TW495673B - Method and apparatus for transmitting registered data onto PCI bus - Google Patents
Method and apparatus for transmitting registered data onto PCI bus Download PDFInfo
- Publication number
- TW495673B TW495673B TW89123934A TW89123934A TW495673B TW 495673 B TW495673 B TW 495673B TW 89123934 A TW89123934 A TW 89123934A TW 89123934 A TW89123934 A TW 89123934A TW 495673 B TW495673 B TW 495673B
- Authority
- TW
- Taiwan
- Prior art keywords
- data
- signal
- multiplexer
- address
- aforementioned
- Prior art date
Links
Landscapes
- Bus Control (AREA)
Abstract
Description
4W6734W673
【發明領域】 減少^發明係關於傳送資料的方法與裝置,特別是關於可 送出传號的延遲時間且不需大幅增加電路複雜度之傳 、 暫存之資料至PCI匯流排之裝置與方法。 【習知技術】[Field of the Invention] The invention and method of reducing data are related to a method and a device for transmitting data, especially a device and a method for transmitting and temporarily storing data to a PCI bus without delaying the transmission of a signal without greatly increasing circuit complexity. [Learning technology]
之問ί知系統通常採用一個或更多週邊匯流排使各種裝置 協二2ί ί聯繫。這些裝置(agent)之間的聯繫一般依據 流2 A,者仃。其中一項協定是互相連接週邊元件(Pc 1)匯 =寬i當脈衝頻率在66MHZ時’允許每秒266Mb之資料通訊 基本上同時會有很多裝置連接至Pc I匯流排。 ^匯流排動作(bUS tranSaCti〇n)至其他接收端時/, 垃/ 排即執行資料運送。請求端被稱為"主控端,,,而 要收端,稱為"受控端"。當主控端輸出位址/命令至受控 =ί次ΐ主控端之命令是"寫入"動作,則主控端接著輸出 貝;’同時受控端將位址/命令解碼並接收資料;若 主,端之命令是”讀取,,動作,則受控端回送資料。因此,Question systems usually use one or more peripheral buses to connect various devices. The connection between these agents is generally based on flow 2 A, or 仃. One of the agreements is to connect peripheral components (Pc 1) to each other = wide i. When the pulse frequency is 66MHZ, 266 Mbps of data communication is allowed. Basically, many devices will be connected to the Pc I bus at the same time. ^ When the bus action (bUS tranSaCti0n) is / to other receivers, the data transfer is performed. The requesting end is called " the master, and the receiving end is called " the controlled end ". When the master terminal outputs the address / command to the controlled = ί times, the command of the master terminal is " write ", the master terminal then outputs the shell; 'At the same time the controlled terminal decodes the address / command and Receive data; if the command of the master and the terminal is "read ,, action, the controlled terminal sends back the data. Therefore,
;排動作包含一個位址相位(address phase)以及之後 的個或多個資料相位(data phase)。 ^ PCI資料傳送是由FRAME#、IRDY#及TRDY#等三個信號 來控制frame#信號是由主控端驅動,藉以表示傳送動作 ^起始和結束。IRDY#信號亦由主控端驅動,藉以將主控 端準備傳送資料的訊息告知受控端。而TRDY#信號是由受 控端驅動’藉以將受控端準備傳送資料的訊息告知主控The sorting action includes an address phase and one or more data phases thereafter. ^ PCI data transmission is controlled by three signals: FRAME #, IRDY # and TRDY #. The frame # signal is driven by the master to indicate the transfer action. ^ Start and end. The IRDY # signal is also driven by the master, thereby notifying the controlled end of the message that the master is ready to send data. The TRDY # signal is driven by the controlled end ’to inform the master of the message that the controlled end is ready to send data.
