TW575809B - Method and system for calculating dynamic burst length - Google Patents

Method and system for calculating dynamic burst length Download PDF

Info

Publication number
TW575809B
TW575809B TW91116102A TW91116102A TW575809B TW 575809 B TW575809 B TW 575809B TW 91116102 A TW91116102 A TW 91116102A TW 91116102 A TW91116102 A TW 91116102A TW 575809 B TW575809 B TW 575809B
Authority
TW
Taiwan
Prior art keywords
data
memory unit
length
module
main memory
Prior art date
Application number
TW91116102A
Other languages
Chinese (zh)
Inventor
Kun-Ying Tsai
Original Assignee
Rdc Semiconductor Co Ltd
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 Rdc Semiconductor Co Ltd filed Critical Rdc Semiconductor Co Ltd
Priority to TW91116102A priority Critical patent/TW575809B/en
Application granted granted Critical
Publication of TW575809B publication Critical patent/TW575809B/en

Links

Landscapes

  • Data Exchanges In Wide-Area Networks (AREA)

Description

575809 五、發明說明α) :【發明領域】 一種動態觸發長度計算方法以及系統,更詳而言之, -係有關於一種透過即時計算緩衝器内資料長度,用以決定 自該緩衝器讀出資料位元組之合理值之動態觸發長度計算 :方法以及系統。 【發明背景】 由於電子資訊相關科技的研發日新月異,許多功能強 大且價格合理的產品紛紛問世,就以電腦通訊設備為例, 不論是大型的超級電腦、伺服器主機乃至於個人電腦以及 g記型電腦等等,都早已成為人們資訊流通所不可或缺之 重要管道與工具。再加上現今網路通訊系統的軟硬體設備 與以往相較,可謂有相當大幅度的成長與改善,因此,許 多的大型企業或是公司行號都先後的將組織網路化,不論 是網際網路(Internet)、企業網路(Intranet)或是企 業間網路(Extranet),當然包括兩種以上甚至三種網路 通訊系統均擁有的亦不在少數。除了上述之企業或是、公司 行號的組織網路化之外,學校、家庭乃至於個人而言電腦 設備以及網路通訊系統,也早已成為我們生活中傳送或接 收各方面資訊所不可或缺的工具。 Φ 一旦要利用各種電腦設備並透過網路通訊系統來傳送 以及接收訊息資料,各個資料傳送接收節點(Node )例如 終端裝置或伺服器等之間必須有一個固定的資料傳輸協 定。以目前習知之乙太網路(E t h e r n e t )架構為例,其媒 體介質之存取方法,即為載波感知多重進出碰撞偵測575809 V. Description of the invention α): [Field of invention] A dynamic trigger length calculation method and system, more specifically,-it relates to a method for calculating the data length in a buffer through real-time calculation to determine the read from the buffer Dynamic trigger length calculation of reasonable values of data bytes: methods and systems. [Background of the Invention] Due to the rapid development of electronic information-related technologies, many powerful and affordable products have come out. Take computer communication equipment as an example, whether it is a large supercomputer, a server host, or even a personal computer and g-type. Computers, etc., have already become indispensable important channels and tools for people's information circulation. In addition, compared with the past, the hardware and software equipment of the current network communication system can be said to have considerable growth and improvement. Therefore, many large enterprises or companies have successively networked their organizations, whether it is The Internet, intranet, or extranet, of course, includes more than two or even three types of network communication systems. In addition to the networking of the above-mentioned enterprises or organizations, computer equipment and network communication systems in schools, homes and even individuals have long been indispensable for transmitting or receiving all aspects of information in our lives. Tool of. Φ Once various computer equipment is to be used to transmit and receive message data through the network communication system, there must be a fixed data transmission agreement between each data transmission and receiving node (Node) such as a terminal device or server. Taking the current known Ethernet network (E t h e r n e t) architecture as an example, the access method of the media is carrier-aware multiple in-out collision detection

16641.ptd 第6頁 575809 五、發明說明(2) (Carrier Sense Multiple Access/Co11ision16641.ptd Page 6 575809 V. Description of the Invention (2) (Carrier Sense Multiple Access / Co11ision

Detection; CSMA/CD)亦即 IEEE 802. 3 通訊協定。要言 之,當網路上每個節點要傳輸資料時必須先偵測網路纜線 上是否有其他節點正在使用網路(Carrier Sense)。若 偵測到纜線閒置,節點即可將資料往纜線上傳送,並且繼 續監聽纜線訊號;由於乙太網路是屬於一種廣播 (Broadcast )媒介,並沒有優先等級之觀念,因此隨時 可能有兩個以上之節點,將資料送往纜線(Mu 11 i p 1 e A c c e s s )。假若複數個不同之節點同時送出資料,資料碰 撞(Collision)於是發生,靠近碰撞發生所在地最近的 網路節點,將偵測此一碰撞並送出奎塞(Jamming)信 號,碰撞的複數個節點於是停止資料傳送(Co 1 1 i s i on D e t e c t i ο η )。碰撞之複數個節點,將等待一段時間再重 新傳輸,避免連續碰撞發生,每個節點所等待之時間將隨 機產生,同樣的訊框(Frame)最多可重傳16次,若仍然 無法完成則放棄,並送出錯誤訊息至上層之通訊協定..。 承前所述,當資料在網路中進行傳送時,係以封包之 型態進行傳送,一般而言,透過上述通訊協定之資料封包 大小為64至1518個位元組,其中目的位址(Destination Address)來源位址(Source Address)各有7個位元組、 長度欄位有2個位元組、資料(D a t a )攔位有4 6至1 5 0 0個 位元組以及訊框順序檢查(Frame Check Sequence)攔位 4個位元組。 當網路上各該節點欲接收自網路上傳送的資料封包Detection; CSMA / CD) is the IEEE 802.3 protocol. In other words, when each node on the network wants to transmit data, it must first detect if other nodes on the network cable are using the network (Carrier Sense). If the cable is detected to be idle, the node can send data to the cable and continue to monitor the cable signal. Since Ethernet is a broadcast medium, there is no concept of priority, so there may be Two or more nodes send data to the cable (Mu 11 ip 1 e A ccess). If a plurality of different nodes send data at the same time, a data collision occurs, and the nearest network node near the place where the collision occurs will detect this collision and send a Jamming signal, and the collided nodes will stop. Data transmission (Co 1 1 isi on Detecti ο η). Collision of multiple nodes will wait for a period of time and then retransmit to avoid continuous collisions. The time each node waits will be randomly generated. The same frame can be retransmitted up to 16 times. If it still cannot be completed, it will be abandoned. , And send an error message to the upper layer protocol .. According to the previous description, when data is transmitted on the network, it is transmitted in the form of a packet. Generally, the size of a data packet through the above communication protocol is 64 to 1518 bytes, of which the destination address (Destination Address) Source address (7 bytes each), Length field (2 bytes), Data (data) block (46 to 1500 bytes) and frame order The check (Frame Check Sequence) blocks 4 bytes. When each node on the network wants to receive data packets sent from the network

