Intel Patent | Real-time temporally consistent object segmented style transfer in media and gaming

Patent: Real-time temporally consistent object segmented style transfer in media and gaming

Patent PDF: 加入映维网会员获取

Publication Number: 20220383580

Publication Date: 20221201

Assignee: Intel Corporation (Santa Clara, Ca, Us)

Abstract

One embodiment provides a method comprising, at a runtime library executed by a processor of a data processing system, receiving an input frame having objects to be stylized via a style transfer network associated with the runtime library, wherein the style transfer network is a neural network model trained to apply one or more visual styles to an input frame, performing instance segmentation on the input frame to generate one or more instance masks to identify one or more objects to be stylized, generating one or more stylized frames for each style to transfer to the input frame, and merging, via the one or more instance masks, stylized objects from one or more stylized frames with un-stylized content from the input frame to generate an output frame with per-instance stylization.

Claims

What is claimed is:

Description

CROSS-REFERENCE

The application claims priority to India Provisional Application Number 202141023763, filed May 28, 2021, which is hereby incorporated by reference.

FIELD

Embodiments described herein generally relate to computers. More particularly, embodiments are described for facilitating real-time temporally consistent style transfer of artistic images onto media and gaming.

BACKGROUND OF THE DISCLOSURE

Filtering techniques for images and videos are standard in the film, broadcast, social media, and many other industries. The drawback of these techniques is that they are not able to capture the high-level features of the styles to transform the imagery appropriately. Such techniques also lack temporal stability, thus the output that is generated is not temporally consistent across multiple frames. While techniques have been published to use machine learning models to transfer artistic styles from a style image into a target image, those techniques are computationally expensive and either cannot be performed in real time, lack temporal stability, or produce output images having visual artifacts. Furthermore, there are not any known techniques for generating segmentation masks for object segmented stylization using output from a 3D game engine.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention is illustrated by way of example and not limitation in the figures of the accompanying drawings in which like references indicate similar elements, and in which:

FIG. 1 is a block diagram of a processing system, according to an embodiment;

FIG. 2A-2D illustrate computing systems and graphics processors provided by embodiments described herein;

FIG. 3A-3C illustrate block diagrams of additional graphics processor and compute accelerator architectures provided by embodiments;

FIG. 4 is a block diagram of a graphics processing engine of a graphics processor in accordance with some embodiments;

FIG. 5A-5B illustrate thread execution logic including an array of processing elements employed in a graphics processor core according to embodiments;

FIG. 6 illustrates an additional execution unit, according to an embodiment;

FIG. 7 is a block diagram illustrating a graphics processor instruction formats according to some embodiments;

FIG. 8 is a block diagram of a graphics processor according to another embodiment;

FIG. 9A-9B illustrate a graphics processor command format and command sequence, according to some embodiments;

FIG. 10 illustrates exemplary graphics software architecture for a data processing system according to some embodiments;

FIG. 11A-11D illustrate an integrated circuit package assembly, according to an embodiment;

FIG. 12 is a block diagram illustrating an exemplary system on a chip integrated circuit, according to an embodiment;

FIG. 13A-13B is a block diagram illustrating an additional exemplary graphics processor;

FIG. 14 illustrates a machine learning software stack, according to an embodiment;

FIG. 15A-15B illustrate layers of exemplary deep neural networks;

FIG. 16 illustrates an exemplary recurrent neural network;

FIG. 17 illustrates training and deployment of a deep neural network;

FIG. 18 is a block diagram illustrating distributed learning;

FIG. 19 illustrates a data processing system, according to an embodiment;

FIG. 20 illustrates a style transfer network, according to an embodiment, according to an embodiment;

FIG. 21 illustrates style transfer for video media frames and 3D game scenes, according to an embodiment;

FIG. 22A-22B illustrate embodiments of a ground truth rendered frame and object-based segmentation;

FIG. 23 illustrates style transfer via the use of object masks derived from game engine data, according to an embodiment;

FIG. 24 illustrates one embodiment of a flow for localizing a primary object based on heuristics;

FIG. 25 illustrates an operational flow for parallel transfer of multiple styles within a media frame, according to an embodiment;

FIG. 26 illustrates a data processing system for game streaming and/or cloud gaming system, according to various embodiments;

FIG. 27 illustrates an exemplary image for which per-instance stylization is applied via a cloud compute platform, according to an embodiment;

FIG. 28 illustrates a method of performing per-instance stylization, according to embodiments;

FIG. 29 is a block diagram of a data processing system, according to an embodiment; and

FIG. 30 is a block diagram of a computing device including a graphics processor, according to an embodiment.

DETAILED DESCRIPTION

A graphics processing unit (GPU) is communicatively coupled to host/processor cores to accelerate, for example, graphics operations, machine-learning operations, pattern analysis operations, and/or various general-purpose GPU (GPGPU) functions. The GPU may be communicatively coupled to the host processor/cores over a bus or another interconnect (e.g., a high-speed interconnect such as PCIe or NVLink). Alternatively, the GPU may be integrated on the same package or chip as the cores and communicatively coupled to the cores over an internal processor bus/interconnect (i.e., internal to the package or chip). Regardless of the manner in which the GPU is connected, the processor cores may allocate work to the GPU in the form of sequences of commands/instructions contained in a work descriptor. The GPU then uses dedicated circuitry/logic for efficiently processing these commands/instructions.

Current parallel graphics data processing includes systems and methods developed to perform specific operations on graphics data such as, for example, linear interpolation, tessellation, rasterization, texture mapping, depth testing, etc. Traditionally, graphics processors used fixed function computational units to process graphics data. However, more recently, portions of graphics processors have been made programmable, enabling such processors to support a wider variety of operations for processing vertex and fragment data.

In the following description, numerous specific details are set forth. However, embodiments, as described herein, may be practiced without these specific details. In other instances, well-known circuits, structures and techniques have not been shown in detail in order not to obscure the understanding of this description.

Embodiments provide for an algorithm for real-time temporally consistent style transfer of artistic images onto media and gaming. An approach for semantically stylizing the localized parts of a video with different styles is extended. Previous style transfer networks are enhanced by addition of an alternate normalization layer and a corresponding activation function. In the gaming scenarios, the approach can help to stylize the character or specific parts or the full frame of the scene. For media streams, the segmentation is generated using a segmentation algorithm, whereas, the object level mask from the rendering engine is utilized in the gaming scenario. We also explore heuristics on localizing the character for stylizing a small region (windowing) instead of the complete frame.

It is contemplated that terms like “request”, “query”, “job”, “work”, “work item”, and “workload” may be referenced interchangeably throughout this document. Similarly, an “application” or “agent” may refer to or include a computer program, a software application, a game, a workstation application, etc., offered through an application programming interface (API), such as a free rendering API, such as Open Graphics Library (OpenGL®), Open Computing Language (OpenCL®), CUDA®, DirectX® 11, DirectX® 12, etc., where “dispatch” may be interchangeably referred to as “work unit” or “draw” and similarly, “application” may be interchangeably referred to as “workflow” or simply “agent”. For example, a workload, such as that of a three-dimensional (3D) game, may include and issue any number and type of “frames” where each frame may represent an image (e.g., sailboat, human face). Further, each frame may include and offer any number and type of work units, where each work unit may represent a part (e.g., mast of sailboat, forehead of human face) of the image (e.g., sailboat, human face) represented by its corresponding frame. However, for the sake of consistency, each item may be referenced by a single term (e.g., “dispatch”, “agent”, etc.) throughout this document.

In some embodiments, terms like “display screen” and “display surface” may be used interchangeably referring to the visible portion of a display device while the rest of the display device may be embedded into a computing device, such as a smartphone, a wearable device, etc. It is contemplated and to be noted that embodiments are not limited to any particular computing device, software application, hardware component, display device, display screen or surface, protocol, standard, etc. For example, embodiments may be applied to and used with any number and type of real-time applications on any number and type of computers, such as desktops, laptops, tablet computers, smartphones, head-mounted displays and other wearable devices, and/or the like. Further, for example, rendering scenarios for efficient performance using this novel technique may range from simple scenarios, such as desktop compositing, to complex scenarios, such as 3D games, augmented reality applications, etc.

System Overview

FIG. 1 is a block diagram of a processing system 100, according to an embodiment. Processing system 100 may be used in a single processor desktop system, a multiprocessor workstation system, or a server system having a large number of processors 102 or processor cores 107. In one embodiment, the processing system 100 is a processing platform incorporated within a system-on-a-chip (SoC) integrated circuit for use in mobile, handheld, or embedded devices such as within Internet-of-things (IoT) devices with wired or wireless connectivity to a local or wide area network.

In one embodiment, processing system 100 can include, couple with, or be integrated within: a server-based gaming platform; a game console, including a game and media console; a mobile gaming console, a handheld game console, or an online game console. In some embodiments the processing system 100 is part of a mobile phone, smart phone, tablet computing device or mobile Internet-connected device such as a laptop with low internal storage capacity. Processing system 100 can also include, couple with, or be integrated within: a wearable device, such as a smart watch wearable device; smart eyewear or clothing enhanced with augmented reality (AR) or virtual reality (VR) features to provide visual, audio or tactile outputs to supplement real world visual, audio or tactile experiences or otherwise provide text, audio, graphics, video, holographic images or video, or tactile feedback; other augmented reality (AR) device; or other virtual reality (VR) device. In some embodiments, the processing system 100 includes or is part of a television or set top box device. In one embodiment, processing system 100 can include, couple with, or be integrated within a self-driving vehicle such as a bus, tractor trailer, car, motor or electric power cycle, plane or glider (or any combination thereof). The self-driving vehicle may use processing system 100 to process the environment sensed around the vehicle.

In some embodiments, the one or more processors 102 each include one or more processor cores 107 to process instructions which, when executed, perform operations for system or user software. In some embodiments, at least one of the one or more processor cores 107 is configured to process a specific instruction set 109. In some embodiments, instruction set 109 may facilitate Complex Instruction Set Computing (CISC), Reduced Instruction Set Computing (RISC), or computing via a Very Long Instruction Word (VLIW). One or more processor cores 107 may process a different instruction set 109, which may include instructions to facilitate the emulation of other instruction sets. Processor core 107 may also include other processing devices, such as a Digital Signal Processor (DSP).

In some embodiments, the processor 102 includes cache memory 104. Depending on the architecture, the processor 102 can have a single internal cache or multiple levels of internal cache. In some embodiments, the cache memory is shared among various components of the processor 102. In some embodiments, the processor 102 also uses an external cache (e.g., a Level-3 (L3) cache or Last Level Cache (LLC)) (not shown), which may be shared among processor cores 107 using known cache coherency techniques. A register file 106 can be additionally included in processor 102 and may include different types of registers for storing different types of data (e.g., integer registers, floating point registers, status registers, and an instruction pointer register). Some registers may be general-purpose registers, while other registers may be specific to the design of the processor 102.

In some embodiments, one or more processor(s) 102 are coupled with one or more interface bus(es) 110 to transmit communication signals such as address, data, or control signals between processor 102 and other components in the processing system 100. The interface bus 110, in one embodiment, can be a processor bus, such as a version of the Direct Media Interface (DMI) bus. However, processor busses are not limited to the DMI bus, and may include one or more Peripheral Component Interconnect buses (e.g., PCI, PCI express), memory busses, or other types of interface busses. In one embodiment the processor(s) 102 include an integrated memory controller 116 and a platform controller hub 130. The memory controller 116 facilitates communication between a memory device and other components of the processing system 100, while the platform controller hub (PCH) 130 provides connections to I/O devices via a local I/O bus.

The memory device 120 can be a dynamic random-access memory (DRAM) device, a static random-access memory (SRAM) device, flash memory device, phase-change memory device, or some other memory device having suitable performance to serve as process memory. In one embodiment the memory device 120 can operate as system memory for the processing system 100, to store data 122 and instructions 121 for use when the one or more processors 102 executes an application or process. Memory controller 116 also couples with an optional external graphics processor 118, which may communicate with the one or more graphics processors 108 in processors 102 to perform graphics and media operations. In some embodiments, graphics, media, and or compute operations may be assisted by an accelerator 112 which is a coprocessor that can be configured to perform a specialized set of graphics, media, or compute operations. For example, in one embodiment the accelerator 112 is a matrix multiplication accelerator used to optimize machine learning or compute operations. In one embodiment the accelerator 112 is a ray-tracing accelerator that can be used to perform ray-tracing operations in concert with the graphics processor 108. In one embodiment, an external accelerator 119 may be used in place of or in concert with the accelerator 112.

In some embodiments a display device 111 can connect to the processor(s) 102. The display device 111 can be one or more of an internal display device, as in a mobile electronic device or a laptop device or an external display device attached via a display interface (e.g., DisplayPort, etc.). In one embodiment the display device 111 can be a head mounted display (HMD) such as a stereoscopic display device for use in virtual reality (VR) applications or augmented reality (AR) applications.

In some embodiments the platform controller hub 130 enables peripherals to connect to memory device 120 and processor 102 via a high-speed I/O bus. The I/O peripherals include, but are not limited to, an audio controller 146, a network controller 134, a firmware interface 128, a wireless transceiver 126, touch sensors 125, a data storage device 124 (e.g., non-volatile memory, volatile memory, hard disk drive, flash memory, NAND, 3D NAND, 3D XPoint, etc.). The data storage device 124 can connect via a storage interface (e.g., SATA) or via a peripheral bus, such as a Peripheral Component Interconnect bus (e.g., PCI, PCI express). The touch sensors 125 can include touch screen sensors, pressure sensors, or fingerprint sensors. The wireless transceiver 126 can be a Wi-Fi transceiver, a Bluetooth transceiver, or a mobile network transceiver such as a 3G, 4G, 5G, or Long-Term Evolution (LTE) transceiver. The firmware interface 128 enables communication with system firmware, and can be, for example, a unified extensible firmware interface (UEFI). The network controller 134 can enable a network connection to a wired network. In some embodiments, a high-performance network controller (not shown) couples with the interface bus 110. The audio controller 146, in one embodiment, is a multi-channel high definition audio controller. In one embodiment the processing system 100 includes an optional legacy I/O controller 140 for coupling legacy (e.g., Personal System 2 (PS/2)) devices to the system. The platform controller hub 130 can also connect to one or more Universal Serial Bus (USB) controllers 142 connect input devices, such as keyboard and mouse 143 combinations, a camera 144, or other USB input devices.

It will be appreciated that the processing system 100 shown is exemplary and not limiting, as other types of data processing systems that are differently configured may also be used. For example, an instance of the memory controller 116 and platform controller hub 130 may be integrated into a discreet external graphics processor, such as the external graphics processor 118. In one embodiment the platform controller hub 130 and/or memory controller 116 may be external to the one or more processor(s) 102. For example, the processing system 100 can include an external memory controller 116 and platform controller hub 130, which may be configured as a memory controller hub and peripheral controller hub within a system chipset that is in communication with the processor(s) 102.

For example, circuit boards (“sleds”) can be used on which components such as CPUs, memory, and other components are placed are designed for increased thermal performance. In some examples, processing components such as the processors are located on a top side of a sled while near memory, such as DIMMs, are located on a bottom side of the sled. As a result of the enhanced airflow provided by this design, the components may operate at higher frequencies and power levels than in typical systems, thereby increasing performance. Furthermore, the sleds are configured to blindly mate with power and data communication cables in a rack, thereby enhancing their ability to be quickly removed, upgraded, reinstalled, and/or replaced. Similarly, individual components located on the sleds, such as processors, accelerators, memory, and data storage drives, are configured to be easily upgraded due to their increased spacing from each other. In the illustrative embodiment, the components additionally include hardware attestation features to prove their authenticity.

A data center can utilize a single network architecture (“fabric”) that supports multiple other network architectures including Ethernet and Omni-Path. The sleds can be coupled to switches via optical fibers, which provide higher bandwidth and lower latency than typical twisted pair cabling (e.g., Category 5, Category 5e, Category 6, etc.). Due to the high bandwidth, low latency interconnections and network architecture, the data center may, in use, pool resources, such as memory, accelerators (e.g., GPUs, graphics accelerators, FPGAs, ASICs, neural network and/or artificial intelligence accelerators, etc.), and data storage drives that are physically disaggregated, and provide them to compute resources (e.g., processors) on an as needed basis, enabling the compute resources to access the pooled resources as if they were local.

A power supply or source can provide voltage and/or current to processing system 100 or any component or system described herein. In one example, the power supply includes an AC to DC (alternating current to direct current) adapter to plug into a wall outlet. Such AC power can be renewable energy (e.g., solar power) power source. In one example, power source includes a DC power source, such as an external AC to DC converter. In one example, power source or power supply includes wireless charging hardware to charge via proximity to a charging field. In one example, power source can include an internal battery, alternating current supply, motion-based power supply, solar power supply, or fuel cell source.

FIG. 2A-2D illustrate computing systems and graphics processors provided by embodiments described herein. The elements of FIGS. 2A-2D having the same reference numbers (or names) as the elements of any other figure herein can operate or function in any manner similar to that described elsewhere herein, but are not limited to such.

FIG. 2A is a block diagram of an embodiment of a processor 200 having one or more processor cores 202A-202N, an integrated memory controller 214, and an integrated graphics processor 208. Processor 200 can include additional cores up to and including additional core 202N represented by the dashed lined boxes. Each of processor cores 202A-202N includes one or more internal cache units 204A-204N. In some embodiments each processor core also has access to one or more shared cached units 206. The internal cache units 204A-204N and shared cache units 206 represent a cache memory hierarchy within the processor 200. The cache memory hierarchy may include at least one level of instruction and data cache within each processor core and one or more levels of shared mid-level cache, such as a Level 2 (L2), Level 3 (L3), Level 4 (L4), or other levels of cache, where the highest level of cache before external memory is classified as the LLC. In some embodiments, cache coherency logic maintains coherency between the various cache units 206 and 204A-204N.

In some embodiments, processor 200 may also include a set of one or more bus controller units 216 and a system agent core 210. The one or more bus controller units 216 manage a set of peripheral buses, such as one or more PCI or PCI express busses. System agent core 210 provides management functionality for the various processor components. In some embodiments, system agent core 210 includes one or more integrated memory controllers 214 to manage access to various external memory devices (not shown).

In some embodiments, one or more of the processor cores 202A-202N include support for simultaneous multi-threading. In such embodiment, the system agent core 210 includes components for coordinating and operating cores 202A-202N during multi-threaded processing. System agent core 210 may additionally include a power control unit (PCU), which includes logic and components to regulate the power state of processor cores 202A-202N and graphics processor 208.

In some embodiments, processor 200 additionally includes graphics processor 208 to execute graphics processing operations. In some embodiments, the graphics processor 208 couples with the set of shared cache units 206, and the system agent core 210, including the one or more integrated memory controllers 214. In some embodiments, the system agent core 210 also includes a display controller 211 to drive graphics processor output to one or more coupled displays. In some embodiments, display controller 211 may also be a separate module coupled with the graphics processor via at least one interconnect, or may be integrated within the graphics processor 208.

In some embodiments, a ring-based interconnect 212 is used to couple the internal components of the processor 200. However, an alternative interconnect unit may be used, such as a point-to-point interconnect, a switched interconnect, or other techniques, including techniques well known in the art. In some embodiments, graphics processor 208 couples with the ring-based interconnect 212 via an I/O link 213.

The exemplary I/O link 213 represents at least one of multiple varieties of I/O interconnects, including an on package I/O interconnect which facilitates communication between various processor components and a high-performance embedded memory module 218, such as an eDRAM module. In some embodiments, each of the processor cores 202A-202N and graphics processor 208 can use embedded memory modules 218 as a shared Last Level Cache.

In some embodiments, processor cores 202A-202N are homogenous cores executing the same instruction set architecture. In another embodiment, processor cores 202A-202N are heterogeneous in terms of instruction set architecture (ISA), where one or more of processor cores 202A-202N execute a first instruction set, while at least one of the other cores executes a subset of the first instruction set or a different instruction set. In one embodiment, processor cores 202A-202N are heterogeneous in terms of microarchitecture, where one or more cores having a relatively higher power consumption couple with one or more power cores having a lower power consumption. In one embodiment, processor cores 202A-202N are heterogeneous in terms of computational capability. Additionally, processor 200 can be implemented on one or more chips or as an SoC integrated circuit having the illustrated components, in addition to other components.

FIG. 2B is a block diagram of hardware logic of a graphics processor core 219, according to some embodiments described herein. Elements of FIG. 2B having the same reference numbers (or names) as the elements of any other figure herein can operate or function in any manner similar to that described elsewhere herein, but are not limited to such. The graphics processor core 219, sometimes referred to as a core slice, can be one or multiple graphics cores within a modular graphics processor. The graphics processor core 219 is exemplary of one graphics core slice, and a graphics processor as described herein may include multiple graphics core slices based on target power and performance envelopes. Each graphics processor core 219 can include a fixed function block 230 coupled with multiple sub-cores 221A-221F, also referred to as sub-slices, that include modular blocks of general-purpose and fixed function logic.

In some embodiments, the fixed function block 230 includes a geometry/fixed function pipeline 231 that can be shared by all sub-cores in the graphics processor core 219, for example, in lower performance and/or lower power graphics processor implementations. In various embodiments, the geometry/fixed function pipeline 231 includes a 3D fixed function pipeline (e.g., 3D pipeline 312 as in FIG. 3A and FIG. 4, described below) a video front-end unit, a thread spawner and thread dispatcher, and a unified return buffer manager, which manages unified return buffers (e.g., unified return buffer 418 in FIG. 4, as described below).

