US 12,147,804 B2
Systems, methods, and apparatuses for tile matrix multiplication and accumulation
Robert Valentine, Kiryat Tivon (IL); Zeev Sperber, Zichron Yackov (IL); Mark J. Charney, Lexington, MA (US); Bret L. Toll, Hillsboro, OR (US); Rinat Rappoport, Haifa (IL); Stanislav Shwartsman, Haifa (IL); Dan Baum, Haifa (IL); Igor Yanover, Yokneam Illit (IL); Elmoustapha Ould-Ahmed-Vall, Chandler, AZ (US); Menachem Adelman, Haifa (IL); Jesus Corbal, King City, OR (US); Yuri Gebil, Nahariya (IL); and Simon Rubanovich, Haifa (IL)
Assigned to Intel Corporation, Santa Clara, CA (US)
Filed by Intel Corporation, Santa Clara, CA (US)
Filed on Jul. 22, 2021, as Appl. No. 17/382,917.
Application 17/382,917 is a continuation of application No. 16/487,787, granted, now 11,086,623, previously published as PCT/US2017/040548, filed on Jul. 1, 2017.
Claims priority of provisional application 62/473,732, filed on Mar. 20, 2017.
Prior Publication US 2021/0349720 A1, Nov. 11, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 9/30 (2018.01); G06F 7/485 (2006.01); G06F 7/487 (2006.01); G06F 7/76 (2006.01); G06F 9/38 (2018.01); G06F 17/16 (2006.01)
CPC G06F 9/30036 (2013.01) [G06F 7/485 (2013.01); G06F 7/4876 (2013.01); G06F 7/762 (2013.01); G06F 9/3001 (2013.01); G06F 9/30032 (2013.01); G06F 9/30043 (2013.01); G06F 9/30109 (2013.01); G06F 9/30112 (2013.01); G06F 9/30134 (2013.01); G06F 9/30145 (2013.01); G06F 9/30149 (2013.01); G06F 9/3016 (2013.01); G06F 9/30185 (2013.01); G06F 9/30196 (2013.01); G06F 9/3818 (2013.01); G06F 9/3836 (2013.01); G06F 17/16 (2013.01); G06F 2212/454 (2013.01)] 21 Claims
OG exemplary drawing
 
1. A processor comprising:
decode circuitry to decode an instance of a single instruction having fields for an opcode, an identifier for a first source multidimensional matrix operand, an identifier of a second source multidimensional matrix operand, and an identifier for a source/destination multidimensional matrix operand; and
execution circuitry to execute the decoded instance of the single instruction to multiply the identified first source multidimensional matrix operand by the identified second source multidimensional matrix operand, add a result of the multiplication to the identified source/destination multidimensional matrix operand, and store a result of the addition in the identified source/destination multidimensional matrix operand.