On Oct 17, 2010, at 2:07 PM, Kyle Sluder wrote: > On Sun, Oct 17, 2010 at 1:29 PM, Velocityboy <velocity...@rodentia.net> wrote: >> This is actually not always true. If you have an non-static inline in a >> header file, the compiler is pretty smart about it. You can see this in >> action if you generate the assembly and look it it. This: > > IIRC there's no guarantee that a two references to the same non-static > inline function will have the same address even within the same > translation unit, depending on optimization level, order of appearance > in the source file, and context in the call site. But that might be a > Microsoftism, or just a complete fabrication of my mind. > > --Kyle Sluder
Actually it's worse than that - I went back and looked again and when the compiler converts an inline into a non-inline function for the purposes of getting a pointer to it, the function it creates is not anonymous. So if you take the address of an inline in more than once translation unit, you end up with a multiply defined symbol at link time. I haven't ever come across this situation in non-hypothetical code, so I don't remember what the spec has to say about it, and I don't have a copy handy. Moral of the story is that it's not safe to take the address of an inline. _______________________________________________ Cocoa-dev mailing list (Cocoa-dev@lists.apple.com) Please do not post admin requests or moderator comments to the list. Contact the moderators at cocoa-dev-admins(at)lists.apple.com Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/cocoa-dev/archive%40mail-archive.com This email sent to arch...@mail-archive.com
