JPS60247767A - Simultaneous input and output system for reading-out and writing-in operation of shared memory - Google Patents
Simultaneous input and output system for reading-out and writing-in operation of shared memoryInfo
- Publication number
- JPS60247767A JPS60247767A JP10516484A JP10516484A JPS60247767A JP S60247767 A JPS60247767 A JP S60247767A JP 10516484 A JP10516484 A JP 10516484A JP 10516484 A JP10516484 A JP 10516484A JP S60247767 A JPS60247767 A JP S60247767A
- Authority
- JP
- Japan
- Prior art keywords
- data
- memory
- write
- shared memory
- line
- 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
Links
- 238000000034 method Methods 0.000 claims description 6
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000003292 diminished effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/14—Handling requests for interconnection or transfer
- G06F13/16—Handling requests for interconnection or transfer for access to memory bus
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multi Processors (AREA)
Abstract
Description
【発明の詳細な説明】
〔技術分野〕
本発明は共有メモリの読出し訃込み動作の同時入出力方
式に関する。DETAILED DESCRIPTION OF THE INVENTION [Technical Field] The present invention relates to a simultaneous input/output method for read and write operations of a shared memory.
従来、マルチプロセッサ方式において共有メモVt使用
する場合、1つのプロセッサが共有メモリに対して絖出
し動作を行っていた場合、他のプロセッサが共有メモリ
に対してデータを書込もうとしても、共有メモリは使用
状態にあるためデータt4込もうとしたプロセッサは待
状態となシ、処理を続行することができない。マルチプ
ロセッサ方式により処理の分散化を計っていても、共有
データを効率よく受渡しができないと、各プロセッサに
おける処理が渋滞し分散処理の効果が薄らいでくる。Conventionally, when using a shared memory Vt in a multiprocessor system, if one processor performs a start-up operation on the shared memory, even if another processor attempts to write data to the shared memory, the shared memory Since t4 is in use, the processor that attempted to input data t4 is in a waiting state and cannot continue processing. Even if processing is distributed using a multiprocessor system, if shared data cannot be transferred efficiently, processing in each processor will become congested and the effectiveness of distributed processing will be diminished.
本発明の目的は、従来マルチプロセッサ方式における共
有メモリ使用時における待ちとなるプロセッサを減らし
、総合的な分散処理効果を高めるために、共有メモリ側
に薔込みアドレスレジスタとデータレジスタを設け、共
有メモリ読み出し動作中でも共有メモリに対して他のプ
ロセッサがデータ書込みができるようにした共有メモリ
の絖出し書込み動作の同時入出力方式を提供するにある
。An object of the present invention is to reduce the number of waiting processors when using shared memory in a conventional multiprocessor system, and to improve the overall distributed processing effect. An object of the present invention is to provide a simultaneous input/output method for a start-up write operation of a shared memory, which allows another processor to write data to the shared memory even during a read operation.
本発明によると、共有メそりに誉込み用アドレスレジス
タと書込み用データレジスタを設け、1つのプロセッサ
ーがメモリ読出しを行っている時に他のプロセッサがメ
モリ番込みを行う場合畳込みアドレスとデータは一時書
込み用アドレスレジスタとデータレジスタに舌込み、メ
モリ読み出しが終了した時に書込みデータをメモリに書
込むことを特徴とする共有メモリの読出しl書込み動作
の同時入出力方式が得られる。According to the present invention, an address register for writing and a data register for writing are provided in a shared memory, and when one processor is reading memory and another processor is writing memory, the convolution address and data are temporarily stored. A simultaneous input/output method for read and write operations of a shared memory is obtained, which is characterized in that the write address register and the data register are entered and the write data is written to the memory when the memory read is completed.
次に本発明の実施例を図面ヲ診照して説明する。 Next, embodiments of the present invention will be described with reference to the drawings.
