TWI227836B - A computer system with multi-function input/output interface architecture - Google Patents

A computer system with multi-function input/output interface architecture Download PDF

Info

Publication number
TWI227836B
TWI227836B TW92130578A TW92130578A TWI227836B TW I227836 B TWI227836 B TW I227836B TW 92130578 A TW92130578 A TW 92130578A TW 92130578 A TW92130578 A TW 92130578A TW I227836 B TWI227836 B TW I227836B
Authority
TW
Taiwan
Prior art keywords
bus
chip
multiplexed
interface structure
computer system
Prior art date
Application number
TW92130578A
Other languages
Chinese (zh)
Other versions
TW200515183A (en
Inventor
Chun-Jung Hsu
I-Tzuo Chang
Chia-Hsing Yu
Original Assignee
Via Tech Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Via Tech Inc filed Critical Via Tech Inc
Priority to TW92130578A priority Critical patent/TWI227836B/en
Application granted granted Critical
Publication of TWI227836B publication Critical patent/TWI227836B/en
Publication of TW200515183A publication Critical patent/TW200515183A/en

Links

Landscapes

  • Bus Control (AREA)

Abstract

A computer system with multi-function input/output interface architecture. The computer system uses the bus of a plurality of multi-function IO interface between central processing unit (CPU), memory modules, peripheral components and chipset units. The computer system includes a CPU; a first chip connected to CPU via a front side bus; a system memory connected to the first chip via a bus of a multi-ratio bandwidth memory architecture; a graphic card module connected to the first chip via a bus of a multi-function IO interface architecture and a second chip connected to the first chip, the peripheral components and a plurality of connecting ports via a bus of a multi-function IO interface architecture.

Description

1227836 五、發明說明(1) 一、【發明所屬之技術領域】 本舍明係於電細糸統架構中連接匯流排使用一具有多 工輸出入介面架構之電腦系統,特別是有關於一種使 用P C I - E X p r e s S匯流排介面之電腦系統架構。 一、【先别技術】 現行或習用技術中,因為中央處理單元(Centr«a i Pirocessing Unit, CPU)的數據處理速度遠大於此中央處 理單元與系統記憶體(即隨機動態存取記憶體,DRAM)之 匯流排資料傳輸速度,而中央處理單元處理資料時,須隨 時對該系統記憶體中作存取的動作,故欲提昇整體電腦系 統效能,可以針對資料傳輸之匯流排作改進。 以現行普遍之架構而言,請參閱第一圖習用技術電腦 系統内部架構第一示意圖,如圖所示,電腦系統架構包括 有處理整體電腦系統各樣資訊數據運算之中央處理單元1 1 ,中央處理單元11藉第一晶片1 2與系統記憶體1 3與其他週 邊連接,此第一晶片1 2於一般個人電腦架構中之北橋晶片 (north bridge),而中央處理單元η與第一晶片12之第 一系統匯流排101稱之為前端系統匯流排(front side bus),第一晶片1 2中設置有記憶體控制器,以第二系統 匯流排1 0 2連接糸統§己憶體1 3 ’現行大多以6 4位元之雙倍 資料傳輸率(Double Data Rate, DDR)匯流排來連接, 系統記憶體1 3即以雙倍資料傳輸率記憶體模組(Doub 1 e1227836 V. Description of the invention (1) 1. [Technical field to which the invention belongs] The present invention relates to the connection of buses in the electrical system architecture using a computer system with a multiplexed I / O interface architecture, and particularly relates to a use Computer system architecture of PCI-EX pres S bus interface. I. [Other technologies] In current or conventional technologies, the data processing speed of the central processing unit (Centr «ai Pirocessing Unit, CPU) is much faster than this central processing unit and system memory (ie, random dynamic access memory, DRAM ) Of the bus data transmission speed, and the central processing unit must access the system memory at any time when processing the data, so to improve the overall computer system performance, you can improve the data transmission bus. In terms of the current general architecture, please refer to the first diagram of the internal architecture of the conventional computer system. As shown in the figure, the computer system architecture includes a central processing unit 1 1 that processes various information and data operations of the overall computer system. The processing unit 11 is connected to other peripherals through the first chip 12 and the system memory 13. This first chip 12 is in a north bridge in a general personal computer architecture, and the central processing unit η and the first chip 12 The first system bus 101 is called the front side bus. The first chip 12 is provided with a memory controller. The second system bus 1 0 2 is connected to the system § self-memory 1 3 'Currently, most of them are connected by 64-bit double data rate (Double Data Rate, DDR) bus. The system memory 1 3 is a double data rate memory module (Doub 1 e

Data Rate_Synchronous DRAM,DDR-SDRAM)來實施,第 一晶片1 2更藉第三系統匯流排1 0 3來存取資料至顯示卡模Data Rate_Synchronous DRAM (DDR-SDRAM), the first chip 12 and the third system bus 1 103 are used to access the data to the graphics card module

