# Invocation command line:
# /export/bmk2/johnh/cpu2006/bin/runspec -c oct16a-speed --reportable fp --nopreenv --note-preenv
# output_root was not used for this run
############################################################################
ext       = oct16a-speed

flagsurl000         = http://www.spec.org/cpu2006/flags/Oracle-Solaris-Studio12.2-SPARC.xml
tune                = all
iterations          = 1        # in a real run, would be 3 or more
expand_notes        = 1
ignore_errors       = 1
log_line_width      = 4090     # to avoid choking vi
output_format       = asc,cfgfile,csv,html
mean_anyway         = 1
#minimize_rundirs   = 1
#minimize_builddirs = 1
preenv              = 1

#######################  Processor usage #######################
# Note that the first instance of a "named section" will
# probably occur with the "include" just below.  Therefore,
# any variables that belong in the header section should be above
# this point.

makeflags           = -j32   # how many jobs specmake should spawn
XJOBS               = 32     # how many CPUs ipo can use (at link time)

use_submit_for_speed = 1  # needed to bind the primary thread
command_add_redirect = 1  # useful with submit
submit = echo 'pbind -b $BIND \$\$ >> pbind.out; $command' > dobmk; sh dobmk

#include: $[top]/config/BIND-speed.inc
#  ----- Begin inclusion of '/export/bmk2/johnh/cpu2006/config/BIND-speed.inc'
############################################################################

%define allthreads          128
%define allcores            64
%define onechipcores         4
%define twochipcores         8
%define threechipcores      12

preENV_OMP_NUM_THREADS  = %{allcores}
preENV_SUNW_MP_THR_IDLE = SPIN

%ifdef %{scatter}
preENV_SUNW_MP_PROCBIND = 127  95  63  31 119  87  55  23 111  79  47  15 103  71  39   7 125  93  61  29 117  85  53  21 109  77  45  13 101  69  37   5 123  91  59  27 115  83  51  19 107  75  43  11  99  67  35   3 121  89  57  25 113  81  49  17 105  73  41   9  97  65  33   1 126  94  62  30 118  86  54  22 110  78  46  14 102  70  38   6 124  92  60  28 116  84  52  20 108  76  44  12 100  68  36   4 122  90  58  26 114  82  50  18 106  74  42  10  98  66  34   2 120  88  56  24 112  80  48  16 104  72  40   8  96  64  32   0
%else
preENV_SUNW_MP_PROCBIND = 127 125 123 121 119 117 115 113 111 109 107 105 103 101 99 97 95 93 91 89 87 85 83 81 79 77 75 73 71 69 67 65 63 61 59 57 55 53 51 49 47 45 43 41 39 37 35 33 31 29 27 25 23 21 19 17 15 13 11 9 7 5 3 1 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100 102 104 106 108 110 112 114 116 118 120 122 124 126
%endif

# Use this thread for main thread, needs to be on same chip as 1st thread in PROCBIND
bind = 124



# ---- End inclusion of '/export/bmk2/johnh/cpu2006/config/BIND-speed.inc'

#################### Compiler invocation ####################################

default=default=default=default:

%define compiler_showversion  1
%define compiler_verbose      1

%ifdef %{compiler_showversion}
    CVERBOSE   = -V
    CXXVERBOSE = -verbose=version
    FCVERBOSE  = -V
%endif
%ifdef %{compiler_verbose}
    CVERBOSE   = -V -\#
    CXXVERBOSE = -verbose=diags,version
    FCVERBOSE  = -V -v
%endif

SPECLANG = /export/solstudio12.2/bin/
CC  = $(SPECLANG)cc  -xjobs=$(XJOBS) $(CVERBOSE)
CXX = $(SPECLANG)CC  -xjobs=$(XJOBS) $(CXXVERBOSE) -xdepend
FC  = $(SPECLANG)f90 -xjobs=$(XJOBS) $(FCVERBOSE)

