High performance code generation for VLIW digital signal processors
VLIW (Very Long Instruction Word) architecture has been widely adopted in latest digital signal processor designs to meet the ever increasing need of computing power in, e.g. multimedia applications. In this paper, we present an efficient and retargetable code generation tool for VLIW based DSPs. To...
Gespeichert in:
| Veröffentlicht in: | 2000 IEEE Workshop on SiGNAL PROCESSING SYSTEMS. SiPS 2000. Design and Implementation (Cat. No.00TH8528) S. 683 - 692 |
|---|---|
| Hauptverfasser: | , |
| Format: | Tagungsbericht |
| Sprache: | Englisch |
| Veröffentlicht: |
IEEE
2000
|
| Schlagworte: | |
| ISBN: | 0780364880, 9780780364882 |
| ISSN: | 1520-6130 |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Abstract | VLIW (Very Long Instruction Word) architecture has been widely adopted in latest digital signal processor designs to meet the ever increasing need of computing power in, e.g. multimedia applications. In this paper, we present an efficient and retargetable code generation tool for VLIW based DSPs. To make the code generation tool retargetable, we first developed a versatile coding constraint model to faithfully characterize the target machine's limitations on hardware resource, pipeline execution and other specific instruction usage. Second, since loop executions account for most of the time consumption, the software pipelining technique was employed to overlap the execution of successive iterations. To generate efficient code for real time applications, a simulated evolution (SE) based code generation module was introduced to derive the steady state loop scheduling. Effective heuristics subject to various coding constraints were developed. In addition, to alleviate the scheduling overhead in checking coding/resource constraints repetitively, bit-parallel verification schemes was devised as well. Several test benches on TI TMS320C62/67X DSP have been conducted to verify the effectiveness of our tool and preliminary results show that it can achieve near hand optimized quality code. |
|---|---|
| AbstractList | VLIW (Very Long Instruction Word) architecture has been widely adopted in latest digital signal processor designs to meet the ever increasing need of computing power in, e.g. multimedia applications. In this paper, we present an efficient and retargetable code generation tool for VLIW based DSPs. To make the code generation tool retargetable, we first developed a versatile coding constraint model to faithfully characterize the target machine's limitations on hardware resource, pipeline execution and other specific instruction usage. Second, since loop executions account for most of the time consumption, the software pipelining technique was employed to overlap the execution of successive iterations. To generate efficient code for real time applications, a simulated evolution (SE) based code generation module was introduced to derive the steady state loop scheduling. Effective heuristics subject to various coding constraints were developed. In addition, to alleviate the scheduling overhead in checking coding/resource constraints repetitively, bit-parallel verification schemes was devised as well. Several test benches on TI TMS320C62/67X DSP have been conducted to verify the effectiveness of our tool and preliminary results show that it can achieve near hand optimized quality code. |
| Author | Yin-Tsung Hwang Ying-Chou Chuang |
| Author_xml | – sequence: 1 surname: Yin-Tsung Hwang fullname: Yin-Tsung Hwang organization: Inst. of Electron. & Inf. Eng., Nat. Yunlin Univ. of Sci. & Technol., Taiwan – sequence: 2 surname: Ying-Chou Chuang fullname: Ying-Chou Chuang |
| BookMark | eNotj09LwzAYhwNOcJu7i6d8gdY3SZM0RynqCgWF-ec42vRNjWxNSXrx21uYp-fw_HjgtyGrMYxIyB2DnDEwD4f67ZBzAMjLUmmlrsgGdAlCFWUJK7JmkkOmmIAbskvpZxkyzqQ2Zk2qvR--6YTRhXhuR4vUhh7pgCPGdvZhpIugn039RXs_-Lk90eSHccEUg8WUQky35Nq1p4S7f27Jx_PTe7XPmteXunpsMs9BzJnpOBZOG5RaGLCl6TpECZ3uLXeWLdIyrfpCtVxIhxJdZznrDefGomZabMn9pesR8ThFf27j7_FyWfwBYPBMkA |
