SPEC Seal of Reviewal CINT2000 Result
Copyright © 1999-2004 Standard Performance Evaluation Corporation
Fujitsu Siemens Computers
PRIMEPOWER800/1000/2000 (788MHz)
SPECint2000 = 591    
SPECint_base2000 = 505    
SPEC license # 22 Tested by: Fujitsu Limited Test date: Jun-2002 Hardware Avail: Sep-2002 Software Avail: May-2002
Benchmark Reference
Time
Base
Runtime
Base
Ratio
Runtime Ratio Graph Scale
164.gzip 1400 309    453     271    517     164.gzip base result bar (453)
164.gzip peak result bar (517)
175.vpr 1400 318    441     278    504     175.vpr base result bar (441)
175.vpr peak result bar (504)
176.gcc 1100 282    390     210    523     176.gcc base result bar (390)
176.gcc peak result bar (523)
181.mcf 1800 346    520     301    597     181.mcf base result bar (520)
181.mcf peak result bar (597)
186.crafty 1000 200    501     153    654     186.crafty base result bar (501)
186.crafty peak result bar (654)
197.parser 1800 345    522     310    580     197.parser base result bar (522)
197.parser peak result bar (580)
252.eon 1300 210    618     189    687     252.eon base result bar (618)
252.eon peak result bar (687)
253.perlbmk 1800 316    569     262    688     253.perlbmk base result bar (569)
253.perlbmk peak result bar (688)
254.gap 1100 385    285     376    292     254.gap base result bar (285)
254.gap peak result bar (292)
255.vortex 1900 242    785     175    1087      255.vortex base result bar (785)
255.vortex peak result bar (1087)
256.bzip2 1500 281    534     259    579     256.bzip2 base result bar (534)
256.bzip2 peak result bar (579)
300.twolf 3000 487    616     444    676     300.twolf base result bar (616)
300.twolf peak result bar (676)
SPECint_base2000 505      
  SPECint2000 591      

Hardware
Hardware Vendor: Fujitsu Siemens Computers
Model Name: PRIMEPOWER800/1000/2000 (788MHz)
CPU: SPARC64 GP
CPU MHz: 788
FPU: Integrated
CPU(s) enabled: 1 core, 1 chip, 1 core/chip
CPU(s) orderable: 4 to 16/4 to 32/8 to 128
Parallel: None
Primary Cache: 128KBI+128KBD on chip
Secondary Cache: 8MB(I+D) off chip, per CPU
L3 Cache: None
Other Cache: None
Memory: 8192MB
Disk Subsystem: 1 x 18.2GB SCSI (10025rpm)
Other Hardware: Ethernet
Software
Operating System: Solaris 8 2/02 with patches 108434-07 and 108435-07.
Compiler: Fujitsu Parallelnavi 1.0.2 with patches 911403-01 and 911746-01.
Sun ONE Studio 7 with patches 111704-01, 111705-01, 111706-01,
111708-01, 111709-01, 111715-01 and 111716-01
Sun Performance Library 7
File System: ufs
System State: single user
Notes / Tuning Information
Baseline (except 252.eon, for Parallelnavi 1.0.2): -Kfast_GP=3,largepage
fdo_pre0=rm -rf `pwd`/*.fbk
PASS1=-Kpg
PASS2=-Kpu=$(EXEBASE).fbk
(252.eon, for Sun ONE Studio 7): -fast -xcrossfile -xtarget=ultra3
fdo_pre0=rm -rf `pwd`/../feedback.profile `pwd`/SunWS_cache
PASS1=-xprofile=collect:`pwd`/../feedback
PASS2=-xprofile=use:`pwd`/../feedback

