Ok in my ever increasing search to figure out why EQ2 is getting a unhandled exception and also a no ipixel error. (I am convinced this is two separate bugs now)
Bug One :
This is the unhandled exception: If I have EQ2 run in anything but win98 mode (which I don't think is supported anymore by EQ2) wine dies with a unhandled exception. If I set to win98 it ends with a no ipixel format error message box (the second bug) after it begins to draw the window. The other thing is the msg box is not a typical msg box saying that something isn't supported but looks to be an exception in ClientApp.cpp which means they are getting some unhandled exception while drawing the window.
What I think might be going on based on this :
trace:win:WIN_SetWindowLong 0x2002c 0 0 W trace:win:WIN_SetWindowLong Fell Through trace:win:WIN_SetWindowLong Release Ptr trace:win:WIN_DestroyWindow 0x2002c trace:win:WIN_DestroyWindow 0x1002e trace:win:WIN_DestroyWindow 0x10030 trace:win:WIN_DestroyWindow 0x10032 trace:win:WIN_DestroyWindow 0x10034 trace:win:WIN_DestroyWindow 0x10036 trace:win:WIN_DestroyWindow 0x10038 trace:win:WIN_DestroyWindow 0x1003a trace:win:WIN_DestroyWindow 0x1003c trace:win:WIN_DestroyWindow 0x1003e trace:win:WIN_DestroyWindow 0x10040 trace:win:WIN_DestroyWindow 0x10042 trace:win:WIN_DestroyWindow 0x10044 trace:win:WIN_SetWindowLong 0x2002c 0 0 W
==================================================================== These are my trace additions in SetWindowLong to help me follow the execution of the code
trace:win:WIN_SetWindowLong Fell Through trace:win:WIN_SetWindowLong Release Ptr trace:win:WIN_SetWindowLong 0x2002c 12 0 W trace:win:WIN_SetWindowLong Fell Through trace:win:WIN_SetWindowLong Server Start Request trace:win:WIN_SetWindowLong Default trace:win:WIN_SetWindowLong no call Err trace:win:WIN_SetWindowLong Default get and Set Win Data trace:win:WIN_SetWindowLong Release Ptr trace:win:WIN_SetWindowLong Returning 1871960
CRASH! Unhandled Exception
I think the window is being destroyed in reverse for some reason. Thats just my gut first feeling based on the above trace and the exception output. This is where my knowledge on Windows behavior ends. Does 98 destroy the windows chain differently than XP? This would explain why the exception doesn't occur in 98 but does in XP.
Second Bug: This is the ipixel problem which I haven't figured out how to debug whats going on there, but think it might be related to this error:
libGL error: drmMap of framebuffer failed (Cannot allocate memory) libGL error: reverting to (slow) indirect rendering libGL error: drmMap of framebuffer failed (Cannot allocate memory) libGL error: reverting to (slow) indirect rendering
Which I am not sure why this is occurring yet but one bug at a time. Where would someone look to trace along the window draw chain to see where it dies. Some of the debug flags cause my machine to lock up which I know is a driver issue and other lockups could be caused by the above exception messing something up in X as the disk still runs when that lockup occurs but X is DOA.
Thoughts????? Ideas???? Suggestions??????
Chris