# Invocation command line: # /root/cpu2006-1.1/bin/runspec -c amd1104il-rate-revA-Dell.cfg --rate -T all int -o asc,html --nopreenv --note-preenv # output_root was not used for this run ############################################################################ # AMD1104 SPEC CPU2006 V1.1 Rate Configuration File for 64-bit Linux # # # CPU2006 Version 1.1 # 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: No # # Important! Please run with your stack size set to 'unlimited'. # Failure to do so may cause some benchmarks 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 AMD1104il-rate-revX.inc file.) # This line should be commented out during builds. # ##################################################################### ext = amd1104-rate-revA ignore_errors = no tune = base,peak output_format = all flagsurl000 = http://www.spec.org/cpu2006/flags/x86-open64-425-flags-rate-revA.xml $[top]/amd1104-platform-rate-revA.xml size = test,train,ref check_md5 = yes reportable = yes env_vars = no teeout = yes mean_anyway = yes teerunout = yes post_setup = sync #################################################################### # Include file containing the commonly changed fields #################################################################### %define 2nd_inc amd1104il-bind.inc %define inc_file_name AMD1104il-rate-revA.inc #include: %{inc_file_name} # ----- Begin inclusion of 'AMD1104il-rate-revA.inc' ############################################################################ ##################################################################### # Paths ##################################################################### # Set location of run time libraries for runs. preENV_LD_LIBRARY_PATH = $[top]/amd1104-rate-libs-revA/32:$[top]/amd1104-rate-libs-revA/64 # Set this path to your SmartHeap Library for builds. SMARTHEAP_DIR = /root/work/libraries/SmartHeap-10/lib #################################################################### # Tester information #################################################################### license_num = 55 prepared_by = Dell Inc. tester = Dell Inc. test_sponsor = Dell Inc. hw_vendor = Dell Inc. hw_model000 = PowerEdge R415 hw_model001 = (AMD Opteron 4234, 3.10 GHz) #################################################################### # Hardware information #################################################################### hw_avail = Jan-2012 hw_cpu_name = AMD Opteron 4234 hw_cpu_char = AMD Turbo CORE technology up to 3.50 GHz hw_cpu_mhz = 3100 hw_fpu = Integrated hw_nchips = 2 hw_ncores = 12 hw_ncoresperchip = 6 hw_nthreadspercore = 1 hw_ncpuorder = 1,2 chips hw_pcache000 = 192 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 = 6 MB I+D on chip per chip, 2 MB shared / 2 cores hw_tcache = 8 MB I+D on chip per chip hw_ocache = None hw_disk = 2 x 73 GB SAS, 15000 RPM hw_memory000 = 32 GB (4 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_015 =The x86 Open64 Compiler Suite is only available from (and supported by) AMD at notes_020 =http://developer.amd.com/cpu/open64 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 =Large pages were not enabled for this run notes_os_015 =Binaries were compiled on a system with 2x AMD Opteron 6276 chips + 128GB Memory using RHEL 6.1 %if '%{num_system_huge_pages}' ne '0' notes_os_020 =Set kernel/randomize_va_space=0 in /etc/sysctl.conf %endif #num_system_huge_pages # ---- End inclusion of '/root/cpu2006-1.1/config/AMD1104il-rate-revA.inc' #include: %{2nd_inc} # ----- Begin inclusion of 'amd1104il-bind.inc' ############################################################################ %define num_sockets 2 %define cores_per_socket 6 %define cores_per_node 6 %define mem_size 32 %define mem_sticks 4 copies = 12 %define num_nodes 2 %ifdef %{autopar} preENV_OMP_NUM_THREADS=6 %endif #autopar bind0 = numactl -m 0 --physcpubind=0 bind1 = numactl -m 1 --physcpubind=1 bind2 = numactl -m 0 --physcpubind=2 bind3 = numactl -m 1 --physcpubind=3 bind4 = numactl -m 0 --physcpubind=4 bind5 = numactl -m 1 --physcpubind=5 bind6 = numactl -m 0 --physcpubind=6 bind7 = numactl -m 1 --physcpubind=7 bind8 = numactl -m 0 --physcpubind=8 bind9 = numactl -m 1 --physcpubind=9 bind10 = numactl -m 0 --physcpubind=10 bind11 = numactl -m 1 --physcpubind=11 # ---- End inclusion of '/root/cpu2006-1.1/config/amd1104il-bind.inc' # Required for SPEC CPU v1.1 and earlier submissions parallel_test = 1 #################################################### # Auto-configured settings based on your .inc file # #################################################### %if '%{GB_mem_per_core}' eq '4' preENV_HUGETLB_LIMIT = 896 %elif '%{GB_mem_per_core}' eq '3' preENV_HUGETLB_LIMIT = 480 %elif '%{GB_mem_per_core}' eq '2' #No huge pages should be configured, so do nothing %endif #GB_mem_per_core ################################ # End auto-configured settings # ################################ submit = echo "$command" > run.sh ; $BIND bash run.sh ############################ Software Info ############################ fp=default=default=default: sw_base_ptrsize = 64-bit sw_peak_ptrsize = 32/64-bit sw_avail = Jul-2011 sw_compiler = x86 Open64 4.2.5.2 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_compiler000 = C/C++: Version 4.2.5.2 of x86 Open64 Compiler sw_compiler001 = 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 -INLINE:aggressive=on -IPA:plimit=8000:small_pu=100 -HP:bd=2m:heap=2m -mso -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:unroll_fb_req=on:movext_icmp=off -HP:bd=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 -mso 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:bd=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 -LNO:full_unroll=10 -IPA:pu_reorder=2 -HP:bd=2m:heap=2m -WOPT:sib=on 462.libquantum=peak=default: COPTIMIZE = -Ofast -mso -OPT:unroll_size=512:unroll_times_max=16 -LNO:prefetch=2:prefetch_ahead=4:pf2=0 -CG:local_sched_alg=1 -INLINE:aggressive=on -IPA:plimit=15000:small_pu=100 -HP:bdt=2m:heap=2m,limit=300 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: basepeak = 1 473.astar=peak=default: CXXOPTIMIZE = -Ofast -TENV:frame_pointer=off -WOPT:if_conv=0:sib=on -CG:divrem_opt=on -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 -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 447.dealII=peak=default: # Needed to avoid -DSPEC_CPU_LP64 since peak is built with -m32 PORTABILITY = 450.soplex=peak=default: # Needed to avoid -DSPEC_CPU_LP64 PORTABILITY = 481.wrf=default=default: PORTABILITY = -DSPEC_CPU_LINUX -DSPEC_CPU_CASE_FLAG -DSPEC_CPU_LP64 FPORTABILITY = -fno-second-underscore ################################################################# # # # SPECfp Tuning # # # ################################################################# fp=base: COPTIMIZE = -Ofast -OPT:malloc_alg=1 -HP:bd=2m:heap=2m -IPA:plimit=8000:small_pu=100 -mso CXXOPTIMIZE = -Ofast -static -CG:load_exe=0 -OPT:malloc_alg=1 -INLINE:aggressive=on -HP:bd=2m:heap=2m -D__OPEN64_FAST_SET FOPTIMIZE = -Ofast -LNO:blocking=off -OPT:rsqrt=2:unroll_size=256 -HP:bd=2m:heap=2m -mso sw_base_ptrsize = 64-bit sw_peak_ptrsize = 32/64-bit ##################################################################### # 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 410.bwaves=peak=default: FOPTIMIZE = -Ofast -OPT:Ofast:treeheight=on -LNO:blocking=off: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 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 -mso feedback = 0 434.zeusmp=peak=default: FOPTIMIZE = -Ofast -LNO:blocking=off:interchange=off -HP:bdt=2m:heap=2m feedback = 0 435.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 -CG:locs_shallow_depth=1:load_exe=0 -WOPT:sib=on COPTIMIZE = $(FOPTIMIZE) 437.leslie3d=peak=default: FOPTIMIZE = -Ofast -CG:pre_minreg_level=2 -LNO:simd=0:fusion=2 -HP:bdt=2m:heap=2m -mso 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 -D__OPEN64_FAST_SET -static -INLINE:aggressive=on -LNO:opt=0:simd=0 -fno-emit-exceptions -m32 -OPT:unroll_times_max=8:unroll_size=256:unroll_level=2 -HP:bdt=2m:heap=2m -GRA:unspill=on -CG:cmp_peep=on:movext_icmp=off -TENV:frame_pointer=off 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 -m32 -HP:bdt=2m:heap=2m -WOPT:sib=on 453.povray=peak=default: CXXOPTIMIZE = -Ofast -CG:pre_local_sched=off -INLINE:aggressive=on -HP:bd=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 feedback = 0 465.tonto=peak=default: FOPTIMIZE = -Ofast -OPT:alias=no_f90_pointer_alias -LNO:blocking=off -CG:load_exe=1 -IPA:plimit=525 -HP:bdt=2m:heap=2m feedback = 0 470.lbm=peak=default: COPTIMIZE = -Ofast -CG:cmp_peep=on -OPT:unroll_times_max=8:unroll_size=256:unroll_level=2:keep_ext=on -HP:bdt=2m:heap=2m -IPA:plimit=8000:small_pu=100 -mso feedback = 0 481.wrf=peak=default: FOPTIMIZE = -Ofast -LNO:blocking=off -LANG:copyinout=off -IPA:callee_limit=5000 -GRA:prioritize_by_density=on -CG:load_exe=1 -HP -WOPT:sib=on COPTIMIZE = $(FOPTIMIZE) feedback = 0 482.sphinx3=peak=default: COPTIMIZE = -Ofast -OPT:malloc_alg=2 -CG:cmp_peep=on:local_sched_alg=2 -INLINE:aggressive=on -LNO:prefetch=2:prefetch_ahead=4 -mso # 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: flagsurl001 = http://www.spec.org/cpu2006/flags/amd1104-platform-rate-revA.xml notes_000 =Environment variables set by runspec before the start of the run: notes_005 =LD_LIBRARY_PATH = "/root/cpu2006-1.1/amd1104-rate-libs-revA/32:/root/cpu2006-1.1/amd1104-rate-libs-revA/64" notes_010 =