495673 五、發明說明(2) 端。若FRAME#和IRDY#信號兩者被不致能時,介面是處在 閒置狀態。當FRAME#信號被致能時,被視為位址相位,且 位址/匯流排命令碼在第一脈衝邊緣被送出。當IRDY#和 TRDY#兩者被致能時,資料在之後的資料相位期間被送 出。當主控端結束最後的資料傳送,FRAME#信號被不致 能。在受控端完成最後的資料傳送之後,TRDY#信號也是 不致能,且介面回到閒置狀態。 圖1顯示習知主控端的輸出控制邏輯連接至pc I匯流排 之功能方塊圖。PC I匯流排之規格中定義超過47個腳位, 但是在本圖中僅標示出AD匯流排AD[31:〇]1〇l。目前相位 資料指標105、目前相位位址指標1 〇6、以及位址相位選擇 #號1 0 7是裝置的内部狀態信號。目前相位資料指標丨〇 5和 目前相位位址指標1 〇 6是由輸出動作控制的指標,藉以分 別^換儲存在緩衝器113和112之資料和位址。位址S相位選 擇信號1 0 7是用來選擇位址相位或資料相位。多工器丨丨J和 I/O緩衝器110是用來驅動信號至PCI匯流排。資料緩衝器 和位址緩衝器的可以是任何形式實施。 圖2是描述圖1習知主控端之運作的時序圖。本圖上半 部顯示主控端之内部信號的波形’以及下半部顯示由主控 端驅動輸出信號至PCI匯流排的波形。在第三脈衝週期,馨 該主控端利用多工器111根據位址相位選擇信號1〇7選擇從 位址緩衝器112所輸出之位址a』· ’且經過該多工器“I與 i/Ο緩衝器110之延遲D2後,將位址\驅動至PCI匯流排、。 在下一個脈衝週期時,位址相位選擇信號1〇7被不致能,495673 V. Description of invention (2). If both the FRAME # and IRDY # signals are disabled, the interface is in an idle state. When the FRAME # signal is enabled, it is regarded as the address phase, and the address / bus command code is sent out at the edge of the first pulse. When both IRDY # and TRDY # are enabled, data is sent during the subsequent data phase. When the master ends the final data transmission, the FRAME # signal is disabled. After the controlled terminal completes the final data transmission, the TRDY # signal is also disabled and the interface returns to the idle state. Figure 1 shows the functional block diagram of the conventional master control output connected to the PC I bus. There are more than 47 pins defined in the specifications of the PC I bus, but only AD [31: 〇] 101 is shown in this figure. The current phase data indicator 105, the current phase address indicator 1 06, and the address phase selection #No. 1 7 are internal status signals of the device. The current phase data index 丨 05 and the current phase address index 106 are indexes controlled by the output action, so that the data and addresses stored in the buffers 113 and 112 are changed respectively. The address S phase selection signal 1 0 7 is used to select the address phase or data phase. The multiplexer J and the I / O buffer 110 are used to drive signals to the PCI bus. The data buffer and address buffer can be implemented in any form. FIG. 2 is a timing diagram describing the operation of the conventional master terminal in FIG. 1. The upper part of the figure shows the waveform of the internal signal of the master terminal 'and the lower part shows the waveform of the output signal driven by the master to the PCI bus. In the third pulse period, the master uses the multiplexer 111 to select the address a output from the address buffer 112 according to the address phase selection signal 107, and passes the multiplexer "I and After the delay D2 of the I / O buffer 110, the address \ is driven to the PCI bus. At the next pulse period, the address phase selection signal 107 is disabled,
第5頁 495673Page 5 495673
=主控端利用多工器11 1根據位址相位選擇信號i 〇 7選擇從 資料緩衝器11 3所輸出之資料Dj,且經過該多工器丨i J與 I/O緩衝|§11〇之延遲])3後,將資料])厂驅動至?(:1匯流排。 因為無資料轉換產生在第六脈衝週期,Dj是維持驅動至 pci匯流排。在第六脈衝週期之後,trdy# 信號兩 者同時被致能,目前相位資料指標1〇5的指數】將改變為 。此外,由多工器lu和1/0緩衝器11()產生輸出延遲, 貧料相位改變從Dj到Dj+1也導致一些延遲。 二此外,如圖1所示主控端在驅動AD信號至PCI匯流排之 =丄將包含大量的延遲,例如Μ和〇2。該延遲將減低受控 ^接收同—PC 1匯流排信號之可利用時間。當時序週期 =:越土 2短時’將難以使受控端有效地處理信號。舉例 =,s時脈頻率在66MHz且時序週期為15ns時,受控 將無法在此短或甚至更短時序期間内處理信號。 ^ 為了解決上述延遲問題,圖3揭示出具有兩層管路 /^典^)之其他電路。圖3所示之功能方塊圖係描述習 T二裝置的輸出控制邏輯。在這電路結構中,在發 排之5 ’先將輸出信號暫存,且在信號被傳 =資二βΡ A 1料緩衝器310提前一個週期準備資料, 顯然地,該電路必須使用其他 貝料輸出於PC 1匯流排之前儲存初值 二位貝枓。如圖3所示’該電路使用二個多工器3ιι和 樣ψ &六個正反器3 1 3和31 4來保持初值資料直到資料 傳达成功。然而,圖3的電路相當複雜。 