On Fri, Apr 06, 2007 at 11:38:50AM +0100, Rob Quill wrote:
> So the general concensus is that's it's not worth doing?
> Hypothetically, if I did it and it didn't make much difference, would
> it be worth submitting a patch? Or should I just give up before I
> start?
It might be worth doing. I think that, in addition to a patch,
I'd like to see measurements (maybe just the size increase in
libstdc++.{a,so}). If the cost is small, I will not object.
Seeing what the debugging sessions look like before and after
will help assure that we're getting it right, especially at
the boundary: if we have
23 Foo foo = func();
24 int bar = 0;
then if a breakpoint is set at line 23, foo should not be in scope,
but at line 24, it should be.