Back To Schedule
Friday, October 30 • 11:15am - 12:00pm
Optimizing LLVM for GPGPU

Sign up or log in to save this to your schedule, view media, leave feedback and see who's attending!

This talk presents Google’s effort of optimizing LLVM for CUDA. When we started this effort, LLVM was well-tuned for CPUs but there had been little public work on improving its GPU performance. We developed, tuned, and augmented several general and CUDA-specific optimization passes. As a result, our LLVM-based compiler generates better code than nvcc on key end-to-end internal benchmarks and is on par with nvcc on a variety of open-source benchmarks.

avatar for Jingyue Wu

Jingyue Wu

Software Engineer, Google Inc.

Friday October 30, 2015 11:15am - 12:00pm PDT
Salon III & Salon IV

Attendees (0)