February 2008 - Wine-Devel - list.winehq.org

Is wine reporting the right value for MaxVertexBlendMatrices ?
by Julio Fernandez Feb. 9, 2008

Feb. 9, 2008

Hi, I don't know if this is the right place to post this, but sorry if not, and I'd like to be pointed to the right one I have been working last weeks on making the game Everquest to work under wine, since 0.9.49 it was working flawlesly, but most models were not shown properly. This corresponds with bug http://bugs.winehq.org/show_bug.cgi?id=8357. These are my findings: The main executable is eqgame.exe it typically accesses D3D through a custom dll eqgraphicsdx9 which interacts with D3D itself or through d3dx9_30.dll Using channels and adding some extra messages into certain source files I got to track a possible reason: The game was declaring vertex lists with a vertex of type '0xfe' and +189 offset Sadly, I could not get any extra information tracking just from source code, so I had to run the game under Windows on a computer and using windbg to debug it, while running the same time on another computer the game under Debian etch using winedbg to debug, and so tracing both programs in assambler (I burnt my eyelashes there!), the fact that I could have d3d dll symbols availables on the windows box was the key here, I really missed winedbg to be able to read pdb files :( After many hours I was able to track down the reason, and to my surprise it seems its not wine's fault (for the most of it) The 'troublesome' vertex are apparently Blended ones on a Mesh setup. When d3dx9_30 calls ConvertToIndexedBlendedMesh, it calls GetDeviceInfo to get the device capabilites, in the end wine returns the capabilites through IWineD3DImpl_GetDeviceCaps in wine/dlls/wined3d/directx.c Here comes the fun thing, the attribute MaxVertexBlendMatrices in D3DCAPS9, from (MSDN): "Maximum number of matrices that this device can apply when performing multimatrix vertex blending. For a given physical device, this capability may vary across Direct3D devices depending on the parameters supplied to IDirect3D9::CreateDevice." At the moment is set in wine as *pCaps->MaxVertexBlendMatrices = GL_LIMITS(blends) which results in a value of 0 (at least in my Linux computer/build), while under Windows (XPSP2 DX9c) results in a value of 4 d3dx9_30 has a check right after capabilites are returned comparing that value with 2, resulting in different code paths for the remaining of MaxVertexBlendMatrices function, when the value was 0 both Linux and Windows (hacking the returned value to be 0) make those odd '0xfe' vertexes to be inserted (in Linux that makes the models to not be shown, in Windows it makes eqgame.exe to crash) I proceeded then to try to 'hack' wine to return a value of 4, adding the line pCaps->MaxVertexBlendMatrices=0x4; Before returning from function IDirect3DDevice9Impl_GetDeviceCaps(LPDIRECT3DDEVICE9 iface, D3DCAPS9* pCaps) at wine/dlls/d3d9/directx.c The result: MaxVertexBlendMatrices runs the same path as Windows does, the odd '0xfe' vertexes are not generated and the models work are now shown as expected My apologies for the long post, but I thought it would be handy to tell the full story to justify calling your atention on this - d3dx9_30 ConvertToIndexedBlendedMesh shows an 'odd' behavior when MaxVertexBlendMatrices capabilites in D3DCAPS9 is 0 - Is wine reporting the right value for MaxVertexBlendMatrices ??? As it seems the game works flawlesly in wine when it's hacked to be 4 (I don't know if this 'hack' would affect other programs in any other way

3 3

No win95 tests after commit "tests: Add a helper routine to wait for a child process and propagate its result correctly."
by Paul Vriens Feb. 9, 2008

Feb. 9, 2008

Hi, This commit: 05c7174909e975fc7d0ebf9a59bb07160c09d66d introduces a new helper script but uses InterlockedExchangeAdd that is not available on win95. This means that none of the tests run on win95 anymore. -- Cheers, Paul.

2 2

Is wine reporting the right value for MaxVertexBlendMatrices ?
by Julio Fernandez Feb. 9, 2008

