Deprecated: The each() function is deprecated. This message will be suppressed on further calls in /home/zhenxiangba/zhenxiangba.com/public_html/phproxy-improved-master/index.php on line 456
JPH0664584B2 - Vector sum operation unit - Google Patents
[go: Go Back, main page]

JPH0664584B2 - Vector sum operation unit - Google Patents

Vector sum operation unit

Info

Publication number
JPH0664584B2
JPH0664584B2 JP59279609A JP27960984A JPH0664584B2 JP H0664584 B2 JPH0664584 B2 JP H0664584B2 JP 59279609 A JP59279609 A JP 59279609A JP 27960984 A JP27960984 A JP 27960984A JP H0664584 B2 JPH0664584 B2 JP H0664584B2
Authority
JP
Japan
Prior art keywords
vector
addition
elements
storage means
sum
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.)
Expired - Lifetime
Application number
JP59279609A
Other languages
Japanese (ja)
Other versions
JPS61160175A (en
Inventor
康博 中井
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.)
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 JP59279609A priority Critical patent/JPH0664584B2/en
Publication of JPS61160175A publication Critical patent/JPS61160175A/en
Publication of JPH0664584B2 publication Critical patent/JPH0664584B2/en
Anticipated expiration legal-status Critical
Expired - Lifetime 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
    • G06F15/76Architectures of general purpose stored program computers
    • G06F15/80Architectures of general purpose stored program computers comprising an array of processing units with common control, e.g. single instruction multiple data processors
    • G06F15/8053Vector processors

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Theoretical Computer Science (AREA)
  • Computing Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Complex Calculations (AREA)

Description

【発明の詳細な説明】 産業上の利用分野 本発明は、ベクトル演算装置に関し、特に、ベクトル要
素の総和演算をなす演算装置に関する。
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vector operation device, and more particularly, to an operation device for performing a sum operation of vector elements.

従来の技術 一般に、ベクトル演算装置にあつては、複数のベクトル
の間での要素対応の演算をなすように構成されている。
従つて、1つのベクトルについてその要素の総和を求め
る場合には、通常のベクトル要素対応の加算手段の他に
特別な加算手段を設けるか、あるいは通常のベクトル要
素対応の加算手段を用いることになる。
2. Description of the Related Art Generally, a vector arithmetic device is configured to perform element-corresponding arithmetic operations among a plurality of vectors.
Therefore, when the total sum of the elements of one vector is obtained, special addition means is provided in addition to the addition means corresponding to the normal vector element, or the addition means corresponding to the normal vector element is used. .

発明が解決しようとする問題点 その為に、このベクトル加算手段に対するデータとして
1つのベクトル内の互いに別の要素を同時に供給する必
要があり、そのための制御手段、データの出力手段を設
ける必要が生ずることとなる。その結果、演算装置のハ
ードウエアが大きくなると共に制御が複雑となる。又、
後者の場合でベクトル格納手段より別の要素を同時に出
力するためには一般に出力に長い時間が必要であり、通
常の複数のベクトル間での要素対応の演算よりも演算手
段占有時間が長くなり、ベクトル演算装置の性能低下の
要因ともなる。
Problems to be solved by the invention Therefore, it is necessary to simultaneously supply different elements in one vector as data to the vector addition means, and it is necessary to provide control means and data output means therefor. It will be. As a result, the hardware of the arithmetic unit becomes large and the control becomes complicated. or,
In the latter case, in order to output another element from the vector storage means at the same time, it generally requires a long time to output, and the operation means occupancy time becomes longer than the operation corresponding to an element between a plurality of ordinary vectors, It also causes a decrease in the performance of the vector operation device.

本発明は従来の技術に内在する上記問題点を解消する為
になされたものであり、従つて本発明の目的は、通常の
ベクトル要素対応の加算と加算手段を共用できる構成を
もち、簡単な構成で性能低下を招来することなくベクト
ル要素の総和を1つの命令で実行できる新規なベクトル
演算装置を提供することにある。
The present invention has been made in order to solve the above-mentioned problems inherent in the prior art. Therefore, the object of the present invention is to have a structure in which the addition and the addition means corresponding to normal vector elements can be shared, It is an object of the present invention to provide a novel vector operation device capable of executing the total sum of vector elements with one instruction without causing performance deterioration in the configuration.

