TWI261968B - Shifting apparatus for bit-shifting manipulation in a digital processor device - Google Patents
Shifting apparatus for bit-shifting manipulation in a digital processor device Download PDFInfo
- Publication number
- TWI261968B TWI261968B TW90123122A TW90123122A TWI261968B TW I261968 B TWI261968 B TW I261968B TW 90123122 A TW90123122 A TW 90123122A TW 90123122 A TW90123122 A TW 90123122A TW I261968 B TWI261968 B TW I261968B
- Authority
- TW
- Taiwan
- Prior art keywords
- displacement
- movement
- amount
- bit
- information
- Prior art date
Links
Landscapes
- Executing Machine-Instructions (AREA)
Abstract
Description
12619681261968
五、發明說明( 發明之範圍 曰本毛月大致係有關於_種數位資料處理裝置,特別 疋有關於可在數位邏輯裝置,諸如半導體數位邏輯元件 之中進订有彈性且有效率的,資料字元(data w〇rds) 的位元位移操作(b i t · s h丨f t丨η㊂〇 p e r & t丨〇 n $ )的一種位 元位移處理裝置。 請 先 閱 讀 背 面 之 注 經濟部智慧財產局員工消費合作社印製 發明之背景 貝料子元在邏輯電路之中的位元位移(bit shifting) 動作,是為各種數位資訊處理裝置,諸如半導體數位資 訊處理元件(semiconduct〇r digital inf〇rmati〇n processing devices)之中,所必須進行的一種重要的 基礎性資料處理操作。此些數位元件典型地包含有微控 制 器 (microcontroller) , 微 處 理 器 (microprocessor), 與各種微電子裝置 (microelectronic devices)所需要的微電腦 (microcomputer),以及數位信號處理器(DSP, Digital Signal Processor),與迷你及主電腦(mini_ and mainframe computers)系統之中的中央處理器 等。 當進行程式規劃(programming)以控制此些處理器 裝置的操作時,所有的南階程式語言(h i g h -1 e v e 1 programming language),皆會被編譯(compile)或 被解譯(interpret)成為低階的機器語言(machine \\BMTPENT002\CHlNESE\ching\b08054a.doc 項再填寫本頁)V. INSTRUCTIONS OF THE INVENTION (The scope of the invention 曰 This month is roughly related to _ kinds of digital data processing devices, especially for the flexible and efficient data can be customized among digital logic devices, such as semiconductor digital logic elements. A bit shift processing device for the bit shift operation of the character (data w〇rds) (bit · sh丨ft丨η三〇per & t丨〇n $ ). Please read the note on the back of the Ministry of Economic Affairs. The employee shift cooperatives print the background of the invention. The bit shifting action in the logic circuit is for various digital information processing devices, such as semiconductor digital information processing components (semiconductor digital inf〇rmati〇n An important basic data processing operation that must be performed in processing devices. Such digital components typically include a microcontroller, a microprocessor, and various microelectronic devices. Required microcomputer, and digital signal processor (DSP, Digital Si Gnal Processor), with the central processing unit in the mini and mainframe computers system, etc. When programming to control the operation of these processor devices, all the southern programming languages (high -1 eve 1 programming language), will be compiled (compile) or interpreted (interpret) into a low-level machine language (machine \\BMTPENT002\CHlNESE\ching\b08054a.doc and fill this page)
-----Γ---:訂---------線 I‘-----Γ---:Book---------Line I‘
V 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公爱) 1261968 A7 B7 五、發明說明( 經濟部智慧財產局員工消費合作社印製 language)。當一組程式直 妓田力y ^ 破執仃時,機器碼乃是直 接用采操控處理為動作的指八 上 7。在諸如 FORTRAN莖 的高階語言之中,與位元位移有關的,對資料 進 :的複雜操作指令’可以在單獨一行的指令之 。 當此些高階指令行被編譯成為哭 70 擁π任_ 士 > …、機σ口馬,亦即硬體階層的 機為ό口 § k,在此歧數付声田壯 (. -數位處理裝置的機器語言指令集 (instruction a -, )之史的位凡位移指令,通常可以被 區分為兩種型態。在1第一錄 t 牡/、弟種型恶之中,位移的方向是 由指令簡字碼(i n s t r u c t i ( Ctlon mnem〇nic)直接指明的, 而其第二種型態的指令’則是使用由其所處理於其上的 异量(operand)的正負號位元(sign bu),以動態方式 决疋的’間接指明式的位移方向指令。 在第種型恶的位元位移指令之中,其指令的執行 牽涉到至少一個算量的處理操作,該至少一個算量係指 月所要私動的位元位置數(am〇unt 〇f bit places)。 其位移的方向,不論是向左或向右,皆是由指令簡字碼 本身明確標示出來。 例如’ 一個典型的組合語言指令係如下列:SHIFT-RIGHT ACC,R0 此位移指令的簡字碼“5///厂7-穴/6//厂,明確地標示出, 資料字元將在d CC記錄器之中,被向右移動由記錄器 及〇的内容所指定的位元位置數。如此,下面的一個類 3 WBMTPENT002\CHINESE\ching\b08054a.doc (請先閲讀背面之注意事項再填寫本頁)V This paper scale applies to China National Standard (CNS) A4 specification (210 X 297 public) 1261968 A7 B7 V. Invention description (Ministry of Commerce, Intellectual Property Bureau, Consumer Cooperatives, printed language). When a group of programs is straightforward, the machine code is directly used to control the action as the finger. In higher-order languages such as FORTRAN stems, the complex manipulation instructions associated with bit shifts can be on a single line of instructions. When these high-order instruction lines are compiled into crying 70 π 任 任 士 士 士 士 士 士 士 士 士 士 士 士 士 士 士 士 士 士 士 士 士 士 士 士 士 士 士The displacement instruction of the history of the machine language instruction set (instruction a -, ) of the processing device can usually be divided into two types. In the first recording of the t-mu, the younger type, the direction of displacement It is directly indicated by the instruction short code (instructi ( Ctlon mnem〇nic), and the second type of instruction ' is the sign bit of the operand (sign) used by it. Bu), the indirect specified displacement direction command in a dynamic manner. In the first type of bit shift instruction, the execution of the instruction involves at least one arithmetic processing operation, the at least one calculation Refers to the number of bit positions (am〇unt 〇f bit places) that are to be privately moved in the month. The direction of the displacement, whether left or right, is clearly indicated by the instruction short code itself. For example, 'a typical The combined language command is as follows: SHIFT-RIGHT ACC, R0 The shorthand code for this displacement command "5///factory 7-hole/6//factory, clearly indicates that the data character will be in the d CC recorder and moved to the right by the contents of the recorder and the device. The specified number of bit positions. Thus, the following class 3 WBMTPENT002\CHINESE\ching\b08054a.doc (please read the notes on the back and fill out this page)
本紙張尺度適用中國國家標準(CNs)A4規格(210 X 297公釐This paper scale applies to the Chinese National Standard (CNs) A4 specification (210 X 297 mm)
