WO2019197866A2 - Parallélisme au niveau de fils de discussion - Google Patents
Parallélisme au niveau de fils de discussion Download PDFInfo
- Publication number
- WO2019197866A2 WO2019197866A2 PCT/IB2018/052432 IB2018052432W WO2019197866A2 WO 2019197866 A2 WO2019197866 A2 WO 2019197866A2 IB 2018052432 W IB2018052432 W IB 2018052432W WO 2019197866 A2 WO2019197866 A2 WO 2019197866A2
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- parallel
- thread
- execution
- execute
- main processor
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
Definitions
- thread which is a sequence of program instructions is divided into a set of independent tasks and such division can either be done by the user or by the compiler dividing the program instructions into set of independent program instructions which can execute in parallel but identifies the sequence of execution of such sets of independent program instructions to get the output.
- the main thread running on the main processor or the Central Processing Unit will execute a special fork instruction and a parallel thread is spawned on the coprocessor with the context provided by the main processor and the parallel thread starts running at the instruction address provided by the special fork instruction.
- Main processor will execute special fork instruction for as many independent parallel tasks or instruction sets which can execute in parallel also taking into account the number of coprocessors which can help it for parallel execution.
- All execution results of the parallel thread running on the coprocessor is buffered and it is not part of the program state.
- the main thread After executing special fork instruction for running parallel tasks or set of instructions, the main thread continues with its execution on the main processor. Now when the main thread reaches the instruction address where one of the parallel task or set of instructions were spawned by it for parallel execution, it will read the buffered execution results and skip the execution of parallel task or set of instructions.
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Advance Control (AREA)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/IB2018/052432 WO2019197866A2 (fr) | 2018-04-08 | 2018-04-08 | Parallélisme au niveau de fils de discussion |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/IB2018/052432 WO2019197866A2 (fr) | 2018-04-08 | 2018-04-08 | Parallélisme au niveau de fils de discussion |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2019197866A2 true WO2019197866A2 (fr) | 2019-10-17 |
Family
ID=68164893
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/IB2018/052432 WO2019197866A2 (fr) | 2018-04-08 | 2018-04-08 | Parallélisme au niveau de fils de discussion |
Country Status (1)
Country | Link |
---|---|
WO (1) | WO2019197866A2 (fr) |
-
2018
- 2018-04-08 WO PCT/IB2018/052432 patent/WO2019197866A2/fr active Application Filing
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CY1122108T1 (el) | Διαταξη συστοιχιας ηλεκτρονικων υπολογιστων για επεξεργασια ενος υπολογιστικου εργου και μεθοδος λειτουργιας αυτης | |
PH12017550124A1 (en) | Decoupled processor instruction window and operand buffer | |
US20070180438A1 (en) | Stall prediction thread management | |
WO2012138950A3 (fr) | Instructions de charge conditionnelle dans un microprocesseur d'exécution hors service | |
WO2016103091A9 (fr) | Parallélisation dynamique de code ayant surveillance de séquences d'instructions répétitives | |
EP1124182A3 (fr) | Communication des résultats dans un processeur, et procédés de compilation | |
JP2014513373A5 (fr) | ||
JP2016207232A5 (ja) | プロセッサ | |
EP2778907A3 (fr) | Boucles de parallélisation en présence d'alias possibles de mémoire | |
RU2014118858A (ru) | Команда векторного типа для поиска равнозначного элемента | |
WO2012078735A3 (fr) | Réalisation d'appels de fonctions au moyen de registres à instruction unique et données multiples (simd) | |
TW200731739A (en) | Cryptography system and elliptic curve operation method involved thereof | |
WO2015124116A3 (fr) | Système et procédé permettant d'isoler l'exécution des entrées et sorties grâce a un compilateur et à un système d'exploitation | |
WO2007008519A3 (fr) | Calcul par machine a elements actifs | |
WO2013144734A3 (fr) | Optimisation de fusion d'instructions | |
WO2009076324A3 (fr) | Matériel informatique à base de fil et programme « strandware » (logiciel) à optimisation dynamique pour un système de microprocesseur haute performance | |
JP2010204979A5 (ja) | コンパイル方法 | |
WO2017052811A3 (fr) | Processeurs, procédés, systèmes et instructions pour exponentiation modulaire sécurisée | |
TW200709051A (en) | Real-time control apparatus having a multi-thread processor | |
WO2015017129A4 (fr) | Conduite de gpu multifil | |
WO2006083046A3 (fr) | Procedes et appareils permettant d'obtenir une interface de programmation a changement de taches | |
WO2017053022A8 (fr) | Scalarisation spéculative dans le traitement vectoriel | |
EP1770516A4 (fr) | Procede de commande 'l' destine a commander l'execution d'un programme/instruction et d'une architecture, et processeur associe | |
WO2019197866A2 (fr) | Parallélisme au niveau de fils de discussion | |
TW201614481A (en) | Flexible instruction execution in a processor pipeline |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18914382 Country of ref document: EP Kind code of ref document: A2 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 18914382 Country of ref document: EP Kind code of ref document: A2 |