JPS6119064B2 - - Google Patents
Info
- Publication number
- JPS6119064B2 JPS6119064B2 JP55070807A JP7080780A JPS6119064B2 JP S6119064 B2 JPS6119064 B2 JP S6119064B2 JP 55070807 A JP55070807 A JP 55070807A JP 7080780 A JP7080780 A JP 7080780A JP S6119064 B2 JPS6119064 B2 JP S6119064B2
- Authority
- JP
- Japan
- Prior art keywords
- central processing
- sub
- program
- memory device
- processing unit
- Prior art date
- Legal status (The legal status 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 status listed.)
- Expired
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/38—Concurrent instruction execution, e.g. pipeline or look ahead
- G06F9/3836—Instruction issuing, e.g. dynamic instruction scheduling or out of order instruction execution
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/38—Concurrent instruction execution, e.g. pipeline or look ahead
- G06F9/3867—Concurrent instruction execution, e.g. pipeline or look ahead using instruction pipelines
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/38—Concurrent instruction execution, e.g. pipeline or look ahead
- G06F9/3885—Concurrent instruction execution, e.g. pipeline or look ahead using a plurality of independent parallel functional units
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multi Processors (AREA)
- Advance Control (AREA)
Description
【発明の詳細な説明】
本発明は副中央処理装置を使用してプログラム
を並列処理し、実行時間を短縮させた情報処理装
置のプログラム並列処理方式に関する。DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a program parallel processing method for an information processing apparatus that uses a sub-central processing unit to process programs in parallel and shortens execution time.
一般に中央処理装置とメモリ装置とで構成され
る情報処理装置は、処理実行用のプログラムをメ
モリ装置に予め蓄積しておき、それを読出し解析
して所定のデータ処理を実行している。このとき
中央処理装置は単一であつて、蓄積されているプ
ログラムはその当初から順次に処理を進めなけれ
ばならない。したがつてデータを一緒に処理する
ことでなく単にプログラムについてそれを並列的
に処理を進めて貰い処理終了後に再び単一装置で
処理した方が時間的に早くできる見込みがあつて
も具体的に処理できる装置が存在しなかつた。 2. Description of the Related Art In general, an information processing device including a central processing unit and a memory device stores a program for executing processing in the memory device in advance, reads and analyzes the program, and executes predetermined data processing. At this time, there is a single central processing unit, and the stored programs must be processed sequentially from the beginning. Therefore, rather than processing data together, it is possible to simply have the program process it in parallel, and then process it again on a single device after the processing is completed, which would be faster in terms of time. There was no equipment that could process it.
本発明の目的は前述の欠点を改善し、副中央処
理装置を設けることによりプログラムを並列処理
させ、実行時間を短縮する情報処理装置のプログ
ラム並列処理方式を提供することにある。 SUMMARY OF THE INVENTION An object of the present invention is to provide a program parallel processing method for an information processing apparatus that improves the above-mentioned drawbacks and reduces execution time by processing programs in parallel by providing a sub-central processing unit.
以下図面に示す本発明の実施例について説明す
る。第1図は本発明の第1実施例のブロツク構成
図であつて、主中央処理装置MCCと主メモリ装
置MMとで構成される情報処理装置が従来の一般
装置を示し、1点鎖線で囲む部分が本発明により
追加した装置である。SCC1,SCC2,…SCCn
がn個の副中央処理装置、SCTは副中央処理装
置制御部、ADAはアドレス演算部、CTは制御
部、MCNはメモリ競合回路を示している。情報
処理を開始する前にプログラムについて解析を行
ない、分割でき且つそれを独立して処理できるも
のを見出し、並列処理単位に分割しておく。次に
そのプログラムのローデイングアドレス、プログ
ラム識別名、並列処理可能識別フラグを設定し、
それらについて主メモリ装置MM内にローデイン
グを行なう。この動作は主として制御部CTが行
ない、制御部CTになお共通データの配置位置を
アドレス演算部ADAに渡しておく。主中央処理
装置MCCは主メモリ装置MMにおけるプログラ
ムによつて動作を開始した後、並列処理可能プロ
グラムを実行することとなつたとき、主中央処理
装置MCCは制御部CTに対し並列処理プログラム
を実行する旨通知する。制御部CTは副中央処理
装置SCCに起動をかけるため並列処理プログラ
ム識別名をアドレス演算部ADAに渡す。アドレ
ス演算部ADAはプログラム識別名と主メモリ装
置MM中のローデイングアドレスの対応づけを行
ない、副中央処理装置制御部SCTにアドレスを
報告する。副中央処理装置制御部SCTは副中央
処理装置SCCの空状態となつているものをハー
ドウエアにより探し出し、それに対し前記ローデ
イングアドレスを渡して起動する、即ち主メモリ
装置MMのアドレスを知りプログラムを読出し、
次の動作に入る。メモリ競合回路MCNは主中央
処理装置MCCと副中央処理装置SCCが複数動作
して、同時に主メモリ装置MMにアクセスするこ
とがあつたとき予め定めている優先順位に従つて
処理を行なう。このように副中央処理装置SCC
が動作しているとき、主中央処理装置MCCの予
定処理が終了すると副中央処理装置の処理待ちと
なるから、待時間をできるだけ少なくするため当
該副中央処理装置の処理を他の副中央処理装置の
処理より優先させる。通常は複数の副中央処理装
置SCCの処理が同時に或いは順次に終了し、終
了したことを主中央処理装置MCCに報告するか
ら、主中央処理装置は同期を保つて次の処理に入
り必要なときに再び副中央処理装置が動作する。 Embodiments of the present invention shown in the drawings will be described below. FIG. 1 is a block configuration diagram of a first embodiment of the present invention, in which the information processing device consisting of a main central processing unit MCC and a main memory device MM is a conventional general device, and is surrounded by a chain line. This section is a device added according to the present invention. SCC1, SCC2,...SCCn
is n sub-central processing units, SCT is a sub-central processing unit control section, ADA is an address calculation section, CT is a control section, and MCN is a memory contention circuit. Before starting information processing, the program is analyzed, programs that can be divided and processed independently are found, and the programs are divided into parallel processing units. Next, set the loading address, program identification name, and parallel processing flag for that program.
