TWI311721B - - Google Patents

Download PDF

Info

Publication number
TWI311721B
TWI311721B TW095107574A TW95107574A TWI311721B TW I311721 B TWI311721 B TW I311721B TW 095107574 A TW095107574 A TW 095107574A TW 95107574 A TW95107574 A TW 95107574A TW I311721 B TWI311721 B TW I311721B
Authority
TW
Taiwan
Prior art keywords
micro
control
instruction
core logic
control system
Prior art date
Application number
TW095107574A
Other languages
Chinese (zh)
Other versions
TW200734921A (en
Inventor
Chuan Po Ling
Original Assignee
Padauk Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Padauk Technology Co Ltd filed Critical Padauk Technology Co Ltd
Priority to TW095107574A priority Critical patent/TW200734921A/en
Priority to KR1020060119678A priority patent/KR20070092093A/en
Publication of TW200734921A publication Critical patent/TW200734921A/en
Application granted granted Critical
Publication of TWI311721B publication Critical patent/TWI311721B/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/22Microcontrol or microprogram arrangements
    • G06F9/28Enhancement of operational speed, e.g. by using several microcontrol devices operating in parallel
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/22Microcontrol or microprogram arrangements
    • G06F9/26Address formation of the next micro-instruction ; Microprogram storage or retrieval arrangements
    • G06F9/262Arrangements for next microinstruction selection
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30003Arrangements for executing specific machine instructions
    • G06F9/30007Arrangements for executing specific machine instructions to perform operations on data operands
    • G06F9/30018Bit or string instructions
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/30145Instruction analysis, e.g. decoding, instruction word fields
    • G06F9/30156Special purpose encoding of instructions, e.g. Gray coding
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/38Concurrent instruction execution, e.g. pipeline, look ahead
    • G06F9/3885Concurrent instruction execution, e.g. pipeline, look ahead using a plurality of independent parallel functional units
    • G06F9/3893Concurrent instruction execution, e.g. pipeline, look ahead using a plurality of independent parallel functional units controlled in tandem, e.g. multiplier-accumulator

Description

1311721 九、發明說明: 【發明所屬之技術領域】 本發明係有關-種微控㈣統之指令及其執行方法,_是有關一種 多重微控制系統之指令及其執行方法。 【先前技術】 隨_科技的進步,電腦能夠提供的服務與魏也越來越全面,若將中1311721 IX. Description of the invention: [Technical field to which the invention pertains] The present invention relates to an instruction of a micro-control (four) system and an execution method thereof, and _ is an instruction related to a multi-micro control system and an execution method thereof. [Prior Art] With the advancement of _ technology, the services that computers can provide and Wei are more and more comprehensive.

