Transcription
SPEC CPU 2017 Floating Point Rate Resultspec Copyright 2017-2021 Standard Performance Evaluation CorporationCisco SystemsCisco UCS C225 M6 (AMD EPYC 7453 28-Core)CPU2017 License: 9019Test Sponsor:Cisco SystemsTested by:Cisco SystemsCopies 0 100 200503.bwaves r112507.cactuBSSN r112508.namd r112SPECrate 2017 fp base 402SPECrate 2017 fp peak 417Test Date:Oct-2021Hardware Availability: Jun-2021Software Availability: 50016001800693593334334510.parest r206112207511.povray r501112503112519.lbm r56112521.wrf r56177179350408481526.blender r 112482421527.cam4 r 1124251790538.imagick r 1121790544.nab r665112672549.fotonik3d r554.roms r1125611256242245141181SPECrate 2017 fp base (402)SPECrate 2017 fp peak (417)HardwareCPU Name:Max MHz:Nominal:Enabled:Orderable:Cache L1:L2:L3:Other:Memory:Storage:Other:Page 1AMD EPYC 74533450275056 cores, 2 chips, 2 threads/core1,2 chips32 KB I 32 KB D on chip per core512 KB I D on chip per core64 MB I D on chip per chip,16 MB shared / 7 coresNone2 TB (16 x 128 GB 4Rx4 PC4-3200AA-L)1 x 1.9 TB SSD SATANoneSoftwareOS:SUSE Linux Enterprise Server 15 SP3 (x86 64)kernel :File System:System State:Base Pointers:Peak Pointers:Other:Power Management:C/C /Fortran: Version 3.0.0 of AOCCNoVersion 4.2.1c released Aug-2021xfsRun level 3 (multi-user)64-bit64-bitjemalloc: jemalloc memory allocator library v5.1.0BIOS and OS set to prefer performance at the costof additional power usageStandard Performance Evaluation Corporation (info@spec.org)https://www.spec.org/
SPEC CPU 2017 Floating Point Rate Resultspec Copyright 2017-2021 Standard Performance Evaluation CorporationCisco SystemsCisco UCS C225 M6 (AMD EPYC 7453 28-Core)SPECrate 2017 fp base 402SPECrate 2017 fp peak 417CPU2017 License: 9019Test Sponsor:Cisco SystemsTested by:Cisco SystemsTest Date:Oct-2021Hardware Availability: Jun-2021Software Availability: Jun-2021Results TableBenchmark503.bwaves r507.cactuBSSN r508.namd r510.parest r511.povray r519.lbm r521.wrf r526.blender r527.cam4 r538.imagick r544.nab r549.fotonik3d r554.roms 121121121121121121121121121121121621 6931621 6931621 6931121621 6931621 6931621 693240 592239 594239 593112240 592239 594239 593319 333319 334319 334112318 334318 334319 3341423 2061425 2061431 2051121412 2071422 2061409 208521 502525 498522 501112520 503518 505520 503668 177670 176667 17756331 179331 178331 179715 351718 350721 34856308 407308 408308 408355 481354 482356 479112352 484355 481354 482464 422465 421468 419112460 426461 425465 421156 1790155 1790156 1790112155 1790155 1790155 1790284 665283 665283 665112280 672280 673280 6721803 2421803 2421803 24256890 245890 245890 2451258 1411262 1411264 14156494 180491 181490 182SPECrate 2017 fp base 402SPECrate 2017 fp peak 417Results appear in the order in which they were run. Bold underlined text indicates a median measurement.Compiler NotesThe AMD64 AOCC Compiler Suite is available athttp://developer.amd.com/amd-aocc/Submit NotesThe config file option 'submit' was used.'numactl' was used to bind copies to the cores.See the configuration file for details.Operating System Notes'ulimit -s unlimited' was used to set environment stack size limit'ulimit -l 2097152' was used to set environment locked pages in memory limitruncpu command invoked through numactl i.e.:numactl --interleave all runcpu etc 'echo 8 /proc/sys/vm/dirty ratio' run as root to limit dirty cache to 8% ofmemory.'echo 1 /proc/sys/vm/swappiness' run as root to limit swap usage to minimumnecessary.'echo 1 /proc/sys/vm/zone reclaim mode' run as root to free node-local memoryand avoid remote memory usage.(Continued on next page)Page 2Standard Performance Evaluation Corporation (info@spec.org)https://www.spec.org/
