Core Tech Blog on TiJet

Hugo Image Optimization: Cutting an 8.6MB Background to 310KB

Tue, 04 Aug 2026 00:00:00 +0000

1. Why Optimize Images

Images are usually the biggest culprit for slow static sites:

Reason	Explanation
Large file size	A 2560×1440 PNG can be 8-10MB
Inefficient format	PNG is lossless — great for screenshots, bad for photos
No lazy loading	Page loads all images at once
No caching	Re-downloads on every visit

Core idea: make the browser download smaller files.

2. Hugo Image Processing Pipeline

Hugo has built-in image processing that runs at build time:

Jenkins CI/CD Hands-On: Building an Automated Pipeline from Scratch

Tue, 04 Aug 2026 00:00:00 +0000

1. Why Jenkins

Jenkins is an open-source CI/CD tool. Its core purpose: after code is pushed to a Git repository, automatically run “fetch code → build → test → package → deploy”.

2. Installation

Prerequisite: Java

Jenkins requires Java 17 or 21. Newer versions (e.g., Java 26) fail to start:

Running with Java 26 ... not yet fully supported. Supported Java versions are: [17, 21]

wget https://get.jenkins.io/war-stable/latest/jenkins.war -O /tmp/jenkins.war
/opt/jdk-21/bin/java -jar /tmp/jenkins.war --httpPort=9090

Recommended: Docker

docker run -d \
 --name jenkins \
 -p 9090:8080 \
 -v jenkins_home:/var/jenkins_home \
 jenkins/jenkins:lts-jdk21

3. Initial Setup

Visit http://localhost:9090:

Secure Boot and Firmware Signing: A Practical Guide for Industrial IoT Devices

Wed, 22 Jul 2026 00:00:00 +0000

1. Why Bother?

In industrial control, a compromised firmware image isn’t just a data breach — it’s a physical safety hazard. A tampered firmware could disable watchdog timers, override thermal shutdown thresholds, or silently exfiltrate proprietary motion-control parameters.

Yet most industrial IoT devices ship with JTAG left open and firmware images verified by nothing more than a CRC32 checksum. CRC32 catches bit flips; it does nothing against a motivated attacker.

Taming Linux Jitter: Achieving Sub-50µs Cycle Times with PREEMPT_RT and Xenomai

Tue, 21 Jul 2026 00:00:00 +0000

1. The Challenge: Soft PLC on a $30 SoC

Tainuo’s next-generation motion controller needed to run a soft-PLC runtime doing 20 kHz closed-loop PID control on six servo axes simultaneously. The target hardware: a $30 Allwinner A40i quad-core Cortex-A7. No FPGA, no dedicated DSP — just plain ARM cores running Linux.

The problem was obvious: standard Linux, even with SCHED_FIFO, routinely exhibits 200–500µs scheduling jitter on this class of hardware. For a 20 kHz control loop (50µs period), that’s a non-starter. Missing the deadline means a phase error that accumulates and destabilizes the mechanical system.

FPGA-Accelerated Visual Inspection: How We Cut Latency from 200ms to 8µs

Mon, 20 Jul 2026 00:00:00 +0000

1. The Problem: Software-Based Visual Inspection Hits a Wall

In Tainuo’s PCB printing pipeline, one of the most critical steps is real-time solder-paste inspection. Each board passing through the production line has less than 50ms total for defect detection — any slower and the line stalls.

For years we ran a GPU-accelerated OpenCV pipeline on an x86 edge node. It worked — mostly. But under peak load with high-resolution 8K images, the pipeline occasionally hit 180–220ms per frame. That 180ms of jitter meant missed defects, false passes, and production-line backpressure. The root cause was architectural: a general-purpose GPU pipeline, no matter how optimized, introduces unpredictable scheduling latency and PCIe transfer overhead.

Rust on Bare Metal: Rewriting Our Bootloader Without a Single Page Fault

Sun, 19 Jul 2026 00:00:00 +0000

1. The Bootloader Problem

Tainuo’s industrial control boards ship with an MCU that manages power sequencing, watchdog timers, and over-the-air (OTA) firmware updates. The bootloader — roughly 8,000 lines of C — had been stable for years, but “stable” in the embedded sense means “the bugs are known and we have workarounds.”

Every 6–9 months, a new OTA edge case would surface: a buffer overflow during firmware image verification, an off-by-one in the flash erase routine, a use-after-free in the CAN bus command dispatcher. Each of these bugs required a full recall-cycle OTA fix, which on a deployed fleet of industrial controllers is a multi-week ordeal.

From Proprietary Protocols to gRPC: A Major Leap in Tainuo's Heterogeneous Industrial Control Communication Architecture

Sat, 18 Jul 2026 00:00:00 +0000

1. Why We Had to Abandon Our Legacy Proprietary Protocols

For a long time, Tainuo’s industrial control boards communicated with the host computer and print control terminals using self-built proprietary protocols. These protocols were indeed lightweight, flexible, and easy to modify during single-category development. However, as Tainuo gradually moves toward Industry 5.0 standards, our system architecture is evolving into a “universal industrial-grade heterogeneous computing and intelligent control platform.”

Against this backdrop, the legacy proprietary protocols exposed three irreconcilable pain points: prohibitively high multi-language SDK maintenance costs (C#, C++/Rust heterogeneous endpoints required manual parser rewriting), lack of strong type safety boundaries, and difficulty in flexibly extending RPC services across heterogeneous systems.