Compiler-Assisted Floating-Point Error Analysis and Profiling with FPChecker · Tutorial

© 2026. All rights reserved.

Tutorial<br>on Floating-Point Analysis Tools

Compiler-Assisted Floating-Point Error Analysis and Profiling with FPChecker

ISC High Performance 2026 (ISC26), Hamburg, Germany

June 22, 2026

Time: 9:00am - 1pm Europe/Berlin (half-day tutorial)

Location: CCH – Congress Center Hamburg, Room: Hall X12 - 1st Floor

Description

Floating-point arithmetic is central to scientific computing, but small rounding<br>effects can accumulate into significant numerical errors, especially when<br>applications are ported to lower precision for performance and energy savings.<br>This tutorial introduces compiler-assisted analysis tools built on the<br>FPChecker framework (https://fpchecker.org/) and clang/LLVM to instrument C/C++<br>code and expose floating-point behavior.

Participants will learn how to evaluate dynamic range and precision<br>requirements, track rounding and relative error propagation across code<br>regions, and detect infinities and NaNs caused by numerical limits. Hands-on<br>examples, including linear solvers, finite-difference methods, and other C/C++<br>codes, will run on participants’ laptops. In the advanced session, attendees<br>can apply the tools to applications or widely used numerical<br>libraries. All tools and examples will be distributed through Conda, with<br>pre-configured AWS instances available for participants who cannot use it.

Presenter

Ignacio Laguna, Lawrence Livermore National Laboratory

AWS Instances

https://fpchecker.org/usernames

Presentation Slides

The tutorial slides are here: slides

Repositories:

FPChecker: https://github.com/LLNL/FPChecker

Documentation