SPEC2017 FPrate: LTO -O2 - Different x86_64 levels
-flto -O2
AMD EPYC 7543P (zen3 based Milan)
Run-times
Benchmark
x86-64
x86-64-v2
x86-64-v3
x86-64-v2
x86-64-v3
503.bwaves_r
100.00%
100.00%
102.27%
100.00%
102.27%
507.cactuBSSN_r
100.00%
100.83%
97.29%
100.00%
96.49%
508.namd_r
100.00%
100.00%
85.10%
100.00%
85.10%
510.parest_r
100.00%
102.06%
113.99%
100.00%
111.69%
511.povray_r
100.00%
101.42%
97.97%
100.00%
96.59%
519.lbm_r
100.00%
99.60%
94.98%
100.00%
95.36%
521.wrf_r
100.00%
100.00%
95.77%
100.00%
95.77%
526.blender_r
100.00%
100.00%
99.19%
100.00%
99.19%
527.cam4_r
100.00%
99.60%
95.79%
100.00%
96.18%
538.imagick_r
100.00%
93.78%
80.32%
100.00%
85.65%
544.nab_r
100.00%
100.00%
95.18%
100.00%
95.18%
549.fotonik3d_r
100.00%
98.19%
87.55%
100.00%
89.16%
554.roms_r
100.00%
100.00%
98.18%
100.00%
98.18%
| Benchmark | x86-64 | x86-64-v2 | x86-64-v3 | x86-64-v2 | x86-64-v3 |
|---|---|---|---|---|---|
| 503.bwaves_r | 100.00% | 100.00% | 102.27% | 100.00% | 102.27% |
| 507.cactuBSSN_r | 100.00% | 100.83% | 97.29% | 100.00% | 96.49% |
| 508.namd_r | 100.00% | 100.00% | 85.10% | 100.00% | 85.10% |
| 510.parest_r | 100.00% | 102.06% | 113.99% | 100.00% | 111.69% |
| 511.povray_r | 100.00% | 101.42% | 97.97% | 100.00% | 96.59% |
| 519.lbm_r | 100.00% | 99.60% | 94.98% | 100.00% | 95.36% |
| 521.wrf_r | 100.00% | 100.00% | 95.77% | 100.00% | 95.77% |
| 526.blender_r | 100.00% | 100.00% | 99.19% | 100.00% | 99.19% |
| 527.cam4_r | 100.00% | 99.60% | 95.79% | 100.00% | 96.18% |
| 538.imagick_r | 100.00% | 93.78% | 80.32% | 100.00% | 85.65% |
| 544.nab_r | 100.00% | 100.00% | 95.18% | 100.00% | 95.18% |
| 549.fotonik3d_r | 100.00% | 98.19% | 87.55% | 100.00% | 89.16% |
| 554.roms_r | 100.00% | 100.00% | 98.18% | 100.00% | 98.18% |
Naturally, the smaller the run-times, the better. On the other hand, rates and especially their geometric means are better when they are greater.
See also continuous gcc tuning benchmarking at https://lnt.opensuse.org/db_default/v4/SPEC/spec_report/tuning.
Non-debug executable sizes
Click to display/hide
Benchmark
x86-64
x86-64-v2
x86-64-v3
x86-64-v2
x86-64-v3
503.bwaves_r
100.00%
100.04%
96.52%
100.00%
96.48%
507.cactuBSSN_r
100.00%
100.07%
96.85%
100.00%
96.79%
508.namd_r
100.00%
99.27%
93.04%
100.00%
93.72%
510.parest_r
100.00%
100.16%
100.15%
100.00%
99.99%
511.povray_r
100.00%
99.38%
96.97%
100.00%
97.57%
519.lbm_r
100.00%
99.77%
93.01%
100.00%
93.23%
521.wrf_r
100.00%
100.13%
100.33%
100.00%
100.19%
526.blender_r
100.00%
99.76%
99.61%
100.00%
99.86%
527.cam4_r
100.00%
99.99%
99.84%
100.00%
99.84%
538.imagick_r
100.00%
98.97%
97.11%
100.00%
98.11%
544.nab_r
100.00%
99.94%
99.21%
100.00%
99.27%
549.fotonik3d_r
100.00%
99.86%
99.02%
100.00%
99.16%
554.roms_r
100.00%
99.97%
99.35%
100.00%
99.38%
Smaller binary sizes are always better.