1227836 五、發明說明(2) 組1 4,此第三系統匯流排1 〇 3現行多為繪圖加速連接淳( Accelerated Graphic Port,AGP)之匯流排實施,此綠 圖加速連接埠是電腦主機板上介面上插槽的一種,專為^會 圖加速連接埠顯示卡而設計,功能在於輸送往返於中央^ 理單元與繪圖顯示卡之間的影像資訊。而第一晶片丨2更藉 第四系統匯流排1 0 4連接至第二晶片! 6,此第四系統匯流 排1 04現行以週邊元件連接介面匯流排(per ipheral C〇mp0nent interconnect, pci)實施,此第四系統匯流 排1 0 4不僅連接第一曰曰片1 2與第二晶片1 6,更為電腦系統 各樣擴充介面卡之傳輸匯流排,如聲效卡、網路卡等。此 第一晶片1 6習用為晶片組中俗稱之南橋晶片(s〇uth bridge) ’並藉第二晶片丨6連接控制各週邊,如鍵盤、滑 鼠U邊裝置17,如磁碟控制器18,用以連接硬碟控制器 串^控制15等週邊儲存裝置’更有複數種連接蟑19,如 ==阜(senal port)、平行埠(paralUl 用串列匯流排(USB)等。 P rt) :¾ 太圓ΐ,ΐ係為習用技術電腦系統内部架構第二示咅圖, 本”憶體控制器由第一晶片12移至中:』單 Ϊ接中ί :第六糸統匯流排1 06連接控制系統記恃體1 3, 連接中央處理單元i i之第一曰'。己u體Ν 105連接一圖框暫存琴. 日曰 第五糸統匯流排 ^ ^ ^ ^ ν frame buffer) 21» w ^ ^ ^ ^ 匯流排105現行使用64位元之雔1此弟五糸統 ,圖框暫存器2 1係ffl认 又。、"專輸率匯流排實施 ... ° 1係用於内建於電腦主機板上之黠-θ ϋ , 為使用繪圖加速連接埠 ^上之顯不晶片 早匯/瓜排之顯不卡杈組1 4之另外選擇1227836 V. Description of the invention (2) Group 1 4. The third system bus 1 03 is currently implemented by the Accelerated Graphic Port (AGP) bus. The green map acceleration port is the computer motherboard. One of the slots on the upper interface is specially designed for the graphics acceleration port graphics card. Its function is to transmit the image information between the central processing unit and the graphics card. The first chip 丨 2 is also connected to the second chip by the fourth system bus 1 0 4! 6. The fourth system bus 1 04 is currently implemented with peripheral component connection interface buses (Peripheral C0mp0nent interconnect, pci). The fourth system bus 1 0 4 not only connects the first chip 12 and the first. The two chips 16 are more expansion buses for various computer systems, such as sound cards and network cards. This first chip 16 is commonly used as a south bridge in the chipset, and it is controlled by the second chip 丨 6 to control various peripherals, such as a keyboard, a mouse U-side device 17, such as a disk controller 18 , Used to connect the hard disk controller string ^ control 15 and other peripheral storage devices, there are multiple types of connection cockroach 19, such as = = fu (senal port), parallel port (paraalUl serial bus (USB), etc.) P rt ): ¾ Taiyuan, this is the second diagram of the internal architecture of the computer system of conventional technology. This "memory controller" was moved from the first chip 12 to the center: "Single connection": the sixth system bus 1 06 is connected to the control system recording body 1 3, and is connected to the first processing unit of the central processing unit ii. 己 u body N 105 is connected to a frame for temporary storage of the piano. ) 21 »w ^ ^ ^ ^ The bus 105 currently uses 64-bit units. This is the fifth system. The frame register 2 is a ffl system. &Quot; Special input rate bus implementation ... ° 1 is for 黠 -θ 内 built into the motherboard of the computer. It is an additional option for the display chip early remittance / melon display card set 1 on the graphics acceleration port ^.

第7頁 1227836 —~—--Page 7 1227836-~ ----

五、發明說明(3) 電 貲 多年來週邊元件連接介面匯流排 腦内資料傳送的主流技術標準,作PCI) 一直是掌管 。而由美國英特爾(INTEL)電 級PCI速度所費不 之現行電腦系統加以改良後成為一第'、統晶片公司將上述 thiM generati〇n 1/〇,3GI〇),=代串、接介面(V. Description of the invention (3) For many years, the mainstream technology standard for data transfer in the brain (peripheral component connection interface bus, PCI) has been in charge. The current computer system, which is costly for Intel-level PCI speed in the United States, has been improved to become the first, and the chip company will take the above thiM generati0n 1 / 〇, 3GI〇), = generation string, interface (

流排資料傳輪速度可望比最高速的f二代串接介面的匯 :之多,進而提升各種應用的執行1技術(二1])快六 習用技術電腦系統内部架構第三示音文=。各請茶閱第三圖 ,11藉第-晶片12與系統記憶冑13:;=其中中央處理單 107連接,此第七系統匯流排107係為依第七糸,匯流排 流排,孫A τ日> 4 α 供組C DDR-SDRAM)匯 SDRAM) ^ ( 加了备二 在不增加自身基準頻率的條件下增 統記愔、、、_ 5己憶體系統數據頻寬,而解決中央處理單元與系 ^思?之間傳輸線路影響和習用電腦系統中儲存系統執 Γ/度慢的問題。而第一晶片1 2與第二晶片1 6係以第三代 接f面之第八系統匯流排108連接。Streaming data transfer speed is expected to be faster than the highest speed of the second-generation serial interface: the second, which will improve the execution of various applications 1 technology (two 1) faster six conventional technology computer system internal architecture third audio =. Please read the third picture, 11 by the first chip-12 and the system memory 胄 13 :; = of which the central processing unit 107 is connected, this seventh system bus 107 is based on the seventh bus, the bus, Sun A τday > 4 α for group C DDR-SDRAM) sink SDRAM) ^ (Added the second to increase the system bandwidth without adding its own reference frequency. 5 ,,, and _ 5 memory system data bandwidth, and solve The problem of the transmission line between the central processing unit and the system and the slow implementation of the storage system in the conventional computer system. The first chip 12 and the second chip 16 are connected to the f-plane by the third generation. Eight system buses 108 are connected.

