# Invocation command line: # /autobench/sources/speccpu2006/bin/runspec --config speccpu_at_new.cfg --action run --rate 1 --copies 128 --reportable --nobuild --tune base --iterations 3 --define 32c750=1 --define distro=rhel64 --define make_j=32 --define huge_pages=8448 --extension rhel64-at-604-new int # output_root was not used for this run ############################################################################ ####################################################################### # # SPEC CPU2006 Benchmark Config file IBM eServer pSeries Linux # ####################################################################### # THIS CONFIG FILE IS FOR BUILDING on P7 # # --define (-S) options that are passed to the runspec command # that dynamically affect this config file # # mybaseopt=x Allows override where x=3,4,5 (-O3, -O4, -O5) for base runs # # debug Adds debug ('-g') compiler option # # rhel64 # sles11sp2 # noL3 # mymode64 # The following setting was inserted automatically as a result of # post-run basepeak application. basepeak = 1 ##################################################################### # System Under Test (SUT) Section # Replace the setting below with the ones that match your # machine. ##################################################################### sysinfo_program = # System Description %ifdef %{48c470} %if '%{distro}' eq 'rhel64' hw_model = IBM Flex System p470 (3.4 GHz, 48 core, RHEL, GCC) %endif %if '%{distro}' eq 'sles11sp2' hw_model = IBM Flex System p470 (3.4 GHz, 48 core, SLES, GCC) %endif hw_avail = May-2013 hw_memory = 512 GB (16x32 GB) DDR3 1066 MHz hw_disk = 2 x 372.11 GB Raid0 SATA SSD hw_cpu_mhz = 3416 hw_nchips = 4 hw_ncores = 48 hw_ncoresperchip = 12 hw_nthreadspercore = 4 hw_fpu = Integrated hw_ncpuorder = 32, 48 cores hw_ocache = None hw_other = None hw_cpu_char000 = Intelligent Energy Optimization enabled, hw_cpu_char001 = up to 3.727 GHz notes_plat_000 = This Compute Node is housed in an "IBM Flex System Enterprise Chassis" notes_plat_005 = notes_plat_010 = The Maximum Power Limit for this Compute Node was set according to notes_plat_015 = recommendation on "IBM Chassis Management Module" %endif %ifdef %{24c270} %if '%{distro}' eq 'rhel64' hw_model = IBM Flex System p270 (3.4 GHz, 24 core, RHEL, GCC) %endif %if '%{distro}' eq 'sles11sp2' hw_model = IBM Flex System p270 (3.4 GHz, 24 core, SLES, GCC) %endif hw_avail = May-2013 hw_memory = 256 GB (16 x 16 GB) DDR3 1066 MHz hw_disk = 2 x 177 GB Raid0 SATA SSD 1.8" hw_cpu_mhz = 3416 hw_nchips = 2 hw_ncores = 24 hw_ncoresperchip = 12 hw_nthreadspercore = 4 hw_fpu = Integrated hw_ncpuorder = 24 cores hw_ocache = None hw_other = None hw_cpu_char000 = Intelligent Energy Optimization enabled, hw_cpu_char001 = up to 3.727 GHz notes_plat_000 = This Compute Node is housed in an "IBM Flex System Enterprise Chassis" notes_plat_005 = notes_plat_010 = The Maximum Power Limit for this Compute Node was set according to notes_plat_015 = recommendation on "IBM Chassis Management Module" %endif %ifdef %{16c24cn} %if '%{distro}' eq 'rhel64' hw_model = IBM Flex System p24L (4.1 GHz, 16 core, RHEL, GCC) %endif %if '%{distro}' eq 'sles11sp2' hw_model = IBM Flex System p24L (4.1 GHz, 16 core, SLES, GCC) %endif hw_avail = Jun-2013 hw_memory = 128 GB (16 x 8 GB) DDR3 1066 MHz hw_disk = 1 x 600 GB SAS SFF 15K RPM hw_cpu_mhz = 4116 hw_nchips = 2 hw_ncores = 16 hw_ncoresperchip = 8 hw_nthreadspercore = 4 hw_fpu = Integrated hw_ncpuorder = 8, 16 cores hw_ocache = None hw_other = None hw_cpu_char000 = Intelligent Energy Optimization enabled, hw_cpu_char001 = up to 4.340 GHz notes_plat_000 = This Compute Node is housed in an "IBM Flex System Enterprise Chassis" notes_plat_005 = notes_plat_010 = The Maximum Power Limit for this Compute Node was set according to notes_plat_015 = recommendation on "IBM Chassis Management Module" %endif %ifdef %{48c760} %if '%{distro}' eq 'rhel64' hw_model = IBM Power 760 (3.4 GHz, 48 core, RHEL, GCC) %endif %if '%{distro}' eq 'sles11sp2' hw_model = IBM Power 760 (3.4 GHz, 48 core, SLES, GCC) %endif hw_avail = Mar-2013 hw_memory = 512 GB (64 x 8 GB) DDR3 1066 MHz hw_disk = 3 x 146.8 GB Raid0 SAS SFF 15K RPM hw_cpu_mhz = 3416 hw_nchips = 4 hw_ncores = 48 hw_ncoresperchip = 12 hw_nthreadspercore = 4 hw_fpu = Integrated hw_ncpuorder = 12, 24, 36, 48 cores hw_ocache = None hw_other = None hw_cpu_char000 = Intelligent Energy Optimization enabled, hw_cpu_char001 = up to 3.787 GHz %endif %ifdef %{32c750} %if '%{distro}' eq 'rhel64' hw_model000= IBM PowerLinux 7R4 (4.0 GHz, 32 core, RHEL, hw_model001 = GCC) %endif %if '%{distro}' eq 'sles11sp2' hw_model = IBM Power 750 Express (4.0 GHz, 32 core, SLES, GCC) %endif hw_avail = Aug-2013 hw_memory = 256 GB (64 x 4 GB) DDR3 1066 MHz hw_disk = 1 x 300 GB SAS SFF 15K