# Invocation command line: # /root/work/cpu2006v1.2/bin/runspec -c amd1104il-speed-revA.cfg -T all fp # output_root was not used for this run ############################################################################ # AMD SPEC CPU2006 V1.2 Speed Configuration File for 64-bit Linux # # CPU2006 Version 1.2 # Compiler name/version: Open64 4.2.5 # Operating system version: 64-bit Linux # Hardware: AMD Opteron (Orochi) # FP Base Pointer Size: 64-bit only # FP Peak Pointer Size: 32/64-bit # INT Base Pointer Size: 32/64-bit # INT Peak Pointer Size: 32/64-bit # Auto Parallization: Yes # # Important! Please run with your stack size set to 'unlimited'. # Failure to do so may cause 483.xalancbmk to get a stack overflow during # execution. # Using csh: unlimit # Using bash: ulimit -s unlimited # # Your LD_LIBRARY_PATH must be set to the location of the dependent # runtime libraries. (Though this should be taken care of by the # preENV_LD_LIBRARY_PATH found in the AMD1002mc-speed-revX.inc file.) # This line should be commented out during builds. # ##################################################################### ext = amd1104-speed-revA ignore_errors = no tune = base,peak output_format = all flagsurl000 = http://www.spec.org/cpu2006/flags/x86-open64-425-flags-speed-revA.xml $[top]/amd-platform-speed-revA.xml size = test,train,ref check_md5 = yes reportable = yes env_vars = no teeout = yes mean_anyway = yes teerunout = yes post_setup = sync use_submit_for_speed = 1 sysinfo_program = specperl $[top]/Docs/sysinfo -f -p #################################################################### # Include file containing the commonly changed fields #################################################################### %define inc_file_name AMD1104il-speed-revA.inc #include: %{inc_file_name} # ----- Begin inclusion of 'AMD1104il-speed-revA.inc' ############################################################################ ##################################################################### # Paths ##################################################################### # Set location of run time libraries for runs. preENV_LD_LIBRARY_PATH = $[top]/amd1104-speed-libs-revA/32:$[top]/amd1104-speed-libs-revA/64 # Set this path to your SmartHeap Library for builds. SMARTHEAP_DIR = /root/work/libraries/SmartHeap-10/lib ##################################################################### # Machine Specific Settings ##################################################################### # If your machine orders nodes sequentially, you have 8GB/socket # memory, and your system has huge pages configured, most settings # will be correctly configured based on the below two settings. See # README.amd1104il-speed for more information. # How many sockets does your test system have? (1, 2, or 4) %define num_sockets 2 # How many cores per socket does your system have? (4, 8, 12, or 16) %define cores_per_socket 4 ######################################################################## # You should not need to change anything in this section unless you have # a non-standard configuration. See README.amd1104il-speed for more # information. ######################################################################## %if '%{cores_per_socket}' eq '16' % define cores_per_node 8 % define total_I_cache 512 %elif '%{cores_per_socket}' eq '12' % define cores_per_node 6 % define total_I_cache 384 %elif '%{cores_per_socket}' eq '8' % define cores_per_node 4 % define total_I_cache 256 %elif '%{cores_per_socket}' eq '4' % define cores_per_node 2 % define total_I_cache 128 %endif #cores_per_socket %if ('%{num_sockets}' eq '1') % define mem_sticks 4 % define memory_size 32 % define num_system_huge_pages 2000 %elif ('%{num_sockets}' eq '2') % define mem_sticks 8 % define memory_size 64 % define num_system_huge_pages 4000 %elif ('%{num_sockets}' eq '4') % define mem_sticks 16 % define memory_size 128 % define num_system_huge_pages 8000 %endif #num_sockets %define total_L2_cache %{cores_per_socket} %warning Please ensure you have %{num_system_huge_pages} huge pages configured on this system. This message was generated #################################################################### # Tester information #################################################################### license_num = 49 prepared_by = AMD Performance Lab tester = Advanced Micro Devices test_sponsor = Advanced Micro Devices hw_vendor = Supermicro hw_model000 = Supermicro A+ Server 1022G-NTF, hw_model001 = AMD Opteron 6204 #################################################################### # Hardware information #################################################################### hw_avail = Nov-2011 hw_cpu_name = AMD Opteron 6204 hw_cpu_mhz = 3300 hw_fpu = Integrated hw_nchips = 2 hw_ncores = 8 hw_ncoresperchip = 4 hw_nthreadspercore = 1 %if ('%{num_sockets}' eq '1') hw_ncpuorder = 1 chip %elif ('%{num_sockets}' eq '2') hw_ncpuorder = 1,2 chips %elif ('%{num_sockets}' eq '4') hw_ncpuorder = 2,4 chips %endif #num_sockets hw_pcache000 = 128 KB I on chip per chip, hw_pcache001 = 64 KB I shared / 2 cores; hw_pcache002 = 16 KB D on chip per core hw_scache = 4 MB I+D on chip per chip, 2 MB shared / 2 cores hw_tcache = 16 MB I+D on chip per chip, 8 MB shared / 2 cores hw_ocache = None hw_disk = 1 x 500 GB SATA, 7200 RPM hw_memory = 64 GB (8 x 8 GB 2Rx4 PC3-12800R-11, ECC) hw_other = None sw_file = ext3 sw_os000 = Red Hat Enterprise Linux Server release 6.1, sw_os001 = Kernel 2.6.32-131.0.15.el6.x86_64 sw_state = Run level 3 (Full multiuser with network) ##################################################################### # Notes ##################################################################### notes_030 =The x86 Open64 Compiler Suite is only available from (and supported by) AMD at notes_035 =http://developer.amd.com/cpu/open64 notes_040 = notes_045 =Binaries were compiled on a system with 2x AMD Opteron 6220 chips + 64GB Memory using RHEL 6.1 notes_submit_000 ='numactl' was used to bind copies to the cores. notes_submit_005 =See the configuration file for details. notes_os_000 ='ulimit -s unlimited' was used to set environment stack size notes_os_005 ='ulimit -l 2097152' was used to set environment locked pages in memory limit notes_os_010 = notes_os_015 =Set transparent_hugepage=never as a boot parameter in /boot/grub/menu.lst notes_os_020 =Set kernel/randomize_va_space=0 in /etc/sysctl.conf # RHEL notes_os_025 =cpuspeed stop was used to set the CPU frequency to its maximum. # SLES #notes_os_030 =powersave -f was used to set the CPU frequency to its maximum. %if '%{num_system_huge_pages}' ne '0' notes_os_030 = notes_os_035 =Set vm/nr_hugepages=4000 in /etc/sysctl.conf notes_os_040 =mount -t hugetlbfs nodev /mnt/hugepages %else notes_os_035 = notes_os_040 =Huge pages were not configured for this run. %endif #num_system_huge_pages # ---- End inclusion of '/root/work/cpu2006v1.2/config/AMD1104il-speed-revA.inc' #################################################### # Auto-configured settings based on your .inc file # #################################################### preENV_O64_OMP_SPIN_USER_LOCK = true preENV_O64_OMP_SPIN_COUNT = 800000 # Submit commands for INT and general settings %if '%{cores_per_socket}' eq '16' % if '%{num_sockets}' eq '1' # Settings for 1P with IL-16 preENV_O64_OMP_AFFINITY_MAP = 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15 submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15 bash run.sh % elif '%{num_sockets}' eq '2' # Settings for 2P with IL-16 preENV_O64_OMP_AFFINITY_MAP = 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31 submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31 bash run.sh % elif '%{num_sockets}' eq '4' # Settings for 4P with IL-16 preENV_O64_OMP_AFFINITY_MAP = 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63 submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63 bash run.sh % else #num_sockets % error Please define "num_sockets" to either '1', '2' or '4' in %{inc_file_name}. This error was generated % endif #num_sockets %elif '%{cores_per_socket}' eq '12' % if '%{num_sockets}' eq '1' # Settings for 1P with IL-12 preENV_O64_OMP_AFFINITY_MAP = 0,1,2,3,4,5,6,7,8,9,10,11 submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11 bash run.sh % elif '%{num_sockets}' eq '2' # Settings for 2P with IL-12 preENV_O64_OMP_AFFINITY_MAP = 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23 submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23 bash run.sh % elif '%{num_sockets}' eq '4' # Settings for 4P with IL-12 preENV_O64_OMP_AFFINITY_MAP = 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47 submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47 bash run.sh % else #num_sockets % error Please define "num_sockets" to either '1', '2' or '4' in %{inc_file_name}. This error was generated % endif #num_sockets %elif '%{cores_per_socket}' eq '8' % if '%{num_sockets}' eq '1' # Settings for 1P with IL-8 preENV_O64_OMP_AFFINITY_MAP = 0,1,2,3,4,5,6,7 submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7 bash run.sh % elif '%{num_sockets}' eq '2' # Settings for 2P with IL-8 preENV_O64_OMP_AFFINITY_MAP = 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15 submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15 bash run.sh % elif '%{num_sockets}' eq '4' # Settings for 4P with IL-8 preENV_O64_OMP_AFFINITY_MAP = 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31 submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31 bash run.sh % else #num_sockets % error Please define "num_sockets" to either '1', '2' or '4' in %{inc_file_name}. This error was generated % endif #num_sockets %elif '%{cores_per_socket}' eq '4' % if '%{num_sockets}' eq '1' # Settings for 1P with IL-4 preENV_O64_OMP_AFFINITY_MAP = 0,1,2,3 submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3 bash run.sh % elif '%{num_sockets}' eq '2' # Settings for 2P with IL-4 preENV_O64_OMP_AFFINITY_MAP = 0,1,2,3,4,5,6,7 submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7 bash run.sh % elif '%{num_sockets}' eq '4' # Settings for 4P with IL-4 preENV_O64_OMP_AFFINITY_MAP = 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15 submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15 bash run.sh % else #num_sockets % error Please define "num_sockets" to either '1', '2' or '4' in %{inc_file_name}. This error was generated % endif #num_sockets %else #cores_per_socket % error Please define "cores_per_socket" to either '4', '8', '12' or '16' in %{inc_file_name}. This error was generated %endif #cores_per_socket ################################ # End auto-configured settings # ################################ ########################## Software Info ############################ fp=default=default=default: sw_base_ptrsize = 64-bit sw_peak_ptrsize = 32/64-bit sw_avail = Jul-2011 sw_compiler000 = C/C++/Fortran: Version 4.2.5.2 of x86 Open64 sw_compiler001 = Compiler Suite (from AMD) sw_other = None int=default=default=default: sw_base_ptrsize = 32/64-bit sw_peak_ptrsize = 32/64-bit sw_avail = Jul-2011 sw_compiler = C/C++: Version 4.2.5.2 of x86 Open64 Compiler Suite (from AMD) sw_other = SmartHeap 10.0 32-bit Library for Linux default=default=default=default: ##################################################################### # Compiler selection # # CC = opencc -march=bdver1 CXX = openCC -march=bdver1 FC = openf95 -march=bdver1 ##################################################################### # Portability for default 64-bit code generation ##################################################################### default=default=default: PORTABILITY = -DSPEC_CPU_LP64 ##################################################################### # INT Portability ##################################################################### 400.perlbench=default=default: CPORTABILITY = -DSPEC_CPU_LINUX_X64 403.gcc=peak=default: PORTABILITY = 429.mcf=peak=default: PORTABILITY = 462.libquantum=default=default: CPORTABILITY = -DSPEC_CPU_LINUX 471.omnetpp=default=default: # Needed to avoid -DSPEC_CPU_LP64 on -m32 C++ base codes PORTABILITY = 473.astar=default=default: # Needed to avoid -DSPEC_CPU_LP64 on -m32 C++ base codes PORTABILITY = 483.xalancbmk=default=default: CXXPORTABILITY = -DSPEC_CPU_LINUX # Needed to avoid -DSPEC_CPU_LP64 on -m32 C++ base codes PORTABILITY = ################################################################# # # # SPECint Tuning & Notes # # # ################################################################# int=base: COPTIMIZE = -Ofast -CG:local_sched_alg=1:p2align=0 -INLINE:aggressive=on -IPA:plimit=8000:small_pu=100 -HP:bdt=2m:heap=2m -LNO:prefetch=2 CXXOPTIMIZE = -Ofast -m32 -INLINE:aggressive=on -CG:cmp_peep=on -D__OPEN64_FAST_SET EXTRA_CXXLIBS = -L$(SMARTHEAP_DIR) -lsmartheap sw_base_ptrsize = 32/64-bit sw_peak_ptrsize = 32/64-bit ##################################################################### # INT Peak Tuning ##################################################################### int=peak=default: COPTIMIZE = -Ofast -CG:local_sched_alg=1 -HP:bdt=2m:heap=2m CXXOPTIMIZE = -Ofast -m32 -HP:bdt=2m:heap=2m -D__OPEN64_FAST_SET PASS1_CFLAGS = -fb_create fbdata PASS1_CXXFLAGS = -fb_create fbdata PASS1_LDFLAGS = -fb_create fbdata PASS2_CFLAGS = -fb_opt fbdata PASS2_CXXFLAGS = -fb_opt fbdata PASS2_LDFLAGS = -fb_opt fbdata 400.perlbench=peak=default: COPTIMIZE = -Ofast -LNO:prefetch=2:opt=0 -IPA:plimit=20000 -OPT:unroll_times_max=8:unroll_size=256:unroll_level=2:keep_ext=on -WOPT:if_conv=0:sib=on -CG:local_sched_alg=1:load_exe=0:unroll_fb_req=on:movext_icmp=off -HP:bdt=2m:heap=2m 401.bzip2=peak=default: COPTIMIZE = -O3 -LNO:prefetch=2:pf2=0 -OPT:alias=disjoint:goto=off -CG:local_sched_alg=1 -HP:bdt=2m:heap=2m 403.gcc=peak=default: COPTIMIZE = -Ofast -LNO:trip_count=256 -CG:cmp_peep=on:pre_minreg_level=2 -m32 -HP:bdt=2m:heap=2m -GRA:unspill=on -IPA:small_pu=200 -WOPT:sib=on 429.mcf=peak=default: COPTIMIZE = -O3 -OPT:unroll_times_max=5 -ipa -INLINE:aggressive=on -CG:gcm=off -GRA:prioritize_by_density=on -m32 -HP:bdt=2m:heap=2m feedback = 0 445.gobmk=peak=default: COPTIMIZE = -Ofast -OPT:unroll_size=256:unroll_times_max=8:keep_ext=on -IPA:plimit=750:min_hotness=300:pu_reorder=1 -LNO:ignore_feedback=off -WOPT:if_conv=2 -HP:bdt=2m:heap=2m 456.hmmer=peak=default: COPTIMIZE = -Ofast -LNO:prefetch=2 -OPT:alias=disjoint:unroll_times_max=16:unroll_size=512:unroll_level=2:keep_ext=on -CG:cflow=0:cmp_peep=on:pre_local_sched=off -HP:bdt=2m:heap=2m 458.sjeng=peak=default: COPTIMIZE = -Ofast -CG:ptr_load_use=0:divrem_opt=on:movext_icmp=off:locs_best=on:p2align=1 -LNO:full_unroll=10 -IPA:pu_reorder=2 -HP:bdt=2m:heap=2m -WOPT:sib=on 462.libquantum=peak=default: COPTIMIZE = -Ofast -OPT:unroll_size=512:unroll_times_max=8 -LNO:prefetch=2:pf2=0 -CG:local_sched_alg=1 -INLINE:aggressive=on -IPA:plimit=8000:small_pu=100 -HP:bdt=2m:heap=2m,limit=450 -apo feedback = 0 464.h264ref=peak=default: COPTIMIZE = -O3 -OPT:unroll_size=256:unroll_times_max=2 -IPA:plimit=20000 -OPT:alias=disjoint -CG:ptr_load_use=0:local_sched_alg=1 -HP:bdt=2m:heap=2m 471.omnetpp=peak=default: CXXOPTIMIZE = -Ofast -D__OPEN64_FAST_SET -CG:gcm=off -INLINE:aggressive=on -WOPT:if_conv=0:sib=on -m32 -HP:bdt=2m:heap=2m feedback = 0 473.astar=peak=default: CXXOPTIMIZE = -Ofast -TENV:frame_pointer=off -WOPT:if_conv=0:sib=on -CG:divrem_opt=on:p2align=0 -GRA:optimize_boundary=on -OPT:alias=disjoint -INLINE:aggressive=on -IPA:small_pu=3000:plimit=3000 -m32 -HP:bdt=2m:heap=2m 483.xalancbmk=peak=default: CXXOPTIMIZE = -Ofast -LNO:prefetch=2 -OPT:unroll_size=512:unroll_times_max=8 -D__OPEN64_FAST_SET -INLINE:aggressive=on -m32 -CG:cmp_peep=on:local_sched=off:p2align=0 -GRA:unspill=on -TENV:frame_pointer=off -fno-emit-exceptions EXTRA_CXXLIBS = -L$(SMARTHEAP_DIR) -lsmartheap feedback = 0 ##################################################################### # FP Portability ##################################################################### 436.cactusADM=default=default: FPORTABILITY = -fno-second-underscore 450.soplex=peak=default: PORTABILITY = 481.wrf=default=default: PORTABILITY = -DSPEC_CPU_LP64 -DSPEC_CPU_LINUX -DSPEC_CPU_CASE_FLAG FPORTABILITY = -fno-second-underscore ################################################################# # # # SPECfp Tuning # # # ################################################################# fp=base: COPTIMIZE = -Ofast -HP:bdt=2m:heap=2m -apo -mso -OPT:alias=restricted:malloc_alg=2 -LNO:parallel_overhead=10000 CXXOPTIMIZE = -Ofast -static -CG:load_exe=0:p2align=0 -INLINE:aggressive=on -HP:bdt=2m:heap=2m -D__OPEN64_FAST_SET FOPTIMIZE = -Ofast -LNO:blocking=off:fusion_peeling_limit=0:parallel_overhead=10000 -OPT:rsqrt=2:unroll_size=256 -HP:bdt=2m:heap=2m -apo sw_base_ptrsize = 64-bit sw_peak_ptrsize = 32/64-bit #submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63 bash run.sh %if '%{cores_per_socket}' eq '16' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,4,6,8,10,12,14 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,8,10,16,18,24,26 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '12' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,4,6,8,10 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,6,8,12,14,18,20 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '8' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,4,6 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,4,6,8,10,12,14 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '4' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %endif #cores_per_socket ##################################################################### # FP Peak Tuning ##################################################################### fp=peak=default: COPTIMIZE = -Ofast -HP:bdt=2m:heap=2m CXXOPTIMIZE = -Ofast -INLINE:aggressive=on -HP:bdt=2m:heap=2m -D__OPEN64_FAST_SET FOPTIMIZE = -Ofast -HP:bdt=2m:heap=2m PASS1_CFLAGS = -fb_create fbdata PASS2_CFLAGS = -fb_opt fbdata PASS1_CXXFLAGS = -fb_create fbdata PASS2_CXXFLAGS = -fb_opt fbdata PASS1_FFLAGS = -fb_create fbdata PASS2_FFLAGS = -fb_opt fbdata PASS1_LDFLAGS = -fb_create fbdata PASS2_LDFLAGS = -fb_opt fbdata submit = echo "$command" > run.sh ; numactl -l --physcpubind=2 bash run.sh 410.bwaves=peak=default: FOPTIMIZE = -Ofast -apo -OPT:Ofast:treeheight=on -LNO:blocking=off:prefetch=2:pf2=0:prefetch_ahead=3:ignore_feedback=off:fu=4:loop_model_simd=on:simd_rm_unity_remainder=on -WOPT:aggstr=0 -HP:bdt=2m:heap=2m -CG:cmp_peep=on:p2align=0 #submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63 bash run.sh %if '%{cores_per_socket}' eq '16' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '12' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,4,6,8,10 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,6,8,12,14,18,20 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '8' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '4' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %endif #cores_per_socket 416.gamess=peak=default: FOPTIMIZE = -O3 -LNO:fu=6:blocking=0:simd=0 -OPT:Ofast:ro=3:unroll_size=256:unroll_times_max=2 -CG:local_sched_alg=1 -HP:bdt=2m:heap=2m -WOPT:sib=on 433.milc=peak=default: COPTIMIZE = -Ofast -CG:movnti=1:locs_best=on -HP:bdt=2m:heap=2m -IPA:plimit=7000:callee_limit=1200 -OPT:struct_array_copy=2 -OPT:alias=field_sensitive feedback = 0 434.zeusmp=peak=default: FOPTIMIZE = -Ofast -apo -LNO:blocking=off:interchange=off:fusion_peeling_limit=0 -OPT:treeheight=on:unroll_size=256 -CG:cmp_peep=on:compute_to=on -GRA:prioritize_by_density=on -HP:bdt=2m:heap=2m feedback = 0 #submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63 bash run.sh %if '%{cores_per_socket}' eq '16' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,4,6,8,10,12,14 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,4,6,8,10,12,14,16,18,20,22 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '12' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '8' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '4' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %endif #cores_per_socket 435.gromacs=peak=default: FOPTIMIZE = -Ofast -OPT:rsqrt=2 -HP:bdt=2m:heap=2m COPTIMIZE = $(FOPTIMIZE) 436.cactusADM=peak=default: FOPTIMIZE = -Ofast -LNO:blocking=off:prefetch=2 -HP:bdt=2m:heap=2m -CG:locs_shallow_depth=1:load_exe=0 -WOPT:sib=on -apo COPTIMIZE = $(FOPTIMIZE) #submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63 bash run.sh %if '%{cores_per_socket}' eq '16' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '12' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '8' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '4' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %endif #cores_per_socket 437.leslie3d=peak=default: FOPTIMIZE = -Ofast -LNO:prefetch=2:blocking=off -CG:interior_ptrs=on -OPT:unroll_size=256 -GRA:prioritize_by_density=on -HP:bdt=2m:heap=2m feedback = 0 444.namd=peak=default: CXXOPTIMIZE = -Ofast -LNO:ignore_feedback=off -CG:local_sched_alg=2:load_exe=0 -OPT:unroll_size=256 -fno-exceptions -HP:bdt=2m:heap=2m 447.dealII=peak=default: CXXOPTIMIZE = -Ofast -LNO:simd=0 -D__OPEN64_FAST_SET -static -INLINE:aggressive=on -OPT:alias=disjoint:unroll_times_max=8:unroll_size=256:unroll_level=2 -HP:bdt=2m:heap=2m feedback = 0 450.soplex=peak=default: CXXOPTIMIZE = -O3 -INLINE:aggressive=on -OPT:RO=1:IEEE_arith=3:IEEE_NaN_Inf=off:fold_unsigned_relops=on -fno-exceptions -CG:p2align=0 -m32 -HP:bdt=2m:heap=2m -WOPT:sib=on 453.povray=peak=default: CXXOPTIMIZE = -Ofast -CG:pre_local_sched=off -INLINE:aggressive=on -HP:bdt=2m:heap=2m -OPT:transform=2:alias=disjoint -WOPT:aggcm=0 454.calculix=peak=default: FOPTIMIZE = -Ofast -OPT:unroll_size=256 -GRA:optimize_boundary=on -HP:bdt=2m:heap=2m COPTIMIZE = $(FOPTIMIZE) feedback = 0 459.GemsFDTD=peak=default: FOPTIMIZE = -Ofast -OPT:unroll_size=0 -LNO:fission=2 -CG:load_exe=0:local_sched_alg=2 -HP -apo feedback = 0 #submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63 bash run.sh %if '%{cores_per_socket}' eq '16' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,4,6 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,4,6 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '12' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,4 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,4 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '8' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '4' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %endif #cores_per_socket 465.tonto=peak=default: FOPTIMIZE = -Ofast -OPT:alias=no_f90_pointer_alias -LNO:blocking=off -CG:load_exe=1:local_sched_alg=1 -IPA:plimit=525 -HP feedback = 0 470.lbm=peak=default: COPTIMIZE = -Ofast -mso -apo -CG:sse_cse_regs=0 -LNO:prefetch_ahead=4 -CG:locs_shallow_depth=1:cmp_peep=on:compute_to=on -OPT:unroll_times_max=8:unroll_size=256:unroll_level=2:keep_ext=on:alias=restricted -m3dnow -IPA:inline=off feedback = 0 #submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63 bash run.sh %if '%{cores_per_socket}' eq '16' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '12' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '8' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '4' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %endif #cores_per_socket 481.wrf=peak=default: FOPTIMIZE = -Ofast -OPT:unroll_size=256 -LNO:blocking=off -LANG:copyinout=off -IPA:callee_limit=5000 -GRA:prioritize_by_density=on -CG:load_exe=1 -HP -WOPT:sib=on -apo COPTIMIZE = $(FOPTIMIZE) feedback = 0 #submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63 bash run.sh %if '%{cores_per_socket}' eq '16' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,8,10 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,8,10,16,18,24,26 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '12' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,4,6,8,10 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,6,8,12,14,18,20 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '8' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,4,6 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,2,4,6,8,10 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %elif '%{cores_per_socket}' eq '4' % if '%{num_sockets}' eq '1' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3 bash run.sh % elif '%{num_sockets}' eq '2' submit = echo "$command" > run.sh ; numactl -l --physcpubind=0,1,2,3,4,5,6,7 bash run.sh % elif '%{num_sockets}' eq '4' % endif #num_sockets %endif #cores_per_socket 482.sphinx3=peak=default: COPTIMIZE = -Ofast -LNO:loop_model_simd=on:simd_rm_unity_remainder=on -OPT:malloc_alg=2 -CG:cmp_peep=on:local_sched_alg=2:use_incdec=off -INLINE:aggressive=on -WOPT:sib=on -HP # The following settings were obtained by running 'specperl $[top]/Docs/sysinfo -f -p' default: # Sysinfo program /root/work/cpu2006v1.2/Docs/sysinfo # $Rev: 6775 $ $Date:: 2011-08-16 #$ 8787f7622badcf24e01c368b1db4377c # running on smval2p01 Fri Jan 13 23:43:19 2012 # # 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 /proc/cpuinfo # model name : AMD Opteron(tm) Processor 6204 # 2 "physical id"s (chips) # 8 "processors" # cores, siblings (Caution: counting these is hw and system dependent. The # following excerpts from /proc/cpuinfo might not be reliable. Use with # caution.) # cpu cores : 2 # siblings : 4 # physical 0: cores 0 1 # physical 1: cores 0 1 # cache size : 2048 KB # # From /proc/meminfo # MemTotal: 65956296 kB # HugePages_Total: 4000 # Hugepagesize: 2048 kB # # /usr/bin/lsb_release -d # Red Hat Enterprise Linux Server release 6.1 (Santiago) # # From /etc/*release* /etc/*version* # redhat-release: Red Hat Enterprise Linux Server release 6.1 (Santiago) # system-release: Red Hat Enterprise Linux Server release 6.1 (Santiago) # system-release-cpe: cpe:/o:redhat:enterprise_linux:6server:ga:server # # uname -a: # Linux smval2p01 2.6.32-131.0.15.el6.x86_64 #1 SMP Tue May 10 15:42:40 EDT # 2011 x86_64 x86_64 x86_64 GNU/Linux # # run-level 3 Jan 13 13:53 # # SPEC is set to: /root/work/cpu2006v1.2 # Filesystem Type Size Used Avail Use% Mounted on # /dev/sda2 ext3 189G 42G 137G 24% /root/work # # (End of data from sysinfo program) # hw_cpu_name = AMD Opteron 6204 # hw_disk = 189 GB add more disk info here # hw_memory001 = 62.901 GB fixme: If using DDR3, format is: # hw_memory002 = 'N GB (M x N GB nRxn PCn-nnnnnR-n, ECC)' # hw_nchips = 2 # prepared_by = root (is never output, only tags rawfile) # sw_file = ext3 # sw_os001 = Red Hat Enterprise Linux Server release 6.1 (Santiago) # sw_os002 = 2.6.32-131.0.15.el6.x86_64 # sw_state = Run level 3 (add definition here) # 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. 481.wrf: # The following setting was inserted automatically as a result of # post-run basepeak application. basepeak = 1 437.leslie3d: # 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 =LD_LIBRARY_PATH = "/root/work/cpu2006v1.2/amd1104-speed-libs-revA/32:/root/work/cpu2006v1.2/amd1104-speed-libs-revA/64" notes_010 =O64_OMP_AFFINITY_MAP = "0,1,2,3,4,5,6,7" notes_015 =O64_OMP_SPIN_COUNT = "800000" notes_020 =O64_OMP_SPIN_USER_LOCK = "true" notes_025 =