TW507209B - Asynchronous FIFO controller - Google Patents
Asynchronous FIFO controller Download PDFInfo
- Publication number
- TW507209B TW507209B TW90112702A TW90112702A TW507209B TW 507209 B TW507209 B TW 507209B TW 90112702 A TW90112702 A TW 90112702A TW 90112702 A TW90112702 A TW 90112702A TW 507209 B TW507209 B TW 507209B
- Authority
- TW
- Taiwan
- Prior art keywords
- buffer
- effective
- read
- aforementioned
- vector
- Prior art date
Links
Landscapes
- Information Transfer Systems (AREA)
Abstract
Description
507209507209
五、發明說明(l) [發明之技術領域] 本發明係有關於一種FIFO控制器的技術領域,·特 說明.,本發明是有關於一種操作於不同時脈頻域(Cl〇c Domain ),可保持FIFO緩衝器資料讀寫完整性的非 FIFO控制器。 v [發明背景與先前技藝] 在資訊時代中,資料幾乎是以數位袼式來儲存、處理 與傳輸。諸如語音、音樂、影像、圖像等皆被數位化並以 0與1的形式表不成一連串的位元資訊(Bit Inf〇rmati〇n )。電腦系統中包含複數個積體電路(Int egra tedV. Description of the Invention (l) [Technical Field of the Invention] The present invention relates to the technical field of a FIFO controller. Special explanation. The present invention relates to a method that operates in different clock frequency domains (Cloc Domain). A non-FIFO controller that can maintain the read and write integrity of FIFO buffer data. v [Background of the Invention and Prior Art] In the information age, data is stored, processed, and transmitted almost digitally. Bits such as speech, music, video, and images are all digitized and represent a series of bit information (Bit Infomation) in the form of 0 and 1. The computer system contains a plurality of integrated circuits (Int egra ted
Circuit),而每一積體電路則由數百萬個以上的電晶體 所組成,這些電晶體可被"〇 "的信號所關閉,或是被”丨”的 信號所開啟,以完成電腦程式所規劃的工作(Task )。 在積體電路的操作中,「時脈」(cl〇ck)決定其運算處理 的速度。隨著半導體製程技術的提昇,使得電腦系統的性 月&及處理速度亦大幅提昇。同樣地,諸如硬碟、數據機、 印表機等支援電腦系統的週邊設備也變得較快,甚至對於 應用於電腦網路之傳輸媒體如同軸電纜、光纖等而言,傳 送資料之傳輸速度亦快速提昇中。實際上,整個環境是朝 向著提昇時脈速度。 為了提昇時脈速度所衍生而出的狀況是,許多裝置與 傳輸媒體是操作在不同的時脈速度。例如,一電子裝置可 能執打在較慢時脈速度(如400MHz ),然而另一電子裝置 可能執4丁在較快時脈速度(如1 GHz )。在許多的應用中,Circuit), and each integrated circuit is composed of millions of transistors. These transistors can be turned off by the signal of "〇" or turned on by the signal of "丨" to complete Tasks planned by computer programs. In the operation of the integrated circuit, the "clock" (cloc) determines the speed of its arithmetic processing. With the advancement of semiconductor process technology, the performance of computer systems and processing speeds have also increased significantly. Similarly, peripheral devices that support computer systems such as hard disks, modems, printers, etc. have become faster. Even for transmission media used in computer networks such as coaxial cables and optical fibers, the transmission speed of data transmission Quickly improving. In fact, the entire environment is moving towards clock speed. In order to increase the clock speed, the situation is that many devices and transmission media operate at different clock speeds. For example, one electronic device may operate at a slower clock speed (such as 400 MHz), while another electronic device may operate at a faster clock speed (such as 1 GHz). In many applications,
01P0064.ptd 第4頁 507209 五、發明說明(2) ' --- 此兩種電子裝置需要彼此連接。因此,兩電子裝置的介面 連接便會產生了問題,這是因為彼此執行於不同的時脈速 度而造成的不相容問題,尤其是在傳輪資料上彼此傳送盥 接收的速度是不同。因此,必須有一些方法來處理具有不 同時脈速度之裝置與傳輸線之間的資料交換。 上述問題的其中一個因素是由於許多不同工業標準與 2 定(Pr〇t〇c〇is)的規定。這些標準與通訊協定通、 常指定一特定的時脈頻域。而且,一種標準所規定的時脈 頻域通常是不相容於另一種標準的時脈頻域。而且,系統 的升級(Upgrade)也可能引起時脈的衝突(c〇nfHcts )。此外,一種多用途的電腦系統必須要能適應不同的時 脈處理,如此方可連結操作於不同時脈頻域的裝置。為了 使執行不同時脈速度的裝置間具有相容性、升級性與適應 性,應使用介面控制器來處理這些電子裝置之間的資料交 換。 、 如圖一所示,繪示習知技藝處理不同時脈速度的資料 傳运方式。一 F I F0記憶體3介於系統1與系統2的傳送介面 之間’系統1以時脈CLK1的速度將資料11寫入fifo記憶體3 中’而系統2以時脈CLK2的速度將資料12讀出FIFO記憶體 3 °在此習知技藝中,存在的問題是兩時脈CLK1與^^^之 間的相位差(Phase Di f f erence )可能會導致系統2的資 料1 2讀取錯誤。