貝讨= The master uses the multiplexer 11 1 to select the data Dj output from the data buffer 11 3 according to the address phase selection signal i 〇7, and passes through the multiplexer i J and I / O buffer | §11〇 Delay]) after 3, will the data]) factory drive to? (: 1 bus. Because no data conversion occurs in the sixth pulse period, Dj is continuously driven to the PCI bus. After the sixth pulse period, both trdy # signals are enabled at the same time. The current phase data index is 105. The exponent] will change to. In addition, the output delay is caused by the multiplexer lu and the 1/0 buffer 11 (), and the lean phase change from Dj to Dj + 1 also causes some delay. In addition, as shown in Figure 1 The master driving the AD signal to the PCI bus = 之 will include a large amount of delay, such as M and 〇2. This delay will reduce the control ^ receive the same-PC 1 bus signal available time. When the timing cycle = : Yuetu 2 for a short time will make it difficult for the controlled end to effectively process the signal. For example =, when the s clock frequency is 66MHz and the timing period is 15ns, the controlled will not be able to process the signal within this short or even shorter timing period ^ In order to solve the above-mentioned delay problem, FIG. 3 reveals other circuits having two layers of pipelines. The functional block diagram shown in Figure 3 describes the output control logic of the Xi T 2 device. In this circuit structure, the output signal is temporarily stored before the 5 ′ of the transmission row, and the signal is transmitted = data ββ A 1 material buffer 310 to prepare data one cycle in advance. Obviously, the circuit must use other materials The output is stored in the initial two-digit shellfish before the PC 1 bus. As shown in Fig. 3 ', this circuit uses two multiplexers 3 ι and samples ψ & six flip-flops 3 1 3 and 31 4 to maintain the initial value data until the data transmission is successful. However, the circuit of Figure 3 is quite complex. Begging
495673 五、發明說明(4) 【發明概要】 本發明之目的是提供一種可以減少輸出信號延遲時間 且不需大幅增加電路複雜度之傳送已暫存資料至pc z匯流 排之裝置與方法。 根據本發明之傳送已暫存之資料至pci匯流排之裝 置’包—資料緩衝器,用來儲存複數個暫存資料,並 根據目I相位資料指標和下一相位資料指標來分別輸出目 則賃料L號和下一資料信號;一位址緩衝器,用來儲存位 址,並依據目前相位位址指標輸出目前位址丨一第一多工 器’係接收資料緩衝器之目前資料信號和位址緩衝器之位 t、盛ί,據位址相位選擇信號輸出目前資料信號或位址; it =山t二係從ΡΠ匯流排接收1RDY#信號和TRDY#信號, = 換選擇信號;-第二多工器,係接收資料緩 料^ ΐ t仏號和第一多工器輸出的信號,並根據資 :傳c輪出下一資料信號或第一多工器的輸出信 脈衝將#i i,係接收第二多工器的輪出,並根據參考 _ ^ 二夕工器輪出的信號驅動至該PCI匯流排。 下步:種傳送已暫存之資料至PCI匯流排之方法,包含以 '第(a\位址輪出步驟,由位址緩衝器輸出一位址信號到 禾—多工器; 料信u S 2 ?出步驟’ * —資料緩衝11分別輸出目前資 器Γϋ下一育料信號到前述第一多工器和一第二多工 五、發明說明(5) 位選作ί二選擇步驟,由前述第一多工器依據—仇^ » ^輪出前述目前資料信號或前述位址信號;相495673 V. Description of the invention (4) [Summary of the invention] The object of the present invention is to provide a device and method for transmitting temporarily stored data to a pcz bus that can reduce the delay time of the output signal without greatly increasing the complexity of the circuit. According to the present invention, the device 'packet-data buffer' for transmitting temporarily stored data to the PCI bus is used to store a plurality of temporarily stored data, and outputs the targets respectively according to the phase data index and the next phase data index of the target. The lease number L and the next data signal; an address buffer to store the address, and output the current address according to the current phase address index 丨 a first multiplexer 'is the current data signal and The bits t and Sheng of the address buffer output the current data signal or address according to the address phase selection signal; it = the second system receives the 1RDY # signal and the TRDY # signal from the PI bus, = the selection signal is changed;- The second multiplexer receives the data buffer ^ ΐ t 仏 and the signal output by the first multiplexer, and outputs the next data signal or the output signal pulse of the first multiplexer according to the information: ii. It receives the rotation of the second multiplexer and drives it to the PCI bus according to the signal of the reference rotation of the second multiplexer. Next step: A method for transmitting the temporarily stored data to the PCI bus, including the step (a \ address rotation), and the address buffer outputs a one-bit signal to the Wo-multiplexer; Step S2: 'Data buffer 11 outputs the current breeding signal of the current resource Γ to the first multiplexer and a second multiplexer respectively. 5. Description of the invention (5) is selected as the second selection step. Based on the aforementioned first multiplexer—revenge ^ »^ turns out the aforementioned current data signal or the aforementioned address signal;