SPEC CPU 2017 Floating Point Rate Resultspec Copyright 2017-2021 Standard Performance Evaluation CorporationCisco SystemsCisco UCS C225 M6 (AMD EPYC 7453 28-Core)SPECrate 2017 fp base 402SPECrate 2017 fp peak 417CPU2017 License: 9019Test Sponsor:Cisco SystemsTested by:Cisco SystemsTest Date:Oct-2021Hardware Availability: Jun-2021Software Availability: Jun-2021Operating System Notes (Continued)'sync; echo 3 /proc/sys/vm/drop caches' run as root to reset filesystem caches.'sysctl -w kernel.randomize va space 0' run as root to disable address space layoutrandomization (ASLR) to reduce run-to-run variability.'echo always /sys/kernel/mm/transparent hugepage/enabled' and'echo always /sys/kernel/mm/transparent hugepage/defrag' run as root for peakinteger runs and all FP runs to enable Transparent Hugepages (THP).'cpupower frequency-set -g performance' run as root to set the scaling governor toperformance.Environment Variables NotesEnvironment variables set by runcpu before the start of the run:LD LIBRARY PATH "/home/cpu2017/amd rate aocc300 milan B lib/lib;/home/cpu2017/amd rate aocc300 milan B lib/lib32:"MALLOC CONF "retain:true"General NotesBinaries were compiled on a system with 2x AMD EPYC 7742 CPU 1TiB Memory using OpenSUSE 15.2NA: The test sponsor attests, as of date of publication, that CVE-2017-5754 (Meltdown)is mitigated in the system as tested and documented.Yes: The test sponsor attests, as of date of publication, that CVE-2017-5753 (Spectre variant 1)is mitigated in the system as tested and documented.Yes: The test sponsor attests, as of date of publication, that CVE-2017-5715 (Spectre variant 2)is mitigated in the system as tested and documented.jemalloc: configured and built with GCC v4.8.2 in RHEL 7.4 (No options specified)jemalloc 5.1.0 is available /download/5.1.0/jemalloc-5.1.0.tar.bz2Platform NotesBIOS ConfigurationSMT Mode set to AutoNUMA nodes per socket set to NPS4ACPI SRAT L3 Cache As NUMA Domain set to EnabledDRAM Scrub Time set to DisabledDeterminism Slider set to PowerMemory Interleaving set to AutoAPBDIS set to 1(Continued on next page)Page 3Standard Performance Evaluation Corporation (info@spec.org)https://www.spec.org/
SPEC CPU 2017 Floating Point Rate Resultspec Copyright 2017-2021 Standard Performance Evaluation CorporationCisco SystemsCisco UCS C225 M6 (AMD EPYC 7453 28-Core)SPECrate 2017 fp base 402SPECrate 2017 fp peak 417CPU2017 License: 9019Test Sponsor:Cisco SystemsTested by:Cisco SystemsTest Date:Oct-2021Hardware Availability: Jun-2021Software Availability: Jun-2021Platform Notes (Continued)Sysinfo program /home/cpu2017/bin/sysinfoRev: r6622 of 2021-04-07 982a61ec0915b55891ef0e16acafc64drunning on localhost Mon Oct 18 14:18:34 2021SUT (System Under Test) info as seen by some common utilities.For more information on this section, ysinfoFrom /proc/cpuinfomodel name : AMD EPYC 7453 28-Core Processor2 "physical id"s (chips)112 "processors"cores, siblings (Caution: counting these is hw and system dependent. The followingexcerpts from /proc/cpuinfo might not be reliable. Use with caution.)cpu cores : 28siblings : 56physical 0: cores 0 1 2 3 4 5 6 8 9 10 11 12 13 14 16 17 18 19 20 21 22 24 25 26 2728 29 30physical 1: cores 0 1 2 3 4 5 6 8 9 10 11 