petri (Stefan Petri) (10/08/89)
Here are my dhrystone-results for our Targon/35-M50 (cloned pyramid 8910?), 10MHz, 16MB, double cpu ; not single-user-mode, but only one user. 3 > cc -V $Revision: 1.2 $ $Date: 88/04/19 08:14:51 $ 4 > cc -O dry.c -o drynr 5 > cc -O -DREG=register dry.c -o dryr 6 > cc -OG -v -DREG=register dry.c -o dryrOG optimization statistics number expressions reduced = 1 number redundant statements removed = 0 number redundant expressions removed = 3 number statements moved outside loop = 0 number expressions moved outside loop= 0 number long statements optimized = 0 SymTab - entries : 290 vars and params : 55 size of sets : 2 max size of sets : 4 7 > ucb cc -OG -DREG=register dry.c -o dryrOGucb # that's a different libc.a 8 > drynr Dhrystone(1.1) time for 500000 passes = 61 This machine benchmarks at 8178 dhrystones/second 9 > dryr Dhrystone(1.1) time for 500000 passes = 61 This machine benchmarks at 8181 dhrystones/second 10 > dryrOG Dhrystone(1.1) time for 500000 passes = 60 This machine benchmarks at 8207 dhrystones/second 11 > dryrOGucb Dhrystone(1.1) time for 500000 passes = 61 This machine benchmarks at 8194 dhrystones/second ------------- Some results from a Sun3/60 (68020 , 8MB , ? MHz), with cc and gcc 4 > cc -O4 -DREG=register dry.c -o dryrO4 5 > dryrO4 Dhrystone(1.1) time for 50000 passes = 10 This machine benchmarks at 4838 dhrystones/second 6 > gcc -O -DREG=register dry.c -o dryrOgcc 7 > dryrOgcc Dhrystone(1.1) time for 50000 passes = 10 This machine benchmarks at 4958 dhrystones/second S.P. -- Stefan Petri <petri@tubsibr.UUCP> Technische Universitaet Braunschweig, Institut fuer Betriebssysteme und Rechnerverbund, 3300 Braunschweig, W. Germany.