SPEC CPU®2026 Floating Point Rate Result

Copyright 2026 Standard Performance Evaluation Corporation

Cisco Systems

Cisco UCS C245 M8
(AMD EPYC 9845 2.10 GHz Processor)

SPECrate®2026_fp_base = 78800

SPECrate®2026_fp_peak = 78800

CPU2026 License: 9019 Test Date: Feb-2026
Test Sponsor: Cisco Systems Hardware Availability: Feb-2025
Tested by: Cisco Systems Software Availability: Jan-2026
Benchmark results graph
Hardware
CPU Name: AMD EPYC 9845
  Max MHz: 3700
  Nominal: 2100
Enabled: 320 cores, 2 chips, 2 threads/core
Orderable: 1,2 chips
Cache L1: 32 KB I + 48 KB D on chip per core
  L2: 1 MB I+D on chip per core
  L3: 320 MB I+D on chip per chip, 32 MB shared / 16
cores
  Other: None
Memory: 3 TB (24 x 128 GB 2Rx4 PC5-6400B-R, running at
6000)
Storage: 240 GB M.2 SATA SSD
Cooling: Air
Other: None
Software
OS: SUSE Linux Enterprise Server 15 SP6
kernel version
6.4.0-150600.21-default
Compiler: C/C++/Fortran: Version 5.1.0 of AOCC
Compiler Category: Vendor
Firmware: Version 4.3.5c released Dec-2024
4.3.5c.0.1202241033
File System: btrfs
System State: Run level 3 (multi-user)
Base Pointers: 64-bit
Peak Pointers: 64-bit
Other: None
Power Management: BIOS and OS set to prefer performance at the cost
of additional power usage

Results Table

Benchmark Base Peak
Copies Seconds Ratio Seconds Ratio Seconds Ratio Copies Seconds Ratio Seconds Ratio Seconds Ratio
709.cactus_r 320 144 1900 216 1270 219 1250 320 144 1900 216 1270 219 1250
722.palm_r 320 694 609 693 610 694 608 320 694 609 693 610 694 608
731.astcenc_r 320 183 1470 183 1470 183 1470 320 183 1470 183 1470 183 1470
736.ocio_r 320 303 923 304 922 305 919 320 303 923 304 922 305 919
737.gmsh_r 320 206 711 207 710 205 716 320 206 711 207 710 205 716
748.flightdm_r 320 161 1430 161 1430 160 1430 320 161 1430 161 1430 160 1430
749.fotonik3d_r 320 1676 221 1676 221 1677 221 320 1676 221 1676 221 1677 221
765.roms_r 320 1728 292 1731 291 1726 292 320 1728 292 1731 291 1726 292
766.femflow_r 320 231 2030 231 2030 231 2030 320 231 2030 231 2030 231 2030
767.nest_r 320 279 911 276 919 278 912 320 279 911 276 919 278 912
772.marian_r 320 337 1500 335 1510 338 1500 320 337 1500 335 1510 338 1500
782.lbm_r 320 606 303 606 302 606 303 320 606 303 606 302 606 303
SPECrate®2026_fp_base 78800
SPECrate®2026_fp_peak 78800
Results appear in the order in which they were run. Bold underlined text indicates a median measurement.

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

runcpu command invoked through numactl i.e.:
numactl --interleave=all runcpu <etc>

To limit dirty cache to 8% of memory, 'sysctl -w vm.dirty_ratio=8' run as root.
To limit swap usage to minimum necessary, 'sysctl -w vm.swappiness=1' run as root.
To free node-local memory and avoid remote memory usage,
'sysctl -w vm.zone_reclaim_mode=1' run as root.
To clear filesystem caches, 'sync; sysctl -w vm.drop_caches=3' run as root.
To disable address space layout randomization (ASLR) to reduce run-to-run
variability, 'sysctl -w kernel.randomize_va_space=0' run as root.
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.

