SPECINT 2006: LTO -Ofast - Different x86_64 levels
-flto -Ofast
AMD EPYC 7543P (zen3 based Milan)
Run-times
Benchmark
x86-64
x86-64-v2
x86-64-v3
x86-64-v2
x86-64-v3
400.perlbench
100.00%
99.60%
99.60%
100.00%
100.00%
401.bzip2
100.00%
100.41%
101.02%
100.00%
100.61%
403.gcc
100.00%
100.00%
100.00%
100.00%
100.00%
429.mcf
100.00%
99.40%
100.00%
100.00%
100.61%
445.gobmk
100.00%
99.17%
100.41%
100.00%
101.25%
456.hmmer
100.00%
76.17%
71.84%
100.00%
94.31%
458.sjeng
100.00%
100.41%
98.78%
100.00%
98.37%
462.libquantum
100.00%
97.47%
88.82%
100.00%
91.13%
464.h264ref
100.00%
100.00%
100.64%
100.00%
100.64%
471.omnetpp
100.00%
105.43%
104.13%
100.00%
98.76%
473.astar
100.00%
100.00%
100.41%
100.00%
100.41%
483.xalancbmk
100.00%
104.71%
100.86%
100.00%
96.32%
| Benchmark | x86-64 | x86-64-v2 | x86-64-v3 | x86-64-v2 | x86-64-v3 |
|---|---|---|---|---|---|
| 400.perlbench | 100.00% | 99.60% | 99.60% | 100.00% | 100.00% |
| 401.bzip2 | 100.00% | 100.41% | 101.02% | 100.00% | 100.61% |
| 403.gcc | 100.00% | 100.00% | 100.00% | 100.00% | 100.00% |
| 429.mcf | 100.00% | 99.40% | 100.00% | 100.00% | 100.61% |
| 445.gobmk | 100.00% | 99.17% | 100.41% | 100.00% | 101.25% |
| 456.hmmer | 100.00% | 76.17% | 71.84% | 100.00% | 94.31% |
| 458.sjeng | 100.00% | 100.41% | 98.78% | 100.00% | 98.37% |
| 462.libquantum | 100.00% | 97.47% | 88.82% | 100.00% | 91.13% |
| 464.h264ref | 100.00% | 100.00% | 100.64% | 100.00% | 100.64% |
| 471.omnetpp | 100.00% | 105.43% | 104.13% | 100.00% | 98.76% |
| 473.astar | 100.00% | 100.00% | 100.41% | 100.00% | 100.41% |
| 483.xalancbmk | 100.00% | 104.71% | 100.86% | 100.00% | 96.32% |
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
400.perlbench
100.00%
100.37%
101.59%
100.00%
101.21%
401.bzip2
100.00%
94.49%
104.72%
100.00%
110.82%
403.gcc
100.00%
100.07%
100.72%
100.00%
100.65%
429.mcf
100.00%
100.14%
102.58%
100.00%
102.44%
445.gobmk
100.00%
99.94%
100.49%
100.00%
100.56%
456.hmmer
100.00%
100.76%
104.21%
100.00%
103.42%
458.sjeng
100.00%
99.99%
100.07%
100.00%
100.08%
462.libquantum
100.00%
113.67%
119.67%
100.00%
105.28%
464.h264ref
100.00%
99.55%
99.91%
100.00%
100.36%
471.omnetpp
100.00%
99.91%
101.11%
100.00%
101.21%
473.astar
100.00%
99.80%
97.67%
100.00%
97.86%
483.xalancbmk
100.00%
99.97%
101.29%
100.00%
101.32%
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 |
|---|---|---|---|---|---|
| 400.perlbench | 100.00% | 100.37% | 101.59% | 100.00% | 101.21% |
| 401.bzip2 | 100.00% | 94.49% | 104.72% | 100.00% | 110.82% |
| 403.gcc | 100.00% | 100.07% | 100.72% | 100.00% | 100.65% |
| 429.mcf | 100.00% | 100.14% | 102.58% | 100.00% | 102.44% |
| 445.gobmk | 100.00% | 99.94% | 100.49% | 100.00% | 100.56% |
| 456.hmmer | 100.00% | 100.76% | 104.21% | 100.00% | 103.42% |
| 458.sjeng | 100.00% | 99.99% | 100.07% | 100.00% | 100.08% |
| 462.libquantum | 100.00% | 113.67% | 119.67% | 100.00% | 105.28% |
| 464.h264ref | 100.00% | 99.55% | 99.91% | 100.00% | 100.36% |
| 471.omnetpp | 100.00% | 99.91% | 101.11% | 100.00% | 101.21% |
| 473.astar | 100.00% | 99.80% | 97.67% | 100.00% | 97.86% |
