http://bugs.winehq.org/show_bug.cgi?id=11802
--- Comment #7 from Mike Nelson miken700@yahoo.com 2008-03-03 21:20:09 --- (In reply to comment #5)
(In reply to comment #3)
Here are the results of the regression test below. I'm not 100% on this, as it's not 100% reproducible.
If you aren't sure, then try reverting the patch (or compile right before this patch if it won't compile cleanly).
$ git revert c19af910ebafa9b1928b0f67bee68f6b0b237e8e $ ./configure --verbose && make clean && make depend && make
git revert couldn't cleanly un-merge that patch from HEAD and the changes are non-trivial, at least to me. But now it looks like that is probably not the problem patch anyway.
I tested before and after this supposedly bad check-in and I was able to repro the issue on both "good" and "bad" versions. After another regression from that "good" revision back to wine-0.9.53 I found that I could indeed repro it on wine-0.9.53 as well (after many many runs). It just seems to happen a lot more often on the more current versions. I'm a bit at a loss now. This may not be a regression issue after all. Seems like some sort of timing issue during startup.