Computation of correctly rounded floating point summation
US-2024160405-A1 · May 16, 2024 · US
US9753693B2 · US · B2
| Field | Value |
|---|---|
| Publication number | US-9753693-B2 |
| Application number | US-201414208591-A |
| Country | US |
| Kind code | B2 |
| Filing date | Mar 13, 2014 |
| Priority date | Mar 15, 2013 |
| Publication date | Sep 5, 2017 |
| Grant date | Sep 5, 2017 |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
A binary logic circuit is provided for determining a rounded value of px q , where p and q are coprime constant integers with p<q and q≠2 i , i is any integer, and x is an integer variable between 0 and integer M where M≧2q, the binary logic circuit implementing in hardware the optimal solution of the multiply-add operation ax + b 2 k where a, b and k are fixed integers.
Opening claim text (preview).
The invention claimed is: 1. In an integrated circuit for performing mathematical operations, the improvement comprising a binary logic circuit configured to determine, for an input value x, a rounded value of px q by implementing the operation ax + b 2 k , where p and q are coprime constant integers with p<q and q≠2 i , i is an integer, input value x is an integer variable between 0 and integer M where M≧2q, a and b are predetermined fixed integers, and k is the smallest integer that satisfies either: 2 k ( p 2 k ) mod q > q ⌊ M q ⌋ - q + p - 1 ; or ( a ) 2 k ( - p 2 k ) mod q > M - ( ( M + p - 1 ) mod q ) ; ( b ) wherein the binary logic circuit comprises: multiplication and addition logic comprising a binary multiplier array having an input to receive x and configured to perform the multiplication of x by the predetermined fixed integer a and being incapable of performing multiplication of x by a variable factor, wherein the multiplication and addition logic is further configured to perform the addition of b; wherein, in dependence on the values of p and q: if (a) is satisfied by a smaller integer k than (b), then a is predetermined to be an integer given by ⌊ p 2 k q ⌋ and b is predetermined to be an integer in the range [ ⌊ M q ⌋ ( ( p 2 k ) mod q ) ,
Rounding · CPC title
Logic synthesis; Behaviour synthesis, e.g. mapping logic, HDL to netlist, high-level language to RTL or netlist · CPC title
Modular multiplication (G06F7/724, G06F7/727, G06F7/728 take precedence) · CPC title
using signed-digit representation · CPC title
Computations with numbers represented by a non-linear combination of denominational numbers, e.g. rational numbers, logarithmic number system or floating-point numbers {(G06F7/4806, G06F7/4824, G06F7/49, G06F7/491, G06F7/544 take precedence)} · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.