# Invocation command line: # /home/omp2001/bin/runspec -a validate -c gh4985_4P_v2.cfg -n 3 --reportable -i ref -T base,peak medium ############################################################################ ############################################################################ # # configuration file for SPEC OMPM2001 on AMD Tyan s4985 # Aug-2007 # VENDOR = AMD action = validate tune = base ext = gh_2000_1800_psc30_667_081107 input = ref env_vars = 1 reportable = 1 #mean_anyway = 1 output_format = asc,pdf,config teeout = yes teerunout = yes check_md5 = 1 ###### Compiler used ################# default=default: CC=/opt/pathscale/bin/pathcc -gnu3 -mp FC=/opt/pathscale/bin/pathf90 -mp ######## Portability Flags and Environment variables ################## 318.galgel_m=default=default=default: FPORTABILITY = -fixedform -col80 330.art_m=default=default=default: EXTRA_CFLAGS = -DINTS_PER_CACHELINE=16 -DDBLS_PER_CACHELINE=8 notes00108 = Extra art Base Flags notes00109 = 330.art_m (extra): -DINTS_PER_CACHELINE=16 -DDBLS_PER_CACHELINE=8 default=default=default=default: notes41000= Portablility flags: notes41001 = 318.galgel_m : -fixedform -col80 notes41002= notes41003= Base and Peak User Environment (bash): ENV_OMP_NUM_THREADS=16 ENV_OMP_DYNAMIC=FALSE ENV_PSC_OMP_AFFINITY=TRUE ENV_PSC_OMP_STATIC_FAIR=TRUE ENV_PSC_OMP_THREAD_SPIN=1000000000 notes41004= export OMP_NUM_THREADS=16 notes41005= export OMP_DYNAMIC=FALSE notes41006= export PSC_OMP_AFFINITY=TRUE notes41007= export PSC_OMP_STATIC_FAIR=TRUE notes41008= export PSC_OMP_THREAD_SPIN=1000000000 notes41010= ulimit -s unlimited notes41011= notes41012= Default BIOS settings used, with DRAM bank interleave set to auto #notes41013= AMD PowerNow!(TM) disabled by use of command "powersave -f" #################### SPEC OMPM2001 src.alt ############################ #330.art_m=default=default=default: #srcalt = purdue1 #################### Baseline Optimization Flags ###################### medium=base=default=default: COPTIMIZE = -Ofast -mcpu=opteron -mcmodel=medium FOPTIMIZE = -Ofast -mcpu=opteron -OPT:early_mp=on -mcmodel=medium #F77OPTIMIZE = -mp -Ofast -OPT:early_mp=on -mcmodel=medium ONESTEP=yes default=default=default=default: notes120 = notes121 = Base tuning: notes123 = Fortran : pathf90 -mp -Ofast -OPT:early_mp=on -mcmodel=medium notes122 = C : pathcc -mp -Ofast -mcmodel=medium notes124 = ONESTEP=yes ######################### Peak Flags ############################# default=peak=default=default: ONESTEP = yes notes300_0 = #notes300_1 = Peak tuning: #notes300_2 = ONESTEP=yes for all peak tests. notes300_3 = 310.wupwise_m=peak=default=default: srcalt=ompl.32 #basepeak=yes FOPTIMIZE = -mp -Ofast -msse3 -LNO:prefetch_ahead=5:prefetch=3 -OPT:unroll_times_max=8:unroll_size=128:IEEE_NaN_Inf=off:ro=3 -IPA:linear=on:plimit=50000:callee_limit=5000 -INLINE:aggressive=on F77OPTIMIZE = -mp -Ofast -msse3 -LNO:prefetch_ahead=5:prefetch=3 -OPT:unroll_times_max=8:unroll_size=128:IEEE_NaN_Inf=off:ro=3 -IPA:linear=on:plimit=50000:callee_limit=5000 -INLINE:aggressive=on #notes205= 310.wupwise_m: basepeak=yes notes205= 310.wupwise_m: -mp -Ofast -msse3 -LNO:prefetch_ahead=5:prefetch=3 notes206= -OPT:unroll_times_max=8:unroll_size=128:IEEE_NaN_Inf=off:ro=3 notes207= -IPA:linear=on:plimit=50000:callee_limit=5000 notes208: -INLINE:aggressive=on 312.swim_m=peak=default=default: srcalt=ompl.32 FOPTIMIZE = -mp -O3 -OPT:Ofast -m3dnow -mcmodel=medium -LNO:simd=0 F77OPTIMIZE = -mp -O3 -OPT:Ofast -m3dnow -mcmodel=medium -LNO:simd=0 notes210= 312.swim_m: -mp -O3 -OPT:Ofast -m3dnow -mcmodel=medium -LNO:simd=0 314.mgrid_m=peak=default=default: srcalt=ompl basepeak=yes #FOPTIMIZE = -mp -Ofast -mcpu=opteron -OPT:early_mp=on -mcmodel=medium #FOPTIMIZE = -mp -Ofast -CG:gcm=off:cflow=off -LNO:fusion=2:prefetch=0 -msse3 -OPT:early_mp=on:unroll_times_max=8:unroll_size=256:ro=3 -mcmodel=medium #F77OPTIMIZE = -mp -Ofast -CG:gcm=off:cflow=off -LNO:fusion=2:prefetch=0 -msse3 -OPT:early_mp=on:unroll_times_max=8:unroll_size=256:ro=3 -mcmodel=medium notes215= 314.mgrid_m: basepeak=yes #notes215= 314.mgrid_m: -mp -Ofast -mcpu=opteron -OPT:early_mp=on -mcmodel=medium #notes215= 314.mgrid_m: -mp -Ofast -CG:gcm=off:cflow=off -LNO:fusion=2:prefetch=0 -msse3 #notes216= -OPT:early_mp=on:unroll_times_max=8:unroll_size=256:ro=3 #notes217= -mcmodel=medium 316.applu_m=peak=default=default: srcalt=ompl basepeak=yes #FOPTIMIZE = -mp -Ofast -mcmodel=medium -CG:local_fwd_sched=on -LNO:fusion=2:fission=2:full_unroll_size=10000:prefetch=3 -TENV:X=3 #F77OPTIMIZE = -mp -Ofast -mcmodel=medium -CG:local_fwd_sched=on -LNO:fusion=2:fission=2:full_unroll_size=10000:prefetch=3 -TENV:X=3 notes220= 316.applu_m: basepeak=yes #notes220= 316.applu_m: -mp -Ofast -mcmodel=medium -CG:local_fwd_sched=on #notes221= -LNO:fusion=2:fission=2:full_unroll_size=10000:prefetch=3 #notes222= -TENV:X=3 318.galgel_m=peak=default=default: basepeak=yes #FOPTIMIZE = -mp -Ofast -OPT:early_mp=on -OPT:fast_complex=on #notes225= 318.galgel_m: -mp -Ofast -OPT:early_mp=on -OPT:fast_complex=on notes225= 318.galgel_m: basepeak=yes 320.equake_m=peak=default=default: srcalt=ompl.32 COPTIMIZE = -Ofast -mp -OPT:alias=disjoint notes230= 320.equake_m: -Ofast -mp -OPT:alias=disjoint 324.apsi_m=peak=default=default: srcalt=ompl FOPTIMIZE = -mp -O3 -OPT:Ofast -TENV:X=4 -LNO:fusion=2:prefetch=0 F77OPTIMIZE = -mp -O3 -OPT:Ofast -TENV:x=4 -LNO:fusion=2:prefetch=0 notes235= 324.apsi_m: -mp -O3 -OPT:Ofast -TENV:x=4 -LNO:fusion=2:prefetch=0 326.gafort_m=peak=default=default: basepeak=yes #F77OPTIMIZE = -Ofast -OPT:early_mp=on -mp -LNO:vintr=0 #FOPTIMIZE = -Ofast -OPT:early_mp=on -mp -LNO:vintr=0 notes240= 326.gafort_m: basepeak=yes #notes240= 326.gafort_m: -Ofast -OPT:early_mp=on -mp -LNO:vintr=0 328.fma3d_m=peak=default=default: basepeak=yes notes245= 328.fma3d_m: basepeak=yes 330.art_m=peak=default=default: basepeak=yes notes250= 330.art_m: basepeak=yes 332.ammp_m=peak=default=default: use_submit_for_speed=1 submit= PSC_OMP_GUIDED_CHUNK_MAX=700 $command COPTIMIZE = -mp -Ofast -mcpu=opteron -mcmodel=medium #COPTIMIZE = -mp -O3 -OPT:alias=disjoint:unroll_times_max=8:Ofast:ro=3 -fno-math-errno -TENV:X=4 -LNO:prefetch=0 #basepeak=yes notes255= 332.ammp_m: -mp -Ofast -mcpu=opteron -mcmodel=medium #notes255= 332.ammp_m: -mp -O3 -OPT:alias=disjoint:unroll_times_max=8:Ofast:ro=3 #notes256= -fno-math-errno -TENV:X=4 -LNO:prefetch=0 #notes257= use_submit_for_speed=1 #notes258= submit= PSC_OMP_GUIDED_CHUNK_MAX=700 $command # # machine configuration # default=default=default=default: hw_vendor = Advanced Micro Devices hw_model = Tyan s4985 (Opteron(TM) Barcelona) hw_cpu = Opteron(TM) Bareclona hw_cpu_mhz = 2000 hw_fpu = Integrated hw_ncpu = 16 cores, 4 chips, 4 cores/chip hw_ncpuorder= 4 hw_pcache = 64KBI + 64KBD on chip per core hw_scache = 512KB (I+D) on chip per core hw_tcache = 2MB (I+D) on chip per chip hw_ocache = None hw_memory = 16x1GB, PC2-5300 CL5 hw_disk = 150GB 7200 rpm SATA hw_avail = Sep-2007 hw_other = None sw_os = SLES10 SP1 64bit sw_compiler = QLogic PathScale Compiler Suite ver. 3.0 sw_Kernel_Extensions = None sw_file = Linux/Ext3 sw_state = Multi-User sw_avail = Jul-2007 sw_parallel = OpenMP license_num = HPG0017 tester_name = AMD test_date = Aug-2007 test_site = Austin TX company_name= Advanced Micro Devices machine_name= gh4895