-
公开(公告)号：KR1020140033616A
公开(公告)日：2014-03-19
申请号：KR1020120099609
申请日：2012-09-07
Applicant: 삼성전자주식회사
CPC classification number: G06F11/3624 , G06F8/41 , G06F11/3612 , G06F11/3672
Abstract: Disclosed are an apparatus and method for detecting a source code error location in a mixed-mode program, capable of providing source level debugging information without applying complex algorithm. An apparatus for detecting a source code error location in a mixed-mode program according to an embodiment comprises: a compiler unit which generates a first program by compiling a source code in a first mode, and generates a second program by compiling the source code in a second mode; a mapping table generation unit which generates a first mapping table for the first program, and a second mapping table for the second program; a simulation unit which simulates the first program and simulates the second program; a low-level data extraction unit which extracts first low-level data on the first program during the simulation of the first program, and extracts second low-level data on the second program during the simulation of the second program; a comparison data generation unit which maps the first low-level data onto the first mapping table to generate a verification target mapping table, and maps the second low-level data onto the second mapping table to generate a reference mapping table; and an error location detection unit which compares the verification target mapping table with the reference mapping table to determine whether there is an inconsistent record of data, and if so, obtains program location information relating to the record. [Reference numerals] (AA) Start; (BB) End; (S100) Compile step; (S110) Simulation and low level data extraction step; (S120) Comparison data generation step; (S130) Error location detection step
Abstract translation: 公开了一种用于检测混合模式程序中的源代码错误位置的装置和方法，能够提供源级调试信息而不应用复杂算法。 根据实施例的用于检测混合模式程序中的源代码错误位置的装置包括：编译单元，其通过以第一模式编译源代码来生成第一程序，并且通过将源代码编译成 第二种模式; 生成第一程序的第一映射表和第二程序的第二映射表的映射表生成单元， 模拟单元，其模拟第一程序并模拟第二程序; 低级数据提取单元，其在第一程序的模拟期间提取第一程序的第一低级数据，并且在第二程序的模拟期间提取关于第二程序的第二低级数据; 比较数据生成单元，将第一低级数据映射到第一映射表上，生成验证对象映射表，将第二低级数据映射到第二映射表，生成参照映射表; 以及错误位置检测单元，其将验证目标映射表与参考映射表进行比较，以确定是否存在不一致的数据记录，如果是，则获得与记录相关的节目位置信息。 （附图标记）（AA）开始; （BB）结束; （S100）编译步骤; （S110）模拟和低电平数据提取步骤; （S120）比较数据生成步骤; （S130）错误位置检测步骤
-
公开(公告)号：KR1020130051821A
公开(公告)日：2013-05-21
申请号：KR1020110117186
申请日：2011-11-10
Applicant: 삼성전자주식회사
CPC classification number: G06F9/5016 , G06F9/461
Abstract: PURPOSE: A method for transmitting vector register file data and a core system with improved interrupted speed delays are provided to prevent transmission delays generated in a process which stores and recovers the vector register file data based on the hardware data transmitting device when an interruption is generated, thereby improving the interrupted speed delays of the core system. CONSTITUTION: A second memory(120) has large capacity in comparison to a first memory(110). A vector register file is vector register sets. A core(140) determines whether or not a first memory stores the vector register file data of being executed when the interruption is generated. The core generates a first command or a second command which stores the vector register file data corresponding to a determination result. A data transmitting device(150) reads the vector register file data from the vector register file and stores the same in the second memory. [Reference numerals] (110) First memory; (120) Second memory; (130) Vector register file; (140) Core; (150) Data transmitting device; (160) Scalar register file; (170) Data memory controller
Abstract translation: 目的：提供一种用于传输向量寄存器文件数据的方法和具有改进的中断速度延迟的核心系统，以防止在产生中断时基于硬件数据发送设备存储和恢复向量寄存器文件数据的过程中产生的传输延迟 ，从而提高核心系统的中断速度延迟。 构成：与第一存储器（110）相比，第二存储器（120）具有大的容量。 向量寄存器文件是向量寄存器集。 核心（140）确定第一存储器是否存储当产生中断时正在执行的向量寄存器文件数据。 核心产生存储对应于确定结果的向量寄存器文件数据的第一命令或第二命令。 数据发送装置（150）从向量寄存器文件读取向量寄存器文件数据并将其存储在第二存储器中。 （附图标记）（110）第一存储器; （120）第二存储器; （130）向量寄存器文件; （140）核心; （150）数据发送装置; （160）标量寄存器文件; （170）数据存储控制器
-
公开(公告)号：KR1020120040630A
公开(公告)日：2012-04-27
申请号：KR1020100109998
申请日：2010-11-05
Applicant: 삼성전자주식회사
Abstract: PURPOSE: A reconfigurable processor to handle a memory dependent loop and a method thereof are provided to analyze the dependency between memory access commands, thereby allocating the commands to a plurality of processing elements based on the analyzed result. CONSTITUTION: An extracting unit(140) extracts an operation trace from a simulation result. A scheduler(150) analyzes the memory dependency between commands of iterations based on a corresponding trace to a memory access command out of the operation traces. A simulation unit(130) simulates the commands. The scheduler analyzes the memory dependency between the commands of iterations which exist in a generated iteration window.
