Apple Inc.

Jonathan M Redshaw

Justin A Hensley

Ali Rabbani Rankouhi

Steven John Fishwick

Lior Zimet

Luca O Iuliano

Christopher A Burns

Per Hakan Hammarlund

Eran Tamari

Harshavardhan Kaushikkar

Sergio Kolor

Gaurav Garg

James Vash

Sagi Lahav

Steven R Hutsell

Shawn Munetoshi Fukami

Sergio V Tota

Yoong Chert Foo

Ralph Clay Taylor

Karthik Ramani

Karl D Mann

Randall R Rauwendaal

Fang Liu

Terence M Potter

Benjiman L Goodman

Winnie W Yeung

David A Gotwalt

Benjamin Bowman

Tzach Zemer

Fergus W MacGarry

Zelin Zhang

Max J Batley

Alfredo Kostianovsky

Haim Hauzi

Idan Nissel

Leonid Gitelman

Oren Betzalel

Dalia R Haim

David K Li

Ji Rao

A system including a plurality of processor cores, a plurality of graphics processing units, a plurality of peripheral circuits, and a plurality of memory controllers is configured to support scaling of the system using a unified memory architecture.

Scalable system on a chip

A chip design methodology and a set of integrated circuits that are taped out from a common design database are disclosed. The area of a full instance of the integrated circuit is defined, and one or more chop lines are defined to identify portions that will be removed for one or more partial instances. A variety of techniques and mechanisms are defined to permit the tape outs to occur from a common design database, so that the effort to tape out partial instances may be minimized beyond that to tape out the full instance.

Full die and partial die tape outs from common design

Disclosed techniques relate to acceleration data structure for ray intersection testing. In some embodiments, storage circuitry stores node data for a spatially organized acceleration data structure, including to store the following node information for a given node: origin coordinates for the node and, for a given child node of multiple child nodes, child information that includes: quantized bounding region information for a bounding region corresponding to the child node, where the quantized bounding region information encodes bounding region coordinates as offsets relative to the origin coordinates. Traversal circuitry may traverse multiple nodes of the data structure and determine whether a ray intersects a bounding region indicated by given a node of the data structure based on the node information. Disclosed techniques may provide substantial improvements to performance, data size, and power consumption.

Node encoding for spatially-organized ray tracing acceleration data structure

Disclosed embodiments relate to controlling sets of graphics work (e.g., kicks) assigned to graphics processor circuitry. In some embodiments, tracking slot circuitry implements entries for multiple tracking slots. Slot manager circuitry may store, using an entry of the tracking slot circuitry, software-specified information for a set of graphics work, where the information includes: type of work, dependencies on other sets of graphics work, and location of data for the set of graphics work. The slot manager circuitry may prefetch, from the location and prior to allocating shader core resources for the set of graphics work, configuration register data for the set of graphics work. Control circuitry may program configuration registers for the set of graphics work using the prefetched data and initiate processing of the set of graphics work by the graphics processor circuitry according to the dependencies. Disclosed techniques may reduce kick-to-kick transition time, in some embodiments.

Kickslot manager circuitry for graphics processors

Techniques are disclosed relating to eviction control for cache lines that store register data. In some embodiments, memory hierarchy circuitry is configured to provide memory backing for register operand data in one or more cache circuits. Lock circuitry may control a first set of lock indicators for a set of registers for a first thread, including to assert one or more lock indicators for registers that are indicated, by decode circuitry, as being utilized by decoded instructions of the first thread. The lock circuitry may preserve register operand data in the one or more cache circuits, including to prevent eviction of a given cache line from a cache circuit based on an asserted lock indicator. The lock circuitry may clear the first set of lock indicators in response to a reset event. Disclosed techniques may advantageously retain relevant register information in the cache with limited control circuit area.

Cache control to preserve register data

Disclosed techniques relate to forming single-instruction multiple-data (SIMD) groups during ray intersection traversal. In particular, ray intersection circuitry may include dedicated circuitry configured to traverse an acceleration data structure, but may dynamically form a SIMD group to transform ray coordinates when traversing from one level of the data structure to another. This may allow shader processors to execute the SIMD group to perform the transformation. Disclosed techniques may facilitate instancing of graphics models within the acceleration data structure.

SIMD group formation techniques during ray intersection traversal

Techniques are disclosed relating to a shared control bus for communicating between primary control circuitry and multiple distributed graphics processor units. In some embodiments, a set of multiple processor units includes first and second graphics processors, where the first and second graphics processors are coupled to access graphics data via respective memory interfaces. A shared workload distribution bus is used to transmit control data that specifies graphics work distribution to the multiple graphics processing units. The shared workload distribution bus may be arranged in a chain topology, e.g., to connect the workload distribution circuitry to the first graphics processor and connect the first graphics processor to the second graphics processor such that the workload distribution circuitry communicates with the second graphics processor via the shared workload distribution bus connection to the first graphics processor. Disclosed techniques may facilitate graphics work distribution for a scalable number of processors.

