JPH05127901A - Processor - Google Patents

Processor

Info

Publication number
JPH05127901A
JPH05127901A JP29276791A JP29276791A JPH05127901A JP H05127901 A JPH05127901 A JP H05127901A JP 29276791 A JP29276791 A JP 29276791A JP 29276791 A JP29276791 A JP 29276791A JP H05127901 A JPH05127901 A JP H05127901A
Authority
JP
Japan
Prior art keywords
instruction
division
multiplication
instructions
storage device
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
JP29276791A
Other languages
Japanese (ja)
Inventor
Yuji Nakai
祐二 中居
Hiroshi Nakano
拓 中野
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.)
Panasonic Holdings Corp
Original Assignee
Matsushita Electric Industrial 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 Matsushita Electric Industrial Co Ltd filed Critical Matsushita Electric Industrial Co Ltd
Priority to JP29276791A priority Critical patent/JPH05127901A/en
Publication of JPH05127901A publication Critical patent/JPH05127901A/en
Pending legal-status Critical Current

Links

Abstract

PURPOSE:To provide a high speed processor whose performance is not deteriorated by means of an instruction whose processing speed is slow. CONSTITUTION:An instruction reader 2 sequentially reads instructions stored in an instruction storage device 1 with subsequent and plural instructions at every machine cycle, and temporarily stores them in a temporary instruction storage device 3. An execution controller 5 checks a subsequent multiplication instruction stored in the temporary instruction storage device 3 and allocates division instructions decoded by an instruction decoding device 4 to a high speed multiplication/division device 7 or a low speed division device 8.

Description

【発明の詳細な説明】Detailed Description of the Invention

【0001】[0001]

【産業上の利用分野】本発明は高速に処理を行なうプロ
セッサに関するものである。
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a processor for high speed processing.

【0002】[0002]

【従来の技術】近年、加減算装置及び乗算の反復計算に
より除算を行なう乗除算装置を内蔵するプロセッサが主
流となってきている。
2. Description of the Related Art In recent years, processors having a built-in addition / subtraction device and a multiplication / division device for performing division by iterative calculation of multiplication have become mainstream.

【0003】以下図面を参照しながら、従来のプロセッ
サの一例について説明する。図3は従来のプロセッサの
構成図を示すものである。図3において、11は命令を
格納する命令格納装置、12は命令を読み出す命令読み
出し装置、13は読み出された命令を解読する命令解読
装置、14は加算または減算を行なう加減算装置、15
は乗算または除算を行なう乗除算装置、16はデータを
格納するデータ格納装置である。命令読み出し装置1
2、命令解読装置13、加減算装置14、乗除算装置1
5はパイプライン動作し、命令読み出し装置12、命令
解読装置13、加減算装置14は1マシンサイクルで処
理を行ない、乗除算装置15は乗算を1マシンサイク
ル、除算を乗算の反復計算により除算を4マシンサイク
ルで行なう。
An example of a conventional processor will be described below with reference to the drawings. FIG. 3 is a block diagram of a conventional processor. In FIG. 3, 11 is an instruction storage device for storing instructions, 12 is an instruction reading device for reading out instructions, 13 is an instruction decoding device for decoding the read instructions, 14 is an addition / subtraction device for performing addition or subtraction, and 15
Is a multiplication / division device for performing multiplication or division, and 16 is a data storage device for storing data. Instruction reading device 1
2, instruction decoding device 13, addition / subtraction device 14, multiplication / division device 1
5 operates in a pipeline, the instruction reading device 12, the instruction decoding device 13, and the addition / subtraction device 14 perform processing in one machine cycle, and the multiplication / division device 15 performs multiplication in one machine cycle, and division is repeated by iterative calculation of multiplication to obtain four divisions. Do it in a machine cycle.

【0004】以上のように構成されたプロセッサについ
て、以下その動作について説明する。
The operation of the processor configured as described above will be described below.