Feb. 9, 2008

Hi, I don't know if this is the right place to post this, but sorry if not, and I'd like to be pointed to the right one I have been working last weeks on making the game Everquest to work under wine, since 0.9.49 it was working flawlesly, but most models were not shown properly. This corresponds with bug http://bugs.winehq.org/show_bug.cgi?id=8357 These are my findings: The main executable is eqgame.exe it typically accesses D3D through a custom dll eqgraphicsdx9 which interacts with D3D itself or through d3dx9_30.dll Using channels and adding some extra messages into certain source files I got to track a possible reason: The game was declaring vertex lists with a vertex of type '0xfe' and +189 offset Sadly, I could not get any extra information tracking just from source code, so I had to run the game under Windows on a computer and using windbg to debug it, while running the same time on another computer the game under Debian etch using winedbg to debug, and so tracing both programs in assambler (I burnt my eyelashes there!), the fact that I could have d3d dll symbols availables on the windows box was the key here, I really missed winedbg to be able to read pdb files :( After many hours I was able to track down the reason, and to my surprise it seems its not wine's fault (for the most of it) The 'troublesome' vertex are apparently Blended ones on a Mesh setup. When d3dx9_30 calls ConvertToIndexedBlendedMesh, it calls GetDeviceInfo to get the device capabilites, in the end wine returns the capabilites through IWineD3DImpl_GetDeviceCaps in wine/dlls/wined3d/directx.c Here comes the fun thing, the attribute MaxVertexBlendMatrices in D3DCAPS9, from (MSDN): "Maximum number of matrices that this device can apply when performing multimatrix vertex blending. For a given physical device, this capability may vary across Direct3D devices depending on the parameters supplied to IDirect3D9::CreateDevice." At the moment is set in wine as *pCaps->MaxVertexBlendMatrices = GL_LIMITS(blends) which results in a value of 0 (at least in my Linux computer/build), while under Windows (XPSP2 DX9c) results in a value of 4 d3dx9_30 has a check right after capabilites are returned comparing that value with 2, resulting in different code paths for the remaining of MaxVertexBlendMatrices function, when the value was 0 both Linux and Windows (hacking the returned value to be 0) make those odd '0xfe' vertexes to be inserted (in Linux that makes the models to not be shown, in Windows it makes eqgame.exe to crash) I proceeded then to try to 'hack' wine to return a value of 4, adding the line pCaps->MaxVertexBlendMatrices=0x4; Before returning from function IDirect3DDevice9Impl_GetDeviceCaps(LPDIRECT3DDEVICE9 iface, D3DCAPS9* pCaps) at wine/dlls/d3d9/directx.c The result: MaxVertexBlendMatrices runs the same path as Windows does, the odd '0xfe' vertexes are not generated and the models work are now shown as expected My apologies for the long post, but I thought it would be handy to tell the full story to justify calling your atention on this - d3dx9_30 ConvertToIndexedBlendedMesh shows an 'odd' behavior when MaxVertexBlendMatrices capabilites in D3DCAPS9 is 0 - Is wine reporting the right value for MaxVertexBlendMatrices ??? As it seems the game works flawlesly in wine when it's hacked to be 4 (I don't know if this 'hack' would affect other programs in any other way

1 0

re: msvcrt: Fix EOF behavior on read from pipe
by Dan Kegel Feb. 9, 2008

Feb. 9, 2008

Damjan wrote > The problem is more general: consoles *also* break on short reads > (Java 1.4.1's gdb takes 1 command then quits because it thinks it's > the end of file) Try my patch, it should help that case. > I'm sure short read != end of file for ordinary > files either, especially in *nix where read() can easily be > interrupted by a signal. If you can come up with an example of a real app or even a test case that is currently broken, I'll gladly try to fix it. - Dan

2 3

Is wine reporting the right value for MaxVertexBlendMatrices ?
by Julio Fernandez Feb. 9, 2008