-n 1 J,* n ):OJ ϋ ϋ 1 I n 1 ϋ I n ϋ a^i ϋ a^i n ϋ n ϋ ϋ I n n 1 ϋ n I 1261968 A7 五、發明說明() 似的指令 經濟部智慧財產局員工消費合作社印製 SHIFT-LEFT ACC #5 便會將J CC記錄器之中的資料字元向左移動5個位元 位置。 在此種型悲的位移指令之中,代表位元位移量的數 總是被當成是正數。換句話說,代表位移量的記錄 裔的内容,總是被當成是正數來處理。不過,在某 坠It况之下,此種正數值位移量的規定,有時候是會造 成不便的。例如,在牽涉到浮點運算的資料處理應用之 中,浮點資料的指數部份,時f會被利用來代表資料字 =位私备作之中的位移位置量。浮點資料的此個指數部 伤,通常是為某種計算程序所獲致的結果,但該些運算 處理,仍然是可能得出負數指數部份的結果的。在出現 負值指數結果時,程式的執行便可能會遭遇到不可預 期,且可能是錯誤的結果。 旦圖式之中,圖丨顯示能夠以指定的位元位置位移 里’朝=指定方向執行資料字元位移操作的,-種習知 之位私☆。方塊圖之中,以功能方塊所大致標示的一個 又向位私為1 1〇為基礎的此一位移器,具有一個目的 !輸入端111 ’可供接受將被進行位移處理的目的資料 :兀7/立移方向輸入端112可接受由圖中所未顯示 、可解#位移指令,並據以產生出位移方向信號的其 ____ —___ 4 WBMTPENT002\CHINESE\ching\b08054a.doc 本紙張尺度適种gi^^CNs)A4規^.挪 (請先閱讀背面之注意事項再填寫本頁)-n 1 J,* n ):OJ ϋ ϋ 1 I n 1 ϋ I n ϋ a^i ϋ a^in ϋ n ϋ ϋ I nn 1 ϋ n I 1261968 A7 V. Description of invention () The Intellectual Property Bureau employee consumption cooperative printed SHIFT-LEFT ACC #5 will move the data characters in the J CC recorder to the left by 5 digits. In this type of sad displacement command, the number representing the displacement of the bit is always treated as a positive number. In other words, the content of the recorded representative of the displacement is always treated as a positive number. However, under certain conditions, such a positive numerical displacement is sometimes inconvenient. For example, in a data processing application involving floating-point operations, the exponential portion of the floating-point data, time f, is used to represent the amount of displacement position in the data word = bit private. This index of floating-point data is usually the result of a calculation program, but the processing of these operations is still likely to result in a negative index. In the event of a negative index result, the execution of the program may encounter unpredictable and may be the result of an error. In the figure, the figure shows that it is possible to perform the data character shift operation in the specified direction of the position shift in the direction of the specified bit position. In the block diagram, the one of the shifters, which is roughly indicated by the function block, is based on the 1 1〇, and has a purpose! The input terminal 111' can accept the purpose data to be subjected to the displacement processing: The 7/verting direction input terminal 112 can accept the ____ 4 WBMTPENT002\CHINESE\ching\b08054a.doc paper scale which is not shown in the figure and can be solved by the # displacement command and accordingly generates the displacement direction signal. Suitable species gi^^CNs) A4 rules ^. Move (please read the notes on the back and fill out this page)
I I I ΙΊ I I J ^ ·11111111 」I I I ΙΊ I I J ^ ·11111111 ”
1261968 A7 五、發明說明( 他迷輯電路其所發送出來的位移方向控制信號。其第 三個輸入端是為位移量輸入端113,其可以接受用以決 疋以位兀位置夏為計算基礎而表示位移總量的資料。另 有一位移貝料輸出端丨丨4可將經過位移處理後的資 料’傳遞給系統之中的其他邏輯電路部份。(主意到目的 碼輸入端1U,位移後資料輸出端114,以及位移量輸 入端1 1 3 ’各皆是為多位元的輸入端,而位移方向輸入 端1 1 2則可為單位元的輸入端。 在第一種型態的位移指令之中,與位元位移處理有 關的位移:向,並不是由指令簡字碼明確標出,而 是由位移貧訊算量的正負號位元來決定。作為一個 實例,考慮下面的指令: SHIFT ACC, R〇 其中的單純簡字碼指令“如斤,只是簡單地將記錄器 』C?中所存放的貝料字元,朝向由正負號位元,即記 錄器L的内容之最高效位元(MSB)所決定的方向,移 動-個數量的位元位置。上例中的指數攔内的正值正負 號位元,依據-般習慣,係被用來標示向左方向的位 移’而負值則是代表向右的位移。記錄器&内容的絕 對值’則決定以位元位置量所表示的位移總量。 圖2為一方塊圖,其顯示能以指定的位元位置位移 量朝向-個動態指定的方向執行資料字元位移操作的, _____5 ^BMTPENT002CHINHSEWhing\b08054ad〇c 本紙張尺度適用中國國家標準(CNS)A4規格 C請先閲讀背面之注意事項再填寫本頁)1261968 A7 V. Description of the invention (There is a displacement direction control signal sent by the circuit. The third input is the displacement input terminal 113, which can be used to determine the position based on the position of the summer. And the data indicating the total displacement. Another displacement bead output 丨丨4 can transmit the displaced data to other logic circuits in the system. (The idea is to the destination code input 1U, after displacement The data output end 114 and the displacement amount input end 1 1 3 ' are each an input end of the multi-bit, and the displacement direction input end 1 1 2 can be an input end of the unit element. Among the instructions, the displacement related to the bit displacement processing: the direction is not explicitly indicated by the instruction short code, but is determined by the sign bit of the displacement poor calculation. As an example, consider the following instructions: SHIFT ACC, R〇 The simple simple code command “such as Jin, simply simply put the beetle character stored in the recorder” into the most efficient bit of the content of the recorder L. Yuan (MSB) The determined direction, moving - the number of bit positions. The positive and negative bits in the index bar in the above example are used to indicate the displacement in the left direction according to the general habit, while the negative value is Represents the displacement to the right. The absolute value of the recorder & content determines the total amount of displacement represented by the amount of position of the bit. Figure 2 is a block diagram showing the amount of displacement at a specified bit position Dynamically specified direction to perform data character shift operation, _____5 ^BMTPENT002CHINHSEWhing\b08054ad〇c This paper size applies to China National Standard (CNS) A4 specification C. Please read the back note before filling in this page)
經 濟 部 智 慧 財 產 局 員 工 消 費 合 社 印 製 1261968 A7Ministry of Economic Affairs, Zhihui Property Bureau, Staff and Consumers Association, Printing 1261968 A7
第二種型態的習知位移器。此位移器2GG,可以將資料 字兀’朝向-個動態決定的方向,移動一個指定的位移 量。此種型態的位移器200,可以利用與圖i中丨1〇 相似的基本雙向位移器21〇為基礎來進行建構。一個7 之補數/旁通邏輯單元(2,s c〇mplement/bypa = unit) 220,可在其位移f訊輸人端22i上搁截 包含有位移量及位移方向兩種資訊的資料。依據其所收 到的位移資訊,2之補數/旁通邏輯單元22〇可在其位 移量輸出端222上產生出—個位移量資料,以供輸出 至基本位移器210的位移量輸入端213。邏輯單元 220亦在其位移方向輸出端223上產生出—個位移方 向k號,並將之傳遞至基本位移器2丨〇的位移方向輸 入端212。與圖i中的情況類似的,將要被進行位移: 作處理的目的碼,便直接被輸入至基本位移器的 目的碼輸入端2 1 1。 不過,其中仍可能發生所欲進行位移的方向血位移 指令的簡字碼與/或目的碼所標示者相反的情況。作為 -個例子m程纽計者,為了諸如程式執行效率 與/或執行彈性的考量,可能會想要將位移指令中的左 右位移方向的定義反轉過來。換句話說,程式設計者$ 能想要將一個正值的目的碼及",重新定義為代表向右 方向的位移,並將負值定義為代表向左方向的位移。在 f種情況之下,若採用先前描述第二種型態的習知位移 指令,則目的碼的數值,事先便必須預先予以逆 L____ 6 ^MTPENT002\CH,NESEWhingyb08054a 本紙張尺度適用中國國家標準(CNS)A4規格(210 χ 297公釐) ---------, t---------線 (請先閱讀背面之注意事項再填寫本頁)The second type of conventional shifter. The shifter 2GG can move the data word 兀 ' toward a dynamically determined direction by a specified amount of displacement. This type of displacement device 200 can be constructed based on a basic bidirectional displacement device 21 similar to that of Fig. i. A 7's complement/bypass logic unit (2, s c〇mplement/bypa = unit) 220 can intercept data containing both displacement and displacement directions on its displacement f-transport terminal 22i. Based on the displacement information received, the 2's complement/bypass logic unit 22 can generate a displacement amount data on the displacement output terminal 222 for output to the displacement input terminal of the basic shifter 210. 213. The logic unit 220 also generates a displacement direction k number at its displacement direction output end 223 and transmits it to the displacement direction input end 212 of the basic shifter 2A. Similar to the situation in Figure i, the displacement is to be made: the destination code for processing is directly input to the destination code input 2 1 1 of the basic shifter. However, it may still occur that the shorthand code of the direction blood displacement command to be displaced is opposite to the one indicated by the destination code. As an example, the m-meter can be used to reverse the definition of the left-right displacement direction in the displacement command for reasons such as program execution efficiency and/or execution flexibility. In other words, the programmer $ can want to redefine a positive value of the object code and " to represent the displacement to the right and a negative value to represent the displacement to the left. In the case of f, if the conventional displacement command of the second type described above is used, the value of the destination code must be reversed in advance. L^6^MTPENT002\CH, NESEWhingyb08054a This paper scale applies to the Chinese national standard ( CNS) A4 size (210 297 297 mm) ---------, t--------- line (please read the notes on the back and fill out this page)