16641.ptd 第7頁 575809 五 、ό寺 Card 包資 進行 用 流 憶 —— --— 發明說明(3) 須再透 ,判斷 料自串列 處理使用 資源。因 正為別的 單元供中央 緩衝記憶單 —再將該資料 ϊI。承前所 ‘隱ί =會~過 :被填滿4 資料的遺^ 單疋繞皆有 疋内的# % 將不同/ 元,j封包的 偯^疋此資料 以常會發生 了:碰撞後往 封4個2〇0〇個 封包的大,丨、 節點而令f不 次 。亚不 貝料。士 必 之 排 後 過該節 該封包 之型態 。因為 此,資 系統單 處理單 元中, 自該緩 述,網 1 5 1 8 個 皆小於 便將資 且網路 可能。 資料長 點之網路卡 是否為該節 轉換成平行 主記憶單元 料接收進來 元所佔用, 元處理,而 當得到主記 衝記憶單元 路資料封包 ’但是為節 1 5 1 8個位元 料及時傳送 封包長短不 因此如何動 度,及網路 資料混合在一起, 傳輸的一個重要課 如同前述之資料碰 往就會造成異常的 位元組之資料,此 足2 0 0 〇個位元組, 具有一定之意義, 記憶單元匯流排在 (Network 點所有,若 之型態,用 之匯流排為 後,可能因 所以無法馬 是·先將該資 憶單元匯流 中傳送出去 的位元組於 省成本,因 組’若無法 到主記憶單 一,從6 4個 態算出目前 封包之間的 並將資料傳 題。另外, 撞之情形發 封包長度出 時主記憶單 ,則該筆資 亦即該筆資 整個節點的16641.ptd Page 7 575809 Fifth, 寺 Temple Card package capital use flow memory —— --- Description of the invention (3) It must be re-examined, it is expected that the materials should be processed and used in series. Because the central buffer memory list is being provided for other units — and then the data is ϊI. Cheng Qiansuo's Hidden = Will ~ Passed: Filled with 4 pieces of data ^ 疋 疋 疋 有 有 疋 #% will be different / Yuan, j packet 偯 ^ 疋 This data often happens: after collision 4 Each of the 2000 packets is large, and f is inferior. Asia is not expected. The taxi must be ranked after the type of the packet in this section. Because of this, in the system information processing unit, it should be eased from time to time, if the number of nets is less than 158, the net will be funded and the net is possible. If the data card with a longer data length is converted into a parallel master memory unit and received by the unit, the unit will occupy the unit and process it, and when the master memory unit data packet is obtained, it will be a unit of 15 1 8 bits. The packet length is transmitted in a timely manner without mixing the network data. An important lesson of transmission is the same as the aforementioned data, which will cause abnormal byte data, which is 2000 bytes. It has a certain meaning. The memory unit bus is located at the (Network point. If it is of the type, after using the bus, it may not be possible. First, the bytes transmitted in the memory unit bus are sent out. Save costs, because if the group cannot reach the master memory single, calculate the current packet between the 64 states and transfer the data. In addition, if the collision occurs and the packet length is sent out, the master memory list will be used. The amount of the entire node

Interface 是則將這些封 以提供該節點 系統中分享共 主記憶單元匯 上傳送至主記 料暫時儲存於 排的使用權之 寫入主記憶單 正常的狀況卞 此一般緩衝記 在緩衝 元,便 4立元組 位於緩 臨界點 送到主 網路上 生,而 現’假 元預設 料對於 料為一 系統中 記憶單 會造成 到 1 5 1 8 衝記憶 ’以免 記憶單 的資料 在發生 設出現 儲存該 接收端 無用的 是屬於 由於主The Interface writes these seals to provide the shared master-memory unit in the node system to the master record and temporarily stores the right to use temporarily stored in the bank. The master-memory list is normal. This general buffer is recorded in the buffer. The 4 tuples are located at the slow critical point and sent to the main network for survival, and the 'false element preset material is expected to cause a memory sheet in the system to cause 1 5 1 8 flush memory' to avoid the occurrence of the data in the memory sheet. The useless storage of this receiver is due to the main

•Ptd 第8頁 575809 五、發明說明(4) 分享共用之資源,因此當非必要時即不應佔用該主記憶單 元匯流排之資源,以免降低系統的資料處理效率。上述之 問題同樣會發生在其他需透過主記憶單元匯流排傳送資料 之系統運作上,例如在一習知的個人電腦系統中,輸出入 裝置、記憶單元以及中央處理單元間即需透過一匯流排來 作資料傳送與接收之工作,如何有效提升主記憶單元匯流 排的使用效率所涉及的問題不只一端。 目前習知的資料處理系統尚無針對前述之主記憶單元 匯流排使用時所遇到的資源浪費問題,提出一有效提升主 記憶單元匯流排使用效率之解決方案。 【發明目的及概述】 為解決上述習知技術之缺點,本發明之主要目的在於 提供一種動態觸發長度計算方法以及系統,用以提供一動 態即時提升資料處理系統内用以傳送資料之主記憶單元匯 流排資源運用效率之功能。 本發明之另一目的在於提供一種動態觸發長度計.算方 法以及系統,用以提供資料處理系統具有即時將缓衝記憶 單元内之有效資料傳送至主記憶單元中並具有動態即時防 止因為資料之資料位元組長度超過預設之主記憶單元容量 大小而造成系統運作效能降低之功能。 為達成以上所述之目的,本發明之動態觸發長度計算 系統包括有以下構件:一用以提供資料處理系統傳送接收 資料之資料傳送系統連接模組;一預先設定用以接收其他 資料傳送系統所傳送資料之緩衝記憶單元;一用以計算該• Ptd Page 8 575809 V. Description of the invention (4) Sharing shared resources, so when not necessary, the resources of the main memory unit bus should not be occupied, so as not to reduce the data processing efficiency of the system. The above problems also occur in the operation of other systems that need to transmit data through the main memory unit bus. For example, in a conventional personal computer system, a bus is required between the input and output devices, the memory unit, and the central processing unit. For data transmission and reception work, how to effectively improve the use efficiency of the main memory unit bus involves more than one issue. At present, the conventional data processing system has not proposed a solution for effectively improving the use efficiency of the main memory unit bus in response to the waste of resources encountered during the use of the main memory unit bus. [Objective and Summary of the Invention] In order to solve the shortcomings of the above-mentioned conventional technology, the main object of the present invention is to provide a dynamic trigger length calculation method and system for providing a dynamic real-time improvement of a main memory unit for transmitting data in a data processing system. The function of bus resource utilization efficiency. Another object of the present invention is to provide a dynamic trigger length calculation method and system for providing a data processing system with real-time transmission of valid data in a buffer memory unit to a main memory unit and dynamic prevention of data due to data. The function that the length of the data byte exceeds the preset capacity of the main memory unit and reduces the performance of the system. In order to achieve the above-mentioned object, the dynamic trigger length calculation system of the present invention includes the following components: a data transmission system connection module for providing a data processing system to transmit and receive data; a preset setting for receiving other data transmission systems Buffer memory unit for transmitting data; one for calculating the

