A Summary of Markdown Syntax and External Tag Usage
Built-in Markdown Formatting Reference: Markdown 语法图文全面详解(10 分钟学会) Note: This page may occasionally have CSS conflict issues! Code Blocks 示例源码渲染演示\```sh # VSCode终端 hexo clean; hexo s hexo clean; hexo g; hexo d git add .; git commit -m "npm publish"; npm version patch; git push # Cmder终端 hexo clean && hexo s hexo clean && hexo g && hexo d git add . && git commit -m "npm publish" && npm version patch git push \```# VSCode终端 hexo clean; hexo s hexo cl...
How to Ask Questions the Smart Way
Reprint Notice The following content is reposted by Asuna from How-To-Ask-Questions-The-Smart-Way It is intended solely to accelerate access within mainland China. If there is any infringement, please contact us and it will be removed! How To Ask Questions The Smart Way How To Ask Questions The Smart Way Copyright © 2001,2006,2014 Eric S. Raymond, Rick Moen The English version of this guide is copyrighted by Eric S. Raymond and Rick Moen. Original URL: http://www.catb.org/~esr/faqs/smart-qu...
Running KDE on Pad 6S Pro: It Works, but Accept the Cost First
After unlocking the Pad 6S Pro (sheng), flashing back to the official 306 build, and getting root set up, beyond “one more machine I can tinker with,” I wanted it to do some laptop things: LibreOffice, a terminal, a browser, drawing. Let’s be clear about what this is first: this is a chroot, not native Linux. Android keeps running underneath, competing for memory and throttling the CPU, and graphics go through X11 forwarding rather than direct output. Light office work is doable; don’t count ...
Rooting the Xiaomi 13 and Pad 6S Pro: when the patch went on boot, the tool even built it a ramdisk
Both the Xiaomi 13 and the Pad 6S Pro run GKI kernels (5.15.194); the ramdisk lives in init_boot, and boot contains only the kernel — and that is what decides which partition the patch has to go on. The tablet came first. Unlock, flash the official 306 package back, confirm it boots normally, then install SukiSU, put the original boot.img into /sdcard, and pick “Select and patch a file” in the app. The log had these two lines: Adding KernelSU LKM No ramdisk, create by default Flashed it, rebo...
After unlocking Xiaomi 13 / Pad 6S Pro: empty userdata, dual-slot abl, and the location of init_boot
Both devices run HyperOS 3 (309 on the Xiaomi 13, 306 on the Pad 6S Pro). Once the unlock step is done, the screen shows an orange warning, a five-second wait, and then it boots into the system. Up to here everything is normal. Then the problems start. This post documents three sticking points, and what they share is this: the error message points somewhere other than the real cause. “Couldn’t mount file system” doesn’t mean the system flash failed Unlocking wipes data, so the data partition...
Xiaomi 13 (fuxi) OS3.0.309 Bootloader Unlock: Only One of Four Routes Doesn't Gamble with the Motherboard
The device is a Xiaomi 13 (fuxi), HyperOS 3, build number OS3.0.309.0.WMCCNXM, kernel 5.15.194, security patch 2026-08-01. There is only one goal: skip the 168-hour wait and get the BL unlocked. Conclusion first: in the end it came down to a domestic tool from 2026, using a temporary kernel root to write an engineering abl into both slots, then booting into fastboot to run the unlock. Three approaches were tried before this one: two finished tools failed on the spot, and a downgrade approach ...
Read It Once Before Deleting, Read It Again After Deleting
After the archive backend moved from Windows to object storage, that machine still had a whole old chain left on it: scheduled tasks, the H: drive directory, a root directive in Caddy, a stale credential copy, and a WSL build directory created by mistake. The service no longer uses them, but “not used” and “no impact” are two different things. That root * H:/wg in Caddy points at a directory that no longer exists; the config is still running, and the next reload is a real chance of failure. A...
Moving saves into object storage: plan your rollback first
The save bodies have always lived in a Windows storage service, right alongside the game sources in the same bucket. The reason to move them isn’t complicated: that backend can only run on a Windows box with a GUI, and every save of every user is a complete copy, so the same bytes get stored over and over. The goal is to move the bodies into a shared CAS in object storage: the object name is just <sha256>.bin, across users and across works, keeping only one copy of identical bytes. The ...
The response body the debugger hands me isn't necessarily the one on the wire
The thing to prove is very specific: whether a brand-new browser with zero cache can fully download the startup resources of four engines and reach the native title screen. “Brand-new” is provable: four fresh Chromium profiles, with object counts, Service Workers, and Cache Storage all at zero before launch, no pre-filled cache, no reading of existing resource copies, and no purging of the Cloudflare cache. Requests are triggered on demand, one at a time, with at least 5 seconds between start...
Five minutes idle, and that route stops working
On October 28, I ran three SSH attempts against the original entry point of the fixed server localhost-1: two timed out. Switching to the LAN address 192.168.9.29, three consecutive attempts succeeded, in 0.8 to 0.9 seconds. Same tailnet, same machine, one path works and the other doesn’t. The difference is that “direct connection” here means Tailscale’s NAT traversal path, and once traversal succeeds it goes quiet — stay quiet for a few minutes, send one more packet, and it has to punch thro...
