On Thu, Jan 6, 2011 at 9:28 PM, Anthony Williams wrote:

"Hicham Mouline" writes:

...

in Release build in Xcode, gcc 4.2.1,

inline void condition_variable::wait(unique_lock<mutex>& m) { ...  int const res = pthread_cond_wait(&cond, &internal_mutex);  BOOST_ASSERT(!res);  ... }

Unused variable 'res'

is gcc mistaken?

Yes and no :-)

In the release build the assert compiles to nothing, so res is no longer referenced, even though it is referenced in the source.

I actually have a patch for this: https://svn.boost.org/trac/boost/ticket/4818 which I don't think has been merged into (or acted upon) yet. HTH -- Dean Michael Berris about.me/deanberris

On Fri, Jan 7, 2011 at 6:54 AM, Anthony Williams wrote:

Dean Michael Berris writes:

...

I actually have a patch for this: https://svn.boost.org/trac/boost/ticket/4818 which I don't think has been merged into (or acted upon) yet.

I've changed it to throw rather than assert, for consistency with condition_variable_any. This should get rid of the warning too.

Cool! Thanks Anthony! (even though my patch didn't solve it) :-D -- Dean Michael Berris about.me/deanberris