http://bugs.winehq.org/show_bug.cgi?id=17195
--- Comment #123 from Adam Martinson adam.r.martinson@gmail.com 2013-01-03 10:00:18 CST --- (In reply to comment #119)
You can (and should) split it some more. Changing flush deserves its separate patch, and so does changing ioctl.
You mean split patch 5 more?
(In reply to comment #120)
there's a way round the "requirement" that tests be run, which is to add unnecessary code that checks - at runtime - whether the old functionality is to be used or the new functionality is to be used, and to have the choice be made either as a compile-time switch or even as a configuration option.
I don't think Alexandre would go for that without some compelling reason.
(In reply to comment #121)
btw adam did you also check that this patch works on FreeBSD?
No, but there's no reason it shouldn't. I'm not using any system calls, etc that Wine doesn't already use.
(In reply to comment #122)
oh wait! event is also being renamed to event_empty which is highly significant, and buried within the renaming.
Why is that significant? Empty events are the only thing it's used for. In a later patch an event_avail gets added, I just renamed it to avoid confusion. The structs in here are just for the server, the renames don't affect anything outside of that file.