-第四圖係為習用技術電腦系統内部架構第四示意圖。 二二為將§己憶體控制器移至中央處理單元1 1,並以現行雙 ^二料傳輸率記憶體模組或四倍頻寬記憶體技術之第九系 ί 3 $排109連接系統記憶體13,而第一晶片12亦以雙倍 匯二輸率§己憶體模組或四倍頻寬記憶體技術之第十系統 机排1 1 〇連接圖框暫存器2丨,其中係依實際需要而選擇-The fourth diagram is the fourth schematic diagram of the internal architecture of a conventional computer system. The second is to move the §memory controller to the central processing unit 11 and use the current 9th series of dual-data-rate memory module or quad-band memory technology. 3 $ 排 109 connection system Memory 13, and the first chip 12 is also connected to the frame register 2 丨 by the tenth system bank of the memory module or quadruple bandwidth memory technology. Choose according to actual needs

1227836 五、發明說明^ -—— 雙倍資料傳輪率記憶體模組或四倍頻寬記憶體技術二者其 中之一匯流排。第一晶片1 2與第二晶片丨6亦以第三代串接 介面之第八系統匯流排1 〇 8連接。 以上所述係為現行電腦主機板架構,為改善現行電腦 系統架構内部匯流排傳輸速度,本發明係使用四倍頻寬記 憶體技術(QBM)與有多工傳輸功能之第三代串接介面以 改善整體傳輸效能。 三、【發明内容】1227836 V. Description of the invention ^ ----- One of the double data transfer rate memory module or quad bandwidth memory technology. The first chip 12 and the second chip 6 are also connected by the eighth system bus 108 of the third generation serial interface. The above is the current computer motherboard architecture. In order to improve the internal bus transmission speed of the current computer system architecture, the present invention uses a quadruple-bandwidth memory technology (QBM) and a third-generation serial interface with a multiplexing function. To improve overall transmission performance. Third, [invention content]

本發明係為一種具有多工輸出入介面架構之電腦系 ’係於電細糸統之中央處理單元、記憶體與各週邊與曰'片 組之連接匯流排使用複數個多工輸出入介面架構匯流 達到電腦糸統内部匯流排多路傳輸及增加傳輸速度之目、 。此電腦系統包括有··一中央處理單元;一第一曰^片,的 藉一前端系統匯流排連接該中央處理單元;一系統紀慢係 ’係藉一多倍頻寬記憶體架構之匯流排連接該第一曰二體 曰曰 , 一顯示卡模組,係藉一具有多工輸出入介面架構之匯浐 連接該第一晶片,及一第二晶片,係藉具有多工輸出入丄 面架構之匯流排連接該第一晶片,該第二晶片並連 1 系統之週邊裝置與複數個連接蜂。 电月自The present invention is a computer system with a multiplexed I / O interface structure. It is a central processing unit connected to an electrical system, a memory, and a connection bus between each peripheral and a chip set. A plurality of multiplexed I / O interface structures are used. Confluence achieves the purpose of multiplexing the internal bus of the computer system and increasing the transmission speed. The computer system includes: a central processing unit; a first chip, which is connected to the central processing unit by a front-end system bus; a system chronicle system that uses a multi-bandwidth memory architecture to converge Connected to the first and second body, a graphics card module is connected to the first chip by a sink having a multiplexed I / O interface structure, and a second chip is provided by a multiplexed I / O. The bus of the surface structure is connected to the first chip, and the second chip is connected to peripheral devices of system 1 and a plurality of connection bees. Electric month since

有關本發明之詳細内容及技術,兹就配合圖式說明如 四、【實施方式】 構之電腦 有處理整 請參閱第五圖本發明具有多工輸出入介面架 系統第一示意圖,圖中所示之電腦系統架構包括Regarding the detailed content and technology of the present invention, we will explain it in conjunction with the drawings as follows: [Embodiment] The computer has a processing structure. Please refer to the fifth figure. The first schematic diagram of the multiplexed I / O interface rack system of the present invention. The computer system architecture shown below includes

1227836 五、發明說明(5) 體電腦系統各樣資訊數據運算之中央處理單元1 1,中央處 理單元11藉第一晶片1 2與系統記憶體1 3與其他週邊連接& 中央處理單元1 1與第一晶片1 2之第一系統匯流排1 〇丨以前 端系統匯流排實施,第一晶片1 2中設置有記憶體控制器與 系統記憶體連接,本發明係分別以第一匯流排5 1連接第— 糸統記憶體1 3 a ’以弟·一匯流排5 2連接弟一糸統記憶體1 3 b ,此第一匯流排5 1與第二匯流排5 2同時相容習用之6 4位元 雙倍資料傳輸率之記憶體匯流排與一多倍頻寬記憶體技術 之匯流排,此多倍頻寬記憶體技術之匯流排本發明係以四 倍頻寬記憶體技術之64位元雙倍資料傳輸率匯流排來實施 所述之電腦系統,係 主機板匯流排傳輸速度,本發 記憶體1 3之間使用四倍頻寬記 腦系統,此提昇傳輸頻寬之電 資料傳輪率記憶體模組中加以 率的條件下增加系統記憶體系 理單元與系統記憶體之間傳輸 儲存系統執行速度慢的問題。 要更高時脈頻率的記憶體元件 設計為高頻,這使得實現數據 容易,實現了一種高速、經濟 解決了系統中不斷增加的處理 盾0 為改善現行普遍之電腦架構 明於中央處理單元Η與系統 憶體技術提昇傳輸頻寬之電 腦系統係在現行架構之雙倍 改良,在不增加自身基準頻 統數據頻寬,而解決中央處 線路影響和習用電腦系統中 此四倍頻寬記憶體技術不需 ’因此,此電腦系統也無需 匯流排的高速傳輸設計更加 高效、可擴展的解決辦法, 能力和記憶體頻寬之間的矛1227836 V. Description of the invention (5) Central processing unit 11 for computing various information and data of the body computer system, the central processing unit 11 borrows the first chip 12 and the system memory 13 and connects to other peripherals & the central processing unit 1 1 The first system bus 1 with the first chip 12 is implemented as a front-end system bus. A memory controller is provided in the first chip 12 to connect to the system memory. The present invention uses the first bus 5 1 Connected to the first system memory 1 3 a 'Is the first bus 5 2 Connected to the first system memory 1 3 b, this first bus 5 1 is compatible with the second bus 5 2 4-bit memory bus with double data transfer rate and a multi-bandwidth memory technology bus. This multi-bandwidth memory technology bus is based on 64 of the quad-band memory technology. The double-bit data transmission rate bus is used to implement the computer system described above. It is the motherboard bus transmission speed. The quad-band memory system is used between the memory 13 and the memory system to increase the transmission bandwidth of electrical data. Conditions for adding rates in the pass rate memory module Problem of increasing the transmission system between the memory system storage system processing unit and system memory slow execution. The memory components that require higher clock frequencies are designed to be high frequency, which makes it easy to implement data, achieves a high-speed, economical solution to the increasing processing shield in the system. 0 To improve the current general computer architecture, it is stated in the central processing unit The computer system with the system memory technology to increase the transmission bandwidth is a double improvement in the current architecture. It does not increase the data bandwidth of its own reference frequency, but solves the influence of the central line and the quadruple bandwidth memory in the conventional computer system. No technology is needed '. Therefore, this computer system also does not need the high-speed transmission design of the bus. It is a more efficient and scalable solution. The conflict between capacity and memory bandwidth.

