-
公开(公告)号：KR101334189B1
公开(公告)日：2013-11-28
申请号：KR1020130100570
申请日：2013-08-23
Applicant: 삼성전자주식회사
Abstract: 본 발명의 일 실시 예는 멀티 프로세서 시스템 온 칩에서의 메모리 관리 방법에 관한 것으로, 본 발명의 일 실시 예에 따른 메모리 관리 방법은 소정의 함수의 하부에서 사용하는 메모리 영역이 있는지 판단하여, 메모리 영역을 제1 메모리에 할당할 수 있는지 판단하는 단계, 판단 결과, 메모리 영역을 제1 메모리에 할당할 수 없는 경우, 제2 메모리에 할당되는 소정의 크기만큼을 제1 메모리에 저장한 경우의 성능 이득과, 제1 메모리에서 사용 중인 메모리 영역의 크기만큼을 제2 메모리로 복사하고, 복원하는 비용을 비교하는 단계 및 비교 결과, 성능 이득이 큰 경우, 메모리 영역의 크기만큼을 제2 메모리로 복사하고, 메모리 영역의 크기만큼을 제1 메모리에 할당하는 단계를 포함한다.
-
公开(公告)号：KR1020080068468A
公开(公告)日：2008-07-23
申请号：KR1020070006298
申请日：2007-01-19
Applicant: 삼성전자주식회사
CPC classification number: G06F8/4441
Abstract: A memory management method in a multi-processor system on chip is provided to manage a scratch pad memory efficiently and allocate a memory effectively for even a dynamically generated task and a dynamically loaded task. A position at which the code of a variable or a function corresponding to a symbol is stored is determined on the basis of memory access frequency of the code of the variable or function with reference to a predetermined symbol table(S300,S302). The determined position is compared with a previous position to determine whether the two positions are different from each other. The code of a variable or a function, stored in the previous position, is copied to the determined position when the two positions are different from each other(S308).
Abstract translation: 提供了一种多处理器片上系统的存储器管理方法，用于有效地管理临时存储器存储器并有效地为即使动态生成的任务和动态加载的任务分配存储器。 参照预定的符号表，基于变量或函数的代码的存储器访问频率来确定与存储符号对应的变量或功能的代码的位置（S300，S302）。 将确定的位置与先前位置进行比较，以确定两个位置是否彼此不同。 存储在前一位置的变量或功能的代码在两个位置彼此不同时被复制到所确定的位置（S308）。
-
公开(公告)号：KR1020080064577A
公开(公告)日：2008-07-09
申请号：KR1020070001643
申请日：2007-01-05
Applicant: 삼성전자주식회사
CPC classification number: G06F13/24 , G06F9/5016 , G06F2209/504 , Y02D10/22 , G06F9/06 , G06F12/00 , G06F12/06
Abstract: A method and a device for preventing stack overflow in an embedded system are provided to prevent the stack overflow without causing runtime overflow by removing need for continuously checking whether the stack overflow is generated during runtime, as a maximum stack usage of a main function of functions or threads provided from a kernel is stored, and a task is selectively processed by comparing the stored maximum stack usage with the size of an assigned stack or an available memory. A stack usage storing part(110) stores maximum stack usage of functions provided from a kernel. A processing module(140) task processes a requested task selectively based on at least one maximum stack usage needed for executing the requested task among the maximum stack usage of the functions and the size of an available area among stacks for the requested task. The requested task is interrupt processing, at least one function is an interrupt handler function corresponding to an interrupt, and the stack for the requested task is an interrupt stack(170). The processing module includes an interrupt module(142) preventing occurrence of the additional interrupt and calling the interrupt handler function.
Abstract translation: 提供了一种用于防止嵌入式系统中的堆栈溢出的方法和设备，以防止堆栈溢出而不会导致运行时溢出，因为通过消除在运行时间内是否持续检查堆栈溢出是否产生堆栈溢出的需要，作为函数的主要功能的最大堆栈使用 或从内核提供的线程被存储，并且通过将所存储的最大堆栈使用量与所分配的堆栈或可用存储器的大小进行比较来选择性地处理任务。 堆栈使用存储部分（110）存储从内核提供的功能的最大堆栈使用。 处理模块（140）任务基于所请求的任务的堆栈中功能的最大堆栈使用和可用区域的大小，基于在所请求的任务中执行所请求的任务所需的至少一个最大堆栈使用来选择性地处理所请求的任务。 所请求的任务是中断处理，至少一个功能是对应于中断的中断处理函数，并且所请求的任务的堆栈是中断堆栈（170）。 处理模块包括一个中断模块（142），用于防止发生额外的中断并调用中断处理程序功能。
-
公开(公告)号：KR101544889B1
公开(公告)日：2015-08-18
申请号：KR1020070106850
申请日：2007-10-23
Applicant: 삼성전자주식회사
IPC: G06Q50/10
CPC classification number: A63F13/35 , A63F13/12 , A63F13/60 , A63F2300/308 , A63F2300/535 , A63F2300/5526
Abstract: 본발명은가상세계에서의이벤트를관리하는장치및 그방법에관한것으로서, 더욱상세하게는가상세계에서발생되는동적인이벤트를자동적으로감지하여변경된가상세계의정보를실시간으로제공하는장치및 방법에관한것이다. 본발명의일 실시예에따른가상세계에서의이벤트를관리하는장치는, 가상세계에서발생하는이벤트를모니터링하여새로운이벤트의발생여부를감지하는이벤트감지부와, 상기이벤트발생시 변경되는이벤트상황을스냅샷이미지로생성하는스냅샷관리부와, 상기이벤트발생시 생성된스냅샷이미지를실시간으로제공하는제어부를포함한다.