12 13 14 16 17 18 19 20 21 22 24 25 26 2728 29 30From lscpu from util-linux 2.36.2:Architecture:CPU op-mode(s):Byte Order:Address sizes:CPU(s):On-line CPU(s) list:Thread(s) per core:Core(s) per socket:Socket(s):NUMA node(s):Vendor ID:CPU family:Model:Model name:Stepping:Frequency boost:CPU MHz:CPU max MHz:CPU min MHz:BogoMIPS:Virtualization:L1d cache:L1i cache:L2 cache:L3 cache:x86 6432-bit, 64-bitLittle Endian48 bits physical, 48 bits virtual1120-11122828AuthenticAMD251AMD EPYC 7453 28-Core AMD-V1.8 MiB1.8 MiB28 MiB128 MiB(Continued on next page)Page 4Standard Performance Evaluation Corporation (info@spec.org)https://www.spec.org/
SPEC CPU 2017 Floating Point Rate Resultspec Copyright 2017-2021 Standard Performance Evaluation CorporationCisco SystemsCisco UCS C225 M6 (AMD EPYC 7453 28-Core)SPECrate 2017 fp base 402SPECrate 2017 fp peak 417CPU2017 License: 9019Test Sponsor:Cisco SystemsTested by:Cisco SystemsTest Date:Oct-2021Hardware Availability: Jun-2021Software Availability: Jun-2021Platform Notes (Continued)NUMA node0 CPU(s):0-6,56-62NUMA node1 CPU(s):7-13,63-69NUMA node2 CPU(s):14-20,70-76NUMA node3 CPU(s):21-27,77-83NUMA node4 CPU(s):28-34,84-90NUMA node5 CPU(s):35-41,91-97NUMA node6 CPU(s):42-48,98-104NUMA node7 CPU(s):49-55,105-111Vulnerability Itlb multihit:Not affectedVulnerability L1tf:Not affectedVulnerability Mds:Not affectedVulnerability Meltdown:Not affectedVulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled viaprctl and seccompVulnerability Spectre v1:Mitigation; usercopy/swapgs barriers and userpointer sanitizationVulnerability Spectre v2:Mitigation; Full AMD retpoline, IBPB conditional,IBRS FW, STIBP always-on, RSB fillingVulnerability Srbds:Not affectedVulnerability Tsx async abort:Not affectedFlags:fpu vme de pse tsc msr pae mce cx8 apic sep mtrrpge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr optpdpe1gb rdtscp lm constant tsc rep good nopl nonstop tsc cpuid extd apicidaperfmperf pni pclmulqdq monitor ssse3 fma cx16 pcid sse4 1 sse4 2 movbe popcnt aesxsave avx f16c rdrand lahf lm cmp legacy svm extapic cr8 legacy abm sse4amisalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr core perfctr nbbpext perfctr llc mwaitx cpb cat l3 cdp l3 invpcid single hw pstate ssbd mba ibrsibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 invpcid cqm rdt a rdseed adx smapclflushopt clwb sha ni xsaveopt xsavec xgetbv1 xsaves cqm llc cqm occup llccqm mbm total cqm mbm local clzero irperf xsaveerptr wbnoinvd amd ppin arat npt lbrvsvm lock nrip save tsc scale vmcb clean flushbyasid decodeassists pausefilterpfthreshold v vmsave vmload vgif umip pku ospke vaes vpclmulqdq rdpid overflow recovsuccor smcaFrom lscpu --cache:NAME ONE-SIZE ALL-SIZE WAYS TYPELEVEL SETS PHY-LINE COHERENCY-SIZEL1d32K1.8M8 Data164164L1i32K1.8M8 Instruction164164L2512K28M8 Unified2 1024164L316M128M16 Unified3 16384164/proc/cpuinfo cache datacache size : 512 KBFrom numactl --hardwareWARNING: a numactl 'node' might or might not correspond to a physical chip.available: 8 nodes (0-7)(Continued on next page)Page 5Standard Performance Evaluation Corporation (info@spec.org)https://www.spec.org/