1227836 五、發明說明(6) 四倍頻寬記憶體技術結構的原理是··當欲運瞀處理之 數據,系統記憶體(13a,13b)經第一晶片12輪出^至&中^ 處理單元1 1時,其以一傳輸之基準頻率輸出,以一做為開 關之電晶體在現行雙倍資料傳輸率匯流排之傳輸基準頻^ 的9 0度相(phase)位移輸出作切換,係將每個傳輸之相、 位時間分割,使一個有效的數據位插入到前一個數s據位沒 有佔用的時間内,而達到雙倍於現行雙倍資二 貝枓傳輸率匯流 排架構下之四倍頻寬記憶體架構之目的。 第五圖所示之電腦系統架構中,第一晶片丨2藉第一多 工輸出入介面架構之匯流排53連接顯示卡模纟 f 二 多工輸出人功能之第三代串接介面(multi、ple;;4e/;;;^ generation I/O, 3Gl〇)匯流排代替習用繪圖加速連接蜂 (AGP)匯流排’但仍保有習用繪圖加速連接埠匯流排之 相容性,更增加本發明實施之彈性。另外,藉有第^二多工 輸出人介面架構之匯流排54連接第二晶片丨6 f以代^ ^二 之週邊元件連接介面匯流排(p C I ),可達到更大頻寬而 傳輸速度更快之目的’亦保有習用週邊元件連接介^面%匯流 排之相容性,依不同須有採用之。以上所述之第一多工輸 出入介面架構之匯流排5 3與第二多工輸出入介面架構之匯 <排5 4皆可依照主機板設計而變更使用現行匯流排或是具 有夕工輸出入介面架構匯流排,而增加多重選擇性。第二 晶片1 6更可藉另一多工輸出入介面架構之匯流排連接一多 工輸出入介面橋接晶片(3GI0 bridge chip)(圖示於第 七圖)。1227836 V. Description of the invention (6) The principle of the technology structure of the quadruple bandwidth memory is that when the data to be processed is processed, the system memory (13a, 13b) is output through the first chip 12 rounds ^ to & middle ^ When the processing unit 11 is 1, it outputs at a transmission reference frequency, and uses a transistor as a switch to switch at a 90 degree phase shift output of the transmission reference frequency ^ of the current double data transmission rate bus. The phase and bit time of each transmission is divided, so that a valid data bit is inserted into the time that the previous data bit is not occupied, and it is doubled under the current double-rate two-frame transmission rate bus architecture. Purpose of the quadruple bandwidth memory architecture. In the computer system architecture shown in the fifth figure, the first chip 2 is connected to the graphics card module 53 via the bus 53 of the first multiplexing input / output interface structure. , Ple ;; 4e / ;; ^ generation I / O, 3G10) bus instead of the conventional graphics acceleration connection bee (AGP) bus' but still maintains the compatibility of the conventional graphics acceleration port bus, increasing the cost The flexibility of the invention. In addition, the second multiplexed output interface interface bus 54 is connected to the second chip. 6 f replaces the ^ ^ second peripheral component connection interface bus (p CI), which can achieve greater bandwidth and transmission speed. The purpose of "faster" also maintains the compatibility of the conventional peripheral component connection interface% bus, which must be adopted according to different. The buses 5 3 of the first multiplexed I / O interface structure and the buses of the second multiplexed I / O interface structure described above can be changed and used according to the motherboard design. I / O interface architecture bus, increasing multiple selectivity. The second chip 16 can also be connected to a multiplexed I / O interface bridge chip (3GI0 bridge chip) by using a bus of another multiplexed I / O interface structure (shown in Figure 7).