問題点を解決するための手段 本発明のベクトル総和演算装置は、ベクトルに含まれる
M個(Mは正の整数で、M=2:mは正の整数)のベ
クトル要素の総和を求める場合、それぞれ同一要素のベ
クトル要素からなる部分ベクトルが格納される2つのベ
クトル格納手段と、このベクトル格納手段のそれぞれか
ら順次出力される前記ベクトル要素間の加算を順次行う
ベクトル加算手段と、このベクトル加算手段から順次出
力される加算結果ベクトル要素を前記2つのベクトル格
納手段に均等に分配するデータ分配手段と、前記Mから
前記mを求める第1の処理と加算回数2m-1だけ前記ベ
クトル加算手段に加算を行わせる第2の処理と該2m-1
回の加算を終了する度に該mから1を減じて該第2の処
理を行う第3の処理とを行う制御手段とを備える。
Means for Solving Problems When the vector sum operation device of the present invention obtains the sum of M vector elements (M is a positive integer, M = 2 m : m is a positive integer) included in the vector , Two vector storage means for storing partial vectors each consisting of vector elements of the same element, vector addition means for sequentially performing addition between the vector elements sequentially output from each of the vector storage means, and this vector addition Data distribution means for evenly distributing the addition result vector elements sequentially output from the means to the two vector storage means, the first processing for obtaining the m from the M, and the vector addition means for the number of additions 2 m-1 2 m-1 and the second process for
And a control unit that performs a third process of performing the second process by subtracting 1 from the m each time the addition is completed.

発明の実施例 以下本発明をその好ましい一実施例について図面を参照
して詳細に説明する。
BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, a preferred embodiment of the present invention will be described in detail with reference to the drawings.

第1図は本発明の一実施例を示すブロツク構成図であ
る。図において、本実施例はベクトル格納手段100、200
を持ち、それぞれデータバス10、20を通してベクトル要
素がベクトル加算手段300に転送される。又データバス1
1、21はそれぞれ外部からベクトル格納手段100、200に
ベクトル要素を転送するデータバスである。ベクトル加
算手段300はデータバス30によりベクトル要素の加算結
果をデータ分配手段400に転送し、データ分配手段400は
加算結果ベクトルの要素を交互にデータバス40、41を通
してベクトル格納手段100、200に転送する。算出手段50
0は制御バス50、51、52、53を持つ。制御バス50はベク
トル加算動作の有効なベクトル要素数をベクトル格納手
段100、200、ベクトル加算手段300、データ分配手段400
に与え、その要素数に従つてベクトル要素対応の加算を
行わせる。又制御バス51は、ベクトル加算動作の終了を
受け、制御バス52は総和をとるベクトルの有効なベクト
ル要素数を受け、制御バス53はベクトル総和の演算の終
了を知らせるバスである。
FIG. 1 is a block diagram showing an embodiment of the present invention. In the figure, this embodiment shows vector storage means 100, 200.
And vector elements are transferred to the vector addition means 300 through the data buses 10 and 20, respectively. Data bus 1
Reference numerals 1 and 21 are data buses for transferring vector elements from the outside to the vector storage means 100 and 200, respectively. The vector addition means 300 transfers the addition result of the vector elements to the data distribution means 400 via the data bus 30, and the data distribution means 400 alternately transfers the elements of the addition result vector to the vector storage means 100, 200 through the data buses 40, 41. To do. Calculation means 50
0 has control buses 50, 51, 52, 53. The control bus 50 stores the number of effective vector elements in the vector addition operation as vector storage means 100, 200, vector addition means 300, and data distribution means 400.
To the vector element, and the addition corresponding to the vector element is performed according to the number of elements. The control bus 51 receives the end of the vector addition operation, the control bus 52 receives the effective number of vector elements of the vector for which the sum is added, and the control bus 53 is a bus which notifies the end of the calculation of the vector sum.

次に第2図を参照すると、算出手段500はエンコーダ51
0、−1カウンタ520、デコーダ530、“0”比較器540を
含む。エンコーダ510は制御バス52により与えられる総
和をとるベクトルの有効な要素数Mを受け、−1カウン
タ520にM=2なるmをセットし、デコーダ530は−1
カウンタ520の値mにより2m-1を制御バス50よりベクト
ル加算動作の有効なベクトル要素数として出力する機能
を持つ。
Next, referring to FIG. 2, the calculation means 500 includes an encoder 51.
It includes a 0, -1 counter 520, a decoder 530, and a "0" comparator 540. The encoder 510 receives the effective number of elements M of the vector for summing given by the control bus 52, sets m to M = 2 m in the −1 counter 520, and the decoder 530 outputs −1.
It has a function of outputting 2 m -1 from the control bus 50 as the number of effective vector elements for vector addition operation according to the value m of the counter 520.