SPEC CPU 2017 Floating Point Rate Resultspec Copyright 2017-2021 Standard Performance Evaluation CorporationCisco SystemsCisco UCS C225 M6 (AMD EPYC 7453 28-Core)SPECrate 2017 fp base 402SPECrate 2017 fp peak 417CPU2017 License: 9019Test Sponsor:Cisco SystemsTested by:Cisco SystemsTest Date:Oct-2021Hardware Availability: Jun-2021Software Availability: Jun-2021Platform Notes odenodenodenode0:1:2:3:4:5:6:7:0 cpus: 0 1 2 3 4 50 size: 257854 MB0 free: 257461 MB1 cpus: 7 8 9 10 111 size: 258043 MB1 free: 257676 MB2 cpus: 14 15 16 172 size: 258043 MB2 free: 257679 MB3 cpus: 21 22 23 243 size: 258031 MB3 free: 257424 MB4 cpus: 28 29 30 314 size: 258009 MB4 free: 257679 MB5 cpus: 35 36 37 385 size: 258043 MB5 free: 257664 MB6 cpus: 42 43 44 456 size: 258043 MB6 free: 257694 MB7 cpus: 49 50 51 527 size: 258042 MB7 free: 257705 MBdistances:0123410 12 12 12 3212 10 12 12 3212 12 10 12 3212 12 12 10 3232 32 32 32 1032 32 32 32 1232 32 32 32 1232 32 32 32 126 56 57 58 59 60 61 6212 13 63 64 65 66 67 68 6918 19 20 70 71 72 73 74 75 7625 26 27 77 78 79 80 81 82 8332 33 34 84 85 86 87 88 89 9039 40 41 91 92 93 94 95 96 9746 47 48 98 99 100 101 102 103 10453 54 55 105 106 107 108 109 110 1210From /proc/meminfoMemTotal:2113649060 kBHugePages Total:0Hugepagesize:2048 kB/sys/devices/system/cpu/cpu*/cpufreq/scaling governor hasperformanceFrom /etc/*release* /etc/*version*os-release:NAME "SLES"VERSION "15-SP3"(Continued on next page)Page 6Standard Performance Evaluation Corporation (info@spec.org)https://www.spec.org/
SPEC CPU 2017 Floating Point Rate Resultspec Copyright 2017-2021 Standard Performance Evaluation CorporationCisco SystemsCisco UCS C225 M6 (AMD EPYC 7453 28-Core)SPECrate 2017 fp base 402SPECrate 2017 fp peak 417CPU2017 License: 9019Test Sponsor:Cisco SystemsTested by:Cisco SystemsTest Date:Oct-2021Hardware Availability: Jun-2021Software Availability: Jun-2021Platform Notes (Continued)VERSION ID "15.3"PRETTY NAME "SUSE Linux Enterprise Server 15 SP3"ID "sles"ID LIKE "suse"ANSI COLOR "0;32"CPE NAME "cpe:/o:suse:sles:15:sp3"uname -a:Linux localhost 5.3.18-57-default #1 SMP Wed Apr 28 10:54:41 UTC 2021 (ba3c2e9) x86 64x86 64 x86 64 GNU/LinuxKernel self-reported vulnerability status:CVE-2018-12207 (iTLB Multihit):CVE-2018-3620 (L1 Terminal Fault):Microarchitectural Data Sampling:CVE-2017-5754 (Meltdown):CVE-2018-3639 (Speculative Store Bypass):Not affectedNot affectedNot affectedNot affectedMitigation: Speculative StoreBypass disabled via prctl andseccompCVE-2017-5753 (Spectre variant 1):Mitigation: usercopy/swapgsbarriers and user pointersanitizationCVE-2017-5715 (Spectre variant 2):Mitigation: Full AMD retpoline,IBPB: conditional, IBRS FW, STIBP:always-on, RSB fillingCVE-2020-0543 (Special Register Buffer Data Sampling): Not affectedCVE-2019-11135 (TSX Asynchronous Abort):Not affectedrun-level 3 Oct 18 06:36SPEC is set to: /home/cpu2017FilesystemType Size Used Avail Use% Mounted on/dev/nvme0n1p3 xfs1.5T12G 1.5T1% /From /sys/devices/virtual/dmi/idVendor:Cisco Systems al information from dmidecode 3.2 follows. WARNING: Use caution when youinterpret this section. The 'dmidecode' program reads system data which is "intended toallow hardware to be accurately determined", but the intent may not be met, as there arefrequent changes to hardware, firmware, and the "DMTF SMBIOS" standard.Memory:16x 0xCE00 M386AAG40AM3-CWE 128 GB 4 rank 3200(Continued on next page)Page 7Standard Performance Evaluation Corporation (info@spec.org)https://www.spec.org/