、’視為σ電腦的,。臟,則電腦程式則為電腦的思想邏輯,藉由中 錢理器啟動所有硬體並同時進行讀取程式,可使電腦提供㈣者解決問 題的方法,或是完献时欲達成之任務。 而指揮電腦執行-連串工作的處理程序與步驟的程式,係是由一系列 指令或敘述的組合所構成,其中,機器語言是最基本的指令,其係由運算 碼”運算7G所共同合成,運算碼與運算元之間的關係乃是以運算元的資料 作運算碼的動作,同時,為了簡化程式的編輯,已發展出—套可將指令與 敛述以簡碼方式表示的方法,例如,在縣元中㈣料可能為位址資訊, 或是單純的熱資訊,在運算碼中収提供各種動作啸訊,妓資料轉 移、算數運算、運算、位元運算献程式分支軸作資訊,請先參考 下列式⑴所示,其係為習知之_指令及其包含之運算碼與運算元: MOV A, REGA ' 、’ MOV為運算碼’其係表示「資料之搬移」’而a與嶋則分別為兩個 心其係各自代表控制器中的「累加器」與「暫存器A」,因此,這個 指令所提_整_作可娜為「錄巾,㈣抑A中之資料存入 累加器中」。細’在電腦執行指令前,必須先通過—連串的解碼與分析的 工作’以將運算元與運算 的單元—動作的執行,二:=新解讀— ,與第-⑹圖•,其係分_知的:=時參考第 法’在進行式⑴指令之執行時,首先,在步驟测中二編之執行方 匯流排12讀取記憶體i〇 :制益16從資料 一__ 指令解 16中’把暫存器A162中之資料存·器161+」二控制' 4λ:Α.| 3S 1 λ ., . Τ」’步驟 S03 中,依據 ^ ⑽卿’進行咖Α 162中敝讀 入累加器161中,以完成式⑴指令的執行。 待 根據上述的控«歧其齡之執行枝可得知,财的齡設計皆 以系統中僅有-個控(或僅有—個主要控制器)為前提,並且須經由 ^ 一的控版糊令術娜執行,鄉細制系統内 有一個以上可執行程式之微控制核心邏輯(且可能各自具有對應之暫存器 或其他相關電路單元)時,前述習知之指令謝式,將難以達成平行之 多工處理;即使勉強而為,軟體編寫也勢必十分複雜。對於多重微控制系 統而言,當編膽複程式時,讓不晴制單元皆可彼此分享内部資 源、讀取程式執行織以及助㈣,㈣重要,如私但可以提昇多重 微控制祕平行處理與及時反應醜力,也可降低系統成本。 有鑑於此’本發明係為針對可解決上述問題之一多重微控制系統,提 出-種適用於多重微控制系統之指令及其執行方法,透過多重微控制系統 執行此指令’以達成平行多工處理的效能,並有效提昇控制系統運算處理 1311721 的能力。 【發明内容】 本發明之主要目的’係在提供-種多重難㈣統之齡,其包含— 運算碼(opemtion code)與至少-多重微控制核心邏輯選擇運算元 (operand for selection a_g mUltiple MQJ c〇re 以⑹),且當有 y ’個或⑽以下之微控雜心邏輯時,需要至少N個位元數以選擇此^個微 • 控制核心邏輯,其中,此N個位元數可為連續之位元數,亦或是不連續之 位元數,而運算碼係可用以提供指令執行程式時之運算動作,多重微控制 核心邏輯選擇運算元則是在執行指令時,可指定出多重微控織統内,執 行指令時被獅之微控繼心邏輯及其子運算元(恤―。啊⑷之資訊; 利用此^7,不需透過編寫繁複的程式碼,即可簡單且明確地指示出多重 微控制系統中欲選擇之微控制核心邏輯及其子運算元。 本發明之另—目的’係在提供—種可以根據上述指令雜作的多重微 _ 控制系統。 ' 本發明之又—目的’係在提供—種多錄控⑽統之指令執行方法, 透過主運算微控制單元提取指令並完成解碼,利用指令中之多重微控制核 〜邏輯選擇運算①之指定’除可令被選擇之微控制核心、邏輯配合進行運算 碼之動作外’更可_取_被選擇之微控制核心、邏輯内部之資料並直接寫入 另一被選擇之微控制核心邏輯之指定位置。藉由此指令及方法,使得多重 微控制祕内不同的微控制核心邏輯可彼此分享資源以及互相控制,實現 平行處理數個程式的能力,並提昇其運作之效能。 1311721 算元包含「M3」與「A」兩個部份,意指微控制器核心邏輯M3,其内部之一 位置A而此子位置A即疋對應出第三累加器,因此,此多重微控制核心 邏輯選擇運算元腿係對應出「在第三微控制器核心邏輯的第三累加器」; 而在多重微控制核心邏輯選擇運算元_G2A中,則是同時包含微控制器 =心邏輯的錄址M2,與細部之子⑽_,峨子位置瞧即是 才曰不出暫存n 2A ’故’此多麵控偷心邏輯讎運算元卿隐係對應 魯4「在第二微控制器核心邏輯的暫存器2A」。因此,這個指令所提供的整體 動物釋為「把第二微控制器核心邏輯的暫存器2八之資料存入第三微控 制器核心邏輯的第三累加器中」。 上述係僅為本發明之-種指令實施態樣,除了式⑵中的資料轉移運算 動作以外’-般常見於運算碼中之其他運算動作,如·算數運算、邏輯運 鼻、位娜_樹_峨,歧侧㈣M輯中執行 ^算動作’皆侧在她懈中。同時,姆輸制核心邏 %選娜魏的-峨元中,此概除了爾來源位址、目的 p的貝赠’柯_供—_訊,以提_微_心邏輯選 •擇運算元所對應之指令動作,例如式⑶帽提供之指令: MOV M1REGIB, #55f{ (3) 其中,MOV依然為運算碼,盆孫矣_ v ^貝料之搬移」,而M1REG1B與#55H則 :多錄控_順_喊繼,其巾,同理於上述有關 =一的在多重微控制核心邏輯選擇運算元麵⑽係對應出「在 一微控制器核心邏輯的暫存器1B」,而運算元_則是一立即常數資料, 1311721 其係表示「」的資料資訊。因此,這個指令所提供的整體動作可解釋為 「把55H搬移至第-微控制器核心邏輯的暫存$ ib中」。 上述式⑵係為運算碼搭3&二多麵控術細賴麵運算元的組 合;式⑶係為-運算碼搭配-多重微控制核心邏輯選擇運算元與一其他運 算元的組合;除此之外’僅有-運算竭搭配—多重微控制核心邏輯選擇運 算元的组合,當然也屬可能,例如式⑷中所提供之指令:, 'Think of σ computer. Dirty, the computer program is the computer's logic of thinking. By starting all the hardware and reading the program at the same time, the computer can provide (4) the method to solve the problem, or the task to be fulfilled when finished. The program that directs the computer to execute the series of processing procedures and steps is composed of a series of instructions or a combination of statements. The machine language is the most basic instruction, which is synthesized by the operation code "7G". The relationship between the opcode and the operand is the operation of the operand with the data of the operand. At the same time, in order to simplify the editing of the program, a method of expressing the command and the concatenation in a shortcode manner has been developed. For example, in the county element (4), it may be address information, or simple thermal information, and provide various motion tweets in the operation code, data transfer, arithmetic operation, operation, and bit operation. Please