http://bugs.winehq.org/show_bug.cgi?id=6155
------- Additional Comments From ben(a)atomnet.co.uk 2007-13-05 14:38 -------
Either learn how to code and fix it yourself or pay a professional to do it for
you. No-one here is going to fix something faster because some random person
asked them to.
--
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.
http://bugs.winehq.org/show_bug.cgi?id=8383
------- Additional Comments From ryanshow(a)gmail.com 2007-13-05 14:36 -------
Created an attachment (id=6270)
--> (http://bugs.winehq.org/attachment.cgi?id=6270&action=view)
Game debug with GLSL shaders enabled
I did not have GLSL shaders enabled previously, but enabling them by setting
HKEY_CURRENT_USER\Software\Wine\Direct3D\UseGLSL -> "enabled"
with regedit and then rerunning the game with "wine white.exe" does not work as
well. The game now crashes right before the video sequence. And the main menu
still displays no text. I also have "Allow Pixel Shader" checked in winecfg
with "Vertex Shader Support" set to Hardware.
--
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.
http://bugs.winehq.org/show_bug.cgi?id=8130
------- Additional Comments From rob(a)codeweavers.com 2007-13-05 14:28 -------
Yes, some research has revealed your patch is somewhat correct. I'll post an improved version in a bit.
--
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.
http://bugs.winehq.org/show_bug.cgi?id=8376
------- Additional Comments From focht(a)gmx.net 2007-13-05 14:20 -------
Hello,
--- quote ---
0009:trace:commdlg:PageSetupDlgA (0x941ee4): hwndOwner = (nil), hDevMode =
(nil), hDevNames = (nil)
--- quote ---
As i thought ... could you please try my two-liner fix from
http://bugs.winehq.org/show_bug.cgi?id=8033 and report back?
Copied it here (look for "FIX:")
--- snip dlls/comdlg32/printdlg.c ---
BOOL WINAPI PageSetupDlgA(LPPAGESETUPDLGA setupdlg)
{
.....
/* Initialize default printer struct. If no printer device info is specified
retrieve the default printer data. */
memset(&pdlg,0,sizeof(pdlg));
pdlg.lStructSize = sizeof(pdlg);
if (setupdlg->hDevMode && setupdlg->hDevNames) {
pdlg.hDevMode = setupdlg->hDevMode;
pdlg.hDevNames = setupdlg->hDevNames;
} else {
pdlg.Flags = PD_RETURNDEFAULT;
bRet = PrintDlgA(&pdlg);
if (!bRet){
if (!(setupdlg->Flags & PSD_NOWARNING)) {
char errstr[256];
LoadStringA(COMDLG32_hInstance, PD32_NO_DEFAULT_PRINTER,
errstr, 255);
MessageBoxA(setupdlg->hwndOwner, errstr, 0, MB_OK |
MB_ICONERROR);
}
return FALSE;
}
/* FIX: there was no input hDevMode and hDevNames: give it to client! */
setupdlg->hDevMode = pdlg.hDevMode;
setupdlg->hDevNames = pdlg.hDevNames;
}
....
}
-- snip dlls/comdlg32/printdlg.c ---
Regards
--
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.
http://bugs.winehq.org/show_bug.cgi?id=4146
------- Additional Comments From winebugger(a)piments.com 2007-13-05 14:15 -------
I'd be prepared to contribute if comments like this were not dismissed without
thought. There is absolutely no point in spending time producing patches for
something that has been dismissed as a non-issue and duped to a bug which has
no bearing it's contents.
If the bug is open and confirmed there would be a need for a patch. If it is
closed your call for patches makes not sense at all so don't try to turn the
issue round.
It shows how little you have read or understood of this bug before dismissing
it a second time since the response to wine command is clearly program and not
documentation.
Since you say this has already been discussed many times I suppose we have to
conclude that the consensus of wine devs is that when a user enters wine --help
then most useful response you can provide is : "print this page".
At least that should help him understand that wine is not windows.
I'd still be interested to know how you see this topic as a dupe of a meta bug.
Must be some subtle derivative of reverse polish logic.
--
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.
http://bugs.winehq.org/show_bug.cgi?id=8130
------- Additional Comments From focht(a)gmx.net 2007-13-05 14:11 -------
Hello,
--- quote ---
Perhaps this is just a race that is rarely hit on Windows because of different
code in ole32 or perhaps there
is something when running on Windows that keeps the DLL loaded.
I think the patch that apparantly caused the regression is correct though.
--- quote ---
Sorry, but this is neither a race condition nor "bad" code.
I wrote a simple test client (http://bugs.winehq.org/show_bug.cgi?id=8159) to
simulate the problem.
Basically said: You cant unload ole servers when destroying apartment without
looking at the ole server refcount itself.
A client *may* still hold and use a reference.
That's an information only the ole server has.
One solution would be to ask the server in COMPOBJ_DllList_ReleaseRef().
Beware: This is meant as quick fix, non-intrusive as possible (but working).
--- snip dlls/ole32/compobj.c ---
static void COMPOBJ_DllList_ReleaseRef(OpenDll *entry)
{
if (!InterlockedDecrement(&entry->refs))
{
EnterCriticalSection(&csOpenDllList);
list_remove(&entry->entry);
LeaveCriticalSection(&csOpenDllList);
if( entry->DllCanUnloadNow() == S_OK)
{
TRACE("freeing %p\n", entry->library);
FreeLibrary(entry->library);
}
else
{
WARN("unload skipped, client still holds refs to %s (%p)\n",
debugstr_w(entry->library_name), entry->library);
}
HeapFree(GetProcessHeap(), 0, entry->library_name);
HeapFree(GetProcessHeap(), 0, entry);
}
}
--- snip dlls/ole32/compobj.c ---
Though the general problem remains: with the last apartment data/list (module
entries) gone, there is no way to free com servers thereafter (except if the
client forces creation/destruction of apartment again with all interface
refcounts released)
To overcome this problem, ole server data structures which are used to track
lifetime have to be managed across apartments (e.g. at global scope).
Regards
--
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.
http://bugs.winehq.org/show_bug.cgi?id=6155
------- Additional Comments From ultragun(a)gmail.com 2007-13-05 14:07 -------
This trouble is very important for me. Please, fix his faster.
How can i help?
--
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.
http://bugs.winehq.org/show_bug.cgi?id=8194
------- Additional Comments From stefandoesinger(a)gmx.at 2007-13-05 14:01 -------
Do those applications use the same engine? Are they related to each other? The
crashes look different at first glance, but the first one is clearly a heap
curruption, the 2nd one may be one too.
The only thing I saw that makes me suspicious are a number
of "0009:err:ddraw:IDirectDrawSurfaceImpl_GetSurfaceDesc Impossible/Strange
struct size 0" messages. I will write a test case for this.
Are there downloadable demos for these apps?
--
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.
http://bugs.winehq.org/show_bug.cgi?id=8185
------- Additional Comments From saulius.krasuckas(a)elst.vtu.lt 2007-13-05 14:01 -------
Skype v2.0.0.97 is affected by this bug also.
Someone please add "regression" keyword here.
Wine-0.9.2 works OK, while Wine-0.9.15 do not.
--
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.