JPS62267830A - Arithmetic control system for vector processing built-in function in programming language - Google Patents
Arithmetic control system for vector processing built-in function in programming languageInfo
- Publication number
- JPS62267830A JPS62267830A JP11048986A JP11048986A JPS62267830A JP S62267830 A JPS62267830 A JP S62267830A JP 11048986 A JP11048986 A JP 11048986A JP 11048986 A JP11048986 A JP 11048986A JP S62267830 A JPS62267830 A JP S62267830A
- Authority
- JP
- Japan
- Prior art keywords
- program
- vector processing
- function
- built
- language
- 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
- 230000006870 function Effects 0.000 title claims abstract description 39
- 238000012545 processing Methods 0.000 title claims abstract description 35
- 239000013598 vector Substances 0.000 title claims abstract description 30
- 238000000034 method Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Landscapes
- Complex Calculations (AREA)
- Devices For Executing Special Programs (AREA)
Abstract
Description
【発明の詳細な説明】
ミング言語の組込み関数の演算制御方法に関するもので
ある。DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a calculation control method for built-in functions of the Mining language.
組込み関数を有するプログラミング言語において、コン
ノクイラカr4クトル処理化の分析を行い。In a programming language with built-in functions, we analyzed the processing of Konnoki R4 vectors.
ベクトル処理化が可能と判定された組込み関数を実行す
るときに、それがノ・−ドウエアの持つベクトル処理命
令に展開できない場合、従来、上記組込み関数ルーチン
がプログラム領域中に存在し。Conventionally, when an intrinsic function that has been determined to be vector-processable cannot be developed into a vector processing instruction of the node, the aforementioned intrinsic function routine exists in the program area.
そこに・ぐラメータとともに制御を移しその中で関
。The control is transferred there along with the parameter, and the
.
数演算全行い、実行結果を関数の引用側に返すことによ
シベクトル処理関数の演算制御を行っていた。The calculations of the vector processing function were controlled by performing all numerical operations and returning the execution results to the quoting side of the function.
上述した従来の方式では1組込み関数ルーチンに制御を
移行するに際して、戻り番地の記憶、ノクラメータリス
トの作成、関数ルーチンへの制御の移行。In the conventional method described above, when transferring control to one built-in function routine, a return address is stored, a nomura meter list is created, and control is transferred to the function routine.
関数ルーチン内で使用するレジスタ類の退避/回復。Saving/restoring registers used in function routines.
関数の演算結果の受は渡し、関数の引用側への復帰等の
動作が必要となシ、関数本来の演算以外の処理が必要と
なってしまう。Operations such as receiving and passing the result of a function's operation and returning to the quoting side of the function are required, and processing other than the function's original operation is required.
本発明の目的は2以上述べてきた従来技術での問題点を
解決するベクトル処理組込み関数の演算制御方式を提供
することにある。SUMMARY OF THE INVENTION An object of the present invention is to provide an arithmetic control system for vector processing built-in functions that solves the above-mentioned problems in the prior art.
本発明によるベクトル処理組込み関数の演算制御方式は
、プログラミング言語に組込み関数をそなえ、ベクトル
処理命令を有する計算機システムにおいて、プログラム
の制御の流れを解析する手段と、ベクトル処理化の分析
を行いベクトル処理化を行う手段と、プログラムの最適
化を行う手段と2組込み関数のインライン展開を行う手
段を有している。The arithmetic control method for vector processing built-in functions according to the present invention provides a means for analyzing the control flow of a program and an analysis of vector processing in a computer system that has built-in functions in a programming language and has vector processing instructions. It has means for optimizing the program, means for optimizing the program, and means for inline expansion of two built-in functions.
