[Bug tree-optimization/52272] [4.9/5/6 regression] Performance regression of 410.bwaves on x86.

[amker at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=52272

--- Comment #31 from amker at gcc dot gnu.org ---
(In reply to wuxiang from comment #30)
> (In reply to amker from comment #29)
> > I think this is fixed.
> 
> Hi, I saw your patch, and patched it on gcc-4.8.4，but no improvement of
> 410.bwaves on x86_64 with gcc-4.8.4. 
> > The patch improves 410.bwaves by 3-4% on x86_64.
> Does this just test with gcc trunk before and afer the patch?

Yes, it was tested against trunk.  The code/data-structure have been changed a
lot since gcc 4.8, it may need some non-trivial change to rebase against 4.8,
but I think the basic idea described by comment#5 still holds.