> > If the initializer is constant, but the member value that's being > initialized has a > non-trivial constructor with a side effect, your patch will inhibit the > warning > but the program will not behave the same as if reordering had not happened. > > Peter >
Yes. It sounds unlikely. But not impossible of course. I could also make sure the member variables are POD types before I inhibit the warning. I just have no idea how I check if a member is POD. But I could investigate this. I think, this will still remove most of the -Wreorder warnings that I get. Best regards, Daniel
