The patch looks reasonable to me.  I've added Yuri to nosy since he's responsible for the signature code in inspect, and he might have an opinion on the appriateness of the fix.

I think this special case needs to be documented, since currently the name is documented as needing to be a valid python identifier.

Sounds like we don't have a good answer to Ned's question about a 2.7 workaround.  On the other hand, "fragile byte code inspection" is not really fragile, since the 2.7 bytecode will never change.  The real question is if someone cares enough to do the work, since as you say (and Ned admits) the benefit is small.