PARALLEL    = -xautopar -xreduction

#################################################################
#### Allow Choice of STLport4 or Apache Standard C++ Library ####

default:
APACHE_INCL     = /export/bmk2/apache/stdcxx-4.2.1/include
APACHE_BLD_INCL = /export/bmk2/apache/stdcxx-4.2.1/build/include
APACHE_LIBDIR   = /export/bmk2/apache/stdcxx-4.2.1/build/lib
APACHE_LIB      = std8d

APACHE_EXTRA_CXXFLAGS = -library=no%Cstd -I$(APACHE_INCL) -I$(APACHE_BLD_INCL)
APACHE_EXTRA_CXXLIBS  = -library=no%Cstd -L$(APACHE_LIBDIR) -R$(APACHE_LIBDIR) -l$(APACHE_LIB)

default:
EXTRA_CXXFLAGS  = -library=stlport4
EXTRA_CXXLIBS   = -library=stlport4

# fp base - enable next 3 lines for stdcxx in base, or comment out to use stlport4
fp=base:
EXTRA_CXXFLAGS  = $(APACHE_EXTRA_CXXFLAGS)
EXTRA_CXXLIBS   = $(APACHE_EXTRA_CXXLIBS)

# control deal peak here; deal base above with rest of fp base
447.dealII=peak:
EXTRA_CXXFLAGS  = $(APACHE_EXTRA_CXXFLAGS)
EXTRA_CXXLIBS   = $(APACHE_EXTRA_CXXLIBS)

# Apply the alternate source unconditionally.  It is harmless (and legal) whether apache used or not.
447.dealII:
srcalt          = apache_stdcxx_4_2_1


#################### Base Int #########################
int=base:
OPTIMIZE        = -fast -fma=fused -xipo=2 -xpagesize=4M -xprefetch_level=2
COPTIMIZE       = -xalias_level=std $(PARALLEL)
CXXOPTIMIZE     = -xalias_level=compatible
OPTIMIZATION_CXXLIBS = -lfast

EXTRA_CLIBS     = -ll2amm

#################### Peak Int #########################
int=peak:
OPTIMIZE        = -fast -fma=fused -xpagesize=4M
CXXOPTIMIZE     = -xalias_level=compatible

400.perlbench=peak:
EXTRA_OPTIMIZE  = -xipo=2 -xalias_level=std -xrestrict
EXTRA_LIBS      = -lfast

401.bzip2=peak:
EXTRA_OPTIMIZE  = -xalias_level=strong

403.gcc=peak:
EXTRA_OPTIMIZE  = -xipo=2 -xalias_level=std -xprefetch_level=2
EXTRA_LIBS      = -ll2amm
feedback        = 0

429.mcf=peak:
OPTIMIZE        = -fast -fma=fused -xipo=2 -xpagesize=4M -xprefetch_level=2 -xalias_level=std
EXTRA_OPTIMIZE  = $(PARALLEL) -W2,-Aparallel:nthreads=%{threechipcores}
EXTRA_LIBS      = -ll2amm
feedback        = 0

445.gobmk=peak:
EXTRA_OPTIMIZE  = -xalias_level=std -xrestrict

456.hmmer=peak:
OPTIMIZE  = -fast -fma=fused -xpagesize=4M -xipo=2

458.sjeng=peak:
OPTIMIZE  = -fast -fma=fused -xpagesize=4M -xipo=2

462.libquantum=peak:
OPTIMIZE        = -fast -fma=fused -xipo=2 -xpagesize=4M -xprefetch_level=3 -xalias_level=std $(PARALLEL) -xprefetch_auto_type=indirect_array_access
EXTRA_LIBS      = -ll2amm
feedback        = 0

464.h264ref=peak:
EXTRA_OPTIMIZE  = -xipo=2 -xalias_level=std -xarch=sparcvis2 -xprefetch=no
EXTRA_LIBS      = -ll2amm

471.omnetpp=peak:
EXTRA_OPTIMIZE  = -xipo=2 -xprefetch_level=2 -Qoption cg -Qlp-av=0
EXTRA_LIBS      = -lfast

