[Bug 5007] Star Wars ep I Racer "No 3D hardware found"

[Wine Bugs]

http://bugs.winehq.org/show_bug.cgi?id=5007


stefandoesinger(a)gmx.at changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |stefandoesinger(a)gmx.at




------- Additional Comments From stefandoesinger(a)gmx.at  2006-17-06 09:30 -------
I've debugged that problem a bit and I have the impression that windows 
DirectDraw blatantly violates the COM rules(and the ddraw documentation is 
wrong). Basically what ep1 racer does is

1) CreateDirectDraw to create a DirectDraw 1 interface
2) QueryInterface the DDraw 1 interface for a DDraw 4 interface
3) Release the DDraw 1 interface and expect the returned refcount to be 0
4) Continue working with the DD4 interface

The COM rules say that the DD1 and DD4 interface refer to the same object and 
that QueryInterface increses the refcount of the object of which the interface 
is returned. So step 2 increases the refcount to 2 for both DD1 and DD4 and 
accordingly 3 has to return 1, not 0. A quick and dirty test showed that 
windows returns 0 instead.

This needs a proper test case and then an idea how to implement that. I'd say 
this is a programming sloppyness in ep1 racer:

hr = CreateDirectDraw(..., &DD1, ...);
if(hr != DD_OK) cry_bloody_murder();

hr = QueryInterface(DD1, DD4, ...);
if(hr != DD_OK) cry_bloody_murder();

hr = GetCaps(DD1);
if(hr != DD_OK) cry_bloody_murder();

hr = Release(DD1);
if(hr != DD_OK) cry_bloody_murder(); <--- DD_OK is (HRESULT) 0, but release 
returns a ULONG, not a HRESULT

hr = SetCooperativeLevel(DD4);
if(hr != DD_OK) cry_bloody_murder();

etc...

Can you try to patch the game to the newest version?


-- 
Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.