送選擇作選擇步驟,由前述第二多工器俊據一U資%L 的广缺〇儿輪出前述下一資料信號或前述第一多工3§认傳 的化就;及 夕工态輪出Send the selection as the selection step, and the aforementioned second multiplexer will use the vacancy of U %% L to rotate the next data signal or the first multiplexed 3§ acknowledgment; and the working state Turn out
Te^驅動步驟,由一正反器根據工作脈衝送出針、+ -夕”信號到PCI匯流排。 别述第 ,藉由運用1〇Υ#信號和TRDYMf號之組合來士 個!料緩衝器所輸出之目前資料與下-資料:以 乍週期將位址或資料傳送到正反器輸入 最則 工作脈衝送出資料或位址到PCI取後再 貝科或位址只會受到正反器之延遲。 故 【實施例] 以下參考圖式詳細說明本發明傳送已暫存之資料 PCI匯流排之裝置與方法,#以在不大幅增加電路複雜声 下,減少操作送出信號的延遲時間。 ”又 方地口4顯/太本Λ明應用於主控端之輸出控制邏輯的功能 方,圖。在本貫細例巾’主控端40 0包含儲存複數資料之 一:俺貝料緩衝器410、—儲存複數個位址之位 。該2flw資料緩衝器410依據目前相位資料指 標407和下-相位:貝料指標406分別輪出目前資料4〇4和下 一資該位址資料緩衝器411依據目前相位 位:412丨P Ϊ址4〇5。該主控端400更包含一第 一多工态412和一第二多工器413,盆、 , ,精以選擇目前資料In the Te ^ driving step, a flip-flop sends the needle, +-"signal to the PCI bus according to the working pulse. Not to mention the first, by using a combination of the 10Υ # signal and the TRDYMf number, a material buffer is provided. The output current data and the next-data: the address or data is sent to the flip-flop at the first cycle. The most working pulse is sent to send the data or address to the PCI, and then the Beco or address will only be affected by the flip-flop. Delay. Therefore, [Embodiment] The following describes in detail the device and method for transmitting temporarily stored data PCI bus of the present invention with reference to the drawings, so as to reduce the delay time of the operation sending signal without greatly increasing the complexity of the circuit. " Fangdikou 4 display / Taiben Λming function applied to the output control logic of the main control terminal, picture. In the present example, the main control terminal 40 0 includes one of storing a plurality of data: a shellfish buffer 410, which stores a plurality of bits. The 2flw data buffer 410 is based on the current phase data index 407 and the bottom-phase: the shell material index 406 respectively rotates the current data 404 and the next data. The address data buffer 411 is based on the current phase bit: 412 丨 P address 40%. The main control terminal 400 further includes a first multiplexing mode 412 and a second multiplexer 413, basin,, and so on, so as to select the current data.
495673495673
五、發明說明(6) 404、下一資料403或目前位址405。該第一多工器412連接 至2R1W資料缓衝器410與1R1W位址緩衝器411,並依據位址 相位選擇信號409來選擇目前資料4 04或目前位址4〇5。同 時’第二多工器413連接至2R1W資料緩衝器和第一多工 器412 ’並依據由或閘415致能之資料傳送選擇信號4qi來 選擇下一資料403或第一多工器412所輸出之信號Γ或閘 415接收IRDY##^421和TRDY#信號422,並進行或(qr)運 异。主控端400使用一正反器414將第二多工器413輸出之5. Description of the invention (6) 404, next data 403 or current address 405. The first multiplexer 412 is connected to the 2R1W data buffer 410 and the 1R1W address buffer 411, and selects the current data 404 or the current address 405 according to the address phase selection signal 409. At the same time, 'the second multiplexer 413 is connected to the 2R1W data buffer and the first multiplexer 412' and selects the next data 403 or the first multiplexer 412 according to the data transmission selection signal 4qi enabled by the OR gate 415. The output signal Γ or gate 415 receives the IRDY ## ^ 421 and TRDY # signals 422 and performs an OR (qr) operation. The main control terminal 400 uses a flip-flop 414 to output the second multiplexer 413.
Pre-AD信號驅動到PCI匯流排。該正反器414是由脈衝424The Pre-AD signal is driven to the PCI bus. The flip-flop 414 is composed of pulses 424
的上緣觸發。 本發明的主控端與習知主控端之主要不同處為(如圖 :示)使用2R1W資料緩衝器410。該資料緩衝器41〇根據目 前相位資料指標407和下一相位資料指標4〇6分別輸出 資料404和下一資料4〇3。The upper edge of the trigger. The main difference between the main control terminal of the present invention and the conventional main control terminal is (as shown in the figure) using a 2R1W data buffer 410. The data buffer 410 outputs data 404 and next data 403 according to the current phase data index 407 and the next phase data index 406, respectively.