SPEC CPU 2017 Floating Point Rate Resultspec Copyright 2017-2021 Standard Performance Evaluation CorporationCisco SystemsCisco UCS C225 M6 (AMD EPYC 7453 28-Core)SPECrate 2017 fp base 402SPECrate 2017 fp peak 417CPU2017 License: 9019Test Sponsor:Cisco SystemsTested by:Cisco SystemsTest Date:Oct-2021Hardware Availability: Jun-2021Software Availability: Jun-2021Platform Notes e:Revision:Cisco Systems, Inc.C225M6.4.2.1c.0.080621134908/06/20215.22(End of data from sysinfo program)Compiler Version Notes C 519.lbm r(base, peak) 538.imagick r(base, peak) 544.nab r(base, --------------------------------AMD clang version 12.0.0 (CLANG: AOCC 3.0.0-Build#78 2020 12 10) (based onLLVM Mirror.Version.12.0.0)Target: x86 64-unknown-linux-gnuThread model: posixInstalledDir: --------- C 508.namd r(base, peak) 510.parest r(base, --------------------------------AMD clang version 12.0.0 (CLANG: AOCC 3.0.0-Build#78 2020 12 10) (based onLLVM Mirror.Version.12.0.0)Target: x86 64-unknown-linux-gnuThread model: posixInstalledDir: --------- C , C 511.povray r(base, peak) 526.blender r(base, --------------------------------AMD clang version 12.0.0 (CLANG: AOCC 3.0.0-Build#78 2020 12 10) (based onLLVM Mirror.Version.12.0.0)Target: x86 64-unknown-linux-gnuThread model: posixInstalledDir: /opt/AMD/aocc-compiler-3.0.0/binAMD clang version 12.0.0 (CLANG: AOCC 3.0.0-Build#78 2020 12 10) (based onLLVM Mirror.Version.12.0.0)Target: x86 64-unknown-linux-gnuThread model: posixInstalledDir: ----------(Continued on next page)Page 8Standard Performance Evaluation Corporation (info@spec.org)https://www.spec.org/
SPEC CPU 2017 Floating Point Rate Resultspec Copyright 2017-2021 Standard Performance Evaluation CorporationCisco SystemsCisco UCS C225 M6 (AMD EPYC 7453 28-Core)CPU2017 License: 9019Test Sponsor:Cisco SystemsTested by:Cisco SystemsSPECrate 2017 fp base 402SPECrate 2017 fp peak 417Test Date:Oct-2021Hardware Availability: Jun-2021Software Availability: Jun-2021Compiler Version Notes (Continued) C , C, Fortran 507.cactuBSSN r(base, --------------------------------AMD clang version 12.0.0 (CLANG: AOCC 3.0.0-Build#78 2020 12 10) (based onLLVM Mirror.Version.12.0.0)Target: x86 64-unknown-linux-gnuThread model: posixInstalledDir: /opt/AMD/aocc-compiler-3.0.0/binAMD clang version 12.0.0 (CLANG: AOCC 3.0.0-Build#78 2020 12 10) (based onLLVM Mirror.Version.12.0.0)Target: x86 64-unknown-linux-gnuThread model: posixInstalledDir: /opt/AMD/aocc-compiler-3.0.0/binAMD clang version 12.0.0 (CLANG: AOCC 3.0.0-Build#78 2020 12 10) (based onLLVM Mirror.Version.12.0.0)Target: x86 64-unknown-linux-gnuThread model: posixInstalledDir: --------- Fortran 503.bwaves r(base, peak) 549.fotonik3d r(base, peak) 554.roms r(base, --------------------------------AMD clang version 12.0.0 (CLANG: AOCC 3.0.0-Build#78 2020 12 10) (based onLLVM Mirror.Version.12.0.0)Target: x86 64-unknown-linux-gnuThread model: posixInstalledDir: --------- Fortran, C 521.wrf r(base, peak) 527.cam4 r(base, --------------------------------AMD clang version 12.0.0 (CLANG: AOCC 3.0.0-Build#78 2020 12 10) (based onLLVM Mirror.Version.12.0.0)Target: x86 64-unknown-linux-gnuThread model: posixInstalledDir: /opt/AMD/aocc-compiler-3.0.0/binAMD clang version 12.0.0 (CLANG: AOCC 3.0.0-Build#78 2020 12 10) (based onLLVM Mirror.Version.12.0.0)Target: x86 64-unknown-linux-gnuThread model: posixInstalledDir: ----------Page 9Standard Performance Evaluation Corporation (info@spec.org)https://www.spec.org/