V -I n H ϋ 1261968 Λ7 B7 五、發明說明( 轉 個 下面是執行此程序的 negate r0 shift ACC, R〇 指令步驟短程序: 當記錄器^的内容之正負浐 3 令係將該正負號位元加以反轉,=負料,第-個指 已反轉的方向而進行位移動 二:指令則接著依據 知者’與單一個指令相 。如冋本技藝中所廣為習 ^ η π 下,兩個指令會需要兩仵的 執::間,此外’程式的儲存記憶空間亦會增加的 相較之下,若上列的兩步、、 下之單一指令·· S 7可以被減縮成為如V -I n H ϋ 1261968 Λ7 B7 V. Invention description (The following is the negate r0 shift ACC for executing this program, R〇 instruction step short program: When the content of the recorder ^ is positive or negative 浐 3 is the sign The bit is inverted, = negative, the first refers to the direction that has been reversed and the bit is moved. 2: The instruction is then based on the 'information' with a single instruction. For example, the technique is widely used in the art. η π Next, the two instructions will require two :::, in addition, the program's storage memory space will also increase, if the two steps above, the single instruction, S 7 can be reduced Become like
RR
shift ACC, ::程:的彈性以及程式的執行效率便可以獲得改善。另 卜=於種簡化’執行程式所需要的記憶儲存 可以獲致節約。 發明之概要 因此,本發明之一目的即在於提供一種位移器裝 置叩,、可以有彈性且有效率的方式,在處理器裝置 的早-個機器指令週期之中,執行方向反轉的位元 位移操作。 WBMTPENT002\CHINESE\ching\b08054a.doc 本紙張尺度適用中國國家標準(CNS)A4規格(21〇 297公釐) 1261968 Λ7 B7Shift ACC, ::C: The flexibility of the program and the execution efficiency of the program can be improved. In addition, it is possible to save money by simplifying the memory storage required to execute the program. SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a shifter device that can perform a direction-reversed bit in an early-machine command cycle of a processor device in a flexible and efficient manner. Displacement operation. WBMTPENT002\CHINESE\ching\b08054a.doc This paper scale applies to China National Standard (CNS) A4 specification (21〇 297 mm) 1261968 Λ7 B7
本發明之另-目的係在 可以有彈性且有效率的方式種位移為裝置,其 器指令進行程式規劃,以^行=理器裝置的單一行相 作。 向反轉的位元位移操 本發明之又-目的係在於提供—種位 可相容於處理器裝置的指 α扁置,其 向定義的位移指令。 木之中所包含的明示固定方 為達成前述及其他目的,本發明提 卢理:Γ數位處理器農置之中執行位元位移處 =裝置係用以執行-位移指令,以將—資料字元: 兀私動里,其中的移動 二由:移指令的至少-個算量來指定。該 :生出的:個支援邏輯電路,根據該至少-算量而 產生出-位移該,_動態位移方向,與— 轉控制信號。位移|| f f g Π 秒。°衣置包含有一絕對值邏輯單元,可 利用獲取該位移資訊的絕對值而產生該移動量;一 XOR ,其一輸入端接受代表該動態位移方向的一作 :二且其另一輸入端接受位移方向反轉控制信號,並於 其輸出端產生該位移方向;與一雙向位移器,具有一目 的碼輸入端’可接受該資料字元,一移動量輸入端,可 接欠該移動量’與一移動方向輸入端,可接受來自於該 閘之輸出端的該移動方向信號,#中該雙向位移 器將該資料字元朝向該移動方向移動該移動量所指定之 位元位置。 I_ 8 \\BMTPENT002\CHINESE\ching\b08054a.doc 本紙張尺度刺中國國家鮮(CNS)A4祕(21G x 297公)'_Another object of the present invention is to shift the device into a device that can be flexibly and efficiently, and the device commands the program to program a single row of the device. Displacement to Inverted Bits A further object of the present invention is to provide a finger-alignment that is compatible with the processor device, to a defined displacement command. The explicit fixing party included in the wood is to achieve the foregoing and other purposes. The present invention provides a method for performing a bit displacement in the digital processor of the digital processor = the device is used to execute the - displacement command to Yuan: In the private movement, the movement 2 is specified by at least one calculation of the shift instruction. The generated: a supporting logic circuit generates a displacement-displacement, a dynamic displacement direction, and a rotation control signal according to the at least-calculus. Displacement || f f g Π seconds. ° The garment includes an absolute value logic unit, and the movement amount can be generated by obtaining an absolute value of the displacement information; an XOR, one input receiving a representation of the dynamic displacement direction: two and the other input receiving the displacement The direction reversing the control signal, and generating the displacement direction at the output end thereof; and a bidirectional displacement device having a destination code input terminal 'accepting the data character, and a movement amount input terminal, the owing the movement amount' A moving direction input end accepts the moving direction signal from the output end of the gate, and the two-way shifter moves the data character toward the moving direction by the bit position specified by the moving amount. I_ 8 \\BMTPENT002\CHINESE\ching\b08054a.doc This paper scales the Chinese National Fresh (CNS) A4 Secret (21G x 297))_
----^---Ί ^--------- (請先閱讀背面之注音心事項再填寫本頁)----^---Ί ^--------- (Please read the notes on the back and fill in this page)
1261968 A7 五、發明說明( Γ37 濟 消 費 圖式之簡要說明 本發明之前述目的及其他特 圖式而於後面的說明文字之中,配;Λ 在參考所附 的季乂乜貫施例進行詳細說明 疋性貝 解。圖式之中: 田了更易於獲得瞭 圖1顯示能夠以指定的位元位 方向執行資料字元位移操作的-種習知位:哭朝向W 量圖二為7方塊圖,其顯示能以指定的位元位置位移 -種;態指定的方向執行資料字元位移操作的 η · *為方塊圖,其顯示本發明位移裝置之一實施圖4為一方塊圖,其顯示本發明位移 施例;與 < 力貫 圖5顯示可支援方向反轉指令,供⑽用途的本發 <私衣置,其在程式執行的過程中的資料記 作情形。 ^發明之詳細說明 在進行本發明位移器裝置之詳細說明之前,為了本 發明乾疇之清楚界定之目的,有必要再將數位處理裝置 的扣7集,其中之資料字元位移指令予以列明。特定而 吕,能執行單純位移操作以外之其他額外選擇性位移方 (請先閱讀背面之注音?事項再填寫本頁)1261968 A7 V. INSTRUCTIONS INSTRUCTIONS ( Γ 之 之 之 之 之 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要 简要Explain the sloppy Becky. In the figure: The field is easier to obtain. Figure 1 shows the ability to perform the data character shift operation in the specified bit position: the crying direction W is the second figure is 7 squares. FIG. 4 is a block diagram showing that one of the displacement devices of the present invention is implemented as a block diagram, which is a block diagram of the displacement of the specified character position; The displacement example of the present invention is shown; and the force diagram 5 shows that the direction reversal command can be supported, and the data for the purpose of (10) is set, and the data in the process of program execution is recorded as a case. DETAILED DESCRIPTION Before carrying out the detailed description of the displacing device of the present invention, for the purpose of clearly defining the scope of the present invention, it is necessary to further delineate the digital processing device, in which the data character displacement command is specified. Lu, can perform other optional displacements other than the simple displacement operation (please read the phonetic on the back? Please fill out this page again)
