[plt-scheme] concurrency bug in tcp-connect

> I modified your patch, slightly: I moved the strlen() test back to the
> top of the function (since one branch of the test escapes), but I kept
> the strcpy() under the lock.

Oops.  The result of my cowboy fix-bug-at-any-cost mode.  You know, I looked
at that branch and wondered whether I needed to worry about the deallocation
of the GHBN_Rec structure, but then noticed it wasn't being explicitly
deallocated later anyway (the wonders of conservative collection, I take
it).  But I completely spaced on the fact that the lock would be kept in the
event of an early return.  I'm off to shave my head...

Anton