COMPUTER ARCHITECTURES AND INSTRUCTIONS FOR MULTIPLICATION

A method and apparatus is described, comprising: receiving potential partial products for a first element of a first input vector or scalar; receiving a second input, wherein the second input is a scalar, a vector or a matrix; and for each term of an output vector or scalar: selecting one or more of...

Full description

Saved in:
Bibliographic Details
Main Authors AIROLDI, Roberto, ROUNIOJA, Kim, JIVANESCU, Mihaela, GUEVORKIAN, David, KAJIRU, Kajiru
Format Patent
LanguageEnglish
French
German
Published 08.05.2019
Subjects
CALCULATING
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:A method and apparatus is described, comprising: receiving potential partial products for a first element of a first input vector or scalar; receiving a second input, wherein the second input is a scalar, a vector or a matrix; and for each term of an output vector or scalar: selecting one or more of the potential partial products based on a term of the second input corresponding to the respective term of the output vector or scalar to provide a set of selected partial products; and shifting and adding the selected partial products to the respective term of the output vector or scalar.
Bibliography:Application Number: EP20170199898