RPM hw_cpu_mhz = 4060 hw_nchips = 8 hw_ncores = 32 hw_ncoresperchip = 4 hw_nthreadspercore = 4 hw_fpu = Integrated hw_ncpuorder = 16, 32 cores hw_ocache = None hw_other = None hw_cpu_char000 = Intelligent Energy Optimization enabled, hw_cpu_char001 = up to 4.431 GHz %endif %ifdef %{16c730} %if '%{distro}' eq 'rhel64' hw_model = IBM Power 730 Express (4.2 GHz, 16 core, RHEL, GCC) %endif %if '%{distro}' eq 'sles11sp2' hw_model = IBM Power 730 Express (4.2 GHz, 16 core, SLES, GCC) %endif hw_avail = Feb-2013 hw_memory = 128 GB (16 x 8 GB) DDR3 1066 MHz hw_disk = 5 x 146.8 GB Raid0 SAS SFF 15K RPM hw_cpu_mhz = 4228 hw_nchips = 2 hw_ncores = 16 hw_ncoresperchip = 8 hw_nthreadspercore = 4 hw_fpu = Integrated hw_ncpuorder = 8, 12, 16 cores hw_ocache = None hw_other = None hw_cpu_char000 = Intelligent Energy Optimization enabled, hw_cpu_char001 = up to 4.540 GHz %endif %ifdef %{16c740} %if '%{distro}' eq 'rhel64' hw_model = IBM Power 740 (4.2 GHz, 16 core, RHEL, GCC) %endif %if '%{distro}' eq 'sles11sp2' hw_model = IBM Power 740 (4.2 GHz, 16 core, SLES, GCC) %endif hw_avail = Feb-2013 hw_memory = 128 GB (32 x 4 GB) DDR3 1066 MHz hw_disk = 5 x 146.8 GB Raid0 SAS SFF 15K RPM hw_cpu_mhz = 4228 hw_nchips = 2 hw_ncores = 16 hw_ncoresperchip = 8 hw_nthreadspercore = 4 hw_fpu = Integrated hw_ncpuorder = 6, 8, 12, 16 cores hw_ocache = None hw_other = None hw_cpu_char000 = Intelligent Energy Optimization enabled, hw_cpu_char001 = up to 4.540 GHz %endif %ifdef %{16c7R2} %if '%{distro}' eq 'rhel64' hw_model = IBM PowerLinux 7R2 (4.2 GHz, 16 core, RHEL, GCC) %endif %if '%{distro}' eq 'sles11sp2' hw_model = IBM PowerLinux 7R2 (4.2 GHz, 16 core, SLES, GCC) %endif hw_avail = Feb-2013 hw_memory = 128 GB (16 x 8 GB) DDR3 1066 MHz hw_disk = 5 x 146.8 GB Raid0 SAS SFF 15K RPM hw_cpu_mhz = 4228 hw_nchips = 2 hw_ncores = 16 hw_ncoresperchip = 8 hw_nthreadspercore = 4 hw_fpu = Integrated hw_ncpuorder = 16 cores hw_ocache = None hw_other = None hw_cpu_char000 = Intelligent Energy Optimization enabled, hw_cpu_char001 = up to 4.540 GHz %endif hw_vendor = IBM Corporation # CPU description hw_cpu_name = POWER7+ hw_pcache = 32 KB I + 32 KB D on chip per core hw_scache = 256 KB I+D on chip per core hw_tcache = 10 MB I+D on chip per core # Tester description - edit the date, put your license number here! license_num = 11 prepared_by = -- test_sponsor = IBM Corporation tester = IBM Corporation # Operating system, file system sw_avail = May-2013 %if '%{distro}' eq 'rhel64' sw_os000 = Red Hat Enterprise Linux Server release 6.4 sw_os001 = (ppc64) kernel 2.6.32-358.6.1.el6.ppc64 sw_file = ext4 %endif %if '%{distro}' eq 'sles11sp2' sw_os000 = SUSE Linux Enterprise Server 11 SP2 sw_os001 = (ppc64) kernel 3.0.58-0.6.6-ppc64 sw_file = ext3 %endif sw_compiler000 = C/C++: Version 4.7.3 of IBM sw_compiler001 = Advance Toolchain 6.0-4 gcc/g++ compiler %if '%{distro}' eq 'rhel64' notes_comp_iler_invocation_000: For more information about IBM Advance Toolchain, including support, see notes_comp_iler_invocation_005: ftp://ftp.unicamp.br/pub/linuxpatch/toolchain/at/redhat/RHEL6/at6.0/release_notes.at6.0-6.0-4.html %endif %if '%{distro}' eq 'sles11sp2' notes_comp_iler_invocation_000: For more information about IBM Advance Toolchain, including support, see notes_comp_iler_invocation_005: ftp://ftp.unicamp.br/pub/linuxpatch/toolchain/at/suse/SLES_11/at6.0/release_notes.at6.0-6.0-4.html %endif notes_comp_iler_000 =For more information about IBM Advance Toolchain, including support, see notes_comp_iler_005 =ftp://ftp.unicamp.br/pub/linuxpatch/toolchain/at/redhat/RHEL6/at6.0/release_notes.at6.0-6.0-4.html sw_other000 = -IBM Advance Toolchain 6.0-4 sw_other001 = -IBM Mathematical Acceleration sw_other002 = Subsystem (MASS) libraries 7.1.0.2 sw_base_ptrsize = 32-bit sw_peak_ptrsize = 32-bit sw_parallel_other = -- sw_state = Run level 3 (multi-user) notes_submit_000 = The config file option 'submit' was used notes_submit_005 = to assign benchmark copy to specific kernel thread using notes_submit_010 = the "numactl" command (see flags file for details). notes_os_000 = ulimit -s (stack) set to 1048576. notes_os_005 = notes_os_010 = Large pages reserved as follows by root user: %ifdef %{huge_pages} notes_os_015 = echo 8448 > /proc/sys/vm/nr_hugepages %endif notes_os_020 = notes_os_025 = The Mathematical Acceleration Subsystem libraries notes_os_030 = are shipped with IBM XL C/C++ version 12.1 and notes_os_035 = IBM XL Fortran version 14.1 compiler products. notes_os_040 = ###################################################################### # Environment variables ###################################################################### preENV_TCMALLOC_MEMFS_MALLOC_PATH=/libhugetlbfs preENV_HUGETLB_VERBOSE=0 preENV_HUGETLB_MORECORE=yes preENV_HUGETLB_ELFMAP=RW