专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US09442759B2 Concurrent execution of independent streams in multi-channel time slice groups 有权
标题翻译：在多通道时间片组中并发执行独立流
公开(公告)号：US09442759B2
公开(公告)日：2016-09-13
申请号：US13316334
申请日：2011-12-09
申请人： Samuel H. Duncan , Lacky V. Shah , Sean J. Treichler , Daniel Elliot Wexler , Jerome F. Duluk, Jr. , Philip Browning Johnson , Jonathon Stuart Ramsay Evans
发明人： Samuel H. Duncan , Lacky V. Shah , Sean J. Treichler , Daniel Elliot Wexler , Jerome F. Duluk, Jr. , Philip Browning Johnson , Jonathon Stuart Ramsay Evans
IPC分类号： G06F9/46 , G06F13/00 , G06F9/48 , G06F9/38
CPC分类号： G06F9/4881 , G06F9/3851 , G06F9/461
摘要： A time slice group (TSG) is a grouping of different streams of work (referred to herein as “channels”) that share the same context information. The set of channels belonging to a TSG are processed in a pre-determined order. However, when a channel stalls while processing, the next channel with independent work can be switched to fully load the parallel processing unit. Importantly, because each channel in the TSG shares the same context information, a context switch operation is not needed when the processing of a particular channel in the TSG stops and the processing of a next channel in the TSG begins. Therefore, multiple independent streams of work are allowed to run concurrently within a single context increasing utilization of parallel processing units.
摘要翻译：时间片组（TSG）是共享相同上下文信息的不同工作流（本文称为“信道”）的分组。属于TSG的信道集合以预定的顺序被处理。然而，当通道在处理过程中停顿时，可以切换具有独立工作的下一个通道来完全加载并行处理单元。重要的是，由于TSG中的每个信道共享相同的上下文信息，当TSG中的特定信道的处理停止并且TSG中的下一个信道的处理开始时，不需要上下文切换操作。因此，允许多个独立的工作流在单个上下文中同时运行，增加并行处理单元的利用率。

2. 发明申请

US20130298133A1 TECHNIQUE FOR COMPUTATIONAL NESTED PARALLELISM 有权
标题翻译：计算并行平行技术
公开(公告)号：US20130298133A1
公开(公告)日：2013-11-07
申请号：US13462649
申请日：2012-05-02
申请人： Stephen JONES , Philip Alexander Cuadra , Daniel Elliot Wexler , Ignacio Llamas , Lacky V. Shah , Jerome F. Duluk, JR. , Christopher Lamb
发明人： Stephen JONES , Philip Alexander Cuadra , Daniel Elliot Wexler , Ignacio Llamas , Lacky V. Shah , Jerome F. Duluk, JR. , Christopher Lamb
IPC分类号： G06F9/50
CPC分类号： G06F9/5027 , G06F9/522 , G06F2209/483 , G06T1/20
摘要： One embodiment of the present invention sets forth a technique for performing nested kernel execution within a parallel processing subsystem. The technique involves enabling a parent thread to launch a nested child grid on the parallel processing subsystem, and enabling the parent thread to perform a thread synchronization barrier on the child grid for proper execution semantics between the parent thread and the child grid. This technique advantageously enables the parallel processing subsystem to perform a richer set of programming constructs, such as conditionally executed and nested operations and externally defined library functions without the additional complexity of CPU involvement.
摘要翻译：本发明的一个实施例提出了一种用于在并行处理子系统内执行嵌套的内核执行的技术。该技术涉及使父线程启动并行处理子系统上的嵌套子网格，并使父线程能够在子网格上执行线程同步屏障，以在父线程和子网格之间实现正确的执行语义。该技术有利地使得并行处理子系统能够执行更丰富的编程结构集合，诸如条件执行和嵌套操作以及外部定义的库函数，而不会增加CPU参与的复杂性。