Click to display/hide
| Benchmark | x86-64 | x86-64-v2 | x86-64-v3 | x86-64-v2 | x86-64-v3 |
|---|---|---|---|---|---|
| 503.bwaves_r | 100.00% | 100.04% | 96.52% | 100.00% | 96.48% |
| 507.cactuBSSN_r | 100.00% | 100.07% | 96.85% | 100.00% | 96.79% |
| 508.namd_r | 100.00% | 99.27% | 93.04% | 100.00% | 93.72% |
| 510.parest_r | 100.00% | 100.16% | 100.15% | 100.00% | 99.99% |
| 511.povray_r | 100.00% | 99.38% | 96.97% | 100.00% | 97.57% |
| 519.lbm_r | 100.00% | 99.77% | 93.01% | 100.00% | 93.23% |
| 521.wrf_r | 100.00% | 100.13% | 100.33% | 100.00% | 100.19% |
| 526.blender_r | 100.00% | 99.76% | 99.61% | 100.00% | 99.86% |
| 527.cam4_r | 100.00% | 99.99% | 99.84% | 100.00% | 99.84% |
| 538.imagick_r | 100.00% | 98.97% | 97.11% | 100.00% | 98.11% |
| 544.nab_r | 100.00% | 99.94% | 99.21% | 100.00% | 99.27% |
| 549.fotonik3d_r | 100.00% | 99.86% | 99.02% | 100.00% | 99.16% |
| 554.roms_r | 100.00% | 99.97% | 99.35% | 100.00% | 99.38% |
Intel Cascade Lake Xeon
Run-times
Benchmark
x86-64
x86-64-v2
x86-64-v3
x86-64-v2
x86-64-v3
503.bwaves_r
100.00%
100.00%
97.18%
100.00%
97.18%
507.cactuBSSN_r
100.00%
100.00%
92.56%
100.00%
92.56%
508.namd_r
100.00%
100.00%
76.02%
100.00%
76.02%
510.parest_r
100.00%
100.40%
98.39%
100.00%
97.99%
511.povray_r
100.00%
100.82%
95.91%
100.00%
95.13%
519.lbm_r
100.00%
100.00%
93.35%
100.00%
93.35%
521.wrf_r
100.00%
100.41%
101.02%
100.00%
100.61%
526.blender_r
100.00%
101.65%
99.59%
100.00%
97.97%
527.cam4_r
100.00%
100.40%
97.57%
100.00%
97.18%
538.imagick_r
100.00%
90.93%
79.23%
100.00%
87.14%
544.nab_r
100.00%
100.20%
88.51%
100.00%
88.33%
549.fotonik3d_r
100.00%
98.79%
95.98%
100.00%
97.15%
554.roms_r
100.00%
100.00%
96.98%
100.00%
96.98%
| Benchmark | x86-64 | x86-64-v2 | x86-64-v3 | x86-64-v2 | x86-64-v3 |
|---|---|---|---|---|---|
| 503.bwaves_r | 100.00% | 100.00% | 97.18% | 100.00% | 97.18% |
| 507.cactuBSSN_r | 100.00% | 100.00% | 92.56% | 100.00% | 92.56% |
| 508.namd_r | 100.00% | 100.00% | 76.02% | 100.00% | 76.02% |
| 510.parest_r | 100.00% | 100.40% | 98.39% | 100.00% | 97.99% |
| 511.povray_r | 100.00% | 100.82% | 95.91% | 100.00% | 95.13% |
| 519.lbm_r | 100.00% | 100.00% | 93.35% | 100.00% | 93.35% |
| 521.wrf_r | 100.00% | 100.41% | 101.02% | 100.00% | 100.61% |
| 526.blender_r | 100.00% | 101.65% | 99.59% | 100.00% | 97.97% |
| 527.cam4_r | 100.00% | 100.40% | 97.57% | 100.00% | 97.18% |
| 538.imagick_r | 100.00% | 90.93% | 79.23% | 100.00% | 87.14% |
| 544.nab_r | 100.00% | 100.20% | 88.51% | 100.00% | 88.33% |
| 549.fotonik3d_r | 100.00% | 98.79% | 95.98% | 100.00% | 97.15% |
| 554.roms_r | 100.00% | 100.00% | 96.98% | 100.00% | 96.98% |
Naturally, the smaller the run-times, the better. On the other hand, rates and especially their geometric means are better when they are greater.