【0005】命令読み出し装置12は命令格納装置11
に格納された命令を1マシンサイクルごとに順次読み出
す。命令解読装置13は1マシンサイクルで命令読み出
し装置12により読み出された命令を解読し、加算命令
または減算命令を加減算装置14に割り付け、乗算命令
または除算命令を乗除算装置15に割り付ける。加減算
装置14は命令解読装置13により割り付けられた命令
に対して加算または減算を1マシンサイクルで行なう。
同様に、乗除算装置15は乗算または除算を行なうが、
乗算を1マシンサイクル、除算を4マシンサイクルで行
なう。加減算装置14、乗除算装置15で行なう演算は
データ格納装置16に格納されたデータに対して行なわ
れ、演算結果はデータ格納装置16に格納される。除
算、乗算、除算、乗算と命令が続く場合の従来のプロセ
ッサの命令解読から実行までのパイプライン状態図を図
4に示す。ただし、実行する演算で用いるデータは演算
開始時にデータ格納装置16に格納されているものとす
る。
The instruction reading device 12 is an instruction storage device 11.
The instructions stored in are sequentially read every machine cycle. The instruction decoding device 13 decodes the instruction read by the instruction reading device 12 in one machine cycle, allocates an addition instruction or a subtraction instruction to the addition / subtraction device 14, and allocates a multiplication instruction or a division instruction to the multiplication / division device 15. The adder / subtractor 14 adds or subtracts the instructions assigned by the instruction decoder 13 in one machine cycle.
Similarly, the multiplication / division device 15 performs multiplication or division,
Multiplication is performed in one machine cycle and division is performed in four machine cycles. The calculation performed by the adder / subtractor 14 and the multiplication / division device 15 is performed on the data stored in the data storage device 16, and the calculation result is stored in the data storage device 16. FIG. 4 shows a pipeline state diagram from instruction decoding to execution of a conventional processor in the case where division, multiplication, division, multiplication and instructions follow. However, it is assumed that the data used in the calculation to be executed is stored in the data storage device 16 at the start of the calculation.

【0006】[0006]

【発明が解決しようとする課題】しかしながら上記のよ
うな構成では、除算命令に続き乗算命令がある場合、乗
除算装置は除算実行の4マシンサイクルの間、他の処理
を行なうことができないため、後続する乗算を除算終了
後に行わなければならず、プロセッサの性能を落とすと
いう問題点を有していた。
However, in the above configuration, when there is a multiplication instruction following a division instruction, the multiplication / division device cannot perform other processing during the four machine cycles of the division execution. The subsequent multiplication has to be performed after the division is completed, which has a problem of degrading the performance of the processor.

【0007】本発明は上記問題点に鑑み、処理速度の遅
い命令により性能を落とさない高速なプロセッサを提供
するものである。
In view of the above problems, the present invention provides a high-speed processor which does not deteriorate the performance due to an instruction having a slow processing speed.

【0008】[0008]

【課題を解決するための手段】上記問題点を解決するた
めに本発明のプロセッサは、複数の同一種命令に対して
処理を行なう実行速度の異なる複数の実行装置と、複数
の命令を読み出す命令読み出し装置と、この命令読み出
し装置により読み出された複数の命令を一時的に格納す
る命令一時格納装置と、この命令一時格納装置に格納さ
れた命令を解読する命令解読装置と、前記命令一時格納
装置に格納された複数の命令を用いて、前記命令解読装
置により解読された命令を複数の実行装置のうちいずれ
かに割り付ける実行制御装置とを備えたものである。
In order to solve the above problems, a processor according to the present invention comprises a plurality of execution units having different execution speeds for processing a plurality of same-type instructions and an instruction for reading a plurality of instructions. A reading device, an instruction temporary storage device for temporarily storing a plurality of instructions read by the instruction reading device, an instruction decoding device for decoding an instruction stored in the instruction temporary storage device, and the instruction temporary storage An execution control device for allocating an instruction decoded by the instruction decoding device to any one of a plurality of execution devices by using a plurality of instructions stored in the device.

【0009】[0009]

【作用】本発明は上記した構成によって、後続する複数
の命令を用いて処理速度の遅い命令を高速な実行装置と
低速な実行装置のいずれかに割り付けており、処理速度
の遅い命令によるプロセッサの性能低下を最小限にする
ことができる。
With the above-described structure, the present invention allocates an instruction with a slow processing speed to either a high-speed execution device or a low-speed execution device by using a plurality of subsequent instructions, and Performance degradation can be minimized.

【0010】[0010]

【実施例】以下本発明の一実施例のプロセッサについ
て、図面を参照しながら説明する。
DESCRIPTION OF THE PREFERRED EMBODIMENTS A processor according to an embodiment of the present invention will be described below with reference to the drawings.