通常,資料是在每一時脈的前升緣取樣, 倘若CLK2在相位上超前CLK1,則在有些時間點上便會發生 相位差’且相位上的差異會使資料在不同時間上被閂鎖01P0064.ptd Page 4 507209 V. Description of the invention (2) '--- These two electronic devices need to be connected to each other. Therefore, the interface connection between the two electronic devices will cause problems. This is due to incompatibility caused by the different clock speeds executed by each other, especially the speed of transmitting and receiving each other on the wheel data. Therefore, there must be some way to deal with the data exchange between devices with different clock speeds and transmission lines. One of the factors of the above problems is due to the many different industry standards and regulations (Protto). These standards and communication protocols usually specify a specific clock frequency domain. Furthermore, the clock frequency domain specified by one standard is usually incompatible with the clock frequency domain of another standard. In addition, the system upgrade (Upgrade) may also cause clock conflicts (connHcts). In addition, a multi-purpose computer system must be able to adapt to different clock processing so that it can connect devices operating in different clock frequency domains. In order to make the devices with different clock speeds compatible, upgradeable and adaptable, an interface controller should be used to handle the data exchange between these electronic devices. As shown in Figure 1, the method of transferring data with different clock speeds is shown in the conventional techniques. A FI F0 memory 3 is between the transmission interface of system 1 and system 2 'System 1 writes data 11 into fifo memory 3 at clock CLK1' and system 2 writes data 12 at clock CLK 2 Read the FIFO memory 3 ° In this conventional technique, the problem is that the phase difference (Phase Difference) between the two clocks CLK1 and ^^^ may cause the data 12 of the system 2 to be read incorrectly. Generally, the data is sampled at the leading edge of each clock. If CLK2 is ahead of CLK1 in phase, a phase difference will occur at some time points, and the phase difference will cause the data to be latched at different times.
01P0064.ptd 第5頁 五、發明說明(3) (Latch),進而造成資料讀取錯誤。另一方面,資 t =在艾化,所以在時間11的資料可能會不同於時間t 2的 一另一習知技藝係使用有效向量(Valid Vector )來於 憶體之資料組(Bank)㈣狀態,而且使用設定‘ , ΐ〇Γ)與清除向量⑻如VeCt〇r)來指示另一 2 的有效性。此方式雖可達到同時讀/寫的性 1匕疋*貧料組的數量較大時,配置設定向量盥、青除θ 量將造成控制器額外的成本。 I又疋门里一除向 FIFO::哭吴:ί明專利第5, 95 1,6 35號揭露-種非同步 與讀:控;;器::供 使用的容量,其中,寫::數裔來表不FiF0記憶體中可 需以9位元來表示,.,,、碩取控制器的讀/寫指標器皆 而易導致資料誤判。故έ產生競赛現象(―一 Effect), 能處理不同^:=;有3降,成本與設計複雜度,並 料讀寫完整性的介^ B的貝料父換而保持F IF 〇記憶體資 進步性。 〜面控制器’將是具有產業上的利用性與 [發明概述] 本發明的主要曰 需設定向量與清除6 f曰供一種只用—FIFO緩衝器且不 器,以處理不二;::t指示有效性的非同步FIF_ 本發明的π域,間的資料交換。 ,糸提供一種介面控制器,是採用循01P0064.ptd Page 5 Fifth, the description of the invention (3) (Latch), resulting in data reading errors. On the other hand, the data t = in Aihua, so the data at time 11 may be different from that of another conventional art department at time t 2 using the valid vector (Valid Vector) for the bank of memory. State, and use the setting ', ΐ〇Γ) and clear vector ⑻ such as VeCt〇r) to indicate the validity of the other 2. Although this method can achieve simultaneous read / write performance. 