on every retry attempt (this would imply that CPU_USAGE_THRESHOLD is too high, as the timers are firing before the race_func() thread can exit).
  • The timers only fire sometimes (this would imply that sometimes the timers fire before the thread exits, and other times it fires while the thread exits).

To determine whether the timers are firing or not, insert a printf() statement in the SIGUSR1 polling code in free_func(). If you see the message print out, that means the timers fired.

If set correctly, you'll start seeing the "Parent raced too late / too early" messages in the terminal.

PARENT_SETTIME_DELAY_US

PARENT_SETTIME_DELAY_US. This parameter is used by the parent process to hit the 2nd race window inside send_sigqueue() at the same time as the child process. Run the exploit, observe, and modify it as follows:

  • The message "Parent raced too late, readjusting..." appears too often – reduce this parameter.
  • The message "Parent raced too early, readjusting..." appears too often – increase this parameter.

Ideally, you want to see both "raced too late" and "raced too early" being printed out, and the exploit will work within 1 minute. If you only see one occurring more than the other, adjust accordingly.

Potential Improvements

In my cross-cache implementation, I assumed that the kernel is not too busy, and that there haven't been many struct sigqueue allocations. I added a comment in sigqueue_crosscache_preallocs() that explains what you would need to do improve this.

If the kernel is really busy, or there are already some struct sigqueue slab pages on the per-cpu / per-node partial list, then the current cross-cache implementation in exploit.c will fail, and the uaf_sigqueue / realloc_sigqueue won't be re-allocated as a pipe buffer data page.

I purposely chose not to make the cross-cache work in a busy kernel, so that the exploit doesn't get misused :)

Questions

If you have any questions, please contact me via X / Twitter!

Download Tool

Most Popular

View all →

Discover the most used tools by our community.

  • 1

    OsintHunter

    letox4/osinthunter

    🎯 The ultimate OSINT toolkit: investigation graph, confidence scoring, alias families, map, timeline and reports. 49 modules, 400+ sites, no API keys required.

    732
  • 2

    ghostlock-app

    yukonga/ghostlock-app

    GhostLock One-Tap Execution App (CVE-2026-43499)

    506
  • 3

    ARTEX

    baarai/artex

    LLM multi-agent autonomous penetration testing platform with planner/worker agents, asset and exploration graphs, MITM traffic recording, and vulnerability retesting.

    405
  • 4

    redforge

    cybersecurity91/redforge

    Browser-side red team toolkit with 102 offline tools for recon, payload generation, credential attacks, crypto, encoding and engagement reporting.

    342
  • 5

    CyberStrikeAI

    ed1s0nz/cyberstrikeai

    The system of action for AI-native cybersecurity—where intent becomes governed execution, evidence becomes operational memory, and every operation improves the next.

    261
  • 6

    antifox

    antifox-ai/antifox

    AI multi-agent autonomous penetration testing platform that runs recon, attack-surface discovery, exploitation validation and evidence-backed report delivery.

    259

Explore all tools

Browse our collection of tools

View all tools →