# Invocation command line: # /local/home/mcolgrove/ACCELV2/bin/harness/runaccel -c nvhpc_lop_2311 -D --reportable accel # output_root was not used for this run ############################################################################ # Invocation command line: # runaccel -c nvhpc --reportable -T base --define model= accel # # MODEL may be one of the following options: # 1. "acc" - Use OpenACC (default) # 2. "lop" - Use OpenMP with "loop" construct # 3. "tgt" - Use OpenMP with "distribute" construct # 4. "smd" - Use OpenMP with "distribute" and inner "simd" construct # 5. "acp" - Use OpenACC with the "parallel" src.alt # 6. "lops" - Use OpenMP with "loop" construct src.alt (no metadirectives) # 7. "tgts" - Use OpenMP with "distribute" construct src.alt (no metadirectives) # 8. "smds" - Use OpenMP with "distribute" and inner "simd" construct src.alt (no metadirectives) # ############################################################################ # The following setting was inserted automatically as a result of # post-run basepeak application. basepeak = 1 %ifndef %{model} # IF model is not set use acc % define model acc %endif #################################################################### # Tester information #################################################################### license_num = 9045 tester = NVIDIA Corporation test_sponsor = NVIDIA Corporation ###################################################################### # The header section of the config file. Must appear # before any instances of "default=" # # label = how the binaries you generated will be identified # tune = specify "base" or "peak" or "all" label = nvhpc_lop_2311 tune = base output_format = txt,html,pdf teeout = yes teerunout = yes strict_rundir_verify=1 #output_root = /tmp/accel2023 flagsurl000=http://www.spec.org/accel2023/flags/nv2023_flags_v2.xml # Adjust the number of make jobs to use here makeflags=-j 40 #################################################################### # Hardware information #################################################################### #include: SUT.inc # ----- Begin inclusion of 'SUT.inc' ############################################################################ ##### SUT.inc #### ##### Example of required hardware information for the system under test (SUT) #### default: hw_avail = Mar-2023 sw_avail = Nov-2023 hw_cpu_name = CPU Name hw_cpu_max_mhz = 3400 hw_cpu_nominal_mhz = 2000 hw_nchips = 1 hw_ncores = 64 hw_nthreadspercore = 2 hw_ncpuorder = 2 chips hw_pcache = 32 KB I + 48 KB D on chip per core hw_scache = 1280 KB I+D on chip per core hw_tcache = 48 MB I+D on chip per chip hw_ocache = None hw_vendor = Supermciro hw_model = 120GQ-TNRT hw_disk = 1x NVME, 1000 GB, 99999 RPM hw_memory000 = 512 GB (16x 16GB, PC3200 CL3 DDR4) hw_other = None # Accelerator description hw_accel_connect = PCIe 4.0 16x hw_accel_desc = See Notes hw_accel_ecc = Yes hw_accel_model = H100 PCIe 80GB hw_accel_name = Tesla H100 PCIe 80GB hw_accel_type = GPU hw_accel_vendor = NVIDIA sw_accel_driver = NVIDIA UNIX x86_64 Kernel Module 525.60.13 # ---- End inclusion of '/local/home/mcolgrove/ACCELV2/config/SUT.inc' #################################################################### # Software information #################################################################### default: CC = nvc CXX = nvc++ FC = nvfortran CC_VERSION_OPTION = -V CXX_VERSION_OPTION = -V FC_VERSION_OPTION = -V OPTIMIZE = -Ofast -mp=gpu -Mfprelaxed -Mstack_arrays -static-nvidia 403.stencil: PORTABILITY += -DSPEC_NO_NOTHING 457.spC: PORTABILITY += -mcmodel=medium -Wl,--no-relax 459.miniGhost: EXTRA_LDFLAGS += -Mnomain # The following settings were obtained by running the sysinfo_program # 'specperl $[top]/bin/sysinfo' (sysinfo:SHA:ac6f2b38b7dec295b8795065d3a13319ea78b9367f95037a5e6ed5e46e487861) default: notes_plat_sysinfo_000 = notes_plat_sysinfo_005 = Sysinfo program /local/home/mcolgrove/ACCELV2/bin/sysinfo notes_plat_sysinfo_010 = Rev: r6622 of 2021-04-07 b1a7d5f8f71be5aff70a755cad7211a0 notes_plat_sysinfo_015 = running on ice3 Fri Oct 20 17:48:42 2023 notes_plat_sysinfo_020 = notes_plat_sysinfo_025 = SUT (System Under Test) info as seen by some common utilities. notes_plat_sysinfo_030 = For more information on this section, see notes_plat_sysinfo_035 = https://www.spec.org/cpu2017/Docs/config.html#sysinfo notes_plat_sysinfo_040 = notes_plat_sysinfo_045 = From /proc/cpuinfo notes_plat_sysinfo_050 = model name : Intel(R) Xeon(R) Gold 6338 CPU @ 2.00GHz notes_plat_sysinfo_055 = 2 "physical id"s (chips) notes_plat_sysinfo_060 = 128 "processors" notes_plat_sysinfo_065 = cores, siblings (Caution: counting these is hw and system dependent. The following notes_plat_sysinfo_070 = excerpts from /proc/cpuinfo might not be reliable. Use with caution.) notes_plat_sysinfo_075 = cpu cores : 32 notes_plat_sysinfo_080 = siblings : 64 notes_plat_sysinfo_085 = physical 0: cores 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 notes_plat_sysinfo_090 = 25 26 27 28 29 30 31 notes_plat_sysinfo_095 = physical 1: cores 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 notes_plat_sysinfo_100 = 25 26 27 28 29 30 31 notes_plat_sysinfo_105 = notes_plat_sysinfo_110 = From lscpu from util-linux 2.32.1: notes_plat_sysinfo_115 = Architecture: x86_64 notes_plat_sysinfo_120 = CPU op-mode(s): 32-bit, 64-bit notes_plat_sysinfo_125 = Byte Order: Little Endian notes_plat_sysinfo_130 = CPU(s): 128 notes_plat_sysinfo_135 = On-line CPU(s) list: 0-127 notes_plat_sysinfo_140 = Thread(s) per core: 2 notes_plat_sysinfo_145 = Core(s) per socket: 32 notes_plat_sysinfo_150 = Socket(s): 2 notes_plat_sysinfo_155 = NUMA node(s): 2 notes_plat_sysinfo_160 = Vendor ID: GenuineIntel notes_plat_sysinfo_165 = CPU family: 6 notes_plat_sysinfo_170 = Model: 106 notes_plat_sysinfo_175 = Model name: Intel(R) Xeon(R) Gold 6338 CPU @ 2.00GHz notes_plat_sysinfo_180 = Stepping: 6 notes_plat_sysinfo_185 = CPU MHz: 3200.000 notes_plat_sysinfo_190 = CPU max MHz: 3200.0000 notes_plat_sysinfo_195 = CPU min MHz: 800.0000 notes_plat_sysinfo_200 = BogoMIPS: 4000.00 notes_plat_sysinfo_205 = Virtualization: VT-x notes_plat_sysinfo_210 = L1d cache: 48K notes_plat_sysinfo_215 = L1i cache: 32K notes_plat_sysinfo_220 = L2 cache: 1280K notes_plat_sysinfo_225 = L3 cache: 49152K notes_plat_sysinfo_230 = NUMA node0 CPU(s): 0-31,64-95 notes_plat_sysinfo_235 = NUMA node1 CPU(s): 32-63,96-127 notes_plat_sysinfo_240 = Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov notes_plat_sysinfo_245 = pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp notes_plat_sysinfo_250 = lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid notes_plat_sysinfo_255 = aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 notes_plat_sysinfo_260 = xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave notes_plat_sysinfo_265 = avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 invpcid_single ssbd notes_plat_sysinfo_270 = mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad notes_plat_sysinfo_275 = fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq notes_plat_sysinfo_280 = rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw notes_plat_sysinfo_285 = avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total notes_plat_sysinfo_290 = cqm_mbm_local split_lock_detect wbnoinvd dtherm ida arat pln pts avx512vbmi umip pku notes_plat_sysinfo_295 = ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg tme notes_plat_sysinfo_300 = avx512_vpopcntdq la57 rdpid fsrm md_clear pconfig flush_l1d arch_capabilities notes_plat_sysinfo_305 = notes_plat_sysinfo_310 = /proc/cpuinfo cache data notes_plat_sysinfo_315 = cache size : 49152 KB notes_plat_sysinfo_320 = notes_plat_sysinfo_325 = From numactl --hardware notes_plat_sysinfo_330 = WARNING: a numactl 'node' might or might not correspond to a physical chip. notes_plat_sysinfo_335 = available: 2 nodes (0-1) notes_plat_sysinfo_340 = node 0 cpus: 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 notes_plat_sysinfo_345 = 28 29 30 31 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 notes_plat_sysinfo_350 = 89 90 91 92 93 94 95 notes_plat_sysinfo_355 = node 0 size: 257616 MB notes_plat_sysinfo_360 = node 0 free: 126167 MB notes_plat_sysinfo_365 = node 1 cpus: 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 notes_plat_sysinfo_370 = 57 58 59 60 61 62 63 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 notes_plat_sysinfo_375 = 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 notes_plat_sysinfo_380 = node 1 size: 257985 MB notes_plat_sysinfo_385 = node 1 free: 227371 MB notes_plat_sysinfo_390 = node distances: notes_plat_sysinfo_395 = node 0 1 notes_plat_sysinfo_400 = 0: 10 20 notes_plat_sysinfo_405 = 1: 20 10 notes_plat_sysinfo_410 = notes_plat_sysinfo_415 = From /proc/meminfo notes_plat_sysinfo_420 = MemTotal: 527975808 kB notes_plat_sysinfo_425 = HugePages_Total: 0 notes_plat_sysinfo_430 = Hugepagesize: 2048 kB notes_plat_sysinfo_435 = notes_plat_sysinfo_440 = /sbin/tuned-adm active notes_plat_sysinfo_445 = Current active profile: throughput-performance notes_plat_sysinfo_450 = notes_plat_sysinfo_455 = /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor has notes_plat_sysinfo_460 = performance notes_plat_sysinfo_465 = notes_plat_sysinfo_470 = /usr/bin/lsb_release -d notes_plat_sysinfo_475 = Rocky Linux release 8.8 (Green Obsidian) notes_plat_sysinfo_480 = notes_plat_sysinfo_485 = From /etc/*release* /etc/*version* notes_plat_sysinfo_490 = centos-release: Rocky Linux release 8.8 (Green Obsidian) notes_plat_sysinfo_495 = os-release: notes_plat_sysinfo_500 = NAME="Rocky Linux" notes_plat_sysinfo_505 = VERSION="8.8 (Green Obsidian)" notes_plat_sysinfo_510 = ID="rocky" notes_plat_sysinfo_515 = ID_LIKE="rhel centos fedora" notes_plat_sysinfo_520 = VERSION_ID="8.8" notes_plat_sysinfo_525 = PLATFORM_ID="platform:el8" notes_plat_sysinfo_530 = PRETTY_NAME="Rocky Linux 8.8 (Green Obsidian)" notes_plat_sysinfo_535 = ANSI_COLOR="0;32" notes_plat_sysinfo_540 = redhat-release: Rocky Linux release 8.8 (Green Obsidian) notes_plat_sysinfo_545 = rocky-release: Rocky Linux release 8.8 (Green Obsidian) notes_plat_sysinfo_550 = rocky-release-upstream: Derived from Red Hat Enterprise Linux 8.8 notes_plat_sysinfo_555 = system-release: Rocky Linux release 8.8 (Green Obsidian) notes_plat_sysinfo_560 = system-release-cpe: cpe:/o:rocky:rocky:8:GA notes_plat_sysinfo_565 = notes_plat_sysinfo_570 = uname -a: notes_plat_sysinfo_575 = Linux ice3 4.18.0-477.15.1.el8_8.x86_64 #1 SMP Wed Jun 28 15:04:18 UTC 2023 x86_64 notes_plat_sysinfo_580 = x86_64 x86_64 GNU/Linux notes_plat_sysinfo_585 = notes_plat_sysinfo_590 = Kernel self-reported vulnerability status: notes_plat_sysinfo_595 = notes_plat_sysinfo_600 = CVE-2018-12207 (iTLB Multihit): Not affected notes_plat_sysinfo_605 = CVE-2018-3620 (L1 Terminal Fault): Not affected notes_plat_sysinfo_610 = Microarchitectural Data Sampling: Not affected notes_plat_sysinfo_615 = CVE-2017-5754 (Meltdown): Not affected notes_plat_sysinfo_620 = mmio_stale_data: Mitigation: Clear CPU buffers; SMT notes_plat_sysinfo_625 = vulnerable notes_plat_sysinfo_630 = retbleed: Not affected notes_plat_sysinfo_635 = CVE-2018-3639 (Speculative Store Bypass): Mitigation: Speculative Store notes_plat_sysinfo_640 = Bypass disabled via prctl notes_plat_sysinfo_645 = CVE-2017-5753 (Spectre