In one embodiment the fixed function block 230 also includes a graphics SoC interface 232, a graphics microcontroller 233, and a media pipeline 234. The graphics SoC interface 232 provides an interface between the graphics processor core 219 and other processor cores within a system on a chip integrated circuit. The graphics microcontroller 233 is a programmable sub-processor that is configurable to manage various functions of the graphics processor core 219, including thread dispatch, scheduling, and pre-emption. The media pipeline 234 (e.g., media pipeline 316 of FIG. 3A and FIG. 4) includes logic to facilitate the decoding, encoding, pre-processing, and/or post-processing of multimedia data, including image and video data. The media pipeline 234 implement media operations via requests to compute or sampling logic within the sub-cores 221-221F.

In one embodiment the SoC interface 232 enables the graphics processor core 219 to communicate with general-purpose application processor cores (e.g., CPUs) and/or other components within an SoC, including memory hierarchy elements such as a shared last level cache memory, the system RAM, and/or embedded on-chip or on-package DRAM. The SoC interface 232 can also enable communication with fixed function devices within the SoC, such as camera imaging pipelines, and enables the use of and/or implements global memory atomics that may be shared between the graphics processor core 219 and CPUs within the SoC. The SoC interface 232 can also implement power management controls for the graphics processor core 219 and enable an interface between a clock domain of the graphics processor core 219 and other clock domains within the SoC. In one embodiment the SoC interface 232 enables receipt of command buffers from a command streamer and global thread dispatcher that are configured to provide commands and instructions to each of one or more graphics cores within a graphics processor. The commands and instructions can be dispatched to the media pipeline 234, when media operations are to be performed, or a geometry and fixed function pipeline (e.g., geometry and fixed function pipeline 231, geometry and fixed function pipeline 237) when graphics processing operations are to be performed.

The graphics microcontroller 233 can be configured to perform various scheduling and management tasks for the graphics processor core 219. In one embodiment the graphics microcontroller 233 can perform graphics and/or compute workload scheduling on the various graphics parallel engines within execution unit (EU) arrays 222A-222F, 224A-224F within the sub-cores 221A-221F. In this scheduling model, host software executing on a CPU core of an SoC including the graphics processor core 219 can submit workloads one of multiple graphic processor doorbells, which invokes a scheduling operation on the appropriate graphics engine. Scheduling operations include determining which workload to run next, submitting a workload to a command streamer, pre-empting existing workloads running on an engine, monitoring progress of a workload, and notifying host software when a workload is complete. In one embodiment the graphics microcontroller 233 can also facilitate low-power or idle states for the graphics processor core 219, providing the graphics processor core 219 with the ability to save and restore registers within the graphics processor core 219 across low-power state transitions independently from the operating system and/or graphics driver software on the system.

The graphics processor core 219 may have greater than or fewer than the illustrated sub-cores 221A-221F, up to N modular sub-cores. For each set of N sub-cores, the graphics processor core 219 can also include shared function logic 235, shared and/or cache memory 236, a geometry/fixed function pipeline 237, as well as additional fixed function logic 238 to accelerate various graphics and compute processing operations. The shared function logic 235 can include logic units associated with the shared function logic 420 of FIG. 4 (e.g., sampler, math, and/or inter-thread communication logic) that can be shared by each N sub-cores within the graphics processor core 219. The shared and/or cache memory 236 can be a last-level cache for the set of N sub-cores 221A-221F within the graphics processor core 219, and can also serve as shared memory that is accessible by multiple sub-cores. The geometry/fixed function pipeline 237 can be included instead of the geometry/fixed function pipeline 231 within the fixed function block 230 and can include the same or similar logic units.

In one embodiment the graphics processor core 219 includes additional fixed function logic 238 that can include various fixed function acceleration logic for use by the graphics processor core 219. In one embodiment the additional fixed function logic 238 includes an additional geometry pipeline for use in position only shading. In position-only shading, two geometry pipelines exist, the full geometry pipeline within the geometry/fixed function pipeline 238, 231, and a cull pipeline, which is an additional geometry pipeline which may be included within the additional fixed function logic 238. In one embodiment the cull pipeline is a trimmed down version of the full geometry pipeline. The full pipeline and the cull pipeline can execute different instances of the same application, each instance having a separate context. Position only shading can hide long cull runs of discarded triangles, enabling shading to be completed earlier in some instances. For example and in one embodiment the cull pipeline logic within the additional fixed function logic 238 can execute position shaders in parallel with the main application and generally generates critical results faster than the full pipeline, as the cull pipeline fetches and shades only the position attribute of the vertices, without performing rasterization and rendering of the pixels to the frame buffer. The cull pipeline can use the generated critical results to compute visibility information for all the triangles without regard to whether those triangles are culled. The full pipeline (which in this instance may be referred to as a replay pipeline) can consume the visibility information to skip the culled triangles to shade only the visible triangles that are finally passed to the rasterization phase.

In one embodiment the additional fixed function logic 238 can also include machine-learning acceleration logic, such as fixed function matrix multiplication logic, for implementations including optimizations for machine learning training or inferencing.

Within each graphics sub-core 221A-221F includes a set of execution resources that may be used to perform graphics, media, and compute operations in response to requests by graphics pipeline, media pipeline, or shader programs. The graphics sub-cores 221A-221F include multiple EU arrays 222A-222F, 224A-224F, thread dispatch and inter-thread communication (TD/IC) logic 223A-223F, a 3D (e.g., texture) sampler 225A-225F, a media sampler 226A-226F, a shader processor 227A-227F, and shared local memory (SLM) 228A-228F. The EU arrays 222A-222F, 224A-224F each include multiple execution units, which are general-purpose graphics processing units capable of performing floating-point and integer/fixed-point logic operations in service of a graphics, media, or compute operation, including graphics, media, or compute shader/GPGPU programs. The TD/IC logic 223A-223F performs local thread dispatch and thread control operations for the execution units within a sub-core and facilitate communication between threads executing on the execution units of the sub-core. The 3D sampler 225A-225F can read texture or other 3D graphics related data into memory. The 3D sampler can read texture data differently based on a configured sample state and the texture format associated with a given texture. The media sampler 226A-226F can perform similar read operations based on the type and format associated with media data. In one embodiment, each graphics sub-core 221A-221F can alternately include a unified 3D and media sampler. Threads executing on the execution units within each of the sub-cores 221A-221F can make use of shared local memory 228A-228F within each sub-core, to enable threads executing within a thread group to execute using a common pool of on-chip memory.

FIG. 2C illustrates a graphics processing unit (GPU) 239 that includes dedicated sets of graphics processing resources arranged into multi-core groups 240A-240N. The details of multi-core group 240A are illustrated. Multi-core groups 240B-240N may be equipped with the same or similar sets of graphics processing resources.

As illustrated, a multi-core group 240A may include a set of graphics cores 243, a set of tensor cores 244, and a set of ray tracing cores 245. A scheduler/dispatcher 241 schedules and dispatches the graphics threads for execution on the various cores 243, 244, 245. In one embodiment the tensor cores 244 are sparse tensor cores with hardware to enable multiplication operations having a zero value input to be bypassed.

A set of register files 242 can store operand values used by the cores 243, 244, 245 when executing the graphics threads. These may include, for example, integer registers for storing integer values, floating point registers for storing floating point values, vector registers for storing packed data elements (integer and/or floating point data elements) and tile registers for storing tensor/matrix values. In one embodiment, the tile registers are implemented as combined sets of vector registers.

One or more combined level 1 (L1) caches and shared memory units 247 store graphics data such as texture data, vertex data, pixel data, ray data, bounding volume data, etc., locally within each multi-core group 240A. One or more texture units 247 can also be used to perform texturing operations, such as texture mapping and sampling. A Level 2 (L2) cache 253 shared by all or a subset of the multi-core groups 240A-240N stores graphics data and/or instructions for multiple concurrent graphics threads. As illustrated, the L2 cache 253 may be shared across a plurality of multi-core groups 240A-240N. One or more memory controllers 248 couple the GPU 239 to a memory 249 which may be a system memory (e.g., DRAM) and/or a dedicated graphics memory (e.g., GDDR6 memory).

Input/output (I/O) circuitry 250 couples the GPU 239 to one or more I/O devices 252 such as digital signal processors (DSPs), network controllers, or user input devices. An on-chip interconnect may be used to couple the I/O devices 252 to the GPU 239 and memory 249. One or more I/O memory management units (IOMMUs) 251 of the I/O circuitry 250 couple the I/O devices 252 directly to the memory 249. In one embodiment, the IOMMU 251 manages multiple sets of page tables to map virtual addresses to physical addresses in memory 249. In this embodiment, the I/O devices 252, CPU(s) 246, and GPU 239 may share the same virtual address space.

In one implementation, the IOMMU 251 supports virtualization. In this case, it may manage a first set of page tables to map guest/graphics virtual addresses to guest/graphics physical addresses and a second set of page tables to map the guest/graphics physical addresses to system/host physical addresses (e.g., within memory 249). The base addresses of each of the first and second sets of page tables may be stored in control registers and swapped out on a context switch (e.g., so that the new context is provided with access to the relevant set of page tables). While not illustrated in FIG. 2C, each of the cores 243, 244, 245 and/or multi-core groups 240A-240N may include translation lookaside buffers (TLB s) to cache guest virtual to guest physical translations, guest physical to host physical translations, and guest virtual to host physical translations.

In one embodiment, the CPUs 246, GPU 239, and I/O devices 252 are integrated on a single semiconductor chip and/or chip package. The memory 249 may be integrated on the same chip or may be coupled to the memory controllers 248 via an off-chip interface. In one implementation, the memory 249 comprises GDDR6 memory which shares the same virtual address space as other physical system-level memories, although the underlying principles of the invention are not limited to this specific implementation.

In one embodiment, the tensor cores 244 include a plurality of execution units specifically designed to perform matrix operations, which are the fundamental compute operation used to perform deep learning operations. For example, simultaneous matrix multiplication operations may be used for neural network training and inferencing. The tensor cores 244 may perform matrix processing using a variety of operand precisions including single precision floating-point (e.g., 32 bits), half-precision floating point (e.g., 16 bits), integer words (16 bits), bytes (8 bits), and half-bytes (4 bits). In one embodiment, a neural network implementation extracts features of each rendered scene, potentially combining details from multiple frames, to construct a high-quality final image. p In deep learning implementations, parallel matrix multiplication work may be scheduled for execution on the tensor cores 244. The training of neural networks, in particular, requires a significant number matrix dot product operations. In order to process an inner-product formulation of an N×N×N matrix multiply, the tensor cores 244 may include at least N dot-product processing elements. Before the matrix multiply begins, one entire matrix is loaded into tile registers and at least one column of a second matrix is loaded each cycle for N cycles. Each cycle, there are N dot products that are processed.

Matrix elements may be stored at different precisions depending on the particular implementation, including 16-bit words, 8-bit bytes (e.g., INT8) and 4-bit half-bytes (e.g., INT4). Different precision modes may be specified for the tensor cores 244 to ensure that the most efficient precision is used for different workloads (e.g., such as inferencing workloads which can tolerate quantization to bytes and half-bytes).

In one embodiment, the ray tracing cores 245 accelerate ray tracing operations for both real-time ray tracing and non-real-time ray tracing implementations. In particular, the ray tracing cores 245 include ray traversal/intersection circuitry for performing ray traversal using bounding volume hierarchies (BVHs) and identifying intersections between rays and primitives enclosed within the BVH volumes. The ray tracing cores 245 may also include circuitry for performing depth testing and culling (e.g., using a Z buffer or similar arrangement). In one implementation, the ray tracing cores 245 perform traversal and intersection operations in concert with the image denoising techniques described herein, at least a portion of which may be executed on the tensor cores 244. For example, in one embodiment, the tensor cores 244 implement a deep learning neural network to perform denoising of frames generated by the ray tracing cores 245. However, the CPU(s) 246, graphics cores 243, and/or ray tracing cores 245 may also implement all or a portion of the denoising and/or deep learning algorithms.

In addition, as described above, a distributed approach to denoising may be employed in which the GPU 239 is in a computing device coupled to other computing devices over a network or high speed interconnect. In this embodiment, the interconnected computing devices share neural network learning/training data to improve the speed with which the overall system learns to perform denoising for different types of image frames and/or different graphics applications.

In one embodiment, the ray tracing cores 245 process all BVH traversal and ray-primitive intersections, saving the graphics cores 243 from being overloaded with thousands of instructions per ray. In one embodiment, each ray tracing core 245 includes a first set of specialized circuitry for performing bounding box tests (e.g., for traversal operations) and a second set of specialized circuitry for performing the ray-triangle intersection tests (e.g., intersecting rays which have been traversed). Thus, in one embodiment, the multi-core group 240A can simply launch a ray probe, and the ray tracing cores 245 independently perform ray traversal and intersection and return hit data (e.g., a hit, no hit, multiple hits, etc.) to the thread context. The other cores 243, 244 are freed to perform other graphics or compute work while the ray tracing cores 245 perform the traversal and intersection operations.

In one embodiment, each ray tracing core 245 includes a traversal unit to perform BVH testing operations and an intersection unit which performs ray-primitive intersection tests. The intersection unit generates a “hit”, “no hit”, or “multiple hit” response, which it provides to the appropriate thread. During the traversal and intersection operations, the execution resources of the other cores (e.g., graphics cores 243 and tensor cores 244) are freed to perform other forms of graphics work.

In one particular embodiment described below, a hybrid rasterization/ray tracing approach is used in which work is distributed between the graphics cores 243 and ray tracing cores 245.

In one embodiment, the ray tracing cores 245 (and/or other cores 243, 244) include hardware support for a ray tracing instruction set such as Microsoft's DirectX Ray Tracing (DXR) which includes a DispatchRays command, as well as ray-generation, closest-hit, any-hit, and miss shaders, which enable the assignment of unique sets of shaders and textures for each object. Another ray tracing platform which may be supported by the ray tracing cores 245, graphics cores 243 and tensor cores 244 is Vulkan 1.1.85. Note, however, that the underlying principles of the invention are not limited to any particular ray tracing ISA.

In general, the various cores 245, 244, 243 may support a ray tracing instruction set that includes instructions/functions for ray generation, closest hit, any hit, ray-primitive intersection, per-primitive and hierarchical bounding box construction, miss, visit, and exceptions. More specifically, one embodiment includes ray tracing instructions to perform the following functions:

Ray Generation—Ray generation instructions may be executed for each pixel, sample, or other user-defined work assignment.

Closest Hit—A closest hit instruction may be executed to locate the closest intersection point of a ray with primitives within a scene.

Any Hit—An any hit instruction identifies multiple intersections between a ray and primitives within a scene, potentially to identify a new closest intersection point.

Intersection—An intersection instruction performs a ray-primitive intersection test and outputs a result.

Per-primitive Bounding box Construction—This instruction builds a bounding box around a given primitive or group of primitives (e.g., when building a new BVH or other acceleration data structure).

Miss—Indicates that a ray misses all geometry within a scene, or specified region of a scene.

Visit—Indicates the children volumes a ray will traverse.

Exceptions—Includes various types of exception handlers (e.g., invoked for various error conditions).

In one embodiment the ray tracing cores 245 may be adapted to accelerate general-purpose compute operations that can be accelerated using computational techniques that are analogous to ray intersection tests. A compute framework can be provided that enables shader programs to be compiled into low level instructions and/or primitives that perform general-purpose compute operations via the ray tracing cores. Exemplary computational problems that can benefit from compute operations performed on the ray tracing cores 245 include computations involving beam, wave, ray, or particle propagation within a coordinate space. Interactions associated with that propagation can be computed relative to a geometry or mesh within the coordinate space. For example, computations associated with electromagnetic signal propagation through an environment can be accelerated via the use of instructions or primitives that are executed via the ray tracing cores. Diffraction and reflection of the signals by objects in the environment can be computed as direct ray-tracing analogies.

Ray tracing cores 245 can also be used to perform computations that are not directly analogous to ray tracing. For example, mesh projection, mesh refinement, and volume sampling computations can be accelerated using the ray tracing cores 245. Generic coordinate space calculations, such as nearest neighbor calculations can also be performed. For example, the set of points near a given point can be discovered by defining a bounding box in the coordinate space around the point. BVH and ray probe logic within the ray tracing cores 245 can then be used to determine the set of point intersections within the bounding box. The intersections constitute the origin point and the nearest neighbors to that origin point. Computations that are performed using the ray tracing cores 245 can be performed in parallel with computations performed on the graphics cores 243 and tensor cores 244. A shader compiler can be configured to compile a compute shader or other general-purpose graphics processing program into low level primitives that can be parallelized across the graphics cores 243, tensor cores 244, and ray tracing cores 245.

FIG. 2D is a block diagram of general purpose graphics processing unit (GPGPU) 270 that can be configured as a graphics processor and/or compute accelerator, according to embodiments described herein. The GPGPU 270 can interconnect with host processors (e.g., one or more CPU(s) 246) and memory 271, 272 via one or more system and/or memory busses. In one embodiment the memory 271 is system memory that may be shared with the one or more CPU(s) 246, while memory 272 is device memory that is dedicated to the GPGPU 270. In one embodiment, components within the GPGPU 270 and memory 272 may be mapped into memory addresses that are accessible to the one or more CPU(s) 246. Access to memory 271 and 272 may be facilitated via a memory controller 268. In one embodiment the memory controller 268 includes an internal direct memory access (DMA) controller 269 or can include logic to perform operations that would otherwise be performed by a DMA controller.

The GPGPU 270 includes multiple cache memories, including an L2 cache 253, L1 cache 254, an instruction cache 255, and shared memory 256, at least a portion of which may also be partitioned as a cache memory. The GPGPU 270 also includes multiple compute units 260A-260N. Each compute unit 260A-260N includes a set of vector registers 261, scalar registers 262, vector logic units 263, and scalar logic units 264. The compute units 260A-260N can also include local shared memory 265 and a program counter 266. The compute units 260A-260N can couple with a constant cache 267, which can be used to store constant data, which is data that will not change during the run of kernel or shader program that executes on the GPGPU 270. In one embodiment the constant cache 267 is a scalar data cache and cached data can be fetched directly into the scalar registers 262.

During operation, the one or more CPU(s) 246 can write commands into registers or memory in the GPGPU 270 that has been mapped into an accessible address space. The command processors 257 can read the commands from registers or memory and determine how those commands will be processed within the GPGPU 270. A thread dispatcher 258 can then be used to dispatch threads to the compute units 260A-260N to perform those commands. Each compute unit 260A-260N can execute threads independently of the other compute units. Additionally each compute unit 260A-260N can be independently configured for conditional computation and can conditionally output the results of computation to memory. The command processors 257 can interrupt the one or more CPU(s) 246 when the submitted commands are complete.

FIGS. 3A-3C illustrate block diagrams of additional graphics processor and compute accelerator architectures provided by embodiments described herein. The elements of FIGS. 3A-3C having the same reference numbers (or names) as the elements of any other figure herein can operate or function in any manner similar to that described elsewhere herein, but are not limited to such.

FIG. 3A is a block diagram of a graphics processor 300, which may be a discrete graphics processing unit, or may be a graphics processor integrated with a plurality of processing cores, or other semiconductor devices such as, but not limited to, memory devices or network interfaces. In some embodiments, the graphics processor communicates via a memory mapped I/O interface to registers on the graphics processor and with commands placed into the processor memory. In some embodiments, graphics processor 300 includes a memory interface 314 to access memory. Memory interface 314 can be an interface to local memory, one or more internal caches, one or more shared external caches, and/or to system memory.

In some embodiments, graphics processor 300 also includes a display controller 302 to drive display output data to a display device 318. Display controller 302 includes hardware for one or more overlay planes for the display and composition of multiple layers of video or user interface elements. The display device 318 can be an internal or external display device. In one embodiment the display device 318 is a head mounted display device, such as a virtual reality (VR) display device or an augmented reality (AR) display device. In some embodiments, graphics processor 300 includes a video codec engine 306 to encode, decode, or transcode media to, from, or between one or more media encoding formats, including, but not limited to Moving Picture Experts Group (MPEG) formats such as MPEG-2, Advanced Video Coding (AVC) formats such as H.264/MPEG-4 AVC, H.265/HEVC, Alliance for Open Media (AOMedia) VP8, VP9, as well as the Society of Motion Picture & Television Engineers (SMPTE) 421M/VC-1, and Joint Photographic Experts Group (JPEG) formats such as JPEG, and Motion JPEG (MJPEG) formats.

In some embodiments, graphics processor 300 includes a block image transfer (BLIT) engine 304 to perform two-dimensional (2D) rasterizer operations including, for example, bit-boundary block transfers. However, in one embodiment, 2D graphics operations are performed using one or more components of graphics processing engine (GPE) 310. In some embodiments, GPE 310 is a compute engine for performing graphics operations, including three-dimensional (3D) graphics operations and media operations.

In some embodiments, GPE 310 includes a 3D pipeline 312 for performing 3D operations, such as rendering three-dimensional images and scenes using processing functions that act upon 3D primitive shapes (e.g., rectangle, triangle, etc.). The 3D pipeline 312 includes programmable and fixed function elements that perform various tasks within the element and/or spawn execution threads to a 3D/Media subsystem 315. While 3D pipeline 312 can be used to perform media operations, an embodiment of GPE 310 also includes a media pipeline 316 that is specifically used to perform media operations, such as video post-processing and image enhancement.