^ 多工器412動作類似於圖1所示之多工器U1,除了提 月J個脈衝週期(在PC I匯流排之實際ad動作之週期)來觸 反器414藉以清除已暫存之輸出信號。在pci匯流排y 功傳送資料後,即利用多工器413選擇下—資料4〇3 ;否 貝上妯ί :器413選擇多工器412輸出的資料,藉以在1^1匯 ί!傳送ϊ ί ?同資料"匕外’多工器413的控制開關是資 422作钍號4〇1,也就是以1RDY#信號421*TRDY“f^ 後,資粗經入之或閘邏輯415的輸出。當有一資料成功傳主 選擇由目L f1" 410會輸出下—資料。反之,乡工器413# '擇由目則相位資料指標4〇7所指向之目前資料綱,並幸^ The operation of the multiplexer 412 is similar to that of the multiplexer U1 shown in FIG. 1, except that the J pulse period (the cycle of the actual ad operation at the PC I bus) is raised to trigger the inverter 414 to clear the temporarily output. signal. After the data is transmitted by the PCI bus y, the multiplexer 413 is used to select the data—data 403; No: the 413 selects the data output by the multiplexer 412, so that it can be transmitted on the 1 ^ 1 sink! ϊ ί The control switch of the multiplexer 413 with the same data is 422 as the number 401, that is, after the 1RDY # signal 421 * TRDY "f ^, the information is input into the OR logic 415 When there is a data successfully, the host selects the item L f1 " 410 to output the data. Otherwise, the rural worker 413 # 'selects the item of the current data outline pointed by the phase data index 407, and fortunately
第9頁 495673Page 9 495673
五、發明說明(7) 出成Pre - AD信號4 02。在本發明中正反器414和I/O緩衝器 416用來暫存輸出信號。因為輸出正反器414的節點只連"接 至I/O緩衝器416,因此本發明之優點是從正反器414輪出 之AD信號可直接由丨/^緩衝器““專送到”丨匯流排。 圖5是圖4電路運作之時序圖。該波形還包含本發明較 佳實施例之PC I寫入動作。類似圖2之配置,本圖上半部顯 不主控端的内部信號的波形,而下半部顯示由主控端輸出 至P CI匯流排之輸出信號的波形。需注意到,本發明的内 部信號提前一個脈衝週期將資料輸出至正反器4丨4,即比 在PC I匯流排之實際AD動作快一個週期。在本實施例中, 位址相位選擇信號4 〇 9為1時,多工器41 2輸出位址資料, 而當資料傳送選擇信號4〇1為〇時,多工器413輸出下一資 料。以下說明每個週期之各信號動作,但省略第一週期之 說明。 在第二脈衝週期時,位址相位選擇信號4 0 9為1、且位 址缓衝器411輸出位址人』·、以及資料傳送選擇信號4〇1為 1。因此多工器412根據位址相位選擇信號409輸出位址A』 到多工器413,同時多工器413根據資料傳送選擇信號4 01 將位址Α〗輸出至正反器414。故在此週期中,pre-AD信號 402為位址A;。5. Description of the invention (7) Pre-AD signal 4 02 is produced. The flip-flop 414 and the I / O buffer 416 are used to temporarily store output signals in the present invention. Because the node of the output flip-flop 414 is only connected to the I / O buffer 416, the advantage of the present invention is that the AD signal output from the flip-flop 414 can be directly sent to the buffer "" "Bus. Figure 5 is a timing diagram of the operation of the circuit of Figure 4. The waveform also includes the PC I write action of the preferred embodiment of the present invention. Similar to the configuration of Figure 2, the upper part of this figure shows the internal of the master. The waveform of the signal, and the lower half shows the waveform of the output signal output from the master to the P CI bus. It should be noted that the internal signal of the present invention outputs data to the flip-flops 4 丨 1 in advance, that is, One cycle is faster than the actual AD action on the PC I bus. In this embodiment, when the address phase selection signal 4 09 is 1, the multiplexer 41 2 outputs the address data, and when the data transmission selection signal 4 0 When 1 is 0, the multiplexer 413 outputs the next data. The following describes the operation of each signal in each cycle, but the description of the first cycle is omitted. In the second pulse cycle, the address phase selection signal 4 0 9 is 1. And the address buffer 411 outputs the address person ", and the data transmission option The signal 4〇1 is 1. Therefore, the multiplexer 412 outputs the address A ′ to the multiplexer 413 according to the address phase selection signal 409, and at the same time, the multiplexer 413 outputs the address A to the positive signal according to the data transmission selection signal 4 01. Inverter 414. Therefore, in this period, the pre-AD signal 402 is the address A ;.
J 在第三脈衝週期時,位址相位選擇信號4 〇 9為〇且資料 傳送選擇信號4 0 1為1,同時目前資料為。多工器41 2根 據位址相位選擇信號4 〇 9輸出目前資料h到多工器41 3,同 時多工器413根據資料傳送選擇信號4〇1將目前資料Dj輸出J In the third pulse period, the address phase selection signal 4 0 9 is 0 and the data transmission selection signal 4 0 1 is 1, and the current data is 1 at the same time. The multiplexer 41 2 outputs the current data h to the multiplexer 41 3 according to the address phase selection signal 4 〇 9, and at the same time the multiplexer 413 outputs the current data Dj according to the data transmission selection signal 401.