Abstract translation: 目的：提供用于处理存储器相关环路的可重构处理器及其方法，以分析存储器访问命令之间的依赖关系，从而基于分析结果将命令分配给多个处理元件。 构成：提取单元（140）从仿真结果提取操作轨迹。 调度器（150）基于与操作轨迹之间的存储器访问命令的对应轨迹来分析迭代命令之间的存储器依赖性。 模拟单元（130）模拟命令。 调度器分析生成的迭代窗口中存在的迭代命令之间的内存依赖关系。
-
-
公开(公告)号：KR101854671B1
公开(公告)日：2018-05-08
申请号：KR1020110119146
申请日：2011-11-15
Applicant: 삼성전자주식회사
CPC classification number: G06F8/4452
Abstract: 소프트웨어파이프라인(Software Pipeline)이적용된루프(loop)의프롤로그(Prologue)와에필로그(Epilogue)에서루프외부의명령어를실행시키는기술에관한것으로루프외부의명령어들을프롤로그나에필로그에서겹쳐서실행되도록함으로써코드의성능을최대한으로향상시킬수 있다.
-
Patent Agency Ranking
公开(公告)号：KR1020150040663A
公开(公告)日：2015-04-15
申请号：KR1020130119450
申请日：2013-10-07
Applicant: 삼성전자주식회사
CPC classification number: G06F8/4452 , G06F9/38 , G06F9/06 , G06F9/30
Abstract: 본발명의일 측면에따른모듈로스케줄링방법은, 루프에서인접된반복(iteration)들간의시작간격(Initiation Interval)에대하여, 적어도둘 이상의후보의시작간격들을산출하는단계; 상기후보의시작간격들을이용하여명령어들의스케줄들을병렬적으로탐색하는단계; 및상기탐색된스케줄들의유효성에따라서어느하나의스케줄을선택하는단계를포함한다.
Abstract translation: 根据本发明的一个方面，模调度方法包括以下步骤：计算环路中相邻迭代的起始间隔的至少两个候选的起始间隔; 通过使用候选者的启动间隔来并行地搜索指令的时间表; 并根据所搜索的日程表的有效性来选择日程表。
-
公开(公告)号：KR1020140122111A
公开(公告)日：2014-10-17
申请号：KR1020130038791
申请日：2013-04-09
Applicant: 삼성전자주식회사
CPC classification number: G06F9/3887 , G06F9/3009
Abstract: 명시적인 쓰레드 스위칭을 이용한 다중 쓰레드 실행 장치 및 방법에 관한 것으로 일 양상에 따른 다중 쓰레드 실행 장치는 동일한 명령어를 이용하여 동시에 실행 가능한 복수의 쓰레드를 포함하는 하나 이상의 쓰레드 묶음으로 구성된 쓰레드 묶음 대기열을 저장하는 쓰레드 묶음 제공부 및 실행 중인 쓰레드 묶음의 스위칭 시점을 판단하여 쓰레드 묶음의 스위칭을 제어하는 쓰레드 스케줄러를 포함할 수 있다.