1227836 ---- -- 五、發明說明(7) 此有夕工輸出入功能之第三代串接介面即對資 工值处 、丨T休夕 碍痴功此之匯流排,可解決現行習用週邊元件連 =緣圖加速連接埠之瓶頸,龙因為有多工傳輸功能,。 使用在各式介面卡與電腦週邊上,亦可單獨存在於任 5 2,卡與週邊之傳輸。因為電腦主機板上之晶片組^如 邊:1^、第二晶片16等)的晶片針腳有限,若因為週 逐接4夕而加多其針腳數目,則會有增加成本及擴充彈 不大之缺失,若使用此多工輸出入功能之第三代串接介 Ϊ二ί針腳可多工使用,使系統擴充與考慮成本因素而; 正°又计之彈性增加,又有節省成本之優點。 &如第/、圖第二示意圖所示,為第五圖第一實施例之第 二悲樣記憶體控制器亦可設置於中央處理單元11中,再 上述,具有多倍頻寬記憶體技術之第一匯流排5 1連接 一曰系統屺憶體13a,並可同時將記憶體控制器設置於第 體m ’使第二系統記憶體1扑藉具有多倍頻寬記情 體技術之第二匯流排52 己= 藉另一多工輸出入介日日片12。弟一晶片16更可 面橋接晶片(3GI0匯流排連接一多工輸出人介 第七圖係為本發明提昇丄p)(圖示於第七圖)。 圖,與中央處理單元丨丨以多頻寬之電腦系統第二示魚 排5 1連接第一系統記憶體丨3 乂寬記憶體架構之第一匯产 ,體J構之第三匯流排73連接弟-晶片12以多倍頻寬記"· 夕工輸出入介面架構之匯流排,樞暫存器21,另外以第_ 用繪圖加速連接埠匯流排,但連接顯示卡模組1 4取代習 有白用繪圖加速連接埠匯 1227836 五、發明說明(8) 流排之相容性,可以多重選擇使用。第一晶片1 2以第二多 工輸出入介面架構之匯流排5 4連接第二晶片16’更設置有 一獨立的第三代串接介面之第三多工輸出入介面架構之匯 流排7 5連接至一多工輸出入介面橋接晶片70,此多工輸出 入介面橋接晶片7 0係為了提供更多連接週邊裝置選擇而設 置,具有多重擴充性。以上所述之第一多工輸出入介面架 構之匯流排5 3與第二多工輸出入介面架構之匯流排5 4皆可 依照主機板設計而變更使用現行匯流排或是具有多工輸出 入介面架構之匯流排,而增加多重選擇性。 以上為本發明具有多工輸出入介面架構之電腦系統實 施例之詳細說明,於電腦系統中各匯流排以複數個多倍頻 寬記憶體架構與多工傳輸功能之第三代串接介面之多重選 擇性,達到增加電腦系統傳輸頻寬與有擴充彈性之目的與 功效。 唯以上所述者,僅為本發明之較佳實施例而已,當不 能以之限定本發明所實施之範圍。即大凡依本發明申請專 利範圍所作之均等變化與修飾,皆應仍屬於本發明專利涵 蓋之範圍内。1227836 -----V. Description of the invention (7) This third-generation serial interface with the input and output function of the industrial machine is a confluence of the industrial and commercial value department, and it can solve the current problem. It is a bottleneck to accelerate the connection of peripheral ports with conventional peripheral components, which is because of the multiplexing function. It can be used on all kinds of interface cards and computer peripherals, and it can also exist in any of them. Because the chipset on the computer motherboard (such as edge: 1 ^, second chip 16, etc.) has a limited number of chip pins, if the number of pins is increased because it is connected every week, it will increase the cost and expand the bomb. If the third-generation serial interface of this multiplexing input / output function is used, the two pins can be used multiplexed, which makes the system expand and consider the cost factor. The flexibility of the positive angle increases and it has the advantage of cost savings. . & As shown in the second schematic diagram of FIG./ and FIG. 2, the second memory controller of the first embodiment of the fifth diagram may also be provided in the central processing unit 11, and as described above, it has a multi-bandwidth memory. The first bus 5 of the technology is connected to the system memory 13a, and the memory controller can be set to the body m 'at the same time, so that the second system memory 1 can borrow the memory technology with multiple bandwidths. The second bus bar 52 has = borrowed another multiplexed output to enter the daily film 12. The first chip 16 can also be a bridge chip (3GI0 bus connected to a multiplexed output agent. The seventh picture is the improvement of the invention) (the picture is shown in the seventh picture). Figure, with the central processing unit 丨 丨 Multi-bandwidth computer system, the second fish bar 5 1 is connected to the memory of the first system 丨 3 The first bus of the wide memory architecture, the third bus of the J bus 73 Connection brother-chip 12 is recorded with multi-bandwidth " · Xigong I / O interface structure of the bus, pivot register 21, and in addition to accelerate the port bus with graphics, but connected to the graphics card module 1 4 In place of Xi Youbai, the graphics port 1227836 is accelerated. 5. Description of the invention (8) The compatibility of the stream line can be used in multiple choices. The first chip 1 2 has a second multiplexed I / O interface structure bus 5 4 is connected to the second chip 16 'and a third multiplexed I / O interface structure has a separate third-generation serial interface bus 7 5 Connected to a multiplexer I / O interface bridge chip 70, this multiplexer I / O interface bridge chip 70 is set to provide more options for connecting peripheral devices, and has multiple expandability. The bus 5 3 of the first multiplexed I / O interface structure and the bus 5 4 of the second multiplexed I / O interface structure described above can be changed to use the current bus or have multiplexed I / O according to the motherboard design. Interface architecture bus, increasing multiple options. The above is a detailed description of an embodiment of a computer system with a multiplexed I / O interface structure according to the present invention. Each bus in the computer system uses a plurality of multi-bandwidth memory structures and a third-generation serial interface with a multiplexing function. Multiple selectivity, to achieve the purpose and effect of increasing the transmission bandwidth and expansion flexibility of the computer system. The above are only the preferred embodiments of the present invention, and it should not be used to limit the scope of the present invention. That is, all equal changes and modifications made in accordance with the scope of the patent application for the present invention should still fall within the scope of the patent of the present invention.