In some embodiments, media pipeline 316 includes fixed function or programmable logic units to perform one or more specialized media operations, such as video decode acceleration, video de-interlacing, and video encode acceleration in place of, or on behalf of video codec engine 306. In some embodiments, media pipeline 316 additionally includes a thread spawning unit to spawn threads for execution on 3D/Media subsystem 315. The spawned threads perform computations for the media operations on one or more graphics execution units included in 3D/Media subsystem 315.

In some embodiments, 3D/Media subsystem 315 includes logic for executing threads spawned by 3D pipeline 312 and media pipeline 316. In one embodiment, the pipelines send thread execution requests to 3D/Media subsystem 315, which includes thread dispatch logic for arbitrating and dispatching the various requests to available thread execution resources. The execution resources include an array of graphics execution units to process the 3D and media threads. In some embodiments, 3D/Media subsystem 315 includes one or more internal caches for thread instructions and data. In some embodiments, the subsystem also includes shared memory, including registers and addressable memory, to share data between threads and to store output data.

FIG. 3B illustrates a graphics processor 320 having a tiled architecture, according to embodiments described herein. In one embodiment the graphics processor 320 includes a graphics processing engine cluster 322 having multiple instances of the graphics processing engine 310 of FIG. 3A within a graphics engine tile 310A-310D. Each graphics engine tile 310A-310D can be interconnected via a set of tile interconnects 323A-323F. Each graphics engine tile 310A-310D can also be connected to a memory module or memory device 326A-326D via memory interconnects 325A-325D. The memory devices 326A-326D can use any graphics memory technology. For example, the memory devices 326A-326D may be graphics double data rate (GDDR) memory. The memory devices 326A-326D, in one embodiment, are high-bandwidth memory (HBM) modules that can be on-die with their respective graphics engine tile 310A-310D. In one embodiment the memory devices 326A-326D are stacked memory devices that can be stacked on top of their respective graphics engine tile 310A-310D. In one embodiment, each graphics engine tile 310A-310D and associated memory 326A-326D reside on separate chiplets, which are bonded to a base die or base substrate, as described on further detail in FIGS. 11B-11D.

The graphics processor 320 may be configured with a non-uniform memory access (NUMA) system in which memory devices 326A-326D are coupled with associated graphics engine tiles 310A-310D. A given memory device may be accessed by graphics engine tiles other than the tile to which it is directly connected. However, access latency to the memory devices 326A-326D may be lowest when accessing a local tile. In one embodiment, a cache coherent NUMA (ccNUMA) system is enabled that uses the tile interconnects 323A-323F to enable communication between cache controllers within the graphics engine tiles 310A-310D to maintain a consistent memory image when more than one cache stores the same memory location.

The graphics processing engine cluster 322 can connect with an on-chip or on-package fabric interconnect 324. In one embodiment the fabric interconnect 324 includes a network processor, network on a chip (NoC), or another switching processor to enable the fabric interconnect 324 to act as a packet switched fabric interconnect that switches data packets between components of the graphics processor 320. The fabric interconnect 324 can enable communication between graphics engine tiles 310A-310D and components such as the video codec engine 306 and one or more copy engines 304. The copy engines 304 can be used to move data out of, into, and between the memory devices 326A-326D and memory that is external to the graphics processor 320 (e.g., system memory). The fabric interconnect 324 can also couple with one or more of the tile interconnects 323A-323F to facilitate or enhance the interconnection between the graphics engine tiles 310A-310D. The fabric interconnect 324 is also configurable to interconnect multiple instances of the graphics processor 320 (e.g., via the host interface 328), enabling tile-to-tile communication between graphics engine tiles 310A-310D of multiple GPUs. In one embodiment, the graphics engine tiles 310A-310D of multiple GPUs can be presented to a host system as a single logical device.

The graphics processor 320 may optionally include a display controller 302 to enable a connection with the display device 318. The graphics processor may also be configured as a graphics or compute accelerator. In the accelerator configuration, the display controller 302 and display device 318 may be omitted.

The graphics processor 320 can connect to a host system via a host interface 328. The host interface 328 can enable communication between the graphics processor 320, system memory, and/or other system components. The host interface 328 can be, for example a PCI express bus or another type of host system interface. For example, the host interface 328 may be an NVLink or NVSwitch interface. The host interface 328 and fabric interconnect 324 can cooperate to enable multiple instances of the graphics processor 320 to act as single logical device. Cooperation between the host interface 328 and fabric interconnect 324 can also enable the individual graphics engine tiles 310A-310D to be presented to the host system as distinct logical graphics devices.

FIG. 3C illustrates a compute accelerator 330, according to embodiments described herein. The compute accelerator 330 can include architectural similarities with the graphics processor 320 of FIG. 3B and is optimized for compute acceleration. A compute engine cluster 332 can include a set of compute engine tiles 340A-340D that include execution logic that is optimized for parallel or vector-based general-purpose compute operations. In some embodiments, the compute engine tiles 340A-340D do not include fixed function graphics processing logic, although in one embodiment one or more of the compute engine tiles 340A-340D can include logic to perform media acceleration. The compute engine tiles 340A-340D can connect to memory 326A-326D via memory interconnects 325A-325D. The memory 326A-326D and memory interconnects 325A-325D may be similar technology as in graphics processor 320, or can be different. The graphics compute engine tiles 340A-340D can also be interconnected via a set of tile interconnects 323A-323F and may be connected with and/or interconnected by a fabric interconnect 324. Cross-tile communications can be facilitated via the fabric interconnect 324. The fabric interconnect 324 (e.g., via the host interface 328) can also facilitate communication between compute engine tiles 340A-340D of multiple instances of the compute accelerator 330. In one embodiment the compute accelerator 330 includes a large L3 cache 336 that can be configured as a device-wide cache. The compute accelerator 330 can also connect to a host processor and memory via a host interface 328 in a similar manner as the graphics processor 320 of FIG. 3B.

The compute accelerator 330 can also include an integrated network interface 342. In one embodiment the network interface 342 includes a network processor and controller logic that enables the compute engine cluster 332 to communicate over a physical layer interconnect 344 without requiring data to traverse memory of a host system. In one embodiment, one of the compute engine tiles 340A-340D is replaced by network processor logic and data to be transmitted or received via the physical layer interconnect 344 may be transmitted directly to or from memory 326A-326D. Multiple instances of the compute accelerator 330 may be joined via the physical layer interconnect 344 into a single logical device. Alternatively, the various compute engine tiles 340A-340D may be presented as distinct network accessible compute accelerator devices.

Graphics Processing Engine

FIG. 4 is a block diagram of a graphics processing engine 410 of a graphics processor in accordance with some embodiments. In one embodiment, the graphics processing engine (GPE) 410 is a version of the GPE 310 shown in FIG. 3A, and may also represent a graphics engine tile 310A-310D of FIG. 3B. Elements of FIG. 4 having the same reference numbers (or names) as the elements of any other figure herein can operate or function in any manner similar to that described elsewhere herein, but are not limited to such. For example, the 3D pipeline 312 and media pipeline 316 of FIG. 3A are illustrated. The media pipeline 316 is optional in some embodiments of the GPE 410 and may not be explicitly included within the GPE 410. For example and in at least one embodiment, a separate media and/or image processor is coupled to the GPE 410.

In some embodiments, GPE 410 couples with or includes a command streamer 403, which provides a command stream to the 3D pipeline 312 and/or media pipelines 316. Alternatively or additionally, the command streamer 403 may be directly coupled to a unified return buffer 418. The unified return buffer 418 may be communicatively coupled to a graphics core array 414. In some embodiments, command streamer 403 is coupled with memory, which can be system memory, or one or more of internal cache memory and shared cache memory. In some embodiments, command streamer 403 receives commands from the memory and sends the commands to 3D pipeline 312 and/or media pipeline 316. The commands are directives fetched from a ring buffer, which stores commands for the 3D pipeline 312 and media pipeline 316. In one embodiment, the ring buffer can additionally include batch command buffers storing batches of multiple commands. The commands for the 3D pipeline 312 can also include references to data stored in memory, such as but not limited to vertex and geometry data for the 3D pipeline 312 and/or image data and memory objects for the media pipeline 316. The 3D pipeline 312 and media pipeline 316 process the commands and data by performing operations via logic within the respective pipelines or by dispatching one or more execution threads to a graphics core array 414. In one embodiment the graphics core array 414 include one or more blocks of graphics cores (e.g., graphics core(s) 415A, graphics core(s) 415B), each block including one or more graphics cores. Each graphics core includes a set of graphics execution resources that includes general-purpose and graphics specific execution logic to perform graphics and compute operations, as well as fixed function texture processing and/or machine learning and artificial intelligence acceleration logic.

In various embodiments the 3D pipeline 312 can include fixed function and programmable logic to process one or more shader programs, such as vertex shaders, geometry shaders, pixel shaders, fragment shaders, compute shaders, or other shader and/or GPGPU programs, by processing the instructions and dispatching execution threads to the graphics core array 414. The graphics core array 414 provides a unified block of execution resources for use in processing these shader programs. Multi-purpose execution logic (e.g., execution units) within the graphics core(s) 415A-414B of the graphics core array 414 includes support for various 3D API shader languages and can execute multiple simultaneous execution threads associated with multiple shaders.

In some embodiments, the graphics core array 414 includes execution logic to perform media functions, such as video and/or image processing. In one embodiment, the execution units include general-purpose logic that is programmable to perform parallel general-purpose computational operations, in addition to graphics processing operations. The general-purpose logic can perform processing operations in parallel or in conjunction with general-purpose logic within the processor core(s) 107 of FIG. 1 or core 202A-202N as in FIG. 2A.

Output data generated by threads executing on the graphics core array 414 can output data to memory in a unified return buffer (URB) 418. The URB 418 can store data for multiple threads. In some embodiments the URB 418 may be used to send data between different threads executing on the graphics core array 414. In some embodiments the URB 418 may additionally be used for synchronization between threads on the graphics core array and fixed function logic within the shared function logic 420.

In some embodiments, graphics core array 414 is scalable, such that the array includes a variable number of graphics cores, each having a variable number of execution units based on the target power and performance level of GPE 410. In one embodiment the execution resources are dynamically scalable, such that execution resources may be enabled or disabled as needed.

The graphics core array 414 couples with shared function logic 420 that includes multiple resources that are shared between the graphics cores in the graphics core array. The shared functions within the shared function logic 420 are hardware logic units that provide specialized supplemental functionality to the graphics core array 414. In various embodiments, shared function logic 420 includes but is not limited to sampler logic 421, math logic 422, and inter-thread communication (ITC) 423 logic. Additionally, some embodiments implement one or more cache(s) 425 within the shared function logic 420.

A shared function is implemented at least in a case where the demand for a given specialized function is insufficient for inclusion within the graphics core array 414. Instead, a single instantiation of that specialized function is implemented as a stand-alone entity in the shared function logic 420 and shared among the execution resources within the graphics core array 414. The precise set of functions that are shared between the graphics core array 414 and included within the graphics core array 414 varies across embodiments. In some embodiments, specific shared functions within the shared function logic 420 that are used extensively by the graphics core array 414 may be included within shared function logic 416 within the graphics core array 414. In various embodiments, the shared function logic 416 within the graphics core array 414 can include some or all logic within the shared function logic 420. In one embodiment, all logic elements within the shared function logic 420 may be duplicated within the shared function logic 416 of the graphics core array 414. In one embodiment the shared function logic 420 is excluded in favor of the shared function logic 416 within the graphics core array 414.

Execution Units

FIGS. 5A-5B illustrate thread execution logic 500 including an array of processing elements employed in a graphics processor core according to embodiments described herein. Elements of FIGS. 5A-5B having the same reference numbers (or names) as the elements of any other figure herein can operate or function in any manner similar to that described elsewhere herein, but are not limited to such. FIG. 5A-5B illustrates an overview of thread execution logic 500, which may be representative of hardware logic illustrated with each sub-core 221A-221F of FIG. 2B. FIG. 5A is representative of an execution unit within a general-purpose graphics processor, while FIG. 5B is representative of an execution unit that may be used within a compute accelerator.

As illustrated in FIG. 5A, in some embodiments thread execution logic 500 includes a shader processor 502, a thread dispatcher 504, instruction cache 506, a scalable execution unit array including a plurality of graphics execution units 508A-508N, a sampler 510, shared local memory 511, a data cache 512, and a data port 514. In one embodiment the scalable execution unit array can dynamically scale by enabling or disabling one or more execution units (e.g., any of graphics execution units 508A, 508B, 508C, 508D, through 508N-1 and 508N) based on the computational requirements of a workload. In one embodiment the included components are interconnected via an interconnect fabric that links to each of the components. In some embodiments, thread execution logic 500 includes one or more connections to memory, such as system memory or cache memory, through one or more of instruction cache 506, data port 514, sampler 510, and graphics execution units 508A-508N. In some embodiments, each execution unit (e.g., 508A) is a stand-alone programmable general-purpose computational unit that is capable of executing multiple simultaneous hardware threads while processing multiple data elements in parallel for each thread. In various embodiments, the array of graphics execution units 508A-508N is scalable to include any number individual execution units.

In some embodiments, the graphics execution units 508A-508N are primarily used to execute shader programs. A shader processor 502 can process the various shader programs and dispatch execution threads associated with the shader programs via a thread dispatcher 504. In one embodiment the thread dispatcher includes logic to arbitrate thread initiation requests from the graphics and media pipelines and instantiate the requested threads on one or more execution unit in the graphics execution units 508A-508N. For example, a geometry pipeline can dispatch vertex, tessellation, or geometry shaders to the thread execution logic for processing. In some embodiments, thread dispatcher 504 can also process runtime thread spawning requests from the executing shader programs.

In some embodiments, the graphics execution units 508A-508N support an instruction set that includes native support for many standard 3D graphics shader instructions, such that shader programs from graphics libraries (e.g., Direct 3D and OpenGL) are executed with a minimal translation. The execution units support vertex and geometry processing (e.g., vertex programs, geometry programs, vertex shaders), pixel processing (e.g., pixel shaders, fragment shaders) and general-purpose processing (e.g., compute and media shaders). Each of the execution units 508A-508N is capable of multi-issue single instruction multiple data (SIMD) execution and multi-threaded operation enables an efficient execution environment in the face of higher latency memory accesses. Each hardware thread within each execution unit has a dedicated high-bandwidth register file and associated independent thread-state. Execution is multi-issue per clock to pipelines capable of integer, single and double precision floating point operations, SIMD branch capability, logical operations, transcendental operations, and other miscellaneous operations. While waiting for data from memory or one of the shared functions, dependency logic within the graphics execution units 508A-508N causes a waiting thread to sleep until the requested data has been returned. While the waiting thread is sleeping, hardware resources may be devoted to processing other threads. For example, during a delay associated with a vertex shader operation, an execution unit can perform operations for a pixel shader, fragment shader, or another type of shader program, including a different vertex shader. Various embodiments can apply to use execution by use of Single Instruction Multiple Thread (SIMT) as an alternate to use of SIMD or in addition to use of SIMD. Reference to a SIMD core or operation can apply also to SIMT or apply to SIMD in combination with SIMT.

Each execution unit in graphics execution units 508A-508N operates on arrays of data elements. The number of data elements is the “execution size,” or the number of channels for the instruction. An execution channel is a logical unit of execution for data element access, masking, and flow control within instructions. The number of channels may be independent of the number of physical Arithmetic Logic Units (ALUs), Floating Point Units (FPUs), or other logic units (e.g., tensor cores, ray tracing cores, etc.) for a particular graphics processor. In some embodiments, graphics execution units 508A-508N support integer and floating-point data types.

The execution unit instruction set includes SIMD instructions. The various data elements can be stored as a packed data type in a register and the execution unit will process the various elements based on the data size of the elements. For example, when operating on a 256-bit wide vector, the 256 bits of the vector are stored in a register and the execution unit operates on the vector as four separate 54-bit packed data elements (Quad-Word (QW) size data elements), eight separate 32-bit packed data elements (Double Word (DW) size data elements), sixteen separate 16-bit packed data elements (Word (W) size data elements), or thirty-two separate 8-bit data elements (byte (B) size data elements). However, different vector widths and register sizes are possible.

In one embodiment one or more execution units can be combined into a fused execution unit 509A-509N having thread control logic (507A-507N) that is common to the fused EUs. Multiple EUs can be fused into an EU group. Each EU in the fused EU group can be configured to execute a separate SIMD hardware thread. The number of EUs in a fused EU group can vary according to embodiments. Additionally, various SIMD widths can be performed per-EU, including but not limited to SIMD8, SIMD16, and SIMD32. Each fused graphics execution unit 509A-509N includes at least two execution units. For example, fused execution unit 509A includes a first EU 508A, second EU 508B, and thread control logic 507A that is common to the first EU 508A and the second EU 508B. The thread control logic 507A controls threads executed on the fused graphics execution unit 509A, allowing each EU within the fused execution units 509A-509N to execute using a common instruction pointer register.

One or more internal instruction caches (e.g., 506) are included in the thread execution logic 500 to cache thread instructions for the execution units. In some embodiments, one or more data caches (e.g., 512) are included to cache thread data during thread execution. Threads executing on the execution logic 500 can also store explicitly managed data in the shared local memory 511. In some embodiments, a sampler 510 is included to provide texture sampling for 3D operations and media sampling for media operations. In some embodiments, sampler 510 includes specialized texture or media sampling functionality to process texture or media data during the sampling process before providing the sampled data to an execution unit.

During execution, the graphics and media pipelines send thread initiation requests to thread execution logic 500 via thread spawning and dispatch logic. Once a group of geometric objects has been processed and rasterized into pixel data, pixel processor logic (e.g., pixel shader logic, fragment shader logic, etc.) within the shader processor 502 is invoked to further compute output information and cause results to be written to output surfaces (e.g., color buffers, depth buffers, stencil buffers, etc.). In some embodiments, a pixel shader or fragment shader calculates the values of the various vertex attributes that are to be interpolated across the rasterized object. In some embodiments, pixel processor logic within the shader processor 502 then executes an application programming interface (API)-supplied pixel or fragment shader program. To execute the shader program, the shader processor 502 dispatches threads to an execution unit (e.g., 508A) via thread dispatcher 504. In some embodiments, shader processor 502 uses texture sampling logic in the sampler 510 to access texture data in texture maps stored in memory. Arithmetic operations on the texture data and the input geometry data compute pixel color data for each geometric fragment, or discards one or more pixels from further processing.

In some embodiments, the data port 514 provides a memory access mechanism for the thread execution logic 500 to output processed data to memory for further processing on a graphics processor output pipeline. In some embodiments, the data port 514 includes or couples to one or more cache memories (e.g., data cache 512) to cache data for memory access via the data port.

In one embodiment, the execution logic 500 can also include a ray tracer 505 that can provide ray tracing acceleration functionality. The ray tracer 505 can support a ray tracing instruction set that includes instructions/functions for ray generation. The ray tracing instruction set can be similar to or different from the ray-tracing instruction set supported by the ray tracing cores 245 in FIG. 2C.

FIG. 5B illustrates exemplary internal details of an execution unit 508, according to embodiments. A graphics execution unit 508 can include an instruction fetch unit 537, a general register file array (GRF) 524, an architectural register file array (ARF) 526, a thread arbiter 522, a send unit 530, a branch unit 532, a set of SIMD floating point units (FPUs) 534, and in one embodiment a set of dedicated integer SIMD ALUs 535. The GRF 524 and ARF 526 includes the set of general register files and architecture register files associated with each simultaneous hardware thread that may be active in the graphics execution unit 508. In one embodiment, per thread architectural state is maintained in the ARF 526, while data used during thread execution is stored in the GRF 524. The execution state of each thread, including the instruction pointers for each thread, can be held in thread-specific registers in the ARF 526.

In one embodiment the graphics execution unit 508 has an architecture that is a combination of Simultaneous Multi-Threading (SMT) and fine-grained Interleaved Multi-Threading (IMT). The architecture has a modular configuration that can be fine-tuned at design time based on a target number of simultaneous threads and number of registers per execution unit, where execution unit resources are divided across logic used to execute multiple simultaneous threads. The number of logical threads that may be executed by the graphics execution unit 508 is not limited to the number of hardware threads, and multiple logical threads can be assigned to each hardware thread.

In one embodiment, the graphics execution unit 508 can co-issue multiple instructions, which may each be different instructions. The thread arbiter 522 of the graphics execution unit 508 can dispatch the instructions to one of the send unit 530, branch unit 532, or SIMD FPU(s) 534 for execution. Each execution thread can access 128 general-purpose registers within the GRF 524, where each register can store 32 bytes, accessible as a SIMD 8-element vector of 32-bit data elements. In one embodiment, each execution unit thread has access to 4 Kbytes within the GRF 524, although embodiments are not so limited, and greater or fewer register resources may be provided in other embodiments. In one embodiment the graphics execution unit 508 is partitioned into seven hardware threads that can independently perform computational operations, although the number of threads per execution unit can also vary according to embodiments. For example, in one embodiment up to 16 hardware threads are supported. In an embodiment in which seven threads may access 4 Kbytes, the GRF 524 can store a total of 28 Kbytes. Where 16 threads may access 4 Kbytes, the GRF 524 can store a total of 64 Kbytes. Flexible addressing modes can permit registers to be addressed together to build effectively wider registers or to represent strided rectangular block data structures.

