SPEC® CFP2006 Result

Copyright 2006-2017 Standard Performance Evaluation Corporation

Fujitsu

Fujitsu SPARC M12-2S

SPECfp®_rate2006 = 102  

CPU2006 license: 19 Test date: Mar-2017
Test sponsor: Fujitsu Hardware Availability: Apr-2017
Tested by: Fujitsu Software Availability: Jul-2017
Benchmark results graph
Hardware
CPU Name: SPARC64 XII
CPU Characteristics: High Speed Mode up to 4.35 GHz
CPU MHz: 4250
FPU: Integrated
CPU(s) enabled: 1 core, 1 chip, 1 core/chip, 8 threads/core
CPU(s) orderable: 1 to 16 BBs; each BB contains 1 or 2 CPU chips;
the number of orderable total cores is 2, 3, 4,
.. 384
Primary Cache: 64 KB I + 64 KB D on chip per core
Secondary Cache: 512 KB I+D on chip per core
L3 Cache: 32 MB I+D on chip per chip
Other Cache: None
Memory: 512 GB (16 x 32 GB 2Rx4 PC4-2400T-R)
Disk Subsystem: 1 x 600 GB 10K RPM SAS (for system disk)
Other Hardware: None
Software
Operating System: Oracle Solaris 11.3 (with June 2017 SRU)
Compiler: C/C++/Fortran: Version 12.6 of Oracle Developer
Studio
Auto Parallel: No
File System: tmpfs
System State: Default
Base Pointers: 32-bit
Peak Pointers: 32-bit
Other Software: None

Results Table

Benchmark Base Peak
Copies Seconds Ratio Seconds Ratio Seconds Ratio Copies Seconds Ratio Seconds Ratio Seconds Ratio
Results appear in the order in which they were run. Bold underlined text indicates a median measurement.
410.bwaves 8 793 137   791 137   791 137   2 126 216   126 216   126 216  
416.gamess 8 2999 52.2 3015 52.0 3005 52.1 4 1399 56.0 1405 55.7 1447 54.1
433.milc 8 361 203   361 203   361 204   8 317 232   318 231   317 232  
434.zeusmp 8 888 82.0 891 81.7 891 81.7 8 888 82.0 891 81.7 891 81.7
435.gromacs 8 875 65.3 875 65.3 873 65.4 8 816 70.0 815 70.1 815 70.1
436.cactusADM 8 1050 91.0 1044 91.6 1048 91.2 2 166 144   166 144   167 143  
437.leslie3d 8 838 89.8 833 90.3 843 89.2 8 838 89.8 833 90.3 843 89.2
444.namd 8 934 68.7 936 68.5 936 68.5 8 788 81.5 785 81.7 785 81.7
447.dealII 8 888 103   883 104   884 104   8 693 132   693 132   694 132  
450.soplex 8 826 80.8 827 80.7 827 80.6 8 826 80.8 827 80.7 827 80.6
453.povray 8 652 65.3 652 65.3 652 65.3 8 499 85.3 498 85.4 500 85.2
454.calculix 8 725 91.0 728 90.6 725 91.1 8 702 94.0 697 94.7 701 94.1
459.GemsFDTD 8 1128 75.2 1122 75.6 1121 75.7 8 862 98.5 863 98.4 864 98.2
465.tonto 8 1097 71.8 1096 71.8 1094 72.0 8 990 79.5 990 79.5 990 79.5
470.lbm 8 2014 54.6 2010 54.7 2014 54.6 4 382 144   380 145   382 144  
481.wrf 8 1081 82.7 1080 82.7 1080 82.8 8 964 92.7 965 92.6 965 92.6
482.sphinx3 8 1717 90.8 1719 90.7 1722 90.5 8 1669 93.4 1666 93.6 1668 93.5

Submit Notes

 Processes were assigned to specific processors using 'pbind' commands.
 The config file option 'submit' was used, along with a list of
 processors in the 'BIND' variable, to generate the pbind commands.
 (For details, please see the config file.)

Operating System Notes

 Shell Environments:
    ulimit -s 131072 was used to limit the space consumed by the stack
    (and therefore make more space available to the heap).

    The "Logical Domains Manager" service was turned off using the command "svcadm disable ldmd".

