[vtk-developers] Proposed change to vtkSocket.cxx

[Brad King]

Moreland, Kenneth wrote:
>> Good catch.  However, I don't think there needs to be a limited attempt
>> count or a usleep to yield.  It can just try again immediately.  This is
>> a common idiom on posix system calls.
> 
> That makes sense.  There already exists a similar limited attempt
> count and sleep for Windows.  Is this also unnecessary, or are the
> circumstances sufficiently different for the WSAENOBUFS failure?

I don't know enough about the windows API off the top of my head.
You'll have to check the documentation and look for recommended idioms.

-Brad