https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107905
--- Comment #7 from Georgi ---
(In reply to Alexander Monakov from comment #5)
> Not sure what you don't like about the inputs, they appear quite reasonable.
> Perhaps GCC's estimation of bb frequencies is off (with profile feedback we
>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107905
--- Comment #6 from Alexander Monakov ---
Let me add that Clang supports GCC's -fprofile-{generate,use} flags for
compatibility as well.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107905
--- Comment #5 from Alexander Monakov ---
Not sure what you don't like about the inputs, they appear quite reasonable.
Perhaps GCC's estimation of bb frequencies is off (with profile feedback we
achieve good performance).
Georgi: you'll likely
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107905
--- Comment #4 from Andrew Pinski ---
(In reply to Alexander Monakov from comment #3)
> LLVM does a better job at code layout, and massively wins on the amount of
> executed branches (in particular unconditional jumps). With
>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107905
Alexander Monakov changed:
What|Removed |Added
Keywords|ra |
CC|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107905
--- Comment #2 from Georgi ---
> I don't see that causing a 2x slow down though.
Same here, yet, the results prove that not only on Zen2, on top of that on
i5-7200U @2.5GHz and Windows 10, this function takes 203s, cannot figure it
out, could
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107905
Andrew Pinski changed:
What|Removed |Added
Keywords||missed-optimization, ra