473.astar=peak:
EXTRA_OPTIMIZE  = -xipo=2 -xprefetch=no%auto
EXTRA_LIBS      = -lfast

483.xalancbmk=peak:
EXTRA_OPTIMIZE  = -xipo=2 -xprefetch=no
EXTRA_LIBS      = -lfast

#################### Base FP #########################

fp=base:
OPTIMIZE    = -fast -fma=fused -xpagesize=4M -xipo=2 -xprefetch=latx:2
COPTIMIZE   = -xprefetch_level=3 -xprefetch_auto_type=indirect_array_access $(PARALLEL)
CXXOPTIMIZE = -xprefetch_level=2 -xalias_level=compatible
FOPTIMIZE   = -xprefetch_level=2 $(PARALLEL)

#################### Peak FP ############################

fp=peak:
#OPTIMIZE    = -fast -fma=fused -xpagesize=4M -xipo=2 -xprefetch=latx:2
#COPTIMIZE   = -xalias_level=std -xprefetch_level=3 -xprefetch_auto_type=indirect_array_access


410.bwaves=peak:
OPTIMIZE       = -fast -fma=fused -xipo=2
FOPTIMIZE      = $(PARALLEL) -Qoption iropt -Aparallel:nthreads=%{allthreads}

416.gamess=peak:
OPTIMIZE       = -fast -fma=fused -xpagesize=4M -xipo=2 -xprefetch_level=3

433.milc=peak:
OPTIMIZE       = -fast -fma=fused -xpagesize=4M -xipo=2
COPTIMIZE      = -xprefetch_level=2 -W2,-Ainline:rs=400 -xalias_level=std
EXTRA_OPTIMIZE = $(PARALLEL) -W2,-Aparallel:nthreads=%{twochipcores}
feedback       = 0

434.zeusmp=peak:
OPTIMIZE = -fast -fma=fused -xpagesize=4M -xipo=2 $(PARALLEL)

435.gromacs=peak:
OPTIMIZE       = -fast -fma=fused -xipo=2 -xpagesize=4M
FOPTIMIZE      = $(PARALLEL)  -Qoption iropt -Aparallel:nthreads=%{onechipcores}

436.cactusADM=peak:
OPTIMIZE = -fast -fma=fused -xpagesize=4M -xipo=2 $(PARALLEL)

437.leslie3d=peak:
OPTIMIZE       = -fast -fma=fused -xpagesize=4M -xipo=2 $(PARALLEL)

444.namd=peak:
OPTIMIZE       = -fast -fma=fused -xpagesize=4M
CXXOPTIMIZE    = -xalias_level=any
feedback       = 0

447.dealII=peak:
OPTIMIZE       = -fast -fma=fused -xpagesize=4M -xipo=2 -xprefetch_level=2
CXXOPTIMIZE    = -xrestrict -xalias_level=compatible #-qoption iropt -Ainline:inc=300

450.soplex=peak:
OPTIMIZE       = -fast -xpagesize=4M -xipo=2 -xprefetch_level=2 -xalias_level=compatible -xrestrict
EXTRA_OPTIMIZE =  -Qoption cg -Qlp-ol=1-it=3-imb=1 -Qoption iropt -Apf:pdl=3
CXXOPTIMIZE    = -xprefetch_auto_type=indirect_array_access

453.povray=peak:
OPTIMIZE       = -fast -fma=fused -xpagesize=4M -xipo=2
CXXOPTIMIZE    = -xrestrict -xalias_level=compatible #-qoption iropt -Ainline:inc=300

454.calculix=peak:
OPTIMIZE       = -fast -fma=fused -xpagesize=4M -xipo=2 -xprefetch_level=3 -xprefetch=latx:3.0
FOPTIMIZE      = $(PARALLEL)  -Qoption iropt -Aparallel:nthreads=%{onechipcores}
COPTIMIZE      = -xalias_level=std