16641.ptd 第9頁 575809 五、發明說明(5) Μ衝 一… 一 , 〜 ,人貝个1 效^料數j二模組,一用以計算目前等待傳送之資料位元 -組長度之 >、料長度计算模組;一用以判斷該缓衝記憶單元 内有效資料數計算模組所計算出該淫你#陰罝元内所蘄六 之 記憶單元内所暫存之右林咨刹_ & ^昨- ’欢貝枓數之緩衝記憶早TG内有 :一 田 1、/ 4丨枚〜 … 长反‘只丨丨π反圩异镇組;一用以判斷該缓衝記憶單元 有效資料數計算模組所計算出該緩衝記憶單元内所暫存 々有政 >' 料數疋否超過預设值,並於超過時向該主記憶單 元匯流排發出資料傳送使用權之主記憶單元匯流排請求模 組以及一用以判斷觸發資料位元組長度之觸發長度判斷模 組。 、 透過上述之動態觸發長声外筲 舍 •發長度計算之動作時,首m,於’際進行動態 f記憶單元時令一位於暫:資料位元組寫入-緩 一此a -, A 暫存态内之寫入指標位址加一;者 一貧料位兀、、且自該缓衝記憶單元讀出睥則a 一仿% |六田 内之讀出指標位址加一;豆:兀X “貝,卜,於暫存器 資料數計算模組判斷緩衝:己;軍::緩衝,單元内有效 過預設之主記憶單元匯t;::;之有,數是否超 憶單元匯流排請求模組判斷“料:二:貝)再令該主記 單元所帛設儲存該資料的^貝!!長度是否已超過主!憶 單元匯流排之請求直到兮次/,右疋則將不再發出主記憶 常發生在異常資料之出枓傳送結束為止,此種情形通 鞭太小所至。若否,則透之健存該資料的主記憶單 發出主匯流排請求模組匯流;J =元;;排請求模組 長度計算模組判斷傳送中=月纟 令該資料 端,若是則計算出目二访 > 貝料位70組是否為資料之尾 觸發長度判斷模组將=^ =料位兀組之長度;最後,令該 將綾衝匯流排請求模組内有效資料位元16641.ptd Page 9 575809 V. Description of the invention (5) Μ punch one ... One, ~, person shell 1 effect ^ number of materials j two modules, one used to calculate the number of data bits currently waiting to be transmitted-the length of the group >, material length calculation module; one for judging the number of valid data in the buffer memory unit, the calculation module calculates the obscenity you # 阴 罝 元 内 所 蕲 six of the memory unit temporarily stored in the right forest counselor Cha _ & ^ Yesterday-'The buffer memory of Huanbei's number of early TGs includes: Yitian 1, / 4 丨 ~ ~ ... long reflex' only 丨 π reversal different town groups; one is used to judge the buffer The effective data number calculation module of the memory unit calculates the temporary storage capacity of the buffer memory unit > 'Whether the number of materials exceeds the preset value, and sends the data transmission right to the main memory unit bus when it exceeds A main memory unit bus request module and a trigger length determination module for determining the length of a trigger data byte. 1. When the action of long-sounding outer sacrifice • hair length calculation is triggered by the above dynamics, the first m is used to perform a dynamic f memory unit on the occasion. One is temporarily: data byte writing-slow one a-, A The writing index address in the temporary storage state is increased by one; one is poor, and is read from the buffer memory unit; a is a copy of% | the reading index address in Liutian is increased by one; beans: Wu X, "Bei, Bu, judging the buffer data count module buffer: self; army :: buffer, the unit has been over the preset master memory unit t; :: ;; Yes, whether the number exceeds the memory unit The bus request module judges "material: two: shell) and then instructs the master unit to store the shell! ! Has the length exceeded the master! The request for the bus of the memory unit will not be issued until the number of times, and the main memory will no longer issue the main memory. It usually occurs at the end of the transmission of the abnormal data. This situation is too small. If not, then the master memory sheet that stored the data sends the main bus request module to converge; J = Yuan; the length of the request module calculation module judges that the transmission is in progress = the order of the data terminal is calculated, if it is calculated The second visit > Whether the 70 material level is the end of the data triggering the length judgment module will be = ^ = the length of the material level group; finally, it will make the effective data bit in the bus request module

575809575809

五、發明說明(6) 組長度 最小值 貝料長度以及預設觸發長度相互比較並复 作為觸發長度之值。 〃 中之 透過上述之動態觸發長度計算方法以及系 =:動態即時提升資料處理系統之内用以傳送資料之=$ 思單兀匯流排資源運用效率以及防止因為資料之次 〇 组長度超過預設之主記憶體容量大小而造成系作= 之功能。 中 【發明實施例】 睛參閱第1圖’其中顯示本發明之動態觸發長度曾 系統1應用於一網路通訊節點之一較佳實施例的基本架t 構。其中,該網路通訊節點於本實施例中係一電^裝置。 惟須特別說明者,係該電腦裝置復包括其他功能,以下實 施例僅就與本發明有關之部分加以說明。該動態觸發長S 計算生成系統包括以下構件:一中央處理單元1 〇、一 ^二 體匯流棑2 0、一網路通訊系統3 0、一網路通訊系統連接模 組4 0以及一主記憶單元5 〇。 、 該中央處理單元1 0,其係提供該動態觸發長度計算系 統1之構件以及模組間相互運作之用。 該主記憶單元匯流排2 〇 ’於本實施例中,其係提供該 動態觸發長度計算系統1中之各個記憶單元與該中央處理 單元1 0之間傳送資料之硬雜路線;實質上,該主記憶單元 匯流排2 0係一連接該電腦装置内部同系統部門(包括中央 處理單元1 0、磁碟控制卡纪憶體以及輸出入連接埠等(未 圖示))之分享式高速通道’且能使各該系統部門間相互V. Description of the invention (6) Minimum group length The shell material length and the preset trigger length are compared with each other and used as the trigger length value.之 Through the above-mentioned dynamic trigger length calculation method and system =: Dynamic and real-time improvement of the data processing system used to transmit data = $ Si Dan Wu bus resource utilization efficiency and prevent the second group length exceeds the preset because of the data The size of the main memory capacity causes the function =. [Inventive Embodiment] Refer to FIG. 1 ', which shows the basic structure of a preferred embodiment of a network communication node where the dynamic trigger length system 1 of the present invention is applied. The network communication node is an electrical device in this embodiment. It should be noted that the computer device includes other functions. The following embodiments only describe the parts related to the present invention. The dynamic trigger long S calculation and generation system includes the following components: a central processing unit 10, a two-body confluence 20, a network communication system 30, a network communication system connection module 40, and a main memory. Unit 5 0. The central processing unit 10 is used to provide mutual operation between the components of the dynamic trigger length calculation system 1 and the modules. The main memory unit bus 20 ′ in this embodiment provides a hard and complicated route for transmitting data between each memory unit in the dynamic trigger length calculation system 1 and the central processing unit 10; in essence, the The main memory unit bus 20 is a shared high-speed channel connected to the computer device and the system department (including the central processing unit 10, the disk control card memory, and the input and output ports (not shown)). And enable each system department to

—------ 、發明說明⑺ ')專送訊息。 網路通訊 連接的一群電 统30係—乙太 針對連接網路 系統係一習知 該網路通 度計算系統1, 及接收之用。 0為一網路卡 存取。要言之 則將這些封包 提供該網路節 訊糸統連接模 單元内有效資 一主記憶單元 45 〇 系統3 0,該網路通訊系統3 0係指通訊設施相 腦及相關設備。本實施例中,該網路通訊系 網路系統,其中,該乙太網路系統之標準為 所制定之IEEE 802.3標準。由於該乙太網路 之網路連接系統,故於此不另無贅述。 訊系統連接模組4 〇,其係提供該動態觸發長 透過該網路通訊系統3 0進行資料之傳送以 於本實施例中,該網路通訊系統連接模組4 0 ’用以提供該節點電腦裝置進行網路資料之 ’即係判斷該封包是否為該節點所有,若是 資料自串列之型態轉換成平行之型態,用以 點電腦裝置進行處理使用。其中,該網路通 組4 0包括有一缓衝記憶單元4 1、一緩衝記憶 料數計算模組4 2、一資料長度計算模組4 3、 匯流排請求模組4 4以及一觸發長度判斷、模組 該缓衝記憶單元4 1,其資料之存取係受控於該網路通 訊系統連接模組4 0,用以提供一特定之區域以保留等待欲 %過該網路通訊系統3 0傳送以及接收之資料之功能。 該缓衝區内有效資料數計算模組4 2,其係受控於該網 路通訊系統連接模組4 0,用以提供該動態觸發長度計算系 統1計异該緩衝記憶單元内所暫存之有效封包資料數之功 能0—------ 、 Invention note ⑺ ') Dedicated message. Network communication A group of 30 systems connected to the Ethernet—Ethernet is familiar with the connection to the network system, the network throughput calculation system 1, and reception. 0 is a network card access. In other words, these packets are provided to the network connection information system connection module. The effective data in the unit is a main memory unit 45 0 system 30, the network communication system 30 refers to the communication facilities and related equipment. In this embodiment, the network communication is a network system. The standard of the Ethernet system is the IEEE 802.3 standard. Due to the network connection system of the Ethernet, it will not be repeated here. Communication system connection module 40, which provides the dynamic trigger to transmit data through the network communication system 30. In this embodiment, the network communication system connection module 40 'is used to provide the node. The computer equipment's network data judgment is to determine whether the packet is owned by the node. If the data is converted from a serial type to a parallel type, it can be used by a computer device for processing and use. Among them, the network communication group 40 includes a buffer memory unit 41, a buffer memory material number calculation module 4 2, a data length calculation module 4 3, a bus request module 44, and a trigger length judgment. 2. The module buffer memory unit 41, whose data access is controlled by the network communication system connection module 40, is used to provide a specific area to reserve waiting for the network communication system 3 0The function of transmitting and receiving data. The effective data number calculation module 42 in the buffer area is controlled by the network communication system connection module 40 to provide the dynamic trigger length calculation system 1 to account for temporary storage in the buffer memory unit. The function of the number of valid packet data0