【0011】図1は本発明の実施例におけるプロセッサ
の構成図を示すものである。図1において、1は命令を
格納する命令格納装置、2は命令を読み出す命令読み出
し装置、3は読み出された複数の命令を一時的に格納す
る命令一時格納装置、4は命令を解読する命令解読装
置、5は解読された命令を実行装置に割り付ける実行制
御装置、6は加算または減算を行なう加減算装置、7は
乗算または除算を行なう高速乗除算装置、8は除算を行
なう小規模な低速除算装置、9はデータを格納するデー
タ格納装置である。命令読み出し装置2、命令解読装置
4、実行制御装置5、加減算装置6、高速乗除算装置
7、低速除算装置8はパイプライン動作し、命令読み出
し装置2、加減算装置6は1マシンサイクルで処理を行
ない、命令解読装置4と実行制御装置5は並列に1マシ
ンサイクルで処理を行ない、高速乗除算装置7は乗算を
1マシンサイクル、除算を乗算の反復計算により除算を
4マシンサイクルで行ない、低速除算装置8は除算を8
マシンサイクルで行なう。
FIG. 1 is a block diagram of a processor according to an embodiment of the present invention. In FIG. 1, 1 is an instruction storage device for storing instructions, 2 is an instruction reading device for reading out instructions, 3 is an instruction temporary storage device for temporarily storing a plurality of read instructions, and 4 is an instruction for decoding instructions. Decoding device, 5 is an execution control device for allocating the decoded instruction to the execution device, 6 is an addition / subtraction device for performing addition or subtraction, 7 is a high speed multiplication / division device for performing multiplication or division, and 8 is a small speed division for performing division. A device, 9 is a data storage device for storing data. The instruction reading device 2, the instruction decoding device 4, the execution control device 5, the addition / subtraction device 6, the high speed multiplication / division device 7, and the low speed division device 8 operate in pipeline, and the instruction reading device 2 and the addition / subtraction device 6 perform processing in one machine cycle. The instruction decoding device 4 and the execution control device 5 perform the processing in parallel in one machine cycle, and the high speed multiplication / division device 7 performs the multiplication in one machine cycle, and the division is performed by the iterative calculation of multiplication in four machine cycles. The division device 8 divides by 8
Do it in a machine cycle.

【0012】以上のように構成されたプロセッサについ
て、以下その動作について説明する。
The operation of the processor configured as described above will be described below.

【0013】命令読み出し装置2は、命令格納装置1に
格納された命令を後続する複数の命令と共に1マシンサ
イクルごとに順次読み出し命令一時格納装置3に一時的
に格納する。命令一時格納装置3には未実行の連続する
複数の命令が格納され、実行された命令は除去される。
命令解読装置4は命令一時格納装置3に格納された命令
を解読する。実行制御装置5は加算命令または減算命令
を加減算装置6に割り付け、乗算命令を高速乗除算装置
7に割り付け、除算命令を高速乗除算装置7または低速
除算装置8に割り付ける。命令解読装置4と実行制御装
置5は並列に1マシンサイクルで処理を行なう。加減算
装置6は実行制御装置5により割り付けられた命令に対
して加算または減算を1マシンサイクルで行なう。同様
に、高速乗除算装置7は乗算または除算を行なうが、乗
算を1マシンサイクル、除算を4マシンサイクルで行な
う。低速除算装置8は除算を8マシンサイクルで行な
う。
The instruction reading device 2 temporarily stores the instructions stored in the instruction storage device 1 together with a plurality of subsequent instructions in the sequential read instruction temporary storage device 3 every machine cycle. The instruction temporary storage device 3 stores a plurality of unexecuted consecutive instructions, and the executed instructions are removed.
The instruction decoding device 4 decodes the instruction stored in the instruction temporary storage device 3. The execution control device 5 assigns an addition instruction or a subtraction instruction to the addition / subtraction device 6, a multiplication instruction to the high speed multiplication / division device 7, and a division instruction to the high speed multiplication / division device 7 or the low speed division device 8. The instruction decoder 4 and the execution controller 5 perform processing in parallel in one machine cycle. The adder / subtractor 6 adds or subtracts the instructions assigned by the execution controller 5 in one machine cycle. Similarly, the high speed multiplication / division device 7 performs multiplication or division, but the multiplication is performed in one machine cycle and the division is performed in four machine cycles. The low speed division device 8 performs division in eight machine cycles.