1 When the number of lean material groups is large, configuring the vector setting and eliminating the amount of θ will cause additional costs for the controller. I also knocked down the FIFO :: Crying Wu: 明明 Patent No. 5, 95 1,6 35 disclosed-a kind of non-synchronization and read: control ;; device :: capacity for use, where :: Counts can be expressed in 9 bits in FiF0 memory. It is easy to cause misjudgment of data due to the read / write indicator of the master controller. Therefore, there is a competition phenomenon (-Effect), which can handle different ^: =; there are 3 reductions, cost and design complexity, and the read and write integrity of the medium ^ B's parent material to maintain F IF 〇 memory Physical progress. The ~ face controller 'will have industrial applicability and [inventory of the invention] The main purpose of the present invention is to set vectors and clear 6 f for one type only-FIFO buffer and not to deal with the two; :: t indicates the validity of the asynchronous FIF_ data exchange between the π domain of the present invention. , I provide an interface controller
第6頁Page 6
五、發明說明(4) 環式F IJ? 〇方式表示。 本發明的再—目1益f效讀寫的FIFO控制器。 個位元作為交互仑勺係提供一種對每一 F 1 F〇緩衝器只需 器,這在實施非同二Ϊ (Cr〇SS—interfaCe)的FIF0控制 Skew) 。、 ^故計上可避免信號偏斜(Signal 連接於實施例中,-種非同步FIFO控制器, 第二裝置分別操;乍於;輸?徑上,、而f二與 同步FI F0控制器包含· 1緩種一人一日寺脈頻域’前述非 置之間的交換資料;第以暫存第-與第二裝 日辦告彳繁一壯弟,丨面拴制窃,操作於第一時脈頻 D t用Ϊ :置將交換資料以第一時脈速度讀寫缓衝 益,亚使用弟一有效向量以循環FIF0方式記錄第一裝置對 缓衝器的有效讀寫狀態;以及第二介面控制器,操作於第 二時脈頻域,且控制第二裝置將交換資料以第二時脈速度 讀寫緩衝器,並使用第二有效向量以循環fifo方式記錄第 二裝置=緩衝器的有效讀寫狀態;其中,每一介面控制器 以本身操作的時脈頻域同步接收彼此的有效向量,藉以控 制第一與第二骏置有效的讀寫緩衝器。 9 " 本發明非同步FIFO控制器及其諸多優點與特徵將從下 述詳細說明及所附圖式中得到進一步的瞭解。 [圖式標號說明] I、 2 --〜 系統 3 —— 記憶體 II、 12--〜 資料V. Description of the invention (4) Ring F IJ? The present invention provides a FIFO controller that can effectively read and write. Each bit as an interactive system provides a buffer for each F 1 F 0 buffer, which is implemented in the FIF0 control Skew of the Cross-interface. In order to avoid signal skew (Signal is connected to the embodiment, a non-synchronous FIFO controller, the second device operates separately; at first; on the input path, and f and the synchronous FI F0 controller include · 1 slowly planting one person a day in the pulse frequency domain 'the exchange of information between the aforementioned non-position; the first temporary storage of the first-and the second installment of the day to prosecute a strong young man, tied to theft, operating in the first The clock frequency D t is used to read and write the exchange data at the first clock speed, and use the effective vector to record the effective read / write status of the first device to the buffer in a circular FIF0 manner; and Two interface controllers, which operate in the second clock frequency domain, and control the second device to read and write the exchanged data at the second clock speed to the buffer, and use the second effective vector to record the second device = buffer in a cyclic fifo manner The effective read-write status of each interface controller, wherein each interface controller synchronously receives each other's effective vectors in the clock frequency domain of its own operation, thereby controlling the first and second effective read-write buffers. 9 " Synchronous FIFO controller and its many advantages and features will be from below Detailed description and the accompanying drawings further understanding obtained [FIG numeral described formula] I, 2 --~ system 3 - Memory II, 12 - ~ Profile