preENV_XLFRTEOPTS=intrinthds=1 ###################################################################### # End of SUT section # If this config file were to be applied to several SUTs, edits would # be needed only ABOVE this point. ###################################################################### #====================================================================== # Global Options apply to all runs #====================================================================== # # For just a report generated without a raw file or any runs required # Do a "runspec --reportonly" (also --action=report or --fakereport) # # Just set up the build or run directories, use --buildsetup or --setup # on the runspec command # Sets nobuild as the default - you will need -D to force a compile # nobuild = 1 # -D forces a build (recompile) # -N prohibits a build (recompile) #action = validate # -a #tune = base,peak # -T #iterations = 3 # -n output_format = raw,text,html,csv # -o #basepeak = 0 #reportable = 1 # -l turns off #ignore_errors = 0 # -I mean_anyway = 1 allow_extension_override = 1 #ext = LoP-ppc64 # -e #size = ref post_setup = sync parallel_test = 1 # This tells runspec to enable export environment variables of # the form of preENV_xxx preenv = 1 # This tells runspec to set enviornment variables of the form # ENV_xxx if found in here. Forced to 0 for reportable runs env_vars = 0 # Control if the embedded md5 string of compiler options is checked for # recompile if the options have changed check_md5 = 0 # Controls if the log file info is echo'd to the screen teeout = yes # If 1, make copy of the config file before modifing with MD5 # signitures of the compile options. backup_config = 0 # verbose -> -v #verbose = 1 to 99, 5 is default, 91 is good for debugging verbose = 1 %ifdef %{make_j} makeflags = -j%{make_j} %else makeflags =-j4 %endif # We alway use submit to ensure there is a bindprocessor done. # use_submit_for_speed If set, use submit commands for speed as well as rate use_submit_for_speed = 1 # SPEED submit = echo "$command" > run.sh ; numactl --membind=\$SPECCOPYNUM --physcpubind=\$SPECCOPYNUM sh run.sh %if defined(%{48c760}) || defined(%{48c470}) bind = \ 0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 76 80 84 88 92 96 100 104 108 112 116 120 124 128 132 136 140 144 148 152 156 160 164 168 172 176 180 184 188 \ 1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 101 105 109 113 117 121 125 129 133 137 141 145 149 153 157 161 165 169 173 177 181 185 189 \ 2 6 10 14 18 22 26 30 34 38 42 46 50 54 58 62 66 70 74 78 82 86 90 94 98 102 106 110 114 118 122 126 130 134 138 142 146 150 154 158 162 166 170 174 178 182 186 190 \ 3 7 11 15 19 23 27 31 35 39 43 47 51 55 59 63 67 71 75 79 83 87 91 95 99 103 107 111 115 119 123 127 131 135 139 143 147 151 155 159 163 167 171 175 179 183 187 191 copies = 192 submit = let "MYCPU=\$BIND"; MYMEM=`printf '%d' \$((\$MYCPU/24))` ; numactl --physcpubind=\$MYCPU --membind=\$MYMEM $command %endif %ifdef %{32c750} bind = \ 0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 76 80 84 88 92 96 100 104 108 112 116 120 124 \ 1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 101 105 109 113 117 121 125 \ 2 6 10 14 18 22 26 30 34 38 42 46 50 54 58 62 66 70 74 78 82 86 90 94 98 102 106 110 114 118 122 126 \ 3 7 11 15 19 23 27 31 35 39 43 47 51 55 59 63 67 71 75 79 83 87 91 95 99 103 107 111 115 119 123 127 copies = 128 submit = let "MYCPU=\$BIND"; MYMEM=`printf '%d' \$((\$MYCPU/16))` ; numactl --physcpubind=\$MYCPU --membind=\$MYMEM $command %endif %ifdef %{24c270} bind = \ 0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 76 80 84 88 92 \ 1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 \ 2 6 10 14 18 22 26 30 34 38 42 46 50 54 58 62 66 70 74 78 82 86 90 94 \ 3 7 11 15 19 23 27 31 35 39 43 47 51 55 59 63 67 71 75 79 83 87 91 95 copies = 96 submit = let "MYCPU=\$BIND"; MYMEM=`printf '%d' \$((\$MYCPU/24))` ; numactl --physcpubind=\$MYCPU --membind=\$MYMEM $command %endif %if defined(%{16c730}) || defined(%{16c740}) || defined(%{16c7R2}) || defined(%{16c24cn}) bind = \ 0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 \ 1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 \ 2 6 10 14 18 22 26 30 34 38 42 46 50 54 58 62 \ 3 7 11 15 19 23 27 31 35 39 43 47 51 55 59 63 copies = 64 submit = echo "$command" > run.sh; let "MYCPU=\$BIND"; MYMEM=`printf '%d' \$((\$MYCPU/32))` ; numactl --physcpubind=\$MYCPU --membind=\$MYMEM sh run.sh %endif #--prevent log file lines that are too long for vi log_line_width = 180 # Point to where the usr/compiler/OS Flags file is flagsurl000= http://www.spec.org/cpu2006/flags/IBM-Linux-AT.20130813.xml #=============================================================================== # How the compilers are called # The settings used here are for IBM Advance Toolchain C/C++ and Fortran compilers #=============================================================================== default=default=default=default: %define MY_PATH /opt/at6.0/bin CC = %{MY_PATH}/gcc CXX = %{MY_PATH}/g++ FC = %{MY_PATH}/gfortran #====================================================================== # Default Level Optimizations # No limit to number of flags, must be same for all of each Language. #====================================================================== default=default=default=default: # -O3 implies: -finline-functions, -funswitch-loops, -fpredictive-commoning, -fgcse-after-reload, -ftree-vectorize and -fipa-cp-clone # -O3 -mcpu=power7 implies: -ftree-vectorize, -mvsx and -maltivec ########## %define OPT_32BIT -m32 %define OPT_64BIT -m64 %define OPT_ALIAS -fno-strict-aliasing %define OPT_ALIGN_FUNCS %define OPT_ALIGN_JUMPS %define OPT_ALIGN_LOOPS %define OPT_AUTOPAR -ftree-parallelize-loops=6 %define OPT_CMODEL_LARGE -mcmodel=large %define OPT_CMODEL_MEDIUM -mcmodel=medium %define OPT_CMODEL_SMALL -mcmodel=small %define OPT_EXTRA -g %define OPT_FASTMATH -ffast-math %define OPT_LEVEL -O3 %define OPT_LTO -flto -fwhole-program -fuse-linker-plugin %define OPT_MASS %{OPT_VECLIBABI} %define OPT_MATH -fno-associative-math %define OPT_OPENMP -fopenmp %define OPT_PROCESSOR -mcpu=power7 -mtune=power7 %define OPT_PROFILE1 -fprofile-generate %define OPT_PROFILE2 -fprofile-use %define OPT_RECIP -mrecip=rsqrt %define OPT_TOC %define OPT_USUAL -fpeel-loops -funroll-loops %define OPT_USUAL2 -msave-toc-indirect -mno-pointers-to-nested-functions -fno-jump-tables %define OPT_USUAL3 -mpopcntd %define OPT_VECLIBABI -mveclibabi=mass %define OPT_VECTOR -fvect-cost-model #%define OPTIMIZE_NORMAL_NOBIT %{OPT_EXTRA} %{OPT_FASTMATH} %{OPT_LEVEL} %{OPT_LTO} %{OPT_MASS} %{OPT_PROCESSOR} %{OPT_RECIP} %{OPT_USUAL} %{OPT_USUAL2} %{OPT_USUAL3} %{OPT_VECTOR} %define OPTIMIZE_NORMAL_NOBIT %{OPT_FASTMATH} %{OPT_LEVEL} %{OPT_PROCESSOR} %{OPT_RECIP} %{OPT_USUAL} %{OPT_USUAL3} %define OPTIMIZE_NORMAL_32BIT %{OPTIMIZE_NORMAL_NOBIT} %{OPT_32BIT} %define OPTIMIZE_NORMAL_64BIT %{OPTIMIZE_NORMAL_NOBIT} %{OPT_64BIT} %{OPT_CMODEL_MEDIUM} %if %{mymode64} %define OPTIMIZE_NORMAL %{OPTIMIZE_NORMAL_64BIT} %else %define OPTIMIZE_NORMAL %{OPTIMIZE_NORMAL_32BIT} %endif ########## %define LIB_LHBDT -B/usr/share/libhugetlbfs/ -Wl,--hugetlbfs-align %define LIB_LHLIB -lhugetlbfs %define LIB_MALLOC -ltcmalloc -lstdc++ -lpthread %define LIB_MATH -lm %define LIB_MASS_32BIT -lmassvp7 -lmass_simdp7 -lmass %define LIB_MASS_64BIT -lmassvp7_64 -lmass_simdp7_64 -lmass_64 %define LIB_MASS %{LIB_MASS_64BIT} #%define EXTRA_LIBS_NOBIT %{LIB_MALLOC} %{LIB_MATH} %define EXTRA_LIBS_NOBIT %{LIB_MATH} %define EXTRA_LIBS_32BIT %{LIB_MASS_32BIT} %{EXTRA_LIBS_NOBIT} %define EXTRA_LIBS_64BIT %{LIB_MASS_64BIT} %{EXTRA_LIBS_NOBIT} %if %{mymode64} %define EXTRA_LIBS %{EXTRA_LIBS_64BIT} %else %define EXTRA_LIBS %{EXTRA_LIBS_32BIT} %endif ########## %define BASE_DIR /opt/at6.0 %define LD_LIBDIR_COMPILER_32BIT %{BASE_DIR}/lib %define LD_LIBDIR_COMPILER_64BIT %{BASE_DIR}/lib64 %define LD_LIBDIR_32BIT_1 -L %{LD_LIBDIR_COMPILER_32BIT} %define LD_LIBDIR_32BIT_2 -L /opt/ibmcmp/xlmass/7.1/lib %define LD_LIBDIR_32BIT_3 -L /opt/at6.0/lib %define LD_LIBDIR_32BIT %{LD_LIBDIR_32BIT_1} %{LD_LIBDIR_32BIT_2} %{LD_LIBDIR_32BIT_3} %define LD_LIBDIR_64BIT_1 -L %{LD_LIBDIR_COMPILER_64BIT} %define LD_LIBDIR_64BIT_2 -L /opt/ibmcmp/xlmass/7.1/lib64 %define LD_LIBDIR_64BIT_3 -L /opt/at6.0/lib64 %define LD_LIBDIR_64BIT %{LD_LIBDIR_64BIT_1} %{LD_LIBDIR_64BIT_2} %{LD_LIBDIR_64BIT_3} %define LD_RPATH_32BIT_1 -Wl,-rpath,%{LD_LIBDIR_COMPILER_32BIT} %define LD_RPATH_32BIT_2 -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib %define LD_RPATH_32BIT_3 -Wl,-rpath,/opt/at6.0/lib %define LD_RPATH_32BIT %{LD_RPATH_32BIT_1} %{LD_RPATH_32BIT_2} %{LD_RPATH_32BIT_3} %define LD_RPATH_64BIT_1 -Wl,-rpath,%{LD_LIBDIR_COMPILER_64BIT} %define LD_RPATH_64BIT_2 -Wl,-rpath,/opt/ibmcmp/xlmass/7.1/lib64 %define LD_RPATH_64BIT_3 -Wl,-rpath,/opt/at6.0/lib64 %define LD_RPATH_64BIT %{LD_RPATH_64BIT_1} %{LD_RPATH_64BIT_2} %{LD_RPATH_64BIT_3} %define LD_DYNAMIC_LINKER_64BIT -Wl,-dynamic-linker,/opt/at6.0/lib64/ld64.so.1 %define LD_KEEP_RELOCS -Wl,-q %define LD_LINK_MAP -Wl,-Map=link.map,--cref %define EXTRA_LDFLAGS_NOBIT %{LD_KEEP_RELOCS} %{LD_LINK_MAP} %define EXTRA_LDFLAGS_32BIT %{EXTRA_LDFLAGS_NOBIT} %{OPTIMIZE_NORMAL_32BIT} %{LD_LIBDIR_32BIT} %{LD_RPATH_32BIT} %define EXTRA_LDFLAGS_64BIT %{EXTRA_LDFLAGS_NOBIT} %{OPTIMIZE_NORMAL_64BIT} %{LD_LIBDIR_64BIT} %{LD_RPATH_64BIT} %{LD_DYNAMIC_LINKER_64BIT} %if %{mymode64} %define EXTRA_LDFLAGS %{EXTRA_LDFLAGS_64BIT} %else %define EXTRA_LDFLAGS %{EXTRA_LDFLAGS_32BIT} %endif ########## %define FDPRPRO /opt/ibm/fdprpro/bin/fdprpro %define FDPROPTS -O4 -omullX %define FDPRINT -imullX # Language options %define OPT_C %define OPT_CPLUSPLUS --param max-inline-insns-auto=200 %define OPT_FORTRAN #====================================================================== # Default Base Flags #====================================================================== default=base=default=default: %if %{mymode64} PORTABILITY = -DSPEC_CPU_LP64 CPORTABILITY = -DSPEC_CPU_LP64 CXXPORTABILITY = -DSPEC_CPU_LP64 FPPPORTABILITY = -DSPEC_CPU_LP64 %endif #====================================================================== # INT Portability Flags # no optimizations allowed here #====================================================================== int=default=default=default: 400.perlbench=default=default=default: %if %{mymode64} CPORTABILITY = -DSPEC_CPU_LINUX_PPC -DSPEC_CPU_LP64 %else CPORTABILITY = -DSPEC_CPU_LINUX_PPC %endif 403.gcc=default=default=default: %if %{mymode64} CPORTABILITY = -DSPEC_CPU_LP64 %endif 462.libquantum=default=default=default: CPORTABILITY = -DSPEC_CPU_LINUX 464.h264ref=default=default=default: CPORTABILITY = -fsigned-char 483.xalancbmk=default=default=default: CXXPORTABILITY = -DSPEC_CPU_LINUX #====================================================================== # FP Portability Flags # no optimizations allowed here #====================================================================== fp=default=default=default: 410.bwaves=peak=default=default: FPPPORTABILITY = -DSPEC_CPU_LP64 436.cactusADM=peak=default=default: CPORTABILITY = -DSPEC_CPU_LP64 FPPPORTABILITY = -DSPEC_CPU_LP64 447.dealII=default=default=default: CXXPORTABILITY += -DSPEC_CPU_LINUX 459.GemsFDTD=peak=default=default: FPPPORTABILITY = -DSPEC_CPU_LP64 470.lbm=peak=default=default: CPORTABILITY = -DSPEC_CPU_LP64 481.wrf=default=default=default: CPORTABILITY += -DSPEC_CPU_CASE_FLAG -DSPEC_CPU_LINUX_PPC 481.wrf=peak=default=default: CPORTABILITY += -DSPEC_CPU_LP64 FPPPORTABILITY = -DSPEC_CPU_LP64 482.sphinx3=default=default=default: CPORTABILITY += -fsigned-char #====================================================================== # INT and FP Base Flags #====================================================================== int=base=default=default: OPTIMIZE = %{OPTIMIZE_NORMAL} EXTRA_LIBS = %{LIB_LHLIB} %{EXTRA_LIBS} EXTRA_LDFLAGS = %{EXTRA_LDFLAGS} COPTIMIZE = %{OPT_C} %{OPT_LTO} CXXOPTIMIZE = %{OPT_CPLUSPLUS} EXTRA_CXXLIBS = %{LIB_MALLOC} fp=base=default=default: OPTIMIZE = %{OPTIMIZE_NORMAL} %{OPT_VECTOR} %{OPT_MASS} EXTRA_LIBS = %{LIB_LHLIB} %{EXTRA_LIBS} EXTRA_LDFLAGS = %{EXTRA_LDFLAGS} COPTIMIZE = %{OPT_C} CXXOPTIMIZE = %{OPT_CPLUSPLUS} %{OPT_MATH} %{OPT_LTO} FOPTIMIZE = %{OPT_FORTRAN} #====================================================================== # # Default peak flags. # #====================================================================== default=peak=default=default: #%define O3 -O3 %{BASEARCH} %{MODE} %{GCCOPT} %{DEBUG} %define O3 %{OPTIMIZE_NORMAL} int=peak=default=default: OPTIMIZE = %{OPTIMIZE_NORMAL} EXTRA_LIBS = %{EXTRA_LIBS} EXTRA_LDFLAGS = %{EXTRA_LDFLAGS} COPTIMIZE = %{OPT_C} CXXOPTIMIZE = %{OPT_CPLUSPLUS} EXTRA_CXXLIBS = %{LIB_MALLOC} fp=peak=default=default: OPTIMIZE = %{OPTIMIZE_NORMAL} %{OPT_VECTOR} %{OPT_MASS} EXTRA_LIBS = %{LIB_LHLIB} %{EXTRA_LIBS} EXTRA_LDFLAGS = %{EXTRA_LDFLAGS} COPTIMIZE = %{OPT_C} CXXOPTIMIZE = %{OPT_CPLUSPLUS} %{OPT_MATH} FOPTIMIZE = %{OPT_FORTRAN} #====================================================================== # INT peak Flags #====================================================================== int=peak=default=default: 400.perlbench=peak=default=default: # C only Benchmark %ifdef %{debug_copies} copies = %{debug_copies} %endif OPTIMIZE += %{OPT_LTO} #EXTRA_LIBS += %{LIB_LHLIB} %{LIB_MALLOC} EXTRA_LIBS += %{LIB_LHLIB} PASS1_CFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_CFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} 401.bzip2=peak=default=default: # C only Benchmark %ifdef %{debug_copies} copies = %{debug_copies} %endif %define FDPROPTS_BZIP2 -O4 -vrox %define FDPRINT_BZIP2 EXTRA_LIBS += %{LIB_LHLIB} PASS1_CFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_CFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT_BZIP2} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS_BZIP2} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} 403.gcc=peak=default=default: # C only Benchmark %ifdef %{debug_copies} copies = %{debug_copies} %endif %define FDPROPTS_GCC -O4 -nodp -rtb -m power7 -A 1 %define FDPRINT_GCC #OPTIMIZE = %{OPTIMIZE_NORMAL_64BIT} %{OPT_MASS} %{OPT_LTO} %{OPT_USUAL2} #EXTRA_LIBS = %{LIB_LHLIB} %{EXTRA_LIBS_64BIT} #EXTRA_LDFLAGS = %{EXTRA_LDFLAGS_64BIT} #OPTIMIZE += %{OPT_LTO} EXTRA_LIBS += %{LIB_LHLIB} #PASS1_CFLAGS = %{OPT_PROFILE1} #PASS1_LDFLAGS = %{OPT_PROFILE1} #PASS2_CFLAGS = %{OPT_PROFILE2} #PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre1 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT_GCC} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run1 = $command fdo_post1 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS_GCC} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} 429.mcf=peak=default=default: # C only Benchmark %ifdef %{32c750} copies = 32 %endif %ifdef %{debug_copies} copies = %{debug_copies} %endif #%define FDPROPTS_MCF -O3 -m power7 -A 1 -omullX OPTIMIZE += --param max-goto-duplication-insns=16 EXTRA_LIBS += %{LIB_LHLIB} #EXTRA_LIBS += %{LIB_LHLIB} %{LIB_MALLOC} fdo_pre1 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run1 = $command fdo_post1 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} #fdo_post1 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS_MCF} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} 445.gobmk=peak=default=default: # C only Benchmark %ifdef %{debug_copies} copies = %{debug_copies} %endif #%define FDPROPTS_GOBMK -m power7 -A 1 -q -O3 -omullX %define FDPROPTS_GOBMK -O3 %define FDPRINT_GOBMK OPTIMIZE += %{OPT_LTO} EXTRA_LIBS += %{LIB_LHLIB} %{LIB_MALLOC} PASS1_CFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_CFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT_GOBMK} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS_GOBMK} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} 456.hmmer=peak=default=default: # C only Benchmark %ifdef %{debug_copies} copies = %{debug_copies} %endif OPTIMIZE += -finline-limit=11880 #EXTRA_LIBS += %{LIB_LHLIB} %{LIB_MALLOC} EXTRA_LIBS += %{LIB_LHLIB} PASS1_CFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_CFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} 458.sjeng=peak=default=default: # C only Benchmark %ifdef %{debug_copies} copies = %{debug_copies} %endif OPTIMIZE += -misel #EXTRA_LIBS += %{LIB_LHLIB} %{LIB_MALLOC} EXTRA_LIBS += %{LIB_LHLIB} PASS1_CFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_CFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} 462.libquantum=peak=default=default: # C only Benchmark %ifdef %{32c750} copies = 64 %endif %ifdef %{debug_copies} copies = %{debug_copies} %endif # We want just 1 copy per core, not SMT, when we have L3cache %define FDPROPTS_LIBQUANTUM -m power7 -A 1 -bf -dp -lro -nop -RC -RD -tb -tlo -vro -isf 88 -lu 8 -hrf 0.10 -sdp 4 -lun 27 -omullX %if %{noL3} %define FDPROPTS_LIBQUANTUM -ece -pca -RC -RD -tb -tlo -lu 5 -rt 0.75 -hrf 0.75 -sdpla 4 -rcctf 0.55 -rccrf 0.10 -rcaf 2 -omullX %endif OPTIMIZE += %{OPT_LTO} EXTRA_LIBS += %{LIB_LHLIB} %{LIB_MALLOC} PASS1_CFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_CFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS_LIBQUANTUM} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} 464.h264ref=peak=default=default: # C only Benchmark %ifdef %{debug_copies} copies = %{debug_copies} %endif OPTIMIZE += %{OPT_LTO} -fno-branch-count-reg -misel -fprefetch-loop-arrays EXTRA_LIBS += %{LIB_MALLOC} PASS1_CFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_CFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof -O4 -vrox -nodp ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} 471.omnetpp=peak=default=default: # C++ Benchmark %ifdef %{32c750} copies = 64 %endif %ifdef %{debug_copies} copies = %{debug_copies} %endif #%define FDPROPTS_OMNETPP -O3 -lu -1 -nodp -sdp 9 EXTRA_LIBS += %{LIB_LHLIB} %{LIB_MALLOC} #fdo_pre1 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} #fdo_run1 = $command #fdo_post1 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS_OMNETPP} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} 473.astar=peak=default=default: # C++ Benchmark %ifdef %{32c750} copies = 96 %endif %ifdef %{debug_copies} copies = %{debug_copies} %endif #%define FDPROPTS_ASTAR -O4 -m power7 -A 1 OPTIMIZE += %{OPT_LTO} EXTRA_LIBS += %{LIB_LHLIB} %{LIB_MALLOC} #PASS1_CXXFLAGS = %{OPT_PROFILE1} #PASS1_LDFLAGS = %{OPT_PROFILE1} #PASS2_CXXFLAGS = %{OPT_PROFILE2} #PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre1 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run1 = $command fdo_post1 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} #fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS_ASTAR} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} 483.xalancbmk=peak=default=default: # C++ Benchmark %ifdef %{debug_copies} copies = %{debug_copies} %endif #%define FDPROPTS_XALAN -O3 -m power7 #OPTIMIZE += %{OPT_LTO} EXTRA_LIBS += %{LIB_LHLIB} %{LIB_MALLOC} PASS1_CXXFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_CXXFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} #fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS_XALAN} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} 