 System Tunables:
 (/etc/system parameters)
    autoup = 86400
       Causes pages older than the listed number of seconds to be written by fsflush.
    doiflush = 0
       Controls whether file system metadata syncs will be executed during fsflush invocations.
    dopageflush = 0
       Controls whether memory is examined for modified pages during fsflush invocations.
    zfs:zfs_arc_max=1073741824
       Determines the maximum size of the ZFS Adaptive Replacement Cache (ARC).

Platform Notes

Firmware Settings:
(XSCF operations)
   Set High Speed Mode via XSCF command "sethsmode -s on".

   Set 1 core per chip via "setcod -p 0 -s cpu 1".

 Sysinfo program /export/cpu2006/config/sysinfo
 Revision 6993 of 2015-11-06 (5bc7c140478f0d042f37127effc8c1a9)
 running on H2S-256-D0 Thu Mar  2 16:14:16 2017

 This section contains SUT (System Under Test) info as seen by
 some common utilities.  To remove or add to this section, see:
   http://www.spec.org/cpu2006/Docs/config.html#sysinfo

 From /usr/sbin/psrinfo

    SPARC64-XII (chipid 0, clock 4250 MHz)
    1 chips
    8 threads
    4250 MHz

 From kstat:        1 cores

 From prtconf: 521728 Megabytes

 /etc/release:
      Oracle Solaris 11.3 SPARC
 uname -a:
    SunOS H2S-256-D0 5.11 11.3 sun4v sparc sun4v

 SPEC is set to: /export/cpu2006

 disk: df -h  /export/cpu2006
 Filesystem             Size   Used  Available Capacity  Mounted on
 rpool/export           547G   5.6G       269G     3%    /export

 (End of data from sysinfo program)

General Notes

 The Building Block (BB) is just a Fujitsu SPARC M12-2S that is the basic
 unit to be expanded as if stacking up children's blocks.

 File System:
  tmpfs: output_root was used to put run directories in /tmp/cpu2006
  zfs: operating system

 SPEC CPU2006 benchmark:
  Updated with runspec --update

Base Compiler Invocation

C benchmarks:

 cc 

C++ benchmarks:

 CC 

Fortran benchmarks:

 f90 

Benchmarks using both Fortran and C:

 cc   f90 

Base Portability Flags

447.dealII:  -DBOOST_NO_COMPILER_CONFIG 

Base Optimization Flags

C benchmarks:

 -std=c99   -m32   -fast   -xtarget=sparc64xii   -xipo=2   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xalias_level=std   -xprefetch_level=2 

C++ benchmarks:

 -m32   -fast   -xtarget=sparc64xii   -xipo=2   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xalias_level=compatible   -library=stlport4 

Fortran benchmarks:

 -m32   -fast   -xtarget=sparc64xii   -xipo=2   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xvector=no%lib 

Benchmarks using both Fortran and C:

 -std=c99   -m32   -fast(cc)   -fast(f95)   -xtarget=sparc64xii   -xipo=2   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xalias_level=std   -xprefetch_level=2   -xvector=no%lib 

Base Other Flags

C benchmarks:

 -xjobs=8 

C++ benchmarks:

 -xjobs=8 

Fortran benchmarks:

 -xjobs=8 

Benchmarks using both Fortran and C:

 -xjobs=8 

Peak Compiler Invocation

C benchmarks:

 cc 

C++ benchmarks:

 CC 

Fortran benchmarks:

 f90 

Benchmarks using both Fortran and C:

 cc   f90 

Peak Portability Flags

447.dealII:  -DBOOST_NO_COMPILER_CONFIG 

Peak Optimization Flags

C benchmarks:

433.milc:  -std=c99   -m32   -fast   -xtarget=sparc64xii   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xipo=2   -xalias_level=std   -fsimple=1   -W2,-Ainline:rs=400   -Qoption cg -Qms_pipe+alldoall   -W2,-Asac   -xthroughput=no 
470.lbm:  -std=c99   -m32   -fast   -xtarget=sparc64xii   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xipo=2   -xalias_level=std   -xprefetch_level=2   -xpagesize=256M   -xsegment_align=256M   -xthroughput=no   -lbsdmalloc 
482.sphinx3:  -std=c99   -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -m32   -fast   -xtarget=sparc64xii   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xO4   -xipo=2   -xprefetch=latx:0.6   -xinline_param=level:1   -xprefetch=no%auto   -lbsdmalloc 