次に以上の構成を持つ本実施例の動作を詳細に説明す
る。第3図を参照すると、ベクトルAは総和をとる対象
となるベクトルを表わす。今、ベクトル要素数が“8”
である場合を例にとると、第4図に示すように、ベクト
ル格納手段100、200それぞれに部分ベクトルであるベク
トルB、ベクトルCとしてセツトされ、−1カウンタ52
0には“3”がセツトされる。ここで、デコード手段530
よりベクトル加算動作の有効なベクトル要素数“4”が
制御バス50に出力され、ベクトル加算動作が実行された
あと加算結果ベクトルの要素がデータ分配手段400によ
つて交互に分配されてベクトル格納手段100、200にそれ
ぞれベクトルB′、ベクトルC′に示すようにセツトさ
れる。又同時に−1カウンタ520はカウントを行い、カ
ウンタの値が“2”に更新される(ここで、ai,jはベク
トル要素aiとajの加算結果を表わし、同様にai,jとak,l
の加算結果はai,j,k,lという様に表わしていくものとす
る。)。
Next, the operation of this embodiment having the above configuration will be described in detail. Referring to FIG. 3, a vector A represents a vector for which a sum is taken. The number of vector elements is now "8"
For example, as shown in FIG. 4, the vector storing means 100 and 200 are set as the partial vectors vector B and vector C, respectively, and the -1 counter 52
"3" is set to 0. Here, the decoding means 530
More effective vector element number "4" of the vector addition operation is output to the control bus 50, the elements of the addition result vector are alternately distributed by the data distribution means 400 after the vector addition operation is executed, and the vector storage means 100 and 200 are set as shown in vector B'and vector C ', respectively. At the same time, the -1 counter 520 counts and the counter value is updated to "2" (where ai, j represents the addition result of the vector elements ai and aj, and similarly ai, j and ak, l).
The addition result of is expressed as ai, j, k, l. ).

続いて、今と同様にデコード手段530より出力されるベ
クトル加算動作の有効なベクトル要素数により、ベクト
ル加算動作が−1カウンタ520の値が“0”になるまで
くり返して実行される。その結果、第4図のベクトル
B″に示すように、ベクトル格納手段100にこの場合の
ベクトルAの総和であるa1,2,3,4,5,6,7がセツトされ、
“0”比較器540によりベクトルの総和が求められたこ
とが制御バス53を介して外部に対して出力される。
Then, the vector addition operation is repeated until the value of the -1 counter 520 becomes "0" by the effective number of vector elements of the vector addition operation output from the decoding means 530 as in the present case. As a result, as shown by the vector B ″ in FIG. 4, a 1,2,3,4,5,6,7 which is the total sum of the vector A in this case is set in the vector storing means 100,
The fact that the sum of the vectors has been obtained by the "0" comparator 540 is output to the outside via the control bus 53.

通常のベクトル要素対応の加算を実行するには、加算し
ようとするベクトルのベクトル要素をデータバス11、21
を通して外部からそれぞれベクトル格納手段100、200に
格納し、それらの出力についてベクトル加算手段300に
よりベクトル要素対応の加算をして出力し、それを任意
の記憶手段に格納するようにすることは言うまでもない
ことである。
To perform a normal vector element-aware addition, add the vector elements of the vectors you want to add to the data buses 11, 21.
It is needless to say that they are externally stored in the vector storage means 100 and 200 respectively, and their outputs are subjected to addition corresponding to vector elements by the vector addition means 300 and output, and then stored in arbitrary storage means. That is.

発明の効果 以上の結果から判るように、本発明によれば、通常のベ
クトル演算装置にデータ分配手段400、算出手段500を付
加することにより1つの命令であるベクトルAの総和を
求めることができるという効果が得られる。
EFFECTS OF THE INVENTION As can be seen from the above results, according to the present invention, by adding the data distribution means 400 and the calculation means 500 to the ordinary vector operation device, the total sum of the vector A, which is one instruction, can be obtained. The effect is obtained.

