Basic HTML version of Foils prepared October 16 98

Foil 37 JGF Proposed FP Execution Modes

From Integration of Scientific and Technical Computing with Enterprise Information Systems RCI North American Annual Member Executive Conference October 13-15 98 -- October 15 98. by Geoffrey C. Fox
1 Language Chip Architecture
2 Modifier x86 PowerPC SPARC
3
4 strictfp Java 1.0 (no double Java 1.0 Java 1.0
5 rounding on underflow) (no fused mac)
6 default i.e. larger exponent range fused mac can Java 1.0
7 no modifier allowed be used
8 associativefp many optimizations allowed on all platforms
9 Current default is essentially strictfp although not always implemented precisely
10 mac is fused hardware multiply-add (a*b+c)
11 strictfp is a little more than 2 times slower than default on INTEL x86 for a vector dot product
in Table To:

© Northeast Parallel Architectures Center, Syracuse University, npac@npac.syr.edu

If you have any comments about this server, send e-mail to webmaster@npac.syr.edu.

Page produced by wwwfoil on Fri Oct 16 1998