[go: up one dir, main page]

JPS6073766A - Inter-cpu information exchange device - Google Patents

Inter-cpu information exchange device

Info

Publication number
JPS6073766A
JPS6073766A JP58181365A JP18136583A JPS6073766A JP S6073766 A JPS6073766 A JP S6073766A JP 58181365 A JP58181365 A JP 58181365A JP 18136583 A JP18136583 A JP 18136583A JP S6073766 A JPS6073766 A JP S6073766A
Authority
JP
Japan
Prior art keywords
information
header record
opu
receiving
area
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.)
Granted
Application number
JP58181365A
Other languages
Japanese (ja)
Other versions
JPH0463424B2 (en
Inventor
Junichi Takai
純一 高井
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Meidensha Corp
Meidensha Electric Manufacturing Co Ltd
Original Assignee
Meidensha Corp
Meidensha Electric Manufacturing 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 Meidensha Corp, Meidensha Electric Manufacturing Co Ltd filed Critical Meidensha Corp
Priority to JP58181365A priority Critical patent/JPS6073766A/en
Publication of JPS6073766A publication Critical patent/JPS6073766A/en
Publication of JPH0463424B2 publication Critical patent/JPH0463424B2/ja
Granted legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F15/00Digital computers in general; Data processing equipment in general
    • G06F15/16Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs
    • G06F15/163Interprocessor communication
    • G06F15/167Interprocessor communication using a common memory, e.g. mailbox
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F15/00Digital computers in general; Data processing equipment in general
    • G06F15/16Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs
    • G06F15/163Interprocessor communication
    • G06F15/17Interprocessor communication using an input/output type connection, e.g. channel, I/O port

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Multi Processors (AREA)
  • Computer And Data Communications (AREA)
  • Exchange Systems With Centralized Control (AREA)

Abstract

PURPOSE:To attain efficient information exchange between CPUs by writing information in a two-port memory of a receiving side CPU by the originating side CPU and also writing its header record in the special address area of the two-port memory. CONSTITUTION:First, a CPU1 at the originating side writes a information data 5a except a header record in the area 6a of a two-port memory 6, then writes the header record 5b in the area 6b. A CPU2 at the receiving side is provided with an interrupt generation circuit 9 which gives an interrupt signal INT to a processor in the CPU2 after receiving a detection signal of an address decoder 8 which detects that the header record storage area 6b of the two-port memory 6 is accessed from the side of a system bus 3. Therefore, when the CPU1 writes the header record 5b in the area 6b, the interrupt signal INT automatically occurs and the CPU2, upon receiving the signal, fetches information 5a and 5b in the two-port memory 6.

Description

【発明の詳細な説明】 この発明は、複数のopty(中央処理装置)が共通の
システムバスで結合されたマルチOPUシステムにおい
て、OPU相互間で情報を授受するための情報交換装置
に関する。
DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an information exchange device for exchanging information between OPUs in a multi-OPU system in which a plurality of opties (central processing units) are connected via a common system bus.

マルチOPUシステムにおける0PUISQで情報交換
を行なうための手段としては、従来からいくつかの方式
が知られている。代表的な8種の従来方式について、そ
の概念t−第1図(A) (B)(o)にそれぞれ図解
している。
Several methods are conventionally known as means for exchanging information using 0PUISQ in a multi-OPU system. The concepts of eight typical conventional methods are illustrated in FIGS. 1A, 1B, and 1O.

同図(A)は、システムバス3に接続されているシステ
ムメモリ参の所定エリアに発信側の0PTT/がデータ
jを書き込み、受信側の0PUJがこのエリアをスキャ
ニングすることによってデータSを取9込む方式である
。この方式ではデータの授受のために特定の7ラグが使
用されるので、これをフラグスキャニング方式と呼ぶ。
9A shows that 0PTT/ on the sending side writes data j to a predetermined area of the system memory connected to system bus 3, and 0PUJ on the receiving side retrieves data S by scanning this area. This is a method that involves In this method, seven specific lags are used for sending and receiving data, so this is called a flag scanning method.

上記フラグスキャニング方式では、受信側のCPU、2
がフラグをスキャンしたシデータ會取シ扱うためにシス
テムバス3を使用することが大きな欠点である。この使
用#I4度が高いとシステム全体の効率が著しく低下す
る。
In the above flag scanning method, the receiving side CPU, 2
A major drawback is the use of the system bus 3 to handle the system data session that scanned the flag. If this usage #I4 degree is high, the efficiency of the entire system will be significantly reduced.

同図(B)は、受信側のCPUJ内の2ボートメモリぶ
を情報伝送エリアとして使用する方式である。
FIG. 2B shows a method in which two memory ports in the CPUJ on the receiving side are used as the information transmission area.

2ボートメモリ乙はOPU、2の内部プロセッサからは
システムバス3を介さずに直接アクセスされ。
2 boat memory B is accessed directly from the OPU and the internal processor of 2 without going through the system bus 3.

他のOPU/からはシステムバス3を介してアクセスさ
れる。データの授受には先の方式と同様に7ラグを使用
し、OPU/が2ボートメモリ乙に書き込んだデータj
を、0PU2が内部処理でスキャニングして取シ込む。
It is accessed via the system bus 3 from other OPUs. As with the previous method, 7 lags are used for data exchange, and the data written by OPU/ to 2-board memory B is
is scanned and imported by 0PU2 through internal processing.

この方式を2ボートメモリ・フラグスキャニング方式と
呼ぶ。
This method is called a 2-boat memory flag scanning method.

上記2ボートメモリ・フラグスキャニング方式では、デ
ータ受信のために0PU2かシステムバス3t−使用し
ないという点が先の方式よシ勝れている。しかし、これ
ら2つの方式では共通して。
The above-mentioned two-boat memory flag scanning method is superior to the previous method in that it does not use 0PU2 or system bus 3t for data reception. However, these two methods have the same thing in common.

受信側のOPUλは上1己フラグを常時スキャンするこ
とで受信すべきデータSを検出するので、このスキャニ
ング処理のためにOPU、2の能力の多くが裂かれる。
Since the OPU λ on the receiving side detects the data S to be received by constantly scanning the upper 1 flag, much of the capability of the OPU 2 is depleted due to this scanning process.

したがって、他の処理に0アU2の能力を充分に発揮す
ることができないという問題がある。
Therefore, there is a problem that the ability of 0A U2 cannot be fully utilized for other processing.

同図(0)は、OPU/が0PU2内の2ボートメモリ
4にデータjを誓き込んだ後、特定の工10レジスタ7
(ボート)をアクセスすることによって0PU2に割シ
込み信号INTを与える方式である。この割込み信号工
NTを受けて0PU2はデータjを取り込む。この方式
を割込み方式と呼ぶ。
In the same figure (0), after OPU/ commits data j to 2-vote memory 4 in 0PU2, a specific device 10 register 7
In this method, an interrupt signal INT is given to 0PU2 by accessing the 0PU2. In response to this interrupt signal NT, 0PU2 takes in data j. This method is called an interrupt method.

上記割込み方式では、発信側のOPU/において、デー
タをメモリ乙にセットする処理の他に、割込み信号を発
生する処理が付加されるので、その分だけ処理効率が低
下する。また、発信側か受信側のいずれかのOPHに割
込み用の回路(工10レジスタ7)が必要になる。
In the above-mentioned interrupt method, in addition to the process of setting data in memory B, the process of generating an interrupt signal is added to the OPU/ on the originating side, so the processing efficiency decreases accordingly. Further, an interrupt circuit (register 7) is required on either the transmitting side or the receiving side OPH.

この発明は前述し九従来の問題点に鑑みなされたもので
あり、その目的は、マルチOPUシステムにおけるCP
U間のデータの授受を、発信側および受信側ともに効率
よく処理することがで自るようにした情報交換装置を提
供することにある。
This invention was made in view of the nine conventional problems mentioned above, and its purpose is to improve the CPU speed in a multi-OPU system.
It is an object of the present invention to provide an information exchange device in which both the transmitting side and the receiving side can efficiently process the exchange of data between U's.

上記の目的を達成するために、この発明は、発信側のO
PUが受信側CPU内の2ボートメモリに情報データを
書き込み、続いてその情報のヘッダレコードを上記2ボ
ートメモリの予め決められたアドレスエリアに書き込む
こととし、受信側OPHには上記の特定アドレスにヘッ
ダレコードを書き込む操作があったことを検出して割込
み信号を発生する回路を設けたことを特徴とする。
In order to achieve the above object, the present invention provides O
The PU writes information data to the 2-boat memory in the receiving side CPU, and then writes the header record of the information to a predetermined address area of the 2-boat memory, and the receiving side OPH writes the information data to the above specified address. The present invention is characterized in that it includes a circuit that detects that there is an operation to write a header record and generates an interrupt signal.

以下、この発明の実施例を図面に基づいて詳細に説明す
る。
Hereinafter, embodiments of the present invention will be described in detail based on the drawings.