SPEC CPU 2017 Floating Point Rate Resultspec Copyright 2017-2021 Standard Performance Evaluation CorporationCisco SystemsCisco UCS C225 M6 (AMD EPYC 7453 28-Core)SPECrate 2017 fp base 402SPECrate 2017 fp peak 417CPU2017 License: 9019Test Sponsor:Cisco SystemsTested by:Cisco SystemsTest Date:Oct-2021Hardware Availability: Jun-2021Software Availability: Jun-2021Base Compiler InvocationC benchmarks:clangC benchmarks:clang Fortran benchmarks:flangBenchmarks using both Fortran and C:flang clangBenchmarks using both C and C :clang clangBenchmarks using Fortran, C, and C :clang clang flangBase Portability Flags503.bwaves r: -DSPEC LP64507.cactuBSSN r: -DSPEC LP64508.namd r: -DSPEC LP64510.parest r: -DSPEC LP64511.povray r: -DSPEC LP64519.lbm r: -DSPEC LP64521.wrf r: -DSPEC CASE FLAG -Mbyteswapio -DSPEC LP64526.blender r: -funsigned-char -D BOOL DEFINED -DSPEC LP64527.cam4 r: -DSPEC CASE FLAG -DSPEC LP64538.imagick r: -DSPEC LP64544.nab r: -DSPEC LP64549.fotonik3d r: -DSPEC LP64554.roms r: -DSPEC LP64Base Optimization FlagsC benchmarks:-m64 -flto -Wl,-mllvm -Wl,-region-vectorize-Wl,-mllvm -Wl,-function-specialize-Wl,-mllvm -Wl,-align-all-nofallthru-blocks 6-Wl,-mllvm -Wl,-reduce-array-computations 3 -O3 -ffast-math-march znver3 -fveclib AMDLIBM -fstruct-layout 5(Continued on next page)Page 10Standard Performance Evaluation Corporation (info@spec.org)https://www.spec.org/
SPEC CPU 2017 Floating Point Rate Resultspec Copyright 2017-2021 Standard Performance Evaluation CorporationCisco SystemsCisco UCS C225 M6 (AMD EPYC 7453 28-Core)SPECrate 2017 fp base 402SPECrate 2017 fp peak 417CPU2017 License: 9019Test Sponsor:Cisco SystemsTested by:Cisco SystemsTest Date:Oct-2021Hardware Availability: Jun-2021Software Availability: Jun-2021Base Optimization Flags (Continued)C benchmarks (continued):-mllvm -unroll-threshold 50 -mllvm -inline-threshold 1000-fremap-arrays -mllvm -function-specialize -flv-function-specialization-mllvm -enable-gvn-hoist -mllvm -global-vectorize-slp true-mllvm -enable-licm-vrp -mllvm -reduce-array-computations 3 -z muldefs-lamdlibm -ljemalloc -lflang -lflangrtiC benchmarks:-m64 -std c 98 -mno-adx -mno-sse4a-Wl,-mllvm -Wl,-x86-use-vzeroupper false -flto-Wl,-mllvm -Wl,-region-vectorize -Wl,-mllvm -Wl,-function-specialize-Wl,-mllvm -Wl,-align-all-nofallthru-blocks 6-Wl,-mllvm -Wl,-reduce-array-computations 3 -O3 -ffast-math-march znver3 -fveclib AMDLIBM -mllvm -enable-partial-unswitch-mllvm -unroll-threshold 100 -finline-aggressive-flv-function-specialization -mllvm -loop-unswitch-threshold 200000-mllvm -reroll-loops -mllvm -aggressive-loop-unswitch-mllvm -extra-vectorizer-passes -mllvm -reduce-array-computations 3-mllvm -global-vectorize-slp true -mllvm -convert-pow-exp-to-int false-z muldefs -lamdlibm -ljemalloc -lflang -lflangrtiFortran benchmarks:-m64 -Wl,-mllvm -Wl,-enable-X86-prefetching-Wl,-mllvm -Wl,-enable-licm-vrp -flto -Wl,-mllvm -Wl,-region-vectorize-Wl,-mllvm -Wl,-function-specialize-Wl,-mllvm -Wl,-align-all-nofallthru-blocks 6-Wl,-mllvm -Wl,-reduce-array-computations 3 -Hz,1,0x1 -O3 -ffast-math-march znver3 -fveclib AMDLIBM -Kieee -Mrecursive-mllvm -fuse-tile-inner-loop -funroll-loops-mllvm -extra-vectorizer-passes -mllvm -lsr-in-nested-loop-mllvm -enable-licm-vrp -mllvm -reduce-array-computations 3-mllvm -global-vectorize-slp true -z muldefs -lamdlibm -ljemalloc-lflang -lflangrtiBenchmarks using both Fortran and C:-m64 -Wl,-mllvm -Wl,-enable-X86-prefetching-Wl,-mllvm -Wl,-enable-licm-vrp -flto -Wl,-mllvm -Wl,-region-vectorize-Wl,-mllvm -Wl,-function-specialize-Wl,-mllvm -Wl,-align-all-nofallthru-blocks 6-Wl,-mllvm -Wl,-reduce-array-computations 3 -O3 -ffast-math-march znver3 -fveclib AMDLIBM -fstruct-layout 5-mllvm -unroll-threshold 50 -mllvm -inline-threshold 1000-fremap-arrays -mllvm -function-specialize -flv-function-specialization-mllvm -enable-gvn-hoist -mllvm -global-vectorize-slp true-mllvm -enable-licm-vrp -mllvm -reduce-array-computations 3 -Hz,1,0x1-Kieee -Mrecursive -mllvm -fuse-tile-inner-loop -funroll-loops(Continued on next page)Page 11Standard Performance Evaluation Corporation (info@spec.org)https://www.spec.org/