refer to the following formula (1), which is a conventional _ instruction and its included opcodes and operands: MOV A, REGA ', 'MOV is the opcode' which means "data transfer" and a And 嶋 为 为 两个 两个 两个 两个 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自 各自Data is stored in the accumulator . Fine 'before the computer executes the instruction, it must pass the series of decoding and analysis work' to execute the operation unit and the operation unit - the second: = new interpretation - and the - (6) diagram Points _ know: = when referring to the first method 'in the execution of the instruction (1) instruction, first, in the step of the test, the second party's executive bus 12 reads the memory i: the benefit 16 from the data one __ instruction In the solution 16, 'put the data store 161+ in the register A162' to control '4λ: Α.| 3S 1 λ ., . Τ"', in step S03, according to ^(10)卿', the curry 162 is 敝The accumulator 161 is read in to complete the execution of the instruction of equation (1). According to the above-mentioned control, the age of the design is based on the premise that there is only one control (or only one main controller) in the system, and it must be controlled by ^1. When the implementation of the micro-control core logic of one or more executable programs in the township system (and possibly each of the corresponding scratchpads or other related circuit units), the above-mentioned conventional instructions will be difficult to achieve. Parallel multiplex processing; even if it is reluctant, software writing is bound to be very complicated. For multiple micro-control systems, when the program is programmed, the unclear units can share internal resources, read program execution and help (4), and (4) important, such as private but can improve the parallel processing of multiple micro-controls. Responding to uglyness in a timely manner can also reduce system costs. In view of the above, the present invention is directed to a multi-micro control system that can solve the above problems, and proposes an instruction suitable for a multiple micro-control system and an execution method thereof, and executes the instruction through multiple micro-control systems to achieve parallel multiple The efficiency of processing, and effectively improve the ability of the control system operation processing 1311721. SUMMARY OF THE INVENTION The main object of the present invention is to provide a multi-difficult (four) system age, including - opemtion code and at least - multiple micro control core logic selection operation elements (operand for selection a_g mUltiple MQJ c 〇re to (6)), and when there are y ' or (10) below the micro control clutter logic, at least N number of bits are needed to select the micro control core logic, wherein the N number of bits can be It is a continuous number of bits, or a number of discontinuous bits, and the arithmetic code can be used to provide an operation when the instruction executes the program. The multiple micro-control core logic selection operation element can be specified when the instruction is executed. In the multi-micro-control woven system, when the command is executed, the lion's micro-control relay logic and its sub-operational elements (shirts. ah (4) information; using this ^7, without writing complicated code, can be simple and The micro-control core logic and its sub-operating elements to be selected in the multiple micro-control system are explicitly indicated. The