Chained multiply accumulate using an unrounded product

The processing circuitry comprises chained-floating-point multiply accumulate circuitry 602 configured to: generate an unrounded product based on multiplying a first floating-point operand and a second floating-point operand; generate a first rounding increment based on the unrounded product; genera...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Javier Diaz Bruguera, Harsha Valsaraju, David Raymond Lutz
Format: Patent
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Javier Diaz Bruguera
Harsha Valsaraju
David Raymond Lutz
description The processing circuitry comprises chained-floating-point multiply accumulate circuitry 602 configured to: generate an unrounded product based on multiplying a first floating-point operand and a second floating-point operand; generate a first rounding increment based on the unrounded product; generate a sum based on adding the unrounded product, a value based on the first rounding increment, and the third floating-point operand; determine a second rounding increment based on the sum; and perform rounding based on the second rounding increment. The rounding increment may be used to round the numbers using an injection rounding approach. Instead of rounding the product of the first and second FP operands and then separately adding the third FP operand, the product is effectively rounded at the same time as adding the third FP operand, and the rounding of the product does not need to delay preliminary steps for preparing for addition of the third FP operand.
format Patent
fullrecord <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_GB2617436A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>GB2617436A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_GB2617436A3</originalsourceid><addsrcrecordid>eNrjZDBzzkjMzEtNUcgtzSnJLMipVEhMTi4FchJLUhVKizPz0hUS8xRK84ryS_NSgMoKivJTSpNLeBhY0xJzilN5oTQ3g7yba4izh25qQX58anFBYnJqXmpJvLuTkZmhuYmxmaMxYRUAGOUtHQ</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Chained multiply accumulate using an unrounded product</title><source>esp@cenet</source><creator>Javier Diaz Bruguera ; Harsha Valsaraju ; David Raymond Lutz</creator><creatorcontrib>Javier Diaz Bruguera ; Harsha Valsaraju ; David Raymond Lutz</creatorcontrib><description>The processing circuitry comprises chained-floating-point multiply accumulate circuitry 602 configured to: generate an unrounded product based on multiplying a first floating-point operand and a second floating-point operand; generate a first rounding increment based on the unrounded product; generate a sum based on adding the unrounded product, a value based on the first rounding increment, and the third floating-point operand; determine a second rounding increment based on the sum; and perform rounding based on the second rounding increment. The rounding increment may be used to round the numbers using an injection rounding approach. Instead of rounding the product of the first and second FP operands and then separately adding the third FP operand, the product is effectively rounded at the same time as adding the third FP operand, and the rounding of the product does not need to delay preliminary steps for preparing for addition of the third FP operand.</description><language>eng</language><subject>CALCULATING ; COMPUTING ; COUNTING ; ELECTRIC DIGITAL DATA PROCESSING ; PHYSICS</subject><creationdate>2023</creationdate><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&amp;date=20231011&amp;DB=EPODOC&amp;CC=GB&amp;NR=2617436A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,780,885,25564,76547</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&amp;date=20231011&amp;DB=EPODOC&amp;CC=GB&amp;NR=2617436A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>Javier Diaz Bruguera</creatorcontrib><creatorcontrib>Harsha Valsaraju</creatorcontrib><creatorcontrib>David Raymond Lutz</creatorcontrib><title>Chained multiply accumulate using an unrounded product</title><description>The processing circuitry comprises chained-floating-point multiply accumulate circuitry 602 configured to: generate an unrounded product based on multiplying a first floating-point operand and a second floating-point operand; generate a first rounding increment based on the unrounded product; generate a sum based on adding the unrounded product, a value based on the first rounding increment, and the third floating-point operand; determine a second rounding increment based on the sum; and perform rounding based on the second rounding increment. The rounding increment may be used to round the numbers using an injection rounding approach. Instead of rounding the product of the first and second FP operands and then separately adding the third FP operand, the product is effectively rounded at the same time as adding the third FP operand, and the rounding of the product does not need to delay preliminary steps for preparing for addition of the third FP operand.</description><subject>CALCULATING</subject><subject>COMPUTING</subject><subject>COUNTING</subject><subject>ELECTRIC DIGITAL DATA PROCESSING</subject><subject>PHYSICS</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2023</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZDBzzkjMzEtNUcgtzSnJLMipVEhMTi4FchJLUhVKizPz0hUS8xRK84ryS_NSgMoKivJTSpNLeBhY0xJzilN5oTQ3g7yba4izh25qQX58anFBYnJqXmpJvLuTkZmhuYmxmaMxYRUAGOUtHQ</recordid><startdate>20231011</startdate><enddate>20231011</enddate><creator>Javier Diaz Bruguera</creator><creator>Harsha Valsaraju</creator><creator>David Raymond Lutz</creator><scope>EVB</scope></search><sort><creationdate>20231011</creationdate><title>Chained multiply accumulate using an unrounded product</title><author>Javier Diaz Bruguera ; Harsha Valsaraju ; David Raymond Lutz</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_GB2617436A3</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>eng</language><creationdate>2023</creationdate><topic>CALCULATING</topic><topic>COMPUTING</topic><topic>COUNTING</topic><topic>ELECTRIC DIGITAL DATA PROCESSING</topic><topic>PHYSICS</topic><toplevel>online_resources</toplevel><creatorcontrib>Javier Diaz Bruguera</creatorcontrib><creatorcontrib>Harsha Valsaraju</creatorcontrib><creatorcontrib>David Raymond Lutz</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Javier Diaz Bruguera</au><au>Harsha Valsaraju</au><au>David Raymond Lutz</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Chained multiply accumulate using an unrounded product</title><date>2023-10-11</date><risdate>2023</risdate><abstract>The processing circuitry comprises chained-floating-point multiply accumulate circuitry 602 configured to: generate an unrounded product based on multiplying a first floating-point operand and a second floating-point operand; generate a first rounding increment based on the unrounded product; generate a sum based on adding the unrounded product, a value based on the first rounding increment, and the third floating-point operand; determine a second rounding increment based on the sum; and perform rounding based on the second rounding increment. The rounding increment may be used to round the numbers using an injection rounding approach. Instead of rounding the product of the first and second FP operands and then separately adding the third FP operand, the product is effectively rounded at the same time as adding the third FP operand, and the rounding of the product does not need to delay preliminary steps for preparing for addition of the third FP operand.</abstract><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier
ispartof
issn
language eng
recordid cdi_epo_espacenet_GB2617436A
source esp@cenet
subjects CALCULATING
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
PHYSICS
title Chained multiply accumulate using an unrounded product
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-02T08%3A08%3A09IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-epo_EVB&rft_val_fmt=info:ofi/fmt:kev:mtx:patent&rft.genre=patent&rft.au=Javier%20Diaz%20Bruguera&rft.date=2023-10-11&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3EGB2617436A%3C/epo_EVB%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true