[Bug tree-optimization/82604] [8 Regression] SPEC CPU2006 410.bwaves ~50% performance regression with trunk@253679 when ftree-parallelize-loops is used

[rguenther at suse dot de]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82604

--- Comment #17 from rguenther at suse dot de <rguenther at suse dot de> ---
On Thu, 18 Jan 2018, amker at gcc dot gnu.org wrote:

> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82604
> 
> --- Comment #16 from amker at gcc dot gnu.org ---
> I hope it's possible to break the dependence by reordering passes so that
> graphite/parallelization could be moved earlier.  There are several issues like
> this IIRC.

Certainly not for GCC 8.  Also given that parallelization involves
outlining of code this always introduces issues with dependence
analysis due to extra indirections and us not being able to transition
all relevant information to the outlined copy.  So I'd even move
it later... (after vectorization for example).

For this particular case dependence analysis should be enhanced to
also handle memory builtins of course.