other object of the present invention is to provide a multiple micro-control system which can be mixed according to the above instructions. The purpose--the purpose is to provide a multi-recording control (10) system instruction execution method, and extract the instruction through the main operation micro-control unit and complete the decoding, and use the multiple micro-control kernel in the instruction to the logic selection operation 1 to specify The selected micro-control core and logic cooperate to perform the operation of the operation code, and the data of the selected micro-control core and the logic are directly written to the designated position of the other selected micro-control core logic. With this instruction and method, different micro-control core logics in multiple micro-control secrets can share resources and control each other, realize the ability to process several programs in parallel, and improve the performance of the operation. 1311721 The arithmetic element contains "M3" The two parts of "A" mean the core logic M3 of the microcontroller, and one of the internal positions A and the sub-position A corresponds to the third accumulator. Therefore, the multi-micro control core logic selects the operation leg. Corresponding to "the third accumulator in the third microcontroller core logic"; and in the multi-micro control core logic selection operand _G2A, it also contains the micro-control Controller = heart logic address M2, and the child of the detail (10) _, the position of the scorpion is that it can not be temporarily stored n 2A 'so' this multi-face control stealing heart logic 雠 operation Yuanqing hidden system corresponds to Lu 4 "in The second microcontroller core logic register 2A". Therefore, the overall animal provided by this instruction is interpreted as "storing the data of the second microcontroller core logic register 2-8 into the third microcontroller core. The logic is the third accumulator. The above is only the implementation of the invention, except for the data transfer operation in equation (2), which is commonly used in other computational operations, such as arithmetic operations. Logic, nose, _ tree _ 峨, 歧 ( (4) M in the implementation of the ^ calculations are all in the side of her slack. At the same time, the core of the core system of the selection of Na Wei - 峨元中, this is in addition to The source address, the destination p, the gift of 'K_____, to mention the instruction action corresponding to the operation element, such as the instruction provided by the formula (3) cap: MOV M1REGIB, #55f{ ( 3) Among them, MOV is still the opcode, the pottery 矣 _ v ^ shell material moved, and M1REG1B and #55H: multi-record Control _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ is an immediate constant data, 1311721 which is the information of "". Therefore, the overall action provided by this instruction can be interpreted as "moving the 55H to the temporary storage $ib of the -microcontroller core logic." The above formula (2) is a combination of the operation code lap 3 & two multi-face control operation thin surface operation unit; the equation (3) is a combination of the operation code collocation-multiple micro control core logic selection operation element and one other operation element; In addition to the 'only-computing-matching-multiple micro-control core logic selection operand combination, of course, is also possible, such as the instructions provided in equation (4):