第13頁 1227836_ 圖式簡單說明 【圖式簡單說明】 第一圖係為習用電腦系統内部架構之第一示意圖; 第二圖係為習用電腦系統内部架構之第二示意圖; 第三圖係為習用電腦系統内部架構之第三示意圖; 第四圖係為習用電腦系統内部架構之第四示意圖; 第五圖係為本發明具有多工輸出入介面架構之電腦系 統的第一示意圖; 第六圖係為本發明具有多工輸出入介面架構之電腦系 統的第二不意圖,以及 第七圖係為本發明具有多工輸出入介面架構之電腦系 統的第三示意圖。 【符號說明】 Π-中央處理單元;12-第一晶片; 1 3-系統記憶體;1 3a-第一系統記憶體; 1 3b-第二系統記憶體;1 4-顯示卡模組; 1 5 -週邊元件連接介面;1 6 -第二晶片; 1 7 -週邊裝置;1 8 -磁碟控制器; 1 9 -連接埠;1 (Π -第一系統匯流排; 1 0 2 -第二系統匯流排;1 0 3 -第三系統匯流排; 1 0 4 -第四系統匯流排;1 0 5 -第五系統匯流排; 1 0 6 -第六系統匯流排;1 0 7 -第七系統匯流排; 1 0 8 -第八系統匯流排;1 0 9 -第九系統匯流排; 1 1 0 -第十系統匯流排;1 Π -第十一系統匯流排; 2 1 -圖框暫存器;5 1 -第一匯流排;Page 13 1227836_ Brief description of the drawings [Simplified description of the drawings] The first diagram is the first diagram of the internal architecture of a conventional computer system; the second diagram is the second diagram of the internal architecture of a conventional computer system; the third diagram is a conventional diagram The third diagram of the internal architecture of a computer system; the fourth diagram is the fourth diagram of the internal architecture of a conventional computer system; the fifth diagram is the first diagram of a computer system with a multiplexed input / output interface architecture according to the present invention; the sixth diagram is This is a second schematic diagram of a computer system with a multiplexed input / output interface structure according to the present invention, and a seventh diagram is a third schematic diagram of a computer system with a multiplexed input / output interface structure according to the present invention. [Symbol description] Π-central processing unit; 12-first chip; 1 3-system memory; 1 3a-first system memory; 1 3b-second system memory; 1 4-graphics card module; 1 5-Peripheral component connection interface; 1 6-Second chip; 1 7-Peripheral device; 1 8-Disk controller; 1 9-Port; 1 (Π-First system bus; 1 0 2-Second System bus; 103-third system bus; 104-fourth system bus; 105-fifth system bus; 106-sixth system bus; 107-seventh System bus; 1 0-8th system bus; 10 9-9th system bus; 1 10-10th system bus; 1 Π-11th system bus; 2 1-frame temporarily Memory; 5 1-the first bus;

第14頁 1227836 圖式簡單說明 5 2-第二匯流排; 5 3 -第一多工輸出入介面架構之匯流排 5 4 -第二多工輸出入介面架構之匯流排 7 〇-多工輸出入介面橋接晶片; 7 3-第三匯流排; 7 5 -第三多工輸出入介面架構之匯流排Page 14 1227836 Schematic description 5 2-Second bus; 5 3-Bus of the first multiplexed I / O interface structure 5 4-Bus of the second multiplexed I / O interface structure 7 〇-Multiplex output Input interface bridge chip; 7 3-third bus; 7 5-third multiplexed I / O interface architecture bus

第15頁Page 15

Claims (1)