Environment Variables Notes

Environment variables set by runcpu before the start of the run:
LD_LIBRARY_PATH =
     "/home/cpu2026.902/amd_rate_aocc510_znver5_A_lib/lib:/home/cpu2026.902/a
     md_rate_aocc510_znver5_A_lib/lib32:"
MALLOC_CONF = "retain:true"

General Notes

Binaries were compiled on a system with 2x AMD EPYC Venice256 CPU + 2TiB Memory using Ubuntu 24.04

Platform Notes


 BIOS Settings:
 NUMA nodes per socket set to NPS2
 Determinism Slider set to Power
 Enhanced CPU Performance set to Auto

 Sysinfo program /home/cpu2026.902/bin/sysinfo
 Rev: 069f95da7e7f5d81b2ce48a82150e54f
 running on localhost Tue Feb  3 11:46:59 2026

 SUT (System Under Test) info as seen by some common utilities.

 ------------------------------------------------------------
 Table of contents
 ------------------------------------------------------------
  1. uname -srvm
  2. w
  3. Username
  4. ulimit -a
  5. sysinfo process ancestry
  6. /proc/cpuinfo
  7. lscpu
  8. numactl --hardware
  9. /proc/meminfo
 10. who -r
 11. Systemd service manager version: systemd 254 (254.10+suse.84.ge8d77af424)
 12. Services, from systemctl list-unit-files
 13. Linux kernel boot-time arguments, from /proc/cmdline
 14. cpupower frequency-info
 15. tuned-adm active
 16. sysctl
 17. /sys/kernel/mm/transparent_hugepage
 18. /sys/kernel/mm/transparent_hugepage/khugepaged
 19. OS release
 20. Disk information
 21. /sys/devices/virtual/dmi/id
 22. dmidecode
 23. BIOS
 ------------------------------------------------------------

 ------------------------------------------------------------
 1. uname -srvm
   Linux 6.4.0-150600.21-default #1 SMP PREEMPT_DYNAMIC Thu May 16 11:09:22 UTC 2024 (36c1e09) x86_64

 ------------------------------------------------------------
 2. w
    11:46:59 up 21:22,  1 user,  load average: 42.00, 275.27, 452.39
   USER     TTY      FROM             LOGIN@   IDLE   JCPU   PCPU WHAT
   root     tty1     -                Mon14    6:42m  2.03s  0.77s /bin/bash ./amd_rate_aocc510_znver5_A1.sh

 ------------------------------------------------------------
 3. Username
   From environment variable $USER:  root

 ------------------------------------------------------------
 4. ulimit -a
   core file size          (blocks, -c) unlimited
   data seg size           (kbytes, -d) unlimited
   scheduling priority             (-e) 0
   file size               (blocks, -f) unlimited
   pending signals                 (-i) 12381968
   max locked memory       (kbytes, -l) 2097152
   max memory size         (kbytes, -m) unlimited
   open files                      (-n) 1024
   pipe size            (512 bytes, -p) 8
   POSIX message queues     (bytes, -q) 819200
   real-time priority              (-r) 0
   stack size              (kbytes, -s) unlimited
   cpu time               (seconds, -t) unlimited
   max user processes              (-u) 12381968
   virtual memory          (kbytes, -v) unlimited
   file locks                      (-x) unlimited

 ------------------------------------------------------------
 5. sysinfo process ancestry
  /usr/lib/systemd/systemd --switched-root --system --deserialize=42
  login -- root
  -bash
  python3 ./run_amd_rate_aocc510_znver5_A1.py
  /bin/bash ./amd_rate_aocc510_znver5_A1.sh
  runcpu --config amd_rate_aocc510_znver5_A1.cfg --tune base --reportable --iterations 3 fprate
  runcpu --configfile amd_rate_aocc510_znver5_A1.cfg --tune base --reportable --iterations 3 --nopower
    --runmode rate --tune base --size test:train:refrate fprate --nopreenv --note-preenv --logfile
    $SPEC/tmp/CPU2026.008/templogs/preenv.fprate.008.0.log --lognum 008.0 --from_runcpu 2
  specperl $SPEC/bin/sysinfo
 $SPEC = /home/cpu2026.902

 ------------------------------------------------------------
 6. /proc/cpuinfo
     model name      : AMD EPYC 9845 160-Core Processor
     vendor_id       : AuthenticAMD
     cpu family      : 26
     model           : 17
     stepping        : 0
     microcode       : 0xb101025
     bugs            : sysret_ss_attrs spectre_v1 spectre_v2 spec_store_bypass
     TLB size        : 192 4K pages
     cpu cores       : 160
     siblings        : 320
     2 physical ids (chips)
     640 processors (hardware threads)
     physical id 0: core ids 0-159
     physical id 1: core ids 0-159
     physical id 0: apicids 0-319
     physical id 1: apicids 512-831
   Caution: /proc/cpuinfo data regarding chips, cores, and threads is not necessarily reliable, especially for
   virtualized systems.  Use the above data carefully.

