The Single Instruction Multiple Data (SIMD) architecture, supported by various high-
performance computing platforms, efficiently utilizes data-level parallelism. The SIMD model …

The emergence of new architectures create a recurring challenge to ensure that existing
programs still work on them. Manually porting legacy code is often impractical. Static binary …

Vector processing has become commonplace in today's CPU microarchitectures. Vector
instructions improve performance and energy which is crucial for resource-constraint mobile …

System virtualization is a fundamental technology that enables many important applications.
However, existing virtualization techniques suffer from a critical limitation due to their limited …

SIMD (Single Instruction Multiple Data) instructions apply the same operation to multiple
elements simultaneously. Compilers transform codes to exploit SIMD instructions through …

Advanced processors incorporate SIMD extensions to execute data-parallel operations
efficiently. As technology advances, new generations of SIMD extensions evolve with longer …

Production compilers have achieved a high level of maturity in terms of generating efficient
code. Compilers are embedded with numerous code optimization techniques, with special …

随着单指令流多数据流(SIMD) 技术的迅速发展, 近年来许多面向SIMD 扩展部件的自动向量化
编译方法被提出, 有效缓解了程序员手写向量程序的压力, 并发挥了SIMD 扩展部件的加速效能 …

First, we present goSLP, a framework that uses integer linear programming to find a globally
pairwise-optimal statement packing strategy to achieve superior vectorization performance …

Program synthesis is a promising method for building efficient, flexible software by deriving
low-level implementations from high-level specifications. In this thesis, I use programming …