United states graphics processor techniques with split between workload distribution control data on shared control bus and corresponding graphics data on memory interfaces

In an embodiment, a system on a chip (SOC) comprises a semiconductor die on which circuitry is formed, wherein the circuitry comprises a plurality of agents and a plurality of network switches coupled to the plurality of agents. The plurality of network switches are interconnected to form a plurality of physical and logically independent networks. A first network of the plurality of physically and logically independent networks is constructed according to a first topology and a second network of the plurality of physically and logically independent networks is constructed according to a second topology that is different from the first topology. For example, the first topology may a ring topology and the second topology may be a mesh topology. In an embodiment, coherency may be enforced on the first network and the second network may be a relaxed order network.

Multiple independent on-chip interconnect

Disclosed techniques relate to memory space management for graphics processing. In some embodiments, first and second graphics cores are configured to execute instructions for multiple threadgroups. In some embodiments, the threads groups include a first threadgroup with multiple single-instruction multiple-data (SIMD) groups configured to execute a first shader program and a second threadgroup with multiple SIMD groups configured to execute a second, different shader program. Control circuitry may be configured to provide access to data stored in memory circuitry according to a shader memory space. The shader memory space may be accessible to threadgroups executed by the first graphics shader core, including the first and second threadgroups, but is not accessible to threadgroups executed by the second graphics shader core. Disclosed techniques may reduce latency, increase bandwidth available to the shader, reduce coherency cost, or any combination thereof.

Graphics memory space for shader core

Disclosed techniques relate to primitive testing associated with ray intersection processing for ray tracing. In some embodiments, shader circuitry executes a first SIMD group that includes a ray intersect instruction for a set of rays. Ray intersect circuitry traverses, in response to the ray intersect instruction, multiple nodes in a spatially organized acceleration data structure (ADS). In response to reaching a node of the ADS that indicates one or more primitives, the apparatus forms a second SIMD group that executes one or more instructions to determine whether a set of rays that have reached the node intersect the one or more primitives. The shader circuitry may execute the first SIMD group to shade one or more primitives that are indicated as intersected based on results of execution of the second SIMD group. Thus, disclosed techniques may use both dedicated ray intersect circuitry and dynamically formed SIMD groups executed by shader processors to detect ray intersection.

SIMD group formation techniques for primitive testing associated with ray intersect traversal

Disclosed techniques relate to grouping rays during traversal of a spatially-organized acceleration data structure (e.g., a bounding volume hierarchy) for ray intersection processing. The grouping may provide temporal locality for accesses to bounding region data. In some embodiments, ray intersect circuitry is configured to group rays based on the node of the data structure that they target next. The ray intersect circuitry may select one or more groups of rays for issuance each clock cycle, e.g., to bounding region test circuitry.

Grouping techniques for ray intersection traversal

Disclosed techniques relate to ray intersection processing for ray tracing. In some embodiments, ray intersection circuitry traverses a spatially organized acceleration data structure and includes bounding region circuitry configured to test, in parallel, whether a ray intersects multiple different bounding regions indicated by a node of the data structure. Shader circuitry may execute a ray intersect instruction to invoke traversal by the ray intersect circuitry and the traversal may generate intersection results. The shader circuitry may shade intersected primitives based on the intersection results. Disclosed techniques that share processing between intersection circuitry and shader processors may improve performance, reduce power consumption, or both, relative to traditional techniques.

Ray intersect circuitry with parallel ray testing

Techniques are disclosed relating to filtering cache accesses. In some embodiments, a control unit is configured to, in response to a request to process a set of data, determine a size of a portion of the set of data to be handled using a cache. In some embodiments, the control unit is configured to determine filtering parameters indicative of a set of addresses corresponding to the determined size. In some embodiments, the control unit is configured to process one or more access requests for the set of data based on the determined filter parameters, including: using the cache to process one or more access requests having addresses in the set of addresses and bypassing the cache to access a backing memory directly, for access requests having addresses that are not in the set of addresses. The disclosed techniques may reduce average memory bandwidth or peak memory bandwidth.

Cache filtering

Techniques are disclosed relating to rendering graphics objects. In some embodiments, a graphics unit is configured to transform graphics objects from a virtual space into a second space according to different transformation parameters for different portions of the second space. This may result in sampling different portions of the virtual space at different sample rates, which may reduce the number of samples required in various stages of the rendering process. In the disclosed techniques, transformation may occur prior to rasterization and shading, which may further reduce computation and power consumption in a graphics unit, improve image quality as displayed to a user, and/or reduce bandwidth usage or latency of video content on a network. In some embodiments, a transformed image may be viewed through a distortion-compensating lens or resampled prior to display.

Growing community of inventors

St. Albans, United Kingdom

Jonathan M Redshaw