459.GemsFDTD=peak:
OPTIMIZE    = -fast -fma=fused -xpagesize=4M -xipo=2 # -xprefetch=latx:2
FOPTIMIZE   = -xprefetch_level=2 $(PARALLEL)

465.tonto=peak:
OPTIMIZE       = -fast -xpagesize=4M -xipo=2 -fma=fused
FOPTIMIZE      = $(PARALLEL)  -Qoption iropt -Aparallel:nthreads=%{onechipcores}
EXTRA_LIBS     = -lumem

470.lbm=peak:
OPTIMIZE = -fast -fma=fused -xpagesize=4M -xipo=2 $(PARALLEL) -W2,-Aparallel:nthreads=%{allthreads}
COPTIMIZE   = -xalias_level=std -xprefetch_level=3 -xprefetch_auto_type=indirect_array_access

481.wrf=peak:
OPTIMIZE       = -fast -fma=fused -xpagesize=4M -xipo=2 -xprefetch_level=3 -xunroll=8
FOPTIMIZE      = $(PARALLEL)  -Qoption iropt -Aparallel:nthreads=%{threechipcores}
feedback=0

482.sphinx3=peak:
OPTIMIZE       =  -fast -xipo=2 -fma=fused $(PARALLEL) -xalias_level=std


###################### Portability #################################

400.perlbench:
CPORTABILITY    = -DSPEC_CPU_SOLARIS_SPARC

403.gcc:
CPORTABILITY    = -DSPEC_CPU_SOLARIS

462.libquantum:
CPORTABILITY    = -DSPEC_CPU_SOLARIS

483.xalancbmk:
CXXPORTABILITY  = -DSPEC_CPU_SOLARIS

###################### Profile Feedback #################################
#
# By default, turns on feedback for everything
#
default=default=default=default:

fdo_pre0    =    rm -rf ./feedback.profile ./SunWS_cache

PASS1_CFLAGS   =         -xprofile=collect:./feedback
PASS1_CXXFLAGS =         -xprofile=collect:./feedback
PASS1_FFLAGS   =         -xprofile=collect:./feedback
PASS1_LDFLAGS  =         -xprofile=collect:./feedback

PASS2_CFLAGS   =         -xprofile=use:./feedback
PASS2_CXXFLAGS =         -xprofile=use:./feedback
PASS2_FFLAGS   =         -xprofile=use:./feedback
PASS2_LDFLAGS  =         -xprofile=use:./feedback

default=default=default=default:

sw_compiler    = Oracle Solaris Studio 12.2

notes_comp_000 = Oracle Solaris Studio 12.2 is distributed with mandatory OS patches
notes_comp_005 =  118683-05  119963-20  120753-08
notes_comp_010 = Oracle Solaris Studio 12.2 and patches are available at
notes_comp_015 =  http://oracle.com/goto/solarisstudio
notes_comp_020 =
notes_comp_025 =The Apache C++ Standard Library V4.2.1 was installed from
notes_comp_030 =   http://stdcxx.apache.org/download.html using:
notes_comp_035 =     alias gmake=specmake
notes_comp_040 =     gmake BUILDTYPE=8d CONFIG=sunpro.config

notes_submit_000 =
notes_submit_005 = Processes were assigned to specific processors using 'pbind'
notes_submit_010 = commands.  The config file option 'submit' was used, along
notes_submit_015 = with a list of processors in the 'BIND' variable, to generate
notes_submit_020 = the pbind commands.  (For details, please see the config file.)
notes_submit_025 =


default=default=default=default:
#include: $[top]/config/SUT.inc
#  ----- Begin inclusion of '/export/bmk2/johnh/cpu2006/config/SUT.inc'
############################################################################
default=default=default=default:

