发明名称 Dispatching of instructions for execution by heterogeneous processing engines
摘要 An embodiment of a computing system is configured to process data using a multithreaded SIMD architecture that includes heterogeneous processing engines to execute a program. The program is constructed of various program instructions. A first type of the program instructions can only be executed by a first type of processing engine and a second type of program instructions can only be executed by a second type of processing engine. A third type of program instructions can be executed by the first and the second type of processing engines. An instruction dispatcher is configured to identify and remove program instruction execution conflicts for the heterogeneous processing engines to improve instruction execution throughput.
申请公布号 US9304775(B1) 申请公布日期 2016.04.05
申请号 US200711935266 申请日期 2007.11.05
申请人 NVIDIA Corporation 发明人 Lindholm John Erik;Wierzbicki Jered
分类号 G06F9/38 主分类号 G06F9/38
代理机构 Artegis Law Group, LLP 代理人 Artegis Law Group, LLP
主权项 1. A method for dispatching program instructions for execution in a SIMD (single-instruction multiple-data) architecture with heterogeneous processing engines of different types, the method comprising: loading first dispatch slots with first program instructions that are highest priority program instructions, wherein each of the first program instructions is executable by only a specific type of the heterogeneous processing engines; loading second dispatch slots with second program instructions that can be executed by any type of the heterogeneous processing engines, wherein each of the second dispatch slots corresponds to a different first dispatch slot; and for each of the first program instructions that are loaded into first dispatch slots: determining that an execution conflict exists between the first instruction and other first program instructions, andfor each of the other first program instructions: loading the first program instruction into the corresponding second dispatch slot and loading the second program instruction in the corresponding second slot into the corresponding first dispatch slot.
地址 Santa Clara CA US