They are loaded into the main memory device MM. This operation is mainly performed by the control unit CT, which also passes the location of the common data to the address calculation unit ADA. After the main central processing unit MCC starts operating according to the program in the main memory unit MM, when it is decided to execute a program capable of parallel processing, the main central processing unit MCC executes the parallel processing program for the control unit CT. We will notify you that we will do so. The control unit CT passes the parallel processing program identification name to the address calculation unit ADA in order to start up the sub-central processing unit SCC. The address calculation unit ADA associates the program identification name with the loading address in the main memory device MM, and reports the address to the sub-central processing unit control unit SCT. The sub-central processing unit control unit SCT uses hardware to find an empty sub-central processing unit SCC, passes the loading address to it, and starts the sub-central processing unit. reading,
Start the next action. The memory contention circuit MCN performs processing according to a predetermined priority order when a plurality of main central processing units MCC and sub central processing units SCC operate and access the main memory device MM at the same time. In this way, the sub-central processor SCC
When the main central processing unit (MCC) is running, when the scheduled processing of the main central processing unit (MCC) is completed, it will wait for the processing of the sub-central processing unit. Priority is given to the processing of Normally, the processing of multiple sub-central processing units SCC is completed simultaneously or sequentially, and the completion is reported to the main central processing unit MCC, so the main central processing unit maintains synchronization and starts the next processing when necessary. The sub-central processing unit operates again.
主メモリ装置MMにはランプのような表示装置
を設けておき、副中央処理装置SCCの並列処理
要求がなされ、当該プログラムが読出されたと
き、表示装置を点灯させれば取扱者にとつて便利
である。 It is convenient for the operator if a display device such as a lamp is provided in the main memory device MM, and the display device is turned on when a parallel processing request is made to the sub-central processing unit SCC and the relevant program is read. It is.
次に第2図は第1図中の主メモリ装置MM以外
に、副中央処理装置SCCに対して更に別個の副
メモリ装置SMをも設けた本発明の第2実施例を
示し、副メモリ競合回路SMCNを設け所要の接続
をしておく。副メモリ装置SMには並列処理単位
に分割したプログラムについて格納しておく。こ
のとき副中央処理装置の動作時に主メモリ装置
MMをその都度アクセスせずに副メモリ装置SM
へのアクセスで済む。また処理データを蓄積して
おき後に再処理するような場合有利である。なお
中央処理装置MCCと副中央処理装置SCCと副中
央処理装置SCCとの共用データは主メモリ装置
内に蓄積しておくから、必要の都度副中央処理装
置は主メモリ競合回路MCNを介して参照する。 Next, FIG. 2 shows a second embodiment of the present invention in which, in addition to the main memory device MM shown in FIG. 1, a separate sub-memory device SM is also provided for the sub-central processing unit SCC. Set up the circuit SMCN and make the necessary connections. The sub memory device SM stores programs divided into parallel processing units. At this time, when the sub-central processing unit operates, the main memory device
Secondary memory device SM without accessing MM each time
All you need to do is access. It is also advantageous when processing data is stored and reprocessed later. Note that the data shared by the central processing unit MCC, sub-central processing unit SCC, and sub-central processing unit SCC is stored in the main memory device, so the sub-central processing unit can refer to it via the main memory contention circuit MCN whenever necessary. do.
このようにして本発明によると予め並列処理に
よつて動作可能なプログラムはメモリ装置に格納
されているため、副中央処理装置の動作が主中央
処理装置と或いは他の副中央処理装置と並列的に
行なわれて、プログラム実行時間が短縮され、情
報処理が効率良く行なわれる。 In this way, according to the present invention, since the program that can be executed by parallel processing is stored in the memory device in advance, the operation of the sub-central processing unit can be performed in parallel with the main central processing unit or with other sub-central processing units. As a result, program execution time is shortened and information processing is performed efficiently.
