JPH01194043A - Electronic computer - Google Patents

Electronic computer

Info

Publication number
JPH01194043A
JPH01194043A JP63018828A JP1882888A JPH01194043A JP H01194043 A JPH01194043 A JP H01194043A JP 63018828 A JP63018828 A JP 63018828A JP 1882888 A JP1882888 A JP 1882888A JP H01194043 A JPH01194043 A JP H01194043A
Authority
JP
Japan
Prior art keywords
instruction
processing
instruction queue
tlb
sent
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
JP63018828A
Other languages
Japanese (ja)
Inventor
Yoshiko Miyamoto
佳子 宮本
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.)
NEC Corp
Original Assignee
NEC 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 NEC Corp filed Critical NEC Corp
Priority to JP63018828A priority Critical patent/JPH01194043A/en
Publication of JPH01194043A publication Critical patent/JPH01194043A/en
Pending legal-status Critical Current

Links

Landscapes

  • Memory System Of A Hierarchy Structure (AREA)

Abstract

PURPOSE:To rapidly execute the mishit processing of TLB by using respectively different instruction queues for normal processing and interruption processing by a control means. CONSTITUTION:When a mishit is generated in the TLB 8 during the period of normal processing, an interruption signal (c) is sent to an instruction reading mechanism 3 and an instruction queue control mechanism 7. A mishit processing instruction in the TLB 8 is read out by the mechanism 3, an enqueue signal a2 is sent from the mechanism 7 to a 1st instruction queue 2 and an instruction on an instruction bus 4 is stored in the instruction queue 2. On the other hand, '1' to be a selecting signal (b) is sent from the mechanism 7 to a multiplexer 5 and the instruction of the instruction queue 2 is sent to a decoder 6. At that time, the state of the 2nd instruction queue 1 is held as it is. Consequently, the mishit processing of the TLB can be rapidly executed without flashing the instruction queue or rereading an once read instruction.

Description

【発明の詳細な説明】 〔産業上の利用分野〕 本発明は電子計算機の制御方式に関し、特にTLBのミ
スヒツトによる割込みの制御に関するものである。
DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a control method for an electronic computer, and particularly to control of interrupts caused by TLB mishits.

〔従来の技術〕[Conventional technology]

従来、ページング機構を持つ電子計算機では、TLBの
ミスヒツトが起きると、割込みが発生し、TLBのミス
ヒツトを起こした命令の位置を記憶し、TLBのミスヒ
ツト処理を実行する命令の読出しが始まり、命令キュー
に残っていた命令はフラッシュされ、TLBのミスヒツ
ト処理の命令が命令キューに収納されていた。TLBの
ミスヒツト処理の終了後は元の処理の命令を再び読み出
し命令キューに収納していた。
Conventionally, in electronic computers with a paging mechanism, when a TLB mishit occurs, an interrupt is generated, the position of the instruction that caused the TLB mishit is memorized, reading of the instruction to execute the TLB mishit processing is started, and the instruction queue is The remaining instructions were flushed, and the TLB miss processing instructions were stored in the instruction queue. After the TLB miss processing is completed, the instructions for the original processing are stored in the read instruction queue again.

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

上述した従来の電子計算機の制御方式は、TLBのミス
ヒツトによる割込みが発生すると、命令キューをフラッ
シュしなくてはならず、通常の処理に戻るときに、フラ
ッシュした命令を再度読み出さなくてはならないという
欠点があった。
In the conventional computer control method described above, when an interrupt occurs due to a TLB miss, the instruction queue must be flushed, and when returning to normal processing, the flushed instructions must be read again. There were drawbacks.

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

このような欠点を除去するために本発明は、ページング
方式でメモリを管理する電子計算機において、TLBの
ミスヒツトの処理をするために読み出されたミスヒツト
処理命令を解読・実行のために収納しておく第1の命令
キューと、ミスヒツト処理命令以外の通常の処理をする
ための命令を収納しておく第2の命令キューと、第1と
第2の命令キューについての命令の収納およびデコーダ
への命令の転送を制御する制御手段とを備え、制御手段
により通常の処理と割込み処理とで異なる命令キューを
使用するようにしたものである。
In order to eliminate such drawbacks, the present invention provides a method for storing mishit processing instructions read out for processing TLB misses for decoding and execution in an electronic computer that manages memory using a paging method. a first instruction queue that stores instructions for normal processing other than miss-processing instructions; The control means controls the transfer of instructions, and the control means uses different instruction queues for normal processing and interrupt processing.

