Bug 1987097

So this request is pretty easy to fulfill, although like any engineering request it requires developer and QA time, and there's not enough of that remaining for 8.5, so it would need to be for 8.6 if we were to accept it. That said, it's an odd request because this macro does basically nothing. It's only used to suppress -Wcast-function-type warnings. So the usual solutions would be to just not use it. E.g.:

-    item->update_properties_timeout = g_timeout_add (10, G_SOURCE_FUNC (update_all_properties), item);
+    item->update_properties_timeout = g_timeout_add (10, update_all_properties, item);

The worst that could happen is a -Wcast-function-type warning. If you want to avoid that warning, you could simply build with -Wno-cast-function-type. Or you could do the cast manually:

-    item->update_properties_timeout = g_timeout_add (10, G_SOURCE_FUNC (update_all_properties), item);
+    item->update_properties_timeout = g_timeout_add (10, ((GSourceFunc) (void (*)(void))update_all_properties, item);

Although I haven't tested it, I suspect a much simpler cast would work just fine:

-    item->update_properties_timeout = g_timeout_add (10, G_SOURCE_FUNC (update_all_properties), item);
+    item->update_properties_timeout = g_timeout_add (10, ((GSourceFunc)(gpointer)update_all_properties, item);

Or you could define it in your code:

#ifndef G_SOURCE_FUNC
#define G_SOURCE_FUNC(f) ((GSourceFunc) (void (*)(void)) (f))
#endif

All of these workarounds are extremely easy, so I'm hesitant to ack this one.

I'm going to nak this one for RHEL 8. The new API will be available in RHEL 9 though.

Development Management has reviewed and declined this request. You may appeal this decision by reopening this request.