Feb. 9, 2008

Hi, I don't know if this is the right place to post this, but sorry if not, and I'd like to be pointed to the right one I have been working last weeks on making the game Everquest to work under wine, since 0.9.49 it was working flawlesly, but most models were not shown properly. This corresponds with bug http://bugs.winehq.org/show_bug.cgi?id=8357. These are my findings: The main executable is eqgame.exe, it typically accesses D3D through a custom dll eqgraphicsdx9 which interacts with D3D itself or through d3dx9_30.dll Using channels and adding some extra messages into certain source files I got to track a possible reason: The game was declaring vertex lists with vertex of type '0xfe' and +189 offset Sadly, I could not get any extra information tracking just from source code, so I had to run the game under Windows on a computer and using windbg to debug it, while running at the same time on another computer the game under Debian etch using winedbg to debug, and so tracing both programs in assembler (I've burnt my eyelashes there!), the fact that I could have d3d dll symbols availables on the windows box was the key here, I really missed winedbg was able to read pdb files :( After many hours I was able to track down the reason, and to my surprise it seems it's not wine's fault (for the most of it) The 'troublesome' vertex are apparently Blended ones on a Mesh setup. When d3dx9_30 calls ConvertToIndexedBlendedMesh, it calls GetDeviceInfo to get the device capabilites, in the end wine returns the capabilites through IWineD3DImpl_GetDeviceCaps in wine/dlls/wined3d/directx.c Here comes the fun thing, for the attribute MaxVertexBlendMatrices in D3DCAPS9, from (MSDN): "Maximum number of matrices that this device can apply when performing multimatrix vertex blending. For a given physical device, this capability may vary across Direct3D devices depending on the parameters supplied to IDirect3D9::CreateDevice." At the moment is set in wine as *pCaps->MaxVertexBlendMatrices = GL_LIMITS(blends) which results in a value of 0 (at least in my Linux computer/build), while under Windows (XPSP2 DX9c) results in a value of 4 d3dx9_30 has a check right after capabilites are returned comparing that value with 2, resulting in different code paths for the remaining of MaxVertexBlendMatrices function, when the value was 0 both Linux and Windows (hacking the returned value to be 0) make those odd '0xfe' vertexes to be inserted (in Linux that makes the models not to be shown, in Windows it makes eqgame.exe to crash) I proceeded then to try to 'hack' wine to return a value of 4, adding the line pCaps->MaxVertexBlendMatrices=0x4; Before returning from function IDirect3DDevice9Impl_GetDeviceCaps(LPDIRECT3DDEVICE9 iface, D3DCAPS9* pCaps) at wine/dlls/d3d9/directx.c The result: MaxVertexBlendMatrices runs the same path as Windows does, the odd '0xfe' vertexes are not generated and the models are now shown as expected My apologies for the long post, but I thought it would be handy to tell the full story to justify calling your atention on this - d3dx9_30 ConvertToIndexedBlendedMesh shows an 'odd' behavior when MaxVertexBlendMatrices capabilites in D3DCAPS9 is 0 - Is wine reporting the right value for MaxVertexBlendMatrices ??? As it seems the game works flawlesly in wine when it's hacked to be 4 (I don't know if this 'hack' would affect other programs in any other way

1 0

Re: wine keyboard fix
by Stefan Dösinger Feb. 9, 2008

Feb. 9, 2008

Am Samstag, 9. Februar 2008 04:52:23 schrieb dhliu: > Some input methods can send a much longer string. The original size is > too small. Please consider enlarging it. > Thanks. You should send a patch generated via git to wine-patches. If you do not know how to do this, plese consider filing a bugreport about this issue, together with an example for "Some input methods" which can trigger the issue.

1 0

Is wine reporting the right value for MaxVertexBlendMatrices?
by Julio Fernandez Feb. 9, 2008

Feb. 9, 2008

Hi, I don't know if this is the right place to post this, but sorry if not, and I'd like to be pointed to the right one I have been working last weeks on making the game Everquest to work under wine, since 0.9.49 it was working flawlesly, but most models were not shown properly. This corresponds with bug http://bugs.winehq.org/show_bug.cgi?id=8357. These are my findings: The main executable is eqgame.exe it typically accesses D3D through a custom dll eqgraphicsdx9 which interacts with D3D itself or through d3dx9_30.dll Using channels and adding some extra messages into certain source files I got to track a possible reason: The game was declaring vertex lists with a vertex of type '0xfe' and +189 offset Sadly, I could not get any extra information tracking just from source code, so I had to run the game under Windows on a computer and using windbg to debug it, while running the same time on another computer the game under Debian etch using winedbg to debug, and so tracing both programs in assambler (I burnt my eyelashes there!), the fact that I could have d3d dll symbols availables on the windows box was the key here, I really missed winedbg to be able to read pdb files :( After many hours I was able to track down the reason, and to my surprise it seems its not wine's fault (for the most of it) The 'troublesome' vertexes are apparently blended ones in mesh declarations. When d3dx9_30 calls ConvertToIndexedBlendedMesh, it calls GetDeviceInfo to get the device capabilites, in the end wine returns the capabilites through IWineD3DImpl_GetDeviceCaps in wine/dlls/wined3d/directx.c Here comes the fun thing, the attribute MaxVertexBlendMatrices in D3DCAPS9, from (MSDN): "Maximum number of matrices that this device can apply when performing multimatrix vertex blending. For a given physical device, this capability may vary across Direct3D devices depending on the parameters supplied to IDirect3D9::CreateDevice." At the moment is set in wine as *pCaps->MaxVertexBlendMatrices = GL_LIMITS(blends) which results in a value of 0 (at least in my Linux computer/build), while under Windows (XPSP2 DX9c) results in a value of 4 d3dx9_30 has a check right after the capabilites are returned comparing that value with 2, resulting in different code paths for the remaining of MaxVertexBlendMatrices function, when the value was 0 both Linux and Windows (hacking the returned value to be 0) make those odd '0xfe' vertexes to be inserted (in Linux that makes the models to not be shown, in Windows it makes eqgame.exe to crash) I proceeded then to try to 'hack' wine to return a value of 4, adding the line pCaps->MaxVertexBlendMatrices=0x4; Before returning from function IDirect3DDevice9Impl_GetDeviceCaps(LPDIRECT3DDEVICE9 iface, D3DCAPS9* pCaps) at wine/dlls/d3d9/directx.c The result: MaxVertexBlendMatrices runs the same path as Windows does, the odd '0xfe' vertexes are not generated and the models work are now shown as expected My apologies for the long post, but I thought it would be handy to tell the full story to justify calling your atention on this - d3dx9_30 ConvertToIndexedBlendedMesh shows an 'odd' behavior when MaxVertexBlendMatrices capabilites in D3DCAPS9 is 0 - Is wine reporting the right value for MaxVertexBlendMatrices ??? As it seems the game works flawlesly in wine when it's hacked to be 4 (I don't know if this 'hack' would affect other programs in any other way