In one embodiment, memory operations, sampler operations, and other longer-latency system communications are dispatched via “send” instructions that are executed by the message passing send unit 530. In one embodiment, branch instructions are dispatched to a dedicated branch unit 532 to facilitate SIMD divergence and eventual convergence.

In one embodiment the graphics execution unit 508 includes one or more SIMD floating point units (FPU(s)) 534 to perform floating-point operations. In one embodiment, the FPU(s) 534 also support integer computation. In one embodiment the FPU(s) 534 can SIMD execute up to M number of 32-bit floating-point (or integer) operations, or SIMD execute up to 2M 16-bit integer or 16-bit floating-point operations. In one embodiment, at least one of the FPU(s) provides extended math capability to support high-throughput transcendental math functions and double precision 54-bit floating-point. In some embodiments, a set of 8-bit integer SIMD ALUs 535 are also present, and may be specifically optimized to perform operations associated with machine learning computations.

In one embodiment, arrays of multiple instances of the graphics execution unit 508 can be instantiated in a graphics sub-core grouping (e.g., a sub-slice). For scalability, product architects can choose the exact number of execution units per sub-core grouping. In one embodiment the execution unit 508 can execute instructions across a plurality of execution channels. In a further embodiment, each thread executed on the graphics execution unit 508 is executed on a different channel.

FIG. 6 illustrates an additional execution unit 600, according to an embodiment. The execution unit 600 may be a compute-optimized execution unit for use in, for example, a compute engine tile 340A-340D as in FIG. 3C, but is not limited as such. Variants of the execution unit 600 may also be used in a graphics engine tile 310A-310D as in FIG. 3B. In one embodiment, the execution unit 600 includes a thread control unit 601, a thread state unit 602, an instruction fetch/prefetch unit 603, and an instruction decode unit 604. The execution unit 600 additionally includes a register file 606 that stores registers that can be assigned to hardware threads within the execution unit. The execution unit 600 additionally includes a send unit 607 and a branch unit 608. In one embodiment, the send unit 607 and branch unit 608 can operate similarly as the send unit 530 and a branch unit 532 of the graphics execution unit 508 of FIG. 5B.

The execution unit 600 also includes a compute unit 610 that includes multiple different types of functional units. The compute unit 610 can include an ALU 611, a systolic array 612, and a math unit 613. The ALU 611 includes an array of arithmetic logic units. The ALU 611 can be configured to perform 64-bit, 32-bit, and 16-bit integer and floating point operations across multiple processing lanes and data channels and for multiple hardware and/or software threads. The ALU 611 can perform integer and floating point operations simultaneously (e.g., within the same clock cycle).

The systolic array 612 includes a W wide and D deep network of data processing units that can be used to perform vector or other data-parallel operations in a systolic manner. In one embodiment the systolic array 612 can be configured to perform various matrix operations, including as dot product, outer product, and general matrix-matrix multiplication (GEMM) operations. In one embodiment the systolic array 612 supports 16-bit floating point operations, as well as 8-bit, 4-bit, 2-bit, and binary integer operations. The systolic array 612 can be configured to accelerate specific machine learning operations, in addition to matrix multiply operations. In such embodiments, the systolic array 612 can be configured with support for the bfloat (brain floating point) 16-bit floating point format or a tensor float 32-bit floating point format (TF32) that have different numbers of mantissa and exponent bits relative to Institute of Electrical and Electronics Engineers (IEEE) 754 formats.

The systolic array 612 includes hardware to accelerate sparse matrix operations. In one embodiment, multiplication operations for sparse regions of input data can be bypassed at the processing element level by skipping multiply operations that have a zero value operand. In on embodiment, sparsity within input matrices can be detected and operations having known output values can be bypassed before being submitted to the processing elements of the systolic array 612. Additionally, the loading of zero value operands into the processing elements can be bypassed and the processing elements can be configured to perform multiplications on the non-zero value input elements. Output can be generated in a compressed (e.g., dense) format, with associated decompression or decoding metadata. The output can be cached in the compressed format. The output can be maintained in the compressed format when written to local memory or host system memory. The output may also be decompressed before being written to local memory or host system memory.

In one embodiment, the systolic array 612 includes hardware to enable operations on sparse data having a compressed representation. A compressed representation of a sparse matrix stores non-zero values and metadata that identifies the positions of the non-zero values within the matrix. Exemplary compressed representations include but are not limited to compressed tensor representations such as compressed sparse row (CSR), compressed sparse column (CSC), compressed sparse fiber (CSF) representations. Support for compressed representations enable operations to be performed on input in a compressed tensor format without requiring the compressed representation to be decompressed or decoded. In such embodiment, operations can be performed only on non-zero input values and the resulting non-zero output values can be mapped into an output matrix. In some embodiments, hardware support is also provided for machine-specific lossless data compression formats that are used when transmitting data within hardware or across system busses. Such data may be retained in a compressed format for sparse input data and the systolic array 612 can used the compression metadata for the compressed data to enable operations to be performed on only non-zero values, or to enable blocks of zero data input to be bypassed for multiply operations.

In one embodiment, a math unit 613 can be included to perform a specific subset of mathematical operations in an efficient and lower-power manner than the ALU 611. The math unit 613 can include a variant of math logic that may be found in shared function logic of a graphics processing engine provided by other embodiments (e.g., math logic 422 of the shared function logic 420 of FIG. 4). In one embodiment the math unit 613 can be configured to perform 32-bit and 64-bit floating point operations.

The thread control unit 601 includes logic to control the execution of threads within the execution unit. The thread control unit 601 can include thread arbitration logic to start, stop, and preempt execution of threads within the execution unit 600. The thread state unit 602 can be used to store thread state for threads assigned to execute on the execution unit 600. Storing the thread state within the execution unit 600 enables the rapid pre-emption of threads when those threads become blocked or idle. The instruction fetch/prefetch unit 603 can fetch instructions from an instruction cache of higher-level execution logic (e.g., instruction cache 506 as in FIG. 5A). The instruction fetch/prefetch unit 603 can also issue prefetch requests for instructions to be loaded into the instruction cache based on an analysis of currently executing threads. The instruction decode unit 604 can be used to decode instructions to be executed by the compute units. In one embodiment, the instruction decode unit 604 can be used as a secondary decoder to decode complex instructions into constituent micro-operations.

The execution unit 600 additionally includes a register file 606 that can be used by hardware threads executing on the execution unit 600. Registers in the register file 606 can be divided across the logic used to execute multiple simultaneous threads within the compute unit 610 of the execution unit 600. The number of logical threads that may be executed by the graphics execution unit 600is not limited to the number of hardware threads, and multiple logical threads can be assigned to each hardware thread. The size of the register file 606 can vary across embodiments based on the number of supported hardware threads. In one embodiment, register renaming may be used to dynamically allocate registers to hardware threads.

FIG. 7 is a block diagram illustrating graphics processor instruction formats 700 according to some embodiments. In one or more embodiment, the graphics processor execution units support an instruction set having instructions in multiple formats. The solid lined boxes illustrate the components that are generally included in an execution unit instruction, while the dashed lines include components that are optional or that are only included in a sub-set of the instructions. In some embodiments, the graphics processor instruction format 700 described and illustrated are macro-instructions, in that they are instructions supplied to the execution unit, as opposed to micro-operations resulting from instruction decode once the instruction is processed. Thus, a single instructions may cause hardware to perform multiple micro-operations.

In some embodiments, the graphics processor execution units natively support instructions in a 128-bit instruction format 710. A 64-bit compacted instruction format 730 is available for some instructions based on the selected instruction, instruction options, and number of operands. The native 128-bit instruction format 710 provides access to all instruction options, while some options and operations are restricted in the 64-bit format 730. The native instructions available in the 64-bit format 730 vary by embodiment. In some embodiments, the instruction is compacted in part using a set of index values in an index field 713. The execution unit hardware references a set of compaction tables based on the index values and uses the compaction table outputs to reconstruct a native instruction in the 128-bit instruction format 710. Other sizes and formats of instruction can be used.

For each format, instruction opcode 712 defines the operation that the execution unit is to perform. The execution units execute each instruction in parallel across the multiple data elements of each operand. For example, in response to an add instruction the execution unit performs a simultaneous add operation across each color channel representing a texture element or picture element. By default, the execution unit performs each instruction across all data channels of the operands. In some embodiments, instruction control field 714 enables control over certain execution options, such as channels selection (e.g., predication) and data channel order (e.g., swizzle). For instructions in the 128-bit instruction format 710 an exec-size field 716 limits the number of data channels that will be executed in parallel. In some embodiments, exec-size field 716 is not available for use in the 64-bit compact instruction format 730.

Some execution unit instructions have up to three operands including two source operands, src0 720, src1 722, and one destination 718. In some embodiments, the execution units support dual destination instructions, where one of the destinations is implied. Data manipulation instructions can have a third source operand (e.g., SRC2 724), where the instruction opcode 712 determines the number of source operands. An instruction's last source operand can be an immediate (e.g., hard-coded) value passed with the instruction.

In some embodiments, the 128-bit instruction format 710 includes an access/address mode field 726 specifying, for example, whether direct register addressing mode or indirect register addressing mode is used. When direct register addressing mode is used, the register address of one or more operands is directly provided by bits in the instruction.

In some embodiments, the 128-bit instruction format 710 includes an access/address mode field 726, which specifies an address mode and/or an access mode for the instruction. In one embodiment the access mode is used to define a data access alignment for the instruction. Some embodiments support access modes including a 16-byte aligned access mode and a 1-byte aligned access mode, where the byte alignment of the access mode determines the access alignment of the instruction operands. For example, when in a first mode, the instruction may use byte-aligned addressing for source and destination operands and when in a second mode, the instruction may use 16-byte-aligned addressing for all source and destination operands.

In one embodiment, the address mode portion of the access/address mode field 726 determines whether the instruction is to use direct or indirect addressing. When direct register addressing mode is used bits in the instruction directly provide the register address of one or more operands. When indirect register addressing mode is used, the register address of one or more operands may be computed based on an address register value and an address immediate field in the instruction.

In some embodiments instructions are grouped based on opcode 712 bit-fields to simplify Opcode decode 740. For an 8-bit opcode, bits 4, 5, and 6 allow the execution unit to determine the type of opcode. The precise opcode grouping shown is merely an example. In some embodiments, a move and logic opcode group 742 includes data movement and logic instructions (e.g., move (mov), compare (cmp)). In some embodiments, move and logic group 742 shares the five most significant bits (MSB), where move (mov) instructions are in the form of 0000xxxxb and logic instructions are in the form of 0001xxxxb. A flow control instruction group 744 (e.g., call, jump (jmp)) includes instructions in the form of 0010xxxxb (e.g., 0×20). A miscellaneous instruction group 746 includes a mix of instructions, including synchronization instructions (e.g., wait, send) in the form of 0011xxxxb (e.g., 0×30). A parallel math instruction group 748 includes component-wise arithmetic instructions (e.g., add, multiply (mul)) in the form of 0100xxxxb (e.g., 0×40). The parallel math instruction group 748 performs the arithmetic operations in parallel across data channels. The vector math group 750 includes arithmetic instructions (e.g., dp4) in the form of 0101xxxxb (e.g., 0×50). The vector math group performs arithmetic such as dot product calculations on vector operands. The illustrated opcode decode 740, in one embodiment, can be used to determine which portion of an execution unit will be used to execute a decoded instruction. For example, some instructions may be designated as systolic instructions that will be performed by a systolic array. Other instructions, such as ray-tracing instructions (not shown) can be routed to a ray-tracing core or ray-tracing logic within a slice or partition of execution logic.

Graphics Pipeline

FIG. 8 is a block diagram of another embodiment of a graphics processor 800. Elements of FIG. 8 having the same reference numbers (or names) as the elements of any other figure herein can operate or function in any manner similar to that described elsewhere herein, but are not limited to such.

In some embodiments, graphics processor 800 includes a geometry pipeline 820, a media pipeline 830, a display engine 840, thread execution logic 850, and a render output pipeline 870. In some embodiments, graphics processor 800 is a graphics processor within a multi-core processing system that includes one or more general-purpose processing cores. The graphics processor is controlled by register writes to one or more control registers (not shown) or via commands issued to graphics processor 800 via a ring interconnect 802. In some embodiments, ring interconnect 802 couples graphics processor 800 to other processing components, such as other graphics processors or general-purpose processors. Commands from ring interconnect 802 are interpreted by a command streamer 803, which supplies instructions to individual components of the geometry pipeline 820 or the media pipeline 830.

In some embodiments, command streamer 803 directs the operation of a vertex fetcher 805 that reads vertex data from memory and executes vertex-processing commands provided by command streamer 803. In some embodiments, vertex fetcher 805 provides vertex data to a vertex shader 807, which performs coordinate space transformation and lighting operations to each vertex. In some embodiments, vertex fetcher 805 and vertex shader 807 execute vertex-processing instructions by dispatching execution threads to execution units 852A-852B via a thread dispatcher 831.

In some embodiments, execution units 852A-852B are an array of vector processors having an instruction set for performing graphics and media operations. In some embodiments, execution units 852A-852B have an attached L1 cache 851 that is specific for each array or shared between the arrays. The cache can be configured as a data cache, an instruction cache, or a single cache that is partitioned to contain data and instructions in different partitions.

In some embodiments, geometry pipeline 820 includes tessellation components to perform hardware-accelerated tessellation of 3D objects. In some embodiments, a programmable hull shader 811 configures the tessellation operations. A programmable domain shader 817 provides back-end evaluation of tessellation output. A tessellator 813 operates at the direction of hull shader 811 and contains special purpose logic to generate a set of detailed geometric objects based on a coarse geometric model that is provided as input to geometry pipeline 820. In some embodiments, if tessellation is not used, tessellation components (e.g., hull shader 811, tessellator 813, and domain shader 817) can be bypassed. The tessellation components can operate based on data received from the vertex shader 807.

In some embodiments, complete geometric objects can be processed by a geometry shader 819 via one or more threads dispatched to execution units 852A-852B, or can proceed directly to the clipper 829. In some embodiments, the geometry shader operates on entire geometric objects, rather than vertices or patches of vertices as in previous stages of the graphics pipeline. If the tessellation is disabled the geometry shader 819 receives input from the vertex shader 807. In some embodiments, geometry shader 819 is programmable by a geometry shader program to perform geometry tessellation if the tessellation units are disabled.

Before rasterization, a clipper 829 processes vertex data. The clipper 829 may be a fixed function clipper or a programmable clipper having clipping and geometry shader functions. In some embodiments, a rasterizer and depth test component 873 in the render output pipeline 870 dispatches pixel shaders to convert the geometric objects into per pixel representations. In some embodiments, pixel shader logic is included in thread execution logic 850. In some embodiments, an application can bypass the rasterizer and depth test component 873 and access un-rasterized vertex data via a stream out unit 823.

The graphics processor 800 has an interconnect bus, interconnect fabric, or some other interconnect mechanism that allows data and message passing amongst the major components of the processor. In some embodiments, execution units 852A-852B and associated logic units (e.g., L1 cache 851, sampler 854, texture cache 858, etc.) interconnect via a data port 856 to perform memory access and communicate with render output pipeline components of the processor. In some embodiments, sampler 854, caches 851, 858 and execution units 852A-852B each have separate memory access paths. In one embodiment the texture cache 858 can also be configured as a sampler cache.

In some embodiments, render output pipeline 870 contains a rasterizer and depth test component 873 that converts vertex-based objects into an associated pixel-based representation. In some embodiments, the rasterizer logic includes a windower/masker unit to perform fixed function triangle and line rasterization. An associated render cache 878 and depth cache 879 are also available in some embodiments. A pixel operations component 877 performs pixel-based operations on the data, though in some instances, pixel operations associated with 2D operations (e.g. bit block image transfers with blending) are performed by the 2D engine 841, or substituted at display time by the display controller 843 using overlay display planes. In some embodiments, a shared L3 cache 875 is available to all graphics components, allowing the sharing of data without the use of main system memory.

In some embodiments, media pipeline 830 includes a media engine 837 and a video front-end 834. In some embodiments, video front-end 834 receives pipeline commands from the command streamer 803. In some embodiments, media pipeline 830 includes a separate command streamer. In some embodiments, video front-end 834 processes media commands before sending the command to the media engine 837. In some embodiments, media engine 837 includes thread spawning functionality to spawn threads for dispatch to thread execution logic 850 via thread dispatcher 831.

In some embodiments, graphics processor 800 includes a display engine 840. In some embodiments, display engine 840 is external to processor 800 and couples with the graphics processor via the ring interconnect 802, or some other interconnect bus or fabric. In some embodiments, display engine 840 includes a 2D engine 841 and a display controller 843. In some embodiments, display engine 840 contains special purpose logic capable of operating independently of the 3D pipeline. In some embodiments, display controller 843 couples with a display device (not shown), which may be a system integrated display device, as in a laptop computer, or an external display device attached via a display device connector.

In some embodiments, the geometry pipeline 820 and media pipeline 830 are configurable to perform operations based on multiple graphics and media programming interfaces and are not specific to any one application programming interface (API). In some embodiments, driver software for the graphics processor translates API calls that are specific to a particular graphics or media library into commands that can be processed by the graphics processor. In some embodiments, support is provided for the Open Graphics Library (OpenGL), Open Computing Language (OpenCL), and/or Vulkan graphics and compute API, all from the Khronos Group. In some embodiments, support may also be provided for the Direct3D library from the Microsoft Corporation. In some embodiments, a combination of these libraries may be supported. Support may also be provided for the Open Source Computer Vision Library (OpenCV). A future API with a compatible 3D pipeline would also be supported if a mapping can be made from the pipeline of the future API to the pipeline of the graphics processor.

Graphics Pipeline Programming

FIG. 9A is a block diagram illustrating a graphics processor command format 900 that may be used to program graphics processing pipelines according to some embodiments. FIG. 9B is a block diagram illustrating a graphics processor command sequence 910 according to an embodiment. The solid lined boxes in FIG. 9A illustrate the components that are generally included in a graphics command while the dashed lines include components that are optional or that are only included in a sub-set of the graphics commands. The exemplary graphics processor command format 900 of FIG. 9A includes data fields to identify a client 902, a command operation code (opcode) 904, and a data field 906 for the command. A sub-opcode 905 and a command size 908 are also included in some commands.

In some embodiments, client 902 specifies the client unit of the graphics device that processes the command data. In some embodiments, a graphics processor command parser examines the client field of each command to condition the further processing of the command and route the command data to the appropriate client unit. In some embodiments, the graphics processor client units include a memory interface unit, a render unit, a 2D unit, a 3D unit, and a media unit. Each client unit has a corresponding processing pipeline that processes the commands. Once the command is received by the client unit, the client unit reads the opcode 904 and, if present, sub-opcode 905 to determine the operation to perform. The client unit performs the command using information in data field 906. For some commands an explicit command size 908 is expected to specify the size of the command. In some embodiments, the command parser automatically determines the size of at least some of the commands based on the command opcode. In some embodiments commands are aligned via multiples of a double word. Other command formats can be used.

The flow diagram in FIG. 9B illustrates an exemplary graphics processor command sequence 910. In some embodiments, software or firmware of a data processing system that features an embodiment of a graphics processor uses a version of the command sequence shown to set up, execute, and terminate a set of graphics operations. A sample command sequence is shown and described for purposes of example only as embodiments are not limited to these specific commands or to this command sequence. Moreover, the commands may be issued as batch of commands in a command sequence, such that the graphics processor will process the sequence of commands in at least partially concurrence.

In some embodiments, the graphics processor command sequence 910 may begin with a pipeline flush command 912 to cause any active graphics pipeline to complete the currently pending commands for the pipeline. In some embodiments, the 3D pipeline 922 and the media pipeline 924 do not operate concurrently. The pipeline flush is performed to cause the active graphics pipeline to complete any pending commands. In response to a pipeline flush, the command parser for the graphics processor will pause command processing until the active drawing engines complete pending operations and the relevant read caches are invalidated. Optionally, any data in the render cache that is marked ‘dirty’ can be flushed to memory. In some embodiments, pipeline flush command 912 can be used for pipeline synchronization or before placing the graphics processor into a low power state.

In some embodiments, a pipeline select command 913 is used when a command sequence requires the graphics processor to explicitly switch between pipelines. In some embodiments, a pipeline select command 913 is required only once within an execution context before issuing pipeline commands unless the context is to issue commands for both pipelines. In some embodiments, a pipeline flush command 912 is required immediately before a pipeline switch via the pipeline select command 913.

In some embodiments, a pipeline control command 914 configures a graphics pipeline for operation and is used to program the 3D pipeline 922 and the media pipeline 924. In some embodiments, pipeline control command 914 configures the pipeline state for the active pipeline. In one embodiment, the pipeline control command 914 is used for pipeline synchronization and to clear data from one or more cache memories within the active pipeline before processing a batch of commands.

In some embodiments, commands related to the return buffer state 916 are used to configure a set of return buffers for the respective pipelines to write data. Some pipeline operations require the allocation, selection, or configuration of one or more return buffers into which the operations write intermediate data during processing. In some embodiments, the graphics processor also uses one or more return buffers to store output data and to perform cross thread communication. In some embodiments, the return buffer state 916 includes selecting the size and number of return buffers to use for a set of pipeline operations.

