[tz] Minor (unimportant really) technical UB bug in strftime() ?

Clive D.W. Feather clive at davros.org
Thu Nov 10 10:15:36 UTC 2022

Guy Harris said:
> >    trap  representation.
> Does the C specification allow for a C implementation on a hypothetical load-store architecture machine in which:
> 	a byte is 8 bits plus an "initialized bit";
> 	the "uninitialize" instruction has a memory-location operand, and clears the initialized bit;
> 	a store instruction set the initialized bit on all bytes to which it stores;
> 	a load instruction traps if any of the bytes from which it's loading doesn't have the initialized bit set?


My memory is that we agreed "no hidden bits" when writing that wording.

I do recall that parity bits were discussed and we agreed that they weren't
part of the representation of a value.

Of course, an implementation could do what you suggested provided that
there was a way of disabling it so as to produce a conforming

Clive D.W. Feather          | If you lie to the compiler,
Email: clive at davros.org     | it will get its revenge.
Web: http://www.davros.org  |   - Henry Spencer
Mobile: +44 7973 377646

More information about the tz mailing list