第1図は、本発明の一実施例のブロック図T1?あシ、
図において共有メモリ4内におるメモリデータのアクセ
スを行う場合、まずデータを読み出したいプロセラ+j
lがろると、メモリの読み出しアドレス線15へ読み出
しアドレスを出力し、読み出しデータ線14よシメモリ
内のデータを読みだす。FIG. 1 is a block diagram T1 of an embodiment of the present invention. Ash,
In the figure, when accessing memory data in the shared memory 4, first select the processor +j from which you want to read the data.
When l is cleared, a read address is output to the read address line 15 of the memory, and the data in the memory is read out through the read data line 14.
この時データを書き込みたいプロセンサ2は共有メモリ
に対して書込みアドレスヲ壽込みアドレス線lOへ出力
し、同時に壽込みデータケ書込みデータ線11に出力す
る。この時プロセッサ1によるメモリ読み出し1g号1
3が発生しているので、プロセッサ2よジのメモリ書込
みアドレスはアドレスレジスタ1(フリップフロラフ)
に、メモリ書込みデータはデータレジスタ2(フリップ
フロラフ)にそれぞれ書込まれる。プロセッサ2は、こ
のレジスタに書込んだ時点で共有メモリとの処理は終了
し、更に他の処理へ移る。At this time, the processor 2 that wants to write data outputs a write address to the shared memory to the write address line IO, and at the same time outputs the write data to the write data line 11. At this time, memory read 1g No. 1 by processor 1
3 has occurred, the memory write address of processor 2 is address register 1 (flip flow rough).
Then, the memory write data is written to the data register 2 (flip flow rough), respectively. When the processor 2 writes to this register, processing with the shared memory ends, and the processor 2 moves on to other processing.
各レジスタに貯えられたアドレスとデータはメモリ読出
し信号が終了した時点に書込みアドレスをメモリアドレ
ス線5へ、書込みデータをメモリデータ線6へ出力し、
更にメモリ書込信号12をメモリ書込み線8へ出力し、
1時的に貯えらnたメモリ畳込みデータを共有メモリへ
薔込む。The address and data stored in each register are outputted to the memory address line 5 as the write address and the memory data line 6 when the memory read signal ends.
Furthermore, it outputs the memory write signal 12 to the memory write line 8,
The temporarily stored memory convolution data is transferred to the shared memory.
このようにして共有メモリにデータ読出し、書込み動作
を行えば共有メモリ読出し処理を実行したプロセッサ1
と共有メモリ書込み処理全実施したプロセッサ2は、時
間的に1メモリサイクル内で同時に共有メモリに対して
データの入出力を行ったことになシ、プロセッサ1.プ
ロセッサ2は共有メモリのアクセスに対して待時間を持
つ必要なく、各プロセッサが持つ内部処理を続行するこ
とができる。When data is read and written to the shared memory in this way, the processor 1 that executed the shared memory read process
Processor 2, which has executed all of the shared memory write processing, has simultaneously input and output data to and from the shared memory within one memory cycle. The processor 2 can continue the internal processing of each processor without having to wait for access to the shared memory.
更に誓込みアドレス線10.4込みデータ線11を増や
すことにより、共有メモリに対する各プロセッサからの
重複書込み要求に対しても処理が可能である。Furthermore, by increasing the number of commit address lines 10, 4 and write data lines 11, it is possible to process duplicate write requests from each processor to the shared memory.
本発明は、以上説明したように、共有メモリを見かけ上
の同時入出力を可能とし、各プロセッサのデータ待状態
?減らし分散処理能力の向上に効果がある。As explained above, the present invention enables apparent simultaneous input/output of shared memory, and allows each processor to enter a data waiting state. It is effective in reducing the amount of data and improving distributed processing capacity.
第1図は本発明の一実施例のブロック図でろる。
l・・・・・・書込みアドレスレジスタ、2・・・・・
・書込みデータレジスタ、3・・・・・・メモリ制御信
号ラッチ回路、4・・・・・・メモリ、5・・・・・・
メモリアドレス線、6・・・・・・メモリデータ線、7
・・・・・・詠出しアドレス線、8・・・・・・メモリ
書込信号線、9・・・・・・メモリ読出信号線、10・
・・・・・書込みアドレス線、11・・・・・・書込み
データ線、12・・・・・・メモリ書込信号、13・・
・・・・メモリ読出し信号、14・・・・・・読出しデ
ータ線、15・・・・・・続出しアドレス線。FIG. 1 is a block diagram of one embodiment of the present invention. l...Write address register, 2...