575809 五、發明說明(8) 該資料長度計算模組4 3,其係受控於該網路通訊系統 連接模組4 0,用以提供該動態觸發長度計算系統1計算目 前等待傳送之封包資料位元組長度之資料長度計算模組之 功能。 該主記憶單元匯流排請求模組4 4,其係受控於該網路 通訊系統連接模組4 0,用以提供該動態觸發長度計算系統 1判斷該緩衝記憶單元内有效資料數計算模組所計算出該 缓衝記憶單元内所暫存之有效封包資料數是否超過預設 值;此外該主記憶單元匯流排請求模組4 4復用以提供一判 斷該封包資料長度是否已超過主記憶單元5 〇所預設儲存該 封包的大小’若是則將不再發出主記憶單元匯流排之請求 直到該封包結束為止;若否,則向該主記憶單元匯流排發 出資料傳送使用權之功能。 該觸發長度判斷模組4 5,其係受控於該網路通訊系統 連接模組40,用以提供該動態觸發長度計算系統1判斷觸 發資料位元組長度之觸發長度判斷模組。 該主記憶單元5 0,其係用以提供該網路通訊節點儲存 資料並與該該中央處理單元1 〇相互連接以執行資料運算處 理之功能,該主記憶單元5 0於本實施例中係為一動態隨機 存取記憶單元(Dynamic Random Access Memory; DRAM )〇 透過前述之動態觸發長度計算系統丨,於實施動態觸 發長度計算時,其流程如下: 請參閱第2 ( A )圖,其係一顯示自該網路通訊系統3 〇575809 V. Description of the invention (8) The data length calculation module 43 is controlled by the network communication system connection module 40 to provide the dynamic trigger length calculation system 1 to calculate the packet data currently waiting to be transmitted. Function of byte length data length calculation module. The main memory unit bus request module 44 is controlled by the network communication system connection module 40 to provide the dynamic trigger length calculation system 1 to determine a valid data number calculation module in the buffer memory unit. It is calculated whether the number of valid packet data temporarily stored in the buffer memory unit exceeds a preset value; in addition, the main memory unit bus request module 44 reuses to provide a judgment whether the length of the packet data has exceeded the main memory The size of the packet stored in the unit 50 is preset, if it is, the request of the main memory unit bus will not be issued until the end of the packet; if not, the function of transmitting data usage rights to the main memory unit bus will be issued. The trigger length determination module 45 is controlled by the network communication system connection module 40 to provide a trigger length determination module for the dynamic trigger length calculation system 1 to determine the length of a trigger data byte. The main memory unit 50 is used to provide the network communication node to store data and interconnect with the central processing unit 10 to perform data operation processing functions. The main memory unit 50 is in this embodiment. It is a Dynamic Random Access Memory (DRAM). Through the dynamic trigger length calculation system described above, when implementing the dynamic trigger length calculation, the process is as follows: Please refer to Figure 2 (A), which shows A display from the network communication system 3 〇

16641.ptd 第13頁 575809 “五、發明說明(9) ~----- :所接收之封包資料於該緩衝記憶單元41之示意圖。首先, 田貝料封包6 〇、7 0先後由該網路通訊系統連接模组4 0透 過該,路通訊系統30接收後,將該資料位元組寫入該缓衝 ,憶單元41之内,並令一位於暫存器内之寫入指標於 :每一資料位元組寫入時,將其寫入指標i 〇丨之位址加一; 另 方面,g 為料位元組自该緩衝記憶單元4 1讀出時則 令一位於暫存器内之讀出指標102位址加一。 其次,令該緩衝記憶單元内有效資料數計算模組42判 斷該緩衝記憶單元4 1内之有效封包資料數是否超過預設之 囑單元匯流排凊求門楹’若是再令該主記憶單元貭流 排凊求模組4 4利斷該資料長度是否已超過主記憶單元所預 没儲存該封包資料的大小,若是則將不再發出主記憶單元 匯/;,L排之請求直到該封包資料結束為止,此種情形通常發 生在異常封包資料之出現或所設之儲存該封包資料的主記 憶單元太小所至。若否,則透過該主記憶單元匯流排請求 模組44發出主記憶單元匯流排20使用請求。其中該有效封 包資料數係指該寫入指標1 0 1以及讀出指標1 0 2間之資料位 元組數,而主記憶體匯流排清求門禮’於本實施例中則預 設為1 8個資料位元組數,且設該緩衝區内有效資料位元组 鑷度 為2 0個資料位元組’故令該5己憶體匯流排請求模組4 4 發出記憶體匯流排2 0使用請求。 再者,令該資料長度計算模組4 3判斷傳送中之資料位 凡t是否為資料封包之尾端,若是則將於該暫存器内寫入 一資料尾蠕指標1 0 3,用以讓該網路通訊系統連接模組4 016641.ptd Page 13 575809 "Fifth, the description of the invention (9) ~ -----: The schematic diagram of the received packet data in the buffer memory unit 41. First, Tianbei material packets 6 0, 7 0 by the network The communication module 40 of the communication system passes through it, and after receiving it, the communication system 30 writes the data byte into the buffer, memorizing the unit 41, and makes a writing indicator located in the temporary register in: When each data byte is written, the address of its writing index i 〇 丨 is increased by one; on the other hand, g is the material byte when it is read from the buffer memory unit 41 and one is located in the temporary register. The reading index in the address 102 is incremented by one. Next, the effective data number calculation module 42 in the buffer memory unit 42 is determined to determine whether the number of valid packet data in the buffer memory unit 41 exceeds the preset order unit bus request. If the door is' re-ordered the main memory unit's stream, ask for module 4 4 to determine whether the length of the data has exceeded the size of the packet data that the main memory unit has not stored in advance. If so, the main memory unit sink will no longer be issued. / ;, L row request until the end of the packet information, this situation is common Occurs when the abnormal packet data appears or the main memory unit set to store the packet data is too small. If not, the main memory unit bus request module 44 is used to issue a request for using the main memory unit bus 20 through the main memory unit bus request module 44. The number of valid packet data refers to the number of data bytes between the write index 1 0 and the read index 10 2, and the main memory bus clears the request for salute '. In this embodiment, it is preset to 1 The number of 8 data bytes, and the effective data byte tweezers in the buffer is set to 20 data bytes. Therefore, the 5 memory bus request module 4 4 issues a memory bus 2 0 Use request. In addition, the data length calculation module 4 3 is used to determine whether the data bit t in the transmission is the end of the data packet, and if it is, a data tail creep indicator will be written in the register 1 0 3 for connecting the network communication system to the module 4 0