次に2本発明の実施例について図面を参照して説明する
。Next, two 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構成図である。先ず、翻訳
プログラム1はプログラムを記述した原始プログラム2
t−人力し、第1の中間言語3に変換する。次に、翻訳
グログラム1は第1の中間言語3t−人力として、プロ
グラムの制御の流れを解析する手段4とベクトル処理化
の分析を行いベクトル処理化を行う手段5によりベクト
ル処理を施し、第2の中間言語6に変換する。さらに、
翻訳プログラム1は第2の中間言語6をプログラムの最
適化を行う手段7により第3の中間言語8に変換する。FIG. 1 is a block diagram of an embodiment of the present invention. First, the translation program 1 is the source program 2 that describes the program.
t- manually and convert it into the first intermediate language 3. Next, the translation program 1 is subjected to vector processing by means 4 for analyzing the control flow of the program and means 5 for analyzing the vector processing and converting it into vector processing as a human power of the first intermediate language 3t. into intermediate language 6. moreover,
The translation program 1 converts the second intermediate language 6 into the third intermediate language 8 by means 7 for optimizing the program.
さらに、翻訳プログラム1は第3の中間言語8に対し組
込み関数のインライン展開を行う手段9によジインライ
ン展開処理を行い、最終峠な機械語命令の目的プログラ
ム10t−作成する。Furthermore, the translation program 1 performs inline expansion processing on the third intermediate language 8 by means 9 for inline expansion of built-in functions, thereby creating a final target program 10t of machine language instructions.
上記の実施例について、実際のプログラムの流れに基づ
いて説明する。。The above embodiment will be explained based on the flow of an actual program. .
従来技術では、第3図、第4図に示すように。In the prior art, as shown in FIGS. 3 and 4.
ベクトル処理組込み関数の実行において、フロー20に
おいて組込み関数が必要とする・母うメータリストの作
成を行い、フロー30において戻り番地の記憶動作を行
った後に組込み関数ルーチンに制御を移行しくフロー4
0)、フロー50でレジスタ類の退避、フロー60でノ
クラメータの取シ込み動作を行った後に、フロー70で
関数演算の実行を行い、フロー80でレジスタ類の回復
、フロー90での関数値の返却処理を行った後に、関数
の引用側に制御の復帰を行っていた。In the execution of the vector processing intrinsic function, flow 20 creates a meter list required by the intrinsic function, flow 30 stores the return address, and then flow 4 is executed to transfer control to the intrinsic function routine.
0), after saving the registers in flow 50 and importing the nomura meter in flow 60, the function operation is executed in flow 70, the registers are restored in flow 80, and the function value is restored in flow 90. After performing the return process, control was returned to the quoting side of the function.
本発明では第2図に示すように、フロー100でベクト
ル処理組込み関数のインライン展開による実行を行うだ
けで良いので第3図、第4図のフロー20.30,50
.60,80.90の処理が不要となり、ベクトル処理
組込み関数本来の演算以外のオーバーヘッドの削除が可
能となっている。In the present invention, as shown in FIG. 2, it is only necessary to execute the vector processing built-in function by inline expansion in flow 100.
.. 60, 80, and 90 are no longer necessary, and it is possible to eliminate overhead other than the original calculations of the vector processing built-in function.
以上説明したように2本発明によれば、関数本来の演算
以外の処理が不要になり、ベクトル処理組込み関数を実
行するときのオーバーヘッドを削減することができる。As described above, according to the second aspect of the present invention, processing other than the original calculation of the function becomes unnecessary, and the overhead when executing the vector processing built-in function can be reduced.
また1組込み関数のインライン展開を行うことにより、
コンノぐイラの最適化処理の適用範囲が広がり、プログ
ラムの実行速度を速くすることができる効果がある。Also, by performing inline expansion of one built-in function,
This has the effect of expanding the scope of application of Konnoguira's optimization processing and increasing program execution speed.