| 483.xalancbmk | 100.00% | 99.97% | 101.29% | 100.00% | 101.32% |
Intel Cascade Lake Xeon
Run-times
Benchmark
x86-64
x86-64-v2
x86-64-v3
x86-64-v2
x86-64-v3
400.perlbench
100.00%
99.19%
97.18%
100.00%
97.97%
401.bzip2
100.00%
102.05%
100.41%
100.00%
98.39%
403.gcc
100.00%
99.38%
99.38%
100.00%
100.00%
429.mcf
100.00%
100.61%
98.58%
100.00%
97.98%
445.gobmk
100.00%
99.37%
100.42%
100.00%
101.05%
456.hmmer
100.00%
81.98%
78.80%
100.00%
96.12%
458.sjeng
100.00%
101.03%
101.03%
100.00%
100.00%
462.libquantum
100.00%
98.48%
92.41%
100.00%
93.83%
464.h264ref
100.00%
103.74%
99.78%
100.00%
96.19%
471.omnetpp
100.00%
95.06%
100.00%
100.00%
105.19%
473.astar
100.00%
100.81%
96.55%
100.00%
95.77%
483.xalancbmk
100.00%
101.75%
100.88%
100.00%
99.14%
| Benchmark | x86-64 | x86-64-v2 | x86-64-v3 | x86-64-v2 | x86-64-v3 |
|---|---|---|---|---|---|
| 400.perlbench | 100.00% | 99.19% | 97.18% | 100.00% | 97.97% |
| 401.bzip2 | 100.00% | 102.05% | 100.41% | 100.00% | 98.39% |
| 403.gcc | 100.00% | 99.38% | 99.38% | 100.00% | 100.00% |
| 429.mcf | 100.00% | 100.61% | 98.58% | 100.00% | 97.98% |
| 445.gobmk | 100.00% | 99.37% | 100.42% | 100.00% | 101.05% |
| 456.hmmer | 100.00% | 81.98% | 78.80% | 100.00% | 96.12% |
| 458.sjeng | 100.00% | 101.03% | 101.03% | 100.00% | 100.00% |
| 462.libquantum | 100.00% | 98.48% | 92.41% | 100.00% | 93.83% |
| 464.h264ref | 100.00% | 103.74% | 99.78% | 100.00% | 96.19% |
| 471.omnetpp | 100.00% | 95.06% | 100.00% | 100.00% | 105.19% |
| 473.astar | 100.00% | 100.81% | 96.55% | 100.00% | 95.77% |
| 483.xalancbmk | 100.00% | 101.75% | 100.88% | 100.00% | 99.14% |
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
400.perlbench
100.00%
100.37%
101.59%
100.00%
101.21%
401.bzip2
100.00%
94.49%
104.72%
100.00%
110.82%
403.gcc
100.00%
100.07%
100.72%
100.00%
100.65%
429.mcf
100.00%
100.14%
102.58%
100.00%
102.44%
445.gobmk
100.00%
99.94%
100.49%
100.00%
100.56%
456.hmmer
100.00%
100.76%
104.21%
100.00%
103.42%
458.sjeng
100.00%
99.99%
100.07%
100.00%
100.08%
462.libquantum
100.00%
113.67%
119.67%
100.00%
105.28%
464.h264ref
100.00%
99.55%
99.91%
100.00%
100.36%
471.omnetpp
100.00%
99.91%
101.11%
100.00%
101.21%
473.astar
100.00%
99.80%
97.67%
100.00%
97.86%
483.xalancbmk
100.00%
99.97%
101.29%
100.00%
101.32%
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 |
|---|---|---|---|---|---|
| 400.perlbench | 100.00% | 100.37% | 101.59% | 100.00% | 101.21% |
| 401.bzip2 | 100.00% | 94.49% | 104.72% | 100.00% | 110.82% |
| 403.gcc | 100.00% | 100.07% | 100.72% | 100.00% | 100.65% |
| 429.mcf | 100.00% | 100.14% | 102.58% | 100.00% | 102.44% |
| 445.gobmk | 100.00% | 99.94% | 100.49% | 100.00% | 100.56% |
| 456.hmmer | 100.00% | 100.76% | 104.21% | 100.00% | 103.42% |
| 458.sjeng | 100.00% | 99.99% | 100.07% | 100.00% | 100.08% |
| 462.libquantum | 100.00% | 113.67% | 119.67% | 100.00% | 105.28% |
| 464.h264ref | 100.00% | 99.55% | 99.91% | 100.00% | 100.36% |
| 471.omnetpp | 100.00% | 99.91% | 101.11% | 100.00% | 101.21% |
| 473.astar | 100.00% | 99.80% | 97.67% | 100.00% | 97.86% |
| 483.xalancbmk | 100.00% | 99.97% | 101.29% | 100.00% | 101.32% |