POP M3PC (4) 此指令中,POP為運算碼’表示「將堆疊記髓的内容送至指定位置」,而 c為此才"中之唯運异兀’且為—多重微控繼心邏輯選擇運算元, 意指「第三微控制難心邏輯之程辆數器」,故式⑷賴齡所提供的 整體動作可解釋為「將堆疊記憶_内容送至第三微控私邏輯之程 式計數器中」。 引述各式⑵-⑷中’「多重微控制核心邏輯選擇運算元」均包含表 示微控制核心邏輯的部份(M1,M2, M3等)與表示子位置的子運算元(A, PC等)。但指令中4需要定義微控制核心邏輯,而不需要定 義其子位置,也是可能的。鱗「多紐蝴_輯娜㈣元」便只 包含表示難偷心·的部份,㈣必包含表示子位置的子運算元。例 如’假設撕㈣細心「賴咖__堆疊記憶體 内」,則在町齡中’便不需料含表示子位置的子運算元. PUSHPC M4 _ (5) 此指令表示「«讀㈣核心顧之料計數器_料至堆疊記憶體 1311721 内J ο 此外’細上所勒料,細較高_語法來㈣齡4實上如 果轉譯成低階二進位碼,驗述各式⑵—⑷,將會以—連串的二進位 數字來表示,iux射—雜域來表轉算碼,其他錄麵運算元。 例如假設以16位的二進位碼來表示前述指令,則例如可以使用從最具意義 位元Signifi咖t bit,膽)開始的五位數來作為表示運算碼的位 置,其他位數則作為表示運算元的位置。 心在n二進位之下,每_位元數係可用以選擇二個微控制核心邏輯,因此, 當有2N個或2、以下之微控制核心邏輯時’ f要至少n個位魏來選定所 要的微控繼、邏輯(概念上,這個N位元的二触碼,可峨為微控制 核心邏輯的「識別碼」(identiflcati〇n,ID))。在本發明中,此位元數之 編排係可鱗續之位元,亦歧林賴德元;所使㈣錄,-般情 形下可以放在運算元的位置巾,但也可以放在運算位置中,或是兩者 皆具。例如’假設指令如前所舉例則6位的二進位石馬來表示,且指令集共 有24種運算碼需要定義。由於共有24種運算碼,2< = ^ n 欠需要五恤元絲不’因驗用從最具意義位元開始的五位數來作為表 示運算碼驗置’其他位數則作為表示運算元驗置。此時,假設多重微 控制系統使狀個微控繼心缝(故需要三位數的識㈣來缺所要的 微控制核心邏輯)’則根據本發明,可以使用第六至第讀數,或第Η至 16位數等連續位元來表示所缺賴控繼心、賴,也可以使用例如第 ’、十14 一個不連續的位元來表示所指定的微控制核心邏輯。且因為。 1311721 有24種運算碼’故可令麵至1G1U表示第i至μ種運算,而令ιΐχχχ 表示第24種,並令U_至η出分別絲釺鄉_至树綱核心邏 輯執行第24種運算^此時,在第丨至23種運算的指令中,指定微控制核 :邏輯的三個位元是位於運算元的位置内;而在第24種運算的指令中,指 定微控制核心、邏輯的三個位元則是位於運算碼的位置内。 在前述第24種運算的齡中,雜難_心邏輯識觸是位於運算 碼的位置内,但在概念上,我們也可以將運算碼視為由咖開始的兩位數, 而將多重微控制核心邏輯選擇運算元視為由第三位數開始之後的數字。 以上’係針對多重微控制系統之指令進行介紹,接續,將再提出指令 之執行方法。係、以上述式⑵中的指令為例,並請同時參考第二⑷圖與第 -(b)圖所示,其係分職本發_多重微控織統及其齡之執行方法。 在進行式⑵指令之執行時’首先,在步驟su中,在N個微控制核心邏輯 30 32所共同組成之多重微控制系統中,各微控制核心邏輯内有 暫存器262、263、282、302、322、323、324,利用第一微控制核心邏輯 26以做為主運算微控制單元,並透過資髓流排22 取記‘隨2〇中的 資料’且暫存至指令暫存器24中;步驟S12 +,第-微控制核心邏輯26 將式⑵指令解碼’並判讀出式⑵指令之内容為「把第二微控制核心邏輯 28的暫存器2A282中之資料存入第三微控制核心邏輯30的第三累加器3〇1 中」,步驟S13中,依據第一微控制核心邏輯26所判讀出的結果,由第一 微控制器核心邏輯26命令第二微控制核心邏輯28,以將其中暫存器2A犯2 的資料進行讀取’並將此資料寫入第三微控制核心邏輯30的第三累加器3〇1 13 1311721 【圖式簡單說明】 第一(a)圖為習知的控制系統方塊示意圖。 第一(b)圖為習知的控制系統之指令執行方法。 第二(a)圖為本發明之一實施例之系統方塊示意圖。 第二(b)圖為本發明之一實施例之指令執行方法。 【主要元件符號說明】 10記憶體 12資料匯流排 14指令暫存器 16控制器 162暫存器 161累加器 20記憶體 22資料匯流排 24指令暫存器 26第一微控制器核心邏輯POP M3PC (4) In this command, POP is the opcode 'meaning "send the contents of the stack to the specified location", and c is the only one in the "transportation" and is - multi-micro control The logic selects the operand, which means "the third micro-controls the process of the difficult logic." The overall action provided by the formula (4) Lai Ling can be interpreted as "sending the stack memory_content to the third micro-control private logic. In the program counter." The references to the "multiple micro-control core logic selection operands" in the equations (2)-(4) include the parts representing the micro-control core logic (M1, M2, M3, etc.) and the sub-operation elements (A, PC, etc.) indicating the sub-positions. . However, it is also possible that the instruction 4 needs to define the micro-control core logic without defining its sub-location. The scale "Dou New Butterfly _ Collection Na (four) Yuan" contains only the part that indicates difficulty to steal, and (4) must contain the sub-operating element that represents the sub-position. For example, 'hypothetical tearing (four) careful "Lai __ stacked memory", then in the age of the town, there is no need to include sub-operating elements that represent sub-positions. PUSHPC M4 _ (5) This instruction means "«read (four) core Take care of the counter _ material to the stack memory 1311721 J ο In addition, 'finely on the material, fine higher _ grammar to (four) age 4 in fact, if translated into low-order binary code, test the various equations (2) - (4), It will be represented by a series of binary digits, iux shot------------------------------------------------------------------------------ For example, if you use the 16-bit binary code to represent the above instructions, you can use the most The five-digit number of the meaning bit Signifi is the beginning of the operation code, and the other digits are used to represent the position of the operand. The heart is below the n-digit, and the number of _bits is available. To select two micro-control core logics, therefore, when there are 2N or 2 micro-control core logics, 'f must have at least n bits to select the desired micro-control, logic (conceptually, this N-bit The two-touch code can be regarded as the "identification code" of the micro-control core logic (identi Flcati〇n, ID)). In the present invention, the arrangement of the number of bits is a bit of a continuation of the bit, and is also a dissimilarity of the forest; the (four) record, in the general case can be placed in the position of the operand, but can also be placed in the operation position Or both. For example, assuming that the instruction is as exemplified above, the 6-bit binary stone is represented, and the instruction set has a total of 24 arithmetic codes to be defined. Since there are 24 kinds of opcodes, 2<=^n owes the need for five-element wire. 