variant 1): Mitigation: usercopy/swapgs notes_plat_sysinfo_650 = barriers and __user pointer notes_plat_sysinfo_655 = sanitization notes_plat_sysinfo_660 = CVE-2017-5715 (Spectre variant 2): Mitigation: Enhanced IBRS, IBPB: notes_plat_sysinfo_665 = conditional, RSB filling, notes_plat_sysinfo_670 = PBRSB-eIBRS: SW sequence notes_plat_sysinfo_675 = CVE-2020-0543 (Special Register Buffer Data Sampling): Not affected notes_plat_sysinfo_680 = CVE-2019-11135 (TSX Asynchronous Abort): Not affected notes_plat_sysinfo_685 = notes_plat_sysinfo_690 = run-level 3 Sep 19 12:23 notes_plat_sysinfo_695 = notes_plat_sysinfo_700 = SPEC is set to: /local/home/mcolgrove/ACCELV2 notes_plat_sysinfo_705 = Filesystem Type Size Used Avail Use% Mounted on notes_plat_sysinfo_710 = /dev/mapper/rl_ice33-local xfs 930G 201G 729G 22% /local notes_plat_sysinfo_715 = notes_plat_sysinfo_720 = From /sys/devices/virtual/dmi/id notes_plat_sysinfo_725 = Vendor: Supermicro notes_plat_sysinfo_730 = Product: SYS-120GQ-TNRT notes_plat_sysinfo_735 = Product Family: SMC X12 notes_plat_sysinfo_740 = notes_plat_sysinfo_745 = Cannot run dmidecode; consider saying (as root) notes_plat_sysinfo_750 = chmod +s /usr/sbin/dmidecode notes_plat_sysinfo_755 = notes_plat_sysinfo_760 = BIOS: notes_plat_sysinfo_765 = BIOS Vendor: American Megatrends International, LLC. notes_plat_sysinfo_770 = BIOS Version: 1.4a notes_plat_sysinfo_775 = BIOS Date: 10/11/2022 notes_plat_sysinfo_780 = notes_plat_sysinfo_785 = (End of data from sysinfo program) hw_cpu_name = Intel Xeon Gold 6338 hw_disk = 1TB SATA hw_nchips = 2 prepared_by = mcolgrove (is never output, only tags rawfile) sw_compiler = C/Fortran: Version 23.11 of NVHPC SDK fw_bios = 1.4a 10/11/2022 sw_file = xfs sw_os000 = Rocky Linux release 8.8 (Green Obsidian) sw_os001 = 4.18.0-477.15.1.el8_8.x86_64 sw_state = Run level 3 (multi-user) sw_other = None # End of settings added by sysinfo_program # 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_plat_000 = Information from nvaccelinfo notes_plat_005 = CUDA Driver Version: 12000 notes_plat_010 = NVRM version: NVIDIA UNIX x86_64 Kernel Module 525.60.13 Wed Nov 30 06:39:21 UTC 2022 notes_plat_015 = Device Number: 0 notes_plat_020 = Device Name: NVIDIA H100 PCIe notes_plat_025 = Device Revision Number: 9.0 notes_plat_030 = Global Memory Size: 85021163520 notes_plat_035 = Number of Multiprocessors: 114 notes_plat_040 = Concurrent Copy and Execution: Yes notes_plat_045 = Total Constant Memory: 65536 notes_plat_050 = Total Shared Memory per Block: 49152 notes_plat_055 = Registers per Block: 65536 notes_plat_060 = Warp Size: 32 notes_plat_065 = Maximum Threads per Block: 1024 notes_plat_070 = Maximum Block Dimensions: 1024, 1024, 64 notes_plat_075 = Maximum Grid Dimensions: 2147483647 x 65535 x 65535 notes_plat_080 = Maximum Memory Pitch: 2147483647B notes_plat_085 = Texture Alignment: 512B notes_plat_090 = Clock Rate: 1755 MHz notes_plat_095 = Execution Timeout: No notes_plat_100 = Integrated Device: No notes_plat_105 = Can Map Host Memory: Yes notes_plat_110 = Compute Mode: default notes_plat_115 = Concurrent Kernels: Yes notes_plat_120 = ECC Enabled: Yes notes_plat_125 = Memory Clock Rate: 1593 MHz notes_plat_130 = Memory Bus Width: 5120 bits notes_plat_135 = L2 Cache Size: 52428800 bytes notes_plat_140 = Max Threads Per SMP: 2048 notes_plat_145 = Async Engines: 3 notes_plat_150 = Unified Addressing: Yes notes_plat_155 = Managed Memory: Yes notes_plat_160 = Concurrent Managed Memory: Yes notes_plat_165 = Preemption Supported: Yes notes_plat_170 = Cooperative Launch: Yes notes_plat_175 = Cluster Launch: Yes notes_plat_180 = Unified Function Pointers: Yes notes_plat_185 = Default Target: cc90 notes_os_000 =Shell stacksize set to unlimited via "limit stacksize unlimited"