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

第1図は本発明の一実施例を示すブロツク構成図、第2
図は第1図に示した算出手段500の詳細なブロツク構成
図、第3図は総和を求めるベクトルの図、第4図は第1
図に示した一実施例の動作を説明するための図である。 100、200…ベクトル格納手段、300…ベクトル加算手
段、400…データ分配手段、500…算出手段、510…デコ
ーダ、520…−1カウンタ、530…エンコーダ、540…
“0”比較器、10、11、20、21、30、40、41…データバ
ス、50、51、52、53…制御バス
FIG. 1 is a block diagram showing an embodiment of the present invention, and FIG.
FIG. 4 is a detailed block diagram of the calculation means 500 shown in FIG. 1, FIG. 3 is a diagram of a vector for obtaining the total sum, and FIG.
It is a figure for demonstrating operation | movement of one Example shown in the figure. 100, 200 ... Vector storage means, 300 ... Vector addition means, 400 ... Data distribution means, 500 ... Calculation means, 510 ... Decoder, 520 ... -1 counter, 530 ... Encoder, 540 ...
"0" comparator, 10, 11, 20, 21, 30, 40, 41 ... Data bus, 50, 51, 52, 53 ... Control bus

Claims (1)

【特許請求の範囲】[Claims] 【請求項1】ベクトルに含まれるM個(Mは正の整数
で、M=2:mは正の整数)のベクトル要素の総和を
求める場合、それぞれ同一要素数のベクトル要素からな
る部分ベクトルが格納される2つのベクトル格納手段
と、 このベクトル格納手段のそれぞれから順次出力される前
記ベクトル要素間の加算を順次行うベクトル加算手段
と、 このベクトル加算手段から順次出力される加算結果ベク
トル要素を前記2つのベクトル格納手段に均等に分配す
るデータ分配手段と、 前記Mから前記mを求める第1の処理と加算回数2m-1
だけ前記ベクトル加算手段に加算を行わせる第2の処理
と該2m-1回の加算を終了する度に該mから1を減じて
該第2の処理を行う第3の処理とを行う制御手段とを備
えたことを特徴とするベクトル総和演算装置。
1. When calculating the sum of M vector elements (M is a positive integer and M = 2 m : m is a positive integer) contained in the vector, each sub vector is composed of the same number of vector elements. Are stored in the vector storage means, vector addition means for sequentially performing addition between the vector elements sequentially output from each of the vector storage means, and addition result vector elements sequentially output from the vector addition means. Data distribution means for evenly distributing to the two vector storage means, first processing for obtaining the m from the M, and number of additions 2 m-1
Control for performing a second process for causing the vector adding means to perform the addition and a third process for performing the second process by subtracting 1 from the m each time the 2 m-1 times of addition are completed And a vector sum calculating device.
JP59279609A 1984-12-31 1984-12-31 Vector sum operation unit Expired - Lifetime JPH0664584B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP59279609A JPH0664584B2 (en) 1984-12-31 1984-12-31 Vector sum operation unit

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP59279609A JPH0664584B2 (en) 1984-12-31 1984-12-31 Vector sum operation unit

Publications (2)

Publication Number Publication Date
JPS61160175A JPS61160175A (en) 1986-07-19
JPH0664584B2 true JPH0664584B2 (en) 1994-08-22

Family

ID=17613366

Family Applications (1)

Application Number Title Priority Date Filing Date
JP59279609A Expired - Lifetime JPH0664584B2 (en) 1984-12-31 1984-12-31 Vector sum operation unit

Country Status (1)

Country Link
JP (1) JPH0664584B2 (en)

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS5342583B2 (en) * 1973-12-18 1978-11-13
JPS5532161A (en) * 1978-08-29 1980-03-06 Fujitsu Ltd Integration processing unit
JPS5947643A (en) * 1982-09-13 1984-03-17 Hitachi Ltd Arithmetic processing system

Also Published As

Publication number Publication date
JPS61160175A (en) 1986-07-19

Similar Documents

Publication Publication Date Title
GB1537504A (en) Network computer system
IE56105B1 (en) Inverse discrete cosine transform calculation processor
KR101202445B1 (en) Processor
JPH0664584B2 (en) Vector sum operation unit
JP2958968B2 (en) Discrete cosine arithmetic unit
JPS6226579A (en) Vector sum-total arithmetic unit
JPS6123274A (en) Vector processor
JPS60144875A (en) Inter-vector element arithmetic device
JPH0434191B2 (en)
JPS6349872A (en) Loading system for firmware
JPS6058502B2 (en) information processing system
CA1270954A (en) Apparatus for arithmetic processing
JPH06214754A (en) Rounding operation circuit and image processing apparatus having the same
JPS60144876A (en) Inter-vector element arithmetic device
JPH0784939A (en) Direct memory access controller
JPS61222313A (en) Time changing digital filter
JPH04227589A (en) Device and method for allocating data flow program
JPH0395683A (en) Picture processor
JPS6127775B2 (en)
JPH01108676A (en) Vector processor
JPS6190265A (en) Digital signal processor
JPH07210371A (en) Pipeline type division processor
JPH04156590A (en) Pixel read-out circuit
JPS58166581A (en) Addressing system of memory
JPH02294873A (en) Vector data processing system