--------Ί ^--------- 本紙張反度顧中關家標準(CNS)A4 ⑵G x \\BMTPENT002\CHINESE\ching\b08054a.doc--------Ί ^--------- This paper is counter-measures (CNS) A4 (2)G x \\BMTPENT002\CHINESE\ching\b08054a.doc
I 1261968 A7I 1261968 A7
、發明說明() 向反轉操作的位移指令,當 田以組合浯言的格 時,係如下列: 式加以表示 SHIFT OBJ, SDA} DR 其中 紐π &為4日明貧料位元位移操作的指令簡字 碼。指令的單純簡字碼本身表明了位移 的方向是由指令的算量所決定的。 是為包含有將予以位元位移處理的目的 碼之第-算量。其可為儲存於存有所將 進打位移處理的資料字元之記錄哭 OBJ 〇 疋為傳遞位移相關資訊的第二算量。位 移的量及方向可分別以記錄器内容 的絕對及正負號位元來表示。 是為表明位移處理方向反轉之第三算 量。DR可為正號“ +,,或負號‘、,,。Inventive Note () The displacement command to the reverse operation, when Tian is a combination of rumors, is as follows: Formula to represent SHIFT OBJ, SDA} DR where New π & is the 4th day of the poor material bit displacement The instruction's instruction short code. The simple simple code of the instruction itself indicates that the direction of the displacement is determined by the calculation of the instruction. It is the first calculation amount of the destination code containing the bit displacement processing. It can be a second calculation of the information about the transfer displacement related information stored in the data character in which the displacement processing is performed. The amount and direction of the shift can be represented by the absolute and sign bits of the contents of the recorder, respectively. Is the third measure to indicate the direction of displacement processing. The DR can be a positive sign "+,, or a minus sign ‘,,,.
OBJOBJ
SDASDA
DR (請先閱讀背面之注咅?事項再填寫本頁)DR (Please read the note on the back? Please fill out this page again)
··
經濟部智慧財產局員工消費合作社印製 相對於前述第二種型態的習知位移指令而言,此種 位移指令可以提供額外的程式規劃設計彈性及執行效 率。在適當的設計安排之下,支援包含了上列此種有彈 性位移指令的一個指令集的處理器裝置,其組合語言編 譯器(a s s e m b 1 e r*),同樣亦是可以處理下列之習知指令 10 WBMTPENT002\CHINESE\ching\b08054a.doc ^紙張尺度適用中國國家標準(CNS)A4規格(21〇 X 297公釐) 智 局 員 工 消 印 1261968 五、發明說明( A7 SHIFT ACC, Rn 〇 在此種情況之下,本發明之彈性位移指令袼式可以 SHIFT OBJ,SDA,[DR] 或者為 SHIFT ACC,R〇, [·] 改為 ,:W方格號表示方向反轉算量⑽,或負值符號, _ :透擇性的。例如’組合語言編譯器在 二即方向反轉算量的存在之時,可以將此指 方:反轉的指令。另一方面,若只偵測到兩個算量: ==器便可將此指令當作是不具有選擇性位; 反轉而求的—般習知位移指令來進行處理。 '者’在此有必要須予指出,數位系統之 二疋以其2之補數的形式來儲存的。此可以容許幻 Y系統之中,不論並值為 料進行曾m 即對所有的數值, 種Ή: T 〃’而不需要任何預先進行的處理。在出 理^ 負錢㈣料獲取其絕龍的直接處 驟,乃是獲取其2之補數值。相較之下,為了^ —個正值數值她絕對值,便不需要進行任^ 紙張尺______1 ^BMTPENT002\CHINESE\ching\b08054a.doi ----Μ ^---------線 (請先閱讀背面之注意事項再填寫本頁)Printed by the Intellectual Property Office of the Ministry of Economic Affairs, the Consumer Cooperatives. Compared to the second type of conventional displacement command, this displacement command can provide additional program planning flexibility and execution efficiency. Under appropriate design arrangements, a processor device including an instruction set of such a flexible displacement instruction is supported, and a combination language compiler (assemb 1 er*) is also capable of processing the following conventional instructions. 10 WBMTPENT002\CHINESE\ching\b08054a.doc ^The paper scale is applicable to China National Standard (CNS) A4 specification (21〇X 297 mm). Employees of the Zhi Bureau are printing 1261968. V. Inventions (A7 SHIFT ACC, Rn 〇 In this case) In the following, the elastic displacement command of the present invention can be changed to SHIFT OBJ, SDA, [DR] or SHIFT ACC, R〇, [·], and: W square sign indicates direction reversal calculation (10), or negative value. Symbol, _: Selective. For example, 'combined language compiler can refer to this in the presence of the second direction reversal calculation. Inverted instructions. On the other hand, if only two are detected The calculation: == The device can treat this instruction as a non-selective bit; the reverse-like general-purpose displacement instruction is used for processing. The 'person' needs to be pointed out here, the digital system储存 stored in the form of its 2's complement. This can be In order to allow the illusion Y system, regardless of the value of the material, the m is for all the values, the type Ή: T 〃 ' does not require any pre-processing. In the case of the ^ ^ negative money (four) material to obtain its absolute dragon Directly, it is to obtain the complement value of 2. In contrast, for the positive value of ^ positive value, you do not need to do any paper size ______1 ^BMTPENT002\CHINESE\ching\b08054a.doi - ---Μ ^---------Line (please read the notes on the back and fill out this page)
1261968 五、發明說明() 理’只需取得其原始的數值資❹卩可 建構本發明之位移裝置實施例所需的絕對值邏輯:於 ==)’便可以採用與圖”所描= 知技術2之補數/旁通邏輯單元相似者。 為 參考圖3,其為一方塊圖,其中顯示本發 置之一貫施例。其中,位移裝置300包含有 : 移器31〇,-絕對值邏輯單元32〇,盘—有又向= 咖。此種位移裝置3⑽係適合於多種數位處理穿置^ 用途’可用來為需要進行處理㈣料字元,執行位、 移的操作處理。如同習於本技蓺 之中诉去链-山+ 议-者所熱知的,雖然圖式 的J:# T〜但構成—個完整系統所需整合在内 及週邊邏輯電路,亦可以㈣#地加人於圖 3之位移裝置之中。 口 雙向:移器310可為與圖…中所分別描述之位 夕^ & 21G相類似之習知位移器。與前述圖式之 甲所描述-樣的,將要進行位㈣移處理的目的資料字 疋’係直接被輸人至基本位移器3 1()的目的碼輸入端 川。位移器3 1G亦具有—個位移量輸人端3ι3。由此 輸入端上所接㈣的-餘移量,係制 進行的位移量。位移器310更具有一位移方向輸入; J 12。在此輸入端上所接收到的一個位移方向資訊,係 用以決定位移所必須進行的方向。如此,位移器3置〇 即能夠將所須進行處理的—„料字元,朝向指°定的方 或右方向,移動指定的一個位移量。 12 WBMTPENT002\CHlNESE\ching\b08054a.doc 本紙張尺度適用t國國家標準(CNS)A4規絡(g 297公釐) 1261968 A7 經濟部智慧財產局員工消費合作社印製 五、發明說明() 絕對值邏輯單元32G,係由其位移資訊輸人端32ι 上,接受包含有位移量及方向兩種資訊的資料。根據里 削妾收到的位移資訊,邏輯單元32Q在其位移量輪出 鳊322上產生出一個位移量資料,以供輸入給基本位 移器31G。此位移量資料’係由位移器η。的位移量 輸入立而3 1 3輪入。力麻· [I八μ » 在貝(V、上,位移量必須由可能為正 ,了能,負的位移資訊數值之中抽取出來。不論位移資 訊疋否藏載於正或_信的杳印 夂貝值的貝甙之中,其絕對值皆必須被 抽取出來。 在取可月“x生的情況,亦即’在採用了補數技術來 表不並儲存負值數值資料的數位系統.之中,—負值數值 的絕對值,乃是可以簡單地利 少 门平e们用獲取該負值數值的補數 而取得。相較之下,於此笨备 、寺糸、冼之中,一個正值數值的 、’e、值’很單純的乃是直接等於其數值。除了簡單地令此正值數值旁通通過m何處m卜,要取得正值 數值的絕對值,並不須推) +肩進仃任何的邏輯處理。在正常的 月況之下;^疋2 $丨之補數系統,皆是可以採用 的/在匕圖3中貫施例的絕對值邏輯單元3 2 0便可 以! ’例如’與前述圖2中相類似的-個2之補數/旁 通邏輯單元。 1供應线對值邏輯單元32G之輪人端321的位移資 Λ,如同前述,亦被用央被 用;獲取位移方向。在-個數位系 統之中,其最為直接的—種作法,乃是抽取位移資訊的 Μ氏張d朗巾關家鮮(CNS)A!