hw_avail           = Dec-2010
hw_cpu_mhz         = 3000
hw_cpu_name        = SPARC64 VII+
hw_disk000         = 698 GB mirrored partition on
hw_disk001         = 12 x 146 GB 15K RPM SAS disks
hw_disk002         = in each of 2 StorageTek 2530 Array
hw_disk003         = (24 total disk, 12 in each array)
hw_fpu             = Integrated
hw_memory          = 512 GB (128 x 4 GB, 8-way interleaved)
hw_model           = SPARC Enterprise M8000
hw_nchips          = 16
hw_ncores          = 64
hw_ncoresperchip   = 4
hw_ncpuorder       = 1 to 4 CMUs; each CMU contains 2 or 4 CPU chips
hw_nthreadspercore = 2
hw_ocache          = None
hw_other           = None
hw_pcache          = 64 KB I + 64 KB D on chip per core
hw_scache          = 12 MB I+D on chip per chip
hw_tcache          = None
hw_vendor          = Fujitsu
tester             = Oracle Corporation
test_sponsor       = Fujitsu
license_num        = 19
sw_avail           = Sep-2010
sw_base_ptrsize    = 32-bit
sw_file            = ufs
sw_os              = Oracle Solaris 10 9/10
sw_other           = None
sw_peak_ptrsize    = 32-bit
sw_state           = Default

notes_plat_000     =
notes_plat_005     = Memory is 8-way interleaved by filling all slots with
notes_plat_010     = the same capacity DIMMs.
notes_plat_015     =
notes_plat_020 = This result is measured on a SPARC Enterprise M8000 server from
notes_plat_025 = Oracle.  The SPARC Enterprise M8000 server from Oracle and from
notes_plat_030 = Fujitsu are electrically equivalent.

notes_os_shell_000 = ulimit -s 131072 was used to limit the space
notes_os_shell_005 = consumed by the stack (and therefore make more
notes_os_shell_010 = space available to the heap).
notes_os_sys_000 =
notes_os_sys_005 =  /etc/system parameters
notes_os_sys_010 =     autoup=600
notes_os_sys_015 =        Causes pages older than the listed number of seconds to
notes_os_sys_020 =        be written by fsflush.
notes_os_sys_025 =     tune_t_fsflushr=10
notes_os_sys_030 =        Controls how many seconds elapse between runs of the
notes_os_sys_035 =        page flush daemon, fsflush.
notes_os_sys_040 =     lpg_alloc_prefer=1
notes_os_sys_045 =         Indicates that extra effort should be taken to ensure
notes_os_sys_050 =         that pages are created in the nearby lgroup (NUMA location).
notes_os_sys_055 =  The "webconsole" service was turned off using
notes_os_sys_060 =      svcadm disable webconsole
notes_os_sys_065 =  The system had 75 GB of swap space.
# ---- End inclusion of '/export/bmk2/johnh/cpu2006/config/SUT.inc'




# The following section was added automatically, and contains settings that
# did not appear in the original configuration file, but were added to the
# raw file after the run.

459.GemsFDTD:
# The following setting was inserted automatically as a result of
# post-run basepeak application.
basepeak = 1


410.bwaves:
# The following setting was inserted automatically as a result of
# post-run basepeak application.
basepeak = 1

default:
notes_000 =Environment variables set by runspec before the start of the run:
notes_005 =OMP_NUM_THREADS = "64"
notes_010 =SUNW_MP_PROCBIND = "127 125 123 121 119 117 115 113 111 109 107 105 103
notes_015 =  101 99 97 95 93 91 89 87 85 83 81 79 77 75 73 71 69 67 65 63 61 59 57 55
notes_020 =  53 51 49 47 45 43 41 39 37 35 33 31 29 27 25 23 21 19 17 15 13 11 9 7 5
notes_025 =  3 1 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46
notes_030 =  48 50 52 54 56 58 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94
notes_035 =  96 98 100 102 104 106 108 110 112 114 116 118 120 122 124 126"
notes_040 =SUNW_MP_THR_IDLE = "SPIN"
notes_045 =
notes_050 =447.dealII (peak): "apache_stdcxx_4_2_1" src.alt was used.
notes_055 =447.dealII (base): "apache_stdcxx_4_2_1" src.alt was used.
notes_060 =