2011/4/18 Denis Chertykov <cherty...@gmail.com>: > 2011/4/18 Georg-Johann Lay <a...@gjlay.de>: > Few years ago I have played with early splitting of anything possible > (move,add,sub,and,...). The results was very bad. > It's happened because flow of splitted insns (8bits insns) becomes > unreadable for most of GCC optimisation passes. > May be splitting is appropriate before register allocation or even after > reload pass.
May be something like this (early splitting) happened with the subreg lowering. Denis.