賴 WBMTPENT002\CHINESE\ching\b08054a.doc (請先閱讀背面之注意事項再填寫本頁)1261968 V. INSTRUCTIONS (REC) 'Requires the absolute value logic required to construct the embodiment of the displacement device of the present invention by simply obtaining its original numerical value: at ==) 'can be used as described in the figure> The complement/bypass logic unit of technique 2 is similar. Referring to Figure 3, which is a block diagram showing a consistent embodiment of the present invention, wherein the shifting device 300 includes: a shifter 31 〇, - an absolute value Logic unit 32 〇, disk - has = 咖. This displacement device 3 (10) is suitable for a variety of digital processing through the use of 'use' can be used for processing (four) material characters, perform bit, shift operation processing. In this technology, I want to go to the chain-mountain + the discussion--the one who knows well, although the figure J:#T~ constitutes a complete system that needs to be integrated into and around the logic circuit, it can also (4)# Adding to the displacement device of Fig. 3. The port bidirectional: shifter 310 can be a conventional shifter similar to the one described in Fig. 2, which is described in the above figure - In the same way, the target data word of the bit (four) shift processing will be directly transmitted. The destination code input terminal of the human to basic shifter 3 1(). The shifter 3 1G also has a displacement amount input end 3ι3. The displacement of the (4) connected to the input end is the displacement of the system. The displacement device 310 has a displacement direction input; J 12. A displacement direction information received at the input end is used to determine the direction in which the displacement must be performed. Thus, the displacement device 3 can be set The material character to be processed, moving toward the specified square or right direction, moves the specified displacement. 12 WBMTPENT002\CHlNESE\ching\b08054a.doc This paper scale applies to national standard (CNS) A4 regulation (g 297 mm) 1261968 A7 Ministry of Economic Affairs Intellectual Property Bureau employee consumption cooperative printing 5, invention description () absolute value The logic unit 32G receives the information including the displacement amount and the direction by the displacement information input terminal 32. Based on the displacement information received by the reticle, the logic unit 32Q generates a displacement amount data on its displacement amount wheel 322 for input to the basic shifter 31G. This displacement amount data is taken from the shifter η. The amount of displacement is entered and the 3 1 3 rounds. Li Ma · [I 八 μ » In the shell (V, upper, the displacement must be extracted from the possible negative, energy, negative displacement information values. Whether the displacement information is contained in the positive or _ letter 杳In the case of the shellfish, the absolute value of the shellfish must be extracted. In the case of the "x", that is, the digital system that uses the complement technique to store and store negative numerical data. Among them, the absolute value of the negative value can be obtained by simply reducing the number of complements of the negative value. In contrast, this is a fool, a temple, a scorpion, a The positive value of 'e, the value' is very simple is directly equal to its value. In addition to simply bypassing this positive value by m, to obtain the absolute value of the positive value, do not have to push + shoulders into any logical processing. Under normal monthly conditions; ^ 疋 2 $ 丨 complement system, can be used / in Figure 3, the absolute value of the logical unit 3 2 0 Yes! 'For example, a 2's complement/bypass logic unit similar to the one in Figure 2 above. 1 Supply The displacement of the wheel terminal 321 of the value logic unit 32G, as described above, is also used by the center; the direction of displacement is obtained. Among the digital systems, the most straightforward method is to extract the displacement information. Μ张张dlang towel Guan Jia Xian (CNS) A! Lai WBMTPENT002\CHINESE\ching\b08054a.doc (please read the notes on the back and fill out this page)
----.---T~訂---------線----.---T~Book---------Line
13 Ϊ261968 A7 B713 Ϊ261968 A7 B7
正負號:元’作為動態的位移方向信號。此可以利用, 例如,簡單地探測正負號位元信號線路的信號情況,便 可=得方向資訊。在圖3之實施例之中,此種作法係 以直接由從位移資訊傳遞至邏輯單元3 20的資料匯流 排之中,取出此動態位移方向資訊的方式,來顯現本發 明此種單純的作法。 不過,若不如同目2之習知技術位移器作法一樣 地,直接送至基本雙向位移器31〇的位移方向輸入端 312 ’則此動態位移方向資訊位元,在被傳送至位移器 31〇的輪入端312之前’亦可先由x〇R閘33〇加以 处 特疋而5,由位移資訊之中所抽取出來的此動態 位移方向資訊’係可由其他的控制信號,ώ XOR閘 3 3 〇進行處理,以便進行調節。 動態位移方向資訊位元係被送至X0R閘330的兩 個輸入端中之-,331。職f甲1 3 3 0的另一個輸入端 3 3 =,則被饋以一個額外的位移方向反轉控制信號。此 位私方向反轉信號,係利用圖中所未顯示的適當邏輯電 路,而由位移指令行之中的算量上獲取的。如此, XOR閘3 3 0的定義功能,便可以容許利用一個強迫的 方向反轉彳§號,來改變由位移資訊之中所獲取的動態位 矛夕方向,號的位元數值。此即是說,當施加方向反轉信 5虎(避輯1)時’被抽取出來的動態位移方向信號便會被 轉另 方面,若方向反轉信號並未施加(邏輯〇), 則抽取出來的動態位移方向信號,便維持其原有之位元 本紙張&適(CNS)A4規格 WBMTPENT002\CHINESE\ching\b08054a.docThe sign: element ' acts as a dynamic displacement direction signal. This can be used, for example, to simply detect the signal condition of the positive and negative bit signal lines, and then = direction information. In the embodiment of FIG. 3, this method is used to directly extract the dynamic displacement direction information from the data bus that is transmitted from the displacement information to the logic unit 32, thereby showing the simple method of the present invention. . However, if it is not sent to the displacement direction input terminal 312' of the basic bidirectional displacement device 31, as in the conventional technique shifter of FIG. 2, the dynamic displacement direction information bit is transmitted to the shifter 31. Before the wheel 312, it can also be specially characterized by x〇R gate 33〇5, and the dynamic displacement direction information extracted from the displacement information can be controlled by other control signals, ώ XOR gate 3 3 〇 Process for adjustment. The dynamic displacement direction information bit is sent to -331 of the two inputs of X0R gate 330. The other input 3 3 = of the job f A 1 3 3 0 is then fed with an additional displacement direction reversal control signal. This private direction inversion signal is obtained from the calculation in the displacement command line using the appropriate logic circuit not shown in the figure. Thus, the definition function of the XOR gate 3 3 0 can allow a forced direction reversal of the 彳§ number to change the bit value of the dynamic bit direction obtained by the displacement information. That is to say, when the direction reversal letter 5 (avoidance 1) is applied, the extracted dynamic displacement direction signal will be switched to another aspect. If the direction reversal signal is not applied (logic 〇), then the extraction is performed. The dynamic displacement direction signal is maintained, and the original position paper & (CNS) A4 specification WBMTPENT002\CHINESE\ching\b08054a.doc is maintained.
