Kay Röpke wrote: > > On Mar 8, 2009, at 4:14 AM, Monty Taylor wrote: > >>> ok, i can't resist the urge to feed the troll (sorry, monty, but it's >>> trolling) >> >> Yes yes... I assume people know I'm trolling whenever I mention OSX... > > i know that and neglectfully omitted the ';)' > >> on the other hand, THANK YOU for this useful information that I'm quite >> happy to learn. (I would love to ditch support for 4.0.2 as long as that >> doesn't mean losing OS X devs - troll though I might be, ) > > > just for my benefit: what is the deal with 4.0.2? does it pertain to C++ or > is it of general nature?
A couple of things: - It give some warnings (which for us are errors) which are actually truly spurious and incorrect. I can't think of an example at the moment, but I believe we had some issues a while ago with "this could be used uninitialized" and "this is unreachable" - neither of which were true. :) - 4.2 got a lockless tr1::shared_ptr - 4.0 doesn't have gcc atomics support Other things I'm forgetting... I know we have more than once pushed code that tests fine elsewhere only to have it vomit at the last minute on Brian's Mac laptop. I tested the "force 4.2 on Mac" patch last night... works like a charm! I now owe you a beer. Monty _______________________________________________ Mailing list: https://launchpad.net/~drizzle-discuss Post to : [email protected] Unsubscribe : https://launchpad.net/~drizzle-discuss More help : https://help.launchpad.net/ListHelp
