SPEC CPU®2017 Floating Point Speed Result

Copyright 2017-2021 Standard Performance Evaluation Corporation

ASUSTeK Computer Inc.

ASUS RS520A-E11(KMPA-U16) Server System
3.70 GHz, AMD EPYC 72F3

SPECspeed®2017_fp_base = 89.80

SPECspeed®2017_fp_peak = 94.00

CPU2017 License: 9016 Test Date: Jul-2021
Test Sponsor: ASUSTeK Computer Inc. Hardware Availability: May-2021
Tested by: ASUSTeK Computer Inc. Software Availability: Mar-2021

Benchmark result graphs are available in the PDF report.

Hardware
CPU Name: AMD EPYC 72F3
  Max MHz: 4100
  Nominal: 3700
Enabled: 8 cores, 1 chip, 2 threads/core
Orderable: 1 chip
Cache L1: 32 KB I + 32 KB D on chip per core
  L2: 512 KB I+D on chip per core
  L3: 256 MB I+D on chip per chip, 32 MB per core
  Other: None
Memory: 512 GB (8 x 64 GB 4Rx4 PC4-3200AA-L)
Storage: 1 x 240 GB SATA SSD
Other: None
Software
OS: SUSE Linux Enterprise Server 15 SP2 (x86_64)
Kernel 5.3.18-22-default
Compiler: C/C++/Fortran: Version 3.0.0 of AOCC
Parallel: Yes
Firmware: Version 0401 released Apr-2021
File System: xfs
System State: Run level 3 (multi-user)
Base Pointers: 64-bit
Peak Pointers: 64-bit
Other: jemalloc: jemalloc memory allocator library v5.1.0
Power Management: BIOS and OS set to prefer performance
at the cost of additional power usage.

Results Table

Benchmark Base Peak
Threads Seconds Ratio Seconds Ratio Seconds Ratio Threads Seconds Ratio Seconds Ratio Seconds Ratio
SPECspeed®2017_fp_base 89.80
SPECspeed®2017_fp_peak 94.00
Results appear in the order in which they were run. Bold underlined text indicates a median measurement.
603.bwaves_s 16 150 3940 150 3930 150 3930 16 1500 3940 1500 3930 1500 3930
607.cactuBSSN_s 16 148 1120 149 1120 148 1120 16 1470 1130 1480 1130 1470 1130
619.lbm_s 16 123 42.6 123 42.7 122 42.8 16 82.0 63.9 84.0 62.4 82.8 63.3
621.wrf_s 16 138 96.1 136 97.2 138 95.8 16 1380 96.1 1360 97.2 1380 95.8
627.cam4_s 16 155 57.3 155 57.2 155 57.3 16 1550 57.3 1550 57.2 1550 57.3
628.pop2_s 16 175 68.0 175 67.9 175 67.7 16 1750 68.0 1750 67.9 1750 67.7
638.imagick_s 16 242 59.7 242 59.6 242 59.6 16 2420 59.7 2420 59.6 2420 59.6
644.nab_s 16 163 1070 162 1080 162 1080 16 1630 1070 1620 1080 1620 1080
649.fotonik3d_s 16 134 68.1 133 68.4 134 68.2 16 1340 68.1 1330 68.4 1340 68.2
654.roms_s 16 143 1100 143 1100 143 1100 16 1350 1170 1350 1170 1350 1170

Compiler Notes

The AMD64 AOCC Compiler Suite is available at
http://developer.amd.com/amd-aocc/

Submit Notes

The 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 limit
 OS set to performance mode via cpupower frequency-set -g performance
runcpu 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% of
memory.
'echo 1 > /proc/sys/vm/swappiness' run as root to limit swap usage to minimum
necessary.
'echo 1 > /proc/sys/vm/zone_reclaim_mode' run as root to free node-local memory
and avoid remote memory usage.
'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 layout
randomization (ASLR) to reduce run-to-run variability.
To enable Transparent Hugepages (THP) for all allocations,
'echo always > /sys/kernel/mm/transparent_hugepage/enabled' and
'echo always > /sys/kernel/mm/transparent_hugepage/defrag' run as root.
To enable THP only on request for peak runs of 628.pop2_s, and 638.imagick_s,
'echo madvise > /sys/kernel/mm/transparent_hugepage/enabled' run as root.
To disable THP for peak runs of 627.cam4_s, 644.nab_s, 649.fotonik3d_s, and 654.roms_s,
'echo never > /sys/kernel/mm/transparent_hugepage/enabled' run as root.