The remaining commands in the command sequence differ based on the active pipeline for operations. Based on a pipeline determination 920, the command sequence is tailored to the 3D pipeline 922 beginning with the 3D pipeline state 930 or the media pipeline 924 beginning at the media pipeline state 940.

The commands to configure the 3D pipeline state 930 include 3D state setting commands for vertex buffer state, vertex element state, constant color state, depth buffer state, and other state variables that are to be configured before 3D primitive commands are processed. The values of these commands are determined at least in part based on the particular 3D API library in use. In some embodiments, 3D pipeline state 930 commands are also able to selectively disable or bypass certain pipeline elements if those elements will not be used.

In some embodiments, 3D primitive 932 command is used to submit 3D primitives to be processed by the 3D pipeline. Commands and associated parameters that are passed to the graphics processor via the 3D primitive 932 command are forwarded to the vertex fetch function in the graphics pipeline. The vertex fetch function uses the 3D primitive 932 command data to generate vertex data structures. The vertex data structures are stored in one or more return buffers. In some embodiments, 3D primitive 932 command is used to perform vertex operations on 3D primitives via vertex shaders. To process vertex shaders, 3D pipeline 922 dispatches shader execution threads to graphics processor execution units.

In some embodiments, 3D pipeline 922 is triggered via an execute 934 command or event. In some embodiments, a register write triggers command execution. In some embodiments execution is triggered via a ‘go’ or ‘kick’ command in the command sequence. In one embodiment, command execution is triggered using a pipeline synchronization command to flush the command sequence through the graphics pipeline. The 3D pipeline will perform geometry processing for the 3D primitives. Once operations are complete, the resulting geometric objects are rasterized and the pixel engine colors the resulting pixels. Additional commands to control pixel shading and pixel back end operations may also be included for those operations.

In some embodiments, the graphics processor command sequence 910 follows the media pipeline 924 path when performing media operations. In general, the specific use and manner of programming for the media pipeline 924 depends on the media or compute operations to be performed. Specific media decode operations may be offloaded to the media pipeline during media decode. In some embodiments, the media pipeline can also be bypassed and media decode can be performed in whole or in part using resources provided by one or more general-purpose processing cores. In one embodiment, the media pipeline also includes elements for general-purpose graphics processor unit (GPGPU) operations, where the graphics processor is used to perform SIMD vector operations using computational shader programs that are not explicitly related to the rendering of graphics primitives.

In some embodiments, media pipeline 924 is configured in a similar manner as the 3D pipeline 922. A set of commands to configure the media pipeline state 940 are dispatched or placed into a command queue before the media object commands 942. In some embodiments, commands for the media pipeline state 940 include data to configure the media pipeline elements that will be used to process the media objects. This includes data to configure the video decode and video encode logic within the media pipeline, such as encode or decode format. In some embodiments, commands for the media pipeline state 940 also support the use of one or more pointers to “indirect” state elements that contain a batch of state settings.

In some embodiments, media object commands 942 supply pointers to media objects for processing by the media pipeline. The media objects include memory buffers containing video data to be processed. In some embodiments, all media pipeline states must be valid before issuing a media object command 942. Once the pipeline state is configured and media object commands 942 are queued, the media pipeline 924 is triggered via an execute command 944 or an equivalent execute event (e.g., register write). Output from media pipeline 924 may then be post processed by operations provided by the 3D pipeline 922 or the media pipeline 924. In some embodiments, GPGPU operations are configured and executed in a similar manner as media operations.

Graphics Software Architecture

FIG. 10 illustrates an exemplary graphics software architecture for a data processing system 1000 according to some embodiments. In some embodiments, software architecture includes a 3D graphics application 1010, an operating system 1020, and at least one processor 1030. In some embodiments, processor 1030 includes a graphics processor 1032 and one or more general-purpose processor core(s) 1034. The graphics application 1010 and operating system 1020 each execute in the system memory 1050 of the data processing system.

In some embodiments, 3D graphics application 1010 contains one or more shader programs including shader instructions 1012. The shader language instructions may be in a high-level shader language, such as the High-Level Shader Language (HLSL) of Direct3D, the OpenGL Shader Language (GLSL), and so forth. The application also includes executable instructions 1014 in a machine language suitable for execution by the general-purpose processor core 1034. The application also includes graphics objects 1016 defined by vertex data.

In some embodiments, operating system 1020 is a Microsoft® Windows® operating system from the Microsoft Corporation, a proprietary UNIX-like operating system, or an open source UNIX-like operating system using a variant of the Linux kernel. The operating system 1020 can support a graphics API 1022 such as the Direct3D API, the OpenGL API, or the Vulkan API. When the Direct3D API is in use, the operating system 1020 uses a front-end shader compiler 1024 to compile any shader instructions 1012 in HLSL into a lower-level shader language. The compilation may be a just-in-time (JIT) compilation or the application can perform shader pre-compilation. In some embodiments, high-level shaders are compiled into low-level shaders during the compilation of the 3D graphics application 1010. In some embodiments, the shader instructions 1012 are provided in an intermediate form, such as a version of the Standard Portable Intermediate Representation (SPIR) used by the Vulkan API.

In some embodiments, user mode graphics driver 1026 contains a back-end shader compiler 1027 to convert the shader instructions 1012 into a hardware specific representation. When the OpenGL API is in use, shader instructions 1012 in the GLSL high-level language are passed to a user mode graphics driver 1026 for compilation. In some embodiments, user mode graphics driver 1026 uses operating system kernel mode functions 1028 to communicate with a kernel mode graphics driver 1029. In some embodiments, kernel mode graphics driver 1029 communicates with graphics processor 1032 to dispatch commands and instructions.

IP Core Implementations

One or more aspects of at least one embodiment may be implemented by representative code stored on a machine-readable medium which represents and/or defines logic within an integrated circuit such as a processor. For example, the machine-readable medium may include instructions which represent various logic within the processor. When read by a machine, the instructions may cause the machine to fabricate the logic to perform the techniques described herein. Such representations, known as “IP cores,” are reusable units of logic for an integrated circuit that may be stored on a tangible, machine-readable medium as a hardware model that describes the structure of the integrated circuit. The hardware model may be supplied to various customers or manufacturing facilities, which load the hardware model on fabrication machines that manufacture the integrated circuit. The integrated circuit may be fabricated such that the circuit performs operations described in association with any of the embodiments described herein.

FIG. 11A is a block diagram illustrating an IP core development system 1100 that may be used to manufacture an integrated circuit to perform operations according to an embodiment. The IP core development system 1100 may be used to generate modular, re-usable designs that can be incorporated into a larger design or used to construct an entire integrated circuit (e.g., an SOC integrated circuit). A design facility 1130 can generate a software simulation 1110 of an IP core design in a high-level programming language (e.g., C/C++). The software simulation 1110 can be used to design, test, and verify the behavior of the IP core using a simulation model 1112. The simulation model 1112 may include functional, behavioral, and/or timing simulations. A register transfer level (RTL) design 1115 can then be created or synthesized from the simulation model 1112. The RTL design 1115 is an abstraction of the behavior of the integrated circuit that models the flow of digital signals between hardware registers, including the associated logic performed using the modeled digital signals. In addition to an RTL design 1115, lower-level designs at the logic level or transistor level may also be created, designed, or synthesized. Thus, the particular details of the initial design and simulation may vary.

The RTL design 1115 or equivalent may be further synthesized by the design facility into a hardware model 1120, which may be in a hardware description language (HDL), or some other representation of physical design data. The HDL may be further simulated or tested to verify the IP core design. The IP core design can be stored for delivery to a 3^rd party fabrication facility 1165 using non-volatile memory 1140 (e.g., hard disk, flash memory, or any non-volatile storage medium). Alternatively, the IP core design may be transmitted (e.g., via the Internet) over a wired connection 1150 or wireless connection 1160. The fabrication facility 1165 may then fabricate an integrated circuit that is based at least in part on the IP core design. The fabricated integrated circuit can be configured to perform operations in accordance with at least one embodiment described herein.

FIG. 11B illustrates a cross-section side view of an integrated circuit package assembly 1170, according to some embodiments described herein. The integrated circuit package assembly 1170 illustrates an implementation of one or more processor or accelerator devices as described herein. The package assembly 1170 includes multiple units of hardware logic 1172, 1174 connected to a substrate 1180. The logic 1172, 1174 may be implemented at least partly in configurable logic or fixed-functionality logic hardware, and can include one or more portions of any of the processor core(s), graphics processor(s), or other accelerator devices described herein. Each unit of logic 1172, 1174 can be implemented within a semiconductor die and coupled with the substrate 1180 via an interconnect structure 1173. The interconnect structure 1173 may be configured to route electrical signals between the logic 1172, 1174 and the substrate 1180, and can include interconnects such as, but not limited to bumps or pillars. In some embodiments, the interconnect structure 1173 may be configured to route electrical signals such as, for example, input/output (I/O) signals and/or power or ground signals associated with the operation of the logic 1172, 1174. In some embodiments, the substrate 1180 is an epoxy-based laminate substrate. The substrate 1180 may include other suitable types of substrates in other embodiments. The package assembly 1170 can be connected to other electrical devices via a package interconnect 1183. The package interconnect 1183 may be coupled to a surface of the substrate 1180 to route electrical signals to other electrical devices, such as a motherboard, other chipset, or multi-chip module.

In some embodiments, the units of logic 1172, 1174 are electrically coupled with a bridge 1182 that is configured to route electrical signals between the logic 1172, 1174. The bridge 1182 may be a dense interconnect structure that provides a route for electrical signals. The bridge 1182 may include a bridge substrate composed of glass or a suitable semiconductor material. Electrical routing features can be formed on the bridge substrate to provide a chip-to-chip connection between the logic 1172, 1174.

Although two units of logic 1172, 1174 and a bridge 1182 are illustrated, embodiments described herein may include more or fewer logic units on one or more dies. The one or more dies may be connected by zero or more bridges, as the bridge 1182 may be excluded when the logic is included on a single die. Alternatively, multiple dies or units of logic can be connected by one or more bridges. Additionally, multiple logic units, dies, and bridges can be connected together in other possible configurations, including three-dimensional configurations.

FIG. 11C illustrates a package assembly 1190 that includes multiple units of hardware logic chiplets connected to a substrate 1180. A graphics processing unit, parallel processor, and/or compute accelerator as described herein can be composed from diverse silicon chiplets that are separately manufactured. In this context, a chiplet is an at least partially packaged integrated circuit that includes distinct units of logic that can be assembled with other chiplets into a larger package. A diverse set of chiplets with different IP core logic can be assembled into a single device. Additionally the chiplets can be integrated into a base die or base chiplet using active interposer technology. The concepts described herein enable the interconnection and communication between the different forms of IP within the GPU. IP cores can be manufactured using different process technologies and composed during manufacturing, which avoids the complexity of converging multiple IPs, especially on a large SoC with several flavors IPs, to the same manufacturing process. Enabling the use of multiple process technologies improves the time to market and provides a cost-effective way to create multiple product SKUs. Additionally, the disaggregated IPs are more amenable to being power gated independently, components that are not in use on a given workload can be powered off, reducing overall power consumption.

In various embodiments a package assembly 1190 can include components and chiplets that are interconnected by a fabric 1185 and/or one or more bridges 1187. The chiplets within the package assembly 1190 may have a 2.5D arrangement using Chip-on-Wafer-on-Substrate stacking in which multiple dies are stacked side-by-side on a silicon interposer 1189 that couples the chiplets with the substrate 1180. The substrate 1180 includes electrical connections to the package interconnect 1183. In one embodiment the silicon interposer 1189 is a passive interposer that includes through-silicon vias (TSVs) to electrically couple chiplets within the package assembly 1190 to the substrate 1180. In one embodiment, silicon interposer 1189 is an active interposer that includes embedded logic in addition to TSVs. In such embodiment, the chiplets within the package assembly 1190 are arranged using 3D face to face die stacking on top of the active interposer 1189. The active interposer 1189 can include hardware logic for I/O 1191, cache memory 1192, and other hardware logic 1193, in addition to interconnect fabric 1185 and a silicon bridge 1187. The fabric 1185 enables communication between the various logic chiplets 1172, 1174 and the logic 1191, 1193 within the active interposer 1189. The fabric 1185 may be an NoC interconnect or another form of packet switched fabric that switches data packets between components of the package assembly. For complex assemblies, the fabric 1185 may be a dedicated chiplet enables communication between the various hardware logic of the package assembly 1190.

Bridge structures 1187 within the active interposer 1189 may be used to facilitate a point to point interconnect between, for example, logic or 110 chiplets 1174 and memory chiplets 1175. In some implementations, bridge structures 1187 may also be embedded within the substrate 1180. The hardware logic chiplets can include special purpose hardware logic chiplets 1172, logic or I/O chiplets 1174, and/or memory chiplets 1175. The hardware logic chiplets 1172 and logic or I/O chiplets 1174 may be implemented at least partly in configurable logic or fixed-functionality logic hardware and can include one or more portions of any of the processor core(s), graphics processor(s), parallel processors, or other accelerator devices described herein. The memory chiplets 1175 can be DRAM (e.g., GDDR, HBM) memory or cache (SRAM) memory. Cache memory 1192 within the active interposer 1189 (or substrate 1180) can act as a global cache for the package assembly 1190, part of a distributed global cache, or as a dedicated cache for the fabric 1185.

Each chiplet can be fabricated as separate semiconductor die and coupled with a base die that is embedded within or coupled with the substrate 1180. The coupling with the substrate 1180 can be performed via an interconnect structure 1173. The interconnect structure 1173 may be configured to route electrical signals between the various chiplets and logic within the substrate 1180. The interconnect structure 1173 can include interconnects such as, but not limited to bumps or pillars. In some embodiments, the interconnect structure 1173 may be configured to route electrical signals such as, for example, input/output (I/O) signals and/or power or ground signals associated with the operation of the logic, 110 and memory chiplets. In one embodiment, an additional interconnect structure couples the active interposer 1189 with the substrate 1180.

In some embodiments, the substrate 1180 is an epoxy-based laminate substrate. The substrate 1180 may include other suitable types of substrates in other embodiments. The package assembly 1190 can be connected to other electrical devices via a package interconnect 1183. The package interconnect 1183 may be coupled to a surface of the substrate 1180 to route electrical signals to other electrical devices, such as a motherboard, other chipset, or multi-chip module.

In some embodiments, a logic or I/O chiplet 1174 and a memory chiplet 1175 can be electrically coupled via a bridge 1187 that is configured to route electrical signals between the logic or I/O chiplet 1174 and a memory chiplet 1175. The bridge 1187 may be a dense interconnect structure that provides a route for electrical signals. The bridge 1187 may include a bridge substrate composed of glass or a suitable semiconductor material. Electrical routing features can be formed on the bridge substrate to provide a chip-to-chip connection between the logic or I/O chiplet 1174 and a memory chiplet 1175. The bridge 1187 may also be referred to as a silicon bridge or an interconnect bridge. For example, the bridge 1187, in some embodiments, is an Embedded Multi-die Interconnect Bridge (EMIB). In some embodiments, the bridge 1187 may simply be a direct connection from one chiplet to another chiplet.

FIG. 11D illustrates a package assembly 1194 including interchangeable chiplets 1195, according to an embodiment. The interchangeable chiplets 1195 can be assembled into standardized slots on one or more base chiplets 1196, 1198. The base chiplets 1196, 1198 can be coupled via a bridge interconnect 1197, which can be similar to the other bridge interconnects described herein and may be, for example, an EMIB. Memory chiplets can also be connected to logic or I/O chiplets via a bridge interconnect. I/O and logic chiplets can communicate via an interconnect fabric. The base chiplets can each support one or more slots in a standardized format for one of logic or I/O or memory/cache.

In one embodiment, SRAM and power delivery circuits can be fabricated into one or more of the base chiplets 1196, 1198, which can be fabricated using a different process technology relative to the interchangeable chiplets 1195 that are stacked on top of the base chiplets. For example, the base chiplets 1196, 1198 can be fabricated using a larger process technology, while the interchangeable chiplets can be manufactured using a smaller process technology. One or more of the interchangeable chiplets 1195 may be memory (e.g., DRAM) chiplets. Different memory densities can be selected for the package assembly 1194 based on the power, and/or performance targeted for the product that uses the package assembly 1194. Additionally, logic chiplets with a different number of type of functional units can be selected at time of assembly based on the power, and/or performance targeted for the product. Additionally, chiplets containing IP logic cores of differing types can be inserted into the interchangeable chiplet slots, enabling hybrid processor designs that can mix and match different technology IP blocks.

Exemplary System on a Chip Integrated Circuit

FIGS. 12-13B illustrate exemplary integrated circuits and associated graphics processors that may be fabricated using one or more IP cores, according to various embodiments described herein. In addition to what is illustrated, other logic and circuits may be included, including additional graphics processors/cores, peripheral interface controllers, or general-purpose processor cores.

FIG. 12 is a block diagram illustrating an exemplary system on a chip integrated circuit 1200 that may be fabricated using one or more IP cores, according to an embodiment. Exemplary integrated circuit 1200 includes one or more application processor(s) 1205 (e.g., CPUs), at least one graphics processor 1210, and may additionally include an image processor 1215 and/or a video processor 1220, any of which may be a modular IP core from the same or multiple different design facilities. Integrated circuit 1200 includes peripheral or bus logic including a USB controller 1225, UART controller 1230, an SPI/SDIO controller 1235, and an I²S/I²C controller 1240. Additionally, the integrated circuit can include a display device 1245 coupled to one or more of a high-definition multimedia interface (HDMI) controller 1250 and a mobile industry processor interface (MIPI) display interface 1255. Storage may be provided by a flash memory subsystem 1260 including flash memory and a flash memory controller. Memory interface may be provided via a memory controller 1265 for access to SDRAM or SRAM memory devices. Some integrated circuits additionally include an embedded security engine 1270.

FIGS. 13A-13B are block diagrams illustrating exemplary graphics processors for use within an SoC, according to embodiments described herein. FIG. 13A illustrates an exemplary graphics processor 1310 of a system on a chip integrated circuit that may be fabricated using one or more IP cores, according to an embodiment. FIG. 13B illustrates an additional exemplary graphics processor 1340 of a system on a chip integrated circuit that may be fabricated using one or more IP cores, according to an embodiment. Graphics processor 1310 of FIG. 13A is an example of a low power graphics processor core. Graphics processor 1340 of FIG. 13B is an example of a higher performance graphics processor core. Each of graphics processor 1310 and graphics processor 1340 can be variants of the graphics processor 1210 of FIG. 12.

As shown in FIG. 13A, graphics processor 1310 includes a vertex processor 1305 and one or more fragment processor(s) 1315A-1315N (e.g., 1315A, 1315B, 1315C, 1315D, through 1315N-1, and 1315N). Graphics processor 1310 can execute different shader programs via separate logic, such that the vertex processor 1305 is optimized to execute operations for vertex shader programs, while the one or more fragment processor(s) 1315A-1315N execute fragment (e.g., pixel) shading operations for fragment or pixel shader programs. The vertex processor 1305 performs the vertex processing stage of the 3D graphics pipeline and generates primitives and vertex data. The fragment processor(s) 1315A-1315N use the primitive and vertex data generated by the vertex processor 1305 to produce a frame buffer that is displayed on a display device. In one embodiment, the fragment processor(s) 1315A-1315N are optimized to execute fragment shader programs as provided for in the OpenGL API, which may be used to perform similar operations as a pixel shader program as provided for in the Direct 3D API.

Graphics processor 1310 additionally includes one or more memory management units (MMUs) 1320A-1320B, cache(s) 1325A-1325B, and circuit interconnect(s) 1330A-1330B. The one or more MMU(s) 1320A-1320B provide for virtual to physical address mapping for the graphics processor 1310, including for the vertex processor 1305 and/or fragment processor(s) 1315A-1315N, which may reference vertex or image/texture data stored in memory, in addition to vertex or image/texture data stored in the one or more cache(s) 1325A-1325B. In one embodiment the one or more MMU(s) 1320A-1320B may be synchronized with other MMUs within the system, including one or more MMUs associated with the one or more application processor(s) 1205, image processor 1215, and/or video processor 1220 of FIG. 12, such that each processor 1205-1220 can participate in a shared or unified virtual memory system. The one or more circuit interconnect(s) 1330A-1330B enable graphics processor 1310 to interface with other IP cores within the SoC, either via an internal bus of the SoC or via a direct connection, according to embodiments.

As shown FIG. 13B, graphics processor 1340 includes the one or more MMU(s) 1320A-1320B, cache(s) 1325A-1325B, and circuit interconnect(s) 1330A-1330B of the graphics processor 1310 of FIG. 13A. Graphics processor 1340 includes one or more shader core(s) 1355A-1355N (e.g., 1455A, 1355B, 1355C, 1355D, 1355E, 1355F, through 1355N-1, and 1355N), which provides for a unified shader core architecture in which a single core or type or core can execute all types of programmable shader code, including shader program code to implement vertex shaders, fragment shaders, and/or compute shaders. The unified shader core architecture is also configurable to execute direct compiled high-level GPGPU programs (e.g., CUDA). The exact number of shader cores present can vary among embodiments and implementations. Additionally, graphics processor 1340 includes an inter-core task manager 1345, which acts as a thread dispatcher to dispatch execution threads to one or more shader cores 1355A-1355N and a tiling unit 1358 to accelerate tiling operations for tile-based rendering, in which rendering operations for a scene are subdivided in image space, for example to exploit local spatial coherence within a scene or to optimize use of internal caches.