第10頁 五、發明說明(8) 至正反=414。故在此週期中,pre_A])信號4〇2前 h,而AD信號為位址Aj。 月J貝枓 *屑在第四脈衝週期肖,位址相位選擇信號409為G,且資 =送選擇信號401為1,同時目前資料為Dj。多工器412貝 址相位選擇信號4()9輸出目前資 多 Π多工器413根據資料傳送選擇信號4〇/將目前資料。輸 :至正反器414。故在此週期中…D信號4。2為目前i 枓Dj,而AD信號為目前資料Dj。 貝 在第五週期時’位址相位選擇信號4〇 9為〇,且由於 m號421和TRDY#信號422皆為〇,故 為〇 ’同時下一資料為‘。多卫器413根據資料擇傳^送 =信號4Q1將下—資料心輸出至正反器⑴。故在此週运 資料D. P/e_AM5谠4〇2為下一資料Dj+1 ,而〇信號為目前 、受遥ii六週期時’位址相位選擇信號為〇且資料傳送 ==號401為0,同時下-資料為^2。多玉器413根據資 ,傳迗選擇信號401將下一資料Dj+2輸出至正反器414。故 =此,期中,Pre_AD信號4〇2為下一資料d出,而Μ信號為 下一貪料Dj+1。在第七週期時,AD信號為丁一資料仏。 因此,從圖5即可清楚看到,本發明之主控端^&將… 仏號驅動至p c I匯流排的延遲時間明顯縮短。 圖6顯示本發明應用於受控端之輸出控制邏輯的功能 方塊圖。如該圖所示,受控端之結構相似於主控端之結 構,除了不具有位址信號控制電路。因此,圖6的資料傳 495673 五、發明說明(9)Page 10 V. Description of the invention (8) To positive and negative = 414. Therefore, in this period, the pre_A]) signal is h before 402, and the AD signal is the address Aj. Month J * The chip is in the fourth pulse period, the address phase selection signal 409 is G, and the data selection signal 401 is 1, and the current data is Dj. The multiplexer 412 selects the phase selection signal 4 () 9 to output the current data. The multiplexer 413 transmits the selection signal 40 / to the current data according to the data. Input: to flip-flop 414. So in this period ... D signal 4.2 is the current i 枓 Dj, and AD signal is the current data Dj. In the fifth cycle, the 'address phase selection signal 409 is 0, and since both the m-number 421 and the TRDY # signal 422 are 0, it is 0' and the next data is'. The multi-monitor 413 selects the transmission according to the data = signal 4Q1 outputs the data-data core to the flip-flop ⑴. Therefore, in this week, the data D. P / e_AM5 谠 4〇2 is the next data Dj + 1, and the 〇 signal is the current, remote ii six-period 'address phase selection signal is 〇 and data transmission == No. 401 Is 0, meanwhile the data is ^ 2. The multi-jade 413 transmits the next data Dj + 2 to the flip-flop 414 according to the data, and transmits the selection signal 401. Therefore = this, during the period, the Pre_AD signal 402 is the next data d, and the M signal is the next data Dj + 1. At the seventh cycle, the AD signal is a data frame. Therefore, it can be clearly seen from FIG. 5 that the delay time of the master control terminal ^ & FIG. 6 shows a functional block diagram of the output control logic applied to the controlled terminal according to the present invention. As shown in the figure, the structure of the controlled terminal is similar to that of the master, except that it does not have an address signal control circuit. Therefore, the data transmission of Figure 6 495673 V. Description of the invention (9)
送機構的電路,是類似於圖4的電路。需注意到,主控端 傳送讀取命令時,受控端將對應資料驅動至PCI匯流^, 因此傳送端可以是主控端或受控端。本發明亦可應用在具 有相同匯流排協定之控制器。 圖7是圖6的電路運作之時序圖。該時序圖是關於主控 端向受控端發布讀取指令之情形。在第四脈衝週期,當^ 控端驅動位址或命令於P C I匯流排時,受控端6 〇 〇會對位址 或命令解碼。當受控端600確認為’’讀取”命令時,會在第 四脈衝週期之後開始準備資料。以下分別說明各週期之動 作。 " ^ 在第四脈衝週期時,資料傳送選擇信號6 0 1為1,且目 則資料為Dj,下一資料為Dj+1。因此,多工器6丨}根據資料 傳送選擇信號601將目前資料Dj輸出至正反器612。故在此 週期中,Pre-AD信號602為目前資料Dj ,而…信號不需 要。 a次在第五脈衝週期時,資料傳送選擇信號6 0 1為0,且目 :貝料為Dj,下一資料為Dj+1。因此,多工器611根據資料 迗選擇信號6〇1將下一資料])j+i輸出至正反器612。故在 週期中,Pre —AD信號6〇2為下一資料,而ad信號為資 /第六脈衝週期時,主控端設定IRDY#621致能,資料 )迗選擇信號601為1,且目前資料為,下一資料為 料D因此,多工器611根據資料傳送選擇信號601將目前 、/叶m輸出至正反器612。