| ContentType | Conference Proceeding |
| DBID | 6IE 6IL CBEJK RIE RIL |
| DOI | 10.1109/SIPS.2000.886766 |
| DatabaseName | IEEE Electronic Library (IEL) Conference Proceedings IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume IEEE Xplore All Conference Proceedings IEEE Electronic Library (IEL) IEEE Proceedings Order Plans (POP All) 1998-Present |
| DatabaseTitleList | |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering |
| EndPage | 692 |
| ExternalDocumentID | 886766 |
| GroupedDBID | 6IE 6IL ABLEC AFFNX ALMA_UNASSIGNED_HOLDINGS CBEJK IEGSK RIE RIL |
| ID | FETCH-LOGICAL-i203t-9b2e4f79e57390c89bbee50b7dc2fc12e4c176d46a235fe5efbc21d9229ce7173 |
| IEDL.DBID | RIE |
| ISBN | 0780364880 9780780364882 |
| ISICitedReferencesCount | 1 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000165893400068&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1520-6130 |
| IngestDate | Tue Aug 26 18:17:20 EDT 2025 |
| IsPeerReviewed | false |
| IsScholarly | false |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-i203t-9b2e4f79e57390c89bbee50b7dc2fc12e4c176d46a235fe5efbc21d9229ce7173 |
| PageCount | 10 |
| ParticipantIDs | ieee_primary_886766 |
| PublicationCentury | 2000 |
| PublicationDate | 20000000 |
| PublicationDateYYYYMMDD | 2000-01-01 |
| PublicationDate_xml | – year: 2000 text: 20000000 |
| PublicationDecade | 2000 |
| PublicationTitle | 2000 IEEE Workshop on SiGNAL PROCESSING SYSTEMS. SiPS 2000. Design and Implementation (Cat. No.00TH8528) |
| PublicationTitleAbbrev | SIPS |
| PublicationYear | 2000 |
| Publisher | IEEE |
| Publisher_xml | – name: IEEE |
| SSID | ssj0001215799 ssj0020401 |
| Score | 1.3059651 |
| Snippet | VLIW (Very Long Instruction Word) architecture has been widely adopted in latest digital signal processor designs to meet the ever increasing need of computing... |
| SourceID | ieee |
| SourceType | Publisher |
| StartPage | 683 |
| SubjectTerms | Computer aided instruction Computer architecture Digital signal processing Digital signal processors Multimedia computing Pipeline processing Process design Signal design Signal generators VLIW |
| Title | High performance code generation for VLIW digital signal processors |
| URI | https://ieeexplore.ieee.org/document/886766 |
| WOSCitedRecordID | wos000165893400068&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV27bsIwFLUK6tAubSlV3_LQNZDY8WtGRUWqEBJ9sKHEvkYsgCD0--tHKFTq0ilOrCiRbdk699xzLkJP1qQFt4ImANwBFMkhUaqwCVc5cEYkYxCLTYjhUE4malT7bActDACE5DPo-Gbg8s1Sb32orCslF5w3UEMIHqVaB-GUjAWCqsZabm1Gq9SAjmgaELv0pJtbsLXxzu6e7OjLVHXHg9E4yFc68WO_iq6EM6d_9q-_PUftvXYPj35OpQt0BIsWOj2wHbxEPZ_cgVd7yQD2wnY8Cw7UfqKw68Afr4NPbOYzX1UE-zQPd1lFWcFyvWmj9_7zW-8lqYspJHOS0ipRJYHcCgVMUJVqqcoSgKWlMJpYnblOnQlucl4QyiwwsKUmmVGEKA2eqr9CzcVyAdcIu32JSpNbTRVz6E6rsqCQlVS4V4lDlDeo5Udjuop-GdM4ELd_Pr1DJ1Hc7oMa96hZrbfwgI71VzXfrB_DHH8DlEegOw |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3LTsJAFJ0omqgbFTG-nYXbQjvTea2NBCISElDZkXZ6h7ABwsPvdx5FMHHjqo9J02bupJNz7z3nIPRkijjjRtAIgFuAIjlESmUm4ioFzohkDILZhOh25XCoeqXOtufCAIBvPoO6O_W1_GKm1y5V1pCSC8730YEzzirJWjsJlYT5ElWJtuzqDGKpHh_R2GN26cpudsmW0juba7IpYMaq0W_3-p7AUg-v-2W74ned5um_vvcM1bbsPdz72ZfO0R5Mq-hkR3jwAj279g4835IGsKO247HXoHahwnYAf3Tan7iYjJ2vCHaNHvYwD8SC2WJZQ-_Nl8FzKyrtFKIJiekqUjmB1AgFTFAVa6nyHIDFuSg0MTqxgzoRvEh5RigzwMDkmiSFIkRpcMX6S1SZzqZwhbD9M1FZpEZTZUMhtcozCklOhX2UWEx5japuNkbzoJgxChNx8-fdR3TUGrx1Rp129_UWHQequ0tx3KHKarGGe3Sov1aT5eLBx_sbJYujhA |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=2000+IEEE+Workshop+on+SiGNAL+PROCESSING+SYSTEMS.+SiPS+2000.+Design+and+Implementation+%28Cat.+No.00TH8528%29&rft.atitle=High+performance+code+generation+for+VLIW+digital+signal+processors&rft.au=Yin-Tsung+Hwang&rft.au=Ying-Chou+Chuang&rft.date=2000-01-01&rft.pub=IEEE&rft.isbn=9780780364882&rft.issn=1520-6130&rft.spage=683&rft.epage=692&rft_id=info:doi/10.1109%2FSIPS.2000.886766&rft.externalDocID=886766 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1520-6130&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1520-6130&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1520-6130&client=summon |

