December 2007 - Wine-Devel - list.winehq.org

Re: ntdll: Initialize APC call and result variables to suppress valgrind warnings
by Alexandre Julliard Dec. 11, 2007

Dec. 11, 2007

Dmitry Timoshkov <dmitry(a)codeweavers.com> writes: > Alexandre said that it's OK to use memset for server requests because of > backwards compatibility. It's OK for requests, but why do you want to initialize the result structure too? -- Alexandre Julliard julliard(a)winehq.org

2 2

Wine Weekly News
by Kai Blin Dec. 11, 2007

Dec. 11, 2007

Hi Zach, thanks for getting a new WWN issue out there. Finally there's more than one release after the other. I think WWN is a really valuable ressource for our users. Also, I hope it'll be a platform to get some GSoC exposure on the front page, especially without me having to take care of it. ;) Thanks, Kai -- Kai Blin WorldForge developer http://www.worldforge.org/ Wine developer http://wiki.winehq.org/KaiBlin Samba team member http://www.samba.org/samba/team/ -- Will code for cotton.

12 19

Re: kernel32: cpu: Detect AMD 64bit CPUs
by Alexandre Julliard Dec. 11, 2007

Dec. 11, 2007

Alexander Nicolaysen Sørnes <alex(a)thehandofagony.com> writes: > kernel32: cpu: Detect AMD 64bit CPUs Are you really supposed to return a 64bit CPU in 32-bit mode? Do you have a test case? -- Alexandre Julliard julliard(a)winehq.org

3 2

Work legalities
by Nathan Williams Dec. 11, 2007

Dec. 11, 2007

Hey everyone, I have been planning to do some work on wine for a while now, but after I started working I got myself a new programming job. I'm worried about the copyright of any external work I do, so I need a little advice. What do I need from my employer to clear me to work on wine? Is something verbal ok, or should I get it in writing? -- Nathan

10 12

Re: winex11drv: Treat CapsLock and NumLock keys as normal keys. They should toggle on button down.
by Alexandre Julliard Dec. 11, 2007

Dec. 11, 2007

Vitaliy Margolen <wine-patches(a)kievinfo.com> writes: > On winxp both keys toggle on button down not button up. So no need for any extra treatment. > Fixes bug 10446. You still have to handle the case when the state has been toggled outside of Wine, you can't just get rid of the whole thing. -- Alexandre Julliard julliard(a)winehq.org

2 2

Re: [2/5] WineD3D: Remove 24/32 bit hacks
by Chris Robinson Dec. 11, 2007

Dec. 11, 2007

On Monday 10 December 2007 03:24:04 pm Stefan Dösinger wrote: > devmode.dmFields = DM_BITSPERPEL | DM_PELSWIDTH | DM_PELSHEIGHT; > - devmode.dmBitsPerPel = (bpp >= 24) ? 32 : bpp; /* Stupid XVidMode cannot > change bpp */ Did you mean to leave dmBitsPerPel at 0?

2 1

Re: shlwapi: Fix memory leak in test
by Zac Brown Dec. 10, 2007

Dec. 10, 2007

Alistair Leslie-Hughes wrote: > Hi, > Changelog: > shlwapi: Fix memory leak in test. > > Best Regards > Alistair Leslie-Hughes > > > No patch attached? -Zac

1 0

re: COM
by Dan Kegel Dec. 10, 2007

Dec. 10, 2007

Adelle wrote: > I have some COM DLLs which work fine with VB6 applications on wine. > Is it possible to instantiate a windowsy COM object from a linuxy > application written in some form of C++ (eg, for compilation with gcc)? Erm, maybe you could write a winelib app, but you might have to use the C interface to COM. Why not keep using the VB6 app? - Dan -- Wine for Windows ISVs: http://kegel.com/wine/isv

1 0

deleting AppDB deleted
by Richard S Dec. 10, 2007

Dec. 10, 2007

Hi Im one of the maintainers of the iTunes 6.x (was) On Saturday I received this email ' Subject: [AppDB] Version '6.x' of 'iTunes' deleted Date: Sat, 08 Dec 2007 12:15:45 -0600 (18:15 GMT) Version '6.x' of 'iTunes' deleted ------------------------------------------------------- The action was performed by Wesley Hamel Reasons given Cleanup ' Is it standard practice to remove all old versions of a program except for the most recent? IF this is true, WHY? Rich

4 3

Bugzilla rights
by Vitaliy Margolen Dec. 10, 2007

Dec. 10, 2007

Bugzilla admin, please disallow people to add/remove e-mail addresses to/from bugs unless they've been given that right. Or at least have rights to modify any aspect of the bug. Thanks, Vitaliy.

