Cache-Aware Lock-Free Ring Buffer

Engineered wait-free synchronisation using std::atomic with explicit acquire/release semantics, eliminating all mutex contention and priority inversion risks.

Applied 64-byte cache-line padding to head/tail indices to prevent false sharing, validated via perf stat L1-cache-miss reduction on x86-64.

Modelled producer/consumer state transitions with UML sequence and class diagrams before implementation, enforcing invariant correctness at the design stage.