'Because the test uses the five-digit number starting from the most significant bit as the indication of the operation code check', the other digits are used as the representation operand. Inspection. At this point, it is assumed that the multiple micro-control system makes a micro-control center-sewed seam (so three-digit knowledge (four) is needed to lack the micro-control core logic). According to the present invention, the sixth to the first reading, or the A continuous bit such as a 16-bit number is used to indicate the default control, and a non-contiguous bit such as the first and the eleventh may be used to represent the specified micro-control core logic. And because. 1311721 There are 24 kinds of arithmetic code 'so you can make the face to 1G1U to represent the i-th to μ-th operation, and let ιΐχχχ denote the 24th kind, and let U_ to η out the silky township_to the tree core logic to execute the 24th kind Operation ^ At this time, in the instructions from the 23rd to the 23rd operations, the micro-control kernel is specified: the three bits of the logic are located in the position of the operand; and in the instruction of the 24th operation, the micro-control core is specified, The three bits of logic are located in the location of the opcode. In the age of the 24th operation mentioned above, the miscellaneous _ heart logic is located in the position of the opcode, but conceptually, we can also regard the opcode as a two-digit number starting from the coffee, and The control core logic selects the operand as the number after the start of the third digit. The above is an introduction to the instructions of the multiple micro-control system, and the method of execution of the instructions will be further proposed. For example, the instruction in the above formula (2) is taken as an example, and please refer to the second (4) diagram and the - (b) diagram at the same time, which are divided into two parts: the multi-micro control system and the execution method of its age. In the execution of the instruction of the equation (2), first, in the step su, in the multiple micro-control system composed of the N micro-control core logics 30 32, each micro-control core logic has a register 262, 263, 282 therein. , 302, 322, 323, 324, using the first micro-control core logic 26 as the main operation micro-control unit, and taking the data in the '2' through the medullary flow row 22 and temporarily storing it to the instruction temporary storage In step S12+, the first-micro-control core logic 26 decodes the instruction of the equation (2) and determines that the content of the read-only (2) instruction is "storing the data in the register 2A282 of the second micro-control core logic 28 into the first The third micro-controller core logic 30 of the third accumulator 3〇1, in step S13, according to the result of the first micro-control core logic 26 read out, the first micro-controller core logic 26 commands the second micro-control core Logic 28, to read the data in which the scratchpad 2A commits 2' and write this data to the third accumulator of the third micro control core logic 30. 3〇1 13 1311721 [Simplified description] First ( a) The figure is a block diagram of a conventional control system. The first (b) diagram shows the instruction execution method of the conventional control system. The second (a) diagram is a block diagram of a system according to an embodiment of the present invention. The second (b) diagram is an instruction execution method according to an embodiment of the present invention. [Main component symbol description] 10 memory 12 data bus 14 instruction register 16 controller 162 register 161 accumulator 20 memory 22 data bus 24 instruction register 26 first microcontroller core logic