1 0

Code quality (was Re: comctl32: Fix invalid syntax)
by Bang Jun-young Feb. 9, 2008

Feb. 9, 2008

On 2/7/08, Michael Stefaniuc <mstefani(a)redhat.com> wrote: > Bang Jun-young wrote: > > On 2/7/08, Alexandre Julliard <julliard(a)winehq.org> wrote: > >> "Bang Jun-young" <junyoung(a)mogua.com> writes: > >> > >>> 18 hours passed, and it looks like Alexandre decided to ignore this...why? > >>> > >>> This fix is required for Wine to be built with VC2005/2008 (although > >>> Alexandre doesn't seem to care about it). > >> There's no need to count the hours, or to ask every day, especially not > >> with a mail to wine-patches. I have lots of pending patches from my > >> vacation, I'll get to it eventually, and if I don't you should resubmit > >> it after a week has passed. > > > > That's because you wanted yourself to be the only committer in the > > project. Why don't you give commit privilege to other Wine > > contributors as well? > Huh? Why would you need that? Because I'm concerned with the code quality of Wine. When Alexandre commits what he doesn't understand, it's a red sign to the project. Months ago I posted a patch similar to this one to wine-patches, and he committed it: typedef struct _OLEOBJECTVTBL { - void CALLBACK *(*QueryProtocol)(_LPOLEOBJECT,LPCOLESTR16); + void * (CALLBACK *QueryProtocol)(_LPOLEOBJECT,LPCOLESTR16); How did he know what the patch does from the same message "Fix invalid syntax?" IIRC, he didn't tell anything about what he doesn't understand. In fact, this is a well known mistake many newbie Win32 developers make (and fix in minutes). It shouldn't have been in the tree in the first place if he actually have read the patch. There are a lot of easily catchable bugs in the tree, for example, potential security holes like buffer overrun, meaningless comparison of unsigned < 0 (or > 0), misuse of BOOL vs. HRESULT, misuse of functions such as strcasecmp(), use of non-portable syntax, etc. Most of them (if not all) could be filtered out if he have read the patches carefully. That's the main reason why Wine keeps crashing every time I give it a try with my Windows apps. Since 1993, Wine has never gotten to the point where everybody could rely on it for his daily work. It has as awfully many bugs as Win95. I see something fundamentally wrong with development process. > Wine uses git, a distributed SCM. You are > free to publish your own Wine tree without waiting for Alexandre. 99% of the world won't benefit from what I have done in my own local repository. People usually don't have a much time to keep track of repositories of hundreds of developers around the world. That's why every open source project has the main repository even if developers usually maintain their own repostories at home. > People prefer to follow Alexandre's tree but nobody forces them to use it. If winehq.org is Alexandre's own property, this is something fundametally wrong again.

6 5

Bugzilla: Remove obsolete components
by Austin English Feb. 9, 2008

Feb. 9, 2008

Howdy, I just finished moving over the last of the obsolete components. _obsolete_binary, _obsolete_directx, and _obsolete_gui can all be removed now. -Austin

3 2

Re: user32: Implement GetWindowModuleFileName with tests
by Dmitry Timoshkov Feb. 8, 2008

