Xavier Leroy, personal communication. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Optimizing Direct Threaded Code By Selective Inlining - Piumarta, Riccardi (1998) (20 citations) (Correct)
....of dynamically generated macro opcodes. Figure 12 shows the performance of the benchmarks after inlining, relative to the original performance without inlining. It is important to note that the Objective Caml bytecode set has already been optimized statically, as described in Section 2. 4 [Ler98]. Any further improvements are therefore due mainly to the elimination of dispatch overhead in common sequences that are particular to each application. Virtual machines whose bytecode sets have not been statically optimized in this way would benefit more from our technique. We can see from the ....
Xavier Leroy, personal communication.
Standard ML of New Jersey - Appel, MacQueen (1991) (152 citations) (Correct)
....SML NJ 0.65 x 8.5 0.3 1.4 0.1 LeLisp 15.23 4.1 1.4 SunOS 3.5, cc O 4.35 gcc 1.37.1, gcc O 4.22 Ultrix 4.0, cc O2 0. 90 Figure 2: Comparison of several di#erent compilers Xavier Leroy translated Gerard Huet s Knuth Bendix program into several di#erent languages, and ran them on two di#erent machines[21]. This table shows non gc run time and gc time in seconds for each version of the program. Since the program uses higher order functions, Leroy had to do manual lambda lifting to write the program in Lisp and C, and in some places had to use explicit closures (structures containing ....
Xavier Leroy. INRIA, personal communication, 1991.
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