・Write data register, 3...Memory control signal latch circuit, 4...Memory, 5...
Memory address line, 6...Memory data line, 7
. . . Emit address line, 8 . . . Memory write signal line, 9 . . . Memory read signal line, 10.
...Write address line, 11...Write data line, 12...Memory write signal, 13...
...Memory read signal, 14...Read data line, 15...Continuous address line.
Claims (1)
タレジスタを設け、1つのプロセッサーがメモリ挽出し
金行っている時に他のプロセッサがメモリ書込みを行う
場合薔込みアドレスとデータは一時書込み用アドレスレ
ジスタとデータレジスタに薔込み、メモリ読み出しが終
了した時に舊込みデータ會メモリに簀込むことを特徴と
する共有メモリの読出しl書込み動作の同時入出力方式
。A write address register and a write data register are provided in the shared memory, and when one processor writes memory while another processor writes to the memory, the write address and data are temporarily written to the write address register and data register. A simultaneous input/output method for read and write operations of a shared memory, which is characterized in that the data is stored in the memory when the memory read is completed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP10516484A JPS60247767A (en) | 1984-05-24 | 1984-05-24 | Simultaneous input and output system for reading-out and writing-in operation of shared memory |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP10516484A JPS60247767A (en) | 1984-05-24 | 1984-05-24 | Simultaneous input and output system for reading-out and writing-in operation of shared memory |
Publications (1)
Publication Number | Publication Date |
---|---|
JPS60247767A true JPS60247767A (en) | 1985-12-07 |
Family
ID=14400047
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP10516484A Pending JPS60247767A (en) | 1984-05-24 | 1984-05-24 | Simultaneous input and output system for reading-out and writing-in operation of shared memory |
Country Status (1)
Country | Link |
---|---|
JP (1) | JPS60247767A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS6384212A (en) * | 1986-09-29 | 1988-04-14 | Matsushita Electric Ind Co Ltd | Road-side communication equipment |
JPH01199261A (en) * | 1987-10-26 | 1989-08-10 | Matsushita Electric Works Ltd | Shared memory control system |
-
1984
- 1984-05-24 JP JP10516484A patent/JPS60247767A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS6384212A (en) * | 1986-09-29 | 1988-04-14 | Matsushita Electric Ind Co Ltd | Road-side communication equipment |
JPH01199261A (en) * | 1987-10-26 | 1989-08-10 | Matsushita Electric Works Ltd | Shared memory control system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JPS6113629B2 (en) | ||
JPS60247767A (en) | Simultaneous input and output system for reading-out and writing-in operation of shared memory | |
JPS58201166A (en) | Multiprocessor system | |
JPS61165170A (en) | Bus control method | |
JPH0546527A (en) | Dual port memory circuit | |
JPS61256478A (en) | parallel processing computer | |
JPS60195661A (en) | Data processing system | |
JPS63142455A (en) | Semiconductor storage device | |
JPH0522939B2 (en) | ||
JPS6391756A (en) | Partial write instruction processing system for storage device | |
JPS5897199A (en) | Memory controller | |
JPH0214741B2 (en) | ||
JPH03252856A (en) | Processing system for program | |
JPS5839364A (en) | Graphic information processing unit | |
JPS59114657A (en) | Memory interface circuit of microcomputer | |
JPS6113267B2 (en) | ||
JPH05204866A (en) | Data conversion system for multi-cpu system | |
JPS63201810A (en) | Time system of information processing system | |
JPH04104355A (en) | Multiprocessing method | |
JPS5836380B2 (en) | Direct memory access method in multiprocessor systems | |
JPH02257241A (en) | Memory access competition improving system | |
JPH04346155A (en) | Multiprocessor system | |
JPS6112306B2 (en) | ||
JPH0685154B2 (en) | Intermediate buffer control method | |
JPH03172960A (en) | Arithmetic unit |