Environment Variables Notes

Environment variables set by runcpu before the start of the run:
GOMP_CPU_AFFINITY = "0-15"
LD_LIBRARY_PATH =
     "/cpu118/amd_speed_aocc300_milan_B_lib/64;/cpu118/amd_speed_aocc300_mila
     n_B_lib/32:"
MALLOC_CONF = "retain:true"
OMP_DYNAMIC = "false"
OMP_SCHEDULE = "static"
OMP_STACKSIZE = "128M"
OMP_THREAD_LIMIT = "16"

Environment variables set by runcpu during the 607.cactuBSSN_s peak run:
GOMP_CPU_AFFINITY = "0-15"

Environment variables set by runcpu during the 619.lbm_s peak run:
GOMP_CPU_AFFINITY = "0-15"

Environment variables set by runcpu during the 654.roms_s peak run:
GOMP_CPU_AFFINITY = "0-15"

General Notes

Binaries were compiled on a system with 2x AMD EPYC 7742 CPU + 1TiB Memory using OpenSUSE 15.2

NA: 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 here:
https://github.com/jemalloc/jemalloc/releases/download/5.1.0/jemalloc-5.1.0.tar.bz2

Platform Notes

BIOS Configuration:
DLWM Support = Disabled
SVM Mode = Disabled
NUMA nodes per socket = NPS2
APBDIS = 1
Fix SOC P-state = P0
Engine Boost = Enabled
IOMMU = Disabled

 Sysinfo program /cpu118/bin/sysinfo
 Rev: r6622 of 2021-04-07 982a61ec0915b55891ef0e16acafc64d
 running on localhost Mon Jul  5 13:41:01 2021

 SUT (System Under Test) info as seen by some common utilities.
 For more information on this section, see
    https://www.spec.org/cpu2017/Docs/config.html#sysinfo

 From /proc/cpuinfo
    model name : AMD EPYC 72F3 8-Core Processor
       1  "physical id"s (chips)
       16 "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 : 8
       siblings  : 16
       physical 0: cores 0 1 2 3 4 5 6 7

 From lscpu from util-linux 2.33.1:
      Architecture:        x86_64
      CPU op-mode(s):      32-bit, 64-bit
      Byte Order:          Little Endian
      Address sizes:       48 bits physical, 48 bits virtual
      CPU(s):              16
      On-line CPU(s) list: 0-15
      Thread(s) per core:  2
      Core(s) per socket:  8
      Socket(s):           1
      NUMA node(s):        2
      Vendor ID:           AuthenticAMD
      CPU family:          25
      Model:               1
      Model name:          AMD EPYC 72F3 8-Core Processor
      Stepping:            1
      CPU MHz:             1787.926
      CPU max MHz:         3700.0000
      CPU min MHz:         1500.0000
      BogoMIPS:            7385.96
      Virtualization:      AMD-V
      L1d cache:           32K
      L1i cache:           32K
      L2 cache:            512K
      L3 cache:            32768K
      NUMA node0 CPU(s):   0-3,8-11
      NUMA node1 CPU(s):   4-7,12-15
      Flags:               fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov
      pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm
      constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf pni pclmulqdq
      monitor ssse3 fma cx16 pcid sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand
      lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw
      ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb
      cat_l3 cdp_l3 invpcid_single hw_pstate ssbd mba ibrs ibpb stibp vmmcall fsgsbase
      bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a rdseed adx smap clflushopt clwb sha_ni
      xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local
      clzero irperf xsaveerptr wbnoinvd arat npt lbrv svm_lock nrip_save tsc_scale
      vmcb_clean flushbyasid decodeassists pausefilter pfthreshold v_vmsave_vmload vgif
      umip pku ospke vaes vpclmulqdq rdpid overflow_recov succor smca