Feb. 8, 2008

"Maarten Lankhorst" <m.b.lankhorst(a)gmail.com> wrote: > UINT WINAPI GetWindowModuleFileNameA( HWND hwnd, LPSTR lpszFileName, UINT cchFileNameMax) > { > - FIXME("GetWindowModuleFileNameA(hwnd %p, lpszFileName %p, cchFileNameMax %u) stub!\n", > - hwnd, lpszFileName, cchFileNameMax); > - return 0; > + LPWSTR filenameW = NULL; > + UINT retval; > + > + if (cchFileNameMax) > + { > + filenameW = HeapAlloc(GetProcessHeap(), 0, cchFileNameMax); > + if (!filenameW) > + { > + SetLastError(ERROR_NOT_ENOUGH_MEMORY); > + return 0; > + } > + } Probably GetWindowModuleFileNameA should simply allocate a buffer of size MAX_PATH on the stack and fail with ERROR_FILENAME_EXCED_RANGE if unicode counterpart returns a string longer than MAX_PATH. kernel32 APIs behave that way. > + retval = GetWindowModuleFileNameW(hwnd, filenameW, cchFileNameMax); > + if (retval) > + { > + DWORD lasterror = GetLastError(); > + WideCharToMultiByte(CP_ACP, 0, filenameW, -1, lpszFileName, cchFileNameMax, NULL, NULL); > + SetLastError(lasterror); What's the point of saving/restoring last error value? If WideCharToMultiByte fails you need to return an error in that case, not silently continue. Also GetWindowModuleFileNameA also should be fixed to return correct value, and not assume that the buffer is '\0' terminated. > UINT WINAPI GetWindowModuleFileNameW( HWND hwnd, LPWSTR lpszFileName, UINT cchFileNameMax) > { > - FIXME("GetWindowModuleFileNameW(hwnd %p, lpszFileName %p, cchFileNameMax %u) stub!\n", > - hwnd, lpszFileName, cchFileNameMax); > - return 0; > + PROCESS_BASIC_INFORMATION pbi; > + DWORD pid, retval; > + HANDLE hproc; > + DWORD lasterror = GetLastError(); > + UNICODE_STRING *pathname; > + TRACE("GetWindowModuleFileNameW(hwnd %p, lpszFileName %p, cchFileNameMax %u)\n", hwnd, lpszFileName, > cchFileNameMax); TRACE/WARN/ERR/etc already print the caller's name, there is no need to duplicate it. > + if (!GetWindowThreadProcessId(hwnd, &pid)) > + { > + SetLastError(ERROR_INVALID_WINDOW_HANDLE); > + return 0; > + } GetWindowThreadProcessId already sets the error to the correct value, there is no need to guess why it failed. > + if (!cchFileNameMax) > + { /* Fail silently without setting filename */ > + SetLastError(lasterror); > + return 0; > + } Saving/restoring last error value doesn't look correct at all. > + hproc = OpenProcess(PROCESS_QUERY_INFORMATION, 0, pid); > + if (!hproc) > + { > + FIXME("Is this correct?"); > + SetLastError(ERROR_INVALID_WINDOW_HANDLE); > + return 0; > + } There is no need to guess why OpenProcess failed. > + retval = NtQueryInformationProcess(hproc, ProcessBasicInformation, &pbi, sizeof(pbi), NULL); > + pathname = &pbi.PebBaseAddress->ProcessParameters->ImagePathName; > + if (pathname->Length / sizeof(WCHAR) < cchFileNameMax) > + { > + memcpy(lpszFileName, pathname->Buffer, pathname->Length); > + lpszFileName[pathname->Length / sizeof(WCHAR)] = 0; > + retval = pathname->Length / sizeof(WCHAR) + 1; > + } > + else > + { > + memcpy(lpszFileName, pathname->Buffer, cchFileNameMax * sizeof(WCHAR)); > + retval = cchFileNameMax; > + } Both branches above should be merged, that will simplify the code. Just '\0' terminate the result if there is enough space in the provided buffer. > + CloseHandle(hproc); > + SetLastError(lasterror); > + FIXME("--> %u %s\n", retval, debugstr_w(lpszFileName)); Left-over from debugging? -- Dmitry.

3 7

AppDB Issues
by Jeremy Newman Feb. 8, 2008

Feb. 8, 2008

As many of you have noticed we've been having a bit of server trauma this past month. The issue boils down to a bug somewhere in mysqld. We are still trying to find the exact cause of the bug. We do have some leads. I just wanted to do a brain dump on wine-devel here before I go on Vacation for a week. The problem: mysqld will randomly eat up all resources on the system until the server is completely unresponsive to anything but pings. My band-aids thus far: 1. mysql is now running at a nice of 1. This should allow us to at least ssh in and kill it if it gets out of control again. 2. I've removed InnoDB table support to save on memory. Not many tables where using it anyway, and nothing is using transaction support at this time. 3. I've patched the appdb to use memcached for session management. This will reduce some load on the DB as the session table was busiest and slowest table of the appdb. These fixes should give longer uptime, but they are not solving the core issue. While I'm gone, if the server goes offline again, please contact Alexandre or Jeremy White. They can kick proverbial butt of the server to get it going again. Lastly, sorry for the issues. -Newman "We do what we must because we can" - GLaDOS

1 0

Re: More correctly handle the presence of the default ime inputwindow on Ime open and close
by Dmitry Timoshkov Feb. 8, 2008