999.specrand=peak=default=default: # C only Benchmark - Must run, but not reported unless fails %ifdef %{debug_copies} copies = %{debug_copies} %endif #============================================================ # FP peak Flags #============================================================ fp=peak=default=default: 410.bwaves=peak=default=default: # Fortran Benchmark %ifdef %{32c750} copies = 32 %endif %ifdef %{debug_copies} copies = %{debug_copies} %endif OPTIMIZE = %{OPTIMIZE_NORMAL_64BIT} %{OPT_VECTOR} %{OPT_MASS} %{OPT_LTO} EXTRA_LIBS = %{LIB_LHLIB} %{EXTRA_LIBS_64BIT} EXTRA_LDFLAGS = %{EXTRA_LDFLAGS_64BIT} FOPTIMIZE = %{OPT_FORTRAN} PASS1_FFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_FFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} notes_peak_fdpr_410_000= Post-Link optimization tool used for: notes_peak_fdpr_410_005= 410.bwaves notes_peak_fdpr_410_010= with options %{FDPROPTS} for optimization phase, notes_peak_fdpr_410_015= and %{FDPRINT} for instrumentation phase 416.gamess=peak=default=default: # Fortran Benchmark %ifdef %{debug_copies} copies = %{debug_copies} %endif EXTRA_LIBS += %{LIB_MALLOC} PASS1_FFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_FFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} notes_peak_fdpr_416_000= 416.gamess notes_peak_fdpr_416_005= with options %{FDPROPTS} for optimization phase, notes_peak_fdpr_416_010= and %{FDPRINT} for instrumentation phase 433.milc=peak=default=default: # C only Benchmark %ifdef %{32c750} copies = 32 %endif %ifdef %{debug_copies} copies = %{debug_copies} %endif OPTIMIZE += %{OPT_LTO} PASS1_CFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_CFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} notes_peak_fdpr_433_000= 433.milc notes_peak_fdpr_433_005= with options %{FDPROPTS} for optimization phase, notes_peak_fdpr_433_010= and %{FDPRINT} for instrumentation phase 434.zeusmp=peak=default=default: # Fortran Benchmark %ifdef %{debug_copies} copies = %{debug_copies} %endif OPTIMIZE += %{OPT_LTO} EXTRA_LIBS += %{LIB_MALLOC} PASS1_FFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_FFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} notes_peak_fdpr_434_000= 434.zeusmp notes_peak_fdpr_434_005= with options %{FDPROPTS} for optimization phase, notes_peak_fdpr_434_010= and %{FDPRINT} for instrumentation phase 435.gromacs=peak=default=default: # Fortran & C Benchmark %ifdef %{debug_copies} copies = %{debug_copies} %endif PASS1_CFLAGS = %{OPT_PROFILE1} PASS1_FFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_CFLAGS = %{OPT_PROFILE2} PASS2_FFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} #fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} #fdo_run2 = $command #fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} 436.cactusADM=peak=default=default: # Fortran & C Benchmark %ifdef %{32c750} copies = 64 %endif %ifdef %{debug_copies} copies = %{debug_copies} %endif OPTIMIZE = %{OPTIMIZE_NORMAL_64BIT} %{OPT_VECTOR} %{OPT_MASS} #OPTIMIZE = %{OPTIMIZE_NORMAL_64BIT} %{OPT_VECTOR} %{OPT_MASS} %{OPT_LTO} EXTRA_LIBS = %{LIB_LHLIB} %{EXTRA_LIBS_64BIT} EXTRA_LDFLAGS = %{EXTRA_LDFLAGS_64BIT} COPTIMIZE = %{OPT_C} FOPTIMIZE = %{OPT_FORTRAN} PASS1_CFLAGS = %{OPT_PROFILE1} PASS1_FFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_CFLAGS = %{OPT_PROFILE2} PASS2_FFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} #fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} #fdo_run2 = $command #fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} 437.leslie3d=peak=default=default: # Fortran Benchmark %ifdef %{32c750} copies = 32 %endif %ifdef %{debug_copies} copies = %{debug_copies} %endif OPTIMIZE += %{OPT_LTO} PASS1_FFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_FFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} notes_peak_fdpr_437_000= 437.leslie3d notes_peak_fdpr_437_005= with options %{FDPROPTS} for optimization phase, notes_peak_fdpr_437_010= and %{FDPRINT} for instrumentation phase 444.namd=peak=default=default: # C++ Benchmark %ifdef %{debug_copies} copies = %{debug_copies} %endif basepeak = 1 447.dealII=peak=default=default: # C++ Benchmark %ifdef %{debug_copies} copies = %{debug_copies} %endif OPTIMIZE += %{OPT_LTO} EXTRA_LIBS += %{LIB_MALLOC} #PASS1_CXXFLAGS = %{OPT_PROFILE1} #PASS1_LDFLAGS = %{OPT_PROFILE1} #PASS2_CXXFLAGS = %{OPT_PROFILE2} #PASS2_LDFLAGS = %{OPT_PROFILE2} #fdo_pre1 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} #fdo_run1 = $command #fdo_post1 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} 450.soplex=peak=default=default: # C++ Benchmark %ifdef %{32c750} copies = 32 %endif %ifdef %{debug_copies} copies = %{debug_copies} %endif OPTIMIZE += %{OPT_LTO} PASS1_CXXFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_CXXFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} #fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} #fdo_run2 = $command #fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} 453.povray=peak=default=default: # C++ Benchmark %ifdef %{debug_copies} copies = %{debug_copies} %endif OPTIMIZE += %{OPT_LTO} EXTRA_LIBS += %{LIB_MALLOC} PASS1_CXXFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_CXXFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} notes_peak_fdpr_453_000= 453.povray notes_peak_fdpr_453_005= with options %{FDPROPTS} for optimization phase, notes_peak_fdpr_453_010= and %{FDPRINT} for instrumentation phase 454.calculix=peak=default=default: # Fortran & C Benchmark %ifdef %{debug_copies} copies = %{debug_copies} %endif COPTIMIZE += %{OPT_LTO} EXTRA_LIBS += %{LIB_MALLOC} fdo_pre1 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run1 = $command fdo_post1 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} notes_peak_fdpr_454_000= 454.calculix notes_peak_fdpr_454_005= with options %{FDPROPTS} for optimization phase, notes_peak_fdpr_454_010= and %{FDPRINT} for instrumentation phase 459.GemsFDTD=peak=default=default: # Fortran Benchmark %ifdef %{32c750} copies = 64 %endif %ifdef %{debug_copies} copies = %{debug_copies} %endif OPTIMIZE = %{OPTIMIZE_NORMAL_64BIT} %{OPT_VECTOR} %{OPT_MASS} #OPTIMIZE = %{OPTIMIZE_NORMAL_64BIT} %{OPT_VECTOR} %{OPT_MASS} %{OPT_LTO} EXTRA_LIBS = %{LIB_LHLIB} %{EXTRA_LIBS_64BIT} EXTRA_LDFLAGS = %{EXTRA_LDFLAGS_64BIT} FOPTIMIZE = %{OPT_FORTRAN} PASS1_FFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_FFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} notes_peak_fdpr_459_000= 459.GemsFDTD notes_peak_fdpr_459_005= with options %{FDPROPTS} for optimization phase, notes_peak_fdpr_459_010= and %{FDPRINT} for instrumentation phase 465.tonto=peak=default=default: # Fortran Benchmark %ifdef %{debug_copies} copies = %{debug_copies} %endif EXTRA_LIBS += %{LIB_MALLOC} PASS1_FFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_FFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} notes_peak_fdpr_465_000= 465.tonto notes_peak_fdpr_465_005= with options %{FDPROPTS} for optimization phase, notes_peak_fdpr_465_010= and %{FDPRINT} for instrumentation phase 470.lbm=peak=default=default: # C Benchmark %ifdef %{32c750} copies = 32 %endif %ifdef %{debug_copies} copies = %{debug_copies} %endif OPTIMIZE = %{OPTIMIZE_NORMAL_64BIT} %{OPT_VECTOR} %{OPT_MASS} %{OPT_LTO} EXTRA_LIBS = %{LIB_LHLIB} %{EXTRA_LIBS_64BIT} EXTRA_LDFLAGS = %{EXTRA_LDFLAGS_64BIT} COPTIMIZE = %{OPT_C} FOPTIMIZE = %{OPT_FORTRAN} PASS1_CFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_CFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} notes_peak_fdpr_470_000= 470.lbm notes_peak_fdpr_470_005= with options %{FDPROPTS} for optimization phase, notes_peak_fdpr_470_010= and %{FDPRINT} for instrumentation phase 481.wrf=peak=default=default: # Fortran & C-for netCDF Benchmark %ifdef %{32c750} copies = 96 %endif %ifdef %{debug_copies} copies = %{debug_copies} %endif OPTIMIZE = %{OPTIMIZE_NORMAL_64BIT} %{OPT_VECTOR} %{OPT_MASS} EXTRA_LIBS = %{LIB_LHLIB} %{EXTRA_LIBS_64BIT} EXTRA_LDFLAGS = %{EXTRA_LDFLAGS_64BIT} COPTIMIZE = %{OPT_C} FOPTIMIZE = %{OPT_FORTRAN} fdo_pre1 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run1 = $command fdo_post1 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} notes_peak_fdpr_481_000= 481.wrf notes_peak_fdpr_481_005= with options %{FDPROPTS} for optimization phase, notes_peak_fdpr_481_010= and %{FDPRINT} for instrumentation phase 482.sphinx3=peak=default=default: # C only Benchmark %ifdef %{32c750} copies = 64 %endif %ifdef %{debug_copies} copies = %{debug_copies} %endif PASS1_CFLAGS = %{OPT_PROFILE1} PASS1_LDFLAGS = %{OPT_PROFILE1} PASS2_CFLAGS = %{OPT_PROFILE2} PASS2_LDFLAGS = %{OPT_PROFILE2} fdo_pre2 = cp ${baseexe} ${baseexe}.orig ; %{FDPRPRO} -a instr %{FDPRINT} ${baseexe}; cp ${baseexe}.instr ${baseexe} fdo_run2 = $command fdo_post2 = cp ${baseexe}.orig ${baseexe} ; %{FDPRPRO} -a opt -f ${baseexe}.nprof %{FDPROPTS} ${baseexe} ; cp ${baseexe}.fdpr ${baseexe} notes_peak_fdpr_429_000= 429.mcf notes_peak_fdpr_429_005= with options %{FDPROPTS} for optimization phase, notes_peak_fdpr_429_010= and %{FDPRINT} for instrumentation phase 998.specrand=peak=default=default: # C only Benchmark - Must run, but not reported unless fails %ifdef %{debug_copies} copies = %{debug_copies} %endif # 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. default: notes_000 =Environment variables set by runspec before the start of the run: notes_005 =HUGETLB_ELFMAP = "RW" notes_010 =HUGETLB_MORECORE = "yes" notes_015 =HUGETLB_VERBOSE = "0" notes_020 =TCMALLOC_MEMFS_MALLOC_PATH = "/libhugetlbfs" notes_025 =XLFRTEOPTS = "intrinthds=1" notes_030 =