Andrew W. Appel and Trevor Jim. Making lambda calculus smaller, faster. Technical Report CS-TR-477-94, Princeton University, November 1994. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Compiling Standard ML to Java - Gammage (1997) (Correct)
....to properly annotate the ML i Rep operation (as for the = operator) 3.3. 3 Optimization In order to prevent unnecessary effort in subsequent stages of the compiler, a series of simple optimizations on the ML i Rep form are performed, roughly following those suggested by Appel and Jim [6, 7, 4]. In the discussion that follows, an expression e is defined to be small if and only if e is a variable, or an integer, real or enumeration constant. An expression e is defined to be safe if its execution cannot cause side effects (such as updating an array, or input output) or cause an exception ....
Andrew W. Appel and Trevor Jim. Making lambda calculus smaller, faster. Technical Report CS-TR-477-94, Princeton University, November 1994.
Compilation by Transformation in Non-Strict Functional Languages - Santos (1995) (12 citations) (Correct)
....and will actually create an overhead (extra closure) We have indeed seen cases where this happens. ffl Reduce the number of iterations needed for the simplifier to reach a fixed point. This can probably be done using a more systematic approach, as we currently do it in a very ad hoc manner. AJ94] describes the approach used to minimise the number of iterations in a similar pass of the SML NJ Compiler, as well as a linear time algorithm to perform it. The algorithm tries to keep track of the usage counts of variables (the occurrence information) during the simplification process, ....
Andrew W. Appel and Trevor Jim. Making lambda calculus smaller, faster. Technical report CS-TR-477-94, November 1994. to appear in the Journal of Functional Programming.
Shrinking Lambda Expressions in Linear Time - Appel, Jim (1997) (17 citations) Self-citation (Appel Jim) (Correct)
....final output will be identical. Therefore in comparing the algorithms, we only have to compare running times, and not the programs produced by the algorithms. We have also proved confluence for shrinking reductions on ordinary lambda calculus where function arguments can be terms, not just atoms [3]. 4 A naive Contract algorithm The Contract phase of our compiler does just the shrinking reductions: deadvariable elimination, record field selection, and inlining of functions called only once. Because we compile ML, our optimizer can assume that programs are well typed, so that no program ....
Andrew W. Appel and Trevor Jim. Making lambda-calculus smaller, faster. Technical Report CS-TR-477-94, Princeton University, November 1994.
Compiling Standard ML to Java - An (Correct)
No context found.
Andrew W. Appel and Trevor Jim. Making lambda calculus smaller, faster. Technical Report CS-TR-477-94, Princeton University, November 1994.
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