第2図はこの発明による情報交換装置を示している。こ
こでは受信側のOPU、2内における2ボートメモリ6
がデータ授受に使用される。2ボートメモリjには、伝
送すべき情報のうちのヘッダレコードを除く実質的な情
報データを格納するエリア6aと、上記ヘッダレコード
を格納する特定アドレスのヘッダレコード格納エリア6
bが割当てられている。ヘッダレコードとは、情報デー
タの数や発信OPUのステーション番号などの情報を含
んだ伝送情報の見出し部分である。
FIG. 2 shows an information exchange device according to the invention. Here, 2-board memory 6 in OPU 2 on the receiving side
is used for data exchange. The two-board memory j includes an area 6a for storing substantial information data excluding header records among the information to be transmitted, and a header record storage area 6 at a specific address for storing the header records.
b is assigned. The header record is a heading portion of transmission information that includes information such as the number of information data and the station number of the originating OPU.

発信側のOPU/は、情報発信に際して、まずヘンダレ
コードを除く情報データjaを2ボートメモリ乙のエリ
アjaK1jき込み、次にヘッダレコードjbを2ボー
トメモリ乙のエリア≦bに書き込む。
When transmitting information, OPU/ on the transmitting side first writes information data ja excluding the hander record into area jaK1j of 2-boat memory B, and then writes header record jb to area ≦b of 2-boat memory B.

受信側のOPUλには、2ボートメモリ6のへ7りL/
コ−)”格納エリア6bがシステムバス3側からアクセ
スされたことを検出するためのアドレスデコーダざの検
出信号を受けてOPU、2内のプロセッサに割込み信号
INTを与える割込み発生回路9が設けられている。
OPUλ on the receiving side has 2-board memory 6 to 7 L/
An interrupt generation circuit 9 is provided which receives a detection signal from an address decoder for detecting that the storage area 6b is accessed from the system bus 3 and supplies an interrupt signal INT to the processor in the OPU 2. ing.

し友がって、OPU/がエリアabにヘッダレコードj
l)を臀き込む操作を行なうと5割込み信号工NTが自
動的に発生し、これを受けてOPUコは2ボートメモリ
6中の情報ja 、jbを取シ込む。
As a friend, OPU/ writes a header record j to area ab.
1), a 5-interrupt signal NT is automatically generated, and in response to this, the OPU inputs information ja and jb in the 2-board memory 6.

この装置において、システムバス31iIIIから見た
メモリマツプを第8図に示している。同図のように、一
連のメモリ空間内に0PUJ内の2ボートメモリ4が存
在し、この部分が上述した情報データ格納エリア6aと
ヘッダレコード格納エリア6bとに分かれている。シス
テムバス3上のOPU、2以外の他のOPUがヘッダレ
コード格納エリアAl)をアクセスすると、自動的にo
 pUJK割込みがかかる。
In this device, the memory map seen from the system bus 31iIII is shown in FIG. As shown in the figure, there are two boat memories 4 in 0PUJ in a series of memory spaces, and this part is divided into the above-mentioned information data storage area 6a and header record storage area 6b. When an OPU other than OPU 2 on system bus 3 accesses the header record storage area Al), OPU automatically
pUJK interrupt occurs.

以上詳細に説明したように、この発明の装置にあっては
、受信gill OP Uはデータ受けとシのためにシ
ステムバスを使用しないとともに、前述した7ラグス中
ヤニングのように常時データの監視処理を行なう必要は
なく1割込みを受けて始めてデータの取9込み処理を行
なう。また発信側OPUは、伝送すべき情報をヘッダレ
コードとその他の情報データとを区別して2ボートメモ
リに書き込むだけで、その他の付加的な処理をまったく
行なわずに、自動的に受信側OPHに割込みがかけられ
る。すなわち発信側および受信側ともに極めて効率よく
データ授受の処理が行なえ、したがってシステム全体と
しての能力が向上する。
As explained above in detail, in the device of the present invention, the receiving gill OP U does not use the system bus for data reception and transmission, and also performs constant data monitoring processing like the aforementioned 7-lag process. There is no need to carry out the data fetching process, and data fetching processing is performed only after receiving one interrupt. In addition, the transmitting OPU simply writes the information to be transmitted into the two-board memory by distinguishing the header record from other information data, and automatically interrupts the receiving OPH without performing any other additional processing. is applied. In other words, both the transmitting side and the receiving side can process data exchange extremely efficiently, thereby improving the performance of the entire system.

【図面の簡単な説明】[Brief explanation of the drawing]

第1図(ム) CB) (0)は従来の代表的な8種の
情報交換方式の説明図、第2図はこの発明の一実施例に
よる情報交換装置の説明図、第8図は第2図の装置にお
けるシステムバス側から見たメモリマツプである。 /・・・発信側0ア■、、2・・・受信側OPU、j・
・・システムバス、ja・・・情報データ、jb・・・
ヘラダレ;l−)”、≦・・・2ボートメモリ、≦b・
・・ヘッダレコード格納エリア、l・・・アドレスデコ
ーダ、?・・・割込み発生回路。 第2図 第3図
FIG. 1 (M) CB) (0) is an explanatory diagram of eight typical conventional information exchange systems, FIG. 2 is an explanatory diagram of an information exchange device according to an embodiment of the present invention, and FIG. 2 is a memory map of the device shown in FIG. 2 as seen from the system bus side. /... Sending side 0a ■,, 2... Receiving side OPU, j.
...System bus, ja...information data, jb...
Helladare;l-)”, ≦...2 boat memory, ≦b・
...Header record storage area, l...Address decoder, ? ...Interrupt generation circuit. Figure 2 Figure 3

