[PATCH] Reassociate GEP operands for loop invariant code motion

[Sanjoy Das]

> Speaking of indvar widening, all I meant was, if the wider bit-width were
> not an issue and we could enable indvar widening in NVPTX, LSR would be more
> effective on our cases because ScalarEvolution needn't worry about sext/zext
> any more.

Makes sense.

-- Sanjoy