01P0064.ptd 507209 五、發明說明(5) 13 --- 第一有效向量 14 --- 第二有效向量 15 --- 第一裝置 16 —— 第二裝置 17 --- 資料匯流排 10 --- 第一介面控制器 20 --- 第二介面控制器 30 -一 緩衝器 [發明之詳細說明] 首先請參考圖二’係顯示本發明非同步F IF 〇控 同時讀寫交換資料的電路方塊圖。在本發明第一種 中’非同步F IF 0控制器係電氣連接於第一裝置1 5與 置1 6之間的資料匯流排1 7上’使兩裝置得以交換資 第一裝置15與第二裝置16分別操作於不同的第」時 與第二時脈頻域,該非同步FIFO控 (Buffer ),電氣連接於資料匯流 弟一裝置1 5, 1 6之間的交換資料; 中,此緩衝器係採用循環式F丨F〇架 1 〇,係操作於第一時脈頻域,而控 料以第一時脈CLK1速度讀寫緩衝器 =量13以循環FIFO方式記錄第一裝 ::寫狀態;以及第二介面控制器2〇 而控制第二裝置1 6將交換資料 域 裝置得 於不同 制器包 排17, 在本發 構;第 制第一 3 〇,並 置1 5對 ’係操 以第二 量14以 制器可 實施例 第二裝 料’且 脈頻域 含:緩衝器3 0 用以暫存第一與 明的較佳實施例 一介面控制器 裝置1 5將 且使用第 緩衝器3 0 作於第二 時脈CLK2 循環F IF0 交換資 一有效 的有效 時脈頻 速度讀 方式記 寫緩衝器30,並且使用第二有效向01P0064.ptd 507209 V. Description of the invention (5) 13 --- First effective vector 14 --- Second effective vector 15 --- First device 16-Second device 17 --- Data bus 10- -The first interface controller 20 --- The second interface controller 30-A buffer [Detailed description of the invention] First, please refer to FIG. 2 'which is a circuit block showing the asynchronous F IF control of the present invention and the simultaneous reading and writing of data. Illustration. In the first type of the present invention, the "asynchronous F IF 0 controller is electrically connected to the data bus 17 between the first device 15 and the device 16", so that the two devices can exchange funds between the first device 15 and the first device. The two devices 16 operate in different time and second clock frequency domains respectively. The non-synchronous FIFO control (Buffer) is electrically connected to the data exchange device 1 to exchange data between the devices 1 and 16. The device adopts the circular F 丨 F〇 frame 1 〇, which is operated in the first clock frequency domain, and the material control reads and writes the buffer at the first clock CLK1 speed = the volume 13 records the first load in a circular FIFO manner :: Write state; and the second interface controller 20 and control the second device 16 to obtain the data exchange domain device from a different device package row 17, in the present structure; the first system 30, juxtaposed 15 pairs of 'system The second quantity 14 can be used to implement the second loading of the embodiment, and the pulse frequency domain includes: a buffer 3 0 for temporarily storing the first and Ming preferred embodiments. An interface controller device 15 will be used. The third buffer 3 0 is used for the second clock CLK2. The cycle F IF0 exchanges a valid effective clock frequency and speed. Mode write buffer 30 and use the second valid direction
〇lP〇〇64.ptd 第8頁 507209 瓦、發明說明(6) 錄第二裝置16對緩衝器30的有效讀寫狀態;其中,第一介 :Ξ ϋ Γ,1。?進—步以第一時脈CU1速度同步接收來自第二 二 + f20之第二有效向量14,並根據第一有效向量13 ^ Ϊ效向$ 14的比較結果,控制第—裝置15進行有效 5貝”.、乍,此外,第二介面控制器2 0進一步以第-時脈 ==接收來自第一介面控制器1〇之第以:量 ί 向量?與第二有效向量“的比較結 主fj弟一哀置1 6進行有效讀寫動作。 々样二二考圖二,係顯不有效向量1 3、1 4以循環F I F0方式 =、'友衝器30中交換資料的有效讀寫狀態之示意圖。每一 ==向113、14具有一頭指標器(Head p〇inter)盥一尾 P〇intei:) ’該頭指標器指向有效向量13、 =應巧器3" 個被寫入的區·,而該尾指標器指 效向113、14對應缓衝器3〇中下一個被讀取的區域。 :二’有效向量13、14的每-位置係分別對 應緩衝态30中規劃的區域A、B、c、D、E、F 1 一 指標器與尾指標器指向有效向量的同一個位,則:頭 緩衝器30中並無"有效,,的讀寫動作;♦ L 、j表不 器不是指向有效向量的同一個位置時,則:1興,指標 執行了有效的寫入動作,所以介面控制器::$衝器3 0中 指向有效向量的位置,將讀取該位置對^ ^尾指標器 域,例* :區域C,之後,尾指標器再指 衝器30下一個被讀取區域的位置,例如放向::緩 控制器要控制裝置對緩衝器30進行有对的音 ’爵’丨面 另双的冩入動作時,介〇lP〇〇64.ptd page 8 507209 watts, description of the invention (6) Record the effective reading and writing state of the buffer 30 by the second device 16; among them, the first medium: Ξ ϋ Γ, 1. ? Further-step synchronously receives the second effective vector 14 from the second two + f20 at the speed of the first clock CU1, and controls the first-device 15 to be effective according to the comparison result of the first effective vector 13 ^ Ϊ to $ 14. "5." At first, in addition, the second interface controller 20 further received the first-clock pulse from the first interface controller 10: the amount of vector? Compared with the second effective vector " The main brother fj set aside 16 for effective reading and writing. For example, the second and second examinations in Figure 2 are schematic diagrams showing the invalid vectors 1 3, 1 and 4 in a circular F I F0 mode =, 'Effective reading and writing state of the