Claims (1)

【特許請求の範囲】[Claims] (1)opty間で2ボートメモリを用いて情報交換す
るものにおいて、受信側OPHに、発信側OPUからの
情報データを格納する情報データ格納エリアと、予めア
ドレスを特定しているヘッダレコード格納エリアと、こ
のヘッダレコード格納エリアを受信側OPUがアクセス
するのを検出して検出信号を出力するアドレスデコーダ
と、上記検出信号によ)受信側OPUに割込信号を与え
る割込発生回路とを設けたことを特徴とするOPU間の
情報交換装置。
(1) In the case where information is exchanged between opties using two-board memory, the receiving side OPH has an information data storage area for storing information data from the sending side OPU, and a header record storage area whose address is specified in advance. , an address decoder that detects when the receiving OPU accesses this header record storage area and outputs a detection signal, and an interrupt generation circuit that provides an interrupt signal to the receiving OPU (based on the detection signal). An information exchange device between OPUs, characterized in that:
JP58181365A 1983-09-29 1983-09-29 Inter-cpu information exchange device Granted JPS6073766A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP58181365A JPS6073766A (en) 1983-09-29 1983-09-29 Inter-cpu information exchange device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP58181365A JPS6073766A (en) 1983-09-29 1983-09-29 Inter-cpu information exchange device

Publications (2)

Publication Number Publication Date
JPS6073766A true JPS6073766A (en) 1985-04-25
JPH0463424B2 JPH0463424B2 (en) 1992-10-09

Family

ID=16099447

Family Applications (1)

Application Number Title Priority Date Filing Date
JP58181365A Granted JPS6073766A (en) 1983-09-29 1983-09-29 Inter-cpu information exchange device

Country Status (1)

Country Link
JP (1) JPS6073766A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS63128464A (en) * 1986-11-18 1988-06-01 Nec Corp Processor circuit
JPH01199260A (en) * 1986-10-29 1989-08-10 United Technol Corp <Utc> Formation of a plurality of multiple processor system
JP2008521114A (en) * 2004-11-24 2008-06-19 コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ Coherent caching of local memory data

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS5387138A (en) * 1977-01-12 1978-08-01 Toshiba Corp Multiplex data processor
JPS5731072A (en) * 1980-07-31 1982-02-19 Mitsubishi Electric Corp Multiprocessor

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS5387138A (en) * 1977-01-12 1978-08-01 Toshiba Corp Multiplex data processor
JPS5731072A (en) * 1980-07-31 1982-02-19 Mitsubishi Electric Corp Multiprocessor

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH01199260A (en) * 1986-10-29 1989-08-10 United Technol Corp <Utc> Formation of a plurality of multiple processor system
JPS63128464A (en) * 1986-11-18 1988-06-01 Nec Corp Processor circuit
JP2008521114A (en) * 2004-11-24 2008-06-19 コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ Coherent caching of local memory data

Also Published As

Publication number Publication date
JPH0463424B2 (en) 1992-10-09

Similar Documents

Publication Publication Date Title
US5991843A (en) Method and system for concurrent computer transaction processing
JPS6113629B2 (en)
JPS6073766A (en) Inter-cpu information exchange device
US6647450B1 (en) Multiprocessor computer systems with command FIFO buffer at each target device
JP3078000B2 (en) Information processing device
JPS59173828A (en) Data processing system
JPS599928B2 (en) Channel control method
JP2735246B2 (en) Test and set method
JPH069036B2 (en) Input/Output Control Unit
JPS60169966A (en) Data processor
JP2001306411A (en) Information processor and its method
JP2504754B2 (en) Central processing unit
KR100580933B1 (en) Cache enabling architecture and magnetical hard disk drive device
JP3219422B2 (en) Cache memory control method
JPH0214741B2 (en)
JPS6379161A (en) semiconductor storage device
JPH02181550A (en) Communication control equipment and atm exchange provided with the same
JPS60247767A (en) Simultaneous input and output system for reading-out and writing-in operation of shared memory
JPH036766A (en) Multi-address communication system in multi-processor
JPH03204755A (en) Data transfer device
JPS5999522A (en) Input and output control system
JPH01279358A (en) Information transfer system
JPH0571116B2 (en)
JPH11119814A (en) Plant control system
JPS63289665A (en) Data transfer system for multi-processor