|interaction between loop scheduling and register allocation email@example.com (1992-08-26)|
|Loop scheduling/Register Allocation refs. firstname.lastname@example.org (1992-08-29)|
|From:||email@example.com (+Sanjay Krishnamurthy)|
|Date:||Sat, 29 Aug 1992 20:26:16 GMT|
|Keywords:||registers, optimize, bibliography|
>[Is there} any published paper about the interaction between
>loop scheduling and register allocation and the algorithms for combining
>these two parts?
You might want to look at:
"Register Allocation for Modulo Scheduled Loops: Strategies,
Algorithms and Heuristics," in the proc. of the ACM SIGPLAN
1992 Conf. on Prog. Lang. Design and Implementation,
San Fransisco, June 1992.
A more detailed version appears as a HP Tech. Report(HPL-92-48).
"Compiler Techniques for Optimizing Memory and Register
Usage on the Cray-2," C. Eisenbeis, W. Jalby and A. Lichnewsky,
Intl. Journal of High Speed Computing,
Vol. 2, No. 2 (1990), pp. 193-222.
A conference version of the same paper appears as:
"Squeezing more CPU Performance out of a Cray-2 by Vector
Block Scheduling," Supercomputing 1988, Florida.
The goals of vector register allocation and allocation
during loop scheduling are identical. Hence, try:
"Vector Register Allocation," Randy Allen and Ken Kennedy,
Rice University Computer Science Dept. Tech. Report,
Somebody more familiar with register allocation than I am
might want to add references to this list.
-Sanjay M. Krishnamurthy
Cray Research Superservers
Return to the
Search the comp.compilers archives again.