Method for the migration of hardware-proximate, subprogram-indep

Data processing: structural design – modeling – simulation – and em – Emulation – Compatibility emulation

Patent

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

703 26, 712234, 717 8, G06F 9455

Patent

active

061285900

DESCRIPTION:

BRIEF SUMMARY
The invention is directed to a method for the migration of hardware-proximate, subprogram-independent program code present for a source hardware into a program code for a destination hardware with an architecture deviating from the source hardware, the program code to be migrated is composed of portable and non-portable program parts.
The increasing demands made of the performance capability of computer systems of any size often require the change to new computer generations, for example to computers having RISC architecture. However, in order to be able to utilize the higher performance capability of these computers without thereby having to forego the extensive and valuable software base that already exists, a transfer of the existing programs for operating systems, applications, voice compilers, etc., is necessary. The parts of these programs present in the source code of a standardized language can usually be migrated onto the new computer (referred to here as destination hardware, by simple recompiling). The hardware-dependent program parts written, for example, in assembler language, by contrast, must either be transformed into corresponding program code for the destination hardware by complicated methods on the assembler or object code level or must be implemented on the new computer at the running time in the framework of an emulation. What is thereby disadvantageous is that the execution speed of a program under an emulation usually turns up many times slower and given the same program recompiled for the destination hardware. Since a majority of the components of a program can be transferred onto destination hardware by recompiling, it is desirable to combine these portable program parts as well as the remaining hardware-specific program parts such to form a program code runnable on the destination hardware that only the non-portable parts of the program must be subjected to an emulation--Kristy Andrews et al., "Migrating a CISC Computer Family onto RISC via Object Code Translation", ACM SIGPLAN NOTICES, 27 (1992), September, No. 9, New York, US, pages 213 through 222.
Previous methods known, for example, from R. L. Sites, A. Chernoff, M. B. Kirk, M. P. Marks, S. G. Robinson, Binary Translation, Comm. ACM, Vol. 3, No. 2, February 1993, are usually based on one of the two following approaches: recompiled sub-programs that should be capable of being directly called in the runnable program by non-portable program parts present as code for the source hardware. At the one time of the migrated program code, these routines then undertake all conversions, switchings, etc., needed for changing the program flow from the recompiled to emulated program parts (or, respectively, vice versa). What is thereby disadvantageous is the usually substantial programming outlay for the implementation of the "shelling" routines. the source hardware are deposited on the destination hardware in specifically protected memory areas, and the recompiled program parts are deposited in correspondingly unprotected areas. When a branch from a recompiled program part into a program part stored in a protected manner in the code of the source hardware then ensues at the run-time, then a signaling (usually a hardware interrupt) is triggered and, thus, an emulator is called that implements the corresponding program part that has been branched to. The emulation is then in turn ended as soon as an address jumps into the course of the program that lies in the unprotected memory area and, thus, belongs to the program code recompiled for the destination hardware.
For this method, the destination hardware (such as, for example, a DEC-alpha hardware) must offer the possibility of protecting specific memory areas such that the attempt to interpret a data sequence stored therein as command sequence and to implement it leads to a signaling. This, however, is far from being the case given every computer. What is thereby also disadvantageous is that the switch between direct implementation and emulation (usually an interrupt handling) that ensues g

REFERENCES:
patent: 5621651 (1997-04-01), Swoboda
patent: 5841670 (1998-11-01), Swoboda
patent: 6041374 (2000-03-01), Postman et al.
S. Goel et al., Supporting Method Migration in a Distributed Object Database System: A Performance Study, Proceedings of the 29th Hawaii Int'l Conference on System Sciences, vol. 1, 1996, pp. 31-40.
T. Afzal et al., Motorola PowerPC Migration Tools--Emulation and Translation, Technologies for the Information Superhighway, Compcon '96, Digest of Papers, 1996, pp. 145-150.
F. Gabbay et al., Smart: An Advanced Shared-Memory Simulator--Towards a System-Level Simulation Environment, Proceedings 5th Int'l Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems, 1997, MASCOTS '97, pp. 131-138.
Z. Jing et al., Migration of Legacy WSR-8D Algorithms and Product Generators to the Open System RPG, Proceedings of the IEEE 1997 National Aerospace and Electronics Conference, NAECON, 1997, vol. 1, pp. 318-324.
K. Chanchio et al., Memory Space Representation for Heterogeneous Network Process Migration, Proceedings of the 1st Merged Int'l Parallel Processing Symposium and Symposium on Parallel and Distributed Processing, IPPS/SPDP, 1998, pp. 801-805.
M. Eisenring et al., Rapid Prototyping of Dataflow Programs on Hardware/Software Architectures, Proceedings of the 31st Hawaii Int'l Conference on System Sciences, vol. 7, 1998, pp. 243-252.
M. Eisenring et al., Domain-Specific Interface Generation from Dataflow Specifications, Proceedings of the 6th Int'l workshop on Hardware/Software Codesign, (CODE/CASHE '98), 1998, pp. 43-47.
Andrews, Kristy et al, Migrating a CISC Computer Family onto RISC via Object Code Translation, ACM Sigplan Notices, vol. 27, No. 9, (1992), pp. 213-222.
Communications of the ACM, vol. 36, No. 2, (1993), Sites et al, "Binary Translation", pp. 69-81.

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Method for the migration of hardware-proximate, subprogram-indep does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Method for the migration of hardware-proximate, subprogram-indep, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for the migration of hardware-proximate, subprogram-indep will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-204436

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.