[PATCH] iphlpapi/tests: Replace a test about a wrong assumption with a comment

[Detlef Riekenberg]

The order of the interfaces is not fixed and can change,
That is already commented at the top of the file:
https://source.winehq.org/git/wine.git/blob/c6ff0e01224e86c7baa31d019fcb0a7bf03e2f96:/dlls/iphlpapi/tests/iphlpapi.c#l26

When upgrading my machine to Windows 10, the Windows-updater failed to install a driver
for the ethernet port and for the wifi port
The loopback was the first interface and is the first interface
after installing the missing drivers from an USB stick

Example test failures
From the 90 failure package after updating from win7 to win10:
http://test.winehq.org/queue/err5iRR8/report

Patch tested:
https://testbot.winehq.org/JobDetails.pl?Key=50532

BTW: With a linux live system, the ethernet port works out of the box without any user input
and the wifi port just needed the wifi credentials.

Linux driver support beats Windows 10 in this configuration 2:0

--
bye bye ...
        ... Detlef

Signed-off-by: Detlef Riekenberg <wine.dev at web.de>
---
 dlls/iphlpapi/tests/iphlpapi.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/dlls/iphlpapi/tests/iphlpapi.c b/dlls/iphlpapi/tests/iphlpapi.c
index 1be9f3c5fa..369d38dadb 100644
--- a/dlls/iphlpapi/tests/iphlpapi.c
+++ b/dlls/iphlpapi/tests/iphlpapi.c
@@ -285,9 +285,10 @@ static void testGetIpAddrTable(void)
           ok (buf->table[i].wType != 0, "Test[%d]: expected wType > 0\n", i);
           trace("Entry[%d]: addr %s, dwIndex %u, wType 0x%x\n", i,
                 ntoa(buf->table[i].dwAddr), buf->table[i].dwIndex, buf->table[i].wType);
-          /* loopback must never be the first when more than one interface is found */
-          if (buf->table[i].dwAddr ==  htonl(INADDR_LOOPBACK))
-              ok(buf->dwNumEntries == 1 || i, "Loopback interface in wrong first position\n");
+          /* a removed test checked, that loopback was not the first interface,
+             when more than one interface was found,
+             but that assumption about the interface-order is wrong
+             (example machine, where the old test failed: dr-i7-HD3000-x64 / -wow) */
         }
       }
       HeapFree(GetProcessHeap(), 0, buf);
--
2.21.0.windows.1