Regarding this question
What are the evaluation order guarantees introduced by C++17?
With this specification
http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2016/p0145r3.pdf
And this text from the specification
Furthermore, we suggest the following additional rule: the order of
evaluation of an expression involving an overloaded operator is
determined by the order associated with the corresponding built-in
operator, not the rules for function calls.
Does this mean that these two expressions are no longer equivalent?
a << b;
operator<<(a, b);
As the second one looks like a function call, hence there is no guaranteed evaluation order in the parameters?
"As the second one looks like a function call, hence there is no guaranteed evaluation order in the parameters?"
Indeed. [expr.call]/5 contains an example specifically covering the difference between the two cases covered in your question [emphasis mine]:
The postfix-expression is sequenced before each expression in the
expression-list and any default argument. The initialization of a
parameter, including every associated value computation and side
effect, is indeterminately sequenced with respect to that of any other
parameter.
...
Note: If an operator function is invoked using operator notation,
argument evaluation is sequenced as specified for the built-in
operator; see
[over.match.oper].
[ Example:
struct S {
S(int);
};
int operator<<(S, int);
int i, j;
int x = S(i=1) << (i=2);
int y = operator<<(S(j=1), j=2);
After performing the initializations, the value of i
is 2
(see
[expr.shift]), but it is unspecified whether the value of j
is
1
or 2
.
— end example ]