 /proc/cpuinfo cache data
    cache size : 512 KB

 From numactl --hardware
 WARNING: a numactl 'node' might or might not correspond to a physical chip.
   available: 2 nodes (0-1)
   node 0 cpus: 0 1 2 3 8 9 10 11
   node 0 size: 257823 MB
   node 0 free: 257578 MB
   node 1 cpus: 4 5 6 7 12 13 14 15
   node 1 size: 258031 MB
   node 1 free: 257559 MB
   node distances:
   node   0   1
     0:  10  12
     1:  12  10

 From /proc/meminfo
    MemTotal:       528235284 kB
    HugePages_Total:       0
    Hugepagesize:       2048 kB

 /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor has
    performance

 From /etc/*release* /etc/*version*
    os-release:
       NAME="SLES"
       VERSION="15-SP2"
       VERSION_ID="15.2"
       PRETTY_NAME="SUSE Linux Enterprise Server 15 SP2"
       ID="sles"
       ID_LIKE="suse"
       ANSI_COLOR="0;32"
       CPE_NAME="cpe:/o:suse:sles:15:sp2"

 uname -a:
    Linux localhost 5.3.18-22-default #1 SMP Wed Jun 3 12:16:43 UTC 2020 (720aeba) x86_64
    x86_64 x86_64 GNU/Linux

 Kernel self-reported vulnerability status:

 CVE-2018-12207 (iTLB Multihit):                        Not affected
 CVE-2018-3620 (L1 Terminal Fault):                     Not affected
 Microarchitectural Data Sampling:                      Not affected
 CVE-2017-5754 (Meltdown):                              Not affected
 CVE-2018-3639 (Speculative Store Bypass):              Mitigation: Speculative Store
                                                        Bypass disabled via prctl and
                                                        seccomp
 CVE-2017-5753 (Spectre variant 1):                     Mitigation: usercopy/swapgs
                                                        barriers and __user pointer
                                                        sanitization
 CVE-2017-5715 (Spectre variant 2):                     Mitigation: Full AMD retpoline,
                                                        IBPB: conditional, IBRS_FW, STIBP:
                                                        always-on, RSB filling
 CVE-2020-0543 (Special Register Buffer Data Sampling): Not affected
 CVE-2019-11135 (TSX Asynchronous Abort):               Not affected

 run-level 3 Jul 5 08:11

 SPEC is set to: /cpu118
    Filesystem     Type  Size  Used Avail Use% Mounted on
    /dev/sda4      xfs   199G   26G  174G  13% /

 From /sys/devices/virtual/dmi/id
     Vendor:         ASUSTeK COMPUTER INC.
     Product:        RS520A-E11-RS24U
     Product Family: Server
     Serial:         333366669999

 Additional information from dmidecode 3.2 follows.  WARNING: Use caution when you
 interpret this section. The 'dmidecode' program reads system data which is "intended to
 allow hardware to be accurately determined", but the intent may not be met, as there are
 frequent changes to hardware, firmware, and the "DMTF SMBIOS" standard.
   Memory:
     8x Samsung M386A8K40DM2-CWE 64 GB 4 rank 3200
     8x Unknown Unknown

 BIOS:
    BIOS Vendor:       American Megatrends Inc.
    BIOS Version:      0401
    BIOS Date:         04/14/2021
    BIOS Revision:     4.1

 (End of data from sysinfo program)

Compiler Version Notes

==============================================================================
C               | 619.lbm_s(base, peak) 638.imagick_s(base, peak)
                | 644.nab_s(base, peak)
------------------------------------------------------------------------------
AMD clang version 12.0.0 (CLANG: AOCC_3.0.0-Build#78 2020_12_10) (based on
  LLVM Mirror.Version.12.0.0)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /opt/AMD/aocc-compiler-3.0.0/bin
------------------------------------------------------------------------------