故在此週期中,pre —AD信號The circuit of the sending mechanism is similar to the circuit of FIG. 4. It should be noted that when the master sends the read command, the controlled end drives the corresponding data to the PCI bus ^, so the transmitting end can be the master or the controlled end. The present invention can also be applied to controllers having the same bus protocol. FIG. 7 is a timing diagram of the operation of the circuit of FIG. 6. This timing diagram is about the situation where the master sends a read instruction to the slave. In the fourth pulse period, when the control terminal drives the address or command on the PCI bus, the control terminal 600 decodes the address or command. When the controlled end 600 confirms the "read" command, it will start to prepare the data after the fourth pulse period. The operation of each period will be described separately below. ^ During the fourth pulse period, the data transmission selection signal 6 0 1 is 1, and the data is Dj, and the next data is Dj + 1. Therefore, the multiplexer 6 丨} outputs the current data Dj to the flip-flop 612 according to the data transmission selection signal 601. Therefore, in this cycle, The Pre-AD signal 602 is the current data Dj, and the ... signal is not required. A time at the fifth pulse period, the data transmission selection signal 6 0 1 is 0, and the purpose is: Dj is the shell material, and Dj + 1 is the next data. Therefore, the multiplexer 611 outputs the next data according to the data selection signal 601]) j + i to the flip-flop 612. Therefore, in the cycle, the Pre-AD signal 602 is the next data, and ad When the signal is the data / sixth pulse period, the master sets IRDY # 621 to enable, data) 迗 The selection signal 601 is 1, and the current data is, the next data is material D. Therefore, the multiplexer 611 selects according to the data transmission The signal 601 outputs the current and / m to the flip-flop 612. Therefore, in this period, the pre-AD signal
五、發明說明(10) 602為目前資料^,而AD信號為資抑。 在第七脈衝週期時,主批嫂θ 資料傳送選擇作=ni\n不設定IR_21致能, AD 。L 虎6〇1為〇,且目前資料為Dj+1,下一資料 1 i\D ^多工器6U根據資料傳送選擇信號601將下 B〇2\ T+:^ " ^Α: V2 ; ^ ^ ^ ^ P.e-AD ^ 勹卜 貝枓Dj+2,而AD信號為資料D 。 號為iiC脈衝週期時,Pre_AD信號二無作用,而〇信 送延2圖5至圖7即可了解到’本發明的較佳實施例,傳 要來的:間D4及D5比習知圖2所示之傳送延遲時間D2及D3 受术的短。此外,本發明的趟伟每 所示之電路複雜。例電路不會如圖3 圍,=i本發明參考實施例,並不因此限定本發明之範 Ξ明請專利範圍覆蓋,所有修正在不背離本 月之精神與料下’該行業者可進行各種變形或變更。V. Description of the invention (10) 602 is the current information ^, and the AD signal is the information suppression. In the seventh pulse period, the main batch 嫂 θ data transmission is selected as = ni \ n without setting IR_21 to enable, AD. L tiger 6〇1 is 0, and the current data is Dj + 1, the next data 1 i \ D ^ multiplexer 6U will send B〇2 \ T + according to the data transmission selection signal 601: ^ " ^ Α: V2; ^ ^ ^ ^ Pe-AD ^ 勹 卜贝 枓 Dj + 2, and the AD signal is data D. When the number is the iiC pulse period, the Pre_AD signal has no effect, and the 0-letter transmission delay 2 can be seen in FIG. 5 to FIG. 7 'The preferred embodiment of the present invention. The transfer delay times D2 and D3 shown in 2 are short. In addition, the circuit shown in the present invention is complicated. The example circuit will not be as shown in Figure 3, = i the reference embodiment of the present invention, which does not limit the scope of the present invention. Please cover the scope of the patent. Various deformations or changes.