SPEC CPU 2017 Floating Point Rate Resultspec Copyright 2017-2021 Standard Performance Evaluation CorporationCisco SystemsCisco UCS C225 M6 (AMD EPYC 7453 28-Core)SPECrate 2017 fp base 402SPECrate 2017 fp peak 417CPU2017 License: 9019Test Sponsor:Cisco SystemsTested by:Cisco SystemsTest Date:Oct-2021Hardware Availability: Jun-2021Software Availability: Jun-2021Base Optimization Flags (Continued)Benchmarks using both Fortran and C (continued):-mllvm -extra-vectorizer-passes -mllvm -lsr-in-nested-loop -z muldefs-lamdlibm -ljemalloc -lflang -lflangrtiBenchmarks using both C and C :-m64 -std c 98 -mno-adx -mno-sse4a-Wl,-mllvm -Wl,-x86-use-vzeroupper false -flto-Wl,-mllvm -Wl,-region-vectorize -Wl,-mllvm -Wl,-function-specialize-Wl,-mllvm -Wl,-align-all-nofallthru-blocks 6-Wl,-mllvm -Wl,-reduce-array-computations 3 -O3 -ffast-math-march znver3 -fveclib AMDLIBM -fstruct-layout 5-mllvm -unroll-threshold 50 -mllvm -inline-threshold 1000-fremap-arrays -mllvm -function-specialize -flv-function-specialization-mllvm -enable-gvn-hoist -mllvm -global-vectorize-slp true-mllvm -enable-licm-vrp -mllvm -reduce-array-computations 3-mllvm -enable-partial-unswitch -mllvm -unroll-threshold 100-finline-aggressive -mllvm -loop-unswitch-threshold 200000-mllvm -reroll-loops -mllvm -aggressive-loop-unswitch-mllvm -extra-vectorizer-passes -mllvm -convert-pow-exp-to-int false-z muldefs -lamdlibm -ljemalloc -lflang -lflangrtiBenchmarks using Fortran, C, and C :-m64 -std c 98 -mno-adx -mno-sse4a-Wl,-mllvm -Wl,-x86-use-vzeroupper false -flto-Wl,-mllvm -Wl,-region-vectorize -Wl,-mllvm -Wl,-function-specialize-Wl,-mllvm -Wl,-align-all-nofallthru-blocks 6-Wl,-mllvm -Wl,-reduce-array-computations 3 -O3 -ffast-math-march znver3 -fveclib AMDLIBM -fstruct-layout 5-mllvm -unroll-threshold 50 -mllvm -inline-threshold 1000-fremap-arrays -mllvm -function-specialize -flv-function-specialization-mllvm -enable-gvn-hoist -mllvm -global-vectorize-slp true-mllvm -enable-licm-vrp -mllvm -reduce-array-computations 3-mllvm -enable-partial-unswitch -mllvm -unroll-threshold 100-finline-aggressive -mllvm -loop-unswitch-threshold 200000-mllvm -reroll-loops -mllvm -aggressive-loop-unswitch-mllvm -extra-vectorizer-passes -mllvm -convert-pow-exp-to-int false-Hz,1,0x1 -Kieee -Mrecursive -mllvm -fuse-tile-inner-loop-funroll-loops -mllvm -lsr-in-nested-loop -z muldefs -lamdlibm-ljemalloc -lflang -lflangrtiBase Other FlagsC nued on next page)Page 12Standard Performance Evaluation Corporation (info@spec.org)https://www.spec.org/
SPEC CPU 2017 Floating Point Rate Resultspec Copyright 2017-2021 Standard Performance Evaluation CorporationCisco SystemsCisco UCS C225 M6 (AMD EPYC 7453 28-Core)SPECrate 2017 fp base 402SPECrate 2017 fp peak 417CPU2017 License: 9019Test Sponsor:Cisco SystemsTested by:Cisco SystemsTest Date:Oct-2021Hardware Availability: Jun-2021Software Availability: Jun-2021Base Other Flags (Continued)C n arks using both Fortran and C:-Wno-unused-command-line-argumentBenchmarks using both C and C :-Wno-unused-command-line-argumentBenchmarks using Fortran, C, and C :-Wno-unused-command-line-argumentPeak Compiler InvocationC benchmarks:clangC benchmarks:clang Fortran benchmarks:flangBenchmarks using both Fortran and C:flang clangBenchmarks using both C and C :clang clangBenchmarks using Fortran, C, and C :clang clang flangPeak Portability FlagsSame as Base Portability FlagsPage 13Standard Performance Evaluation Corporation (info@spec.org)https://www.spec.org/