262暫存器1A 263暫存器1B 261第一累加器 28第二微控制器核心邏輯262 register 1A 263 register 1B 261 first accumulator 28 second microcontroller core logic

282暫存器2A 1311721 281第二累加器 30第三微控制器核心邏輯 302暫存器3A 301第三累加器 - 32第N微控制器核心邏輯282 register 2A 1311721 281 second accumulator 30 third microcontroller core logic 302 register 3A 301 third accumulator - 32 Nth microcontroller core logic

. 322暫存器NA. 322 Register NA

323暫存器NB > 324暫存器NC 321第N累加器323 register NB > 324 register NC 321 Nth accumulator

Claims (1)

1311721 ^ . ^· ;: 十、申請專利範園: 1.種夕重微控制系統之指令執行方法,其係可應用於一多重微控制系統 中’該指令執行方法係包細下步驟: 透^主運算微控制單元提取一指令,且該指令係由一運算碼與至少一 多重微控制核心邏輯選擇運算元構成; 於該主運算微控制單元中進行該指令之一解碼任務; 以及 依據該解碼任務之絲,魅運算微賴單元職稱魏权—運算動 作,於該多重微控制核心邏輯選擇運算元所指定之至少一位址對應之微控 制單元中進行該運算動作之執行。 2·如申請專利範圍第1項所述之多重微控制系統之指令執行方法,其中, 該夕重微控制系統係由複數微控制器核心邏輯所組成,且任一該微控制器 核心邏輯射做為触運算微控制單元。 3. 如申请專她圍第2項所述之多重微控織統之齡執行紐,其中, 該多重微控键心__運算元係顧至少N個連續或科續讀元 數,以選擇對應之2%2N個以下之微控制核心邏輯,其中N為自然數。 4. 如申凊專利範圍第3項所述之多重微控制系統之指令執行方法,其中該 至少N個位元數位於棚,或運算,或—部分位於運算元内而其 他部分位於運算碼内。 5. 如申請專利細第1項所述之多重微控⑽、統之指令執行方法,其中, 該多重碰繼心邏_擇雜元除指I姐職之微控解科,亦 指定該微控制單元内部之一子單元。 6. 如申請專利範圍第2項所述之多重微控制系統之指令執行方法,其中, 17 1311721 ,a I? 任意 .疋之微控制單疋係可透過該指令與任意微控制器核心邏輯或是與 週邊單元進行溝通。 疋’、 種多重微控制系統,包含複數個微控制核心邏輯,每—微控制核心邏 輯具有-以二進位方式編碼之微控_心邏輯識別碼,其中,該系統於執 行與特定微控她心賴相關之指令時,在該齡巾,包含有該特定微控 制核心邏輯之微控制核心邏輯識別碼。1311721 ^ . ^· ;: X. Applying for a patent garden: 1. The instruction execution method of the singular micro-control system can be applied to a multi-micro control system. The main operation micro control unit extracts an instruction, and the instruction is composed of an operation code and at least one multi-micro control core logic selection operation unit; performing a decoding task of the instruction in the main operation micro control unit; According to the silk of the decoding task, the enchantment operation micro-relay unit title Wei right-operation operation performs the operation operation in the micro-control unit corresponding to at least one address specified by the multiple micro-control core logic selection operation element. 2. The instruction execution method of the multiple micro control system according to claim 1, wherein the digital micro control system is composed of a plurality of microcontroller core logics, and any of the microcontroller core logic shots As a touch computing micro control unit. 3. If applying for the multi-micro-control woven age implementation of the second item described in item 2, the multi-micro control key __ operation element is to select at least N consecutive or continuous reading elements to select Corresponding to 2% 2N or less of micro-control core logic, where N is a natural number. 4. The method for executing an instruction of a multiple micro-control system according to claim 3, wherein the at least N number of bits are located in a shed, or an operation, or - part is located in the operation unit and the other part is located in the operation code . 5. The method for executing multiple micro-controls (10) and the command execution method as described in the first paragraph of the patent application, wherein the multi-touch successor _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ One of the subunits inside the control unit. 6. The method for executing an instruction of a multiple micro-control system as described in claim 2, wherein: 17 1311721, a I? arbitrary micro-control unit can pass the instruction with any microcontroller core logic or It is to communicate with the surrounding units.疋', a multi-micro-control system, comprising a plurality of micro-control core logics, each micro-control core logic has a micro-control _ heart logic identification code encoded in a binary manner, wherein the system performs and specific micro-controls her At the time of the relevant instruction, the micro-control core logic identification code of the specific micro-control core logic is included in the age towel. 8. -種多重微控齡統’包含複數個微控伽㈣輯,每—微控制核心邏 輯具有一以二進位方式編碼之微控制核心邏輯識別碼,其中’該系統於執 行與特定微控制核心邏輯相關之指令時,該指令從—個儲存微控制核心邏 輯識別碼的位置取得微控制核心邏輯識別碼。 9.如申請專利範圍第8項所述之多重微控制系統,其中該指令包含運算碼 位置與運算元位置,而該微控制核心邏輯識別碼位於運算元位置内,成運 算碼位置内,或一部分位於運算元位置内而其他部分位於運算碼位置内。8. A multi-micro-control system consists of a plurality of micro-control gamma (four) series, each micro-control core logic has a micro-control core logic identification code encoded in binary mode, where 'the system is executed and specific micro-control When the core logic is associated with an instruction, the instruction obtains the micro control core logic identification code from a location where the micro control core logic identification code is stored. 9. The multiple micro-control system of claim 8, wherein the instruction comprises an opcode position and an operand position, and the micro-control core logic identifier is located within the operand position, within the opcode position, or One part is in the operand position and the other part is in the opcode position.
TW095107574A 2006-03-07 2006-03-07 Multiple micro control system and commands and command execution method thereof TW200734921A (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
TW095107574A TW200734921A (en) 2006-03-07 2006-03-07 Multiple micro control system and commands and command execution method thereof
KR1020060119678A KR20070092093A (en) 2006-03-07 2006-11-30 Multiple microcontroller system, instruction, and instruction execution method for the same