16641.ptd 第14頁 575809 五、發明說明(ίο) 碟定该資料封包之結束點以音 過該資料尾端指標1 〇 3以及读、入該緩衝纪憶單元4 1内,透 目前該資料封包資料位元組貝出指標1 0 2之位址即可計算出 該資料封包60之資料位元細=長度,於本實施例中,則為 送中之資料封包β 〇,其尾端次’、,味參閱第2 ( Β )圖,若傳 記憶單元4 1之緩衝區内,之^貝料位70組尚未寫入該緩衝 於無限遠處。 :該^料尾端指標1 0 3位址設 最後,令該觸發長产刹 + 效資料位元組長度、封包次杈組44將緩衝記憶單元内有 比較並取其中之最小值,二二長度以及預設觸發長度相互 中,設該緩衝記憶單元内有效;::度f:庚於本實施例 一 · 一 n啕政貝枓位疋組長度為2 0個資料 位元組,該封包資料長度則因該資料封包6 〇,其尾端之資 料位元組尚未寫入該緩衝記憶單元4 1内,故該資料尾端指 標1 0 3位址設於無限遠處其值則為無限大,該觸發長度預 設值則為3 0個資料位元矣 元組數為其觸發長度之 %从《·丨* 〇 rwdu ‘ ,透過該觸發長度判斷模組4 4之 判斷,則應取最小值2 0個資料位 值。另-方面,當該資料討包60其之資料位元組尚已 寫入該緩衝記憶單元㈣,則^%%^尾端指標103以 -出目則該負料封包資料位 元組之長f,後再將該資料封包二〇 :::::數與該緩 衝記憶單元内有效資料位元組長二 2 χ長度相互 為觸發長度之值,如此即可避 及讀出指標1 0 2之位址即矸計算 元組之長度,後再將該資3 衝記憶單元内有效資料位元 比較並取其中之最小值,Ί「… 語分仿从* I + - 免位於該資料封包60後且在記憶!相連之資料 封包70被緊接在該資料封包6〇·敏由^ 5己憶早疋匯流排寫16641.ptd Page 14 575809 V. Explanation of the invention (ίο) Set the end point of the data packet to the end of the data by the index 1 0 3 and read and enter the buffer memory unit 41 to reveal the current data The data bit of the packet data byte index 1 102 can be used to calculate the data bit detail of the data packet 60 = length. In this embodiment, it is the data packet β 0 which is being sent. ',, Refer to Figure 2 (B). If the buffer unit of memory unit 41 is transmitted, the 70 groups of material levels have not been written into the buffer at infinity. : The end of the index of the material 10 is set to the last address, so that the triggering of the long brake + effective data byte length and the packet sub-bracket group 44 will be compared in the buffer memory unit and the minimum value will be taken. The length and the preset trigger length are mutually valid, and it is assumed that the buffer memory unit is valid; :: degree f: geng is in the first embodiment of the present invention. The length of a packet is 20 data bytes, and the packet The length of the data is because the data packet 6 0, and the data bytes at the end have not been written into the buffer memory unit 41, so the end of the data index 103 address is set at infinity and its value is infinite. Large, the default value of the trigger length is 30 data bits. The number of tuples is the percentage of its trigger length. From "· 丨 * 〇rwdu ', through the judgment of the trigger length judgment module 44, you should take The minimum value is 20 data bits. On the other hand, when the data byte of the data packet 60 has been written into the buffer memory unit ㈣, the ^ %% ^ trailing index 103 is-the length of the negative packet data byte f, and then the data packet 20 ::::: number and the length of the effective data byte in the buffer memory unit 2 2 χ length is the value of the trigger length, so you can avoid reading the index 1 0 2 The address is to calculate the length of the tuple, and then compare the valid data bits in the three memory units and take the minimum value of them. "... The language points are copied from * I +-free from the data packet 60. And in the memory! The connected data packet 70 is immediately followed by the data packet 60. 敏 由 ^ 5 其 忆 早 疋

16641.ptd 575809 .五、發明說明(11) •-义該節點電腦裝置之主記憶單元内。 請參閱第3圖,其中顯示本發明之動態觸發長度計算 -方法,於執行動態觸發長度計算之流程步驟。 於步驟S 2 0 1中,預設該觸發長度值為3 0個資料位元 .組。 於步驟S 2 0 2中,判斷是否有封包資料位元組寫入該緩 衝記憶單元4 1内,若是則進行步驟S 2 0 4 ;若否則進行步驟 S2 0 6 〇 於步驟S 2 0 3中,判斷是否有封包資料位元組自該緩衝 ϋ憶單元4 1讀出,若是則進行步驟S 2 0 5 ;若否則進行步驟 S2 0 6 〇 於步驟S204中,令該位於暫存器内之寫入指標101之 位址加一,接著進行步驟S206。 於步驟S205中,令一位於暫存器内之讀出指標102位 址加一,接著進行步驟S206。 於步驟S206中,令該緩衝記憶單元内有效封包數計算 模組4 2計算該緩衝記憶單元4 1内之有效封包資料數,接著 進行步驟S 2 0 7。 於步驟S 2 0 7中,令該缓衝記憶單元内有效封包數計算 % 組4 2判斷該缓衝記憶單元4 1内之有效封包資料數是否超 過預設之主記憶單元匯流排請求門檻,若是則進行步驟 S 2 0 8 ;若否則回到步驟S 2 0 6。 於步驟S 2 0 8中,令該主記憶單元匯流排請求模組4 4判 斷該封包資料長度是否已小於主記憶單元所預設儲存該封16641.ptd 575809. V. Description of the Invention (11) •-It means the main memory unit of the node computer device. Please refer to FIG. 3, which shows the dynamic trigger length calculation method of the present invention, in the process steps of performing dynamic trigger length calculation. In step S 201, the trigger length value is preset to 30 data bit groups. In step S202, it is determined whether a packet data byte is written into the buffer memory unit 41. If yes, go to step S204; if not, go to step S202. In step S203 To determine whether any packet data byte is read from the buffer memory unit 41, and if yes, go to step S205; otherwise, go to step S206. In step S204, make it located in the temporary register. The address of the write index 101 is incremented by one, and then step S206 is performed. In step S205, one of the read index 102 located in the register is incremented by one, and then step S206 is performed. In step S206, the effective packet number in the buffer memory unit is calculated. Module 42 calculates the effective packet data number in the buffer memory unit 41, and then proceeds to step S207. In step S207, the effective packet number in the buffer memory unit is calculated as% group 4 2 to determine whether the number of valid packet data in the buffer memory unit 41 exceeds the preset main memory unit bus request threshold, If yes, go to step S 2 0 8; otherwise, go back to step S 2 0 6. In step S208, the main memory unit bus request module 44 is caused to determine whether the length of the packet data is shorter than the preset storage of the main memory unit.

16641.ptd 第16頁 575809 五、發明說明(12) 包資料的大小,若是則進行步驟S2 0 9 ;若否則進行步驟 S21 6。 於步驟S 2 0 9中,令該主§己憶單元匯流排請求模組4 4發 出主記憶單元匯流排2 0使用請求’接著進行步驟S 2 1 6。 於步驟S 2 1 0中,令該資料長度計算模組4 3判斷傳送中 之資料位元組是否為資料封包之尾端,若是則進行步驟 S 2 1 1 ;若否則進行步驟S 2 1 5。 於步驟S211中,於該暫存器内寫入一資料尾端指標 1 〇 3,用以讓該網路通訊系統連接模組4 0確定該資料封包 之結束點以寫入該缓衝記憶單元4 1之緩衝區内,接著進行 步驟S 2 1 2。 於步驟S212中,透過該資料尾端指標ι〇3以及讀出指 標102之位址即可計算出目前該資料封包資料位元组之長 度,接著進行步驟S 2 1 3。 故加上一時脈 於步驟S21 3中,為區隔不同資料封包 延遲,接著進行步驟S 2 1 4。 斷模組44將該缓衝記 包資料長度以及預設 4乍為觸發長度之 於步驟S214中,令該觸發長度列 憶單元4 1内有效資料位元組長度、封 觸發長度相互比較並取其中之最小值 值。 於步驟S 2 1 5中 遠處。 將該f料尾端指襟1〇3位址設於無 限 於步驟S216中,令該主記憶單元 斷該封包資料是否傳送完成若是列红太/敬排請求模組4 4判 〜束流程步驟,·若則跳16641.ptd Page 16 575809 V. Description of the invention (12) The size of the package data. If it is, go to step S2 0 9; otherwise, go to step S21 6. In step S209, the main §memory unit bus request module 44 is issued to use the main memory unit bus 20 request request ', and then step S216 is performed. In step S 2 10, the data length calculation module 43 is caused to determine whether the data byte in transmission is the tail of the data packet. If it is, proceed to step S 2 1 1; otherwise, proceed to step S 2 1 5 . In step S211, a data tail index 1 03 is written in the register to allow the network communication system connection module 40 to determine the end point of the data packet to write to the buffer memory unit. In the buffer area of 41, step S 2 1 2 is performed. In step S212, the current byte length of the data packet data byte can be calculated through the data trailing indicator ι03 and the address of the reading indicator 102, and then proceed to step S 2 1 3. Therefore, a clock is added in step S21 3 to separate different data packet delays, and then step S 2 1 4 is performed. The breaking module 44 compares the buffer packet data length and the preset length as the trigger length in step S214, and causes the trigger length to memorize the valid data byte length and the packet trigger length in the unit 41 and compare them with each other and obtain The minimum value. In step S 2 1 5 at a distance. Set the address of the tail end finger of the f material 10 to infinite in step S216, so that the main memory unit judges whether the packet data has been transmitted. If it is a red list / request request module 4 4 judge ~ bundle process steps , · If you jump