 ------------------------------------------------------------
 7. lscpu

 From lscpu from util-linux 2.39.3:
   Architecture:                         x86_64
   CPU op-mode(s):                       32-bit, 64-bit
   Address sizes:                        52 bits physical, 57 bits virtual
   Byte Order:                           Little Endian
   CPU(s):                               640
   On-line CPU(s) list:                  0-639
   Vendor ID:                            AuthenticAMD
   BIOS Vendor ID:                       Advanced Micro Devices, Inc.
   Model name:                           AMD EPYC 9845 160-Core Processor
   BIOS Model name:                      AMD EPYC 9845 160-Core Processor                Unknown CPU @ 2.1GHz
   BIOS CPU family:                      107
   CPU family:                           26
   Model:                                17
   Thread(s) per core:                   2
   Core(s) per socket:                   160
   Socket(s):                            2
   Stepping:                             0
   Frequency boost:                      enabled
   CPU(s) scaling MHz:                   57%
   CPU max MHz:                          3718.0659
   CPU min MHz:                          1500.0000
   BogoMIPS:                             4194.12
   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 amd_lbr_v2 nopl nonstop_tsc cpuid
                                         extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 pcid
                                         sse4_1 sse4_2 x2apic 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 hw_pstate ssbd mba perfmon_v2
                                         ibrs ibpb stibp ibrs_enhanced vmmcall fsgsbase tsc_adjust bmi1 avx2
                                         smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap
                                         avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt
                                         xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total
                                         cqm_mbm_local user_shstk avx_vnni avx512_bf16 clzero irperf
                                         xsaveerptr rdpru wbnoinvd amd_ppin cppc arat npt lbrv svm_lock
                                         nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter
                                         pfthreshold avic v_vmsave_vmload vgif x2avic v_spec_ctrl vnmi
                                         avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq
                                         avx512_vnni avx512_bitalg avx512_vpopcntdq la57 rdpid bus_lock_detect
                                         movdiri movdir64b overflow_recov succor smca fsrm avx512_vp2intersect
                                         flush_l1d debug_swap
   Virtualization:                       AMD-V
   L1d cache:                            15 MiB (320 instances)
   L1i cache:                            10 MiB (320 instances)
   L2 cache:                             320 MiB (320 instances)
   L3 cache:                             640 MiB (20 instances)
   NUMA node(s):                         4
   NUMA node0 CPU(s):                    0-79,320-399
   NUMA node1 CPU(s):                    80-159,400-479
   NUMA node2 CPU(s):                    160-239,480-559
   NUMA node3 CPU(s):                    240-319,560-639
   Vulnerability Gather data sampling:   Not affected
   Vulnerability Itlb multihit:          Not affected
   Vulnerability L1tf:                   Not affected
   Vulnerability Mds:                    Not affected
   Vulnerability Meltdown:               Not affected
   Vulnerability Mmio stale data:        Not affected
   Vulnerability Reg file data sampling: Not affected
   Vulnerability Retbleed:               Not affected
   Vulnerability Spec rstack overflow:   Not affected
   Vulnerability Spec store bypass:      Mitigation; Speculative Store Bypass disabled via prctl
   Vulnerability Spectre v1:             Mitigation; usercopy/swapgs barriers and __user pointer sanitization
   Vulnerability Spectre v2:             Mitigation; Enhanced / Automatic IBRS; IBPB conditional; STIBP
                                         always-on; RSB filling; PBRSB-eIBRS Not affected; BHI Not affected
   Vulnerability Srbds:                  Not affected
   Vulnerability Tsx async abort:        Not affected

 From lscpu --cache:
      NAME ONE-SIZE ALL-SIZE WAYS TYPE        LEVEL  SETS PHY-LINE COHERENCY-SIZE
      L1d       48K      15M   12 Data            1    64        1             64
      L1i       32K      10M    8 Instruction     1    64        1             64
      L2         1M     320M   16 Unified         2  1024        1             64
      L3        32M     640M   16 Unified         3 32768        1             64