【0014】実行制御装置5は除算命令を高速乗除算装
置7または低速除算装置8に割り付けるが、命令一時格
納装置3に格納された後続する複数の命令を用いていず
れに割り付けるかを決める。後続する4マシンサイクル
の間に乗算命令がない場合は、除算命令を高速乗除算装
置7に割り付ける。後続する4マシンサイクルの間に乗
算命令がある場合は、乗算命令が除算実行により処理を
待たされるのを避けるため除算命令を低速除算装置8に
割り付ける。ただし、後続する乗算命令が解読された除
算命令の結果を用いる場合または低速除算装置8が既に
除算実行中である場合は、除算命令を高速乗除算装置7
に割り付ける。加減算装置6、高速乗除算装置7、低速
除算装置8で行なう演算はデータ格納装置9に格納され
たデータに対して行なわれ、演算結果はデータ格納装置
9に格納される。
The execution control device 5 allocates the division instruction to the high speed multiplication / division device 7 or the low speed division device 8, and determines which one to allocate by using a plurality of subsequent instructions stored in the instruction temporary storage device 3. When there is no multiplication instruction during the following four machine cycles, the division instruction is assigned to the high speed multiplication / division device 7. When there is a multiplication instruction during the following four machine cycles, the division instruction is assigned to the low speed division device 8 in order to avoid the multiplication instruction from being kept waiting for the processing due to the division execution. However, if the subsequent multiplication instruction uses the decoded result of the division instruction, or if the slow division device 8 is already performing division, the division instruction is added to the fast multiplication / division device 7.
Assign to. The operations performed by the adder / subtractor 6, the high-speed multiplier / divider 7, and the low-speed divider 8 are performed on the data stored in the data storage device 9, and the operation result is stored in the data storage device 9.

【0015】除算、乗算、除算、乗算と命令が続く場合
の従来のプロセッサの命令解読から実行までのパイプラ
イン状態図を図2に示す。ただし、実行する演算に用い
るデータは演算開始時にデータ格納装置9に格納されて
いるものとする。
FIG. 2 shows a pipeline state diagram from instruction decoding to execution in a conventional processor in the case where division, multiplication, division, multiplication and instructions follow. However, it is assumed that the data used for the calculation to be executed is stored in the data storage device 9 at the start of the calculation.

【0016】以上のように本実施例のプロセッサによれ
ば、実行制御装置5が後続する乗算命令を用いて除算命
令を高速乗除算装置7と低速除算装置8のいずれかに割
り付けているので、後続する乗算命令が除算実行により
処理を待たされるのを避けることができ、プロセッサの
性能低下を最小限にすることができる。
As described above, according to the processor of this embodiment, the execution control device 5 allocates the division instruction to either the high speed multiplication / division device 7 or the low speed division device 8 by using the subsequent multiplication instruction. Subsequent multiplication instructions can be prevented from waiting for processing due to division execution, and the performance degradation of the processor can be minimized.

【0017】なお、本実施例のプロセッサでは高速乗除
算装置7の他に低速除算装置8を必要とするが、低速乗
除算装置8はビット幅を縮小した小規模な回路で構成す
ることができるため、プロセッサ全体の回路規模は大き
くならない。
Although the processor of this embodiment requires the low-speed multiplication / division device 8 in addition to the high-speed multiplication / division device 7, the low-speed multiplication / division device 8 can be constituted by a small-scale circuit having a reduced bit width. Therefore, the circuit scale of the entire processor does not increase.

【0018】[0018]

【発明の効果】以上のように本発明は、後続する複数の
命令を用いて処理速度の遅い命令を高速な実行装置と低
速な実行装置のいずれかに割り付ける実行制御装置を設
けることにより、処理速度の遅い命令によるプロセッサ
の性能低下を最小限にすることができる。
As described above, according to the present invention, by providing the execution control device for allocating the instruction having the slow processing speed to the high-speed execution device or the low-speed execution device by using the plurality of subsequent instructions, The performance degradation of the processor due to slow instructions can be minimized.

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

【図1】本発明の第1の実施例におけるプロセッサの構
成図
FIG. 1 is a configuration diagram of a processor according to a first embodiment of the present invention.

【図2】同実施例におけるパイプライン状態図FIG. 2 is a pipeline state diagram in the same embodiment.

【図3】従来のプロセッサの構成図FIG. 3 is a block diagram of a conventional processor

【図4】従来のプロセッサのパイプライン状態図FIG. 4 is a pipeline state diagram of a conventional processor.