Γ-.Β先閱讀背面之注意事項再填寫本頁) 297公釐) 1261968 經濟部智慧財產局員工消費合作社印製 A7 五、發明說明() 數值。 雖然大部份的數位系統係採用補數技術來處理負值的數值,但 仍有些系統是採用正負號位元的系統。在此類系統之中,數值資 料系、匕2了個絕對值資料欄以及一個正負號位元的組合格式來 表不的。為了讓本發明能夠被應用於此類系統之中,圖3中位移裝 置300之絕對值邏輯單元320,即變成不是必要的。圖4為-方塊 圖’其顯不本發明位移裝置之另—實關。注意到此種位移裝置 400的差異處係在於其中省略了絕對值邏輯單元。其x〇r閘43〇的 接腳431,與圖3中的實施例相類似地,仍必須控制來自於位移資 說算置的正負號位元。 另一方面’本發明之位移裝置,諸如前面於圖3中 所描述者,亦可適用於習知方向特定位移指令之新應用 用途之中。例如’考量前述的位移指令 與。此等方向特定位移指令的習知用 途’如同前述,必須要將其位移資訊算量^^解譯作為 正值的數值。 此等習知之方向特定的位移指令,乃是直接可以適 用於本發明之位移裝置之中。位移資訊算量不再必須一 定是一個正值的數值。不論是正值或負值的位移資訊算 量,在本發明之中皆是可以彈性適用的。位移資訊算量 的正負號位元,係被應用來將簡字碼 或S///F77-LEFT7之中的明示特定位移方向加以反轉。 在另一種可能發生的情況之中,指令簡字碼的 與- LEFT符號可以被抽取出來,並被作為前述指令 I 5 WBMTPENT002\CHINESE\ching\b08054a.doc 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) (請先閱讀背面之注意事項再填寫本頁)Γ-.ΒRead the notes on the back and fill out this page) 297 mm) 1261968 Ministry of Economic Affairs Intellectual Property Bureau employee consumption cooperative printed A7 V. Invention description () Value. Although most digital systems use the complement technique to process negative values, some systems use a system of positive and negative bits. Among such systems, the numerical data system, the absolute value data column, and a combination of positive and negative bits are not shown. In order for the present invention to be applied to such a system, the absolute value logic unit 320 of the displacement device 300 in Fig. 3 becomes unnecessary. Fig. 4 is a block diagram showing the other embodiment of the displacement device of the present invention. It is noted that the difference in such a displacement device 400 is that the absolute value logic unit is omitted. The pin 431 of its x〇r gate 43〇, similar to the embodiment of Fig. 3, still has to control the sign bit from the displacement. On the other hand, the displacement device of the present invention, such as those previously described in Figure 3, can also be adapted for use in new applications where conventional direction specific displacement commands are used. For example, 'consider the aforementioned displacement command and . The conventional use of the specific displacement command in these directions is as described above, and the displacement information calculation must be interpreted as a positive value. These conventional direction-specific displacement commands are directly applicable to the displacement device of the present invention. The displacement information calculation no longer has to be a positive value. Whether it is a positive or negative displacement information calculation, it can be flexibly applied in the present invention. The sign bit of the displacement information calculation is applied to invert the explicit specific displacement direction of the simple code or S///F77-LEFT7. In another possible situation, the AND-FTFT symbol of the instruction short code can be extracted and used as the aforementioned instruction I 5 WBMTPENT002\CHINESE\ching\b08054a.doc This paper scale applies to the Chinese National Standard (CNS) A4. Specifications (210 X 297 mm) (Please read the notes on the back and fill out this page)
1261968 A7 五、發明說明() s/z/mcc,心,,_7的第三個算量Η來加以處理。 本發明能夠強迫位移方向反轉的此種位移指令,其 用處可以在,例如,新-代的DSP之中找到應用的場 合。配備有包含了方向可反轉位移指令的指令集的 DSP,其一個應用實例係為壓縮資料的位元流包裝(bh stream packing)用途。下面將說明在其組合語言程式 的相關程式段落之中,執行此等位移操作的一個實例。 經過壓縮的資料,可以被包裝成為一個位元流的格 式。此種經過包裝的資料的整體資料大小可以縮減,且 在儲存與傳輸上,亦可以達到較高的效率。一個壓縮資 料位元流係由一個數量的,各具不同長度的位元攔位資 料所構成。每一個位元攔位資料,各係被儲存於記憶空 間之中,其對應的,接續的記憶位置内。每一個位元攔 位資料可以,例如,擁有少於16位元的位元長度,並 儲存於其指定的1 6位元記憶位置之中。不同長度但短 T 1 6位元的,各自不同的每一個位元攔位資料的經過 壓縮的資料位元,在記憶位置之中皆,例如,向右對 A。母一個圯憶位置中其餘未被佔用的位元空間,則是 以0填充。每一個位元攔位資料的長度資訊,係被儲存 於系統中對應的接續記憶空間的另一個記憶段落之中。 當壓縮的資料要進行處理時,位元欄位資料與其對應的 長度貧訊兩者,皆可由D S p從其儲存記憶位置之中讀取 出來’並加以處理。D S P可以利用加法及位移的操作, 而處理位元攔位資料與長度資訊,以便將此些位元欄位 16 WBMTPENT002\CHINESE\ching\b08054a.doc 本纸張尺度適用中國國家標準(CNS)A4規格(21() χ观公髮 (請先閱讀背面之注意事項再填寫本頁)1261968 A7 V. Invention Description () The third calculation of s/z/mcc, heart, and _7 is handled. The present invention is capable of forcing such a displacement command in which the direction of displacement is reversed, and its use can find an application in, for example, a new generation of DSP. A DSP equipped with a command set containing a direction reversible displacement command, one application example of which is a bit stream packing of compressed data. An example of performing such a shift operation among the relevant program sections of its combined language program will be described below. The compressed data can be packaged into a bit stream format. The overall size of such packaged data can be reduced, and higher efficiency can be achieved in storage and transmission. A compressed data bit stream consists of a number of bit-blocking data of different lengths. Each bit is blocked by the data, and each line is stored in the memory space, corresponding to the subsequent memory location. Each bit block data can, for example, have a bit length of less than 16 bits and be stored in its designated 16-bit memory location. Compressed data bits of different lengths but short T 1 6 bits, each of which is different in each bit, are in the memory position, for example, rightward to A. The remaining unoccupied bit space in the parent's memory location is padded with 0s. The length information of each bit block data is stored in another memory segment of the corresponding contiguous memory space in the system. When the compressed data is to be processed, both the bit field data and its corresponding length information can be read and processed by D S p from its stored memory location. The DSP can use the addition and shift operations to process the bit block data and length information so that the bit fields 16 WBMTPENT002\CHINESE\ching\b08054a.doc apply to the Chinese National Standard (CNS) A4. Specifications (21() χ观公发(Please read the notes on the back and fill out this page)
---------,.訂---------線. 1261968 五、發明說明( 資料包裝成λ π a & 取為不具重複多餘性的位元 率的儲存及傳於 ^ 爪乞式,以利有效 作: #輪°下面所列程式即係、執行這樣的-種操---------,.--------- Line. 1261968 V. Description of the invention (packaged into λ π a & is taken as a bit rate storage without repeatability and Passed in ^ Claw-style, in order to effectively work: #轮°The following programs are listed, perform such a kind of exercise
LD 將t所指明之位址處 中的位元攔位長度資訊 移入尤〃,並再將位址 將及2所指明之位LD will transfer the bit length information in the address specified by t to You, and then the address will be 2 and the specified position.