6 6

File a bug report?
by Gilles May Dec. 10, 2007

Dec. 10, 2007

Hi everybody. After updating WINE to from 0.9.49 to 0.9.50 I noticed major slowdown in Company of Heroes, to the point that I get like 1 Frame every 2 seconds. Wine spams the console with: fixme:d3d_shader:IWineD3DVertexShaderImpl_CompileShader Vertexshader 0x4d20b78 recompiled more than 50 times Trying to find out the cause for the slowdown I did my first attempt at using git bisect and I could trace the slowdown to this patch: 959212304dfaf0e7459374593b20b74fe22a0b57 is first bad commit commit 959212304dfaf0e7459374593b20b74fe22a0b57 Author: Stefan Dösinger <stefan(a)codeweavers.com> Date: Tue Nov 20 21:14:10 2007 +0100 wined3d: Track vertex declaration changes on vertex shaders. If an attribute has type D3DDECLTYPE_D3DCOLOR, the red and blue channels are swizzled in the shader. Since the attribute is stored in the vertex declaration and not the vertex shader, it can change by setting a new vertex declaration. If this happens, we have to recompile the shader with the swizzling of that specific attribute turned on or off. :040000 040000 327661a4d0f8a02bdc18380084589c891b64af9c e35a2bcbf9514e4f261d8b7a93dcb24c0b3c89a7 M dlls Now without that patch there is slight texture corruption in the game, with the patch it's much much less but it renders the game unplayable. The reason I didn't file a bug report is that I am not sure if it is really a bug or not since obviously the patch fixes in-game texture corruption. Regards, Gilles

2 1

Read Application Memory
by Kai Lauterbach Dec. 10, 2007

Dec. 10, 2007

Hi, at the moment I'm writing a little Linux based program to read some variables from a Windows application. But my problem is, that i can't access to the virtual memory of the Win App running under wine. I don't know how to do this. c/c++ isn't the problem but the right way to do it. I've already read the wineserver's main.c and fd.c but I think that it's not a good way to put a code into, which give's me the chance to read something. I need sth to connect, like /dev/mem or a network port. My second idea was the wine debugger. Is it possible to get access the application memory at runtime from this? I mean, without stopping the application? I hope you can help me and I've described my problem as clear as possible ;) Greetings, Kai -- Kai Lauterbach <klaute(a)gmx.de>

3 2

Re: [PATCH 1/3] qmgr: Added DllGetClassObject and DllCanUnloadNow.
by Alexandre Julliard Dec. 10, 2007

Dec. 10, 2007

Roy Shea <roy(a)cs.hmc.edu> writes: > Revision 2. Made dll_ref static for now. > --- > dlls/qmgr/qmgr.spec | 2 ++ > dlls/qmgr/qmgr_main.c | 16 ++++++++++++++++ > 2 files changed, 18 insertions(+), 0 deletions(-) > > diff --git a/dlls/qmgr/qmgr.spec b/dlls/qmgr/qmgr.spec > index 1a63b32..c534b4f 100644 > --- a/dlls/qmgr/qmgr.spec > +++ b/dlls/qmgr/qmgr.spec > @@ -1,2 +1,4 @@ > +@ stdcall -private DllCanUnloadNow() > +@ stdcall -private DllGetClassObject(ptr ptr ptr) > @ stub DllRegisterServer > @ stub DllUnregisterServer These don't exist in the Windows dll. I don't think they make much sense in a service dll. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: user32: Test the initial state of the cursor.
by Alexandre Julliard Dec. 10, 2007

Dec. 10, 2007

"Andrew Riedi" <andrewriedi(a)gmail.com> writes: > +static void test_initial_cursor(void) > +{ > + HCURSOR cursor, cursor2; > + DWORD error; > + > + cursor = GetCursor(); > + > + /* Check what handle GetCursor() returns if a cursor is not set yet. */ > + SetLastError(0xdeadbeef); > + cursor2 = LoadCursor(NULL, IDC_WAIT); > + todo_wine { > + ok(cursor == (HCURSOR) 0x00010015, "Cursor handle is %p, not 0x00010015.\n", cursor); It doesn't make sense to hardcode handle values, it's not a behavior we want to replicate. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: sfnt2fnt: Use code from gdi32/freetype to do a better job findingthe TTAG_EBLC to deal with older version of freetype without having tofall back on unreliable hacks
by Dmitry Timoshkov Dec. 10, 2007