data exchanged in the friend punch 30. Each == has a pointer pointer (Head p〇inter) and a tail pointer: 113, 14 :) 'The head pointer points to the effective vector 13, = Yingqiao 3 " written areas ·, The tail pointer refers to the next read area in buffer 30, corresponding to 113,14. : Each-position of the two 'effective vectors 13, 14 corresponds to the areas A, B, c, D, E, F 1 planned in the buffer state 30, respectively. A pointer and a tail pointer point to the same bit of the effective vector, then : There is no "effective," read or write action in the head buffer 30. ♦ When the L and j table are not pointing to the same position of the effective vector, then: 1 X, the index has performed an effective write action, so Interface controller :: The position of the effective vector in $ 冲 器 30, which will read the position of the tail indicator field, such as ^^ region C, after that, the tail indicator will be read next to the pointer 30. Take the position of the area, for example, to: When the slow controller wants to control the device to perform the paired sound of the sound "Jiao" on the buffer 30, it will introduce
01p〇〇64.Ptd 第9頁 五、發明謂 、/ ____ 面控制器會根據頭指標器 入該位置對應緩衝器3。d:量的位置,將資料寫 指標器再指向有效向量中綉2,例如:區域Η,之後,頭 置,例如:區域Α。由於緩衝V;30/ 一個被寫入區域的位 所以當頭尾指標器依序指到右循環FIF0組架構, 便會再重新指向有效向量η!的最後-個位置後’ 如丄 u里的弟一個位置。 在本發明非同步FIF0 第二介面控制器20具有相德第-介面控制器10與 工作内容,這些工作包含的電路佈局並同時執行相同的 —、同步化·· 第二:面二:控制态η以第-時脈CU1速度同步接收來自 2。4面,器20的第二有效向量“,而第二介面控= 第-有速度同步接收來自第-介面控制器10的 〜斷緩衝器的有效讀寫動作: 内容::5 ϋ為根據第一有效向量13與第二有效向量“的 入時,§碩指標器所指向兩有效向量的位置中皆可有效寫 的F η表不目前並無裝置對緩衝器3 0中由頭指標器所於二01p〇〇64.Ptd Page 9 5. Invention means that / ____ The surface controller will enter the corresponding buffer 3 according to the head pointer. d: the position of the quantity, write the data and the pointer points to the embroidery 2 in the effective vector, for example: area Η, after that, the head position, for example: area A. Because the buffer V; 30 / a bit written to the area, when the head and tail indicators point to the right-cycle FIF0 group structure in sequence, they will point to the last position of the effective vector η! Again. A position. In the present invention, the non-synchronous FIF0 second interface controller 20 has the phase-first interface controller 10 and work content. These tasks include the circuit layout and perform the same at the same time. Synchronization ... Second: Surface 2: Control state η synchronously receives the second effective vector from the 2.4-plane, the second effective vector "2" at the speed of the-clock CU1, and the second interface control = the speed-received synchronous reception of the ~~ buffer from the-interface controller 10 Valid read and write actions: Content :: 5 ϋ is based on the time of the first effective vector 13 and the second effective vector ", F η that can be effectively written in the position of the two effective vectors pointed by the § master indicator is not currently combined. No device to buffer 30 in the second by the pointer indicator
•: f效寫入動作;反之,表示另一介面控制器I τ _ 3 0執行了有效寫入動作。 效讀此2 ’當尾指標器所指向兩有效向量的位置中皆可有 指向Ζ ^ ’表示目前並無裝置對缓衝器3 0中由尾指標器所 器對》^域執行有效讀取動作;反之,表示另一介面控制 、灵衝益3 0執行了有效讀取動作。•: f effect write operation; otherwise, it means that another interface controller I τ _ 30 has performed a valid write operation. Effectively read this 2 'When the tail pointer points to two effective vectors, there can be pointers to ^ ^' means that there is currently no device to perform a valid read of the "^^" field in the buffer 30 by the tail pointer. Action; on the contrary, it means that another interface control, Ling Chongyi 30 has performed an effective reading action.