 ------------------------------------------------------------
 8. numactl --hardware
 NOTE: a numactl 'node' might or might not correspond to a physical chip.
   available: 4 nodes (0-3)
   node 0 cpus: 0-79,320-399
   node 0 size: 773450 MB
   node 0 free: 770091 MB
   node 1 cpus: 80-159,400-479
   node 1 size: 774102 MB
   node 1 free: 771207 MB
   node 2 cpus: 160-239,480-559
   node 2 size: 774102 MB
   node 2 free: 771379 MB
   node 3 cpus: 240-319,560-639
   node 3 size: 773863 MB
   node 3 free: 771070 MB
   node distances:
   node   0   1   2   3
     0:  10  12  32  32
     1:  12  10  32  32
     2:  32  32  10  12
     3:  32  32  12  10

 ------------------------------------------------------------
 9. /proc/meminfo
    MemTotal:       3169810728 kB

 ------------------------------------------------------------
 10. who -r
   run-level 3 Feb 2 14:24

 ------------------------------------------------------------
 11. Systemd service manager version: systemd 254 (254.10+suse.84.ge8d77af424)
   Default Target  Status
   multi-user      running

 ------------------------------------------------------------
 12. Services, from systemctl list-unit-files
   STATE            UNIT FILES
   enabled          YaST2-Firstboot YaST2-Second-Stage apparmor auditd cron getty@ irqbalance issue-generator
                    kbdsettings klog lvm2-monitor nscd postfix purge-kernels rollback rsyslog smartd sshd
                    systemd-pstore wicked wickedd-auto4 wickedd-dhcp4 wickedd-dhcp6 wickedd-nanny
   enabled-runtime  systemd-remount-fs
   disabled         autofs autoyast-initscripts blk-availability boot-sysctl ca-certificates chrony-wait
                    chronyd console-getty cups cups-browsed debug-shell ebtables exchange-bmc-os-info
                    firewalld fsidd gpm grub2-once haveged hv_fcopy_daemon hv_kvp_daemon hv_vss_daemon
                    hwloc-dump-hwdata ipmi ipmievd issue-add-ssh-keys kexec-load ksm kvm_stat lunmask
                    man-db-create multipathd nfs nfs-blkmap rpcbind rpmconfigcheck rsyncd rtkit-daemon
                    serial-getty@ smartd_generate_opts snmpd snmptrapd svnserve systemd-boot-check-no-failures
                    systemd-confext systemd-network-generator systemd-sysext systemd-time-wait-sync
                    systemd-timesyncd tuned udisks2
   indirect         systemd-userdbd wickedd

 ------------------------------------------------------------
 13. Linux kernel boot-time arguments, from /proc/cmdline
   BOOT_IMAGE=/boot/vmlinuz-6.4.0-150600.21-default
   root=UUID=d291fb80-24f2-4857-9e58-e9b1970af383
   splash=silent
   mitigations=auto
   quiet
   security=apparmor

 ------------------------------------------------------------
 14. cpupower frequency-info
   analyzing CPU 32:
     current policy: frequency should be within 1.50 GHz and 2.10 GHz.
                     The governor "performance" may decide which speed to use
                     within this range.
     boost state support:
       Supported: yes
       Active: yes

 ------------------------------------------------------------
 15. tuned-adm active
   Current active profile: throughput-performance

 ------------------------------------------------------------
 16. sysctl
   kernel.numa_balancing               1
   kernel.randomize_va_space           0
   vm.compaction_proactiveness        20
   vm.dirty_background_bytes           0
   vm.dirty_background_ratio          10
   vm.dirty_bytes                      0
   vm.dirty_expire_centisecs        3000
   vm.dirty_ratio                      8
   vm.dirty_writeback_centisecs      500
   vm.dirtytime_expire_seconds     43200
   vm.extfrag_threshold              500
   vm.min_unmapped_ratio               1
   vm.nr_hugepages                     0
   vm.nr_hugepages_mempolicy           0
   vm.nr_overcommit_hugepages          0
   vm.swappiness                       1
   vm.watermark_boost_factor       15000
   vm.watermark_scale_factor          10
   vm.zone_reclaim_mode                1

