The first argument to the function is an ASCII string. The result of normalizing
an ASCII string is always the same as the input string, regardless of the
normalization form used. 

The spec says:

(XPath 3.1.5): a function need not evaluate an argument if the function can
evaluate its body without evaluating that argument.

(XPath 2.3.4): If, at such an intermediate stage of evaluation, a processor is
able to establish that there are only two possible outcomes of evaluating Q,
namely the value V or an error, then the processor may deliver the result V
without evaluating further items in the operand E. 

In this case it is not necessary to evaluate the second argument once it is
known that the first argument consists entirely of characters that are
unaffected by any of the normalization algorithms.

This might seem rather pedantic, but an implementation that doesn't report an
error here is conformant under these rules, so it would be better to change the
first argument to a non-ASCII string to ensure interoperable results.