JPH04175941A - Method and device for controlling common memory - Google Patents

Method and device for controlling common memory

Info

Publication number
JPH04175941A
JPH04175941A JP2304440A JP30444090A JPH04175941A JP H04175941 A JPH04175941 A JP H04175941A JP 2304440 A JP2304440 A JP 2304440A JP 30444090 A JP30444090 A JP 30444090A JP H04175941 A JPH04175941 A JP H04175941A
Authority
JP
Japan
Prior art keywords
cpus
control circuit
memory
cpu
shared memory
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.)
Pending
Application number
JP2304440A
Other languages
Japanese (ja)
Inventor
Akira Hoshino
星野 公
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.)
Fuji Electric Co Ltd
Fuji Facom Corp
Original Assignee
Fuji Electric Co Ltd
Fuji Facom Corp
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 Fuji Electric Co Ltd, Fuji Facom Corp filed Critical Fuji Electric Co Ltd
Priority to JP2304440A priority Critical patent/JPH04175941A/en
Publication of JPH04175941A publication Critical patent/JPH04175941A/en
Pending legal-status Critical Current

Links

Abstract

PURPOSE:To efficiently access the common memory which is accessed from plural CPUs by operating a memory control circuit with the same clock with an accessing CPU. CONSTITUTION:This device is equipped with the CPUs 1-1 - 1-n which operate with 1st individual operation clocks CLK1 - CLKn and the common memory 2 which can be accessed by the respective CPUs 1-1 - 1-n. Then the device is provided with a clock selecting circuit 4 which inputs access request signals REQ1 - REQn of the respective CPUs 1-1 - 1-n and the respective operation clocks CLK1 - CLKn and supplies the memory control circuit 3 with the operation clock CLK of a CPU selected by the memory control circuit 3 as an operation clock CLKX. Consequently, the memory 2 that the CPUs 1-1 - 1-n share is accessed with most efficiency matching the speeds of the CPUs.

Description

【発明の詳細な説明】[Detailed description of the invention]

【産業上の利用分野1 本発明は複数のCPUからアクセスされる共有メモリ(
マルチボートメモリともいう)を効率よくアクセスする
ための共有メモリの制御方法およびその装置に関する。 なお以下各図において同一の符号は同一もしくは相当部
分を示す。 【従来の技術】 第2図はこの種の共有メモリの制御回路の構成例を示す
。同図において、1 (1−1〜1−n)はマイクロプ
ロセッサ(CPUとも記す)、2は共有メモリ(単にメ
モリともいう)、3はCPU 1〜1〜1−nとメモリ
2との間に設けられたメモリ制御回路である。 ここで1−1〜1−nは異なるマイクロプロセンサ(C
PU)を示し、CL K (CLKI、CLに2−−−
CLkn)は、各々CP 01−1. CP Ul−2
・・CP Ul−nに与えられる動作クロックである。 この動作り07りCLKI、CLK2・・CLKnは多
くの場合、周波数がそれぞれ異なる。そのためCP U
l−1,CP Ul−2・・CP Ul−nの動作速度
もそれぞれ異なる。 メモリ制御回路3は各CPUIからのメモリアクセスが
メモリ2上で衝突しないように調停する機能を持つ。ア
クセス要求信号RE Q (REQI、REQ2゜−−
−−REQn)は、それぞれCP U 1−1〜1−n
が共有メモリ2をアクセスしようとする信号である。こ
こでメモリ制御回路3はアクセス要求信号REQを出力
したCPUIの1つを予め定められた優先順位の決定方
法に従って選択し、その選択されたCPUIのアクセス
要求信号REQを共有メモリ2に与え、該CPUIが共
有メモリ2をアクセスできるようにする。 このような構成では、動作の速いCPUの性能を確保す
るため、メモリ制御回路3の動作クロックCLKXには
、各CPUの動作クロックCLKI 。 CLK2・・・CLKnのうち最も周波数の高いものを
用いることがある。
[Industrial Application Field 1] The present invention is a shared memory accessed by multiple CPUs (
The present invention relates to a shared memory control method and device for efficiently accessing shared memory (also called multi-board memory). Note that in the following figures, the same reference numerals indicate the same or corresponding parts. 2. Description of the Related Art FIG. 2 shows an example of the configuration of a control circuit for this type of shared memory. In the figure, 1 (1-1 to 1-n) is a microprocessor (also referred to as CPU), 2 is a shared memory (also simply referred to as memory), and 3 is a link between CPUs 1 to 1 to 1-n and memory 2. This is a memory control circuit provided in the. Here, 1-1 to 1-n are different microprosensors (C
PU) and CL K (CLKI, CL to 2---
CLkn) are each CP 01-1. CP Ul-2
. . . An operating clock given to the CP Ul-n. During this operation, CLKI, CLK2, . . . CLKn often have different frequencies. Therefore, CPU
The operation speeds of CP Ul-1, CP Ul-2, . . . CP Ul-n are also different. The memory control circuit 3 has a function of arbitrating so that memory accesses from each CPUI do not conflict on the memory 2. Access request signal REQ (REQI, REQ2゜--
--REQn) are CPU 1-1 to 1-n, respectively.
is a signal that attempts to access shared memory 2. Here, the memory control circuit 3 selects one of the CPUIs that has outputted the access request signal REQ according to a predetermined priority determination method, provides the access request signal REQ of the selected CPUI to the shared memory 2, and sends the access request signal REQ of the selected CPUI to the shared memory 2. Allow the CPUI to access the shared memory 2. In such a configuration, in order to ensure the performance of a fast-operating CPU, the operating clock CLKX of the memory control circuit 3 is set to the operating clock CLKI of each CPU. The highest frequency among CLK2...CLKn may be used.

【発明が解決しようとする課題】[Problem to be solved by the invention]

前述のようにメモリ制御回路3の動作クロックCLKX
として、各CPUの動作クロック中の最高周波数のクロ
ックを用いた場合、制御回路3゜メモリ2とも最も高い
周波数で動作することが要求され、これらの部品が高価
となるが、この最も高い周波数で動作するCPUIの共
有メモリ2へのアクセス頻度が低い時には、高価な部品
が充分活用されないという問題が生じる。またCPU動
作クロりクCLKI 、 CLK2・・・CLKnの内
から周波数の低いものを選ぶと、動作の速いCPUには
、待ち時間が増え、CPUの性能低下が生じる。 そこで本発明は複数CPUが共有するメモリを、そのC
PUの速度に合わせて最も効率よくアクセスできるよう
にすることができる共有メモリの制御方法およびその装
置を提供することを課題とする。
As mentioned above, the operation clock CLKX of the memory control circuit 3
If the highest frequency clock among the operating clocks of each CPU is used, both the control circuit 3 and the memory 2 are required to operate at the highest frequency, which makes these parts expensive. When the operating CPUI accesses the shared memory 2 infrequently, a problem arises in that expensive components are not fully utilized. Furthermore, if one of the CPU operation clocks CLKI, CLK2, . Therefore, the present invention provides memory that is shared by multiple CPUs.
An object of the present invention is to provide a shared memory control method and device that can provide the most efficient access in accordance with the speed of a PU.

【課題を解決するための手段】[Means to solve the problem]

