Eliminate Branches by Melding IR Instructions

TL;DR

MERIT is a compiler technique that eliminates branches by merging similar IR instructions from divergent paths, improving performance especially on architectures where traditional if-conversion struggles.

Contribution

It introduces a novel IR-level merging approach using sequence alignment and operand guarding, enabling branch elimination without hardware predication.

Findings

Achieves 10.9% average speedup on benchmark programs

Peak improvements reach up to 32x over hardware branch predictors

Reduces static instruction overhead compared to traditional methods

Abstract

Branch mispredictions cause catastrophic performance penalties in modern processors, leading to performance loss. While hardware predictors and profile-guided techniques exist, data-dependent branches with irregular patterns remain challenging. Traditional if-conversion eliminates branches via software predication but faces limitations on architectures like x86. It often fails on paths containing memory instructions or incurs excessive instruction overhead by fully speculating large branch bodies. This paper presents Melding IR Instructions (MERIT), a compiler transformation that eliminates branches by aligning and melding similar operations from divergent paths at the IR instruction level. By observing that divergent paths often perform structurally similar operations with different operands, MERIT adapts sequence alignment to discover merging opportunities and employs safe operand-level guarding to ensure semantic correctness without hardware predication. Implemented as an LLVM pass and evaluated on 102 programs from four benchmark suites, MERIT achieves a geometric mean speedup of 10.9% with peak improvements of 32x compared to hardware branch predictor, demonstrating the effectiveness with reduced static instruction overhead.