第13頁 495673 圖式簡單說明 圖1是習知主控端輸出控制邏輯之功能方塊圖。 圖2是描述圖1習知主控端實際操作之時序圖。 圖3是習知主控端的另一輸出控制邏輯之功能方塊 圖。 圖4是本發明應用於主控端之輸出控制邏輯的功能方 塊圖。 圖5是描述圖4實際操作的電路之時序圖。 圖6是本發明應用於受控端之輸出控制邏輯的功能方 塊圖。 圖7是描述圖6實際操作的電路之時序圖。 【圖式編號】 400〜主控端 401、601〜資料傳送選擇信號 402 、 602〜Pre-AD 信號 403、 603〜下一資料 404、 604〜目前資料 4 0 5〜位址 406、6 0 6〜下一相位資料指標 4 0 7、6 0 7〜目前相位資料指標 4 0 8〜目前相位位址指標 4 0 9〜位址相位選擇信號 410、610〜2R1W資料缓衝器 411〜1R1W位址緩衝器Page 13 495673 Brief description of the diagram Figure 1 is a functional block diagram of the output control logic of the conventional master control terminal. FIG. 2 is a timing diagram describing the actual operation of the master control terminal of FIG. 1. Fig. 3 is a functional block diagram of another output control logic of the conventional master control terminal. Fig. 4 is a functional block diagram of the output control logic applied to the main control terminal according to the present invention. FIG. 5 is a timing diagram illustrating the actual operation of the circuit of FIG. 4. Fig. 6 is a functional block diagram of the output control logic applied to the controlled terminal according to the present invention. FIG. 7 is a timing diagram describing the actual operation of FIG. 6. FIG. [Pattern number] 400 ~ Master 401, 601 ~ Data transmission selection signal 402, 602 ~ Pre-AD signal 403, 603 ~ Next data 404, 604 ~ Current data 4 0 5 ~ Address 406, 6 0 6 ~ Next phase data index 4 0 7, 6 0 7 ~ Current phase data index 4 0 8 ~ Current phase address index 4 0 9 ~ Address phase selection signal 410, 610 ~ 2R1W data buffer 411 ~ 1R1W address buffer
第14頁 495673 圖式簡單說明 4 1 2 、41 3、6 11〜多工器 414、612〜正反器 41 5、6 1 3〜或閘 416、417 > 418、614、615、616 〜I/O 緩衝器 420 、 620〜AD[31:0] 421、 621〜IRDY# 信號 422、 622〜TRDY# 信號 60 0〜受控端Page 14 495673 Brief description of the diagram 4 1 2 、 41 3, 6 11 ~ Multiplexer 414, 612 ~ Flip-flop 41 5, 6 1 3 ~ OR Gate 416, 417 > 418, 614, 615, 616 ~ I / O buffers 420, 620 ~ AD [31: 0] 421, 621 ~ IRDY # signal 422, 622 ~ TRDY # signal 60 0 ~ controlled end
第15頁Page 15
Claims (1)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW89123934A TW495673B (en) | 2000-11-10 | 2000-11-10 | Method and apparatus for transmitting registered data onto PCI bus |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW89123934A TW495673B (en) | 2000-11-10 | 2000-11-10 | Method and apparatus for transmitting registered data onto PCI bus |
Publications (1)
Publication Number | Publication Date |
---|---|
TW495673B true TW495673B (en) | 2002-07-21 |
Family
ID=21661923
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW89123934A TW495673B (en) | 2000-11-10 | 2000-11-10 | Method and apparatus for transmitting registered data onto PCI bus |
Country Status (1)
Country | Link |
---|---|
TW (1) | TW495673B (en) |
-
2000
- 2000-11-10 TW TW89123934A patent/TW495673B/en not_active IP Right Cessation
Similar Documents
Publication | Publication Date | Title |
---|---|---|
TWI257550B (en) | Method and apparatus for local and distributed data memory access (""DMA"") control | |
US5327121A (en) | Three line communications method and apparatus | |
TW200842592A (en) | Improved DMAc to handle transfers of unknown lengths | |
US5379386A (en) | Micro channel interface controller | |
US7502377B2 (en) | PCI to PCI express protocol conversion | |
JPS62210560A (en) | Dma transmission method and apparatus for data collection system | |
US20100131692A1 (en) | Bus bridge apparatus and bus bridge system | |
US5388223A (en) | 1-bit token ring arbitration architecture | |
Bainbridge et al. | Asynchronous macrocell interconnect using MARBLE | |
TW201106669A (en) | Interface method for data TX/RX system using data stream | |
US20040221075A1 (en) | Method and interface for improved efficiency in performing bus-to-bus read data transfers | |
JPS6057454A (en) | Data transmission system and operation thereof | |
EP1145132A2 (en) | Improved inter-device serial bus protocol | |
EP2819019A1 (en) | Latency probe | |
TW495673B (en) | Method and apparatus for transmitting registered data onto PCI bus | |
US5418930A (en) | Circuit for interfacing asynchronous to synchronous communications | |
KR100924989B1 (en) | Transfer acknowledgement for a mobile scalable linkmsl architecture | |
CN1909434B (en) | Data bus mechanism for dynamic source synchronized sampling adjustment | |
US5590372A (en) | VME bus transferring system broadcasting modifiers to multiple devices and the multiple devices simultaneously receiving data synchronously to the modifiers without acknowledging the modifiers | |
US7549074B2 (en) | Content deskewing for multichannel synchronization | |
TW457420B (en) | Single-step debugging card device applied in PCI interface and the method thereof | |
TW381222B (en) | Dynamic retry implementation for the PCI bus based on posted transactions on the PCI bus | |
US20020112106A1 (en) | Backplane physical layer controller with an internal bus reset | |
JPH10150522A (en) | Host interface circuit for image-forming device | |
TW548544B (en) | Method for single-step interruption debug for PCI bus cycle and the device thereof |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
GD4A | Issue of patent certificate for granted invention patent | ||
MM4A | Annulment or lapse of patent due to non-payment of fees |