SPEC2017 INTrate: PGO 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
500.perlbench_r
100.00%
100.88%
99.56%
100.00%
98.70%
502.gcc_r
100.00%
100.00%
100.00%
100.00%
100.00%
505.mcf_r
100.00%
100.00%
100.67%
100.00%
100.67%
520.omnetpp_r
100.00%
100.00%
99.76%
100.00%
99.76%
523.xalancbmk_r
100.00%
100.90%
100.00%
100.00%
99.11%
525.x264_r
100.00%
97.70%
96.68%
100.00%
98.95%
531.deepsjeng_r
100.00%
96.16%
93.13%
100.00%
96.85%
541.leela_r
100.00%
100.00%
101.53%
100.00%
101.53%
548.exchange2_r
100.00%
100.00%
94.77%
100.00%
94.77%
557.xz_r
100.00%
100.00%
99.58%
100.00%
99.58%
| Benchmark | x86-64 | x86-64-v2 | x86-64-v3 | x86-64-v2 | x86-64-v3 |
|---|---|---|---|---|---|
| 500.perlbench_r | 100.00% | 100.88% | 99.56% | 100.00% | 98.70% |
| 502.gcc_r | 100.00% | 100.00% | 100.00% | 100.00% | 100.00% |
| 505.mcf_r | 100.00% | 100.00% | 100.67% | 100.00% | 100.67% |
| 520.omnetpp_r | 100.00% | 100.00% | 99.76% | 100.00% | 99.76% |
| 523.xalancbmk_r | 100.00% | 100.90% | 100.00% | 100.00% | 99.11% |
| 525.x264_r | 100.00% | 97.70% | 96.68% | 100.00% | 98.95% |
| 531.deepsjeng_r | 100.00% | 96.16% | 93.13% | 100.00% | 96.85% |
| 541.leela_r | 100.00% | 100.00% | 101.53% | 100.00% | 101.53% |
| 548.exchange2_r | 100.00% | 100.00% | 94.77% | 100.00% | 94.77% |
| 557.xz_r | 100.00% | 100.00% | 99.58% | 100.00% | 99.58% |
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
500.perlbench_r
100.00%
100.00%
99.92%
100.00%
99.92%
502.gcc_r
100.00%
99.99%
100.18%
100.00%
100.19%
505.mcf_r
100.00%
100.04%
105.01%
100.00%
104.97%
520.omnetpp_r
100.00%
99.99%
100.66%
100.00%
100.67%
523.xalancbmk_r
100.00%
100.00%
100.67%
100.00%
100.67%
525.x264_r
100.00%
98.32%
95.86%
100.00%
97.49%
531.deepsjeng_r
100.00%
100.00%
99.97%
100.00%
99.98%
541.leela_r
100.00%
100.06%
101.39%
100.00%
101.33%
548.exchange2_r
100.00%
100.03%
104.78%
100.00%
104.74%
557.xz_r
100.00%
99.87%
100.32%
100.00%
100.45%
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 |
|---|---|---|---|---|---|
| 500.perlbench_r | 100.00% | 100.00% | 99.92% | 100.00% | 99.92% |
| 502.gcc_r | 100.00% | 99.99% | 100.18% | 100.00% | 100.19% |
| 505.mcf_r | 100.00% | 100.04% | 105.01% | 100.00% | 104.97% |
| 520.omnetpp_r | 100.00% | 99.99% | 100.66% | 100.00% | 100.67% |
| 523.xalancbmk_r | 100.00% | 100.00% | 100.67% | 100.00% | 100.67% |
| 525.x264_r | 100.00% | 98.32% | 95.86% | 100.00% | 97.49% |
| 531.deepsjeng_r | 100.00% | 100.00% | 99.97% | 100.00% | 99.98% |
| 541.leela_r | 100.00% | 100.06% | 101.39% | 100.00% | 101.33% |
| 548.exchange2_r | 100.00% | 100.03% | 104.78% | 100.00% | 104.74% |
| 557.xz_r | 100.00% | 99.87% | 100.32% | 100.00% | 100.45% |