-
公开(公告)号：KR101334176B1
公开(公告)日：2013-11-28
申请号：KR1020070006298
申请日：2007-01-19
Applicant: 삼성전자주식회사
CPC classification number: G06F8/4441
Abstract: 본 발명은 멀티 프로세서 시스템 온 칩에서의 메모리 관리 방법에 관한 것으로, 본 발명은 심볼 테이블을 참조하여 심볼에 대응하는 변수 또는 함수의 코드의 메모리 접근 빈도 수를 기초로 저장 위치를 결정하고, 결정된 저장 위치와 이전에 저장되어 있던 위치를 비교하여, 결정된 저장 위치가 다른 위치인지 판단하여 다른 위치인 경우, 결정된 저장 위치에 이전 위치에 저장된 변수 또는 함수의 코드를 복사함으로써, 메모리의 지역성을 반영하고, 한정된 메모리 자원을 효과적으로 사용할 수 있다.
스크래치 패드 메모리, 복사, 스택Abstract translation: 存储本发明涉及在系统级芯片多处理器的存储器管理方法中，本发明确定的存储位置的变量或基于存储器访问到对应于该符号与参考符号表中的函数的代码的频率的数量，所确定 通过比较存储在位置作为旧的位置，在该情况下在不同的位置所确定的存储位置，以确定是否其他位置，通过从存储在先前的位置到确定的存储位置的变量或函数复制代码，反映存储器的局部性， 有限的内存资源可以有效使用。
-
Patent Agency Ranking
公开(公告)号：KR1020090041247A
公开(公告)日：2009-04-28
申请号：KR1020070106850
申请日：2007-10-23
Applicant: 삼성전자주식회사
IPC: G06Q50/10
CPC classification number: A63F13/35 , A63F13/12 , A63F13/60 , A63F2300/308 , A63F2300/535 , A63F2300/5526 , G06Q50/10
Abstract: An apparatus and a method for managing an event for generating an event of virtual world by generating snapshot image are provided to provide the information of the changed physical world by directly sensing dynamic event generated in the virtual world. An event sensor(120) monitors and senses an event generated in the physical world through a monitor. A snapshot manager(140) produces the changed event situation in the event triggered to the snapshot image. A controller(160) provides the generated snapshot image on a real time basis. The event is the event generated in experience or the physical world of the avatar existing within the physical world. An information generator(130) detect and stores information about the generated event.
Abstract translation: 提供一种用于通过生成快照图像来管理用于生成虚拟世界的事件的事件的装置和方法，以通过直接感知在虚拟世界中生成的动态事件来提供改变的物理世界的信息。 事件传感器（120）通过监视器监视和感测在物理世界中产生的事件。 快照管理器（140）在触发到快照图像的事件中产生改变的事件情况。 控制器（160）实时地提供生成的快照图像。 事件是在物理世界中存在的头像的经验或物理世界中产生的事件。 信息发生器（130）检测并存储关于生成的事件的信息。
-
公开(公告)号：KR1020080050117A
公开(公告)日：2008-06-05
申请号：KR1020060120953
申请日：2006-12-01
Applicant: 삼성전자주식회사
Abstract: A method and a system for reinforcing reliability of embedded software are provided to detect an error of a source code, and process the error based on detected error information, thereby recognizing an error of an operating system, middleware, and applications of an embedded computer system. A system for reinforcing reliability of embedded software includes an error detector(110) detecting an error of a source code(100) in accordance with an error detecting syntax added to the source code of the software, and an error handler(120) receiving the detected error information from the error detector for processing the error based on the error information.
Abstract translation: 提供了一种用于增强嵌入式软件可靠性的方法和系统，用于检测源代码的错误，并根据检测到的错误信息处理错误，从而识别嵌入式计算机系统的操作系统，中间件和应用程序的错误 。 一种用于增强嵌入式软件的可靠性的系统，包括根据添加到软件的源代码的错误检测语法来检测源代码（100）的错误的错误检测器（110），以及错误处理器（120），其接收 从误差检测器检测出的错误信息，用于基于错误信息处理错误。
-
公开(公告)号：KR101345373B1
公开(公告)日：2013-12-24
申请号：KR1020070001644
申请日：2007-01-05
Applicant: 삼성전자주식회사
IPC: H04L12/859
CPC classification number: H04L47/564 , H04L47/19 , H04L47/24 , H04L47/50
Abstract: 본발명은실시간시스템의송신방법및 장치, 수신방법및 장치에관한것으로, 본발명에따른실시간시스템의송신방법은소켓에지정된소정송신옵션에기초하여소켓에속하는태스크를스케쥴링하는단계및 스케쥴링된태스크에의해생성된패킷을소정송신옵션에기초하여송신하는단계를포함하는것을특징으로한다. 본발명에따르면, 네트워크통신의실시간성을보장할수 있으며, 시스템의자원을효율적으로사용하면서, 요구되는송수신특성에따라데이터를송수신할수 있다.