 ------------------------------------------------------------
 17. /sys/kernel/mm/transparent_hugepage
   defrag          [always] defer defer+madvise madvise never
   enabled         [always] madvise never
   hpage_pmd_size  2097152
   shmem_enabled   always within_size advise [never] deny force

 ------------------------------------------------------------
 18. /sys/kernel/mm/transparent_hugepage/khugepaged
   alloc_sleep_millisecs   60000
   defrag                      1
   max_ptes_none             511
   max_ptes_shared           256
   max_ptes_swap              64
   pages_to_scan            4096
   scan_sleep_millisecs    10000

 ------------------------------------------------------------
 19. OS release
   From /etc/*-release /etc/*-version
   os-release SUSE Linux Enterprise Server 15 SP6

 ------------------------------------------------------------
 20. Disk information
 SPEC is set to: /home/cpu2026.902
   Filesystem     Type   Size  Used Avail Use% Mounted on
   /dev/sda2      btrfs  224G   21G  200G  10% /home

 ------------------------------------------------------------
 21. /sys/devices/virtual/dmi/id
     Vendor:         Cisco Systems Inc
     Product:        UCSC-C245-M8SX
     Product Family: Cisco UCS Rack Server
     Serial:         WZP2750Z0CS

 ------------------------------------------------------------
 22. dmidecode
   Additional information from dmidecode 3.4 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:
     24x 0xCE00 M321RAJA0MB2-CCPKC 128 GB 2 rank 6400, configured at 6000


 ------------------------------------------------------------
 23. BIOS
 (This section combines info from /sys/devices and dmidecode.)
    BIOS Vendor:       Cisco Systems, Inc.
    BIOS Version:      C245M8.4.3.5c.0.1202241033
    BIOS Date:         12/02/2024
    BIOS Revision:     5.35

Compiler Version Notes

==============================================================================
C       | 782.lbm_r(base)
------------------------------------------------------------------------------
AMD clang version 17.0.6 (CLANG: AOCC_5.1.0-Build#1994 2025_12_23)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /opt/AMD/aocc/aocc-compiler-5.1.0/bin
------------------------------------------------------------------------------

==============================================================================
C++     | 731.astcenc_r(base) 736.ocio_r(base) 748.flightdm_r(base)
        | 766.femflow_r(base) 767.nest_r(base) 772.marian_r(base)
------------------------------------------------------------------------------
AMD clang version 17.0.6 (CLANG: AOCC_5.1.0-Build#1994 2025_12_23)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /opt/AMD/aocc/aocc-compiler-5.1.0/bin
------------------------------------------------------------------------------

==============================================================================
C++, C  | 709.cactus_r(base) 737.gmsh_r(base)
------------------------------------------------------------------------------
AMD clang version 17.0.6 (CLANG: AOCC_5.1.0-Build#1994 2025_12_23)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /opt/AMD/aocc/aocc-compiler-5.1.0/bin
------------------------------------------------------------------------------

==============================================================================
Fortran | 722.palm_r(base) 749.fotonik3d_r(base) 765.roms_r(base)
------------------------------------------------------------------------------
AMD clang version 17.0.6 (CLANG: AOCC_5.1.0-Build#1994 2025_12_23)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /opt/AMD/aocc/aocc-compiler-5.1.0/bin
------------------------------------------------------------------------------

Base Compiler Invocation

C benchmarks:

 clang 

C++ benchmarks:

 clang++ 

Fortran benchmarks:

 flang 

Benchmarks using both C and C++:

 clang++   clang 

Base Portability Flags

709.cactus_r:  -DSPEC_LP64 
722.palm_r:  -DSPEC_LP64 
731.astcenc_r:  -DSPEC_LP64 
736.ocio_r:  -fno-finite-math-only   -DSPEC_LP64 
737.gmsh_r:  -fno-fast-math   -DSPEC_LP64 
748.flightdm_r:  -fno-reciprocal-math   -DSPEC_LP64 
749.fotonik3d_r:  -DSPEC_LP64 
765.roms_r:  -DSPEC_LP64 
766.femflow_r:  -DSPEC_LP64 
767.nest_r:  -fno-finite-math-only   -DSPEC_LP64 
772.marian_r:  -DSPEC_LP64 
782.lbm_r:  -DSPEC_LP64 

