US 11,880,925 B2
Atomic memory update unit and methods
Steven J. Clohset, San Francisco, CA (US); Jason R. Redgrave, Mountain View, CA (US); and Luke T. Peterson, San Francisco, CA (US)
Assigned to Imagination Technologies Limited, Kings Langley (GB)
Filed by Imagination Technologies Limited, Kings Langley (GB)
Filed on Jan. 7, 2022, as Appl. No. 17/571,104.
Application 17/571,104 is a continuation of application No. 15/275,645, filed on Sep. 26, 2016, granted, now 11,257,271.
Application 15/275,645 is a continuation of application No. 14/494,496, filed on Sep. 23, 2014, granted, now 9,466,091, issued on Oct. 11, 2016.
Claims priority of provisional application 61/955,116, filed on Mar. 18, 2014.
Claims priority of provisional application 61/955,086, filed on Mar. 18, 2014.
Claims priority of provisional application 61/882,755, filed on Sep. 26, 2013.
Prior Publication US 2022/0230378 A1, Jul. 21, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06T 15/00 (2011.01); G06F 12/0804 (2016.01); G06T 15/06 (2011.01); G06T 17/10 (2006.01); G06T 5/00 (2006.01); G06T 1/60 (2006.01); G06F 12/0862 (2016.01); G06F 12/126 (2016.01)
CPC G06T 5/005 (2013.01) [G06F 12/0804 (2013.01); G06T 1/60 (2013.01); G06T 15/06 (2013.01); G06T 17/10 (2013.01); G06F 12/0862 (2013.01); G06F 12/126 (2013.01); G06F 2212/1024 (2013.01); G06F 2212/302 (2013.01); G06F 2212/452 (2013.01); G06F 2212/455 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A machine-implemented method of ray tracing, comprising:
concurrently executing, in a programmable computation unit, one or more threads of computation corresponding to a group of computation tasks identified for concurrent execution, wherein the programmable computation unit executes threads of computation over a set of time frames, the programmable computation unit being configured to concurrently execute the one or more threads of computation by executing the one or more threads of computation in the same time frame, and wherein executing each thread of the one or more threads comprises executing an instruction causing issuance of an operation code including data that identifies a ray; and
at a logic module arranged to execute independently of the programmable computation unit and operable to perform a predetermined set of operations, reading the operation code and performing an operation specified by the operation code for the ray.