C++ benchmarks:

444.namd:  -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -m32   -fast   -xtarget=sparc64xii   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xalias_level=compatible   -xprefetch=no%auto   -Wc,-Qms_pipe+alldoall 
447.dealII:  -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -m32   -fast   -xtarget=sparc64xii   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xtarget=sparc64xplus   -xipo=1   -xalias_level=compatible   -xrestrict   -xprefetch=no%auto   -Qoption cg -Qiselect-funcalign=64   -xthroughput=yes   -library=stdcxx4   -template=extdef 
450.soplex:  basepeak = yes 
453.povray:  -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -m32   -fast   -xtarget=sparc64xii   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xO4   -xtarget=sparc64xplus   -xipo=2   -xalias_level=compatible   -xlinkopt=2   -xprefetch=no%auto   -xunroll=7   -Qoption iropt -Ainline:rs=1024   -Qoption iropt -Ainline:cs=1024   -Qoption iropt -Ainline:inc=900   -lfast 

Fortran benchmarks:

410.bwaves:  -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -m32   -fast   -xtarget=sparc64xii   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xipo=2   -xunroll=4   -xvector=%none   -xprefetch=no%auto 
416.gamess:  -m32   -fast   -xtarget=sparc64xii   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xvector=no%simd   -xprefetch=latx:0.1 
434.zeusmp:  basepeak = yes 
437.leslie3d:  basepeak = yes 
459.GemsFDTD:  -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -m32   -fast   -xtarget=sparc64xii   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xunroll=9   -xprefetch=latx:0.2   -xprefetch_level=3   -Qoption cg -Qlp-av=128   -Qoption iropt -Rujam 
465.tonto:  -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -m32   -fast   -xtarget=sparc64xii   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xipo=1   -xO4   -xunroll=3   -xprefetch=no%auto   -xthroughput=no   -lbsdmalloc 

Benchmarks using both Fortran and C:

435.gromacs:  -std=c99   -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -m32   -fast(cc)   -fast(f95)   -xtarget=sparc64xii   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xalias_level=strong   -Wc,-Qicache-chbab=1   -Wc,-Qiselect-rsqrta=2   -Wc,-Qiselect-rsqrta1x=2   -qoption cg -Qicache-chbab=1   -qoption cg -Qiselect-rsqrta=2   -qoption cg -Qiselect-rsqrta1x=2 
436.cactusADM:  -std=c99   -m32   -fast(cc)   -fast(f95)   -xtarget=sparc64xii   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xtarget=sparc64xplus   -xunroll=10   -xprefetch=latx:2.0   -xpagesize=256M   -xsegment_align=256M   -xthroughput=no   -lbsdmalloc 
454.calculix:  -std=c99   -xprofile=collect:./feedback(pass 1)   -xprofile=use:./feedback(pass 2)   -m32   -fast(cc)   -fast(f95)   -xtarget=sparc64xii   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xtarget=sparc64xplus   -xipo=1   -Wc,-Qiselect-funcalign=64   -xinline_param=level:3   -Qoption cg -Qiselect-funcalign=64 
481.wrf:  -std=c99   -m32   -fast(cc)   -fast(f95)   -xtarget=sparc64xii   -xpagesize=4M   -xsegment_align=4M   -xthroughput   -xunroll=9   -xprefetch=latx:0.4   -Qoption iropt -Rujam   -xO4   -xthroughput=no 

Peak Other Flags

C benchmarks:

 -xjobs=8 

C++ benchmarks:

 -xjobs=8 

Fortran benchmarks:

 -xjobs=8 

Benchmarks using both Fortran and C:

 -xjobs=8 

The flags files that were used to format this result can be browsed at
http://www.spec.org/cpu2006/flags/Oracle-Developer-Studio12.6.html,
http://www.spec.org/cpu2006/flags/Fujitsu-M12-2S.html.

You can also download the XML flags sources by saving the following links:
http://www.spec.org/cpu2006/flags/Oracle-Developer-Studio12.6.xml,
http://www.spec.org/cpu2006/flags/Fujitsu-M12-2S.xml.