Incorrect consideration of global and local variables in resolving waitfor mutex object.

[Thierry Delisle]

The following code :

#include <monitor.hfa>
monitor M { condition e; } m;
void rtn( M & mutex m );
void bar( M & mutex m ) {
 	waitfor( rtn, m ); // not ambiguous, select parameter
}

Produces the following error:

cfa test.c
CFA Version 1.0.0 (debug)
test.c:92:1 error: Ambiguous function in call to waitfor

If mangling is removed for mutex then the following code encounters widening problems:

//----------
struct monitor_desc {};

struct monitor_guard_t {};
void ?{}( monitor_guard_t & this, monitor_desc ** m, int count, void (*func)() );

struct monitor_dtor_guard_t {};
void ?{}( monitor_dtor_guard_t & this, monitor_desc ** m, void (*func)() );

//----------
forall( dtype T | sized(T) | { void ^?{}( T & ); } ) void delete( T * ptr );

//----------
monitor test {};

int main() {
	test * t = 0;
	delete(t);
}

Note the monitor_guard_t and monitor_dtor_guard_t structs are needed because they are expected by the code handling the mutex and monitor keywords.

[a3moss]

The source of this bug is that the local parameter m does not hide the global m. This is because the local m mangles to a different name because it is mutex. Having some ability to overload based on mutex is useful, so we can't necessarily just take it out of the mangler, but it should in principle be possible to modify the mangler so that top-level mutex qualifiers don't affect mangling.

[Thierry Delisle]

The first part of fixing this is to remove the mangling on mutex, I believe mangling that way can only lead to problems. After that however, there is a resolver bug that requires exact matching for assertion regardless of widening rules and I believe that needs to be changed.