==============================================================================
C++, C, Fortran | 607.cactuBSSN_s(base, peak)
------------------------------------------------------------------------------
AMD clang version 12.0.0 (CLANG: AOCC_3.0.0-Build#78 2020_12_10) (based on
  LLVM Mirror.Version.12.0.0)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /opt/AMD/aocc-compiler-3.0.0/bin
AMD clang version 12.0.0 (CLANG: AOCC_3.0.0-Build#78 2020_12_10) (based on
  LLVM Mirror.Version.12.0.0)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /opt/AMD/aocc-compiler-3.0.0/bin
AMD clang version 12.0.0 (CLANG: AOCC_3.0.0-Build#78 2020_12_10) (based on
  LLVM Mirror.Version.12.0.0)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /opt/AMD/aocc-compiler-3.0.0/bin
------------------------------------------------------------------------------

==============================================================================
Fortran         | 603.bwaves_s(base, peak) 649.fotonik3d_s(base, peak)
                | 654.roms_s(base, peak)
------------------------------------------------------------------------------
AMD clang version 12.0.0 (CLANG: AOCC_3.0.0-Build#78 2020_12_10) (based on
  LLVM Mirror.Version.12.0.0)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /opt/AMD/aocc-compiler-3.0.0/bin
------------------------------------------------------------------------------

==============================================================================
Fortran, C      | 621.wrf_s(base, peak) 627.cam4_s(base, peak)
                | 628.pop2_s(base, peak)
------------------------------------------------------------------------------
AMD clang version 12.0.0 (CLANG: AOCC_3.0.0-Build#78 2020_12_10) (based on
  LLVM Mirror.Version.12.0.0)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /opt/AMD/aocc-compiler-3.0.0/bin
AMD clang version 12.0.0 (CLANG: AOCC_3.0.0-Build#78 2020_12_10) (based on
  LLVM Mirror.Version.12.0.0)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /opt/AMD/aocc-compiler-3.0.0/bin
------------------------------------------------------------------------------

Base Compiler Invocation

C benchmarks:

 clang 

Fortran benchmarks:

 flang 

Benchmarks using both Fortran and C:

 flang   clang 

Benchmarks using Fortran, C, and C++:

 clang++   clang   flang 

Base Portability Flags

603.bwaves_s:  -DSPEC_LP64 
607.cactuBSSN_s:  -DSPEC_LP64 
619.lbm_s:  -DSPEC_LP64 
621.wrf_s:  -DSPEC_CASE_FLAG   -Mbyteswapio   -DSPEC_LP64 
627.cam4_s:  -DSPEC_CASE_FLAG   -DSPEC_LP64 
628.pop2_s:  -DSPEC_CASE_FLAG   -Mbyteswapio   -DSPEC_LP64 
638.imagick_s:  -DSPEC_LP64 
644.nab_s:  -DSPEC_LP64 
649.fotonik3d_s:  -DSPEC_LP64 
654.roms_s:  -DSPEC_LP64 

Base Optimization Flags

C benchmarks:

 -m64   -mno-adx   -mno-sse4a   -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   -march=znver3   -fveclib=AMDLIBM   -ffast-math   -flto   -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   -z muldefs   -DSPEC_OPENMP   -fopenmp   -fopenmp=libomp   -lomp   -lamdlibm   -ljemalloc   -lflang   -lflangrti 

Fortran benchmarks:

 -m64   -mno-adx   -mno-sse4a   -Wl,-mllvm -Wl,-enable-X86-prefetching   -Wl,-mllvm -Wl,-enable-licm-vrp   -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   -march=znver3   -fveclib=AMDLIBM   -ffast-math   -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   -DSPEC_OPENMP   -fopenmp   -fopenmp=libomp   -lomp   -lamdlibm   -ljemalloc   -lflang   -lflangrti 

Benchmarks using both Fortran and C:

 -m64   -mno-adx   -mno-sse4a   -Wl,-mllvm -Wl,-enable-X86-prefetching   -Wl,-mllvm -Wl,-enable-licm-vrp   -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   -march=znver3   -fveclib=AMDLIBM   -ffast-math   -flto   -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   -Mrecursive   -mllvm -fuse-tile-inner-loop   -funroll-loops   -mllvm -extra-vectorizer-passes   -mllvm -lsr-in-nested-loop   -z muldefs   -DSPEC_OPENMP   -fopenmp   -fopenmp=libomp   -lomp   -lamdlibm   -ljemalloc   -lflang   -lflangrti 