Dec. 10, 2007

"Aric Stewart" <aric(a)codeweavers.com> wrote: > + /* If the FT_Load_Sfnt_Table function is there we'll use it */ > +#ifdef HAVE_FT_LOAD_SFNT_TABLE > + err = FT_Load_Sfnt_Table(ft_face, table, offset, buf, len); > +#endif > +#ifdef HAVE_FREETYPE_INTERNAL_SFNT_H Probably using #elif instead of #endif+#ifdef above would be more correct. -- Dmitry.

1 0

directX 10 and hlsl2glsl
by Sònia Xapellí Brú Dec. 10, 2007

Dec. 10, 2007

I know there was some discussion concerning the use of hlsl2glsl and that it is not directly useful as it seems that the conversion to assembler is necessary in d3d9. Nevertheless, when it comes to d3d10,Shader Model 4 relies only in source code and forbids the programmers from implementing directly the assembler tokens. Therefore, I think that in this case the best solution, maximising the use of the graphic card capabilities goes mandatory to the use of the trasnlator and leaving up to the graphic card driver the compilation of GLSL. What do you think?

2 1

Re: wined3d: improve thread safety in IWineD3DSurfaceImpl_BltOverride
by Roderick Colenbrander Dec. 10, 2007

Dec. 10, 2007

Hi, The patch itself could be correct in the sense that indeed there might be a missing ActivateContext. But we should avoid making recursive ENTER_GL's. Between the ENTER_GL block there is a call to IWineD3DDevice_Clear which also calls ENTER_GL. You could put a LEAVE_GL before it and an ENTER_GL after it but there is something else which you couldn't have known. A number of places in surface.c are messing with glDrawBuffer. These days activatecontext is also handling this, so in a number of places glDrawBuffer is redundant. This might be one of those places but I haven't checked that in detail. Stefan might be able to answer that. Regards, Roderick -- GMX FreeMail: 1 GB Postfach, 5 E-Mail-Adressen, 10 Free SMS. Alle Infos und kostenlose Anmeldung: http://www.gmx.net/de/go/freemail

1 0

Read RAM from Process
by Kai Lauterbach Dec. 10, 2007

Dec. 10, 2007

Hi, at the moment I'm writing a little Linux based program to read some variables from a Windows application. But my problem is, that i can't access to the virtual memory of the Win App running under wine. I don't know how to do this. c/c++ isn't the problem but the right way to do it. I've already read the wineserver's main.c and fd.c but I think that it's not a good way to put a code into, which give's me the chance to read something. I need sth to connect, like /dev/mem or a network port. My second idea was the wine debugger. Is it possible to get access the application memory at runtime from this? I mean, without stopping the application? I hope you can help me and I've described my problem as clear as possible ;) Greetings, Kai -- Kai Lauterbach <lauterbach.kai(a)web.de>

2 1

Re: wined3d: improve thread safety in IWineD3DSurfaceImpl_BltOverride
by Stefan Dösinger Dec. 10, 2007

Dec. 10, 2007

Am Montag, 10. Dezember 2007 12:39:51 schrieb Alexander Dorofeyev: > Fixes crash in Aliens vs Predator 2 loading screen (bug # 4513 > http://bugs.winehq.org/show_bug.cgi?id=4513) > > --- > dlls/wined3d/surface.c | 5 +++++ > 1 files changed, 5 insertions(+), 0 deletions(-) Looks good to me

1 0

[press] Phoronix compares 7 versions of Wine
by Jonathan Ernst Dec. 9, 2007

Dec. 9, 2007

Hello, What do you think about their tests and results ? http://www.phoronix.com/vr.php?view=11539 "The performance in the fourteen 3DMark 2001 SE and 3DMark 2003 tests were fairly consistent in the releases between WINE 0.9.44 and WINE 0.9.50 -- generally only swaying a few frames in either direction -- though the performance overall has slightly declined and we were surprised by a few of the results."