Base Optimization Flags

C benchmarks:

 -m64   -std=c18   -Wl,-mllvm -Wl,-align-all-nofallthru-blocks=6   -Wl,-mllvm -Wl,-reduce-array-computations=3   -Wl,-mllvm -Wl,-ldist-scalar-expand   -fenable-aggressive-gather   -ffast-math   -O3   -march=znver5   -fveclib=AMDLIBM   -fno-PIE   -no-pie   -flto   -fstruct-layout=7   -mllvm -unroll-threshold=50   -mllvm -inline-threshold=1000   -fremap-arrays   -fstrip-mining   -mllvm -reduce-array-computations=3   -zopt   -lamdlibm   -lamdalloc   -lflang 

C++ benchmarks:

 -m64   -std=c++17   -Wl,-mllvm -Wl,-align-all-nofallthru-blocks=6   -Wl,-mllvm -Wl,-reduce-array-computations=3   -Wl,-mllvm -Wl,-x86-use-vzeroupper=false   -Wl,-mllvm -Wl,-extra-inliner   -ffast-math   -O3   -march=znver5   -fveclib=AMDLIBM   -flto   -mllvm -unroll-threshold=100   -mllvm -loop-unswitch-threshold=200000   -mllvm -reduce-array-computations=3   -zopt   -lamdlibm   -lamdalloc   -lflang 

Fortran benchmarks:

 -m64   -Mstandard   -Wl,-mllvm -Wl,-align-all-nofallthru-blocks=6   -Wl,-mllvm -Wl,-reduce-array-computations=3   -Wl,-mllvm -Wl,-enable-X86-prefetching   -Wl,-mllvm -Wl,-enable-aggressive-gather=true   -Wl,-mllvm -Wl,-enable-masked-gather-sequence=false   -ffast-math   -O3   -march=znver5   -fveclib=AMDLIBM   -flto   -Mrecursive   -funroll-loops   -mllvm -lsr-in-nested-loop   -mllvm -reduce-array-computations=3   -fepilog-vectorization-of-inductions   -zopt   -lamdlibm   -lamdalloc   -lflang 

Benchmarks using both C and C++:

 -m64   -std=c++17   -std=c18   -Wl,-mllvm -Wl,-align-all-nofallthru-blocks=6   -Wl,-mllvm -Wl,-reduce-array-computations=3   -Wl,-mllvm -Wl,-x86-use-vzeroupper=false   -Wl,-mllvm -Wl,-extra-inliner   -ffast-math   -O3   -march=znver5   -fveclib=AMDLIBM   -fno-PIE   -no-pie   -flto   -fstruct-layout=7   -mllvm -unroll-threshold=50   -mllvm -inline-threshold=1000   -fremap-arrays   -fstrip-mining   -mllvm -reduce-array-computations=3   -zopt   -mllvm -unroll-threshold=100   -mllvm -loop-unswitch-threshold=200000   -lamdlibm   -lamdalloc   -lflang 

Peak Optimization Flags

C benchmarks:

782.lbm_r:  basepeak = yes 

C++ benchmarks:

731.astcenc_r:  basepeak = yes 
736.ocio_r:  basepeak = yes 
748.flightdm_r:  basepeak = yes 
766.femflow_r:  basepeak = yes 
767.nest_r:  basepeak = yes 
772.marian_r:  basepeak = yes 

Fortran benchmarks:

722.palm_r:  basepeak = yes 
749.fotonik3d_r:  basepeak = yes 
765.roms_r:  basepeak = yes 

Benchmarks using both C and C++:

709.cactus_r:  basepeak = yes 
737.gmsh_r:  basepeak = yes 

The flags files that were used to format this result can be browsed at
http://www.spec.org/cpu2026/results/flags/aocc-flags.2026-05-04.00.html,
http://www.spec.org/cpu2026/results/flags/Cisco-Platform-Settings-AMD-Turin-v1.1-revG.html.

You can also download the XML flags sources by saving the following links:
http://www.spec.org/cpu2026/results/flags/aocc-flags.2026-05-04.00.xml,
http://www.spec.org/cpu2026/results/flags/Cisco-Platform-Settings-AMD-Turin-v1.1-revG.xml.