Abstract translation: 本发明涉及使用显式线程切换执行多个线程的装置和方法。 根据本发明的一个方面，一种用于执行多个线程的装置包括：线程池提供单元，其存储由包括多个线程的一个或多个线程池组成的线程池队列，所述线程池可以通过使用多个线程同时执行 同样的指示 以及线程调度器，其确定正在执行的线程池的切换定时，并控制线程池的切换。
-
公开(公告)号：KR1020130053632A
公开(公告)日：2013-05-24
申请号：KR1020110119147
申请日：2011-11-15
Applicant: 삼성전자주식회사
CPC classification number: G06F9/30032 , G06F9/3828
Abstract: PURPOSE: A device for minimizing overheads caused by communication between clusters and a method thereof are provided to improve the performance of codes used in a processor including a cluster structure by minimizing commands moving or copying a value between clusters. CONSTITUTION: A suggestive command generating unit(110) generates suggestive commands including copy commands to be executed for general commands in a basic block. An operand providing unit(130) inserts an operand providing command for providing an operand value of a copy command to be executed according to the execution of the suggestive command. The suggestive command includes an operand including the copy command to be executed with the general command. The operand including the copy command is formed of a bit for expressing the copy command according to settings of a value of 0 or 1. [Reference numerals] (110) Suggestive command generating unit; (120) Scheduling unit; (130) Operand providing unit
Abstract translation: 目的：提供一种用于最小化由群集之间的通信引起的开销的设备及其方法，以通过最小化在群集之间移动或复制值的命令来提高包括群集结构的处理器中使用的代码的性能。 构成：暗示命令生成单元（110）生成暗示命令，包括要在基本块中的一般命令执行的复制命令。 操作数提供单元（130）插入用于根据暗示命令的执行来提供要执行的复制命令的操作数值的操作数提供命令。 暗示命令包括一个操作数，其中包括要用一般命令执行的复制命令。 包括复制命令的操作数由用于根据值0或1的设置表示复制命令的位形成。[附图标记]（110）指示命令生成单元; （120）调度单位; （130）操作数提供单元
-
公开(公告)号：KR102244135B1
公开(公告)日：2021-04-23
申请号：KR1020140119362
申请日：2014-09-05
Applicant: 삼성전자주식회사 , 서울대학교산학협력단
Abstract: 루프(loop)가포함하는명령어들간의의존관계를포함하는데이터의존성그래프(dependence graph)를획득하고데이터의존성그래프에서, 사이클을형성하는명령어들의집합인리커런스(recurrence)를탐색하고, 리커런스의처리순서를결정하는스케줄링하고, 리커런스를탐색하는것은이터레이션디스턴스(Iteration Distance)의합이소정의임계값보다작은리커런스만을탐색하고, 이터레이션디스턴스는리커런스에포함된소정의명령어를처리한후 소정의명령어의처리결과에의존하는다른명령어를처리할수 있을때까지소요되는이터레이션횟수인모듈로스케줄링방법이개시된다.
-
公开(公告)号：KR101700406B1
公开(公告)日：2017-01-31
申请号：KR1020100114044
申请日：2010-11-16
Applicant: 삼성전자주식회사
CPC classification number: G06F15/7867 , G06F8/452 , G06F9/38 , G06F9/3853 , G06F9/3897 , G06F11/3409
Abstract: 재구성가능어레이의실행모드를동적으로결정하기위한장치및 방법이제공된다. 본발명의일 양상에따르면, 어떤루프의실행전에또는실행도중에루프에관한성능정보가얻어진다. 성능정보는루프를 VLIW 모드에서실행했을때와 CGA 모드에서실행했을때 어떤것이유리한지를나타낸다. 성능정보에서변수는루프의루프반복횟수이다. 그루프의루프반복횟수가정해지면성능정보에기초하여유리한모드를선택한다. 만약, 루프의루프반복회수가정해지지않는경우루프의실행시간에대한예측값을이용하여유리한모드를선택한다. 예측값은루프의실행시간을측정하고측정된값과이전의예측값에가중치를적용하여누적적으로합한값을이용할수 있다.
Abstract translation: 提供了用于动态地确定可重构阵列的执行模式的装置和方法。 可以在循环执行之前和/或期间获得循环的性能信息。 性能信息可用于确定是否以非常长的指令字（VLIW）模式或粗粒度阵列（CGA）模式操作装置。
-
-
-
-
-
-
-
-
-
Search Results
32
records
(took 0.023 seconds)