°1Ρ0064. Ptd 第10頁 507209 五、發明說明(8) 三、裝置執行有效讀取: 當一介面控制器執行讀取動作時,會碟認另一介面控 制器已對尾指標器所對應的緩衝器3 〇區域執行了有效寫入 動作後,根據尾指標器指向本身有效向量的位置,此介面 控制器會控制裝置讀取該位置對應緩衝器3 0的區域内資 料,同時設定本身有效向量的該位置被有效讀取,且使尾 指標器指向本身有效向量中緩衝器3 0下一個被讀取區域的 位置。 四、裝置執行有效寫入: 當一介面控制器執行寫入動作時,會確認另一介面控 制器已對頭指標器所對應的缓衝器3 0區域執行了有效讀取 後,根據頭指標器指向本身有效向量的位置,此介面控制 器會控制裝置寫入資料到該位置對應緩衝器3 0的區域,同 時設定該位置被有效寫入,且使頭指標器指向本身有效向 量中缓衝器3 〇下一被寫入區域的位置。 其中’有效向量的位置是被有效讀取或是有效寫入的 設定,只需以一個位元表示,而由設定成1值或0值表示該 位置對應緩衝器30的區域内資料是被有效寫入或有效讀 取,因此,本發明實施例可避免競賽現象影響F I ρ 〇讀寫性 能。 制器可 第二種 I置15 以將資 最後請參考圖四,係顯示本發明非同步F丨F 〇控 同時讀寫交換資料的電路方塊圖。圖四為本發明之 實施例,其中非同步FIFO控制器係電氣連接於第〜 至第二裝置1 6的資料匯流排1 7上,使第一袭置丨5得° 1Ρ0064. Ptd Page 10 507209 V. Description of the invention (8) III. Effective reading of the device: When one interface controller performs the reading action, it will recognize that the other interface controller has corresponding to the tail indicator. After the effective writing operation is performed in the buffer 30 area, the interface controller will control the device to read the data in the area corresponding to the buffer 30 according to the tail pointer and set the effective vector at the same time. This position is effectively read, and the tail pointer is directed to the position of the next read area of the buffer 30 in its effective vector. Fourth, the device performs effective writing: When one interface controller performs a write operation, it will confirm that the other interface controller has performed a valid read on the buffer 30 area corresponding to the head pointer, and then according to the head pointer Point to the position of the effective vector itself. This interface controller will control the device to write data to the area corresponding to the buffer 30. At the same time, set the position to be effectively written and make the head pointer point to the buffer in the effective vector itself. 3 The position of the next written area. Among them, the position of the effective vector is a setting that is effectively read or written, and only needs to be expressed by one bit, and a value set to 1 or 0 indicates that the data in the area corresponding to the position 30 is valid. Write or read effectively, therefore, the embodiment of the present invention can avoid the race phenomenon from affecting FI ρ 〇 read and write performance. The controller can be set to 15 for the second type. Finally, please refer to FIG. 4, which is a block diagram showing a circuit for asynchronously reading and writing data in the synchronous F 丨 F0 controller of the present invention. FIG. 4 is an embodiment of the present invention, in which the asynchronous FIFO controller is electrically connected to the data bus 17 of the first to second devices 16 to make the first attack
0lP0064.ptd 第11頁0lP0064.ptd Page 11
507209 五、發明說明(9) 第一裝置15與第二 與第二時脈頻域, (Buffer),電氣 弟一裝置15, 16之 環式F I F 0組架構; 頻域,而控制第一 衝器3 0讀取資料, 式記錄第一裝置1 5 介面控制器20,係 16以第二時脈CLK2 向量14循環FIF〇方 入狀態;其中,第 速度同步接收第二 第二有效向量1 4的 取動作;而第二介 同步接收第一有效 有效向量1 4的比較 〇 裝置1 6分別操 該非同步FIFO 連接於資料匯 間的交換資 第一介面控制 裝置1 5將資料 並且使用第一 對緩衝 操作於 速度寫 器3 0的 第二時 入緩衝 第二裝 控制器 量1 4, 果,控 器20進 ’並根 控制第 式記錄 一介面 有效向 比較結 面控制 向量13 結果’ 料傳輸至第二裝置16,且 作於不同的第一時脈頻域 控制器包含:一緩衝器3 〇 流排1 7,用以暫存第一與 料,且此組緩衝器採用循 器1 〇,係操作於第一時脈 以第一時脈CLK1速度自緩 有效向量13以循環FIFO方 有效讀取狀態;以及第二 脈頻域,而控制第二裝置 器3 0 ’並且使用第二有效 置16對緩衝器30的有效寫 10進一步以第一時脈CLK1 並根據第一有效向量1 3與 制第一裝置1 5進行有效讀 一步以第二時脈CLK2速度 據第一有效向量13與第二 二裝置1 6進行有效寫入動 在詳細說明本發明的較佳實施例之後,熟悉該項技術 人士可清楚的瞭解,並在不脫離下述申請專利範圍與精神 下可進行各種變化與改變’而且本發明亦不受限於說明書 之實施例的實施方式。 [發明功效] 根據本發明較佳實施例所實施的非同步F I F 〇控制器,507209 V. Description of the invention (9) Ring FIF 0 group architecture of the first device 15 and the second and second clock frequency domain, (Buffer), electric device one and 15, 16; frequency domain, and control the first impulse The device 30 reads the data and records the first device 15. The interface controller 20, the system 16 uses the second clock CLK2 vector and 14 cycles of the FIF square-in state; among them, the second speed receives the second effective vector 1 4 synchronously. And the second medium synchronously receives the comparison of the first effective vector 14; the device 16 operates the asynchronous FIFO and exchanges data between the data sinks; the first interface control device 15 stores the data and uses the first pair The buffering operation is performed at the second time of the speed writer 30, and the second controller is installed. As a result, the controller 20 enters the “parallel control formula” to record an interface that is effective to compare the control vector 13 with the result. The second device 16, which is used for different first clock frequency domain controllers, includes: a buffer 30 streamer 17 for temporarily storing the first data, and this set of buffers uses a cycler 10, Operating on the first clock to the first clock CLK1 The self-slowing effective vector 13 reads the state effectively in the circular FIFO side; and the second pulse frequency domain, and controls the second device 3 0 ′ and uses the second valid set 16 to effectively write the buffer 30 to 10 further to the first The clock CLK1 is effectively read according to the first effective vector 1 3 and the first device 15 for one step at the second clock CLK2 speed. The effective writing is performed according to the first effective vector 13 and the second two device 16 in detail. After a preferred embodiment of the present invention, those skilled in the art can clearly understand it, and can make various changes and modifications without departing from the scope and spirit of the patent application described below, and the present invention is not limited to the embodiments of the description. Implementation. [Effect of the invention] Asynchronous F I F 〇 controller implemented according to the preferred embodiment of the present invention,
01P0064.Ptd 第 12 頁 507209 五、發明說明(ίο) 僅需要有效向量來記錄緩衝 )I ( λ — 1 ηΗ"Λν*·Ρο/^^ 介面信 以一個 響,且 雜度降 成本, F I F 0 控 組 F IF 0 綜 術在實 法,完 價值, 審並賜 號(cross-interface 位元表示指標器的有 採用循環FIFO架構等 低且介面控制間的聯 同時時脈頻域可以獲 制器在達到同時讀寫 緩衝器。 上所述,本發明具有 務上與應用上之缺失 成實用可靠之系統, 實已符合發明專利< 准專利權盈保障,以 器的有效讀寫狀態,其中交互 signal)對於每一緩衝器只需 效讀寫動作,避免競赛現象影 特徵,使得FIFO控制器的閘複 絡信號也簡化,谁而分 间% 進而減少製造 侍,昇。此外,本發明非同步 的咼效能時,其架構中僅需一 諸多優良特性,並解決習知技 與不便,提出有效之解決方 進而達成新穎且附經濟效益之 申請要件,懇請鈞局能予詳 優惠民生實感德便。01P0064.Ptd Page 12 507209 V. Description of the invention (ίο) Only the effective vector is needed to record the buffer) I (λ — 1 ηΗ " Λν * · Ρο / ^^ The interface letter is one ring, and the complexity is reduced to cost. The control group F IF 0 comprehensive operation in practice, complete value, review and give the number (cross-interface bit indicates that the indicator has a circular FIFO structure and other low-speed interface clock interface can be obtained by the simultaneous clock frequency domain As described above, the present invention has a practical and reliable system that is practically and practically lacking, and has been in line with the invention patent < (Interactive signal) For each buffer, only read and write operations are required to avoid the characteristics of the race phenomenon, so that the complex signal of the FIFO controller is also simplified. In the case of asynchronous performance, only a lot of excellent characteristics are needed in its architecture, and it solves the conventional problems and inconveniences, and proposes effective solutions to achieve novel and economically beneficial application requirements. For more favorable to the people's livelihood will be a real sense of ethics.