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW095107574A TW200734921A (en) 2006-03-07 2006-03-07 Multiple micro control system and commands and command execution method thereof

Publications (2)

Publication Number Publication Date
TW200734921A TW200734921A (en) 2007-09-16
TWI311721B true TWI311721B (en) 2009-07-01

Family

ID=38689660

Family Applications (1)

Application Number Title Priority Date Filing Date
TW095107574A TW200734921A (en) 2006-03-07 2006-03-07 Multiple micro control system and commands and command execution method thereof

Country Status (2)

Country Link
KR (1) KR20070092093A (en)
TW (1) TW200734921A (en)

Also Published As

Publication number Publication date
KR20070092093A (en) 2007-09-12
TW200734921A (en) 2007-09-16

Similar Documents

Publication Publication Date Title
CN106547518B (en) The device and method that low latency for accelerator calls
CN106293640B (en) Hardware processor, method and the hardware device of Heterogeneous Computing for close-coupled
CN103562856B (en) The pattern that strides for data element is assembled and the scattered system of the pattern that strides of data element, device and method
JP5646656B2 (en) Mapping between registers used by multiple instruction sets
TWI525533B (en) Systems, apparatuses, and methods for performing mask bit compression
CN104813281B (en) The device and method of quick failure handling for instruction
CN104838357B (en) Vectorization method, system and processor
CN104011647B (en) Floating-point rounding treatment device, method, system and instruction
CN104011645B (en) For generating integer phase difference constant integer span wherein in continuous position and smallest positive integral is from the processor of the integer sequence of zero offset integer shifts, method, system and medium containing instruction
CN107003843A (en) Method and apparatus for performing about reducing to vector element set
CN107077329A (en) Method and apparatus for realizing and maintaining the stack of decision content by the stack synchronic command in unordered hardware-software collaborative design processor
CN106406818A (en) Packed data operation mask concatenation processor, method, system and instruction
TW201738758A (en) Instructions and logic to provide base register swap status verification functionality
CN110321159A (en) For realizing the system and method for chain type blocks operation
CN108804137A (en) For the conversion of double destination types, the instruction of cumulative and atomic memory operation
TW202246973A (en) Hardware processor and processor
JP3819686B2 (en) Matrix operation device and digital signal processing device having matrix operation function
TW201344573A (en) Vector frequency expand instruction
TW201732544A (en) Aggregate scatter instructions
CN109196467A (en) Source packet data is divided into processor, method, system and the instruction of access
CN109992299A (en) System and method for calculating the scalar product of the nibble in two blocks operation numbers
CN108139911A (en) In the same execution packet of vliw processor specification is performed using having ready conditions for the instruction for expansion slot of having ready conditions
JPS5621242A (en) Pipeline control method for computer operation
TWI490781B (en) Apparatus and method for selecting elements of a vector computation
CN108292249A (en) The device and method of execution for reserved bit