第17頁 16641.ptd 575809 五、發明說明(13) ψ \至步驟S 2 0 1重新開始流程步驟。 承前所述,透過本發明之動態觸發長度計算方法以及 系統,當任一資料封包透過該網路通訊系統3 0而被該網路 通訊系統連接模組4 0所接收,且該資料封包之資料位元組 數(假設為2 0 0 0個資料位元組)大於預設之緩衝記憶單元 4 1容量大小(假設為1 5 1 8個資料位元組)時,因該超過緩 衝記憶單元4 1容量大小的資料封包已屬於無效的封包資料 位元組,此時即將該寫入指標1 0 1暫停,一直等到該讀出 指標102於暫存器内之位址值與該寫入指標101相等,由於 #該寫入指標1 0 1與該讀出指標1 0 2位址值相等時,該動態 觸發長度計算系統即不會要求該主記憶單元匯流排2 0同意 自該網路通訊系統連接模組4 0將資料傳送出來,因此不會 再令該主記憶單元匯流排2 0傳送不具意義的封包資料,進 而造成該節點電腦裝置之系統之異常運作。 以上所述僅為本發明之動態觸發長度計算方法以及系 統之較佳實施例而已,並非用以限定本發明之實質技術内 容之範圍。本發明之動態觸發長度計算方法以及系統其實 質技術内容係廣義地定義於下述之申請專利範圍中,任何 他人所完成之技術實體或方法,若是與下述之申請專利範 β所定義者完全相同,或是為同一等效之變更,均將被視 為涵蓋於此專利範圍之中。Page 17 16641.ptd 575809 V. Description of the invention (13) ψ \ Go to step S 2 0 1 to restart the process steps. According to the foregoing, through the dynamic trigger length calculation method and system of the present invention, when any data packet is received by the network communication system connection module 40 through the network communication system 30, and the data of the data packet is When the number of bytes (assuming 20,000 data bytes) is larger than the preset buffer memory unit 4 1 capacity (assuming 1 5 1 8 data bytes), the buffer memory unit should be exceeded A data packet with a capacity of 1 already belongs to an invalid packet data byte. At this time, the write index 1 0 1 will be suspended until the address value of the read index 102 in the register and the write index 101 are suspended. Equal, because #the write index 1 0 1 and the read index 1 102 address values are equal, the dynamic trigger length calculation system will not require the main memory unit bus 2 to agree to the network communication system The connection module 40 sends out the data, so the main memory unit bus 20 will not be allowed to send meaningless packet data, which will cause the node computer device's system to operate abnormally. The above is only a preferred embodiment of the dynamic trigger length calculation method and system of the present invention, and is not intended to limit the scope of the substantial technical content of the present invention. The dynamic trigger length calculation method and system of the present invention are broadly defined in the scope of the following patent applications. Any technical entity or method completed by others is completely the same as defined in the following patent scope β. Changes that are the same or equivalent are considered to be covered by this patent.

16641.ptd 第18頁 575809 圖式簡單說明 【簡單圖式說明】 為讓本發明之上述和其它目的、特徵以及優點能更明 顯易懂,將與較佳實施例,並配合所附圖式,詳細說明本 發明之實施例,所附圖式之内容簡述如下: 第1圖為一系統方塊圖,其中顯示本發明之動態觸發 長度計算系統之一較佳實施例的基本架構; 第2圖為一示意圖,其中顯示自網路通訊系統所接收 之封包資料於該緩衝記憶單元之示意圖;以及 第3圖為一流程圖,其中顯示本發明之動態觸發長度 計算方法,於執行動態觸發長度計算之流程步驟。 【圖式標號】 1 動 態 觸 發 長 度 計 算 系 統 10 中 央 處 理 單 元 2 0 主 記 憶 ασ 早 元 匯 流 排 30 網 路 通 訊 系 統 40 網 路 通 訊 系 統 連 接 模 組 41 緩 衝 記 憶 單 元 42 緩 衝 記 憶 早 元 内 有 效 資料數計算模 組 43 資 料 長 度 計 算 模 組 44 主 記 憶 體 匯 流 排 請 求 模組 45 觸 發 長 度 判 斷 模 組 50 主 記 憶 單 元 60 資 料 封 包 6(Γ 資 料 封 包 70 資 料 封 包 10 1 寫 入 指 標 102 讀 出 指 標 103 資 料 尾 端 指 標16641.ptd Page 18 575809 Brief description of the drawings [Simplified description of the drawings] In order to make the above and other objects, features, and advantages of the present invention more comprehensible, it will be combined with the preferred embodiments and the accompanying drawings, The embodiments of the present invention are described in detail. The contents of the drawings are briefly described as follows: FIG. 1 is a system block diagram showing the basic structure of a preferred embodiment of the dynamic trigger length calculation system of the present invention; FIG. 2 It is a schematic diagram showing the packet data received from the network communication system in the buffer memory unit; and FIG. 3 is a flowchart showing the dynamic trigger length calculation method of the present invention in performing the dynamic trigger length calculation Process steps. [Schematic labels] 1 Dynamic trigger length calculation system 10 Central processing unit 2 0 Main memory ασ Early element bus 30 Network communication system 40 Network communication system connection module 41 Buffer memory unit 42 Buffer memory effective number of data in the early element Calculation module 43 Data length calculation module 44 Main memory bus request module 45 Trigger length judgment module 50 Main memory unit 60 Data packet 6 (Γ Data packet 70 Data packet 10 1 Write index 102 Read index 103 Data Trailing index

16641.ptd 第19頁16641.ptd Page 19

Claims (1)