507209 圖式簡單說明 圖一為習知技術處理不同時脈速度間資料傳送的電路 方塊圖。 圖二為本發明非同步F I F0控制器可同時讀寫交互資料 的電路方塊圖。 圖二為有效向量以循ί哀F I F 0方式記錄緩衝之不意 圖。 圖四為本發明非同步F I F0控制器第二種實施例的電路 方塊圖。507209 Brief description of the diagrams Figure 1 is a block diagram of a conventional technique for processing data transmission between different clock speeds. Figure 2 is a circuit block diagram of the asynchronous F I F0 controller of the present invention that can simultaneously read and write interactive data. Figure 2 is a schematic diagram of the effective vector recording buffer in a cycle F I F 0 manner. Fig. 4 is a circuit block diagram of the second embodiment of the asynchronous F I F0 controller of the present invention.
01P0064.ptd 第14頁01P0064.ptd Page 14
Claims (1)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW90112702A TW507209B (en) | 2001-05-25 | 2001-05-25 | Asynchronous FIFO controller |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW90112702A TW507209B (en) | 2001-05-25 | 2001-05-25 | Asynchronous FIFO controller |
Publications (1)
Publication Number | Publication Date |
---|---|
TW507209B true TW507209B (en) | 2002-10-21 |
Family
ID=27621846
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW90112702A TW507209B (en) | 2001-05-25 | 2001-05-25 | Asynchronous FIFO controller |
Country Status (1)
Country | Link |
---|---|
TW (1) | TW507209B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7689739B2 (en) | 2005-07-11 | 2010-03-30 | Via Technologies, Inc. | Spread spectrum receiver, apparatus and method of a circular buffer for multirate data |
TWI792972B (en) * | 2022-01-27 | 2023-02-11 | 瑞昱半導體股份有限公司 | Methods for controlling asynchronous fifo memory and data transmission system utilizing the same |
-
2001
- 2001-05-25 TW TW90112702A patent/TW507209B/en not_active IP Right Cessation
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7689739B2 (en) | 2005-07-11 | 2010-03-30 | Via Technologies, Inc. | Spread spectrum receiver, apparatus and method of a circular buffer for multirate data |
TWI792972B (en) * | 2022-01-27 | 2023-02-11 | 瑞昱半導體股份有限公司 | Methods for controlling asynchronous fifo memory and data transmission system utilizing the same |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US5430847A (en) | Method and system for extending system buses to external devices | |
US20030126336A1 (en) | Reordering unrelated transactions from an ordered interface | |
CN101568904A (en) | High speed fanned out system architecture and input/output circuits for non-volatile memory | |
CN101310339A (en) | Memory device and method having multiple internal data buses and memory bank interleaving | |
US8423703B2 (en) | Data transfer in memory card system | |
US20040107306A1 (en) | Ordering rule controlled command storage | |
CN101344870A (en) | FIFO control module with strong reusability and method for managing internal memory | |
US8559576B2 (en) | Adaptive synchronization circuit | |
CN114721983B (en) | DDR4 accelerating read-write device | |
TW507209B (en) | Asynchronous FIFO controller | |
CN105573947B (en) | A kind of SD/MMC card control methods based on APB buses | |
US6477607B1 (en) | Duplexing structure of switching system processor and method thereof | |
US20030005344A1 (en) | Synchronizing data with a capture pulse and synchronizer | |
CN111198837B (en) | SDIO interface system based on FPGA and controller bridging method | |
CN107958438A (en) | A kind of OpenGL creates display listing circuitry | |
CN116383132A (en) | SOC chip and data storage method thereof | |
CN1153217C (en) | Asynchronous FIFO controller | |
TWI325532B (en) | Controlling circuit and method for power saving | |
CN116340217A (en) | Data processing method and related device | |
CN100365606C (en) | Direct memory access method for data of secure digital memory card and interface circuit therefor | |
US7861026B2 (en) | Signal relay device and method for accessing an external memory via the signal relay device | |
US8473665B2 (en) | Universal serial bus (USB) transaction translator and a universal serial bus (USB) isochronous-in transaction method | |
CN100365639C (en) | First-in first-out analog unit and logic verification analog system | |
US4238834A (en) | Apparatus for coordinating real time transfer of data from a processor to a magnetic media device | |
CN218068841U (en) | Configurable APB protocol converter and SOC system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
GD4A | Issue of patent certificate for granted invention patent | ||
MK4A | Expiration of patent term of an invention patent |