Peak 
(for Sun ONE Studio 7)
 fdo_pre0=rm -rf `pwd`/../feedback.profile `pwd`/SunWS_cache
 PASS1=-xprofile=collect:`pwd`/../feedback
 PASS2=-xprofile=use:`pwd`/../feedback
164.gzip: -xO5 -xtarget=ultra3 -xalias_level=std -W2,-whole -xcrossfile 
    -W2,-Ainline -xprefetch -Wc,-Qgsched-trace_late=1,-Qgsched-spec_load=1 -ll2amm
175.vpr: -fast -xarch=v8plusb -xalias_level=std -xcrossfile -xsfpconst
    -xdepend -W2,-whole,-Mt600,-Mr4000 -Wc,-Qeps:enabled=1,-Qeps:do_spec_load=1,-Qeps:rp_filtering_margin=100
    -xregs=syst -lprism32 -lmopt -lm
176.gcc: -fast -xtarget=ultra3 -xcrossfile -W2,-whole -Wc,-Qgsched-trace_late=1,-Qgsched-T4 -ll2amm 
186.crafty: -fast -xtarget=ultra3 -xarch=v8plus -xF -xinline= -xcrossfile -Wc,-Qgsched-spec_load=1,-Qiselect-funcalign=64
    -xalias_level=strong -xregs=syst -W2,-Ashort_ldst,-Aivel:duplicate_loops -xprefetch=auto,latx:5.0
197.parser: -fast -xarch=v8plusb -xdepend -xprefetch=no%auto
    -xcrossfile -xregs=syst -Wc,-Qgsched-trace_late=1,-Qgsched-T4 
    -xalias_level=strong -Wc,-Qipa:valueprediction
    -W2,-Ashort_ldst,-Mt5000 -Wc,-Qiselect-funcalign=32 -lprism32
252.eon: -fast -xtarget=ultra3 -xcrossfile -xalias_level=compatible -xsafe=mem -Qoption iropt
    -Mt2000,-xrestrict -Qoption cg -Qgsched-trace_spec_load=1,-Qgsched-trace_late=1
253.perlbmk: -xO5 -xtarget=ultra3 -xarch=v8plusb -xcrossfile -xalias_level=strong -xsafe=mem 
    -Wc,-Qgsched-trace_late=1,-Qgsched-T4,-Qgsched-trace_spec_load=1 -Wc,-Qinline_memcpy=32 
    -Wc,-Qiselect-funcalign=32,-Qicache-chbab=1 -Wc,-Qiselect-sw_pf_tbl_th=20 -W2,-Adata_access -xprefetch=auto,latx:5.0 -ll2amm -dn
255.vortex: -fast -xtarget=ultra3 -xcrossfile -W2,-Aheap,-reroll=1,-Aunroll,-Ms1,-Mt600,-Mr13000,-crit
    -Wc,-Qdepgraph-early_cross_call=1 -Wc,-Qiselect-funcalign=32 -Wc,-Qpeep-Sh0
    -xrestrict -xdepend -W2,-Amemopt 
    -ll2amm -lprism32
256.bzip2: -fast -xtarget=ultra3 -W2,-whole,-crit
    -xcrossfile -xalias_level=strong -Wc,-Qiselect-funcalign=32
    -xdepend -xregs=syst -xsfpconst -Wc,-Qgsched-trace_spec_load=1 -xsafe=mem -ll2amm -lprism32

(for Parallelnavi 1.0.2):
fdo_pre0=rm -rf `pwd`/*.fbk
PASS1=-Kpg
PASS2=-Kpu=$(EXEBASE).fbk
181.mcf: -Kfast_GP=2,GREG,eval,preex,popt,unroll=2,prefetch=4,largepage,preload -x-
254.gap: -Kfast_GP=3,popt,eval,cfunc,largepage,xi=10
300.twolf: -Kfast_GP=5,eval,GREG,popt,cfunc,staticclump,use_rodata,xi=10,largepage,bcopy,nounroll,prefetch=4

Portability:
 176.gcc: -Dalloca=__builtin_alloca -DHOST_WORDS_BIG_ENDIAN
 186.crafty: -DSUN
 252.eon: -library=iostream
 253.perlbmk: -DSPEC_CPU2000_SOLARIS
 254.gap: -DSYS_IS_USG -DSYS_HAS_TIME_PROTO -DSYS_HAS_SIGNAL_PROTO -DSYS_HAS_CALLOC_PROTO

Note:
 System Tunables: (for /etc/system)
  consistent_coloring=1, tune_t_fsflushr=86400, autoup=86400,
  shmsys:shminfo_shmmax=1477846784, shmsys:shminfo_shmmni=1024, shmsys:shminfo_shmseg=1024,shminfo_shmmin=1
 (for /etc/opt/FJSVpnrm/lpg.conf)
  TSS=512M, SHMSEGSIZE=256M
 Shell Environments:
   LD_LIBRARY_PATH="/usr/lib:/opt/FD7/SUNWspro/lib/v8plusb:/opt/FD7/SUNWspro/prod/lib/v8plusb:/opt/FSUNf90/lib"
   LD_LIBRARY_PATH_64="/usr/lib/64:/opt/FD7/SUNWspro/lib/v9:/opt/FD7/SUNWspro/prod/lib/v9"
   PRISM_HEAP=268435456
   PRISM_MODE=2
 ONESTEP=yes was set for all baseline and peak benchmarks.
 Feedback directed optimization was used for all baseline and peak benchmarks.
 System board used with only one CPU present.


For questions about this result, please contact the tester.
For other inquiries, please contact webmaster@spec.org
Copyright © 1999-2004 Standard Performance Evaluation Corporation

First published at SPEC.org on 30-Jul-2002

Generated on Fri Mar 25 15:37:38 2005 by SPEC CPU2000 HTML formatter v1.01