[Bug target/84201] 549.fotonik3d_r from SPEC2017 fails verification with recent Intel and AMD CPUs

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84201

--- Comment #4 from Martin Liška <marxin at gcc dot gnu.org> ---
I see it failing with 4 years old revision r216027 on the Haswell machine.
Looking at portability issues:
https://www.spec.org/cpu2017/Docs/benchmarks/549.fotonik3d_r.html

```
It is perhaps worth mentioning that some calculations generate 'subnormal'
numbers (wikipedia) which may cause slower operation than normal numbers on
some hardware platforms. On such platforms, performance may be improved if
"flush to zero on underflow" (FTZ) is enabled. During SPEC's testing of
Fotonik3d, the output validated correctly whether or not FTZ was enabled.
```

Maybe that's causing the -Ofast issues?