On Mo, 2006-11-20 at 11:22 +0100, Florian Echtler wrote:
Welcome to wine.
the software was sending UDP packets to IP _254_.255.255.255 (note the first octet). After some experimentation, I noticed that while Windows just sends this as a broadcast packet,
Are you sure, that the Function on Windows also get this wrong IP? If that is the case, then a fix is needed for the customized tracking software. There are various API-Hook Tools on the Net.
Wine is also a great Software to make Application-Bugs visible.
Did you already create a simple test-App, that used the IP 254.* and verified, that Windows interpreted that as Broadcast? A test for the wine testsuite is always prefered to verify, what Windows does.
I therefore build a small patch which checks for these strange adresses and replaces them by broadcast adresses.
You modify the beavior for a complete "/4" - Subnet. Is this really the same thing, what Windows does? Reimplement a Windows-Bug will require more tests.
It's also Possible, that a different Function in wine malformed the IP. You need some tests to verify this.
However, the question now is whether it makes sense to include this patch into Wine?
Even, when tests verified, that Windows is not RFC compliant for this case, the tracking software should be fixed.
Your Patch need some polish: - SPACE/TAB mixing - No C++ comments allowed - fprintf is wrong here. We have macros for this: (include/wine/debug.h) -
as I have not subscribed to the list.
Please subscribe, it's easier for you to follow the discussion.
-
Detlef Riekenberg