Machine Learning Overview

A machine learning algorithm is an algorithm that can learn based on a set of data. Embodiments of machine learning algorithms can be designed to model high-level abstractions within a data set. For example, image recognition algorithms can be used to determine which of several categories to which a given input belong; regression algorithms can output a numerical value given an input; and pattern recognition algorithms can be used to generate translated text or perform text to speech and/or speech recognition.

An exemplary type of machine learning algorithm is a neural network. There are many types of neural networks; a simple type of neural network is a feedforward network. A feedforward network may be implemented as an acyclic graph in which the nodes are arranged in layers. Typically, a feedforward network topology includes an input layer and an output layer that are separated by at least one hidden layer. The hidden layer transforms input received by the input layer into a representation that is useful for generating output in the output layer. The network nodes are fully connected via edges to the nodes in adjacent layers, but there are no edges between nodes within each layer. Data received at the nodes of an input layer of a feedforward network are propagated (i.e., “fed forward”) to the nodes of the output layer via an activation function that calculates the states of the nodes of each successive layer in the network based on coefficients (“weights”) respectively associated with each of the edges connecting the layers. Depending on the specific model being represented by the algorithm being executed, the output from the neural network algorithm can take various forms.

Before a machine learning algorithm can be used to model a particular problem, the algorithm is trained using a training data set. Training a neural network involves selecting a network topology, using a set of training data representing a problem being modeled by the network, and adjusting the weights until the network model performs with a minimal error for all instances of the training data set. For example, during a supervised learning training process for a neural network, the output produced by the network in response to the input representing an instance in a training data set is compared to the “correct” labeled output for that instance, an error signal representing the difference between the output and the labeled output is calculated, and the weights associated with the connections are adjusted to minimize that error as the error signal is backward propagated through the layers of the network. The network is considered “trained” when the errors for each of the outputs generated from the instances of the training data set are minimized.

The accuracy of a machine learning algorithm can be affected significantly by the quality of the data set used to train the algorithm. The training process can be computationally intensive and may require a significant amount of time on a conventional general-purpose processor. Accordingly, parallel processing hardware is used to train many types of machine learning algorithms. This is particularly useful for optimizing the training of neural networks, as the computations performed in adjusting the coefficients in neural networks lend themselves naturally to parallel implementations. Specifically, many machine learning algorithms and software applications have been adapted to make use of the parallel processing hardware within general-purpose graphics processing devices.

FIG. 14 is a generalized diagram of a machine learning software stack 1400. A machine learning application 1402 can be configured to train a neural network using a training dataset or to use a trained deep neural network to implement machine intelligence. The machine learning application 1402 can include training and inference functionality for a neural network and/or specialized software that can be used to train a neural network before deployment. The machine learning application 1402 can implement any type of machine intelligence including but not limited to image recognition, mapping and localization, autonomous navigation, speech synthesis, medical imaging, or language translation.

Hardware acceleration for the machine learning application 1402 can be enabled via a machine learning framework 1404. The machine learning framework 1404 can provide a library of machine learning primitives. Machine learning primitives are basic operations that are commonly performed by machine learning algorithms. Without the machine learning framework 1404, developers of machine learning algorithms would be required to create and optimize the main computational logic associated with the machine learning algorithm, then re-optimize the computational logic as new parallel processors are developed. Instead, the machine learning application can be configured to perform the necessary computations using the primitives provided by the machine learning framework 1404. Exemplary primitives include tensor convolutions, activation functions, and pooling, which are computational operations that are performed while training a convolutional neural network (CNN). The machine learning framework 1404 can also provide primitives to implement basic linear algebra subprograms performed by many machine-learning algorithms, such as matrix and vector operations.

The machine learning framework 1404 can process input data received from the machine learning application 1402 and generate the appropriate input to a compute framework 1406. The compute framework 1406 can abstract the underlying instructions provided to the GPGPU driver 1408 to enable the machine learning framework 1404 to take advantage of hardware acceleration via the GPGPU hardware 1410 without requiring the machine learning framework 1404 to have intimate knowledge of the architecture of the GPGPU hardware 1410. Additionally, the compute framework 1406 can enable hardware acceleration for the machine learning framework 1404 across a variety of types and generations of the GPGPU hardware 1410.

Machine Learning Neural Network Implementations

The computing architecture provided by embodiments described herein can be configured to perform the types of parallel processing that is particularly suited for training and deploying neural networks for machine learning. A neural network can be generalized as a network of functions having a graph relationship. As is known in the art, there are a variety of types of neural network implementations used in machine learning. One exemplary type of neural network is the feedforward network, as previously described.

A second exemplary type of neural network is the Convolutional Neural Network (CNN). A CNN is a specialized feedforward neural network for processing data having a known, grid-like topology, such as image data. Accordingly, CNNs are commonly used for compute vision and image recognition applications, but they also may be used for other types of pattern recognition such as speech and language processing. The nodes in the CNN input layer are organized into a set of “filters” (feature detectors inspired by the receptive fields found in the retina), and the output of each set of filters is propagated to nodes in successive layers of the network. The computations for a CNN include applying the convolution mathematical operation to each filter to produce the output of that filter. Convolution is a specialized kind of mathematical operation performed by two functions to produce a third function that is a modified version of one of the two original functions. In convolutional network terminology, the first function to the convolution can be referred to as the input, while the second function can be referred to as the convolution kernel. The output may be referred to as the feature map. For example, the input to a convolution layer can be a multidimensional array of data that defines the various color components of an input image. The convolution kernel can be a multidimensional array of parameters, where the parameters are adapted by the training process for the neural network.

Recurrent neural networks (RNNs) are a family of feedforward neural networks that include feedback connections between layers. RNNs enable modeling of sequential data by sharing parameter data across different parts of the neural network. The architecture for a RNN includes cycles. The cycles represent the influence of a present value of a variable on its own value at a future time, as at least a portion of the output data from the RNN is used as feedback for processing subsequent input in a sequence. This feature makes RNNs particularly useful for language processing due to the variable nature in which language data can be composed.

The figures described below present exemplary feedforward, CNN, and RNN networks, as well as describe a general process for respectively training and deploying each of those types of networks. It will be understood that these descriptions are exemplary and non-limiting as to any specific embodiment described herein and the concepts illustrated can be applied generally to deep neural networks and machine learning techniques in general.

The exemplary neural networks described above can be used to perform deep learning. Deep learning is machine learning using deep neural networks. The deep neural networks used in deep learning are artificial neural networks composed of multiple hidden layers, as opposed to shallow neural networks that include only a single hidden layer. Deeper neural networks are generally more computationally intensive to train. However, the additional hidden layers of the network enable multistep pattern recognition that results in reduced output error relative to shallow machine learning techniques.

Deep neural networks used in deep learning typically include a front-end network to perform feature recognition coupled to a back-end network which represents a mathematical model that can perform operations (e.g., object classification, speech recognition, etc.) based on the feature representation provided to the model. Deep learning enables machine learning to be performed without requiring hand crafted feature engineering to be performed for the model. Instead, deep neural networks can learn features based on statistical structure or correlation within the input data. The learned features can be provided to a mathematical model that can map detected features to an output. The mathematical model used by the network is generally specialized for the specific task to be performed, and different models will be used to perform different task.

Once the neural network is structured, a learning model can be applied to the network to train the network to perform specific tasks. The learning model describes how to adjust the weights within the model to reduce the output error of the network. Backpropagation of errors is a common method used to train neural networks. An input vector is presented to the network for processing. The output of the network is compared to the desired output using a loss function and an error value is calculated for each of the neurons in the output layer. The error values are then propagated backwards until each neuron has an associated error value which roughly represents its contribution to the original output. The network can then learn from those errors using an algorithm, such as the stochastic gradient descent algorithm, to update the weights of the of the neural network.

FIG. 15A-15B illustrate an exemplary convolutional neural network. FIG. 15A illustrates various layers within a CNN. As shown in FIG. 15A, an exemplary CNN used to model image processing can receive input 1502 describing the red, green, and blue (RGB) components of an input image. The input 1502 can be processed by multiple convolutional layers (e.g., first convolutional layer 1504, second convolutional layer 1506). The output from the multiple convolutional layers may optionally be processed by a set of fully connected layers 1508. Neurons in a fully connected layer have full connections to all activations in the previous layer, as previously described for a feedforward network. The output from the fully connected layers 1508 can be used to generate an output result from the network. The activations within the fully connected layers 1508 can be computed using matrix multiplication instead of convolution. Not all CNN implementations make use of fully connected layers 1508. For example, in some implementations the second convolutional layer 1506 can generate output for the CNN.

The convolutional layers are sparsely connected, which differs from traditional neural network configuration found in the fully connected layers 1508. Traditional neural network layers are fully connected, such that every output unit interacts with every input unit. However, the convolutional layers are sparsely connected because the output of the convolution of a field is input (instead of the respective state value of each of the nodes in the field) to the nodes of the subsequent layer, as illustrated. The kernels associated with the convolutional layers perform convolution operations, the output of which is sent to the next layer. The dimensionality reduction performed within the convolutional layers is one aspect that enables the CNN to scale to process large images.

FIG. 15B illustrates exemplary computation stages within a convolutional layer of a CNN. Input to a convolutional layer 1512 of a CNN can be processed in three stages of a convolutional layer 1514. The three stages can include a convolution stage 1516, a detector stage 1518, and a pooling stage 1520. The convolution layer 1514 can then output data to a successive convolutional layer. The final convolutional layer of the network can generate output feature map data or provide input to a fully connected layer, for example, to generate a classification value for the input to the CNN.

In the convolution stage 1516 performs several convolutions in parallel to produce a set of linear activations. The convolution stage 1516 can include an affine transformation, which is any transformation that can be specified as a linear transformation plus a translation. Affine transformations include rotations, translations, scaling, and combinations of these transformations. The convolution stage computes the output of functions (e.g., neurons) that are connected to specific regions in the input, which can be determined as the local region associated with the neuron. The neurons compute a dot product between the weights of the neurons and the region in the local input to which the neurons are connected. The output from the convolution stage 1516 defines a set of linear activations that are processed by successive stages of the convolutional layer 1514.

The linear activations can be processed by a detector stage 1518. In the detector stage 1518, each linear activation is processed by a non-linear activation function. The non-linear activation function increases the nonlinear properties of the overall network without affecting the receptive fields of the convolution layer. Several types of non-linear activation functions may be used. One particular type is the rectified linear unit (ReLU), which uses an activation function defined as θ(x)=max (0, x), such that the activation is thresholded at zero.

The pooling stage 1520 uses a pooling function that replaces the output of the second convolutional layer 1506 with a summary statistic of the nearby outputs. The pooling function can be used to introduce translation invariance into the neural network, such that small translations to the input do not change the pooled outputs. Invariance to local translation can be useful in scenarios where the presence of a feature in the input data is more important than the precise location of the feature. Various types of pooling functions can be used during the pooling stage 1520, including max pooling, average pooling, and 12-norm pooling. Additionally, some CNN implementations do not include a pooling stage. Instead, such implementations substitute and additional convolution stage having an increased stride relative to previous convolution stages.

The output from the convolutional layer 1514 can then be processed by the next layer 1522. The next layer 1522 can be an additional convolutional layer or one of the fully connected layers 1508. For example, the first convolutional layer 1504 of FIG. 15A can output to the second convolutional layer 1506, while the second convolutional layer can output to a first layer of the fully connected layers 1508.

FIG. 16 illustrates an exemplary recurrent neural network. In a recurrent neural network (RNN), the previous state of the network influences the output of the current state of the network. RNNs can be built in a variety of ways using a variety of functions. The use of RNNs generally revolves around using mathematical models to predict the future based on a prior sequence of inputs. For example, an RNN may be used to perform statistical language modeling to predict an upcoming word given a previous sequence of words. The illustrated RNN 1600 can be described as having an input layer 1602 that receives an input vector, hidden layers 1604 to implement a recurrent function, a feedback mechanism 1605 to enable a ‘memory’ of previous states, and an output layer 1606 to output a result. The RNN 1600 operates based on time-steps. The state of the RNN at a given time step is influenced based on the previous time step via the feedback mechanism 1605. For a given time step, the state of the hidden layers 1604 is defined by the previous state and the input at the current time step. An initial input (x₁) at a first time step can be processed by the hidden layer 1604. A second input (x₂) can be processed by the hidden layer 1604 using state information that is determined during the processing of the initial input (x₁). A given state can be computed as s_t=θ(Ux_t+Ws_t−1), where U and W are parameter matrices. The function ƒ is generally a nonlinearity, such as the hyperbolic tangent function (Tanh) or a variant of the rectifier function θ(x)=max(0, x). However, the specific mathematical function used in the hidden layers 1604 can vary depending on the specific implementation details of the RNN 1600.

In addition to the basic CNN and RNN networks described, variations on those networks may be enabled. One example RNN variant is the long short-term memory (LSTM) RNN. LSTM RNNs are capable of learning long-term dependencies that may be necessary for processing longer sequences of language. A variant on the CNN is a convolutional deep belief network, which has a structure similar to a CNN and is trained in a manner similar to a deep belief network. A deep belief network (DBN) is a generative neural network that is composed of multiple layers of stochastic (random) variables. DBNs can be trained layer-by-layer using greedy unsupervised learning. The learned weights of the DBN can then be used to provide pre-train neural networks by determining an optimal initial set of weights for the neural network.

FIG. 17 illustrates training and deployment of a deep neural network. Once a given network has been structured for a task the neural network is trained using a training dataset 1702. Various training frameworks have been developed to enable hardware acceleration of the training process. For example, the machine learning framework 1404 of FIG. 14 may be configured as a training framework 1704. The training framework 1704 can hook into an untrained neural network 1706 and enable the untrained neural net to be trained using the parallel processing resources described herein to generate a trained neural network 1708. To start the training process the initial weights may be chosen randomly or by pre-training using a deep belief network. The training cycle then be performed in either a supervised or unsupervised manner.

Supervised learning is a learning method in which training is performed as a mediated operation, such as when the training dataset 1702 includes input paired with the desired output for the input, or where the training dataset includes input having known output and the output of the neural network is manually graded. The network processes the inputs and compares the resulting outputs against a set of expected or desired outputs. Errors are then propagated back through the system. The training framework 1704 can adjust to adjust the weights that control the untrained neural network 1706. The training framework 1704 can provide tools to monitor how well the untrained neural network 1706 is converging towards a model suitable to generating correct answers based on known input data. The training process occurs repeatedly as the weights of the network are adjusted to refine the output generated by the neural network. The training process can continue until the neural network reaches a statistically desired accuracy associated with a trained neural network 1708. The trained neural network 1708 can then be deployed to implement any number of machine learning operations to generate an inference result 1714 based on input of new data 1712.

Unsupervised learning is a learning method in which the network attempts to train itself using unlabeled data. Thus, for unsupervised learning the training dataset 1702 will include input data without any associated output data. The untrained neural network 1706 can learn groupings within the unlabeled input and can determine how individual inputs are related to the overall dataset. Unsupervised training can be used to generate a self-organizing map, which is a type of trained neural network 1708 capable of performing operations useful in reducing the dimensionality of data. Unsupervised training can also be used to perform anomaly detection, which allows the identification of data points in an input dataset that deviate from the normal patterns of the data.

Variations on supervised and unsupervised training may also be employed. Semi-supervised learning is a technique in which in the training dataset 1702 includes a mix of labeled and unlabeled data of the same distribution. Incremental learning is a variant of supervised learning in which input data is continuously used to further train the model. Incremental learning enables the trained neural network 1708 to adapt to the new data 1712 without forgetting the knowledge instilled within the network during initial training.

Whether supervised or unsupervised, the training process for particularly deep neural networks may be too computationally intensive for a single compute node. Instead of using a single compute node, a distributed network of computational nodes can be used to accelerate the training process.

FIG. 18 is a block diagram illustrating distributed learning. Distributed learning is a training model that uses multiple distributed computing nodes to perform supervised or unsupervised training of a neural network. The distributed computational nodes can each include one or more host processors and one or more of the general-purpose processing nodes. As illustrated, distributed learning can be performed model parallelism 1802, data parallelism 1804, or a combination of model and data parallelism 1804.

In model parallelism 1802, different computational nodes in a distributed system can perform training computations for different parts of a single network. For example, each layer of a neural network can be trained by a different processing node of the distributed system. The benefits of model parallelism include the ability to scale to particularly large models. Splitting the computations associated with different layers of the neural network enables the training of very large neural networks in which the weights of all layers would not fit into the memory of a single computational node. In some instances, model parallelism can be particularly useful in performing unsupervised training of large neural networks.

In data parallelism 1804, the different nodes of the distributed network have a complete instance of the model and each node receives a different portion of the data. The results from the different nodes are then combined. While different approaches to data parallelism are possible, data parallel training approaches all require a technique of combining results and synchronizing the model parameters between each node. Exemplary approaches to combining data include parameter averaging and update-based data parallelism. Parameter averaging trains each node on a subset of the training data and sets the global parameters (e.g., weights, biases) to the average of the parameters from each node. Parameter averaging uses a central parameter server that maintains the parameter data. Update based data parallelism is similar to parameter averaging except that instead of transferring parameters from the nodes to the parameter server, the updates to the model are transferred. Additionally, update-based data parallelism can be performed in a decentralized manner, where the updates are compressed and transferred between nodes.

Combined model and data parallelism 1806 can be implemented, for example, in a distributed system in which each computational node includes multiple GPUs. Each node can have a complete instance of the model with separate GPUs within each node are used to train different portions of the model.

Distributed training has increased overhead relative to training on a single machine. However, the parallel processors and GPGPUs described herein can each implement various techniques to reduce the overhead of distributed training, including techniques to enable high bandwidth GPU-to-GPU data transfer and accelerated remote data synchronization.

Exemplary Machine Learning Applications

Machine learning can be applied to solve a variety of technological problems, including but not limited to computer vision, autonomous driving and navigation, speech recognition, and language processing. Computer vision has traditionally been one of the most active research areas for machine learning applications. Applications of computer vision range from reproducing human visual abilities, such as recognizing faces, to creating new categories of visual abilities. For example, computer vision applications can be configured to recognize sound waves from the vibrations induced in objects visible in a video. Parallel processor accelerated machine learning enables computer vision applications to be trained using significantly larger training dataset than previously feasible and enables inferencing systems to be deployed using low power parallel processors.

Parallel processor accelerated machine learning has autonomous driving applications including lane and road sign recognition, obstacle avoidance, navigation, and driving control. Accelerated machine learning techniques can be used to train driving models based on datasets that define the appropriate responses to specific training input. The parallel processors described herein can enable rapid training of the increasingly complex neural networks used for autonomous driving solutions and enables the deployment of low power inferencing processors in a mobile platform suitable for integration into autonomous vehicles.

Parallel processor accelerated deep neural networks have enabled machine learning approaches to automatic speech recognition (ASR). ASR includes the creation of a function that computes the most probable linguistic sequence given an input acoustic sequence. Accelerated machine learning using deep neural networks have enabled the replacement of the hidden Markov models (HMMs) and Gaussian mixture models (GMMs) previously used for ASR.

Parallel processor accelerated machine learning can also be used to accelerate natural language processing. Automatic learning procedures can make use of statistical inference algorithms to produce models that are robust to erroneous or unfamiliar input. Exemplary natural language processor applications include automatic machine translation between human languages.

The parallel processing platforms used for machine learning can be divided into training platforms and deployment platforms. Training platforms are generally highly parallel and include optimizations to accelerate multi-GPU single node training and multi-node, multi-GPU training, while deployed machine learning (e.g., inferencing) platforms generally include lower power parallel processors suitable for use in products such as cameras, autonomous robots, and autonomous vehicles.

Object Segmented Style Transfer

Embodiments described herein provide techniques to enable real-time, temporally consistent object segmented style transfer in media and gaming. Style transfer can be performed in a segmented manner on a per-region or per-object basis. A user can apply a selected style to, for example, a player character in a scene, a skybox, or specific objects. Region specific stylization can be applied, for example, to the background of the scene. In one embodiment, style transfer can be performed based on information provided by a game engine. In one embodiment, style transfer can be performed via a shim layer that intercepts certain calls to the graphics API that is used by the game engine. In one embodiment, style transfer can be performed for a scene by performing object recognition on the scene to detect specific objects or segments for which a style should be applied. Object recognition can also be used to apply stylization to a video frame, which may lack specific markup data. A scene can also be segmented within a video, which lacks markup data, using region specific stylization or by generating an object map of items in the scene using object recognition techniques. Object recognition can be performed using a variety of machine learning models. In one embodiment, general object recognition models can be used. In one embodiment, domain-specific models are used that are trained for the domain in which recognition will be applied.

FIG. 19 illustrates one embodiment of a computing device 1900 comprising a style transfer mechanism 1910. For example, in one embodiment, style transfer mechanism 1910 may be employed or hosted by computing device 1900. Computing device 1900 represents a communication and data processing device including or representing any number and type of smart devices, such as (without limitation) smart command devices or intelligent personal assistants, home/office automation system, home appliances (e.g., washing machines, television sets, etc.), mobile devices (e.g., smartphones, tablet computers, etc.), gaming devices, handheld devices, wearable devices (e.g., smartwatches, smart bracelets, etc.), virtual reality (VR) devices, head-mounted display (HMDs), Internet of Things (IoT) devices, laptop computers, desktop computers, server computers, set-top boxes (e.g., Internet-based cable television set-top boxes, etc.), global positioning system (GPS)-based devices, etc.

