中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

16 records (took 0.016 seconds)

    Graphics Processor with Non-Blocking Concurrent Architecture
    2.
    发明申请
    Graphics Processor with Non-Blocking Concurrent Architecture 有权

    公开(公告)号：US20210065425A1

    公开(公告)日：2021-03-04

    申请号：US17098089

    申请日：2020-11-13

    Applicant: Imagination Technologies Limited

    Inventor： Luke T. Peterson ,  James A. McCombe ,  Steven J. Clohset ,  Jason R. Redgrave

    IPC: G06T15/00 ,  G06F9/50 ,  G06F15/80 ,  G06F9/52 ,  G06T1/20 ,  G06T1/60 ,  G06T15/06

    Abstract: In some aspects, systems and methods provide for forming groupings of a plurality of independently-specified computation workloads, such as graphics processing workloads, and in a specific example, ray tracing workloads. The workloads include a scheduling key, which is one basis on which the groupings can be formed. Workloads grouped together can all execute from the same source of instructions, on one or more different private data elements. Such workloads can recursively instantiate other workloads that reference the same private data elements. In some examples, the scheduling key can be used to identify a data element to be used by all the workloads of a grouping. Memory conflicts to private data elements are handled through scheduling of non-conflicted workloads or specific instructions and/or deferring conflicted workloads instead of locking memory locations.

    Atomic Memory Update Unit & Methods
    3.
    发明申请
    Atomic Memory Update Unit & Methods 有权
    原子内存更新单位和方法

    公开(公告)号：US20150089156A1

    公开(公告)日：2015-03-26

    申请号：US14494496

    申请日：2014-09-23

    Applicant: Imagination Technologies Limited

    Inventor： Steven J. Clohset ,  Jason R. Redgrave ,  Luke T. Peterson

    IPC: G06F12/08

    CPC classification number: G06T15/005 ,  G06F12/0804 ,  G06F12/0862 ,  G06F12/126 ,  G06F2212/1024 ,  G06F2212/302 ,  G06F2212/452 ,  G06F2212/455 ,  G06T1/60 ,  G06T15/06 ,  G06T17/10

    Abstract: In an aspect, an update unit can evaluate condition(s) in an update request and update one or more memory locations based on the condition evaluation. The update unit can operate atomically to determine whether to effect the update and to make the update. Updates can include one or more of incrementing and swapping values. An update request may specify one of a pre-determined set of update types. Some update types may be conditional and others unconditional. The update unit can be coupled to receive update requests from a plurality of computation units. The computation units may not have privileges to directly generate write requests to be effected on at least some of the locations in memory. The computation units can be fixed function circuitry operating on inputs received from programmable computation elements. The update unit may include a buffer to hold received update requests.

    Abstract translation: 在一方面，更新单元可以基于条件评估来评估更新请求中的条件并更新一个或多个存储器位置。 更新单元可以原子地操作以确定是否实现更新并进行更新。 更新可以包括递增和交换值中的一个或多个。 更新请求可以指定预定的一组更新类型之一。 某些更新类型可能是有条件的，而其他更新类型可能是无条件的。 更新单元可以被耦合以从多个计算单元接收更新请求。 计算单元可以不具有直接生成要在存储器中的至少一些位置上实现的写请求的权限。 计算单元可以是对从可编程计算元件接收的输入进行操作的固定功能电路。 更新单元可以包括用于保存接收到的更新请求的缓冲器。

    Atomic memory update unit and methods
    6.
    发明授权
    Atomic memory update unit and methods 有权

    公开(公告)号：US11257271B2

    公开(公告)日：2022-02-22

    申请号：US15275645

    申请日：2016-09-26

    Applicant: Imagination Technologies Limited

    Inventor： Steven J. Clohset ,  Jason R. Redgrave ,  Luke T. Peterson

    IPC: G06T15/00 ,  G06F12/0804 ,  G06T15/06 ,  G06T17/10 ,  G06T1/60 ,  G06F12/0862 ,  G06F12/16 ,  G06F12/126

    Abstract: In an aspect, an update unit can evaluate condition(s) in an update request and update one or more memory locations based on the condition evaluation. The update unit can operate atomically to determine whether to effect the update and to make the update. Updates can include one or more of incrementing and swapping values. An update request may specify one of a pre-determined set of update types. Some update types may be conditional and others unconditional. The update unit can be coupled to receive update requests from a plurality of computation units. The computation units may not have privileges to directly generate write requests to be effected on at least some of the locations in memory. The computation units can be fixed function circuitry operating on inputs received from programmable computation elements. The update unit may include a buffer to hold received update requests.

    Graphics Processor with Non-Blocking Concurrent Architecture
    7.
    发明申请
    Graphics Processor with Non-Blocking Concurrent Architecture 有权

    公开(公告)号：US20250148688A1

    公开(公告)日：2025-05-08

    申请号：US19017672

    申请日：2025-01-12

    Applicant: Imagination Technologies Limited

    Inventor： Luke T. Peterson ,  James A. McCombe ,  Steven J. Clohset ,  Jason R. Redgrave

    IPC: G06T15/00 ,  G06F9/50 ,  G06F9/52 ,  G06F15/80 ,  G06T1/20 ,  G06T1/60 ,  G06T15/06

    Abstract: In some aspects, systems and methods provide for forming groupings of a plurality of independently-specified computation workloads, such as graphics processing workloads, and in a specific example, ray tracing workloads. The workloads include a scheduling key, which is one basis on which the groupings can be formed. Workloads grouped together can all execute from the same source of instructions, on one or more different private data elements. Such workloads can recursively instantiate other workloads that reference the same private data elements. In some examples, the scheduling key can be used to identify a data element to be used by all the workloads of a grouping. Memory conflicts to private data elements are handled through scheduling of non-conflicted workloads or specific instructions and/or deferring conflicted workloads instead of locking memory locations.