3. 发明授权

US09513975B2 Technique for computational nested parallelism 有权
标题翻译：计算嵌套并行性技术
公开(公告)号：US09513975B2
公开(公告)日：2016-12-06
申请号：US13462649
申请日：2012-05-02
申请人： Stephen Jones , Philip Alexander Cuadra , Daniel Elliot Wexler , Ignacio Llamas , Lacky V. Shah , Jerome F. Duluk, Jr. , Christopher Lamb
发明人： Stephen Jones , Philip Alexander Cuadra , Daniel Elliot Wexler , Ignacio Llamas , Lacky V. Shah , Jerome F. Duluk, Jr. , Christopher Lamb
IPC分类号： G06F9/46 , G06F9/52 , G06T1/20
CPC分类号： G06F9/5027 , G06F9/522 , G06F2209/483 , G06T1/20
摘要： One embodiment of the present invention sets forth a technique for performing nested kernel execution within a parallel processing subsystem. The technique involves enabling a parent thread to launch a nested child grid on the parallel processing subsystem, and enabling the parent thread to perform a thread synchronization barrier on the child grid for proper execution semantics between the parent thread and the child grid. This technique advantageously enables the parallel processing subsystem to perform a richer set of programming constructs, such as conditionally executed and nested operations and externally defined library functions without the additional complexity of CPU involvement.
摘要翻译：本发明的一个实施例提出了一种用于在并行处理子系统内执行嵌套的内核执行的技术。该技术涉及使父线程启动并行处理子系统上的嵌套子网格，并使父线程能够在子网格上执行线程同步屏障，以在父线程和子网格之间实现正确的执行语义。该技术有利地使得并行处理子系统能够执行更丰富的编程结构集合，诸如条件执行和嵌套操作以及外部定义的库函数，而不会增加CPU参与的复杂性。

4. 发明授权

US09147224B2 Method for handling state transitions in a network of virtual processing nodes 有权
标题翻译：处理虚拟处理节点网络状态转换的方法
公开(公告)号：US09147224B2
公开(公告)日：2015-09-29
申请号：US13294983
申请日：2011-11-11
申请人： Sean J. Treichler , Lacky V. Shah , Daniel Elliot Wexler
发明人： Sean J. Treichler , Lacky V. Shah , Daniel Elliot Wexler
IPC分类号： G06T1/20
CPC分类号： G06T1/20 , G06F9/461 , G06T15/005
摘要： One embodiment of the present invention sets forth a technique for receiving versions of state objects at one or more stages in a processing pipeline. The method includes receiving a first version of a state object at a first stage in the processing pipeline, determining that the first version of the state object is relevant to the first stage, incrementing a first reference counter associated with the first version of the state object, assigning the first version of the state object to work requests that arrive at the first stage subsequent to the receipt of the first version of the state object, and transmitting the first version of the state object to a second stage in the processing pipeline.
摘要翻译：本发明的一个实施例提出了一种用于在处理流水线中的一个或多个阶段接收状态对象的版本的技术。该方法包括在处理流水线中的第一阶段接收状态对象的第一版本，确定状态对象的第一版本与第一阶段相关，增加与状态对象的第一版本相关联的第一引用计数器将状态对象的第一版本分配给在接收状态对象的第一版本之后到达第一阶段的工作请求，并且将状态对象的第一版本发送到处理流水线中的第二阶段。

5. 发明申请

US20130120413A1 METHOD FOR HANDLING STATE TRANSITIONS IN A NETWORK OF VIRTUAL PROCESSING NODES 有权
标题翻译：用于处理虚拟处理节点网络中的状态转换的方法
公开(公告)号：US20130120413A1
公开(公告)日：2013-05-16
申请号：US13294983
申请日：2011-11-11
申请人： Sean J. TREICHLER , Lacky V. Shah , Daniel Elliot Wexler
发明人： Sean J. TREICHLER , Lacky V. Shah , Daniel Elliot Wexler
IPC分类号： G06T1/20
CPC分类号： G06T1/20 , G06F9/461 , G06T15/005
摘要： One embodiment of the present invention sets forth a technique for receiving versions of state objects at one or more stages in a processing pipeline. The method includes receiving a first version of a state object at a first stage in the processing pipeline, determining that the first version of the state object is relevant to the first stage, incrementing a first reference counter associated with the first version of the state object, assigning the first version of the state object to work requests that arrive at the first stage subsequent to the receipt of the first version of the state object, and transmitting the first version of the state object to a second stage in the processing pipeline.
摘要翻译：本发明的一个实施例提出了一种用于在处理流水线中的一个或多个阶段接收状态对象的版本的技术。该方法包括在处理流水线中的第一阶段接收状态对象的第一版本，确定状态对象的第一版本与第一阶段相关，增加与状态对象的第一版本相关联的第一引用计数器将状态对象的第一版本分配给在接收状态对象的第一版本之后到达第一阶段的工作请求，并且将状态对象的第一版本发送到处理流水线中的第二阶段。

