I am still curious if the mentioned crash RtlCaptureStackBackTrace is the core issue Blue Protocol hits or rather some other bug leads to invalid unwind info. So I tried to look at that with current Proton Experimental with things from here (https://github.com/ValveSoftware/Proton/issues/9093#issuecomment-3422275851) on top without this MR (only coremessaging MR, limiting address space for syscall emulation to work with the game) and the game starts for me this way without this MR. Do I need to do anything specific to trigger the mentioned crash? Regardless, if Windows indeed triggers and catches the exceptions something between the lines of this MR might be worth it I guess, but we still need to know where exactly the crash is supposed to be caught (RtlVirtualUnwind2? WalkStackTrace?) and probably have a test for that as a part of Wine testsuite included in the patches. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/9218#note_119088