Gabriel Paubert wrote:
For all architectures I use, it is rather "if used as an operand
of an arithmetic instruction", but the values can be copied
around without ever generating a trap. Even negating or taking
the absolute value never traps since those are not considered
"arithmetic" instructions.
On some architectures the load causes a trap, and that means
for example that you cannot use fpt load/store to move non-fpt
stuff. But yes, of course the description is approximate, since
the behavior of signalling nan's is unspecifed by the standard!
And even then you have to explicitly enable the trap for invalid
operation (on systems using IEEE754), otherwise it is simply
propagated as a QNaN.
That's not guaranteed
