[racket] Problems trying to do non-blocking I/O and a reactor loop

Matthew Flatt wrote at 08/04/2012 12:33 PM:
> At Fri, 3 Aug 2012 20:51:37 -0400, Greg Hendershott wrote:
>    
>> 3. Bad news: I just got it to happen again, with 5.3.0.16:
>>      
> I see how that could happen, and I've pushed a repair --- mostly by
> improving the guarantees about progress evts. (I think it's too late to
> include this repair in v5.3, though.)

If Greg's "bug-example.rkt" is perfectly legitimate, and, to survive a 
one-hour "ab" burn-in test, the test needs a Racket fix that will not be 
in the 5.3 release... would it make sense to have a 5.3.1 urgent bugfix 
release?  (Or is Greg's code doing something unusual, and there is a 
reasonable code change he can make to avoid the problem?)

Reason I ask is that I might want to move some production servers to 5.3 
to get those other bugfixes (not yet sure), but I similarly might also 
need this latest bugfix of yours.  (For production servers, released 
versions are easier to specify and reproduce; non-released versions 
might sometimes be necessary, and some organizations have extra 
infrastructure and process set up for reproducibility with their own 
custom builds, but it's not something one wants to do if one doesn't 
have to.)

Neil V.