Re: Undefined Behavior Optimizations in C

On Tue, 10 Jan 2023 17:32:28 +0100, David Brown
<david.brown@hesbynett.no> wrote:


[ UB example: generated code assumes bool parameter is only 0 or 1 and
mishandles anything else, but misdeclared call passes 2 ]


> This is clearly wrong - clearly undefined behaviour. And the result
> would be a formatted disk. But there is nothing wrong with the
> compiler's generated code. I have seen other occasions when compiler's
> have made code with booleans that appear to be both true and false, or
> neither true nor false, as a result of undefined behaviour setting the
> underlying memory to something other than 0 or 1, simply because that
> was the result of the most efficient code.
>
FWIW -- back in the 80s the VAX FORTRAN compiler checked only the _low
bit_ of a LOGICAL variable (stored as a byte or a 4-byte word, I
forget which) because this was faster. People often reused (scarce and
expensive) memory then and such a variable might accidentally get set
to a value other than 0 or 1, producing surprising/confusing results.


In those days DEC was vigorously opposing C (they had BLISS instead)
and Unix, so there was no DEC C compiler, though I'm pretty sure there
was a DECUS (user group) one, and of course there was not yet a C
standard at all much less one including bool.

Return to the comp.compilers page.
Search the comp.compilers archives again.