On Thu, 11 Sep 2014, Marc Glisse wrote: > On Thu, 11 Sep 2014, Richard Biener wrote: > > > + /* We can't reassociate floating-point or fixed-point plus or minus > > + because of saturation to +-Inf. */ > > + (if (!FLOAT_TYPE_P (type) && !FIXED_POINT_TYPE_P (type)) > > Do you remember if there was a particular reason not to add > || flag_associative_math > when this was in forwprop?
No idea honestly. The initial drop-in was supposed to mitigate somewhat that tree-ssa-reassoc.c doesn't associate TYPE_OVERFLOW_UNDEFINED expressions at all, so it was designed as "integer only" Richard.
