[Bug 46099] Star Citizen not loading after implementation of WaitOnAddress () in wine

[wine-bugs at winehq.org]

https://bugs.winehq.org/show_bug.cgi?id=46099

--- Comment #7 from Tool <codedonewell at gmail.com> ---
I have been trying to explore the problem - hoping that I might even be able to
contribute here.  My last approach was to compare the working log for a thread
in 3.18 to the log for the same thread that does hang in 3.19.
I have attached thread logs for the same thread at the same phase of loading
Star Citizen.
I have identified these threads by following what threads are created by the
StarCitizen.exe launch process. These logs are for the thread that is launched
for the same purpose.  By following the various activities (exception traces,
SRWLock activity, Virtual memory activity) I believe I have found a different
method being used in each version to achieve the same goal.
The 3.19 version shows the hang at the end, however, the first bit that I need
help with is that I cannot figure out:
- why there were no WaitOnAddress "stub" messages in the 3.18 log; and
- how the different function was selected in each version.
I have 20+ years experience with C (OK a fair chunk of Java in there), but only
about a year of that with Win32.  I am totally willing to keep trying things to
try to find the problem, but I'm not sure where to take it - and my confidence
is somewhat dented by not being able to figure out the above 2 items.
If anyone has theories that they would like me to test - I'll do it!
Need different logs? Done! (Warning the full log in the hang case is 1.4GB
uncompressed)
My tests were conducted with WINEDEBUG=fixme+all,trace+all,trace-heap,+relay
I can provde my RelayExcludes if asked - they are big.
The main log was split into thread logs with an awk script (also available on
request), and some extra relay messages have been filtered by that.

-- 
Do not reply to this email, post in Bugzilla using the
above URL to reply.
You are receiving this mail because:
You are watching all bug changes.