In some embodiments, computing device 1900 may include (without limitation) autonomous machines or artificially intelligent agents, such as a mechanical agents or machines, electronics agents or machines, virtual agents or machines, electro-mechanical agents or machines, etc. Examples of autonomous machines or artificially intelligent agents may include (without limitation) robots, autonomous vehicles (e.g., self-driving cars, self-flying planes, self-sailing boats, etc.), autonomous equipment (self-operating construction vehicles, self-operating medical equipment, etc.), and/or the like. Further, “autonomous vehicles” are not limed to automobiles but that they may include any number and type of autonomous machines, such as robots, autonomous equipment, household autonomous devices, and/or the like, and any one or more tasks or operations relating to such autonomous machines may be interchangeably referenced with autonomous driving.

Further, for example, computing device 1900 may include a cloud computing platform consisting of a plurality of server computers, where each server computer employs or hosts a multifunction perceptron mechanism. For example, automatic ISP tuning may be performed using component, system, and architectural setups described earlier in this document. For example, some of the aforementioned types of devices may be used to implement a custom learned procedure, such as using field-programmable gate arrays (FPGAs), etc.

Further, for example, computing device 1900 may include a computer platform hosting an integrated circuit (“IC”), such as a system on a chip (“SoC” or “SOC”), integrating various hardware and/or software components of computing device 1900 on a single chip.

As illustrated, in one embodiment, computing device 1900 may include any number and type of hardware and/or software components, such as (without limitation) graphics processing unit 1920 (“GPU” or simply “graphics processor”), graphics driver 1906 (also referred to as “GPU driver”, “graphics driver logic”, “driver logic”, user-mode driver (UMD), UMD, user-mode driver framework (UMDF), UMDF, or simply “driver”), central processing unit 1922 (“CPU” or simply “application processor”), memory 1924, network devices, drivers, or the like, as well as input/output (I/O) sources 1928, such as touchscreens, touch panels, touch pads, virtual or regular keyboards, virtual or regular mice, ports, connectors, etc. Computing device 1900 may include operating system (OS) serving as an interface between hardware and/or physical resources of computing device 1900 and a user.

It is to be appreciated that a lesser or more equipped system than the example described above may be preferred for certain implementations. Therefore, the configuration of computing device 1900 may vary from implementation to implementation depending upon numerous factors, such as price constraints, performance requirements, technological improvements, or other circumstances.

Embodiments may be implemented as any or a combination of: one or more microchips or integrated circuits interconnected using a parentboard, hardwired logic, software stored by a memory device and executed by a microprocessor, firmware, an application specific integrated circuit (ASIC), and/or a field programmable gate array (FPGA). The terms “logic”, “module”, “component”, “engine”, and “mechanism” may include, by way of example, software or hardware and/or a combination thereof, such as firmware.

In one embodiment, as illustrated, style transfer mechanism 1910 may be hosted by graphics driver 1906 of computing device 1900. In another embodiment, style transfer mechanism 1910 may be hosted by or be part of operating system 1904 of computing device 1900. In another embodiment, style transfer mechanism 1910 may be hosted or facilitated by memory 1924. For example, the style transfer mechanism 1910 can be provided by a software development kit (SDK). In yet another embodiment, style transfer mechanism 1910 may be hosted by or part of graphics processing unit 1920 (“GPU” or simply graphics processor”) or firmware of graphics processor 1920. For example, style transfer mechanism 1910 may be embedded in or implemented as part of the processing hardware of graphics processor 1920. Similarly, in yet another embodiment, style transfer mechanism 1910 may be hosted by or part of central processing unit 1922 (“CPU” or simply “application processor”). For example, style transfer mechanism 1910 may be embedded in or implemented as part of the processing hardware of application processor 1922.

In yet another embodiment, style transfer mechanism 1910 may be hosted by or part of any number and type of components of computing device 1900, such as a portion of style transfer mechanism 1910 may be hosted by or part of operating system 1904, another portion may be hosted by or part of graphics processor 1920, another portion may be hosted by or part of application processor 1922, while one or more portions of style transfer mechanism 1910 may be hosted by or part of operating system 1904 and/or any number and type of devices of computing device 1900. It is contemplated that embodiments are not limited to certain implementation or hosting of style transfer mechanism 1910 and that one or more portions or components of style transfer mechanism 1910 may be employed or implemented as hardware, software, or any combination thereof, such as firmware.

The style transfer mechanism 1910 includes a stylizer 1912 to apply a style transfer to a scene or frame and an instance segmentation unit 1914 that enables instance segmentation that enables the generation of a segmented instance mask for the scene or frame. The stylizer 1912 provides temporally consistent, real-time style transfer via the use of an architecture with temporal losses and a feedforward operation. The stylizer 1912 can include one or more machine learning models to apply a style to a portion of a scene or a frame. In one embodiment, the stylizer 1912 includes or communicates with adapted variant of the ReCoNet style transfer network, which is described in further detail in FIG. 20. Style transfer can be applied to video media frames or 3D game scenes, as shown in FIG. 21.

The instance segmentation unit 1920 includes multiple sub-units, including an 3D engine interface 1916, an API shim layer 1917, a window-based stylization unit 1918, and a predictive mask generator 1919. The instance segmentation unit 1920 generates instance masks that are used to apply instance stylization to a rendered 3D scene of a game or a video frame. Stylization can be performed in a temporally consistent manner across multiple frames. The 3D engine interface 1916 enables the generation of instance masks based on data provided by a game engine during the rendering of the 3D scene. In one embodiment, an additional geometry buffer (G-buffer) pass is applied that uses a custom shader to generate the object based semantic map for a scene using layer IDs for a group of objects or based on an individual object ID specified in the scene, as shown in FIG. 22A and FIG. 22B. This object-based semantic map can be used for a specific object masks or multiple object masks if performing multiple object style transfers per frame.

The API shim layer 1917 can be used for game engines that are not configured to interact with the instance segmentation unit 1914. Calls to the 3D API library used by a game engine can pass through the API shim layer 1917, which provides information that the instance segmentation unit 1914 can used to generate object masks for a scene. The API shim layer 1917 can be used to add an additional render pass during the G-buffer phase to capture per-pixel object segmentation for each frame rendered for the game. A separate render target can be created that identifies the pixels associated with a specific object or set of objects for which a particular style will be applied. In one embodiment, the API shim layer 1917 can operate at the device driver interface (DDI) layer, for example, in an application intelligence layer in the user mode driver (e.g., user mode graphics driver 1026) that performs operations for specific game titles. The pipeline for stylizing objects in a frame in this manner is illustrated in FIG. 23. Characters can be visualized using object masks rendered from a game engine to perform stylization in real-time. Additionally, a combination of G-buffer information and a segmented per pixel object ID can be used to further control where the stylization should be performed in the frame. For example, stylization can be applied based on the albedo or specular contribution within the object.

In one embodiment, the instance segmentation unit 1914 includes a window-based stylization unit 1918 to apply window-based (e.g., region-based) stylization to a scene or frame of game or media content. The window-based stylization unit 1918 can configure stylization for a scene to be applied only to a specific portion of the frame or limit the portion of the frame in which object classification routines are applied for specific objects. For example, for the stylization shown in FIG. 24, operations to perform instance-based stylization for a skybox can be focused to the upper portion of the frame, while instance-based stylization for the vehicle can be focused to the lower region of the frame. Where the 3D engine interface 1916 is present and the game engine can provide camera view hints to the style transfer mechanism 1910, the window-based stylization unit 1918 can adjust the heuristics applied for window-based stylization based on changes in the camera view for a rendered scene. In one embodiment, depending on the region being stylized, the window-based stylization unit 1918 can improve the performance of the stylization by cropping the frame to a reduced number of pixels before stylization is applied.

Where game engine information or other sources of markup data for a scene is not present, then the predictive mask generator 1919 can be applied. The predictive mask generator 1919 has particular applicability for media frames. While motion vector data for encoded media can assist in the temporal stability of stylization, metadata may not be present about the objects within the frame to enable instance segmentation for the frame. The predictive mask generator 1919 can apply one or more models from of a suite of neural network models to perform real-time instance segmentation for a frame and apply style transfers based on instance masks produced by the neural network models. Exemplary neural network models for one embodiment include adaptations of the YOLACT and CondInst CNNs. In some scenarios, the U²-Net model can be used to generate an instance mask. However, embodiments are not limited to the use of any specific model for instance segmentation.

In one embodiment, the style transfer mechanism 1910 can be extended to enable per instance stylization based on semantically meaningful correspondence between a style image and the content image for which the style is to be applied. Semantically meaningful correspondences between the style image and the content image can be used to select the style of the style image that will be applied to the content image. For example, in a painting of a city, the building can have a specific style in contrast with other buildings or a river in the painting. The style from the style image that is applied to the content image can differ based on the object that is being stylized in the content image.

FIG. 20 illustrates a style transfer network 2000, according to an embodiment. The style transfer network 2000 can be used by the stylizer 1912 to apply style transfer to an input image according to instance segmentation performed by the instance segmentation unit 1914. In one embodiment, the style transfer network 2000 is a reimplementation of the ReCoNet neural network. However, other embodiments make use of other neural networks to perform style transfer on an input image. The style transfer network 2000 includes three primary modules, an encoder 2010 that converts an input image 2001 (I_t) into encoded feature maps 2002 (F_t), a decoder 2015 that generates stylized output images 2003 (O_t) from the encoded feature maps, and a VGG16 loss network 2025 to capture the perceptual style of the transfer target. An occlusion mask 2004 (M_t) and optical flow 2006 (W_t) are provided as temporal loss guidance to facilitate frame-to-frame temporal consistency. During training, previous frame prediction results 2020 (I_t−1, F_t−1, O_t−1) are used during the training process for the style transfer network 2000. A style image 2018 is input to the VGG loss network 2025. The style image 2018 has a specific artistic style that will be transferred to a target and is used to generate style loss for training the style transfer network 2000 to transfer an artistic style. Content and style losses are computed based on the generated image features to update the weights of the style transfer network 2000. In inference mode, previous frame prediction results 2020 and the VGG lost network 2025 are not used and styled output images 2003 are generated using the encoder 2010 and the decoder 2015.

FIG. 21 illustrates style transfer for video media frames and 3D game scenes, according to an embodiment. Style transfer can be applied for media streams 2110 to generate stylized media 2112 and to game output 2120 to generate stylized game output 2122. In one embodiment, style transfer is applied for media streams 2110 via the use of a segmentation algorithm, while an object level mask from the rendering engine is utilized for game output 2120. In one embodiment, a segmentation algorithm can be applied for game output instead of or in addition to object level masks that are output from the rendering engine. Heuristics can also be applied via the window-based stylization unit 1918 to enable the character-specific localization to a window region instead of a complete frame.

FIG. 22A-22B illustrate embodiments of a ground truth rendered frame and object-based segmentation. In one embodiment, much of the cost of segmenting a frame rendered by a rendering engine for a 3D game can be eliminated using a segmentation mask that is generated by the rendering engine. The segmentation mask is an object based semantic map for a scene that is generated using a layer ID for a group of objects or based on individual object ID specified in the scene. A segmentation mask generated for objects in scene 2210 of FIG. 22A can be used to apply stylization for the scene 2210 to generate a stylized scene 2220 shown in FIG. 22B, in which a selected style is applied for each object or group of objects in the scene. The object-based semantic map can be generated, for example, using the 3D engine interface 1916 or the API shim layer 1917 of the instance segmentation unit 1914, as described above.

FIG. 23 illustrates style transfer via the use of object masks derived from game engine data, according to an embodiment. While rendering an input frame 2310 from a game, object masks 2320 can be generated for objects to be stylized within that frame. The object masks 2320 can be generated using the 3D engine interface 1916 or the API shim layer 1917 of the instance segmentation unit 1914. In one embodiment, a complete stylized frame 2330 is generated based a selected style for which the style transfer network 2000 is trained. The object masks 2320 can then be used to compose an output frame 2340 that includes elements from the input frame 2310 and stylized pixels from the complete stylized frame 2330. In one embodiment, the window-based stylization unit 1918 can be used to limit the stylized portion of the complete stylized frame 2330 to a window region surrounding the object masks 2320.

FIG. 24 illustrates one embodiment of a flow for localizing a primary object based on heuristics. In one embodiment, the window-based stylization unit 1918 can apply heuristics for localizing a primary object for stylization. For example, the style transfer mechanism 1910 can be configured to stylize a car in an auto racing game that renders output at 1080p. A 1080 frame rendered by the game can be provided as an input frame 2402. Heuristics applied by the window-based stylization unit 1918 can crop the frame to a 720p car frame 2404 that contains the car to be stylized. The 720p car frame 2404 can be generated by selecting a window of the input frame 2402 that includes the portions of the frame in which the car will be rendered. For example, for the auto racing game, it may be known beforehand that the car will generally be rendered in the lower-center portion of the frame and will not be rendered, for example within the portion of the frame that includes the sky. Similar heuristics can be applied, for example, for first person and third person shooters in which objects near the center of the screen may be stylized. Heuristics applied via the window-based stylization unit 1918 are of particular benefit when game engine information is not available to generate instance maps and the predictive mask generator 1919 is used. In one embodiment, in addition to using a heuristically cropped frame based on operations performed by the window-based stylization unit 1918, the inference time for segmentation performed by the predictive mask generator 1919 can be reduced by predicting the instance mask at a lower resolution than the input image. Contour estimation techniques can be used to smooth the boundaries of objects when generating segmentation masks at the lower resolution. The generated segmentation mask can then be upscaled to the target resolution.

FIG. 25 illustrates an operational flow for parallel transfer of multiple styles within a media frame, according to an embodiment. In one embodiment, multiple different styles can be applied to a video frame, which is an input image 2001 that is provided as input to the stylizer 1912. The stylizer 1912 can output multiple fully stylized frames 2504A-2504C, each having a different stylization. The instance segmentation unit 1914 can also be configured to output a series of single instance masks 2506A-2506C for each relevant object in the frame, for example, three separate individuals in the frame. The instance masks can then be applied to the stylized images to generate an output frame 2508 having a per instance stylized result in which each object instance has a separate stylization.

Stylization for Streaming and Cloud Gaming

FIG. 26 illustrates a data processing system 2600 for game streaming and/or cloud gaming system 2600, according to various embodiments. The client 2640 may generally be a consumer of content that is housed, processed, produced and/or rendered on the server 2620. Content can be stylized using per-instance stylization before transfer to the client 2640. The client 2640, which may be a desktop computer, notebook computer, tablet computer, convertible tablet, wearable device, mobile internet device, smartphone device, personal digital assistant, media player, etc., includes an NIC 2642 to receive the transmitted video stream from the server 2620. The NIC 2622, may include the physical layer and the basis for the software layer of the network interface in the client 2640 to facilitate communications over the network 2610. The client 2640 may also include a decoder 2644 that employs the same formatting/compression scheme of the encoder 2638. Thus, the decompressed video stream may be provided from the decoder 2644 to a video renderer 2646. The illustrated video renderer 2646 is coupled to a display 2648 that visually presents the graphics content. The client 2640 can conduct real-time interaction with streamed data that includes the collection of user input from an input device 2650 and delivery of the user input to the server 2620 via the network 2610.

In one embodiment, the illustrated server 2620 is a cloud gaming server that has the capacity to provide the graphics content to multiple clients simultaneously (e.g., by leveraging parallel and apportioned processing and rendering resources). In one embodiment, the server 2620 is live streaming server that can broadcast a stream that is received from a live stream device 2605 to multiple instances of the client 2640. In one embodiment, the server 2620 is configured as part of a visual compute cloud that includes end-to-end sample pipelines for graphics, media, analytics, and related activities. The server 2620 includes a graphics processor 2630 (e.g., GPU), a host processor 2624 (e.g., CPU) and a network interface controller (NIC) 2622. The NIC 2622 may receive a request from the client 2640 for graphics content, a request from a live stream device 2605 to broadcast a live stream to the client 2640, or a request from the client 2640 to perform cloud-based visual computing operations.

When configured for cloud gaming, a gaming application can be executed by the host processor 2624 and the graphics processor 2630, with the content streamed to the client 2640. The host processor 2624 may carry out high level operations such as, for example, determining position, collision and motion of objects in a given scene. Based on the high-level operations, the host processor 2624 may generate rendering commands that are combined with the scene data and executed by the graphics processor 2630. The rendering commands may cause the graphics processor 2630 to define scene geometry, shading, lighting, motion, texturing, camera parameters, etc., for scenes to be presented via the client 2640. The illustrated graphics processor 2630 includes a graphics renderer 2632 that executes rendering procedures according to the rendering commands generated by the host processor 2624. The output of the graphics renderer 2632 may be a stream of raw video frames that are provided to a frame capturer 2634. The illustrated frame capturer 2634 is coupled to an encoder 2638, which may compress/format the raw video stream for transmission over the network 2610. The encoder 2638 may use a wide variety of video compression algorithms such as, for example, the H.264 or H.265 standard from the International Telecommunication Union Telecommunication Standardization Sector (ITUT), the MPEG4 Advanced Video Coding (AVC) Standard from the International Organization for the Standardization/International Electrotechnical Commission (ISO/IEC), and so forth. In one embodiment, before a frame is encoded by the encoder 2638, a style transfer mechanism 2636 having components of the style transfer mechanism 1910 of FIG. 19 can apply a selected style to a frame according to techniques described herein.

When configured as a streaming server, the server 2620 can receive a video stream from the live stream device 2605 that is then re-transmitted to numerous instances of the client 2640. In one embodiment, per-instance stylization can be performed by the style transfer mechanism 2636 on a stream received from the live stream device 2605, which in various configurations can include decoding, stylizing, and reencoding the stream received from the live stream device 2605. In one embodiment, the live stream device 2605 includes an instance of the style transfer mechanism 2636 to perform per-instance stylization before the stream is transmitted to the server 2620 for broadcast.

When configured to perform cloud-based visual computing operations, the server 2620 can receive media input and perform visual computing operations on the media input. An exemplary server for cloud-based visual computing operations include a server configured for the Open Visual Cloud (OVC), which is a set of open source software stacks with full end-to-end sample pipelines for media, analytics, graphics, and immersive media that are optimized for cloud native deployment on commercial CPU architectures.

FIG. 27 illustrates an exemplary image for which per-instance stylization is applied via a cloud compute platform, according to an embodiment. In one embodiment, an image of a damaged vehicle can be processed via a cloud-based server and per-instance stylization can be applied to the image to highlight damaged portions of the vehicle to facilitate rapid assessment of damages, for example, for insurance purposes. The vehicle shown in the image of FIG. 27 includes multiple damaged regions, including a damaged hood region 2702, a damaged left grill region 2704, a damaged license plate region 2706, and a damaged bumper region 2708. To assist in the identification of these damaged regions, a cloud-based server can apply one or more stylizations to the damaged regions using instance-based stylization as described herein. Other use cases are also possible.

FIG. 28 illustrates a method 2800 of performing per-instance stylization, according to embodiments. The method 2800 can be used to transfer one or more styles from one or more style images onto a target image in real-time via the use of a style transfer mechanism 1910 as in FIG. 19. The style transfer can be performed on a per-region or per-object basis. Style transfer can be performed in real-time for media or gaming content. In various embodiments, style transfer can be performed for locally generated content, content that is received from a streaming service, or content that is to be streamed to multiple clients via a streaming service. Style transfer can also be enabled for cloud-based visual computing systems. In one embodiment, the style transfer mechanism 1910 is provided via a runtime library that provides functions, primitives, and subroutines to implement the operations described herein.

According to method 2800, the style transfer mechanism 1910 can receive an input frame to be stylized via a style transfer network (2801), such as the style transfer network 2000 of FIG. 20. The style transfer mechanism 1910 can then perform instance segmentation on the input frame to generate instance segmentation masks for content to be stylized (2802). Instance segmentation can be performed via the instance segmentation unit 1914 of FIG. 19. Multiple masks can be generated for each object or group of objects for which a specific style is to be applied. One or more identifiers for objects or groups of objects can be provided as input to the style transfer mechanism 1910 in conjunction with the input frame. For game content to be stylized, instance segmentation can be performed using the 3D engine interface 1916 or via the API shim layer 1917. For media content to be stylized, instance segmentation can be performed using predictive mask generator 1919. Heuristics applied by the window-based stylization unit 1918 can also be used to limit the region for which instance segmentation is to be performed to improve the performance of segmentation mask generation.

The style transfer mechanism 1910 can then use the style transfer network to generate a stylized frame for each style to transfer to the input frame via (2803). Multiple visual or artistic styles can be applied to a frame, with different objects being stylized differently. In one embodiment, multiple intermediate frames can be generated that stylizes an entire frame. In one embodiment, the window-based stylization unit 1918 can be used to limit the stylization region to the subset of the frame that contains the objects for which a specific style is to be applied. The style transfer mechanism 1910 can then merge stylized content with un-stylized content from the input frame to generate an output frame with per-instance stylization (2804). In one embodiment, the instance segmentation masks are used to composite, merge, or blend the stylized objects with un-stylized content in the original frame, allowing a single object or group of objects in the frame to be stylized and for multiple styles to be applied to a frame.

