← Search

Dr.Saman Amarasinghe

2 accepted papers

2019

Compiler Auto-Vectorization with Imitation Learning

NeurIPS 2019poster

Modern microprocessors are equipped with single instruction multiple data (SIMD) or vector instruction sets which allow compilers to exploit fine-grained data level parallelism. To exploit this parallelism, compilers employ auto-vectorization techniques to automatically convert scalar code into vect…

2019

Ithemal: Accurate, Portable and Fast Basic Block Throughput Estimation using Deep Neural Networks

ICML 2019oral

Predicting the number of clock cycles a processor takes to execute a block of assembly instructions in steady state (the throughput) is important for both compiler designers and performance engineers. Building an analytical model to do so is especially complicated in modern x86-64 Complex Instructio…