See also continuous gcc tuning benchmarking at https://lnt.opensuse.org/db_default/v4/SPEC/spec_report/tuning.
Non-debug executable sizes
Click to display/hide
Benchmark
x86-64
x86-64-v2
x86-64-v3
x86-64-v2
x86-64-v3
503.bwaves_r
100.00%
100.04%
96.52%
100.00%
96.48%
507.cactuBSSN_r
100.00%
100.07%
96.85%
100.00%
96.79%
508.namd_r
100.00%
99.27%
93.04%
100.00%
93.72%
510.parest_r
100.00%
100.16%
100.15%
100.00%
99.99%
511.povray_r
100.00%
99.38%
96.97%
100.00%
97.57%
519.lbm_r
100.00%
99.77%
93.01%
100.00%
93.23%
521.wrf_r
100.00%
100.13%
100.33%
100.00%
100.19%
526.blender_r
100.00%
99.76%
99.61%
100.00%
99.86%
527.cam4_r
100.00%
99.99%
99.84%
100.00%
99.84%
538.imagick_r
100.00%
98.97%
97.11%
100.00%
98.11%
544.nab_r
100.00%
99.94%
99.21%
100.00%
99.27%
549.fotonik3d_r
100.00%
99.86%
99.02%
100.00%
99.16%
554.roms_r
100.00%
99.97%
99.35%
100.00%
99.38%
Smaller binary sizes are always better.
Click to display/hide
| Benchmark | x86-64 | x86-64-v2 | x86-64-v3 | x86-64-v2 | x86-64-v3 |
|---|---|---|---|---|---|
| 503.bwaves_r | 100.00% | 100.04% | 96.52% | 100.00% | 96.48% |
| 507.cactuBSSN_r | 100.00% | 100.07% | 96.85% | 100.00% | 96.79% |
| 508.namd_r | 100.00% | 99.27% | 93.04% | 100.00% | 93.72% |
| 510.parest_r | 100.00% | 100.16% | 100.15% | 100.00% | 99.99% |
| 511.povray_r | 100.00% | 99.38% | 96.97% | 100.00% | 97.57% |
| 519.lbm_r | 100.00% | 99.77% | 93.01% | 100.00% | 93.23% |
| 521.wrf_r | 100.00% | 100.13% | 100.33% | 100.00% | 100.19% |
| 526.blender_r | 100.00% | 99.76% | 99.61% | 100.00% | 99.86% |
| 527.cam4_r | 100.00% | 99.99% | 99.84% | 100.00% | 99.84% |
| 538.imagick_r | 100.00% | 98.97% | 97.11% | 100.00% | 98.11% |
| 544.nab_r | 100.00% | 99.94% | 99.21% | 100.00% | 99.27% |
| 549.fotonik3d_r | 100.00% | 99.86% | 99.02% | 100.00% | 99.16% |
| 554.roms_r | 100.00% | 99.97% | 99.35% | 100.00% | 99.38% |