Makes sense. Maybe we can create a test dat file for the windows builds
that only targets the sim configs we care about? Maybe another for a
set of
ARM targets? Thanks for helping me understand what's going on.
I think we should have wpcap working windows. It is an important
feature and it is a bug that it is broken. But I don't think there is
anyone with strong enough motivation to tackle that project right
now. I did the original code, but it was a leverage from some of Adam
Dunkels code and I never got it working originally (although a user
claims to have done so many years ago). It was a major re-design of
Adam's code, of course, since nothing from uIP 1.0 is really
compatible with NuttX, but it was enough of a leverage that I never
really needed to understand the wpcap interface.
That is all bad news. But the good news is that aside from wpcap,
everything looks good
Other good news is that the wpcap errors are only related to collisions
of definitions and declarations in Cygwin header file. I see no hard
coding problems in the logs. There is a lot of discussion about this on
the internet. I won't bore you with a long sequence of "what I found on
Google today", but if you google for "cygwin w32api header file
conflicts" you come up with a lot of discussion of these header file
conflicts. This is not a NuttX problem, it is a programming environment
problem.
It seems that, in general up_wpcap.c should NOT include the standard
networking files like in.h, socket.h, etc. it should rely instead on
the Windows interfaces in /usr/include/w32api. These provide the same
standard (abut incompatible) networking definitions, but customized for
use with the direct Windows networking interface used by wpcap.
Greg
