PARALLEL PROCESSING CONTROL
    1.
    发明申请

    公开(公告)号:US20250110747A1

    公开(公告)日:2025-04-03

    申请号:US18478345

    申请日:2023-09-29

    Applicant: Arm Limited

    Abstract: A method of preparing a command stream for a parallel processor, comprising: analysing the command stream to detect at least a first dependency; generating at least one timeline dependency point responsive to detecting the first dependency; determining a latest action for the first dependency to derive a completion stream timeline point for the first dependency; comparing the completion stream timeline point for the first dependency with a completion stream timeline point for a second dependency to determine a latest stream timeline point; generating at least one command stream synchronization control instruction according to the latest stream timeline point; and providing the command stream and the at least one command stream synchronization control instruction to an execution unit of the parallel processor.

    GRAPHICS PROCESSING
    2.
    发明申请
    GRAPHICS PROCESSING 有权
    图形处理

    公开(公告)号:US20140354669A1

    公开(公告)日:2014-12-04

    申请号:US13906132

    申请日:2013-05-30

    Applicant: ARM Limited

    CPC classification number: G06T15/005 G06T11/001 G09G5/393

    Abstract: In a graphics processing system that includes one or more programmable shading stages, the shader compiler receives a shader program to be compiled (step 40), identifies any run time constant expressions in the shader program (step 41), and removes such expressions from the original shader program, and substitutes them in the original main shader program with appropriate load instructions pointing to where the output results from a pilot shader program will be stored (step 42). The shader compiler then creates a separate pilot shader program for the identified run-time constant expressions (step 43). The pilot shader program is then executed on the graphics processing pipeline (step 44), before the modified main shader program (step 45).

    Abstract translation: 在包括一个或多个可编程着色阶段的图形处理系统中,着色器编译器接收要编译的着色器程序(步骤40),识别着色器程序中的任何运行时常数表达式(步骤41),并从 原始着色器程序,并将它们替换为原始主着色器程序,并使用适当的加载指令指向将存储来自飞行员着色器程序的输出结果的位置(步骤42)。 着色器编译器然后为识别的运行时常数表达式创建单独的导频着色器程序(步骤43)。 然后,在修改的主着色器程序之前,在图形处理流水线(步骤44)上执行导频着色器程序(步骤45)。

    GRAPHICS PROCESSING SYSTEMS
    3.
    发明公开

    公开(公告)号:US20240168804A1

    公开(公告)日:2024-05-23

    申请号:US18491104

    申请日:2023-10-20

    Applicant: Arm Limited

    CPC classification number: G06F9/5005 G06F9/54 G06T1/60

    Abstract: The present disclosure relates to a processing resource for a graphics processing system for performing graphics processing for an application executing on a host processor of the graphics processing system according to a command stream, the command stream being generated by the host processor in response to an API call from the application, the processing resource comprising: a control circuit configured to execute commands from the command stream, wherein the command stream comprises one or more commands relating to a processing task and one or more commands relating to at least one state group associated with the processing task; at least one processing circuit configured to perform processing tasks; a shadow state storage module configured for use by the control circuit to store state information; and a processing state storage module configured for use by the processing circuit to store state information, wherein the control circuit is configured to determine one or more changed states within the at least one state group with respect to a preceding API call, to write state information comprising the one or more changed states to the shadow state storage module, and to assign the processing task to the at least one processing circuit and execute an instruction to transmit the state information from the shadow state storage module to the processing state storage module.

    DATA PROCESSING SYSTEMS
    4.
    发明申请

    公开(公告)号:US20220382587A1

    公开(公告)日:2022-12-01

    申请号:US17705856

    申请日:2022-03-28

    Applicant: Arm Limited

    Abstract: A data processing system is disclosed that includes one or more processors that can perform producer processes to produce work and consumer processes that can consume work produced by a producer process. The system includes a pool of plural communication resources that may be used for communications between a producer process and a consumer process. The system tracks the usage of communication resources of the pool of communication resources, and allocates a communication resource from the pool of communication resources based on the tracking.

