R. B. Dewar, "Indirect threaded code," Communications ACM, vol. 18, pp. 330--331, June 1975.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....Perl and Xlisp inter preters, which behave more like general purpose C programs. These benchmarks can be found at http: www. complang. tuwien. ac. at anton interpreter arch) Gforth 0.4.9 19990617, a Forth system [5] it uses a virtual stack machine. We compiled it for indirect threaded code [3] i.e. there is one additional load instruction between the instruction load and the indirect branch (the direct threaded version would not compile for Simplescalar) The workloads we use are prim 2x, a compiler for a little language, and bench gc, a conservative garbage collector. Ocaml 2.04, ....

Robert B. K. Dewar. Indirect threaded code. Communications of the ACM, 18(6):330-331, June 1975. 405

....program. The early experiences also influenced the modifications that we performed when we generalized the generator tailored for Gforth s needs into a general VM interpreter generator. 6 Gforth requires a few additional twists because its basic implementation model is indirect threaded code [Dew75] the code that deals with that additional complication is small enough that we did not create automatic support for it. April 22, 2001 12 : 40 DRAFT 18 7 Performance In this section, we present basic performance numbers and evaluate the effect of the various optimizations. Many of the ....

Robert B.K. Dewar. Indirect threaded code. Communications of the ACM, 18(6):330--331, June 1975.

....version consists mainly of a 1246 line Forth program. The early experiences also influenced the modifications that we performed when we generalized the generator tailored for Gforth s h Gforth requires a few additional twists because its basic implementation model is indirect threaded code [Dew75] the code that deals with that additional complication is small enough that we did not create automatic support for it. 18 needs into a general VM interpreter generator. 7 Performance In this section, we present basic performance data and evaluate the effect of the various optimizations. ....

Robert B.K. Dewar. Indirect threaded code. Communications of the ACM, 18(6):330--331, June 1975.

....and the Perl and Xlisp interpreters, which behave more like general purpose C programs. These benchmarks can be found at http: www.complang.tuwien.ac.at anton interpreter arch) Gforth 0.4.9 19990617, a Forth system [5] it uses a virtual stack machine. We compiled it for indirect threaded code [3] i.e. there is one additional load instruction between the instruction load and the indirect branch (the direct threaded version would not compile for Simplescalar) The workloads we use are prims2x, a compiler for a little language, and bench gc, a conservative garbage collector. Ocaml 2.04, ....

Robert B.K. Dewar. Indirect threaded code. Communications of the ACM, 18(6):330--331, June 1975.

.... of Integration and Optimization of Interpreted and Compiled Languages A Proposal to the NSF in response to CCR Programs 97 (NPAC Technical Report SCCS 780) Geo rey C. Fox, Xiaoming Li , Yuhong Wen, Guansong Zhang Northeast Parallel Architectures Center at Syracuse University Syracuse, New York 13244 4100 315 443 2163, fgcf,lxm,wen,zgsg npac.syr.edu February 12, 1997 Visiting scholar from Harbin Institute of Technology, China 1 A. ....

....link interpreters with the libraries in our integrated system. Another critical aspect of our computational library is parallel computing support. A parallel library should be a collection of high performance mathematical subroutines used by application programmers to solve large problems [34] 38] From the II CVM compiler, we will get a partially ordered set of interpretive frames. While the parallelism among 8 di erent frames is seen by the scheduler, parallel library routines are aimed at exploiting parallelism within a frame. Another issue or opportunity brought up by the idea of ....

[Article contains additional citation context not shown here]

R. B. K. Dewar, Indirect threaded code, CACM, 18, 330-331 (1975).

....be more than won back by exposing more information to 211 type programs and highlevel PVM optimizers. 3.5 Interpreter Structure There are other methods known for implementing interpreters. Threaded code and indirect threaded code are two techniques which are used in interpreter implementation [2, 5]. While the details of these two methods are beyond the scope of this paper, suffice it to say that both eliminate the fetchexecute loop of classical interpreters. Neither are likely to make any impact on the current PVM, though. It has been shown that the speed differences between the methods ....

R. B. K. Dewar. Indirect threaded code. Communications of the ACM, 18(6):330--331, 1975.

.... of Integration and Optimization of Interpreted and Compiled Languages A Proposal to the NSF in response to CCR Programs 97 (NPAC Technical Report SCCS 780) Geoffrey C. Fox, Xiaoming Li , Yuhong Wen, Guansong Zhang Northeast Parallel Architectures Center at Syracuse University Syracuse, New York 13244 4100 315 443 2163, fgcf,lxm,wen,zgsg npac.syr.edu February 12, 1997 Visiting scholar from Harbin Institute of Technology, China A. Project ....

....link interpreters with the libraries in our integrated system. Another critical aspect of our computational library is parallel computing support. A parallel library should be a collection of high performance mathematical subroutines used by application programmers to solve large problems [34] 38] From the II CVM compiler, we will get a partially ordered set of interpretive frames. While the parallelism among different frames is seen by the scheduler, parallel library routines are aimed at exploiting parallelism within a frame. Another issue or opportunity brought up by the idea of ....

[Article contains additional citation context not shown here]

R. B. K. Dewar, Indirect threaded code, CACM, 18, 330-331 (1975).

....code) slowest methods: combinator reduction. We translated from Alfonzo to combinators via the calculus, a trivial transformation for our subset of Scheme, then from calculus to combinators essentially by Turner s algorithm [24] The combinator code was interpreted by an indirectly threaded [12] graph reduction machine written in standard C. In our implementation of this technique each object was tagged with a pointer to a descriptor holding the addresses of functions providing the output functions, garbage collection support [23] and reduction rules [18] for its class. The interpreter ....

Robert B. K. Dewar. Indirect threaded code. Communications of the ACM, 18(6):330--1, June 1975.

....and 2.6 show MIPS assembly code for the three techniques (direct call threading needed a little source code twisting to get reasonable scheduling) Fig. 2. 7 shows the overhead of these techniques in cycles on two processors, 1 Forth has been traditionally implemented using indirect threading [Dew75, Kog82] which performs an additional indirection. i.e. an instruction is represented by an address of a cell that contains the address of the routine. This is used for eliminating inline arguments; e.g. instead of having a call instruction with an inline argument, the instruction points to a ....

Robert B.K. Dewar. Indirect threaded code. Communications of the ACM, 18(6):330--331, June 1975.

....only such ephemera as the macros, backtracking, the data representation sublanguage, support for separate compilation, and a few minor syntactic luxuries. The s in the name can also stand for small , because Setl s was a very compact system based on Dewar s celebrated indirect threaded code [55] technique, and was written in MINIMAL, the portable assembly language in which the run time system of MACRO SPITBOL [54] was implemented. Jay VandeKopple dropped the hyphen from Setl s and worked on what was then called SETLS in the 1990s [202] while at NYU on sabbatical from Marymount College. ....

Robert B.K. Dewar. Indirect threaded code. Communications of the ACM, 18(6):330--331, June 1975.

No context found.

R. B. Dewar, "Indirect threaded code," Communications ACM, vol. 18, pp. 330--331, June 1975.

No context found.

R. B. K. Dewar, Indirect threaded code, CACM, 18, 330-331 (1975).

No context found.

Dewar, R.B.K., "Indirect Threaded Code," Communications of the ACM, Vol. 18, No. 6, (June 1975), pp. 330-331.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC