Branchless quicksort beats std::sort and pdqsort in C/C++
This blog introduces blqsort, a branch‑less quicksort implementation available as single‑header C and C++ files. Benchmarks on Apple M1 and AMD Ryzen show it sorting 50 million doubles up to 25 % faster than std::sort and 30 % faster than pdqsort, with multi‑threaded variants even quicker. Source code is on GitHub.