Generating code adapted for interlinking legacy scalar code and extended vector code
Abstract
Mechanisms for intermixing code are provided. Source code is received for compilation using an extended Application Binary Interface (ABI) that extends a legacy ABI and uses a different register configuration than the legacy ABI. First compiled code is generated based on the source code, the first compiled code comprising code for accommodating the difference in register configurations used by the extended ABI and the legacy ABI. The first compiled code and second compiled code are intermixed to generate intermixed code, the second compiled code being compiled code that uses the legacy ABI. The intermixed code comprises at least one call instruction that is one of a call from the first compiled code to the second compiled code or a call from the second compiled code to the first compiled code. The code for accommodating the difference in register configurations is associated with the at least one call instruction.
- Inventors:
- Issue Date:
- Research Org.:
- International Business Machines Corp., Armonk, NY (United States)
- Sponsoring Org.:
- USDOE
- OSTI Identifier:
- 1084345
- Patent Number(s):
- 8458677
- Application Number:
- 12/544,258
- Assignee:
- International Business Machines Corporation (Armonk, NY)
- Patent Classifications (CPCs):
-
G - PHYSICS G06 - COMPUTING G06F - ELECTRIC DIGITAL DATA PROCESSING
- DOE Contract Number:
- B554331
- Resource Type:
- Patent
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING
Citation Formats
Gschwind, Michael K. Generating code adapted for interlinking legacy scalar code and extended vector code. United States: N. p., 2013.
Web.
Gschwind, Michael K. Generating code adapted for interlinking legacy scalar code and extended vector code. United States.
Gschwind, Michael K. Tue .
"Generating code adapted for interlinking legacy scalar code and extended vector code". United States. https://www.osti.gov/servlets/purl/1084345.
@article{osti_1084345,
title = {Generating code adapted for interlinking legacy scalar code and extended vector code},
author = {Gschwind, Michael K},
abstractNote = {Mechanisms for intermixing code are provided. Source code is received for compilation using an extended Application Binary Interface (ABI) that extends a legacy ABI and uses a different register configuration than the legacy ABI. First compiled code is generated based on the source code, the first compiled code comprising code for accommodating the difference in register configurations used by the extended ABI and the legacy ABI. The first compiled code and second compiled code are intermixed to generate intermixed code, the second compiled code being compiled code that uses the legacy ABI. The intermixed code comprises at least one call instruction that is one of a call from the first compiled code to the second compiled code or a call from the second compiled code to the first compiled code. The code for accommodating the difference in register configurations is associated with the at least one call instruction.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2013},
month = {6}
}
Works referenced in this record:
Achieving optimum coded diversity with scalar codes
journal, July 2001
- Ganesan, G.; Stoica, P.
- IEEE Transactions on Information Theory, Vol. 47, Issue 5
Scalar result producing method in vector/scalar system by vector unit from vector results according to modifier in vector instruction
patent, March 2008
- Barlow, Stephen; Bailey, Neil; Ramsdale, Timothy
- US Patent Document 7,350,057
Language independent optimal size-based storage allocation
patent, August 1999
- Caldwell, Jeffrey
- US Patent Document 5,940,621
Method and arrangement for pattern recognition on the basis of statistics
patent, March 2001
- Breuer, Thomas; Hanisch, Wilfried; Franke, Jurgen
- US Patent Document 6,205,247
SIMD datapath coupled to scalar/vector/address/conditional data register file with selective subpath scalar processing mode
patent, January 2005
- Gschwind, Michael K.; Hofstee, Harm Peter; Hopkins, Martin E.
- US Patent Document 6,839,828
Linking scalar results directly to scalar operation inputs on a bidirectional databus in a computer which superpositions vector and scalar operations
patent, June 1989
- Cook, Erick M.; Phelps, Andrew Everett; Potash, Hanan
- US Patent Document 4,837,730
Wide instruction word architecture central processor
patent, May 1995
- Babaian, Boris A.; Volkonsky, Vladimir J.; Sakhin, July K.
- US Patent Document 5,418,975
Scalar/Vector Processor
patent-application, October 2005
- Van Berkel, Cornelis Hermanus; Meuwissen, Patrick Peter Elizabeth; Engin, Nur
- US Patent Application 10/515451; 20050240644
Method and apparatus for the inter-operation of differing architectural and run time conventions
patent, March 2001
- Coutant, Cary A.; Buzbee, William B.; Istvan, Anthony F.
- US Patent Document 6,199,202