前記の課題を解決するために、請求項1)の制御方法は
「それぞれ個別の第1の動作クロック(CL Kなど〕
で動作する複数のCPU (1など)と、 前記の各CPUによるアクセスが可能な共有メモリ(2
など)と、 第2の動作クロック(CLKXなと)で動作し、前記の
各CPUのアクセス要求信号(RE Qなど)を入力し
、要求を出したCPUの1つを選択し、該CPUの前記
アクセス要求信号を前記共有メモリに与えるメモリ制御
回路(3など)とを備えたシステムにおいて、 前記第2の動作クロックを前記共有メモリにアクセスす
ることとなった前記CPUの第1の動作クロックに、こ
のアクセスのつど等しくするようにjするものとし、ま
た 請求項2)の制御装置は、rそれぞれ個別の第1の動作
クロック (CLKなと)で動作する複数のCPU (
1など)と、 前記の各CPUによるアクセスが可能な共有メモリ(2
など)と、 第2の動作クロック(CLKXなど)で動作し、前記の
各CPUのアクセス要求信号(REQなど)を入力し、
要求を出したCPUの1つを選択し、該CPUの前記ア
クセス要求信号を前記共有メモリに与えるメモリ制御回
路(3など)とを備えたシステムにおいて、 前記の各CPUからのアクセス要求信号と、この各C,
P、Uの前記第1の動作クロックとを人力し、前記メモ
リ制御回路によって選択されるCPUについての第1の
動作クロックを前記第2の動作りロックとして前記メモ
リ制御回路に与える手段(クロック選択回路4など)を
備えたことを特徴とする共有メモリの制御回路。
In order to solve the above-mentioned problem, the control method of claim 1) uses "separate first operating clocks (CLK, etc.)".
multiple CPUs (such as 1) that operate on a shared memory (2) that can be accessed by each of the CPUs.
etc.), operates with a second operating clock (CLKX etc.), inputs the access request signal (REQ, etc.) of each CPU mentioned above, selects one of the CPUs that issued the request, and In a system comprising a memory control circuit (such as 3) that provides the access request signal to the shared memory, the second operating clock is set to the first operating clock of the CPU that has accessed the shared memory. , so that the accesses are made equal each time, and the control device according to claim 2) includes a plurality of CPUs (r) each operating with an individual first operating clock (CLK).
1 etc.), and a shared memory (2
etc.), operates with a second operating clock (CLKX, etc.), and inputs the access request signal (REQ, etc.) of each of the CPUs,
A system comprising a memory control circuit (such as 3) that selects one of the CPUs that has issued a request and provides the access request signal of the CPU to the shared memory, the access request signal from each of the CPUs; Each of these C,
Means (clock selection) for manually inputting the first operating clocks of P and U and providing the first operating clock for the CPU selected by the memory control circuit to the memory control circuit as the second operating lock. A shared memory control circuit characterized by comprising a circuit (4, etc.).

【作 用】[For use]

各CPUl−1〜l−nのアクセス要求信号REQ1〜
REQnと各動作クロックCLK1〜CLKnとを入力
し、メモリ制御回路3によって選択されるCPUについ
ての動作クロックCLKをメモリ制御回路3に対し動作
クロックCLKXとして与えるクロック選択回路4を設
ける。
Access request signals REQ1 to each CPU l-1 to l-n
A clock selection circuit 4 is provided which inputs REQn and each of the operating clocks CLK1 to CLKn and provides the operating clock CLK for a CPU selected by the memory control circuit 3 to the memory control circuit 3 as an operating clock CLKX.

【実施例】【Example】

第1図は本発明の実施例としての構成を示すブロック回
路図で、第2図に対応するものである。 第1図においては、メモリ制御回路3に入力する動作ク
ロックCLKXを各CP U 1−1〜1−nのクロッ
クCLKI〜CL Knから選択できるようにクロック
選択回路4を新たに設けた。即ち各CPU1−1〜1−
nからのアクセス要求信号REQI〜REQnおよび動
作クロックCLK1〜CLKnは、このクロック選択回
路4に人力されるように構成されており、クロック選択
回路4は、アクセス要求信号REQを出力したC P 
01のうちメモリ制御回路3によって選択されるべきC
P Ulの動作クロックCLKを選択し、動作クロック
CLKXとしてメモリ制御回路3に入力する。
FIG. 1 is a block circuit diagram showing the configuration of an embodiment of the present invention, and corresponds to FIG. 2. In FIG. In FIG. 1, a clock selection circuit 4 is newly provided so that the operating clock CLKX input to the memory control circuit 3 can be selected from among the clocks CLKI to CL Kn of each of the CPUs 1-1 to 1-n. That is, each CPU1-1 to 1-
The access request signals REQI to REQn and the operation clocks CLK1 to CLKn from the clock selection circuit 4 are configured to be manually inputted to the clock selection circuit 4, and the clock selection circuit 4 is configured to input the access request signals REQI to REQn and the operation clocks CLK1 to CLKn from the clock selection circuit 4, which outputs the access request signal REQ.
01 to be selected by the memory control circuit 3
The operating clock CLK of P Ul is selected and input to the memory control circuit 3 as the operating clock CLKX.

【発明の効果】【Effect of the invention】

本発明によれば、各CPUl−1〜1−nからのアクセ
ス要求信号REQ1〜REQnを、クロック選択回路4
に入力し、クロック選択回路4がこの要求REQを出し
たCPUIの動作クロックCLKを選択してメモリ制御
回路3に与えるようにしたので、メモリ制御回路3はア
クセスしたCPU1 と同一のクロックで動作すること
ができる。 そのため、高速のCPUの不要な待ち時間やメモリ制御
回路3および共有メモリ2の常時の過度の高速動作をな
くすことができる。
According to the present invention, the access request signals REQ1 to REQn from each CPU1-1 to 1-n are sent to the clock selection circuit 4.
Since the clock selection circuit 4 selects the operating clock CLK of the CPU that issued this request REQ and supplies it to the memory control circuit 3, the memory control circuit 3 operates with the same clock as the accessed CPU1. be able to. Therefore, unnecessary waiting time of the high-speed CPU and constant excessively high-speed operation of the memory control circuit 3 and shared memory 2 can be eliminated.

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

第1図は本発明の実施例としての構成を示すブロック回
路図、 第2図は第1図に対応する従来の回路図である。 1 (1−1〜1−n) : CP U、2:共有メモ
リ、3゜メモリ制御回路、4:クロツタ選択回路、CL
K(CLKI 〜CLKn)、CLKX :動作りo 
7り、REQ (REQI 〜REQn):アクセス要
求信号。 動f乍クロ・、り 、tlFA
FIG. 1 is a block circuit diagram showing a configuration as an embodiment of the present invention, and FIG. 2 is a conventional circuit diagram corresponding to FIG. 1. 1 (1-1 to 1-n): CPU, 2: Shared memory, 3° memory control circuit, 4: Crochet selection circuit, CL
K (CLKI ~ CLKn), CLKX: Operation o
7. REQ (REQI to REQn): Access request signal. Dynamic f 乍 Kuro, Ri, tlFA

Claims (1)

【特許請求の範囲】 1)それぞれ個別の第1の動作クロックで動作する複数
のCPUと、 前記の各CPUによるアクセスが可能な共有メモリと、 第2の動作クロックで動作し、前記の各CPUのアクセ
ス要求信号を入力し、要求を出したCPUの1つを選択
し、該CPUの前記アクセス要求信号を前記共有メモリ
に与えるメモリ制御回路とを備えたシステムにおいて、 前記第2の動作クロックを前記共有メモリにアクセスす
ることとなった前記CPUの第1の動作クロックに、こ
のアクセスのつど等しくするようにしたことを特徴とす
る共有メモリの制御方法。 2)それぞれ個別の第1の動作クロックで動作する複数
のCPUと、 前記の各CPUによるアクセスが可能な共有メモリと、 第2の動作クロックで動作し、前記の各CPUのアクセ
ス要求信号を入力し、要求を出したCPUの1つを選択
し、該CPUの前記アクセス要求信号を前記共有メモリ
に与えるメモリ制御回路とを備えたシステムにおいて、 前記の各CPUからのアクセス要求信号と、この各CP
Uの前記第1の動作クロックとを入力し、前記メモリ制
御回路によって選択されるCPUについての第1の動作
クロックを前記第2の動作クロックとして前記メモリ制
御回路に与える手段を備えたことを特徴とする共有メモ
リの制御回路。
[Scope of Claims] 1) A plurality of CPUs that each operate with an individual first operating clock, a shared memory that can be accessed by each of the CPUs, and a shared memory that operates with a second operating clock and that each of the CPUs a memory control circuit that inputs an access request signal of the CPU, selects one of the CPUs that has issued the request, and provides the access request signal of the CPU to the shared memory, A method for controlling a shared memory, characterized in that the first operating clock of the CPU that accesses the shared memory is made equal to the first operating clock of the CPU each time the access is made. 2) A plurality of CPUs that each operate with an individual first operating clock, a shared memory that can be accessed by each of the CPUs, and a shared memory that operates with a second operating clock and receives access request signals from each of the CPUs. and a memory control circuit that selects one of the CPUs that has issued a request and provides the access request signal of the CPU to the shared memory, C.P.
and means for inputting the first operating clock of the CPU selected by the memory control circuit to the memory control circuit as the second operating clock. shared memory control circuit.
JP2304440A 1990-11-09 1990-11-09 Method and device for controlling common memory Pending JPH04175941A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2304440A JPH04175941A (en) 1990-11-09 1990-11-09 Method and device for controlling common memory

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2304440A JPH04175941A (en) 1990-11-09 1990-11-09 Method and device for controlling common memory

Publications (1)

Publication Number Publication Date
JPH04175941A true JPH04175941A (en) 1992-06-23

Family

ID=17933037

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2304440A Pending JPH04175941A (en) 1990-11-09 1990-11-09 Method and device for controlling common memory

Country Status (1)

Country Link
JP (1) JPH04175941A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6785833B2 (en) 1993-10-15 2004-08-31 Renesas Technology Corp. Data processing system and image processing system
US7757062B2 (en) 2006-03-13 2010-07-13 Panasonic Corporation Semiconductor integrated circuit apparatus

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6785833B2 (en) 1993-10-15 2004-08-31 Renesas Technology Corp. Data processing system and image processing system
US6789210B2 (en) 1993-10-15 2004-09-07 Renesas Technology Corp. Data processing system having memory including mode register
US7254737B2 (en) 1993-10-15 2007-08-07 Renesas Technology Corp. Data processing system and image processing system
US7711976B2 (en) 1993-10-15 2010-05-04 Renesas Technology Corp. Data processing system and image processing system
US8332683B2 (en) 1993-10-15 2012-12-11 Renesas Electronics Corporation Data processing system and image processing system
US7757062B2 (en) 2006-03-13 2010-07-13 Panasonic Corporation Semiconductor integrated circuit apparatus
US8086814B2 (en) 2006-03-13 2011-12-27 Panasonic Corporation Semiconductor integrated circuit apparatus

Similar Documents

Publication Publication Date Title
US5819096A (en) PCI to ISA interrupt protocol converter and selection mechanism
US5455915A (en) Computer system with bridge circuitry having input/output multiplexers and third direct unidirectional path for data transfer between buses operating at different rates
US7406557B2 (en) Programmable logic device including programmable interface core and central processing unit
US6775727B2 (en) System and method for controlling bus arbitration during cache memory burst cycles
EP1851641B1 (en) Switch matrix system with plural bus arbitrations per cycle via higher-frequency arbiter
CN101923523A (en) Memory system and memory access method
JP2007133527A (en) Clock signal generation circuit, semiconductor integrated circuit, and frequency-division rate control method
US5922076A (en) Clocking scheme for digital signal processor system
JPH04350754A (en) Workstation including interface for data channel or similar data processing system
JP2005293596A (en) Arbitration of data request
US5428746A (en) Integrated microprocessor unit generating separate memory and input-output device control signals
JPH052552A (en) Workstation having burst mode capacity
US20110197008A1 (en) Card host lsi and set device including the lsi
JPH04175941A (en) Method and device for controlling common memory
US5687329A (en) Information handling system including a data bus management unit, an address management unit for isolating processor buses from I/O and memory
EP0473453B1 (en) Work station having a selectable CPU
EP1285340B1 (en) Shared bus interface for digital signal processor
US6862640B2 (en) Arbitration in local system for access to memory in a distant subsystem
JPH04350753A (en) Workstation having direct memory access controller and interface device for data channel
JPH052555A (en) Internal bus for workstation interface device
JPH09153009A (en) Arbitration method for hierarchical constitution bus
US6928575B2 (en) Apparatus for controlling and supplying in phase clock signals to components of an integrated circuit with a multiprocessor architecture
US5933648A (en) Configurable arbitration device for controlling the access of components to an arbiter or the like based on a control input
JPH0973429A (en) Computer system and inter-bus control circuit
US7707450B1 (en) Time shared memory access