Kernel float: unlocking mixed-precision GPU programming

DOI10.1145/3779120MaRDI QIDQ7237071zbMATH OpenFDO

Authors Stijn Heldens, Ben van Werkhoven

Publication date 29 April 2026

Published in ACM Transactions on Mathematical Software (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://doi.org/10.1145/3779120

zbMATH Keywords

CUDAGPU computinghigh-performance computingHIPmixed-precision programming

Mathematics Subject Classification ID

Numerical analysis (65-XX)

Cites work

• <scp>Ginkgo</scp>  : A Modern Linear Operator Algebra Framework for High Performance Computing
• CPFloat: a C library for simulating low-precision arithmetic
• FloatX
• scientific article; zbMATH DE number 3136006 (Why is no real title available?)
• IEEE-754 precision-\(p\) base-\(\beta\) arithmetic implemented in binary
• MPFR
• On Division by Functional Iteration
• Simulating Low Precision Floating-Point Arithmetic
• Sur la détermination des polynomes d'approximation de degré donné.

This page was built for publication: Kernel float: unlocking mixed-precision GPU programming

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7237071)