1227836 六、申請專利範圍 1. 一種具有多工輸出入介面架構之電腦系統,至少包含 一中央處理單元; 一第一晶片,係藉一前端系統匯流排連接該中央處理 單元; 一系統記憶體,係藉一多倍頻寬記憶體架構之匯流排 連接該第一晶片; 一顯示卡模組,係藉一第一多工輸出入介面架構之匯 流排連接該第一晶片;及 一第二晶片,係藉一第二多工輸出入介面架構之匯流 排連接該第一晶片,該第二晶片並連接電腦系統之週邊裝 置與複數個連接埠; 其中,該第一多工輸出入介面架構之匯流排係在第三 代輸出入介面(3GI0)匯流排與周邊元件匯流排(PCI bus) 之間作切換。 2. 如申請專利範圍第1項所述之具有多工輸出入介面架構 之電腦系統,其中該第一晶片為一北橋晶片與該第二晶片 為一南橋晶片。 3. 如申請專利範圍第1項所述之具有多工輸出入介面架構 之電腦系統,其中該系統記憶體與該第一晶片之傳輸匯流 排為可選擇該四倍頻寬記憶體(QBM)架構與一雙倍資料傳 輸率(DDR)之匯流排。 4. 如申請專利範圍第1項所述之具有多工輸出入介面架構 之電腦系統,其中該第一多工輸出入介面架構選擇性切換1227836 6. Scope of patent application 1. A computer system with a multiplexed I / O interface structure, which includes at least a central processing unit; a first chip, which is connected to the central processing unit by a front-end system bus; a system memory, Is connected to the first chip by a bus with a multi-times bandwidth memory structure; a display card module is connected to the first chip by a bus with a first multiplexed I / O interface structure; and a second chip Is connected to the first chip by a bus of a second multiplexed input / output interface structure, and the second chip is connected to peripheral devices of the computer system and a plurality of ports; wherein the first multiplexed input / output interface structure The bus is switched between the third-generation I / O interface (3GI0) bus and the peripheral component bus (PCI bus). 2. The computer system with a multiplexed input / output interface structure as described in item 1 of the scope of the patent application, wherein the first chip is a north bridge chip and the second chip is a south bridge chip. 3. The computer system with a multiplexed I / O interface structure as described in item 1 of the scope of patent application, wherein the transmission bus of the system memory and the first chip is selectable from the quad-band memory (QBM) Architecture and a double data rate (DDR) bus. 4. The computer system with a multiplexed input / output interface structure as described in item 1 of the scope of patent application, wherein the first multiplexed input / output interface structure is selectively switched 第16頁 1227836_ 六、申請專利範圍 一繪圖加速連接埠(AGP )匯流排與多工傳輸功能之第三代 串接介面。 5. 如申請專利範圍第1項所述之具有多工輸出入介面架構 之電腦系統,更包含一第三多工輸出入介面架構匯流排連 接該第一晶片與一多工輸出入介面橋接晶片(3GI0 bridge chip)0 6. —種具有多工輸出入介面架構之電腦系統,至少包含 一中央處理單元; 一第一系統記憶體,係藉具有多倍頻寬記憶體架構之 第一匯流排連接該中央處理單元; 一第一晶片,係藉一前端系統匯流排連接該中央處理 單元; 一第二系統記憶體,係藉具有多倍頻寬記憶體架構之 第二匯流排連接該第一晶片; 一顯示卡模組,係藉一第一多工輸出入介面架構之匯 流排連接該第一晶片;及 一第二晶片,係藉一第二多工輸出入介面架構之匯流 排連接該第一晶片,該第二晶片並連接電腦系統之週邊裝 置與複數個連接琿; 其中,該第一多工輸出入介面架構之匯流排係在第三 代輸出入介面(3GI0)匯流排與周邊元件匯流排(PCI bus) 之間作切換。 7. 如申請專利範圍第6項所述之具有多工輸出入介面架構Page 16 1227836_ VI. Scope of patent application A graphics acceleration port (AGP) third-generation serial interface with bus and multiplex transmission functions. 5. The computer system with a multiplexed I / O interface structure described in item 1 of the scope of patent application, further includes a third multiplexed I / O interface structure bus connecting the first chip with a multiplexed I / O interface bridge chip. (3GI0 bridge chip) 0 6. —A computer system with a multiplexed I / O interface structure that includes at least a central processing unit; a first system memory that borrows the first bus with a multi-bandwidth memory architecture Connected to the central processing unit; a first chip connected to the central processing unit via a front-end system bus; a second system memory connected to the first via a second bus having a multi-bandwidth memory architecture A chip; a graphics card module connected to the first chip via a bus of a first multiplexed I / O interface structure; and a second chip connected to the first chip via a bus of a second multiplexed I / O interface structure The first chip, the second chip and peripheral devices connected to the computer system and a plurality of connections; wherein the bus of the first multiplexed I / O interface structure is in the third-generation output Interface (3GI0) as a switching element between the bus and the peripheral bus (PCI bus). 7. Multiplexed input / output interface structure as described in item 6 of the scope of patent application 第17頁 1227836_ 六、申請專利範圍 之電腦系統,其中該第一晶片為一北橋晶片與該第二晶片 為一南橋晶片。 8. 如申請專利範圍第6項所述之具有多工輸出入介面架構 之電腦系統,其中該多工輸出入介面架構之匯流排為一多 工傳輸功能之第三代串接介面。 9. 如申請專利範圍第6項所述之具有多工輸出入介面架構 之電腦系統,其中該第一系統記憶體與該中央處理單元之 傳輸匯流排為可選擇該多倍頻寬記憶體架構與一雙倍資料 傳輸率之匯流排。 1 0.如申請專利範圍第6項所述之具有多工輸出入介面架 構之電腦系統,其中該第二系統記憶體與該第一晶片之傳 輸匯流排為可選擇該多倍頻寬記憶體架構與一雙倍資料傳 輸率之匯流排。 11.如申請專利範圍第6項所述之具有多工輸出入介面架 構之電腦系統,其中該顯示卡模組與該第一晶片之傳輸匯 流排為可選擇該多工輸出入介面架構與一繪圖加速連接埠 之匯流排。 1 2.如申請專利範圍第6項所述之多工輸出入介面架構之 電腦系統,更包含一第三多工輸出入介面架構麗流排連接 該第一晶片與一多工輸出入介面橋接晶片(3GI0 bridge chip)。 1 3. —種多工輸出入介面架構之電腦系統,至少包含: 一中央處理單元; 一系統記憶體,係藉一多倍頻寬記憶體架構之第一匯Page 17 1227836_ VI. Patent application computer system, wherein the first chip is a north bridge chip and the second chip is a south bridge chip. 8. The computer system with a multiplexed input / output interface structure as described in item 6 of the scope of the patent application, wherein the bus of the multiplexed input / output interface structure is a third-generation serial interface with a multiplexed transmission function. 9. The computer system with a multiplexed input / output interface structure as described in item 6 of the scope of the patent application, wherein the transmission bus of the first system memory and the central processing unit is selectable from the multi-bandwidth memory architecture Bus with double data rate. 10. The computer system having a multiplexed input / output interface structure as described in item 6 of the scope of patent application, wherein the transmission bus of the second system memory and the first chip is selectable from the multi-bandwidth memory Bus of architecture and double data rate. 11. The computer system with a multiplexed input / output interface structure as described in item 6 of the scope of the patent application, wherein the transmission bus of the display card module and the first chip is selected from the multiplexed input / output interface structure and a Graphics acceleration bus for ports. 1 2. The computer system with a multiplexed I / O interface structure as described in item 6 of the scope of the patent application, further comprising a third multiplexed I / O interface structure. The streamer connects the first chip to a multiplexed I / O interface bridge. Chip (3GI0 bridge chip). 1 3. —A computer system with a multiplexed I / O interface architecture, including at least: a central processing unit; a system memory, which is the first sink of a multi-times bandwidth memory architecture 第18頁 1227836 六、申請專利範圍 流排連接該中央處理單元; 一第一晶片,係藉一前端系統匯流排連接該中央處理 單元; 一圖框暫存器,係藉一多倍頻寬記憶體架構之第三匯 流排連接該第一晶片; 一顯示卡模組,係藉一第一多工輸出入介面架構之匯 流排連接該第一晶片; 一第二晶片,係藉第二多工輸出入介面架構之匯流排 連接該第一晶片,該第二晶片並連接電腦系統之週邊裝置 與複數個連接埠;及 一多工輸出入介面橋接晶片,係藉第三多工輸出入介 面架構之匯流排連接至該第一晶片; 藉該多工輸出入介面架構之匯流排取代現行之匯流排 架構以提昇該電腦系統中之傳輸頻寬。 1 4.如申請專利範圍第1 3項所述之多工輸出入介面架構之 電腦系統,其中該第一晶片為一北橋晶片。 1 5.如申請專利範圍第1 3項所述之多工輸出入介面架構之 電腦系統,其中該第二晶片為一南橋晶片。 1 6.如申請專利範圍第1 3項所述之多工輸出入介面架構之 電腦系統,其中該多工輸出入介面架構之匯流排為一多工 傳輸功能之第三代寧接介面。 1 7.如申請專利範圍第1 3項所述之多工輸出入介面架構之 電腦系統,其中該系統記憶體與該中央處理單元之傳輸匯 流排為可選擇該多倍頻寬記憶體架構與一雙倍資料傳輸率Page 18 1227836 VI. Patent application range The bus is connected to the central processing unit; a first chip is connected to the central processing unit by a front-end system bus; a frame register is used to borrow a multi-bandwidth memory A third bus connected to the first chip; a graphics card module connected to the first chip via a first multiplexed input / output interface bus; a second chip connected to the second multiplex The bus of the I / O interface structure is connected to the first chip, the second chip is connected to the peripheral devices of the computer system and a plurality of ports; and a multiplexed I / O interface bridge chip is connected to the third multiplexed I / O interface structure. The bus is connected to the first chip; the current bus structure is replaced by the bus of the multiplexed I / O interface structure to increase the transmission bandwidth in the computer system. 14. The computer system with the multiplexed input / output interface structure described in item 13 of the scope of patent application, wherein the first chip is a Northbridge chip. 1 5. The computer system of the multiplexed input / output interface structure as described in item 13 of the scope of the patent application, wherein the second chip is a south bridge chip. 16. The computer system with a multiplexed input / output interface structure as described in item 13 of the scope of the patent application, wherein the bus of the multiplexed input / output interface structure is a third generation Ning interface with a multiplexed transmission function. 1 7. The computer system with the multiplexed input / output interface structure as described in item 13 of the scope of the patent application, wherein the transmission bus of the system memory and the central processing unit is selectable from the multi-bandwidth memory architecture and Double the data transfer rate 第19頁 1227836_ 六、申請專利範圍 之匯流排。 1 8.如申請專利範圍第1 3項所述之多工輸出入介面架構之 電腦系統,其中該圖框暫存器與該第一晶片之傳輸匯流排 為可選擇該多倍頻寬記憶體架構與一雙倍資料傳輸率之匯 流排。 1 9.如申請專利範圍第1 3項所述之多工輸出入介面架構之 電腦系統,其中該顯示卡模組與該第一晶片之傳輸匯流排 為可選擇該多工輸出入介面架構與一繪圖加速連接琿之匯 流排。 2 0.如申請專利範圍第13項所述之多工輸出入介面架構之 電腦系統,其中該第二晶片與該第一晶片之傳輸匯流排為 可選擇該多工輸出入介面架構與一週邊元件連接介面之匯 流排。Page 19 1227836_ VI. Bus for patent application scope. 1 8. The computer system of the multiplexed input / output interface structure as described in item 13 of the scope of the patent application, wherein the transmission bus of the frame register and the first chip can select the multi-bandwidth memory Bus of architecture and double data rate. 19. The computer system with a multiplexed input / output interface structure as described in item 13 of the scope of patent application, wherein the transmission bus of the display card module and the first chip is selectable from the multiplexed input / output interface structure and A drawing accelerates the connection of the bus. 2 0. The computer system of the multiplexed input / output interface structure as described in item 13 of the scope of the patent application, wherein the transmission bus of the second chip and the first chip is selected from the multiplexed input / output interface structure and a peripheral. Bus for component connection interface. 第20頁Page 20
TW92130578A 2003-10-31 2003-10-31 A computer system with multi-function input/output interface architecture TWI227836B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW92130578A TWI227836B (en) 2003-10-31 2003-10-31 A computer system with multi-function input/output interface architecture

