How is Ultrassembler so fast?

Summary

How is Ultrassembler so fast? Published 2025-08-30 Ultrassembler is a superfast and complete RISC-V assembler library that I'm writing as a component of the bigger Chata signal processing project. Assemblers take in a platform-dependent assembly language and output that platform's native machine code which runs directly on the processor. "Why would you want to do this?" you might ask. First, existing RISC-V assemblers that conform the the entirety of the specification, as and llvm-mc, ship as binaries that you run as standalone programs. This is normally not an issue. However, in Chata's case, it needs to access a RISC-V assembler from within its C++ code. The alternative is to use some ugly C function like system() to run external software as if it were a human or script running a command in a terminal. Here's an example of what I'm talking about: #include <iostream> #include <string> #include <stdlib.h> int main() { std::string command = "riscv64-linux-gnu-as code.s -o code.bin"; int res = std::system(command.data()); if (res != 0) { std::cerr << "Error executing command: " << command << std::endl; } return res; } It gets even worse once you realize you need temporary files and possibly have to artisanally craft the command beforehand. Additionally, invoking the assembler in this manner incurs a significant performance overhead on embedded systems which lack significant processing power. There must be a better way. Enter Ultrassembler. With these two goals of speed and standard conformance in mind, I wrote Ultrassembler as a completely standalone library with GNU as as the speed and standard conformity benchmark. The results are nothing short of staggering. After months of peformance optimization, Ultrassembler can assemble a test file with about 16 thousand RISC-V instructions over 10 times faster than as, and around 20 times faster than llvm-mc. To put it another way, it only takes about 1000 CPU instructions (+-50% depending on platform) to assemble one RISC-V ...