4 4

GetMessage related Jagged Alliance 2 regression
by Alexander Dorofeyev Dec. 9, 2007

Dec. 9, 2007

Hello. I get freezes in Jagged Alliance 2 (seems to be tied to moving the mouse). It used to work in older versions. The game doesn't lock up totally (the music and sounds continue to play normally, for example), but the screen doesn't redraw and it doesn't respond to input. Activating some other window than the game window (if it's not full screen) usually unfreezes it. Bisect pointed to this change: http://source.winehq.org/git/wine.git/?a=commitdiff;h=2896540a34a18b8aff848… server: Set the queue mask directly in get_message to avoid an extra server call. I did a bit of debugging and it looks like the game is getting stuck within a GetMessageW call. It stays in that while loop in GetMessage, mostly in MsgWaitForMultipleObjectsEx. If the mouse is moved, then it awakes, gets to PeekMessageW, from there it goes to peek_message which returns false and so PeekMessageW returns false and the game goes back to waiting within GetMessageW. peek_message returns false, because in a first call to server it gets a MSG_HARDWARE (mouse) message which it drops for some reason, then it clears PM_QS_ flags and so in a next call to server it can't get a WM_TIMER message despite an expired timer seems to be ready in the queue. I'm not sure what would be a proper fix, although I did find out that commenting out this line in peek_message() if (HIWORD(flags)) flags = PM_QS_SENDMESSAGE | LOWORD(flags); makes the game playable. Maybe this clearing of PM_QS_* flags shouldn't be done in case when the message gets dropped? Before that commit nothing was passed in flags high word, which is probably why this problem wasn't happening.

3 3

ALLUSERSPROFILE Patch
by Austin English Dec. 9, 2007

Dec. 9, 2007

While looking at bug 7292 (http://bugs.winehq.org/show_bug.cgi?id=7292) I wrote a patch to implement ALLUSERSPROFILE. I looked on MSDN/Google for the dirid for this shell folder, but can only find dirids for it's subdirectories. For example: 16406 All Users\Start Menu 16407 All Users\Start Menu\Programs 16408 All Users\Start Menu\Programs\Startup 16409 All Users\Desktop 16415 All Users\Favorites 16419 All Users\Application Data Looking for an opinion on which would be better, to map ALLUSERSPROFILE to %10%\profiles\All Users (This is how my current patch does it) Or I could continue looking for the windows implementation of it, which so far hasn't proven very fruitful. Patch is attached.

2 1

msi: fix missing alloc check [try 2]
by Lionel_Debroux Dec. 9, 2007

Dec. 9, 2007

Removed a level of indentation, following feedback of James Hawkins. 2007-12-09 Lionel Debroux <lionel_debroux(a)yahoo.fr> * dlls/msi/dialog.c: msi: fix missing alloc check.

1 0

msi: fix memory leaks (found by Smatch) [try 2]
by Lionel_Debroux Dec. 9, 2007

Dec. 9, 2007

Fixed stupid mistake pointed by James Hawkins. 2007-12-09 Lionel Debroux <lionel_debroux(a)yahoo.fr> * dlls/msi/database.c: msi: fix memory leaks (found by Smatch).

1 0

msi: fix missing alloc check.
by Lionel_Debroux Dec. 9, 2007

Dec. 9, 2007

"msi_dialog_list_box" in dlls/msi/dialog.c gets flagged by Michael Stefaniuc's unfree-wine.pl script. * As far as I can see, "info", allocated line 2203, is not used before the return that leaks it (line 2214), neither in a direct nor an indirect (macro) way. But maybe the "fail-fast" behaviour is intended (because if allocation of "info" fails, there are chances the allocation of "control" would fail, too) ? * While looking at the source of the functions called by "msi_dialog_list_box" to make sure they didn't use "info", I spotted a call to msi_alloc, whose result is not checked. Therefore, the subsequent lines may write to NULL+some_offset. The patch contains only the fix for the latter problem, because I'm not 100% sure that the memory leak can be solved by moving the allocation of info after the call to "msi_dialog_add_control", without ill side effects (e.g. msi_dialog_add_control calls msi_dialog_create_window, which calls CreateWindowExW). 2007-12-09 Lionel Debroux <lionel_debroux(a)yahoo.fr> * dlls/msi/dialog.c: msi: fix missing alloc check.

2 1