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW92130578A TWI227836B (en) 2003-10-31 2003-10-31 A computer system with multi-function input/output interface architecture

Publications (2)

Publication Number Publication Date
TWI227836B true TWI227836B (en) 2005-02-11
TW200515183A TW200515183A (en) 2005-05-01

Family

ID=35667156

Family Applications (1)

Application Number Title Priority Date Filing Date
TW92130578A TWI227836B (en) 2003-10-31 2003-10-31 A computer system with multi-function input/output interface architecture

Country Status (1)

Country Link
TW (1) TWI227836B (en)

Also Published As

Publication number Publication date
TW200515183A (en) 2005-05-01

Similar Documents

Publication Publication Date Title
US10268619B2 (en) Cross-threaded memory system
JP5784582B2 (en) Configurable bandwidth memory device and method
US9542352B2 (en) System and method for reducing command scheduling constraints of memory circuits
JPS59216269A (en) Random access memory for communication
EP1894110B1 (en) Memory micro-tiling
EP1723526A2 (en) Dynamic command and/or address mirroring system and method for memory modules
US20190042500A1 (en) Dimm for a high bandwidth memory channel
TW544690B (en) Semiconductor memory device and memory system for improving bus efficiency
US11699471B2 (en) Synchronous dynamic random access memory (SDRAM) dual in-line memory module (DIMM) having increased per data pin bandwidth
US7761668B2 (en) Processor architecture having multi-ported memory
US20210081338A1 (en) Channel depth adjustment in memory systems
US10964361B2 (en) Memory component with adjustable core-to-interface data rate ratio
US5978869A (en) Enhanced dual speed bus computer system
US8539196B2 (en) Hierarchical organization of large memory blocks
CA2508344A1 (en) Switch/network adapter port coupling a reconfigurable processing element for microprocessors with interleaved memory controllers
TWI227836B (en) A computer system with multi-function input/output interface architecture
WO2023274032A1 (en) Storage access circuit, integrated chip, electronic device and storage access method
JP3384770B2 (en) High bandwidth, narrow I / O memory device with command stacking
US7114019B2 (en) System and method for data transmission
CN215341078U (en) Memory access circuit, integrated chip and electronic equipment
JPH10134576A (en) Semiconductor memory device
TW200805373A (en) A multi-port semiconductor device and method thereof
JP2005346715A (en) Multiport random access memory
US6311247B1 (en) System for bridging a system bus with multiple PCI buses
TW462128B (en) DRAM having a reduced chip size

Legal Events

Date Code Title Description
MK4A Expiration of patent term of an invention patent