    Graphics Processor with Non-Blocking Concurrent Architecture
    8.
    发明申请
    Graphics Processor with Non-Blocking Concurrent Architecture 审中-公开
    具有非阻塞并发架构的图形处理器

    公开(公告)号：US20150339798A1

    公开(公告)日：2015-11-26

    申请号：US14817747

    申请日：2015-08-04

    Applicant: Imagination Technologies Limited

    Inventor： Luke T. Peterson ,  James A. McCombe ,  Steven J. Clohset ,  Jason R. Redgrave

    IPC: G06T1/20 ,  G06F15/80 ,  G06F9/50

    CPC classification number: G06T15/005 ,  G06F9/5033 ,  G06F9/505 ,  G06F9/52 ,  G06F15/8007 ,  G06T1/20 ,  G06T1/60 ,  G06T15/06 ,  G06T2200/28

    Abstract: In some aspects, systems and methods provide for forming groupings of a plurality of independently-specified computation workloads, such as graphics processing workloads, and in a specific example, ray tracing workloads. The workloads include a scheduling key, which is one basis on which the groupings can be formed. Workloads grouped together can all execute from the same source of instructions, on one or more different private data elements. Such workloads can recursively instantiate other workloads that reference the same private data elements. In some examples, the scheduling key can be used to identify a data element to be used by all the workloads of a grouping. Memory conflicts to private data elements are handled through scheduling of non-conflicted workloads or specific instructions and/or deferring conflicted workloads instead of locking memory locations.

    Abstract translation: 在一些方面，系统和方法提供用于形成多个独立指定的计算工作负荷（诸如图形处理工作负载）以及在具体示例中的光线跟踪工作负载的分组。 工作负载包括一个调度密钥，这是可以形成分组的一个基础。 分组在一起的工作负载都可以在一个或多个不同的私有数据元素上从相同的指令来源执行。 这样的工作负载可以递归地实例化引用相同私有数据元素的其他工作负载。 在一些示例中，调度密钥可用于标识要由分组的所有工作负载使用的数据元素。 与私有数据元素的内存冲突通过调度非冲突的工作负载或特定指令来处理和/或延迟冲突的工作负载，而不是锁定内存位置。

    ATOMIC MEMORY UPDATE UNIT AND METHODS
    9.
    发明申请
    ATOMIC MEMORY UPDATE UNIT AND METHODS 有权

    公开(公告)号：US20220230378A1

    公开(公告)日：2022-07-21

    申请号：US17571104

    申请日：2022-01-07

    Applicant: Imagination Technologies Limited

    Inventor： Steven J. Clohset ,  Jason R. Redgrave ,  Luke T. Peterson

    IPC: G06T15/00 ,  G06F12/0804 ,  G06T1/60 ,  G06T15/06 ,  G06T17/10

    Abstract: In an aspect, an update unit can evaluate condition(s) in an update request and update one or more memory locations based on the condition evaluation. The update unit can operate atomically to determine whether to effect the update and to make the update. Updates can include one or more of incrementing and swapping values. An update request may specify one of a pre-determined set of update types. Some update types may be conditional and others unconditional. The update unit can be coupled to receive update requests from a plurality of computation units. The computation units may not have privileges to directly generate write requests to be effected on at least some of the locations in memory. The computation units can be fixed function circuitry operating on inputs received from programmable computation elements. The update unit may include a buffer to hold received update requests.

    Graphics processor with non-blocking concurrent architecture
    10.
    发明授权
    Graphics processor with non-blocking concurrent architecture 有权

    公开(公告)号：US10861214B2

    公开(公告)日：2020-12-08

    申请号：US15219860

    申请日：2016-07-26

    Applicant: Imagination Technologies Limited

    Inventor： Luke T. Peterson ,  James A. McCombe ,  Steven J. Clohset ,  Jason R. Redgrave

    IPC: G06T15/00 ,  G06F9/52 ,  G06F9/50 ,  G06F15/80 ,  G06T1/20 ,  G06T1/60 ,  G06T15/06

    Abstract: In some aspects, systems and methods provide for forming groupings of a plurality of independently-specified computation workloads, such as graphics processing workloads, and in a specific example, ray tracing workloads. The workloads include a scheduling key, which is one basis on which the groupings can be formed. Workloads grouped together can all execute from the same source of instructions, on one or more different private data elements. Such workloads can recursively instantiate other workloads that reference the same private data elements. In some examples, the scheduling key can be used to identify a data element to be used by all the workloads of a grouping. Memory conflicts to private data elements are handled through scheduling of non-conflicted workloads or specific instructions and/or deferring conflicted workloads instead of locking memory locations.

Patent Agency Ranking