Benchmarks using Fortran, C, and C++:

 -m64   -mno-adx   -mno-sse4a   -std=c++98   -Wl,-mllvm -Wl,-x86-use-vzeroupper=false   -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   -march=znver3   -fveclib=AMDLIBM   -ffast-math   -flto   -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   -Mrecursive   -mllvm -fuse-tile-inner-loop   -funroll-loops   -mllvm -lsr-in-nested-loop   -z muldefs   -DSPEC_OPENMP   -fopenmp   -fopenmp=libomp   -lomp   -lamdlibm   -ljemalloc   -lflang   -lflangrti 

Base Other Flags

C benchmarks:

 -Wno-unused-command-line-argument   -Wno-return-type 

Fortran benchmarks:

 -Wno-unused-command-line-argument   -Wno-return-type 

Benchmarks using both Fortran and C:

 -Wno-unused-command-line-argument   -Wno-return-type 

Benchmarks using Fortran, C, and C++:

 -Wno-unused-command-line-argument   -Wno-return-type 

Peak Compiler Invocation

C benchmarks:

 clang 

Fortran benchmarks:

 flang 

Benchmarks using both Fortran and C:

 flang   clang 

Benchmarks using Fortran, C, and C++:

 clang++   clang   flang 

Peak Portability Flags

Same as Base Portability Flags

Peak Optimization Flags

C benchmarks:

619.lbm_s:  -m64   -mno-adx   -mno-sse4a   -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   -ffast-math   -flto   -fstruct-layout=5   -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   -DSPEC_OPENMP   -fopenmp   -fopenmp=libomp   -lomp   -lamdlibm   -ljemalloc   -lflang 
638.imagick_s:  basepeak = yes 
644.nab_s:  basepeak = yes 

Fortran benchmarks:

603.bwaves_s:  basepeak = yes 
649.fotonik3d_s:  basepeak = yes 
654.roms_s:  -m64   -mno-adx   -mno-sse4a   -Wl,-mllvm -Wl,-enable-X86-prefetching   -Wl,-mllvm -Wl,-enable-licm-vrp   -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   -ffast-math   -Mrecursive   -mllvm -reduce-array-computations=3   -mllvm -global-vectorize-slp=true   -mllvm -enable-licm-vrp   -DSPEC_OPENMP   -fopenmp   -fopenmp=libomp   -lomp   -lamdlibm   -ljemalloc   -lflang 

Benchmarks using both Fortran and C:

621.wrf_s:  basepeak = yes 
627.cam4_s:  basepeak = yes 
628.pop2_s:  basepeak = yes 

Benchmarks using Fortran, C, and C++:

 -m64   -mno-adx   -mno-sse4a   -std=c++98   -Wl,-mllvm -Wl,-x86-use-vzeroupper=false   -Wl,-mllvm -Wl,-enable-licm-vrp   -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   -ffast-math   -flto   -fstruct-layout=5   -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   -finline-aggressive   -mllvm -unroll-threshold=100   -mllvm -reroll-loops   -mllvm -aggressive-loop-unswitch   -Mrecursive   -DSPEC_OPENMP   -fopenmp   -fopenmp=libomp   -lomp   -lamdlibm   -ljemalloc   -lflang 

Peak Other Flags

C benchmarks:

 -Wno-unused-command-line-argument   -Wno-return-type 

Fortran benchmarks:

 -Wno-unused-command-line-argument   -Wno-return-type 

Benchmarks using both Fortran and C:

 -Wno-unused-command-line-argument   -Wno-return-type 

Benchmarks using Fortran, C, and C++:

 -Wno-unused-command-line-argument   -Wno-return-type 

The flags files that were used to format this result can be browsed at
http://www.spec.org/cpu2017/flags/ASUSTekPlatform-Settings-AMD-Milan-V1.3.2021-07-06.html,
http://www.spec.org/cpu2017/flags/aocc300-flags-A1.html.

You can also download the XML flags sources by saving the following links:
http://www.spec.org/cpu2017/flags/ASUSTekPlatform-Settings-AMD-Milan-V1.3.2021-07-06.xml,
http://www.spec.org/cpu2017/flags/aocc300-flags-A1.xml.