Hello, On Tue, 1 Aug 2023, Joseph Myers wrote:
> > Only because cmpxchg is defined in terms of memcpy/memcmp. If it were > > defined in terms of the == operator (obviously applied recursively > > member-wise for structs) and simple-assignment that wouldn't be a problem. > > It also wouldn't work for floating point, where I think clearly the atomic > operations should consider positive and negative zero as different, and > should consider different DFP quantum exponents for the same real number > as different - but should also consider the same NaN (same payload, same > choice of quiet / signaling) as being the same. That is all true. But the current wording can't work either. It happily requires copying around memory between types of different representations and sizes, it makes padding observable behaviour, and due to that makes basic algebraic guarantees not be observed (after two values are checked equal with the predicates of the algrabra, they are not then in fact equal with predicates of the same algebra). Ciao, Michael.
