Because it would mess up the integer pipeline, and the alpha designers felt
(probably correctly) that integer divide wasn't important enough to make the
pipelines more complex.
They could have done it with the FPU pipeline which more-or-less has the
required hardware already, but then the FPU pipeline would have needed data
from the integer register file, and that would have messed up the clean
register file separation.
So idivs are done in software, with little (but as you found out, some)
degradation. Thanks to that the rest of the chip is simpler and thus easier
to do a fast implementation,
Linus