[Bug tree-optimization/46590] [14/16 Regression] long compile time with -O2 and many loops

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=46590

--- Comment #60 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Thomas Koenig from comment #59)
> Timing for gcc-8:
> 
> real    38m51.275s
> user    37m49.899s
> sys     1m1.319s
> 
> Timing for gcc-10:
> 
> real    42m20.489s
> user    41m19.053s
> sys     1m1.338s
> 
> Summing up (time in seconds)
> 
> Gcc-8	2331.26
> Gcc-10	2540.48
> Gcc-12	2490.37
> Gcc-13	2459.99
> Gcc-14	13530.79
> Gcc-15	2268.68
> Gcc-16	2520.38
> 
> So, a very large regression for gcc-14, and a small one for gcc-16 vs.
> gcc-15.

Can you provide the full output of the current -ftime-report for GCC 14, 15,
trunk?