Feb. 8, 2008

"Aric Stewart" <aric(a)codeweavers.com> wrote: > diff --git a/dlls/imm32/imm.c b/dlls/imm32/imm.c > index 4acbde3..f3ed6f9 100644 > --- a/dlls/imm32/imm.c > +++ b/dlls/imm32/imm.c > @@ -190,19 +190,17 @@ static LRESULT ImmInternalSendIMENotify(WPARAM notify, LPARAM lParam) > > static void ImmInternalSetOpenStatus(BOOL fOpen) > { > - TRACE("Setting internal state to %s\n",(fOpen)?"OPEN":"CLOSED"); > + TRACE("Setting internal state to %s\n",(fOpen)?"OPEN":"CLOSED"); Wouldn't it be better to change indentation to commonly used 4 spaces instead of a strange mix of 3 and 4 spaces? -- Dmitry.

2 1

Fixing make test - progress update
by Jeremy White Feb. 8, 2008

Feb. 8, 2008

I believe that we have an easy way to find out the 'state of the art' wrt to make test. That is, my patch + request of a few days ago: http://www.winehq.org/pipermail/wine-devel/2008-February/062303.html resulted in a very nice, easy to digest report http://test.winehq.org/data/20080205/ That data has five test runs and shows clear across the board failures in msi:install, shell32:shellink, user32:listbox, and user32:msg. So, we can quickly and easily collate and visually see the make test failures. And, heck, I'm sure that fixing those failures is now a trivial exercise. Now we just need the will. :-/. Not sure how to arrange that, but I think having make test succeed on all systems should be a mandatory requirement for Wine 1.0, so I think we have to work on solving it sooner rather than later. The todo's as I see them are as follows: 1. Revise winetest to not require the 'make dist' step, so that the function of my patch can be implemented more cleanly. 2. Persuade people to run a script on a regular basis so we always have this data. It's trickier than the average bear because the tests require the 'real' console, so you can't just stick it in a headless cron job. Maybe some xauth magic... :-/ 3. (The hard one) Focus in and fix the issues. I'll work on #1 and #2 (although not for a week or two), if everyone else agrees to do #3 <evil grin>. Cheers, Jeremy

3 3

Re: user32:scroll.c: Compulsory enabling/disabling scrollbars after setting scrollbar`s flags.
by Anatoly Lyutin Feb. 8, 2008

Feb. 8, 2008

Oh, Sorry! It crashes some another scrollbars. :( -- Best regards Anatoly Lyutin.

1 0

Re: msvcrt: Fix EOF behavior on read from pipe
by Damjan Jovanovic Feb. 8, 2008

Feb. 8, 2008

On Feb 8, 2008 4:31 PM, Dan Kegel <dank(a)kegel.com> wrote: > Nguyễn Thái Ngọc Duy's patch was almost right, he > forgot a test case, and he neglected to move where > EOF is detected for pipes instead of just removing it. > > Should fix http://bugs.winehq.org/show_bug.cgi?id=11187 > > > > The problem is more general: consoles *also* break on short reads (Java 1.4.1's gdb takes 1 command then quits because it thinks it's the end of file) and I'm sure short read != end of file for ordinary files either, especially in *nix where read() can easily be interrupted by a signal. Damjan

1 0

re: [PATCH] msvcrt: Fix early pipe closing
by Dan Kegel Feb. 8, 2008

Feb. 8, 2008

On Sat Feb 2, Nguyễn Thái Ngọc Duy wrote: > Pipes are not like other normal files. If we do not read enough > bytes, it does not mean it reaches EOF > ... fixes broken pipe on git-fetch-pack (MinGW port) This is the same problem as http://bugs.winehq.org/show_bug.cgi?id=11187 so it's my fault. Thanks for the information about git-fetch-pack failing, and for the proposed fix. Before we can commit a fix like yours, we need a test case, so I improved msvcrt/tests/file.c to test EOF on pipes more thoroughly. It shows your patch was almost right. I've fixed up your patch and will send it in. - Dan

1 0

kernel32: Implement ReplaceFileW
by luis.busquets＠ilidium.com Feb. 8, 2008

Feb. 8, 2008

I have read this patch sent in several times and have realised that every time it contains a large number of "goto"'s. I certainly do not like "goto" in structured code. My question to the forum is if it is allowed to use "goto" in wine or if you prefer not to use it. Even when this will make the code larger but it may give a little nore readability,