575809 六、申請專利範圍 1. 一種動態觸發長度計算方法,係透過一動態觸發長度 計算系統,用以提供一資料處理系統,其中之主記憶 ' 單元匯流排之使用效率提升並維護該資料處理系統運 作不受異常資料之影響,其中,該動態觸發長度計算 . 系統包括以下構件··一用以提供該動態觸發長度計算 系統之構件以及模組間相互運作之中央處理單元、一 _ 用以提供該動態觸發長度計算系統中之各個記憶體單 元與該中央處理單元之間傳送資料之主記憶單元匯流 排、一用以提供該動態觸發長度計算系統進行資料之 φ 傳送以及接收之資料傳送系統連接模組以及一用以提 供該貢料處理糸統儲存貢料以供該中央處理早兀處理 運算之主記憶單元;於實施動態觸發長度計算時其流 程步驟如下: (1 ) 當一資料位元組寫入一緩衝記憶單元時令一 位於暫存器内之寫入指標位址加一;當一資料位元組 自該緩衝記憶單元讀出時則令一位於暫存器内之讀出 指標位址加一; (2 ) 令一緩衝記憶單元内有效封包數計算模組判 斷緩衝記憶單元内之有效資料數是否超過預設之主記 Φ 憶單元匯流排請求門檻,若是則令一主記憶單元匯流 排請求模組判斷該資料長度是否已超過主記憶單元所 預設儲存該封包資料的大小,若是則將不再發出主記 憶早兀匯流排之請求直到該封包貢料結束為止,若否 ,則透過該主記憶單元匯流排請求模組發出主記憶單575809 6. Scope of patent application 1. A dynamic trigger length calculation method is to provide a data processing system through a dynamic trigger length calculation system, in which the use of the main memory 'unit bus is improved and the data processing system is maintained. The operation is not affected by abnormal data. Among them, the dynamic trigger length calculation. The system includes the following components: a central processing unit for providing the components of the dynamic trigger length calculation system and interoperation between modules, a _ for providing A main memory unit bus for transmitting data between each memory unit in the dynamic trigger length calculation system and the central processing unit, and a data transmission system connection for providing the dynamic trigger length calculation system for φ transmission and reception of data A module and a main memory unit for providing the tributary processing system to store tributary materials for the central processing early processing operation; the flow steps when implementing dynamic trigger length calculation are as follows: (1) when a data bit Group writes to a buffer memory unit The index address is increased by one; when a data byte is read from the buffer memory unit, a read index address located in the temporary register is increased by one; (2) the number of effective packets in a buffer memory unit is calculated The module judges whether the number of valid data in the buffer memory unit exceeds the preset master Φ memory unit bus request threshold, and if so, causes a master memory unit bus request module to determine whether the length of the data has exceeded the value predicted by the main memory unit Set the size of the packet data. If it is, the request of the main memory early bus will not be issued until the end of the packet. If not, the main memory unit will send the main memory list through the main memory unit bus request module. 16641.ptd 第20頁 575809 六、申請專利範圍 元匯流排使用請求, (3 ) 令該資料長度計算模組判斷傳送中之資料位 元組是否為資料之尾端,若是則計算出目前該資料位 元組之長度;以及 (4 ) 最後,令該觸發長度判斷模組將緩衝記憶單 元内有效資料位元組長度、資料長度以及預設觸發長 度相互比較並取其中之最小值,作為觸發長度之值。 2.如申請專利範圍第1項之方法,其中,該系統之資料傳 送系統連接模組包括以下模組: 一緩衝記憶單元,其資料之存取係受控於該資料 傳送系統連接模組,用以提供一特定之區域以保留等 待傳送以及接收之資料之功能; 一緩衝記憶單元内有效資料數計算模組,其係受 控於該資料傳送系統連接模組,用以提供該動態觸發 長度計算系統計算該記憶單元之緩衝記憶單元内所暫 存之有效貧料數之功能, 一資料長度計算模組,其係受控於該資料傳送系 統連接模組,用以提供該動態觸發長度計算系統計算 目前等待傳送之資料位元組長度之資料長度計算模組 、 之功能; 一主記憶單元匯流排請求模組,其係受控於該資 料傳送系統連接模組,用以提供該動態觸發長度計算 系統判斷該緩衝區内有效資料數計算模組所計算出該 緩衝記憶單元内所暫存之有效資料數是否超過預設值16641.ptd Page 20 575809 VI. Request for the use of the patent scope yuan bus (3) Make the data length calculation module judge whether the data byte in the transmission is the end of the data, and if so, calculate the current data The length of the byte; and (4) finally, make the trigger length judgment module compare the valid data byte length, the data length and the preset trigger length in the buffer memory unit with each other and take the minimum value as the trigger length Value. 2. The method according to item 1 of the scope of patent application, wherein the data transmission system connection module of the system includes the following modules: a buffer memory unit whose data access is controlled by the data transmission system connection module, It is used to provide a specific area to retain the data waiting to be transmitted and received. A valid data number calculation module in the buffer memory unit is controlled by the data transmission system connection module to provide the dynamic trigger length. The function of the calculation system to calculate the number of effective lean materials temporarily stored in the buffer memory unit of the memory unit. A data length calculation module is controlled by the data transmission system connection module to provide the dynamic trigger length calculation. The function of the data length calculation module that calculates the length of the data bytes currently waiting to be transmitted; a main memory unit bus request module, which is controlled by the data transmission system connection module to provide the dynamic trigger The length calculation system judges the validity of the temporary storage in the buffer memory unit calculated by the effective data number calculation module in the buffer. Whether the number of feed exceeds a preset value 16641.ptd 第21頁 575809 <六、申請專利範圍 ♦ • ,此外復可用以提供一判斷該資料長度是否已超過主 記憶體所預設儲存該資料的大小,若是則將不再發出 主記憶單元匯流排之請求直到該資料結束為止;若否 ,則向該主記憶單元匯流排發出資料傳送使用權之功 能;以及 一觸發長度判斷模組,其係受控於該資料傳送系 統連接模組,用以提供該動態觸發長度計算系統判斷 觸發資料位元組長度之觸發長度判斷模組。 3. 如申請專利範圍第1項之方法,其中,該步驟(3 )傳 • 送中之資料位元組若非為資料封包之尾端,則將該資 料尾端指標位址設於無限遠處。 4. 如申請專利範圍第1項之方法,其中,該步驟(3 )於 計算出目前該封包資料位元組之長度之後,復包括一 流程步驟: (3-1 )於資料封包之尾端加上一時脈延遲,用 以區隔不同資料封包。 5. 如申請專利範圍第1項之方法,其中,該系統之資料傳 送系統連接模組可為一網路卡(N e t w 〇 r k I n t e r f a c e Card ) ° #.如申請專利範圍第5項之方法,其中,該資料傳送系統 連接模組係與一網路通訊系統相互連接。 7. 如申請專利範圍第6項之方法,其中,該網路通訊系統 可為一區域網路系統(Local Area Network ; LAN)。 8. 如申請專利範圍第7項之方法,其中,該區域網路系統16641.ptd Page 21 575809 < Sixth, the scope of patent application ♦ • In addition, it can be used to provide a judgement whether the length of the data has exceeded the size of the data stored in the main memory preset, if it is, the main memory will no longer be issued The request of the unit bus is until the end of the data; if not, the function of transmitting data use rights is issued to the main memory unit bus; and a trigger length judgment module is controlled by the data transmission system connection module To provide a trigger length determination module for the dynamic trigger length calculation system to determine the length of a trigger data byte. 3. For the method of applying for the first item of the patent scope, wherein in step (3), if the data byte being transmitted is not the end of the data packet, set the address of the tail index of the data at infinity . 4. The method according to item 1 of the scope of patent application, wherein step (3) after calculating the length of the current data byte of the packet, further includes a process step: (3-1) at the end of the data packet A clock delay is added to separate different data packets. 5. If the method of the scope of patent application is applied for, the data transmission system connection module of the system may be a network card (Netw rk I nterface Card) ° # If the method of scope of the patent application is applied for, Among them, the data transmission system connection module is interconnected with a network communication system. 7. If the method of claim 6 is applied for, the network communication system may be a local area network (LAN). 8. The method of claim 7 in the scope of patent application, wherein the local area network system 16641.ptd 第22頁 575809 六、申請專利範圍 可為一乙太網路(Ethernet)系統架構以及高速乙太 網路(F a s t E t h e r n e t )系統架構其中一種。 9 . 一種動態觸發長度計算系統,係透過一動態觸發長度 計算方法,用以提供一資料處理系統,其中之主記憶 單元匯流排之使用效率提升並維護該資料處理系統運 作不受異常資料封包之影響其中,該動態觸發長度計 算系統包括以下構件: 一中央處理單元,其係用以提供該動態觸發長度 計算系統之構件以及模組間相互運作之功能; 一主記憶單元匯流排,其係提供該動態觸發長度 計算系統中之各個記憶體單元與該中央處理單元之間 傳送資料之硬體路線; 一資料傳送系統連接模組,用以提供該動態觸發 長度計算系統進行資料之傳送以及接收之功能;以及 一主記憶單元,其係用以提供該資料處理系統儲 存資料並與該該中央處理單元相互連接以執行資料運 算處理之功能。 1 〇.如申請專利範圍第9項之系統,其中,該資料傳送系統 連接模組包括以下模組: 一緩衝記憶單元,其資料之存取係受控於該資料 傳送系統連接模組,用以提供一特定之區域以保留等 待欲透過該網路通訊系統傳送以及接收之資料之功能 , 一緩衝記憶單元内有效資料數計算模組,其係受16641.ptd Page 22 575809 6. Scope of patent application It can be one of an Ethernet system architecture and a high-speed Ethernet (F a s t E t h e r n e t) system architecture. 9. A dynamic trigger length calculation system is provided by a dynamic trigger length calculation method to provide a data processing system in which the use of the main memory unit bus improves the efficiency and maintains the operation of the data processing system from abnormal data packets. Among them, the dynamic trigger length calculation system includes the following components: a central processing unit, which is used to provide the components of the dynamic trigger length calculation system and the function of interoperation between modules; a main memory unit bus, which provides A hardware route for transmitting data between each memory unit in the dynamic trigger length calculation system and the central processing unit; a data transmission system connection module for providing the dynamic trigger length calculation system for transmitting and receiving data Functions; and a main memory unit, which is used to provide the data processing system to store data and interconnect with the central processing unit to perform data operation processing functions. 1 〇. The system of item 9 in the scope of patent application, wherein the data transmission system connection module includes the following modules: A buffer memory unit whose data access is controlled by the data transmission system connection module. In order to provide a specific area to retain the function of waiting for data to be transmitted and received through the network communication system, a valid data number calculation module in a buffer memory unit, which is subject to 16641.ptd 第23頁 575809 六、申請專利範圍 V ' 控於該資料傳送系統連接模組,用以提供該動態觸發 長度計算系統計算該記憶單元之緩衝記憶單元内所暫 存之有效資料數之功能; 一資料長度計算模組,其係受控於該資料傳送系 統連接模組,用以提供該動態觸發長度計算系統計算 目前等待傳送之資料位元組長度之資料長度計算模組 之功能; 一主記憶單元匯流排請求模組,其係受控於該資 料傳送系統連接模組,用以提供該動態觸發長度計算 0^ 糸統判斷該緩衝記憶早元内有效貢料數計鼻权組所計 算出該緩衝記憶單元内所暫存之有效資料數是否超過 預設值,此外復可用以提供一判斷該資料長度是否已 超過主記憶體所預設儲存該資料的大小,若是則將不 再發出主記憶單元匯流排之請求直到該資料結束為止 ;若否,則向該主記憶單元匯流排發出資料傳送使用 權之功能;以及 一觸發長度判斷模組,其係受控於該資料傳送系 統連接模組,用以提供該動態觸發長度計算系統判斷 觸發資料位元組長度之觸發長度判斷模組。 馨丨1.如申請專利範圍第9項之系統,其中,該資料傳送系統 連接模組可為一網路卡(Network Interface Card ) ° 1 2.如申請專利範圍第1 1項之系統,其中,該資料傳送系 統連接模組係與一網路通訊系統相互連接。 1 3.如申請專利範圍第1 1項之系統,其中,該網路通訊系16641.ptd Page 23 575809 6. The scope of patent application V 'is controlled by the data transmission system connection module to provide the dynamic trigger length calculation system to calculate the number of valid data temporarily stored in the buffer memory unit of the memory unit. Functions: A data length calculation module, which is controlled by the data transmission system connection module, to provide the function of the data length calculation module that the dynamic trigger length calculation system calculates the length of the data bytes currently waiting to be transmitted; A main memory unit bus request module, which is controlled by the data transmission system connection module, and is used to provide the dynamic trigger length calculation. The system judges the effective memory count in the buffer memory and calculates the nasal weight group. It is calculated whether the number of valid data temporarily stored in the buffer memory unit exceeds a preset value, and in addition, it can be used to provide a judgment as to whether the length of the data has exceeded the size of the data stored in the main memory by default. Send a request for the main memory unit bus until the data ends; if not, send a data transmission to the main memory unit bus Function use rights; and a trigger length determining module, which is controlled by the data-based transmission system connecting module for providing the computing system dynamically determines the length of the trigger the trigger of the trigger block length data bit length determining module. Xin 丨 1. If the system of item 9 of the patent application is applied, the data transmission system connection module may be a network interface card (° C) ° 1 2. If the system of item 11 of the patent application is applied, where The data transmission system connection module is interconnected with a network communication system. 1 3. The system according to item 11 of the scope of patent application, wherein the network communication is 16641.ptd 第24頁 57580916641.ptd Page 24 575809 16641.ptd 第25頁16641.ptd Page 25
TW91116102A 2002-07-19 2002-07-19 Method and system for calculating dynamic burst length TW575809B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW91116102A TW575809B (en) 2002-07-19 2002-07-19 Method and system for calculating dynamic burst length