〔作用〕[Effect]

本発明による電子計算機においては、TLBのミスヒツ
ト処理において、命令キューをフラッシュしたり、−度
読み出した命令を再度読み出したりすることがなくなる
In the electronic computer according to the present invention, in TLB miss processing, there is no need to flush the instruction queue or read out an instruction that has been read out again.

〔実施例〕〔Example〕

図は、本発明に係わる電子計算機の一実施例を示す系統
図である。図において、1は第2の命令キュー、2は第
1の命令キュー、3は命令読出し機構、4は命令バス、
5はマルチプレクサ、6はデコーダ、7は制御手段とし
ての命令キュー制御機構、8はTLBである。
The figure is a system diagram showing an embodiment of an electronic computer according to the present invention. In the figure, 1 is a second instruction queue, 2 is a first instruction queue, 3 is an instruction reading mechanism, 4 is an instruction bus,
5 is a multiplexer, 6 is a decoder, 7 is an instruction queue control mechanism as a control means, and 8 is a TLB.

通常の処理中は、命令読出し機構3から命令バス4を介
して命令が送られてくると、命令キュー制御機構7から
第2の命令キュー1にエンキュー信号alが送られ、命
令は命令キュー1に収納される。マルチプレクサ5へは
命令キュー制御機構7から選択信号すとして「0」が送
られ、命令キュー1の命令がデコーダ6に送られる。
During normal processing, when an instruction is sent from the instruction reading mechanism 3 via the instruction bus 4, an enqueue signal al is sent from the instruction queue control mechanism 7 to the second instruction queue 1, and the instruction is sent to the instruction queue 1. is stored in. A selection signal of "0" is sent from the instruction queue control mechanism 7 to the multiplexer 5, and the instructions in the instruction queue 1 are sent to the decoder 6.

通常の処理中にTLB8でミスヒツトが起きると、割込
み信号Cが命令読出し機構3と命令キュー制御機構7に
送られる。命令読出し機構3ではTLB8のミスヒツト
処理の命令が読み出され、命令キュー制御機構7から第
1の命令キュー2にエンキュー信号a2が送られ、命令
バス4上の命令が命令キュー2に収納される。マルチプ
レクサ5へは命令キュー制御機構7から選択信号すとし
て「1」が送られ、命令キュー2の命令がデコーダ6に
送られる。このとき、第2の命令キュー1の状態はその
まま保持される。
When a miss occurs in the TLB 8 during normal processing, an interrupt signal C is sent to the instruction reading mechanism 3 and the instruction queue control mechanism 7. The instruction reading mechanism 3 reads out the instruction for miss processing from the TLB 8, the instruction queue control mechanism 7 sends an enqueue signal a2 to the first instruction queue 2, and the instructions on the instruction bus 4 are stored in the instruction queue 2. . A selection signal of "1" is sent from the instruction queue control mechanism 7 to the multiplexer 5, and the instructions in the instruction queue 2 are sent to the decoder 6. At this time, the state of the second instruction queue 1 is maintained as it is.

TLB8のミスヒツト処理の最後の命令がデコーダ6に
送られると、デコーダ6から割込み終了信号dが命令読
出し機構3と命令キュー制御機構7に送られる。選択信
号すは「0」になり、再び命令キュー1の命令がデコー
ダ6に送られるようになる。命令続出し機構3では、割
込み以前に読み出して命令キュー1に送られた命令の続
きである通常の処理の命令が読み出され、命令キュー1
に収納される。
When the last instruction of the TLB 8 miss processing is sent to the decoder 6, an interrupt end signal d is sent from the decoder 6 to the instruction reading mechanism 3 and the instruction queue control mechanism 7. The selection signal becomes "0" and the instructions in the instruction queue 1 are sent to the decoder 6 again. In the instruction succession mechanism 3, instructions for normal processing, which are the continuation of the instructions read and sent to the instruction queue 1 before the interrupt, are read and sent to the instruction queue 1.
is stored in.

