List scheduling in the presence of branches: A theoretical evaluation
The extraction of operation level parallelism from sequential code has become an important problem in compiler research due to the proliferation of superscalar and VLIW architectures. This problem becomes especially hard for code containing a large number of conditional branches. In this paper we ex...
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Buchkapitel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 526 |
|---|---|
| container_issue | |
| container_start_page | 515 |
| container_title | |
| container_volume | |
| creator | Gasperoni, Franco Schwiegelshohn, Uwe |
| description | The extraction of operation level parallelism from sequential code has become an important problem in compiler research due to the proliferation of superscalar and VLIW architectures. This problem becomes especially hard for code containing a large number of conditional branches. In this paper we extend previous work on straight line code scheduling by looking at branching task systems whose control flow graph is acyclic. First, we define an optimality measure based on the probability of the various execution paths. Then, we apply a list scheduling algorithm to these systems and derive a worst case performance guarantee for this method. Finally, we show that there are branching task systems for which this bound is almost tight. |
| doi_str_mv | 10.1007/BFb0024743 |
| format | Book Chapter |
| fullrecord | <record><control><sourceid>springer</sourceid><recordid>TN_cdi_springer_books_10_1007_BFb0024743</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>springer_books_10_1007_BFb0024743</sourcerecordid><originalsourceid>FETCH-LOGICAL-s1393-143f8492435f5b10c3a0bf36e15c8136cb3bcaa7b0aaa7bf29c81fae8b77b1ac3</originalsourceid><addsrcrecordid>eNpFUMtOwzAQNC-JUHrhC3zkEtj1OnbMrVQtIEXiAufINjYNREkVp3w_iYrEHmalmZ3VaBi7QbhDAH3_uHUAQmpJJ2xpdEmFBA2KlDplGSrEnEiaM3Y1CwqV0OqcZUAgcjOZLtkypS-YhoQwRmdsUzVp5MnvwsehbbpP3nR83AW-H0IKnQ-8j9wNtpsO0gNfzVo_hLHxtuXhx7YHOzZ9d80uom1TWP7tBXvfbt7Wz3n1-vSyXlV5QjKUo6RYSiMkFbFwCJ4suEgqYOFLJOUdOW-tdmBnjMJMdLShdFo7tJ4W7Pb4N-2HKWwYatf336lGqOd66v966BcTJ1KO</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>book_chapter</recordtype></control><display><type>book_chapter</type><title>List scheduling in the presence of branches: A theoretical evaluation</title><source>Springer Books</source><creator>Gasperoni, Franco ; Schwiegelshohn, Uwe</creator><contributor>Bougé, Luc ; Robert, Yves ; Fraigniaud, Pierre ; Mignotte, Anne</contributor><creatorcontrib>Gasperoni, Franco ; Schwiegelshohn, Uwe ; Bougé, Luc ; Robert, Yves ; Fraigniaud, Pierre ; Mignotte, Anne</creatorcontrib><description>The extraction of operation level parallelism from sequential code has become an important problem in compiler research due to the proliferation of superscalar and VLIW architectures. This problem becomes especially hard for code containing a large number of conditional branches. In this paper we extend previous work on straight line code scheduling by looking at branching task systems whose control flow graph is acyclic. First, we define an optimality measure based on the probability of the various execution paths. Then, we apply a list scheduling algorithm to these systems and derive a worst case performance guarantee for this method. Finally, we show that there are branching task systems for which this bound is almost tight.</description><identifier>ISSN: 0302-9743</identifier><identifier>ISBN: 3540616276</identifier><identifier>ISBN: 9783540616276</identifier><identifier>EISSN: 1611-3349</identifier><identifier>EISBN: 9783540706366</identifier><identifier>EISBN: 3540706364</identifier><identifier>DOI: 10.1007/BFb0024743</identifier><language>eng</language><publisher>Berlin, Heidelberg: Springer Berlin Heidelberg</publisher><subject>Average Execution Time ; Control Flow Graph ; Execution Path ; List Schedule ; Task System</subject><ispartof>Euro-Par'96 Parallel Processing, 2005, p.515-526</ispartof><rights>Springer-Verlag Berlin Heidelberg 1996</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><relation>Lecture Notes in Computer Science</relation></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://link.springer.com/content/pdf/10.1007/BFb0024743$$EPDF$$P50$$Gspringer$$H</linktopdf><linktohtml>$$Uhttps://link.springer.com/10.1007/BFb0024743$$EHTML$$P50$$Gspringer$$H</linktohtml><link.rule.ids>775,776,780,789,27902,38232,41418,42487</link.rule.ids></links><search><contributor>Bougé, Luc</contributor><contributor>Robert, Yves</contributor><contributor>Fraigniaud, Pierre</contributor><contributor>Mignotte, Anne</contributor><creatorcontrib>Gasperoni, Franco</creatorcontrib><creatorcontrib>Schwiegelshohn, Uwe</creatorcontrib><title>List scheduling in the presence of branches: A theoretical evaluation</title><title>Euro-Par'96 Parallel Processing</title><description>The extraction of operation level parallelism from sequential code has become an important problem in compiler research due to the proliferation of superscalar and VLIW architectures. This problem becomes especially hard for code containing a large number of conditional branches. In this paper we extend previous work on straight line code scheduling by looking at branching task systems whose control flow graph is acyclic. First, we define an optimality measure based on the probability of the various execution paths. Then, we apply a list scheduling algorithm to these systems and derive a worst case performance guarantee for this method. Finally, we show that there are branching task systems for which this bound is almost tight.