SFT A, _B 經濟部智慧財產局員工消費合作社印製 , 叶數加一 SUB B,X0 LD Y〇,R2,+ ., 址處 ;中的位元欄位長度資訊 ;移入7,,並再將位址 ;計數加一,同時並執行 ;厶-七,再將相減結果 ; 存入万中。 ;以邏輯位移操作將」中 ;之位元流向右移動厶所 ;指定之位元位置數,並 ;保留將與新到位元攔位 ;資料相結合時所需要的 ; 位元空間。 ADD A, Y〇 ;利用執行加法y +心而 ;將L内的新位元攔位資 ’料與現有位元流A互相 17 \\BMTPENT002\CHINESE\ching\b08054a.doc 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 29f公髮)------ 1261968 A7 五、發明說明( 組合。SFT A, _B Printed by the Ministry of Economic Affairs Intellectual Property Bureau employee consumption cooperative, the number of leaves plus one SUB B, X0 LD Y〇, R2, + ., address; the length of the bit field in the information; move into 7, and then Add the address; add one to the count and execute it at the same time; 厶-seven, then subtract the result; save it in 10,000. The logical displacement operation shifts the bit stream in the right direction to the right; specifies the number of bit positions, and retains the bit space needed to combine with the new bit position block; data; ADD A, Y〇; use the addition method y + heart; the new bit in L is intercepted and the existing bit stream A is 17 \\BMTPENT002\CHINESE\ching\b08054a.doc This paper scale applies to China National Standard (CNS) A4 specification (210 X 29f public hair) ------ 1261968 A7 V. Invention description (combination.
SFT Ay B 中 以邏輯位移操作將乂 包 ;之位元流移動5所指定 ;之位元位置數(向左),回 ;到其原來位置。 CMPS/ & #7(5 ;檢查為後續之位元流 經濟部智慧財產局員工消費合作社印製 ;裝操作所保留之空置位 ;元空間是否少於1 6位元。 ;若檢查結果為不少於,則 ;回到主程式。 SFL16 A ST R0,+,;若少於或等於16 位元, ;將d /中之已包裝位元流 ;存入及ο所指定之位置, ;再將位置計數增加一。 ;同時,以邏輯位移操作將 ;Z中之位元流向左移動 ;1 6個位元位置,以便移 ;除已包裝之位元流。 JD57 5,;執行加法操作5 以In SFT Ay B, the logical displacement operation will move the packet; the bit stream is shifted by 5; the number of bit positions (to the left), back; to its original position. CMPS/ &#7(5; check for the follow-up of the Ministry of Economic Affairs, Intellectual Property Office, employee consumption cooperatives; vacant positions reserved for loading operations; whether the meta-space is less than 16 bits. Not less than, then; back to the main program. SFL16 A ST R0, +,; if less than or equal to 16 bits, ; the d / in the packaged bit stream; stored in the location specified by ο; Then increase the position count by one. At the same time, operate with logical displacement; the bit flow in Z moves to the left; 16 bit positions to move; except for the packed bit stream. JD57 5,; 5 to
RET GT 18 WBMTPENT002\CHINESE\ching\b08054a.doc (請先閱讀背面之注意事項再填寫本頁)RET GT 18 WBMTPENT002\CHINESE\ching\b08054a.doc (Please read the notes on the back and fill out this page)
本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) 1261968 五、 發明說明( ;便更新d中之空置空間 ;U)。 RET ;回到主程式。 在上列的組合語言程式之中,其操作過程中所牽 到的所有記錄器係說明如下: 涉 經濟部智慧財產局員工消費合作社印製 ^ 0 a2 B y〇 χ〇 R〇 Ri DSP之累增器(acCumulator),40位元,供 儲存已包裝之位元流 d之最低效16位元,其位置可供與現有位元 流與新進入之位元攔位資料進行包裝結合 ^累增器之中效16位元,供儲存現有/已包裝 之位元流 ~ d之最高效8位元,未使用 D S P之另一累增器,4 0位元 留下之可獲用/空置之位元空間 D S P之資料記錄器,1 6位元 入之位元攔位資料 D S P之資料記錄器,1 6位元: 入位元攔位資料之長度資訊 D S P之位址記錄器,1 6位元, 生之已包裝位元流之儲存記憶值置 D S P之位址記錄器,1 6位元,用於儲存新進 入位元攔位資料之長度資訊之儲存記憶位置 19 跡丨細·啊㈣\b_54a.d0( 用於儲存新饋 用於儲存新饋 用於儲存新產 (請先閱讀背面之注意事項再填寫本頁) ----·---Μ訂---------線. 本紙張尺度適用中國國家標準(CNS)A4規格(210¾7公釐) I261968 五、 發明說明(This paper scale applies to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) 1261968 V. Description of the invention (; update the vacant space in d; U). RET ; back to the main program. Among the combination language programs listed above, all the recorders involved in the operation are described as follows: Employees of the Ministry of Economic Affairs, Intellectual Property Bureau, Consumer Cooperatives Printed ^ 0 a2 B y〇χ〇R〇Ri DSP acCulator, 40 bits, for storing the least effective 16-bit of the packed bit stream d, its position can be combined with the existing bit stream and the newly entered bit block data. 16-bit effect, for storing the most efficient 8-bit of the existing/packaged bit stream ~ d, without using another accumulator of DSP, 40 bits left for available/vacant Bit space DSP DSP data logger, 16-bit bit-in-bit block data DSP data logger, 16-bit: Length information of bit-in-bit block data DSP address logger, 16 bits Yuan, the stored memory value of the packed bit stream, the address register of the DSP, 16 bits, the storage memory location for storing the length information of the newly entered bit block data. 19 Traces and details (4) \b_54a.d0 (for storing new feeds for storing new feeds for storing new products (please read the back first) Note: Please fill out this page) ----·---Μ定---------线. This paper scale applies to China National Standard (CNS) A4 specification (2103⁄47 mm) I261968 V. Invention Description (
R D S P之位址記錄器,1 6位元·采 入位元爛位資料之儲存記憶位置 ¥新進 如此,顯然若未應用本發明所揭示之新穎位移器裝 置,亦即,若沒有本發明之方向可反轉位移指令,則上 =程式段落之中的每-行位移指令,便必須以兩行指令 1代’即其中必須先插入一個負值指令, 的方向·· 欠位移Address address recorder of RDSP, storage location of 16-bit element-incorporated bit rogue data. It is obvious that if the novel displacer device disclosed by the present invention is not applied, that is, if there is no direction of the present invention The reversible displacement command, then the per-line displacement instruction in the upper = program paragraph, must be two generations of instructions 1 generation 'that is, the direction in which a negative value instruction must be inserted first · under-displacement
NEG B SFT A,B 先反轉位移方向 並再向右位移。NEG B SFT A, B Reverses the direction of displacement and then shifts to the right.