    Tree-based graphics primitive rendering

    公开(公告)号:US10176627B2

    公开(公告)日:2019-01-08

    申请号:US14639867

    申请日:2015-03-05

    Applicant: ARM Limited

    Abstract: A set of primitives is divided into plural sub-sets of primitives, and a tree representation for representing the set of primitives in which each leaf node of the tree represents one of the sub-sets, and each parent node of the tree represents a sub-set of primitives corresponding to the combination of the sub-sets of primitives of all of its child nodes, is generated. For each node of the tree representation data indicating: the sub-set of primitives that the node represents; the vertices that are used by the primitives in the sub-set of primitives that the node represents; and the volume of space that the sub-set of primitives that the node represents falls within, is determined. The tree representation is then used to determine a set of primitives and a set of vertices to be processed when processing the set of primitives for the output frame.

    Graphics processing
    6.
    发明授权
    Graphics processing 有权
    图形处理

    公开(公告)号:US09189881B2

    公开(公告)日:2015-11-17

    申请号:US13906132

    申请日:2013-05-30

    Applicant: ARM Limited

    CPC classification number: G06T15/005 G06T11/001 G09G5/393

    Abstract: In a graphics processing system that includes one or more programmable shading stages, the shader compiler receives a shader program to be compiled (step 40), identifies any run time constant expressions in the shader program (step 41), and removes such expressions from the original shader program, and substitutes them in the original main shader program with appropriate load instructions pointing to where the output results from a pilot shader program will be stored (step 42). The shader compiler then creates a separate pilot shader program for the identified run-time constant expressions (step 43). The pilot shader program is then executed on the graphics processing pipeline (step 44), before the modified main shader program (step 45).

    Abstract translation: 在包括一个或多个可编程着色阶段的图形处理系统中,着色器编译器接收要编译的着色器程序(步骤40),识别着色器程序中的任何运行时常数表达式(步骤41),并从 原始着色器程序,并将它们替换为原始主着色器程序,并使用适当的加载指令指向将存储来自飞行员着色器程序的输出结果的位置(步骤42)。 着色器编译器然后为识别的运行时常数表达式创建单独的导频着色器程序(步骤43)。 然后,在修改的主着色器程序之前,在图形处理流水线(步骤44)上执行导频着色器程序(步骤45)。

    GRAPHICS PROCESSING SYSTEMS
    7.
    发明申请
    GRAPHICS PROCESSING SYSTEMS 审中-公开
    图形处理系统

    公开(公告)号:US20150269773A1

    公开(公告)日:2015-09-24

    申请号:US14639867

    申请日:2015-03-05

    Applicant: ARM Limited

    CPC classification number: G06T17/005

    Abstract: A set of primitives is divided into plural sub-sets of primitives, and a tree representation for representing the set of primitives in which each leaf node of the tree represents one of the sub-sets, and each parent node of the tree represents a sub-set of primitives corresponding to the combination of the sub-sets of primitives of all of its child nodes, is generated. For each node of the tree representation data indicating: the sub-set of primitives that the node represents; the vertices that are used by the primitives in the sub-set of primitives that the node represents; and the volume of space that the sub-set of primitives that the node represents falls within, is determined. The tree representation is then used to determine a set of primitives and a set of vertices to be processed when processing the set of primitives for the output frame.

    Abstract translation: 一组原语被分成多个原语子集,以及用于表示一组原语的树表示,其中树的每个叶节点表示子集之一,并且树的每个父节点表示子 生成与其所有子节点的原语的子集的组合相对应的原语集合。 对于树的每个节点表示数据,指示:节点表示的基元的子集; 由节点代表的图元子集中的原语使用的顶点; 并且确定节点所代表的基元子集落在其中的空间量。 然后,树表示用于确定在处理输出帧的图元集合时将要处理的一组基元和一组顶点。

Patent Agency Ranking