Tensor Acceleration Logic for Graphics and Machine Learning Workloads

FIG. 29 is a block diagram of a data processing system 2900, according to an embodiment. The data processing system 2900 is a heterogeneous processing system having a processor 2902, unified memory 2910, and a GPGPU 2920 including machine learning acceleration logic. The processor 2902 and the GPGPU 2920 can be any of the processors and GPGPU/parallel processors as described herein. For example, with additional reference to FIG. 1, application processor 2902 can be a variant of and/or share an architecture with the one or more processor(s) 102 or CPU 1922. The GPGPU 2920 can be a variant of and/or share an architecture with graphics processor(s) 108 of FIG. 1 or GPU 1920 of FIG. 19.

The processor 2902 can execute instructions for a compiler 2915 stored in system memory 2912. The compiler 2915 executes on the processor 2902 to compile source code 2914A into compiled code 2914B. The compiled code 2914B can include instructions that may be executed by the processor 2902 and/or instructions that may be executed by the GPGPU 2920. Compilation of instructions to be executed by the GPGPU can be facilitated using shader or compute program compilers, such as shader compiler. During compilation, the compiler 2915 can perform operations to insert metadata, including hints as to the level of data parallelism present in the compiled code 2914B and/or hints regarding the data locality associated with threads to be dispatched based on the compiled code 2914B. The compiler 2915 can include the information necessary to perform such operations or the operations can be performed with the assistance of a runtime library 2916. The runtime library 2916 can also assist the compiler 2915 in the compilation of the source code 2914A and can also include instructions that are linked at runtime with the compiled code 2914B to facilitate execution of the compiled instructions on the GPGPU 2920. The compiler 2915 can also facilitate register allocation for variables via a register allocator (RA) and generate load and store instructions to move data for variables between memory and the register assigned for the variable.

The unified memory 2910 represents a unified address space that may be accessed by the processor 2902 and the GPGPU 2920. The unified memory can include system memory 2912 as well as GPGPU memory 2918. The GPGPU memory 2918 is memory within an address pace of the GPGPU 2920 and can include some or all of system memory 2912. In one embodiment the GPGPU memory 2918 can also include at least a portion of any memory dedicated for use exclusively by the GPGPU 2920. In one embodiment, compiled code 2914B stored in system memory 2912 can be mapped into GPGPU memory 2918 for access by the GPGPU 2920.

The GPGPU 2920 includes multiple compute blocks 2924A-2924N, which can include one or more of a variety of processing resources described herein. The processing resources can be or include a variety of different computational resources such as, for example, execution units, compute units, streaming multiprocessors, graphics multiprocessors, or multi-core groups. In one embodiment the GPGPU 2920 additionally includes a tensor accelerator 2923 (e.g., matrix accelerator), which can include one or more special function compute units that are designed to accelerate a subset of matrix operations (e.g., dot product, etc.). The tensor accelerator 2923 may also be referred to as a tensor accelerator or tensor core. In one embodiment, logic components within the tensor accelerator 2923 may be distributed across the processing resources of the multiple compute blocks 2924A-2924N.

The GPGPU 2920 can also include a set of resources that can be shared by the compute blocks 2924A-2924N and the tensor accelerator 2923, including but not limited to a set of registers 2925, a power and performance module 2926, and a cache 2927. In one embodiment the registers 2925 include directly and indirectly accessible registers, where the indirectly accessible registers are optimized for use by the tensor accelerator 2923. The power and performance module 2926 can be configured to adjust power delivery and clock frequencies for the compute blocks 2924A-2924N to power gate idle components within the compute blocks 2924A-2924N. In various embodiments the cache 2927 can include an instruction cache and/or a lower-level data cache.

The GPGPU 2920 can additionally include an L3 data cache 2930, which can be used to cache data accessed from the unified memory 2910 by the tensor accelerator 2923 and/or the compute elements within the compute blocks 2924A-2924N. In one embodiment the L3 data cache 2930 includes shared local memory 2932 that can be shared by the compute elements within the compute blocks 2924A-2924N and the tensor accelerator 2923.

In one embodiment the GPGPU 2920 includes instruction handling logic, such as a fetch and decode unit 2921 and a scheduler controller 2922. The fetch and decode unit 2921 includes a fetch unit and decode unit to fetch and decode instructions for execution by one or more of the compute blocks 2924A-2924N or the tensor accelerator 2923. The instructions can be scheduled to the appropriate functional unit within the compute block 2924A-2924N or the tensor accelerator via the scheduler controller 2922. In one embodiment the scheduler controller 2922 is an ASIC configurable to perform advanced scheduling operations. In one embodiment the scheduler controller 2922 is a micro-controller or a low energy-per-instruction processing core capable of executing scheduler instructions loaded from a firmware module.

In one embodiment some functions to be performed by the compute blocks 2924A-2924N can be directly scheduled to or offloaded to the tensor accelerator 2923. In various embodiments the tensor accelerator 2923 includes processing element logic configured to efficiently perform matrix compute operations, such as multiply and add operations and dot product operations used by 3D graphics or compute shader programs. In one embodiment the tensor accelerator 2923 can be configured to accelerate operations used by machine learning frameworks. In one embodiment the tensor accelerator 2923 is an application specific integrated circuit explicitly configured to perform a specific set of parallel matrix multiplication and/or addition operations. In one embodiment the tensor accelerator 2923 is a field programmable gate array (FPGA) that provides fixed function logic that can updated between workloads. In one embodiment, the set of compute operations that can be performed by the tensor accelerator 2923 may be limited relative to the operations that can be performed by the compute block 2924A-2924N. However, the tensor accelerator 2923 can perform parallel tensor operations at a significantly higher throughput relative to the compute block 2924A-2924N.

Additional Exemplary Computing Device

FIG. 30 is a block diagram of a computing device 3000 including a graphics processor 3004, according to an embodiment. Versions of the computing device 3000 may be or be included within a communication device such as a set-top box (e.g., Internet-based cable television set-top boxes, etc.), global positioning system (GPS)-based devices, etc. The computing device 3000 may also be or be included within mobile computing devices such as cellular phones, smartphones, personal digital assistants (PDAs), tablet computers, laptop computers, e-readers, smart televisions, television platforms, wearable devices (e.g., glasses, watches, bracelets, smartcards, jewelry, clothing items, etc.), media players, etc. For example, in one embodiment, the computing device 3000 includes a mobile computing device employing an integrated circuit (“IC”), such as system on a chip (“SoC” or “SOC”), integrating various hardware and/or software components of computing device 3000 on a single chip. The computing device 3000 can be a computing device such as the processing system 100 as in of FIG. 1.

The computing device 3000 includes a graphics processor 3004. The graphics processor 3004 represents any graphics processor described herein. In one embodiment, the graphics processor 3004 includes a cache 3014, which can be a single cache or divided into multiple segments of cache memory, including but not limited to any number of L1, L2, L3, or L4 caches, render caches, depth caches, sampler caches, and/or shader unit caches. In one embodiment the cache 3014 may be a last level cache that is shared with the application processor 3006.

In one embodiment the graphics processor 3004 includes a graphics microcontroller that implements control and scheduling logic for the graphics processor. The control and scheduling logic can be firmware executed by the graphics microcontroller 3015. The firmware may be loaded at boot by the graphics driver logic 3022. The firmware may also be programmed to an electronically erasable programmable read only memory or loaded from a flash memory device within the graphics microcontroller 3015. The firmware may enable a GPU OS 3016 that includes device management/driver logic 3017, 3018, and a scheduler 3019. The GPU OS 3016 may also include a graphics memory manager 3020 that can supplement or replace the graphics memory manager 3021 within the graphics driver logic 3022.

The graphics processor 3004 also includes a GPGPU engine 3044 that includes one or more graphics engine(s), graphics processor cores, and other graphics execution resources as described herein. Such graphics execution resources can be presented in the forms including but not limited to execution units, shader engines, fragment processors, vertex processors, streaming multiprocessors, graphics processor clusters, or any collection of computing resources suitable for the processing of graphics resources or image resources or performing general purpose computational operations in a heterogeneous processor. The processing resources of the GPGPU engine 3044 can be included within multiple tiles of hardware logic connected to a substrate, as illustrated in FIG. 11 B-11D. The GPGPU engine 3044 can include GPU tiles 3045 that include graphics processing and execution resources, caches, samplers, etc. The GPU tiles 3045 may also include local volatile memory or can be coupled with one or more memory tiles for example, as shown in FIG. 3B-3C.

The GPGPU engine 3044 can also include one or more special tiles 3046 that include, for example, a non-volatile memory tile 3056, a network processor tile 3057, and/or a general-purpose compute tile 3058. The GPGPU engine 3044 also includes a matrix multiply accelerator 3060. The general-purpose compute tile 3058 may also include logic to accelerate matrix multiplication operations. The non-volatile memory tile 3056 can include non-volatile memory cells and controller logic. The controller logic of the non-volatile memory tile 3056 may be managed by one of device management/driver logic 3017, 3018. The network processor tile 3057 can include network processing resources that are coupled to a physical interface within the input/output (I/O) sources 3010 of the computing device 3000. The network processor tile 3057 may be managed by one or more of device management/driver logic 3017, 3018.

The matrix multiply accelerator 3060 is a modular scalable sparse matrix multiply accelerator as described herein. The matrix multiply accelerator 3060 can includes multiple processing paths, with each processing path including multiple pipeline stages. Each processing path can execute a separate instruction. In various embodiments, the matrix multiply accelerator 3060 can have architectural features of any one of more of the matrix multiply accelerators described herein. For example, in one embodiment, the matrix multiply accelerator 3060 is a four-deep systolic array with a feedback loop that is configurable to operate with a multiple of four number of logical stages (e.g., four, eight, twelve, sixteen, etc.). in one embodiment the matrix multiply accelerator 3060 includes one or more instances of a two-path matrix multiply accelerator with a four stage pipeline or a four-path matrix multiply accelerator with a two stage pipeline. In one embodiment the matrix multiply accelerator 3060 includes processing elements configured as a scalable sparse matrix multiply accelerator that is configured to operate only on non-zero values of at least one input matrix. Operations on entire columns or submatrices can be bypassed where block sparsity is present. The matrix multiply accelerator 3060 can also include any logic based on any combination of these embodiments. The matrix multiply accelerator 3060 is also configurable to perform matrix computations for the machine learning models described herein, including models configured to perform instance segmentation and style transfer. In various embodiments, the matrix multiply accelerator 3060 can accelerate matrix operations performed during training and inference.

As illustrated, in one embodiment, and in addition to the graphics processor 3004, the computing device 3000 may further include any number and type of hardware components and/or software components, including, but not limited to an application processor 3006, memory 3008, and input/output (I/O) sources 3010. The application processor 3006 can interact with a hardware graphics pipeline, as illustrated with reference to FIG. 3A, to share graphics pipeline functionality. Processed data is stored in a buffer in the hardware graphics pipeline and state information is stored in memory 3008. The resulting data can be transferred to a display controller for output via a display device, such as the display device 318 of FIG. 3A. The display device may be of various types, such as Cathode Ray Tube (CRT), Thin Film Transistor (TFT), Liquid Crystal Display (LCD), Organic Light Emitting Diode (OLED) array, etc., and may be configured to display information to a user via a graphical user interface.

The application processor 3006 can include one or processors, such as processor(s) 102 of FIG. 1 and may be the central processing unit (CPU) that is used at least in part to execute an operating system (OS) 3002 for the computing device 3000. The OS 3002 can serve as an interface between hardware and/or physical resources of the computing device 3000 and one or more users. The OS 3002 can include driver logic for various hardware devices in the computing device 3000. The driver logic can include graphics driver logic 3022, which can include the user mode graphics driver 1026 and/or kernel mode graphics driver 1029 of FIG. 10. The graphics driver logic can include a graphics memory manager 3021 to manage a virtual memory address space for the graphics processor 3004.

It is contemplated that in some embodiments the graphics processor 3004 may exist as part of the application processor 3006 (such as part of a physical CPU package) in which case, at least a portion of the memory 3008 may be shared by the application processor 3006 and graphics processor 3004, although at least a portion of the memory 3008 may be exclusive to the graphics processor 3004, or the graphics processor 3004 may have a separate store of memory. The memory 3008 may comprise a pre-allocated region of a buffer (e.g., frame buffer); however, it should be understood by one of ordinary skill in the art that the embodiments are not so limited, and that any memory accessible to the lower graphics pipeline may be used. The memory 3008 may include various forms of random-access memory (RAM) (e.g., SDRAM, SRAM, etc.) comprising an application that makes use of the graphics processor 3004 to render a desktop or 3D graphics scene. A memory controller hub, such as memory controller 116 of FIG. 1, may access data in the memory 3008 and forward it to graphics processor 3004 for graphics pipeline processing. The memory 3008 may be made available to other components within the computing device 3000. For example, any data (e.g., input graphics data) received from various I/O sources 3010 of the computing device 3000 can be temporarily queued into memory 3008 prior to their being operated upon by one or more processor(s) (e.g., application processor 3006) in the implementation of a software program or application. Similarly, data that a software program determines should be sent from the computing device 3000 to an outside entity through one of the computing system interfaces, or stored into an internal storage element, is often temporarily queued in memory 3008 prior to its being transmitted or stored.

The I/O sources can include devices such as touchscreens, touch panels, touch pads, virtual or regular keyboards, virtual or regular mice, ports, connectors, network devices, or the like, and can attach via a platform controller hub 130 as referenced in FIG. 1. Additionally, the I/O sources 3010 may include one or more I/O devices that are implemented for transferring data to and/or from the computing device 3000 (e.g., a networking adapter); or, for a large-scale non-volatile storage within the computing device 3000 (e.g., SSD/HDD). User input devices, including alphanumeric and other keys, may be used to communicate information and command selections to graphics processor 3004. Another type of user input device is cursor control, such as a mouse, a trackball, a touchscreen, a touchpad, or cursor direction keys to communicate direction information and command selections to GPU and to control cursor movement on the display device. Camera and microphone arrays of the computing device 3000 may be employed to observe gestures, record audio and video and to receive and transmit visual and audio commands.

The I/O sources 3010 can include one or more network interfaces. The network interfaces may include associated network processing logic and/or be coupled with the network processor tile 3057. The one or more network interface can provide access to a LAN, a wide area network (WAN), a metropolitan area network (MAN), a personal area network (PAN), Bluetooth, a cloud network, a cellular or mobile network (e.g., 3^rd Generation (3G), 4^th Generation (4G), 5^th Generation (5G), etc.), an intranet, the Internet, etc. Network interface(s) may include, for example, a wireless network interface having one or more antenna(e). Network interface(s) may also include, for example, a wired network interface to communicate with remote devices via network cable, which may be, for example, an Ethernet cable, a coaxial cable, a fiber optic cable, a serial cable, or a parallel cable.

Network interface(s) may provide access to a LAN, for example, by conforming to IEEE 802.11 standards, and/or the wireless network interface may provide access to a personal area network, for example, by conforming to Bluetooth standards. Other wireless network interfaces and/or protocols, including previous and subsequent versions of the standards, may also be supported. In addition to, or instead of, communication via the wireless LAN standards, network interface(s) may provide wireless communication using, for example, Time Division, Multiple Access (TDMA) protocols, Global Systems for Mobile Communications (GSM) protocols, Code Division, Multiple Access (CDMA) protocols, and/or any other type of wireless communications protocols.

It is to be appreciated that a lesser or more equipped system than the example described above may be preferred for certain implementations. Therefore, the configuration of the computing devices described herein may vary from implementation to implementation depending upon numerous factors, such as price constraints, performance requirements, technological improvements, or other circumstances. Examples include (without limitation) a mobile device, a personal digital assistant, a mobile computing device, a smartphone, a cellular telephone, a handset, a one-way pager, a two-way pager, a messaging device, a computer, a personal computer (PC), a desktop computer, a laptop computer, a notebook computer, a handheld computer, a tablet computer, a server, a server array or server farm, a web server, a network server, an Internet server, a work station, a mini-computer, a main frame computer, a supercomputer, a network appliance, a web appliance, a distributed computing system, multiprocessor systems, processor-based systems, consumer electronics, programmable consumer electronics, television, digital television, set top box, wireless access point, base station, subscriber station, mobile subscriber center, radio network controller, router, hub, gateway, bridge, switch, machine, or combinations thereof.

It is to be noted that terms like “node”, “computing node”, “server”, “server device”, “cloud computer”, “cloud server”, “cloud server computer”, “machine”, “host machine”, “device”, “computing device”, “computer”, “computing system”, and the like, may be used interchangeably throughout this document. It is to be further noted that terms like “application”, “software application”, “program”, “software program”, “package”, “software package”, and the like, may be used interchangeably throughout this document. Also, terms like “job”, “input”, “request”, “message”, and the like, may be used interchangeably throughout this document.

Embodiments may be provided, for example, as a computer program product which may include one or more machine-readable media having stored thereon machine-executable instructions that, when executed by one or more machines such as a computer, network of computers, or other electronic devices, may result in the one or more machines carrying out operations in accordance with embodiments described herein. A machine-readable medium may include, but is not limited to, floppy diskettes, optical disks, CD-ROMs (Compact Disc-Read Only Memories), and magneto-optical disks, ROMs, RAMs, EPROMs (Erasable Programmable Read Only Memories), EEPROMs (Electrically Erasable Programmable Read Only Memories), magnetic or optical cards, flash memory, or other type of media/machine-readable medium suitable for storing machine-executable instructions.

Moreover, embodiments may be downloaded as a computer program product, wherein the program may be transferred from a remote computer (e.g., a server) to a requesting computer (e.g., a client) by way of one or more data signals embodied in and/or modulated by a carrier wave or other propagation medium via a communication link (e.g., a modem and/or network connection).

Throughout the document, term “user” may be interchangeably referred to as “viewer”, “observer”, “person”, “individual”, “end-user”, and/or the like. It is to be noted that throughout this document, terms like “graphics domain” may be referenced interchangeably with “graphics processing unit”, “graphics processor”, or simply “GPU” and similarly, “CPU domain” or “host domain” may be referenced interchangeably with “computer processing unit”, “application processor”, or simply “CPU”.

It is to be noted that terms like “node”, “computing node”, “server”, “server device”, “cloud computer”, “cloud server”, “cloud server computer”, “machine”, “host machine”, “device”, “computing device”, “computer”, “computing system”, and the like, may be used interchangeably throughout this document. It is to be further noted that terms like “application”, “software application”, “program”, “software program”, “package”, “software package”, and the like, may be used interchangeably throughout this document. Also, terms like “job”, “input”, “request”, “message”, and the like, may be used interchangeably throughout this document.

References herein to “one embodiment,” “an embodiment,” “an example embodiment,” etc., indicate that the embodiment described may include a particular feature, structure, or characteristic, but every embodiment may not necessarily include the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that it is within the knowledge of one skilled in the art to affect such feature, structure, or characteristic in connection with other embodiments whether explicitly described.

In the various embodiments described above, unless specifically noted otherwise, disjunctive language such as the phrase “at least one of A, B, or C” is intended to be understood to mean either A, B, or C, or any combination thereof (e.g., A, B, and/or C). As such, disjunctive language is not intended to, nor should it be understood to, imply that a given embodiment requires at least one of A, at least one of B, or at least one of C to each be present.

One embodiment provides a method comprising, at a runtime library executed by a processor of a data processing system, receiving an input frame having objects to be stylized via a style transfer network associated with the runtime library, wherein the style transfer network is a neural network model trained to apply one or more visual styles to an input frame, performing instance segmentation on the input frame to generate one or more instance masks to identify one or more objects to be stylized, generating one or more stylized frames for each style to transfer to the input frame, and merging, via the one or more instance masks, stylized objects from one or more stylized frames with un-stylized content from the input frame to generate an output frame with per-instance stylization.

In a further embodiment, one or more identifiers of objects to be stylized are received in conjunction with the input frame. The input frame can be a video frame of video content, which in particular can be streaming media content, and/or can be a frame of a three-dimensional graphics application rendered by a graphics processor. For example, the frame can be a three-dimensional graphics application frame that is embedded in a media frame of a gaming live streamer. The data processing system can also be a cloud gaming server that is configured to stylize cloud gaming output via the runtime library.

Instance segmentation can be performed on the input frame based on object data provided by a render associated with the three-dimensional graphics application. Instance segmentation can be performed on the input frame based on object data provided by a shim layer for an application programming interface (API) used to render the three-dimensional graphics application. The method can additionally include performing instance segmentation on the input frame to generate multiple instance masks to identify multiple objects to be stylized, generating multiple stylized frames, where each of the multiple stylized frame is stylized to a separate style, and merging, via the multiple instance masks, multiple stylized objects from the multiple stylized frames with un-stylized content from the input frame to generate an output frame with per-instance stylization of multiple objects.

The operations performed by the above method can be performed by one or more processors of the data processing system based on instructions provided in a non-transitory machine readable medium. A data processing system configured to execute such instructions can also perform operations of the above-described method.

The specification and drawings are, accordingly, to be regarded in an illustrative rather than a restrictive sense. Those skilled in the art will appreciate that the broad techniques of the embodiments described herein can be implemented in a variety of forms. Therefore, while the embodiments have been described in connection with examples thereof, the true scope of the embodiments should not be so limited since other modifications will become apparent to the skilled practitioner upon a study of the drawings, specification, and following claims.