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW91116102A TW575809B (en) 2002-07-19 2002-07-19 Method and system for calculating dynamic burst length

Publications (1)

Publication Number Publication Date
TW575809B true TW575809B (en) 2004-02-11

Family

ID=32734171

Family Applications (1)

Application Number Title Priority Date Filing Date
TW91116102A TW575809B (en) 2002-07-19 2002-07-19 Method and system for calculating dynamic burst length

Country Status (1)

Country Link
TW (1) TW575809B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI423032B (en) * 2009-04-30 2014-01-11 Ralink Technology Corp Method for enhancing data transmission efficiency
WO2015120825A1 (en) * 2014-02-17 2015-08-20 Mediatek Inc. Data access methods and data access devices utilizing the same

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI423032B (en) * 2009-04-30 2014-01-11 Ralink Technology Corp Method for enhancing data transmission efficiency
US8683094B2 (en) 2009-04-30 2014-03-25 Ralink Technology, Corp. Method for enhancing data transmission efficiency
WO2015120825A1 (en) * 2014-02-17 2015-08-20 Mediatek Inc. Data access methods and data access devices utilizing the same
CN105431831A (en) * 2014-02-17 2016-03-23 联发科技股份有限公司 Data access methods and data access devices utilizing the same
US20160218739A1 (en) * 2014-02-17 2016-07-28 Mediatek Inc. Data access methods and data access devices utilizing the same
CN105431831B (en) * 2014-02-17 2018-10-02 联发科技股份有限公司 Data access method and the data access device for utilizing same procedure

Similar Documents

Publication Publication Date Title
TW312068B (en)
US5802048A (en) Method and apparatus for accelerating arbitration in a serial bus by detection of acknowledge packets
CN101867511B (en) Pause frame sending method, associated equipment and system
JPH06511338A (en) Method and apparatus for parallel packet bus
JP2007172575A (en) Network interface controlling lock operation, packet data communication on-chip interconnect system including the network interface, and method of operating the network interface
TW200818773A (en) Merging multi-line log entries
JP2008086027A (en) Method and device for processing remote request
Khan et al. Integrating hardware limitations in CAN schedulability analysis
TW200814634A (en) Systems and methods for packet forward control
CN101447931B (en) Realizing method and device for exclusive operation
CN112106329A (en) Flow control visibility
US20070030862A1 (en) Request processing between failure windows
US7346725B2 (en) Method and apparatus for generating traffic in an electronic bridge via a local controller
US7454456B2 (en) Apparatus and method of improving network performance using virtual interfaces
KR970009050A (en) Arbitration exchange device and method thereof for parallel common bus type high speed packet exchange system
TW200401538A (en) A method and apparatus for hublink read return streaming
TW575809B (en) Method and system for calculating dynamic burst length
US20090240793A1 (en) Memory Buffer Management Method and System Having Multiple Receive Ring Buffers
AU2013207584B2 (en) Time-locked network and nodes for exchanging secure data packets
WO2022160206A1 (en) System-on-chip abnormality processing method and apparatus, and system on chip
WO2016095340A1 (en) Method and device for determining that data is sent successfully
US20050160206A1 (en) Method and system for calculating dynamic burst length
JP5906078B2 (en) Data transfer apparatus and data transfer method
JPS61125253A (en) Medium access control system
JP2003122706A (en) Data processing system

Legal Events

Date Code Title Description
MM4A Annulment or lapse of patent due to non-payment of fees