In nfs_send, after the ->so_send call, the case of ENOBUFS on a
datagram socket is checked for; in that case a retransmission is
arranged for.

I've also seen, on 3.1, using NFS over TCP, an error 35, which would
be EAGAIN; this is not checked for.  It's not obvious where that error
is coming from, and I'm not entirely sure what it even means in this
context.

I've seen this exactly once in recent memory, so it's not a big deal,
but if it points out some trivially fixable oversight, that'd be nice.

-- 
(let ((C call-with-current-continuation)) (apply (lambda (x y) (x y)) (map
((lambda (r) ((C C) (lambda (s) (r (lambda l (apply (s s) l))))))  (lambda
(f) (lambda (l) (if (null? l) C (lambda (k) (display (car l)) ((f (cdr l))
(C k)))))))    '((#\J #\d #\D #\v #\s) (#\e #\space #\a #\i #\newline)))))