https://bugs.winehq.org/show_bug.cgi?id=57758
--- Comment #7 from Paul Gofman pgofman@codeweavers.com --- Thanks for the info! I studied the logs, and I don't see any clue how the commit can be related to the observed crash in log. The uses of find_basename_module() which was changed through the blamed commit are pretty limited and those cases should always leave traces in logs. I studied probably each one of them and each of those seem to provide expected result. If the previous related commit was blamed I could suspect that the game is doing something behind the scenes with introduced hash links which would obviously not reflected in logs but that can't explain the attribution to the blamed commit.
So one guess is maybe there is something regressed it on top and on current Wine it crashes earlier due to a different reason? So, the additional questions:
1. Does the game work if the blamed commit is reverted on top of current Wine? If it does, could you please attach the same logs with the commit reverted on top?
2. Can we add +system,+win,+x11drv,+event,+msg,+ddraw,+d3d,warn+heap logs to the current set for reproduction case (that would be WINEDEBUG=+loaddll,+pid,+timestamp,+seh,+unwind,+threadname,+module,+ddraw,+system,+win,+x11drv,+event,+msg,+ddraw,+d3d,warn+heap)? Note that warn+heap is not just a logging channel but enables heap debugging; if it changes something with reproduction details that is interesting info by itself but also worth running with that one excluded.