int vs. bool / _Bool (Was: Re: Committed: Fix distribute_loop)

[Joern Rennecke <amylaar at spamcop dot net>]

[This started on gcc-patches]
Quoting Richard Guenther <richard.guenther@gmail.com>:

> On Sat, Jan 23, 2010 at 2:37 PM, Paolo Bonzini <bonzini@gnu.org> wrote:
> > On 01/23/2010 01:36 PM, Richard Guenther wrote:
> > > Ugh. ÂI wonder if we should use _Bool instead of int
> > > if the host compiler supports it.
> >
> > Then we'd have another share of bugs if someone uses
> >
> > Âbool a = <something evaluating to 2>;
> > Âa &= <something evaluating to 1>;
>
> We do have those already.  I run over this recently:
>
>   bool all_critical_edge_p = true;
>   all_critical_edge_p &= e->flags & EDGE_CRITICAL;
>
> with _Bool this would work as expected, with ints it doesn't.  With
> C++ it would work as well.
>
> Another option would be to stop using bool - as our bool doesn't have
> the same semantics as bool in C++ or _Bool in C99.  Which can
> be surprising.

I think a better long-term plan would be to add an option to insert run-time
checks to abort when a integer other than 0 or 1 is converted to bool / _Bool.

We could then use this in gcc builds when enable-checking is active.

Or we could add a warning option for any int -> bool/_Bool conversion, but
this could cost some extra GCC runtime overhead as some places will need to
add some_integer != 0 checks to silence the warning, even though some_integer
might be known to always be either 0 or 1.