kanaodnd

Stellar-Tweaks

22.0guide

Intelligent Android performance daemon written in Rust. Fast-paced hybrid scheduler and real-time telemetry for Root and Non-Root environments.

★97 stars
•Performance & Kernel•by kanaodnd•Apache-2.0•Updated Sep 19, 2026
Platforms:
✓ Magisk✓ KernelSU✓ APatch
Download 22.0Starting download...GitHub Source
Stellar-v22.0-release5.3-VoyagingAstralist-25926.zip

Overview

Traditional Android optimization scripts predominantly consist of shell scripts that execute periodic loops using dumpsys or brute-force sysfs writes. These scripts waste CPU cycles through continuous context switching, introduce micro-stutters during heavy animation sequences, and fight against Android’s native framework scheduler.

Engineered entirely in Rust by kanaodnd, Stellar Tweaks is a deterministic systems daemon designed for low-latency kernel and framework orchestration. Running silently in the background, Stellar integrates directly with Linux userspace and kernel interfaces across Qualcomm Snapdragon, MediaTek, Samsung Exynos, and Unisoc architectures. It replaces uncoordinated scheduler hacks with genuine Fast Energy Aware Scheduling (FEAS), intelligent display rate overrides, and hardware-level battery isolation.

Architecture & Core Innovations

1. Fast Energy Aware Scheduling (FEAS)

  • Real-Time Render Monitoring: Rather than reacting after a frame has already dropped, Stellar tracks the throughput and latency of the active application’s RenderThread. If a sudden complex frame is detected, it ramps CPU frequencies ahead of the draw call deadline.
  • Cluster Affinity Mapping: Automatically pins high-priority game and UI threads to Prime and Big CPU performance clusters while shunting non-critical background jobs to Little efficiency cores.
  • Cooperative Idle Snapping: Gracefully steps hardware into minimum power states when the screen is static without relying on aggressive evdev event loops.

2. Cross-Platform GPU & I/O Control

Stellar provides tailored governor logic across all major mobile graphics architectures—including Qualcomm Adreno (KGSL), ARM Mali, and generic Devfreq nodes:

  • Gaming Profile: Locks high GPU minimum clocks and activates bypass charging.
  • Balanced Profile: Retains fluid 120Hz scrolling while lowering power consumption during reading and static media viewing.
  • Powersaver Profile: Enforces strict frequency ceilings and locks the display panel to 60Hz.

3. Thermal Bypass & Charging Management

During extended gaming sessions, battery heat is the primary catalyst for severe SoC thermal throttling. Stellar activates Bypass Charging (Idle Battery State) on supported hardware, allowing the device to draw power directly from the connected USB charger without routing current into the lithium cell, reducing internal operating temperatures.

4. Multi-Vendor Display Pipeline

OEM display managers (MIUI/HyperOS, One UI, ColorOS/OxygenOS, HiOS/XOS) frequently drop display refresh rates to 60Hz inside games or third-party web browsers. Stellar overrides vendor display lockouts to enforce chosen refresh rates while automatically respecting Android’s system Battery Saver mode.

User Interface (M3 WebUI)

Stellar includes an embedded Material You 3 (M3) WebUI dashboard. Compiled as a self-contained distribution, it requires zero external internet access, runs entirely offline, and renders natively inside KernelSU Manager, APatch Manager, or MMRL.

Users can inspect live CPU/GPU frequencies, toggle operational profiles, manage per-app overrides, and execute component cleanups directly from the interface.

Installation & Best Practices

  1. Ensure no conflicting performance or battery governor modules are installed.
  2. Flash the Stellar Tweaks .zip via your preferred root manager.
  3. Reboot your device to let Stellar inspect display refresh boundaries and kernel sysfs nodes.
  4. Open your root manager’s module panel to launch the Stellar WebUI.
  5. Customize profile rules or allow the auto-adaptive scheduler to manage workloads automatically.

[!WARNING] Because Stellar actively coordinates CPU/GPU governors, EAS task placement, and DVFS curves, running other performance scripts or scheduler modules simultaneously will cause parameter collisions and frame pacing volatility.