Comme=
nt # 16
on bug 10692=
8
from Roland Scheidegger
(In reply to Dave Airlie from comment #14)
> I think Roland's first patch is correct, just ca=
ll fold_alu_op2 if we get
> back 2 sources from fold_assoc. return true is for when we've finished=
all
> folding on that instruction, so I don't think that's correct without c=
alling
> fold_alu_op2
Well it would still be called (in fold_assoc instead) in this case so the o=
nly
difference in the end really is that fold_alu_op3 would return false instea=
d of
true if fold_alu_op2 (called when fold_assoc reduced number of src to 2)
returned false.
If the return value is just indicating if we're finished with folding, then=
I
suppose it does not actually matter for correctness either way
This code was all written by Vadim Girlin, but I don't think he's still aro=
und
here.