6. 发明申请

US20130120412A1 METHOD FOR HANDLING STATE TRANSITIONS IN A NETWORK OF VIRTUAL PROCESSING NODES 有权
标题翻译：用于处理虚拟处理节点网络中的状态转换的方法
公开(公告)号：US20130120412A1
公开(公告)日：2013-05-16
申请号：US13294978
申请日：2011-11-11
申请人： Sean J. TREICHLER , Lacky V. Shah , Daniel Elliot Wexler
发明人： Sean J. TREICHLER , Lacky V. Shah , Daniel Elliot Wexler
IPC分类号： G06T1/20
CPC分类号： G06T1/20
摘要： One embodiment of the present invention sets forth a technique for executing an operation once work associated with a version of a state object has been completed. The method includes receiving the version of the state object at a first stage in a processing pipeline, where the version of the state object is associated with a reference count object, determining that the version of the state object is relevant to the first stage, incrementing a counter included in the reference count object, transmitting the version of the state object to a second stage in the processing pipeline, processing work associated with the version of the state object, decrementing the counter, determining that the counter is equal to zero, and in response, executing an operation specified by the reference count object.
摘要翻译：本发明的一个实施例提出了一旦与状态对象的版本相关联的一次工作已经完成的执行操作的技术。该方法包括在处理流水线中的第一阶段接收状态对象的版本，其中状态对象的版本与引用计数对象相关联，确定状态对象的版本与第一阶段相关，递增包括在引用计数对象中的计数器，将状态对象的版本发送到处理流水线中的第二阶段，处理与状态对象的版本相关联的工作，递减计数器，确定计数器等于零，以及作为响应，执行由引用计数对象指定的操作。

7. 发明授权

US08976185B2 Method for handling state transitions in a network of virtual processing nodes 有权
标题翻译：处理虚拟处理节点网络状态转换的方法
公开(公告)号：US08976185B2
公开(公告)日：2015-03-10
申请号：US13294978
申请日：2011-11-11
申请人： Sean J. Treichler , Lacky V. Shah , Daniel Elliot Wexler
发明人： Sean J. Treichler , Lacky V. Shah , Daniel Elliot Wexler
IPC分类号： G06T1/00 , G06T1/20
CPC分类号： G06T1/20
摘要： One embodiment of the present invention sets forth a technique for executing an operation once work associated with a version of a state object has been completed. The method includes receiving the version of the state object at a first stage in a processing pipeline, where the version of the state object is associated with a reference count object, determining that the version of the state object is relevant to the first stage, incrementing a counter included in the reference count object, transmitting the version of the state object to a second stage in the processing pipeline, processing work associated with the version of the state object, decrementing the counter, determining that the counter is equal to zero, and in response, executing an operation specified by the reference count object.
摘要翻译：本发明的一个实施例提出了一旦与状态对象的版本相关联的一次工作已经完成的执行操作的技术。该方法包括在处理流水线中的第一阶段接收状态对象的版本，其中状态对象的版本与引用计数对象相关联，确定状态对象的版本与第一阶段相关，递增包括在引用计数对象中的计数器，将状态对象的版本发送到处理流水线中的第二阶段，处理与状态对象的版本相关联的工作，递减计数器，确定计数器等于零，以及作为响应，执行由引用计数对象指定的操作。