第1図・第2図は本発明の第1・第2実施例の
ブロツク構成図を示す。
MCC……主中央処理装置、MM……主メモリ
装置、SCC1,SCC2,…SCCn……副中央処理
装置、CT……制御部、MCN……主メモリ競合回
路、ADA……アドレス演算部、SCT……副中央
処理装置制御部、SM……副メモリ装置、SMCN
……副メモリ競合回路。
FIGS. 1 and 2 show block diagrams of first and second embodiments of the present invention. MCC...Main central processing unit, MM...Main memory device, SCC1, SCC2,...SCCn...Sub central processing unit, CT...Control unit, MCN...Main memory contention circuit, ADA...Address calculation unit, SCT ...Sub central processing unit control unit, SM...Sub memory device, SMCN
...Sub-memory contention circuit.
Claims (1)
中央処理装置が順次実行して行く情報処理装置の
プログラム処理方式において、 前記プログラムのうち、並列処理単位に分割で
きるものを予め分割して格納させたメモリ装置
と、前記主中央処理装置間には、制御部と、複数
の副中央処理装置と、メモリ競合回路とを設け、 主中央処理装置が並列処理情報を発したとき、
制御部とメモリ競合部の制御により、副中央処理
装置のうち空状態となつているものが分割された
プログラムを実行すること を特徴とする情報処理装置のプログラム並列処理
方式。 2 並列処理要求情報によりメモリ装置から所定
のプログラムが読出されたとき並列処理中である
ことを表示することをメモリ装置に設けたことを
特徴とする特許請求の範囲第1項記載のプログラ
ム並列処理方式。 3 副中央処理装置が使用する副メモリ装置を設
けたことを特徴とする特許請求の範囲第1項記載
のプログラム並列処理方式。[Claims] 1. In a program processing method for an information processing device in which a main central processing unit sequentially executes programs stored in advance in a memory device, programs that can be divided into parallel processing units are divided in advance. A control unit, a plurality of sub-central processing units, and a memory contention circuit are provided between the memory device storing the parallel processing information and the main central processing unit, and when the main central processing unit issues parallel processing information,
A program parallel processing method for an information processing device, characterized in that under the control of a control unit and a memory contention unit, one of the sub-central processing units that is in an empty state executes a divided program. 2. Program parallel processing according to claim 1, characterized in that the memory device is provided with a display indicating that parallel processing is in progress when a predetermined program is read from the memory device based on parallel processing request information. method. 3. The program parallel processing system according to claim 1, further comprising a sub-memory device used by the sub-central processing unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP7080780A JPS56168272A (en) | 1980-05-29 | 1980-05-29 | Parallel processing system for program of information processor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP7080780A JPS56168272A (en) | 1980-05-29 | 1980-05-29 | Parallel processing system for program of information processor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JPS56168272A JPS56168272A (en) | 1981-12-24 |
| JPS6119064B2 true JPS6119064B2 (en) | 1986-05-15 |
Family
ID=13442194
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP7080780A Granted JPS56168272A (en) | 1980-05-29 | 1980-05-29 | Parallel processing system for program of information processor |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPS56168272A (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4484272A (en) * | 1982-07-14 | 1984-11-20 | Burroughs Corporation | Digital computer for executing multiple instruction sets in a simultaneous-interleaved fashion |
| JPS61245239A (en) * | 1985-04-23 | 1986-10-31 | Toshiba Corp | Logical circuit system |
-
1980
- 1980-05-29 JP JP7080780A patent/JPS56168272A/en active Granted
Also Published As
| Publication number | Publication date |
|---|---|
| JPS56168272A (en) | 1981-12-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US5140682A (en) | Storage control apparatus | |
| DE3687805T2 (en) | INPUT AND CONTROL SYSTEM IN A VIRTUAL MACHINE SYSTEM. | |
| US20070028240A1 (en) | Information processing method and information processing apparatus for processing a plurality of threads | |
| JPS6119064B2 (en) | ||
| US5745757A (en) | Multiprocessor system with microprogrammed means for dispatching processes to processors | |
| CN110837411A (en) | Internal concurrent I/O scheduling method and system for data server side partition | |
| JP3092656B2 (en) | Test program execution control method | |
| JP2952896B2 (en) | Shared memory access method in multitask multiprocessor system | |
| JPH0991098A (en) | Device driver access system | |
| JPH07109586B2 (en) | Program structure material creation processor | |
| JPH035869A (en) | information processing equipment | |
| JPS63157257A (en) | Control system for resources of multi-processor constitution | |
| JPH05165680A (en) | Memory operation tracing device | |
| JPH0973435A (en) | Tightly coupled multiprocessor system test system. | |
| JPS6325731A (en) | Data processor | |
| JPS5822765B2 (en) | Program loading method in electronic computer system | |
| JPH03116261A (en) | Multiprocessor control system | |
| JPS61165147A (en) | Shared resource updating system on storage device | |
| EP0480590A2 (en) | Information processing system | |
| JPH02272650A (en) | Library synchronism checking system | |
| JPH07152574A (en) | Program loading method | |
| JPH04256024A (en) | Dividing and loading system for program | |
| JPH03232051A (en) | Semaphore access control method | |
| JPH02151956A (en) | Command procedure starting system | |
| JP2000122885A (en) | Diagnostic system and recording medium |