【符号の説明】[Explanation of symbols]

1 命令格納装置 2 命令読み出し装置 3 命令一時格納装置 4 命令解読装置 5 実行制御装置 6 加減算装置 7 高速乗除算装置 8 低速除算装置 9 データ格納装置 1 instruction storage device 2 instruction reading device 3 instruction temporary storage device 4 instruction decoding device 5 execution control device 6 addition / subtraction device 7 high speed multiplication / division device 8 low speed division device 9 data storage device

Claims (2)

【特許請求の範囲】[Claims] 【請求項1】複数の同一種命令に対して処理を行なう実
行速度の異なる複数の実行装置と、複数の命令を読み出
す命令読み出し装置と、前記命令読み出し装置により読
み出された複数の命令を一時的に格納する命令一時格納
装置と、前記命令一時格納装置に格納された命令を解読
する命令解読装置と、前記命令一時格納装置に格納され
た複数の命令を用いて、前記命令解読装置により解読さ
れた命令を前記複数の実行装置のうちいずれかに割り付
ける実行制御装置とを備えたことを特徴とするプロセッ
サ。
1. A plurality of execution devices having different execution speeds for processing a plurality of same-type instructions, an instruction reading device for reading a plurality of instructions, and a plurality of instructions read by the instruction reading device temporarily. For temporarily storing an instruction, an instruction decoder for decoding an instruction stored in the instruction temporary storage, and a plurality of instructions stored in the instruction temporary storage for decoding by the instruction decoder An execution control unit that assigns the generated instructions to any of the plurality of execution units.
【請求項2】請求項1記載の複数の実行装置として乗算
または除算を行なう乗除算装置と除算を行なう除算装置
を備えたことを特徴とするプロセッサ。
2. A processor comprising a plurality of execution units as claimed in claim 1 and a multiplication / division unit for performing multiplication or division and a division unit for division.
JP29276791A 1991-11-08 1991-11-08 Processor Pending JPH05127901A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP29276791A JPH05127901A (en) 1991-11-08 1991-11-08 Processor

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP29276791A JPH05127901A (en) 1991-11-08 1991-11-08 Processor

Publications (1)

Publication Number Publication Date
JPH05127901A true JPH05127901A (en) 1993-05-25

Family

ID=17786078

Family Applications (1)

Application Number Title Priority Date Filing Date
JP29276791A Pending JPH05127901A (en) 1991-11-08 1991-11-08 Processor

Country Status (1)

Country Link
JP (1) JPH05127901A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012174114A (en) * 2011-02-23 2012-09-10 Fujitsu Ltd Arithmetic processing unit and slot control method of arithmetic processing unit
JP2014002555A (en) * 2012-06-18 2014-01-09 Fujitsu Ltd Processor

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012174114A (en) * 2011-02-23 2012-09-10 Fujitsu Ltd Arithmetic processing unit and slot control method of arithmetic processing unit
JP2014002555A (en) * 2012-06-18 2014-01-09 Fujitsu Ltd Processor

Similar Documents

Publication Publication Date Title
JPS63225837A (en) System for vector access with distance
JPS58147879A (en) Control system of cache memory
JPH05127901A (en) Processor
EP0240108A2 (en) A data processing system
JPH0630063B2 (en) Microprocessor
JPH05197619A (en) Memory control circuit for multi-cpu
CN112486904A (en) Register file design method and device for reconfigurable processing unit array
JPS63254544A (en) Control system for address conversion
JPH02270032A (en) Loading system
KR100246465B1 (en) Apparatus and method for reducing cycle of microprocessor stack order
JPS6382530A (en) Semiconductor storage device
JPH0594318A (en) Load leveling system of multi cpu system
JPH0731608B2 (en) Information processing equipment
JPS5947668A (en) Information processor
JPS6217777B2 (en)
JPH0736705A (en) Compiler execution system
JPS6269321A (en) Process switching system
JPH02230326A (en) Pipe line information processor
JP2001022581A (en) Data processor and computer readable storage medium
JPH02153444A (en) Memory control circuit
JPS62264345A (en) Processing system in address conversion exception mode of vector processor
JPH04340143A (en) Digital computer
JPS60243743A (en) Multiplex microprogram control system
JPH02254563A (en) Vector data processing system
JPH02272654A (en) Cache memory device