NEG B SFT B 接著再次反轉方向。 並接著再移回(向左) 經濟部智慧財產局員工消費合作社印製 圖5顯示可支援方向反轉指令,#Dsp用 明位移器裝置,1右 ^ 〃在私式執仃的過程中的資料記錄器的 ?木1乍If形。在圖5夕半缺^在並…J 中’所有需要的資料皆各自 射赢 ^ 且位兀亦在A的MSB之處 對;=1。之後在j盘j — 斗二 〃的工置位兀空間之中填入〇。接 元椚你次以中執仃5 u作,以計算進來的位 几欄位 料所需要的 疋工間接者,位元流便被向右 2 0 本紙張尺度適t關家鮮 J:\ching\b08054a.doc (210 x 297 公釐) 1261968NEG B SFT B then reverses the direction again. And then moved back (to the left) Ministry of Economic Affairs Intellectual Property Bureau employees consumption cooperatives printed Figure 5 shows the direction reversal command can be supported, #Dsp with the shifter device, 1 right ^ 〃 in the process of private execution The data logger is made of wood. In Figure 5, half of the lack of ^ in ... J 'all the required information are each shot win ^ and the position is also in the MSB of A; After that, fill in the space in the space of the j-disk j- bucket. After receiving the Yuan 椚 椚 椚 椚 椚 椚 , , 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚 椚\b08054a.doc (210 x 297 mm) 1261968
五、發明說明( 位移,移動5 - Fr - / X"個位兀位置。其後,在 中,步驟2的結果利用將 料與累增器…現有位元:互相::的新位元攔位資 ,4I ffl ^ , 兄有位兀机互相組合而包裝起來。此 係利用執行d 7接你;、去 0私作而達成。在步驟4之 位兀流在累增器3中被產 度玍出木取後,在步驟5 中,新位元流被移動厶_ r / 一 秒勁θ A個位兀位置(向左),回到 ^原來位置,此時即可備妥以供儲細記憶之中,或等 待進行另一個包裝處理的循環。 雖然前面已是本發明特定實施例的完整說明,但其各種 的修改變化,變動的構造及等效者的應用仍是可能的。 因此,前面的描述說明即不應被拿來限定本發明,而其 範疇應以後附之申請專利範圍乙節文字内容來加以界 定0 (請先閱讀背面之注意事項再填寫本頁) 經濟部智慧財產局員工消費合作社印製 WBMTPENT002\CHINESE\ching\b08054a.doc 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐)V. Description of the invention (displacement, movement 5 - Fr - / X" position 兀 position. Thereafter, in the result of step 2, use the material and the accumulator... existing bits: new bits of each other:: Position, 4I ffl ^, brother has a set of machines combined with each other and packaged. This system uses the implementation of d 7 to pick you up; go to 0 private to achieve. In step 4, the turbulence is produced in the accumulator 3 After the wood is taken out, in step 5, the new bit stream is moved 厶 _ r / one second θ A position 向 position (to the left), back to the original position, then ready for In the memory of the memory, or waiting for another cycle of packaging processing. While the foregoing has been a comprehensive description of specific embodiments of the invention, various modifications, variations, and equivalent applications are still possible. Therefore, the above description should not be used to limit the invention, and its scope should be defined in the text of the patent application scope attached to the future. 0 (Please read the note on the back and fill in this page) Property Bureau employee consumption cooperative printed WBMTPENT002\CHINESE\ching\b080 54a.doc This paper size applies to the Chinese National Standard (CNS) A4 specification (210 X 297 mm)
Claims (1)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW90123122A TWI261968B (en) | 2001-09-13 | 2001-09-13 | Shifting apparatus for bit-shifting manipulation in a digital processor device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW90123122A TWI261968B (en) | 2001-09-13 | 2001-09-13 | Shifting apparatus for bit-shifting manipulation in a digital processor device |
Publications (1)
Publication Number | Publication Date |
---|---|
TWI261968B true TWI261968B (en) | 2006-09-11 |
Family
ID=37987058
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW90123122A TWI261968B (en) | 2001-09-13 | 2001-09-13 | Shifting apparatus for bit-shifting manipulation in a digital processor device |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI261968B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9495165B2 (en) | 2009-12-17 | 2016-11-15 | Intel Corporation | Method and apparatus for performing a shift and exclusive or operation in a single instruction |
-
2001
- 2001-09-13 TW TW90123122A patent/TWI261968B/en not_active IP Right Cessation
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9495165B2 (en) | 2009-12-17 | 2016-11-15 | Intel Corporation | Method and apparatus for performing a shift and exclusive or operation in a single instruction |
US9495166B2 (en) | 2009-12-17 | 2016-11-15 | Intel Corporation | Method and apparatus for performing a shift and exclusive or operation in a single instruction |
US9501281B2 (en) | 2009-12-17 | 2016-11-22 | Intel Corporation | Method and apparatus for performing a shift and exclusive or operation in a single instruction |
TWI575455B (en) * | 2009-12-17 | 2017-03-21 | 英特爾股份有限公司 | Method and apparatus for performing a shift and exclusive or operation in a single instruction |
US9747105B2 (en) | 2009-12-17 | 2017-08-29 | Intel Corporation | Method and apparatus for performing a shift and exclusive or operation in a single instruction |
US10684855B2 (en) | 2009-12-17 | 2020-06-16 | Intel Corporation | Method and apparatus for performing a shift and exclusive or operation in a single instruction |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US5819117A (en) | Method and system for facilitating byte ordering interfacing of a computer system | |
JP3556955B2 (en) | Method and apparatus for switching byte order in computer | |
TW490636B (en) | Processor for executing highly efficient VLIW | |
CN103562854B (en) | Systems, devices and methods for the register that aligns | |
TWI502499B (en) | Systems, apparatuses, and methods for performing a conversion of a writemask register to a list of index values in a vector register | |
JPH10207870A (en) | Vector processor programming and computer program generating method for parallel programming of asymmetric double processor provided with vector processor and risc processor | |
CN104185839B (en) | Load data into the method and system of register | |
TWI525533B (en) | Systems, apparatuses, and methods for performing mask bit compression | |
CN104903850B (en) | Instruction for sliding window coding algorithm | |
KR101729829B1 (en) | Method and apparatus for performing a plurality of multiplication operations | |
US7716455B2 (en) | Processor with automatic scheduling of operations | |
CN108804137A (en) | For the conversion of double destination types, the instruction of cumulative and atomic memory operation | |
JPH09512652A (en) | Data processing with multiple instruction sets | |
JP3954171B2 (en) | How to fill a vector with scalar values on a computer | |
CN107851013B (en) | Data processing apparatus and method | |
JPH02149156A (en) | Communication processor | |
CN110119362A (en) | Memory writes log storage processor, method, system and instruction | |
AU2004236860B2 (en) | Instruction for computing a security message authentication code | |
JP2018506094A (en) | Method and apparatus for performing BIG INTEGER arithmetic operations | |
EP1680735B1 (en) | Apparatus and method that accomodate multiple instruction sets and multiple decode modes | |
US20080126763A1 (en) | Method of translating n to n instructions employing an enhanced extended translation facility | |
TWI733718B (en) | Systems, apparatuses, and methods for getting even and odd data elements | |
TWI261968B (en) | Shifting apparatus for bit-shifting manipulation in a digital processor device | |
TW201732574A (en) | Systems, methods, and apparatuses for improving vector throughput | |
TW380237B (en) | Data processing system register control |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
MK4A | Expiration of patent term of an invention patent |