8. 发明授权

US08928677B2 Low latency concurrent computation 有权
标题翻译：低延迟并发计算
公开(公告)号：US08928677B2
公开(公告)日：2015-01-06
申请号：US13357569
申请日：2012-01-24
申请人： Daniel Elliot Wexler , Jeffrey A. Bolz , Jesse David Hall , Philip Alexander Cuadra , Naveen Leekha , Ignacio Llamas
发明人： Daniel Elliot Wexler , Jeffrey A. Bolz , Jesse David Hall , Philip Alexander Cuadra , Naveen Leekha , Ignacio Llamas
IPC分类号： G06F15/80 , G06T1/20
CPC分类号： G06F9/5027 , G06F2209/509
摘要： One embodiment of the present invention sets forth a technique for performing low latency computation on a parallel processing subsystem. A low latency functional node is exposed to an operating system. The low latency functional node and a generic functional node are configured to target the same underlying processor resource within the parallel processing subsystem. The operating system stores low latency tasks generated by a user application within a low latency command buffer associated with the low latency functional node. The parallel processing subsystem advantageously executes tasks from the low latency command buffer prior to completing execution of tasks in the generic command buffer, thereby reducing completion latency for the low latency tasks.
摘要翻译：本发明的一个实施例提出了一种用于在并行处理子系统上执行低等待时间计算的技术。低延迟功能节点暴露于操作系统。低延迟功能节点和通用功能节点被配置为针对并行处理子系统内的相同底层处理器资源。操作系统存储由与低延迟功能节点相关联的低延迟命令缓冲器内由用户应用产生的低延迟任务。并行处理子系统在完成在通用命令缓冲器中执行任务之前有利地执行来自低等待时间命令缓冲器的任务，从而减少低延迟任务的完成等待时间。

9. 发明申请

US20110247018A1 API For Launching Work On a Processor 有权
标题翻译：用于在处理器上启动工作的API
公开(公告)号：US20110247018A1
公开(公告)日：2011-10-06
申请号：US13077806
申请日：2011-03-31
申请人： Timothy Paul Lottes FARRAR , Ignacio Llamas , Daniel Elliot Wexler , Craig Ross Duttweiler
发明人： Timothy Paul Lottes FARRAR , Ignacio Llamas , Daniel Elliot Wexler , Craig Ross Duttweiler
IPC分类号： G06F9/46
CPC分类号： G06F9/4843
摘要： One embodiment of the present invention sets forth a technique for launching work on a processor. The method includes the steps of initializing a first state object within a memory region accessible to a program executing on the processor, populating the first state object with data associated with a first workload that is generated by the program, and triggering the processing of the first workload on the processor according to the data within the first state object.
摘要翻译：本发明的一个实施例提出了一种用于在处理器上启动工作的技术。该方法包括以下步骤：初始化在处理器上执行的程序可访问的存储器区域内的第一状态对象，使用与由程序生成的第一工作负荷相关联的数据填充第一状态对象，并且触发第一状态对象的处理处理器上的工作量根据第一个状态对象内的数据。

10. 发明申请

US20130187935A1 LOW LATENCY CONCURRENT COMPUTATION 有权
标题翻译：低期并发计算
公开(公告)号：US20130187935A1
公开(公告)日：2013-07-25
申请号：US13357569
申请日：2012-01-24
申请人： Daniel Elliot Wexler , Jeffrey A. Bolz , Jesse David Hall , Philip Alexander Cuadra , Naveen Leekha , Ignacio Llamas
发明人： Daniel Elliot Wexler , Jeffrey A. Bolz , Jesse David Hall , Philip Alexander Cuadra , Naveen Leekha , Ignacio Llamas
IPC分类号： G06F15/80
CPC分类号： G06F9/5027 , G06F2209/509
摘要： One embodiment of the present invention sets forth a technique for performing low latency computation on a parallel processing subsystem. A low latency functional node is exposed to an operating system. The low latency functional node and a generic functional node are configured to target the same underlying processor resource within the parallel processing subsystem. The operating system stores low latency tasks generated by a user application within a low latency command buffer associated with the low latency functional node. The parallel processing subsystem advantageously executes tasks from the low latency command buffer prior to completing execution of tasks in the generic command buffer, thereby reducing completion latency for the low latency tasks.
摘要翻译：本发明的一个实施例提出了一种用于在并行处理子系统上执行低等待时间计算的技术。低延迟功能节点暴露于操作系统。低延迟功能节点和通用功能节点被配置为针对并行处理子系统内的相同底层处理器资源。操作系统存储由与低延迟功能节点相关联的低延迟命令缓冲器内由用户应用产生的低延迟任务。并行处理子系统有利地在完成在通用命令缓冲器中执行任务之前从低延迟命令缓冲器执行任务，从而减少低延迟任务的完成等待时间。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式