</description><subject>Average Execution Time</subject><subject>Control Flow Graph</subject><subject>Execution Path</subject><subject>List Schedule</subject><subject>Task System</subject><issn>0302-9743</issn><issn>1611-3349</issn><isbn>3540616276</isbn><isbn>9783540616276</isbn><isbn>9783540706366</isbn><isbn>3540706364</isbn><fulltext>true</fulltext><rsrctype>book_chapter</rsrctype><creationdate>2005</creationdate><recordtype>book_chapter</recordtype><sourceid/><recordid>eNpFUMtOwzAQNC-JUHrhC3zkEtj1OnbMrVQtIEXiAufINjYNREkVp3w_iYrEHmalmZ3VaBi7QbhDAH3_uHUAQmpJJ2xpdEmFBA2KlDplGSrEnEiaM3Y1CwqV0OqcZUAgcjOZLtkypS-YhoQwRmdsUzVp5MnvwsehbbpP3nR83AW-H0IKnQ-8j9wNtpsO0gNfzVo_hLHxtuXhx7YHOzZ9d80uom1TWP7tBXvfbt7Wz3n1-vSyXlV5QjKUo6RYSiMkFbFwCJ4suEgqYOFLJOUdOW-tdmBnjMJMdLShdFo7tJ4W7Pb4N-2HKWwYatf336lGqOd66v966BcTJ1KO</recordid><startdate>20050610</startdate><enddate>20050610</enddate><creator>Gasperoni, Franco</creator><creator>Schwiegelshohn, Uwe</creator><general>Springer Berlin Heidelberg</general><scope/></search><sort><creationdate>20050610</creationdate><title>List scheduling in the presence of branches: A theoretical evaluation</title><author>Gasperoni, Franco ; Schwiegelshohn, Uwe</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-s1393-143f8492435f5b10c3a0bf36e15c8136cb3bcaa7b0aaa7bf29c81fae8b77b1ac3</frbrgroupid><rsrctype>book_chapters</rsrctype><prefilter>book_chapters</prefilter><language>eng</language><creationdate>2005</creationdate><topic>Average Execution Time</topic><topic>Control Flow Graph</topic><topic>Execution Path</topic><topic>List Schedule</topic><topic>Task System</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Gasperoni, Franco</creatorcontrib><creatorcontrib>Schwiegelshohn, Uwe</creatorcontrib></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Gasperoni, Franco</au><au>Schwiegelshohn, Uwe</au><au>Bougé, Luc</au><au>Robert, Yves</au><au>Fraigniaud, Pierre</au><au>Mignotte, Anne</au><format>book</format><genre>bookitem</genre><ristype>CHAP</ristype><atitle>List scheduling in the presence of branches: A theoretical evaluation</atitle><btitle>Euro-Par'96 Parallel Processing</btitle><seriestitle>Lecture Notes in Computer Science</seriestitle><date>2005-06-10</date><risdate>2005</risdate><spage>515</spage><epage>526</epage><pages>515-526</pages><issn>0302-9743</issn><eissn>1611-3349</eissn><isbn>3540616276</isbn><isbn>9783540616276</isbn><eisbn>9783540706366</eisbn><eisbn>3540706364</eisbn><abstract>The extraction of operation level parallelism from sequential code has become an important problem in compiler research due to the proliferation of superscalar and VLIW architectures. This problem becomes especially hard for code containing a large number of conditional branches. In this paper we extend previous work on straight line code scheduling by looking at branching task systems whose control flow graph is acyclic. First, we define an optimality measure based on the probability of the various execution paths. Then, we apply a list scheduling algorithm to these systems and derive a worst case performance guarantee for this method. Finally, we show that there are branching task systems for which this bound is almost tight.</abstract><cop>Berlin, Heidelberg</cop><pub>Springer Berlin Heidelberg</pub><doi>10.1007/BFb0024743</doi><tpages>12</tpages><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0302-9743 |
| ispartof | Euro-Par'96 Parallel Processing, 2005, p.515-526 |
| issn | 0302-9743 1611-3349 |
| language | eng |
| recordid | cdi_springer_books_10_1007_BFb0024743 |
| source | Springer Books |
| subjects | Average Execution Time Control Flow Graph Execution Path List Schedule Task System |
| title | List scheduling in the presence of branches: A theoretical evaluation |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-02T11%3A34%3A15IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-springer&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=bookitem&rft.atitle=List%20scheduling%20in%20the%20presence%20of%20branches:%20A%20theoretical%20evaluation&rft.btitle=Euro-Par'96%20Parallel%20Processing&rft.au=Gasperoni,%20Franco&rft.date=2005-06-10&rft.spage=515&rft.epage=526&rft.pages=515-526&rft.issn=0302-9743&rft.eissn=1611-3349&rft.isbn=3540616276&rft.isbn_list=9783540616276&rft_id=info:doi/10.1007/BFb0024743&rft_dat=%3Cspringer%3Espringer_books_10_1007_BFb0024743%3C/springer%3E%3Curl%3E%3C/url%3E&rft.eisbn=9783540706366&rft.eisbn_list=3540706364&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |