https://gcc.gnu.org/bugzilla/show_bug.cgi?id=34416
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Target Milestone|--- |13.0
Status|NEW |RESOLVED
Resolution|--- |FIXED
--- Comment #6 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
> - scalar cleanup after final inlining should be done before
> computing aliasing (basically, re-run what we do for early
> optimization also after final inlining, but at least run
> copyprop, ccp, forwprop and dce)
That was done with r0-89162-g11b08ee9118d10 (4.4.0)
I also have a start for self-DCE for the vectorizer. See PR116711 (I hope to
finish that up for GCC 16), it will improve compile time.
This is a generic bug without any other clear goal. I do think we need to
re-tuning of the gimple pipeline but I am not sure if there is any obvious way
right now. Plus we already had a few other re-tuning since GCC 4.4 too.