Intel Cascade Lake Xeon
Run-times
Benchmark
x86-64
x86-64-v2
x86-64-v3
x86-64-v2
x86-64-v3
500.perlbench_r
100.00%
100.86%
99.78%
100.00%
98.93%
502.gcc_r
100.00%
100.00%
99.57%
100.00%
99.57%
505.mcf_r
100.00%
99.35%
99.35%
100.00%
100.00%
520.omnetpp_r
100.00%
100.00%
101.66%
100.00%
101.66%
523.xalancbmk_r
100.00%
100.00%
100.43%
100.00%
100.43%
525.x264_r
100.00%
97.07%
93.40%
100.00%
96.22%
531.deepsjeng_r
100.00%
96.38%
92.96%
100.00%
96.45%
541.leela_r
100.00%
100.00%
100.43%
100.00%
100.43%
548.exchange2_r
100.00%
102.18%
100.31%
100.00%
98.17%
557.xz_r
100.00%
103.33%
99.17%
100.00%
95.97%
| Benchmark | x86-64 | x86-64-v2 | x86-64-v3 | x86-64-v2 | x86-64-v3 |
|---|---|---|---|---|---|
| 500.perlbench_r | 100.00% | 100.86% | 99.78% | 100.00% | 98.93% |
| 502.gcc_r | 100.00% | 100.00% | 99.57% | 100.00% | 99.57% |
| 505.mcf_r | 100.00% | 99.35% | 99.35% | 100.00% | 100.00% |
| 520.omnetpp_r | 100.00% | 100.00% | 101.66% | 100.00% | 101.66% |
| 523.xalancbmk_r | 100.00% | 100.00% | 100.43% | 100.00% | 100.43% |
| 525.x264_r | 100.00% | 97.07% | 93.40% | 100.00% | 96.22% |
| 531.deepsjeng_r | 100.00% | 96.38% | 92.96% | 100.00% | 96.45% |
| 541.leela_r | 100.00% | 100.00% | 100.43% | 100.00% | 100.43% |
| 548.exchange2_r | 100.00% | 102.18% | 100.31% | 100.00% | 98.17% |
| 557.xz_r | 100.00% | 103.33% | 99.17% | 100.00% | 95.97% |
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
500.perlbench_r
100.00%
100.00%
99.92%
100.00%
99.92%
502.gcc_r
100.00%
99.99%
100.18%
100.00%
100.19%
505.mcf_r
100.00%
100.04%
105.01%
100.00%
104.97%
520.omnetpp_r
100.00%
99.99%
100.66%
100.00%
100.67%
523.xalancbmk_r
100.00%
100.00%
100.67%
100.00%
100.67%
525.x264_r
100.00%
98.32%
95.86%
100.00%
97.49%
531.deepsjeng_r
100.00%
100.00%
99.97%
100.00%
99.98%
541.leela_r
100.00%
100.06%
101.39%
100.00%
101.33%
548.exchange2_r
100.00%
100.03%
104.78%
100.00%
104.74%
557.xz_r
100.00%
99.87%
100.32%
100.00%
100.45%
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 |
|---|---|---|---|---|---|
| 500.perlbench_r | 100.00% | 100.00% | 99.92% | 100.00% | 99.92% |
| 502.gcc_r | 100.00% | 99.99% | 100.18% | 100.00% | 100.19% |
| 505.mcf_r | 100.00% | 100.04% | 105.01% | 100.00% | 104.97% |
| 520.omnetpp_r | 100.00% | 99.99% | 100.66% | 100.00% | 100.67% |
| 523.xalancbmk_r | 100.00% | 100.00% | 100.67% | 100.00% | 100.67% |
| 525.x264_r | 100.00% | 98.32% | 95.86% | 100.00% | 97.49% |
| 531.deepsjeng_r | 100.00% | 100.00% | 99.97% | 100.00% | 99.98% |
| 541.leela_r | 100.00% | 100.06% | 101.39% | 100.00% | 101.33% |
| 548.exchange2_r | 100.00% | 100.03% | 104.78% | 100.00% | 104.74% |
| 557.xz_r | 100.00% | 99.87% | 100.32% | 100.00% | 100.45% |