-
公开(公告)号：KR1020130114023A
公开(公告)日：2013-10-16
申请号：KR1020130100570
申请日：2013-08-23
Applicant: 삼성전자주식회사
Abstract: PURPOSE: A method for managing a memory in a multi processor system on chip (SOC) is provided to use restricted memory resource effectively as reflecting locality of the memory, by copying a code of a function or a variable stored previously in a determined storage position. CONSTITUTION: It is judged whether a memory area is able to be allocated to a first memory or not, by judging whether there is a memory area used in the lower part of a fixed function (S504). If the memory area is not able to be allocated to the first memory, performance gain in the case that a fixed amount allocated to a second memory is stored in the first memory is compared with the cost of copying/restoring the amount of a memory area used in the first memory to the second memory (S506). If the performance gain is larger than the cost, the amount as big as the memory area is copied to the second memory, and the amount as big as the memory area is allocated to the first memory (S509). [Reference numerals] (AA) Start; (BB,CC,FF) Yes; (DD,EE) No; (GG) End; (S500) Analyze a compiler at the beginning of a function; (S502) Is there any memory only used at the bottom of the function ?; (S504) Can the memory be assigned to a scratch pad memory ?; (S505) Assign to a scratch pad memory; (S506) Calculate an alpha value and a beta value; (S508) Alpha > Beta ?; (S509) Assign to a DRAM; (S510) Copy, function execution, recovery
Abstract translation: 目的：提供一种用于管理多处理器片上系统（SOC）中的存储器的方法，以有效地使用受限存储器资源来反映存储器的局部性，通过将先前存储在确定的存储位置中的函数或变量的代码复制 。 构成：通过判断是否存在在固定功能的下半部分中使用的存储区域来判断存储区域是否能够被分配给第一存储器（S504）。 如果存储区域不能被分配给第一存储器，则在将分配给第二存储器的固定量存储在第一存储器中的情况下的性能增益与复制/恢复存储区域的量的成本进行比较 在第一存储器中用于第二存储器（S506）。 如果性能增益大于成本，则与存储区域一样大的量被复制到第二存储器，并且与存储区域一样大的量被分配给第一存储器（S509）。 （附图标记）（AA）开始; （BB，CC，FF）是; （DD，EE）否; （GG）结束; （S500）在函数开头分析一个编译器; （S502）功能底部是否只有内存使用？ （S504）可以将内存分配给暂存器存储器吗？ （S505）分配到暂存卡存储器; （S506）计算α值和β值; （S508）Alpha> Beta？ （S509）分配给DRAM; （S510）复制，功能执行，恢复
-
公开(公告)号：KR1020080064578A
公开(公告)日：2008-07-09
申请号：KR1020070001644
申请日：2007-01-05
Applicant: 삼성전자주식회사
IPC: H04L12/859
CPC classification number: H04L47/564 , H04L47/19 , H04L47/24 , H04L47/50
Abstract: A transmission method and apparatus and a reception method and apparatus of a real time system are provided to guarantee the characteristics of real time of network communication and transmit/receive required data while effectively using resources of a system. When tasks(160) want to communicate with a network, a socket scheduler(110) generates a socket. A protocol stack(120) controls a flow of a packet to be transmitted or a received packet, adds a header to the packet or separates a header added to the packet. A transmission packet manager(130) transmits a packet generated from a task scheduled by the socket scheduler via a network interface(140). A reception packet manager(150) transfers the packet received via the network interface and the protocol stack to a task belonging to a corresponding socket, namely, a tack of a destination of the received packet, based on options designated for the corresponding socket.
Abstract translation: 提供一种实时系统的传输方法和装置及其接收方法和装置，以保证网络通信的实时性和发送/接收所需数据的特性，同时有效地利用系统的资源。 当任务（160）想要与网络通信时，套接字调度器（110）生成套接字。 协议栈（120）控制要发送的分组的流或接收的分组，向分组添加报头或分离添加到分组的报头。 发送分组管理器（130）经由网络接口（140）发送由套接字调度器调度的任务生成的分组。 接收分组管理器（150）基于为相应的套接字指定的选项，将经由网络接口和协议栈接收到的分组传送到属于对应套接字的任务，即接收分组的目的地的粘性。
-
-
-
-
-
-
-
-
-
Search Results
13
records
(took 0.02 seconds)
