Louis,
Anything i could do to improve to get it in?
Perhaps http://source.winehq.org/patches/ shows that you should get rid of the one apply failure.
Basically, your patch does the right thing, as MSDN says "If times is a NULL pointer, the modification time is set to the current local time." -- like POSIX.
-#if(0) /* FIXME: this test crashes wine, remove this line when wine is fixed */
People here prefer if(0) so that the compiler can still validate the form of the code. I've been even been using if (winetest_debug>1) so I can direct testbot to run the additional test.
Finally, why not merge the two? After all, it's one change, and several tests.
Why do you call it test_utime() since you don't verify the effects of utime()?
Regards, Jörg Höhle