Large multipliers with less DSP blocks

Florent de Dinechin 1 Bogdan Pasca 1
1 ARENAIRE - Computer arithmetic
Inria Grenoble - Rhône-Alpes, LIP - Laboratoire de l'Informatique du Parallélisme
Abstract : Recent computing-oriented FPGAs feature DSP blocks including small embedded multipliers. A large integer multiplier, for instance for a double-precision floating-point multiplier, consumes many of these DSP blocks. This article studies three non-standard implementation techniques of large multipliers: the Karatsuba-Ofman algorithm, non-standard multiplier tiling, and specialized squarers. They allow for large multipliers working at the peak frequency of the DSP blocks while reducing the DSP block usage. Their overhead in term of logic resources, if any, is much lower than that of emulating embedded multipliers. Their latency overhead, if any, is very small. Complete algorithmic descriptions are provided, carefully mapped on recent Xilinx and Altera devices, and validated by synthesis results.
Document type :
Conference papers
Complete list of metadatas

https://hal-ens-lyon.archives-ouvertes.fr/ensl-00356421
Contributor : Florent de Dinechin <>
Submitted on : Tuesday, January 27, 2009 - 2:43:37 PM
Last modification on : Wednesday, November 20, 2019 - 3:06:08 AM
Long-term archiving on : Friday, October 12, 2012 - 10:20:15 AM

File

RR2009-03.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : ensl-00356421, version 1

Collections

Citation

Florent de Dinechin, Bogdan Pasca. Large multipliers with less DSP blocks. Field Programmable Logic and Applications, Aug 2009, Czech Republic. ⟨ensl-00356421⟩

Share

Metrics

Record views

368

Files downloads

4874