[Bug rtl-optimization/78200] [7 Regression] 429.mcf of cpu2006 regresses in GCC trunk for avx2 target.

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78200

--- Comment #6 from Richard Biener <rguenth at gcc dot gnu.org> ---
So the non-canonical GIMPLE is actually created by loop versioning and
update-ssa
replacing uses with new defs but not re-canonicalizing operand order.  Not
gimplification as I speculated.