arXiv Open Access 2016

Customizable Precision of Floating-Point Arithmetic with Bitslice Vector Types

Shixiong Xu David Gregg

Lihat Sumber

Abstrak

Customizing the precision of data can provide attractive trade-offs between accuracy and hardware resources. We propose a novel form of vector computing aimed at arrays of custom-precision floating point data. We represent these vectors in bitslice format. Bitwise instructions are used to implement arithmetic circuits in software that operate on customized bit-precision. Experiments show that this approach can be efficient for vectors of low-precision custom floating point types, while providing arbitrary bit precision.

Topik & Kata Kunci

cs.OH

Penulis (2)

Shixiong Xu

David Gregg

Format Sitasi

APA MLA BibTeX

Xu, S., Gregg, D. (2016). Customizable Precision of Floating-Point Arithmetic with Bitslice Vector Types. https://arxiv.org/abs/1602.04716

Akses Cepat

Lihat di Sumber

Informasi Jurnal

Tahun Terbit: 2016
Bahasa: en
Sumber Database: arXiv
Akses: Open Access ✓