The comment above is correct, but the code to calculate the mask was broken.
Fixes: 0e6581b8 ("nir/algebraic: Reassociate shift-by-constant of shift-by-constant") Cc: @idr