2 1

Patch tracking system - again...
by Steven Edwards Feb. 8, 2008

Feb. 8, 2008

Hi, I'm not sure if this would even really be of use with most of the development being done on git anyway, but after re-reading the governance thread when the point of patch tracking was brought up, It got me thinking. last time around it got shelved as normal due to not wanting to disrupt Alexandre's work flow. Thinking about it a bit more I think I've come up with a solution to the problem that should not disrupt Alexandre's flow if I understand git properly. Let me preface this with saying that I don't. At first I thought we might be able to use the sha1 hashes that are the patch names themselves to track if a commit had be accepted or not, but then I guess that would only work if we were pushing and pulling directly in to each others tree's as I see the commit hash differs from the patch hash in the patches I've reviewed. I can only assume this is because of the way git views looks at the entire repo as one chunk of data rather than dealing with single patches. Then as I dug in to it a bit more I noticed the index's did match provided Alexandre did not make any changes before he committed. I was thinking this could be a use for a tracking system. Lets imagine there is this patch daemon that sits out there and reads wine-patches, storing the diff and parsing the index in to a database. This deamon could do a git fetch every 30 mins or so to compare and see if a patch has been merged in based on the index. It seems like we already have most of the tools we need to do it now if someone really wants the tracking system. We know about the patches and we know if its been merged or not, defining rules for accepted or rejected seem pretty trivial to me as we already have the rule, if its not merged in a week then resend. So here is what it would look like as far as work flow goes, someone submits a patch, it goes to this database and up to a website with a listview showing all the patches. Anything not merged yet still in the two week window would show up in the list as white meaning still pending. The further up the list we would have patches that were merged and accepted as is, like the following (showing up in green) http://www.winehq.org/pipermail/wine-patches/2008-February/049719.html http://source.winehq.org/git/wine.git/?a=commit;h=ac61b90f87a2062b99f6f7068… The following patch would show up in yellow as the index differs for the Makefile part of the patch as Julliard made local changes http://www.winehq.org/pipermail/wine-patches/2008-February/049720.html http://source.winehq.org/git/wine.git/?a=commit;h=5bad12c6bc3072a2676f0d9ed… Any other patches that don't get merged would show up in the list as red after the one week window went by. The tracker would assume they were rejected. The cool thing about this is that we could directly link it in to bugzilla so that when a patch is applied the bug could get automatically updated. Now this all hinges on being able to use the diff index to do the tracking. I imagine there might be conflicts if your trying to use that as the only data point as its not as long as a real hash, but assuming your only doing a compare over as period of a few weeks or a month I think it would work. Am I wack? Is there even still the desire to have a graphical patch tracking system? -- Steven Edwards "There is one thing stronger than all the armies in the world, and that is an idea whose time has come." - Victor Hugo

2 1

re: Code quality (was Re: comctl32: Fix invalid syntax)
by Dan Kegel Feb. 7, 2008

Feb. 7, 2008

Bang Jun-young wrote: >In fact, this is a well known mistake many newbie Win32 developers >make (and fix in minutes). It shouldn't have been in the tree in the >first place if he actually have read the patch. There are a lot of >easily catchable bugs in the tree, for example, potential security >holes like buffer overrun, meaningless comparison of unsigned < 0 (or > > 0), misuse of BOOL vs. HRESULT, misuse of functions such as > strcasecmp(), use of non-portable syntax, etc. Most of them (if not > all) could be filtered out if he have read the patches carefully. >That's the main reason why Wine keeps crashing every time I give it a >try with my Windows apps. I don't think so. In my experience, it's either a missing feature or something more subtle. Outright bugs of the kind you describe are the minority. I think. > Since 1993, Wine has never gotten to the > point where everybody could rely on it for his daily work. It has as > awfully many bugs as Win95. I see something fundamentally > wrong with development process. Nah, Alexandre usually does a good job of screening patches. We're lucky to have such a good maintainer, IMHO. (I say this knowing full well that newbies are put off by how difficult it is to get feedback because I have seen much, much worse maintainers. Imagine if, say, that guy who always yells at the newbies to not paste logs in bug reports was the maintainer. We'd probably lose all our developers in a month.) So you're peeved you don't have commit access. That's understandable, but it's time to get over it. IMHO the single-committer system works pretty well once you get into the swing of it. It actually forces at least one code review on every patch! - Dan

1 0

War on conformance test suite failures!
by Dan Kegel Feb. 7, 2008

Feb. 7, 2008

Great progress was made at Wineconf last year on getting our test suite closer to always passing on wine, but http://test.winehq.org/data/200801301937/#Wine shows that, according to the results collected from people who run programs/winetest on Wine, twelve DLLs' tests are currently failing. (Let alone the test failures on Windows!) http://bugs.winehq.org/show_bug.cgi?id=9916 also talks about the fact that "make test" shows failures. This situation greatly confuses new developers, who assume that any test failure is somehow their fault. And that's a reasonable assumption, let's make it so! It would make validating a new Wine development environment easy -- it'd be great to be able to say If it doesn't pass "make test" with no errors, something's wrong with the build environment. What's the best way to encourage people to fix these problems? Should we designate some upcoming day, week, or month as a conformance test bug squishing party? Should we offer prizes (or at least notoriety) for the person who does the best job fixing conformance test suite bugs (as judged by Alexandre or his designate)? Or is a simple call to arms enough?