SPEC CPU 2017 Floating Point Rate Resultspec Copyright 2017-2021 Standard Performance Evaluation CorporationCisco SystemsCisco UCS C225 M6 (AMD EPYC 7453 28-Core)SPECrate 2017 fp base 402SPECrate 2017 fp peak 417CPU2017 License: 9019Test Sponsor:Cisco SystemsTested by:Cisco SystemsTest Date:Oct-2021Hardware Availability: Jun-2021Software Availability: Jun-2021Peak Optimization FlagsC benchmarks:519.lbm r: -m64 -flto -Wl,-mllvm -Wl,-function-specialize-Wl,-mllvm -Wl,-align-all-nofallthru-blocks 6-Wl,-mllvm -Wl,-reduce-array-computations 3 -Ofast-march znver3 -fveclib AMDLIBM -fstruct-layout 7-mllvm -unroll-threshold 50 -fremap-arrays-flv-function-specialization -mllvm -inline-threshold 1000-mllvm -enable-gvn-hoist -mllvm -global-vectorize-slp true-mllvm -function-specialize -mllvm -enable-licm-vrp-mllvm -reduce-array-computations 3 -lamdlibm -ljemalloc538.imagick r: Same as 519.lbm r544.nab r: -m64 -flto -Wl,-mllvm -Wl,-region-vectorize-Wl,-mllvm -Wl,-function-specialize -Ofast -march znver3-fveclib AMDLIBM -fstruct-layout 7-mllvm -unroll-threshold 50 -fremap-arrays-flv-function-specialization -mllvm -inline-threshold 1000-mllvm -enable-gvn-hoist -mllvm -global-vectorize-slp true-mllvm -function-specialize -mllvm -enable-licm-vrp-mllvm -reduce-array-computations 3 -lamdlibm -ljemallocC benchmarks:508.namd r: -m64 -std c 98 -mno-adx -mno-sse4a-Wl,-mllvm -Wl,-x86-use-vzeroupper false-Wl,-mllvm -Wl,-enable-licm-vrp -flto-Wl,-mllvm -Wl,-function-specialize-Wl,-mllvm -Wl,-align-all-nofallthru-blocks 6-Wl,-mllvm -Wl,-reduce-array-computations 3 -Ofast-march znver3 -fveclib AMDLIBM -finline-aggressive-mllvm -unroll-threshold 100 -flv-function-specialization-mllvm -enable-licm-vrp -mllvm -reroll-loops-mllvm -aggressive-loop-unswitch-mllvm -reduce-array-computations 3-mllvm -global-vectorize-slp true -lamdlibm -ljemalloc510.parest r: -m64 -std c 98 -mno-adx -mno-sse4a-Wl,-mllvm -Wl,-x86-use-vzeroupper false-Wl,-mllvm -Wl,-enable-licm-vrp -flto-Wl,-mllvm -Wl,-suppress-fmas-Wl,-mllvm -Wl,-function-specialize -Ofast -march znver3-fveclib AMDLIBM -finline-aggressive-mllvm -unroll-threshold 100 -flv-function-specialization-mllvm -enable-licm-vrp -mllvm -reroll-loops(Continued on next page)Page 14Standard Performance Evaluation Corporation (info@spec.org)https://www.spec.org/
SPEC CPU 2017 Floating Point Rate Resultspec Copyright 2017-2021 Standard Performance Evaluation CorporationCisco SystemsCisco UCS C225 M6 (AMD EPYC 7453 28-Core)SPECrate 2017 fp base 402SPECrate 2017 fp peak 417CPU2017 License: 9019Test Sponsor:Cisco SystemsTested by:Cisco SystemsTest Date:Oct-2021Hardware Availability: Jun-2021Software Availability: Jun-2021Peak Optimization Flags (Continued)510.parest r (continued):-mllvm -aggressive-loop-unswitch-mllvm -reduce-array-computations 3-mllvm -global-vectorize-slp true -lamdlibm -ljemallocFortran benchmarks:503.bwaves r: basepeak yes549.fotonik3d r: -m64 -Wl,-mllvm -Wl,-enable-X86-prefetching-Wl,-mllvm -Wl,-enable-licm-vrp -flto-Wl,-mllvm -Wl,-function-specialize-Wl,-mllvm -Wl,-align-all-nofallt
Oct 26, 2021
