https://morcules-blog.pages.dev/ Systems programming blog page focused on C, networking, multithreading, and low-level performance programming with real-world benchmarks and optimizations. 2026-05-11T15:01:36+00:00 Morcules https://morcules-blog.pages.dev/ Jekyll © 2026 Morcules /assets/img/favicons/favicon.ico /assets/img/favicons/favicon-96x96.png 2026-05-04T00:00:00+00:00 2026-05-04T00:00:00+00:00 https://morcules-blog.pages.dev/posts/How-programming-languages-really-work/ Morcules

Have you ever wondered how programming languages really work? There are several phases that your code goes through. Some compilers have more phases, but I will only talk about the basic ones. These stages are included in my own compiler. We will talk about a compiler made fully from scratch without using LLVM, to go more into detail. Tokenizer The tokenizer phase is really simple. Basically, ...

2026-05-04T00:00:00+00:00 2026-05-04T00:00:00+00:00 https://morcules-blog.pages.dev/posts/How-hashmaps-really-work-under-the-hood/ Morcules

Have you ever wondered how hashmaps really work under the hood? It might seem very complex, but once you understand it, it’s actually very simple. Hashmaps Are Just Arrays With Smart Indexing Hashmaps are basically arrays. They use generated indexes from the input. How Hashing Creates the Index So how do we generate those indexes? It’s right there in the name, by hashing. The exact algorith...

2026-05-01T00:00:00+00:00 2026-05-01T00:00:00+00:00 https://morcules-blog.pages.dev/posts/Reducing-Atomic-Overhead-in-a-Multithreaded-C-Allocator-by-50/ Morcules

When working with multi-threaded C code, atomic operations often look cheap, until they become your biggest bottleneck. In my case, a small change in how threads acquired memory reduced CPU usage by 40-50%, cutting execution from roughly 2 billion cycles to 1 billion cycles when sending 50MB of data in my networking library. Performance Changes Before: CPU Cycles : 2,273,393,860 User Cpu ...