SPEC2026 FPrate: PGO -O2 generic - LLVM and GCC
-O2 and -fprofile-instr-use vs. -profile-use
Please note that the same "optimization levels" mean different
things to different compilers. For example, LLVM and LLVM-based
compiler tend to be much more aggressive at -O2, often
with the corresponding consequences to size of the generated
binaries and compilation time. GCC is sometimes more aggressive at
-O3. LLVM is also often by default less strict when it
comes to standard compliance by default and performs optimizations
GCC would only do with a special compilation option or
at -Ofast. Keep this is mind when comparing different
compilers.
AMD EPYC 8635P (zen5 based Sorano)
Run-times
Benchmark
LLVM
GCC
709.cactus_r
100.00%
102.64%
722.palm_r
CE
731.astcenc_r
100.00%
142.07%
736.ocio_r
100.00%
100.00%
737.gmsh_r
100.00%
92.93%
748.flightdm_r
100.00%
89.98%
749.fotonik3d_r
CE
765.roms_r
CE
766.femflow_r
100.00%
187.23%
767.nest_r
100.00%
99.38%
772.marian_r
100.00%
234.72%
782.lbm_r
100.00%
114.29%
| Benchmark | LLVM | GCC |
|---|---|---|
| 709.cactus_r | 100.00% | 102.64% |
| 722.palm_r | CE | |
| 731.astcenc_r | 100.00% | 142.07% |
| 736.ocio_r | 100.00% | 100.00% |
| 737.gmsh_r | 100.00% | 92.93% |
| 748.flightdm_r | 100.00% | 89.98% |
| 749.fotonik3d_r | CE | |
| 765.roms_r | CE | |
| 766.femflow_r | 100.00% | 187.23% |
| 767.nest_r | 100.00% | 99.38% |
| 772.marian_r | 100.00% | 234.72% |
| 782.lbm_r | 100.00% | 114.29% |
| LLVM | clang version 23.1.3 (/labs/mjambor/git/llvm 0d261d1ca552c95a8f007e061c787ac7132fbcbc), flang version 23.1.3 (/labs/mjambor/git/llvm 0d261d1ca552c95a8f007e061c787ac7132fbcbc) |
| GCC | GCC 17.0.0 20261015 (experimental) (revision 84348add236100e5f69a3e9ec4f7b1038bf2b83d) |
Non-debug executable sizes
Benchmark
LLVM
GCC
709.cactus_r
100.00%
97.84%
722.palm_r
CE
731.astcenc_r
100.00%
120.25%
736.ocio_r
100.00%
77.97%
737.gmsh_r
100.00%
86.87%
748.flightdm_r
100.00%
86.76%
749.fotonik3d_r
CE
765.roms_r
CE
766.femflow_r
100.00%
76.54%
767.nest_r
100.00%
81.21%
772.marian_r
100.00%
82.58%
782.lbm_r
100.00%
79.69%
Smaller binary sizes are always better.
| Benchmark | LLVM | GCC |
|---|---|---|
| 709.cactus_r | 100.00% | 97.84% |
| 722.palm_r | CE | |
| 731.astcenc_r | 100.00% | 120.25% |
| 736.ocio_r | 100.00% | 77.97% |
| 737.gmsh_r | 100.00% | 86.87% |
| 748.flightdm_r | 100.00% | 86.76% |
| 749.fotonik3d_r | CE | |
| 765.roms_r | CE | |
| 766.femflow_r | 100.00% | 76.54% |
| 767.nest_r | 100.00% | 81.21% |
| 772.marian_r | 100.00% | 82.58% |
| 782.lbm_r | 100.00% | 79.69% |