〔発明の効果〕〔Effect of the invention〕

以上説明したように本発明による電子計算機は、命令キ
ューを2つ持ち、1つを通常の処理に、もう1つをTL
Bのミスヒツト処理に用いるようにしたことにより、T
LBのミスヒット処理を、命令キューをフランシュした
り、−度読み出した命令を再度読み出したりすることな
く、高速に行なうことができる効果がある。
As explained above, the electronic computer according to the present invention has two instruction queues, one for normal processing and the other for TL.
By using it for mishit processing of B, T
This has the advantage that LB miss-hit processing can be performed at high speed without franchising the instruction queue or re-reading instructions that have been read out twice.

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

図は本発明に係わる電子計算機の一実施例を示す系統図
である。 1・・・第2の命令キュー、2・・・第1の命令キュー
、3・・・命令読出し機構、4・・・命令バス、5・・
・マルチプレクサ、6・・・デコーダ、7・・・命令キ
ュー制御機構、8・・・TLB。 特許出願人    日本電気株式会社
The figure is a system diagram showing one embodiment of the electronic computer according to the present invention. DESCRIPTION OF SYMBOLS 1... Second instruction queue, 2... First instruction queue, 3... Instruction reading mechanism, 4... Instruction bus, 5...
- Multiplexer, 6... Decoder, 7... Instruction queue control mechanism, 8... TLB. Patent applicant: NEC Corporation

Claims (1)

【特許請求の範囲】[Claims]  ページング方式でメモリを管理する電子計算機におい
て、TLBのミスヒットの処理をするために読み出され
たミスヒット処理命令を解読・実行のために収納してお
く第1の命令キューと、前記ミスヒット処理命令以外の
通常の処理をするための命令を収納しておく第2の命令
キューと、第1と第2の命令キューについての命令の収
納およびデコーダへの命令の転送を制御する制御手段と
を備え、前記制御手段は通常の処理と割込み処理とで異
なる命令キューを使用するように制御することを特徴と
する電子計算機。
In an electronic computer that manages memory using a paging method, there is provided a first instruction queue in which a mishit processing instruction read out for processing a TLB mishit is stored for decoding and execution; a second instruction queue for storing instructions for normal processing other than processing instructions; and a control means for controlling storage of instructions in the first and second instruction queues and transfer of instructions to a decoder. An electronic computer, characterized in that the control means controls so that different instruction queues are used for normal processing and interrupt processing.
JP63018828A 1988-01-29 1988-01-29 Electronic computer Pending JPH01194043A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP63018828A JPH01194043A (en) 1988-01-29 1988-01-29 Electronic computer

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP63018828A JPH01194043A (en) 1988-01-29 1988-01-29 Electronic computer

Publications (1)

Publication Number Publication Date
JPH01194043A true JPH01194043A (en) 1989-08-04

Family

ID=11982423

Family Applications (1)

Application Number Title Priority Date Filing Date
JP63018828A Pending JPH01194043A (en) 1988-01-29 1988-01-29 Electronic computer

Country Status (1)

Country Link
JP (1) JPH01194043A (en)

Similar Documents

Publication Publication Date Title
US6553487B1 (en) Device and method for performing high-speed low overhead context switch
JP2545482B2 (en) Interface device transfer parameter setting method
JPH0836877A (en) Queue device
JPH0827707B2 (en) Control device and control method for FIFO buffer and device for controlling data transfer
JPS6252655A (en) Common interrupt system
WO2007063858A1 (en) Buffer control device and buffer memory
JPH06103494B2 (en) Vector processor control system
WO2001038970A2 (en) Buffer memories, methods and systems for buffering having seperate buffer memories for each of a plurality of tasks
JPH01194043A (en) Electronic computer
US4816992A (en) Method of operating a data processing system in response to an interrupt
JPS61217834A (en) Data processor
JP2701739B2 (en) Processor switching method
JPH06103065A (en) Instruction prefetch device
JPS58205258A (en) Data processor
JP3900660B2 (en) Sequence controller
JPS62251943A (en) Memory address control system for channel device
JPS60178570A (en) Data receiver
JP3084953B2 (en) Information processing device
JPH06124205A (en) Instruction prefetching device
JPH1055303A (en) Memory system
JPS61156454A (en) Data transfer control device
JPS63155254A (en) Information processor
JPH02173828A (en) Interruption process system
JPS63178335A (en) Microprocessor
JPS6133545A (en) Data processor