rw-r-r-0644 GitHub avatar

bc250-core-unlock

rw-r-r-0644

Enables two disabled CPU cores on bc250 boards via an SMU mailbox exploit, upgrading from 6c/12t to 8c/16t.

Stars

66

7-day growth

No data

Forks

7

Open issues

0

License

MIT

Last updated

2026-07-30

AI repository intelligence
FR-AI / ANALYSIS

Why it is worth attention

It demonstrates a low-level hardware unlock through SMU privilege escalation, increasing core count by 33% with no idle power increase, though with caveats.

Who it is for

  • Owners of bc250 boards with PS5 Oberon die
  • Advanced Linux users comfortable with low-level hardware manipulation
  • Developers interested in AMD power management exploits
  • System administrators needing maximum CPU cores on these boards

Use cases

  • Unlocking extra cores for compute workloads on bc250 systems
  • Testing stability of unlocked cores under load
  • Running multi-threaded applications that benefit from more cores
  • As a proof-of-concept for SMU-based hardware modifications

Strengths

  • Low power impact – idle wall power unchanged (72.68W vs 72.43W)
  • Modest temperature increase under load (44°C to 55°C with 16 threads)
  • Survives warm reboots, avoiding redo after soft reset
  • Includes safety check – only proceeds if core mask reads exactly 0x77

Considerations

  • Does not survive cold boot – must be reapplied after each cold power cycle
  • Tested only on BIOS 3.0 and kernel 6.18.40; compatibility with other versions unverified
  • Suspend (s2idle) may become unresponsive; GPU clock reporting becomes cosmetic (incorrect values)

README quick start

usage

Description

Unlock 2 extra cores on the AMD BC-250

Related repositories

Similar projects matched by category, topics, and programming language.

lopopolo
Featured
lopopolo GitHub avatar

harness-engineering

Harness Engineering is a methodology for improving coding agent outputs by carefully crafting the environment around them—providing curated context, tools, and executable constraints that encode an organization’s nonfunctional requirements and cumulative lessons.

AI & Machine LearningAI Agents
2,390
slvDev
Featured
slvDev GitHub avatar

esp32-ai

A 28.9 million parameter language model runs on an $8 ESP32-S3 microcontroller entirely on-device, generating simple stories at about 9.5 tokens per second.

AI & Machine LearningLarge Language Models
1,960
littledivy
Featured
littledivy GitHub avatar

mimic

mimic captures traffic from any iOS or web app and automatically generates a Python client library that lets you call the app's API like a regular library.

AI & Machine Learning
1,482