# Invocation command line: # /export/bmk2/johnh/cpu2006/bin/runspec -c nov18b-rate --define platform=generic --nobuild --reportable --rate fp --nopreenv --note-preenv # output_root was not used for this run ############################################################################ ### Platform ############################################################## # # Search for 'platform' to see how this is used and possible settings. # To build one set of binaries that will be pretty good on all SPARC64-VII+, # you could set this to 'generic'. # # NOTICE that a key difference among platforms is use of autopar. So, # you need to observe what this variable does for autopar, and you need # to set the number of copies accordingly. # %if defined(%{platform}) % define suffix rate-%{platform} %else % error "please use --define platform" %endif ########################################################################## ext = nov18b-%{suffix} 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-rate.inc # ----- Begin inclusion of '/export/bmk2/johnh/cpu2006/config/BIND-rate.inc' ############################################################################ # m8000 16 chip rate parallel_setup = 16 parallel_setup_prefork = pbind -b $BIND $$ preENV_OMP_NUM_THREADS = 128 preENV_SUNW_MP_THR_IDLE = SPIN preENV_SUNW_MP_PROCBIND = 127 126 125 124 123 122 121 120 119 118 117 116 115 114 113 112 111 110 109 108 107 106 105 104 103 102 101 100 99 98 97 96 95 94 93 92 91 90 89 88 87 86 85 84 83 82 81 80 79 78 77 76 75 74 73 72 71 70 69 68 67 66 65 64 63 62 61 60 59 58 57 56 55 54 53 52 51 50 49 48 47 46 45 44 43 42 41 40 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 %ifndef %{OverRideCopies} # # To cause the command line to win for both base and peak, say: # # runspec --define OverRideCopies --rate n # # If you _don't_ say that, then the order of precedence is: # # named section (highest), about 6 lines down, peak for several benchmarks # command line # the line marked "header section default" just below # copies = 127 # header section default 470.lbm=peak: copies = 1 436.cactusADM,437.leslie3d=peak: copies = 63 401.bzip2,450.soplex,482.sphinx3=peak: copies = 64 # %endif default: # In the bind statement below, the order of assignment is to work # from high to low, alternating LSBs, chips, cores, threads, from # highest to lowest. The order is derived from the 'prtdiag' # description of the machine. bind = <