Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!swrinde!elroy.jpl.nasa.gov!ncar!gatech!prism!msd!rhoward From: rhoward@msd.gatech.edu (Robert L. Howard) Newsgroups: comp.lang.fortran Subject: Re: why does optimization take so long ? Message-ID: Date: 4 Jun 91 21:46:46 GMT References: <1236@nikhefh.nikhef.nl> <1991May22.073127.4540@monu0.cc.monash.edu.au> <1991May28.131224.12410@lpi.liant.com> Sender: news@prism.gatech.EDU Organization: Georgia Tech Research Institute Lines: 38 >In article <1236@nikhefh.nikhef.nl: > t19@nikhefh.nikhef.nl (Geert J v Oldenborgh) writes: >: I am curious whether any of you know why compiling certain routines >: with a -O flag takes a semi-infinite amount of time on certain >: computers (Apollo DN1000, Sun Fortran 1.3.1) >: The code in question is generated by a computer algebra program, Alan Miller responds: >Geert sent me the piece of code... >I compiled it with the Sun f77 compiler (v 1.3.1).... > > Compiler + options Time > f77 -c 16.1u + 1.8s > f77 -O -c Gave up after 37 mins. of CPU time ! Here are the times for Sun f77 1.4. I went through a number of optimization levels... f77 test.f Time=0:22.89 (15.880u, 2.400s) size=596k/1816k avg/max f77 -O1 test.f Time=0:26.06 (20.930u, 2.650s) size=938k/3492k avg/max f77 -O2 test.f Time=7:46.69 (448.790u, 4.880s) size=1595k/3760k avg/max f77 -O3 test.f Time=8:52.24 (480.800u, 6.200s) size=1449k/6244k avg/max ^^ same as -O It takes a long time but it does finish. Robert -- | Robert L. Howard | Georgia Tech Research Institute | | rhoward@msd.gatech.edu | MATD Laboratory | | (404) 528-7165 | Atlanta, Georgia 30332 | ------------------------------------------------------------------------- | "Reality is showing us that perhaps we should do with nuclear | | power the same thing Keloggs is advocating for Corn Flakes - | | Discover it again for the first time." -- John De Armond |