On Fri, Oct 11, 2013 at 4:57 PM, Ian Pilcher <arequipeno@xxxxxxxxx> wrote: > I've run into a situation where I'm getting a spurious (I think) "may be > used uninitialized" warning. > > https://bugzilla.redhat.com/show_bug.cgi?id=1018422 > > In this case, the variable is initialized when its address is passed to > another function. > > This situation made me wonder if there's some sort of function (or > parameter) attribute that I could use to inform the compiler that the > function does, in fact, initialize the variable. This does look like a bug. There is no need for such an attribute. When the compiler sees that a pointer to the variable is passed to an unknown function, it assumes that the variable has been initialized, which is really the only reasonable behaviour. Consider filing your bug report in the GCC bug queue, as described at http://gcc.gnu.org/bugs/ . I have no idea what happens to GCC bugs filed in the redhat bug queue. Ian