第1図は本発明の一実施例の構成図、第2図は本発明に
よってプログラム中におけるベクトル処理組込み関数を
引用した部分の流れ図、第3図は従来技術によって第2
図と同じ処理をベクトル処理組込み関数ルーチンへの制
御の移行で実行した流れ図、第4図は従来技術によるベ
クトル処理組込み関数ルーチンの流れ図である。
−4・・・解析手段、5・・・ベクトル処理化手段、7
・・・最適化手段、9・・・インライン展開手段。
、−J−
箆2図
第3図 馬4図FIG. 1 is a block diagram of an embodiment of the present invention, FIG. 2 is a flowchart of a portion of a program that uses vector processing built-in functions according to the present invention, and FIG.
FIG. 4 is a flowchart in which the same process as shown in the figure is executed by transferring control to a vector processing built-in function routine. FIG. 4 is a flow chart of a vector processing built-in function routine according to the prior art. -4... Analysis means, 5... Vector processing means, 7
...Optimization means, 9...Inline expansion means. , -J- Figure 2, Figure 3, Figure 4, Horse
Claims (1)
ル処理命令を有する計算機システムにおいて、プログラ
ムの制御の流れを解析する手段と、ベクトル処理化の分
析を行いベクトル処理化を行う手段と、プログラムの最
適化を行う手段と、組込み関数のインライン展開を行う
手段とを有することを特徴とするベクトル処理組込み関
数の演算制御方式。1. In a computer system with built-in functions in a programming language and vector processing instructions, a means for analyzing the control flow of a program, a means for analyzing vector processing and converting it into vector processing, and a method for program optimization. 1. An arithmetic control method for vector processing built-in functions, characterized by comprising means for performing inline expansion of the built-in functions, and means for inline expansion of the built-in functions.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP11048986A JPS62267830A (en) | 1986-05-16 | 1986-05-16 | Arithmetic control system for vector processing built-in function in programming language |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP11048986A JPS62267830A (en) | 1986-05-16 | 1986-05-16 | Arithmetic control system for vector processing built-in function in programming language |
Publications (1)
Publication Number | Publication Date |
---|---|
JPS62267830A true JPS62267830A (en) | 1987-11-20 |
Family
ID=14537037
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP11048986A Pending JPS62267830A (en) | 1986-05-16 | 1986-05-16 | Arithmetic control system for vector processing built-in function in programming language |
Country Status (1)
Country | Link |
---|---|
JP (1) | JPS62267830A (en) |
-
1986
- 1986-05-16 JP JP11048986A patent/JPS62267830A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
TW356540B (en) | Method for operating a computer program on a computer readable media for execution by a vector processor and by a dual multiprocessor including a vector processor and a RISC processor | |
JPS62267830A (en) | Arithmetic control system for vector processing built-in function in programming language | |
JP3318051B2 (en) | Translation processing method | |
JPH08180094A (en) | Architecture simulator | |
JPS61245239A (en) | Logical circuit system | |
JPH04273533A (en) | In-line expansion system for procedure of translation program | |
JPS60220468A (en) | Vector arithmetic control system | |
JPS5829051A (en) | Operation processing device | |
JPH04343140A (en) | Parallelized processing system | |
JPH0250730A (en) | Language converting system based upon common machine word | |
JP2682206B2 (en) | Optimization method of target program | |
JPS6297031A (en) | Matching system for bit number of mantissa part | |
JPS62271024A (en) | Optimization processing system | |
JPS63138427A (en) | Processing system for scheduling instruction to array | |
JPS6155738A (en) | Determining system of operation mode | |
JPS5960642A (en) | System for optimizing source program | |
JPS62267831A (en) | Arithmetic control system for build-in function in programming language | |
JPH0377141A (en) | Division processing system for vector operation train | |
JPH0713962A (en) | Compiler device | |
JPS616722A (en) | Method for calculating execution cost of sentence | |
JPH04102925A (en) | Programing/execution system for computer | |
JPS63282552A (en) | Speeding-up system for channel program conversion processing | |
JPS62202264A (en) | Vector operation processing system | |
JPS63120369A (en) | Program transformation device | |
JPS61216085A (en) | Vector processing program producing system |