5 8

Re: comctl32: Fix invalid syntax.
by Bang Jun-young Feb. 7, 2008

Feb. 7, 2008

18 hours passed, and it looks like Alexandre decided to ignore this...why? This fix is required for Wine to be built with VC2005/2008 (although Alexandre doesn't seem to care about it). On 2/6/08, Bang Jun-young <junyoung(a)mogua.com> wrote: > Any reason why this hasn't been committed? >

3 3

Re: gdi32: correct the test message in font.c that was out of sync with the test. [RESEND]
by Alexandre Julliard Feb. 7, 2008

Feb. 7, 2008

"Reece Dunn" <msclrhd(a)googlemail.com> writes: > diff --git a/dlls/gdi32/tests/font.c b/dlls/gdi32/tests/font.c > index 7e560d2..bfdb0af 100644 > --- a/dlls/gdi32/tests/font.c > +++ b/dlls/gdi32/tests/font.c > @@ -604,7 +604,7 @@ static void test_GetGlyphIndices(void) > flags |= GGI_MARK_NONEXISTING_GLYPHS; > charcount = pGetGlyphIndicesW(hdc, testtext, (sizeof(testtext)/2)-1, glyphs, flags); > ok(charcount == 5, "GetGlyphIndices count of glyphs should = 5 not %d\n", charcount); > - ok((glyphs[4] == 0x001f || glyphs[4] == UNICODE_NOCHAR /* Vista */), "GetGlyphIndices should have returned a nonexistent char not %04x\n", glyphs[4]); > + ok((glyphs[4] == 0x001f || glyphs[4] == UNICODE_NOCHAR /* Vista */), "GetGlyphIndices should have returned a nonexistent char or UNICODE_NOCHAR, not %04x\n", glyphs[4]); UNICODE_NOCHAR is a nonexistent char too. And if someone really cares about the exact details they can always look at the test, we don't need to duplicate all of it in the error message. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: user32: make the listbox tests pass in directories other than dlls/user32/tests. [RESEND]
by Alexandre Julliard Feb. 7, 2008

Feb. 7, 2008

"Reece Dunn" <msclrhd(a)googlemail.com> writes: > diff --git a/dlls/user32/tests/listbox.c b/dlls/user32/tests/listbox.c > index 3645190..f1a7b2b 100644 > --- a/dlls/user32/tests/listbox.c > +++ b/dlls/user32/tests/listbox.c > @@ -560,6 +560,8 @@ static void test_listbox_LB_DIR() > char * p; > char driveletter; > > + fclose( fopen( "wtest1.tmp.c", "w" ) ); Using stdio is not really appropriate here, CreateFile() would be a better choice. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: shell32: Fix security issue in SHELL_execute.
by Alexandre Julliard Feb. 7, 2008

Feb. 7, 2008

Robert Shearman <rob(a)codeweavers.com> writes: > execfunc should not be called on files which aren't supposed to be > executable. This doesn't work: ../../../tools/runtest -q -P wine -M shell32.dll -T ../../.. -p shell32_test.exe.so shlexec.c && touch shlexec.ok shlexec.c:971: Test failed: ShellExecuteEx(null, "C:\windows\temp\\test_shortcut_exe.lnk", null, null) failed: rc=31 err=2 shlexec.c:974: Test failed: argvA3 expected 'Lnk', got '(null)' shlexec.c:1010: Test failed: ShellExecuteEx(null, "C:\windows\temp\\test_shortcut_exe.lnk", ""C:\windows\temp\\nonexistent.shlexec"", null) failed: rc=31 err=2 shlexec.c:1010: Test failed: ShellExecuteEx(null, "C:\windows\temp\\test_shortcut_exe.lnk", ""C:\windows\temp\\nonexistent.noassoc"", null) failed: rc=31 err=2 shlexec.c:1010: Test failed: ShellExecuteEx(null, "C:\windows\temp\\test_shortcut_exe.lnk", ""C:\windows\temp\\test file.shlexec"", null) failed: rc=31 err=2 shlexec.c:1010: Test failed: ShellExecuteEx(null, "C:\windows\temp\\test_shortcut_exe.lnk", ""C:\windows\temp\/%nasty% $file.shlexec"", null) failed: rc=31 err=2 shlexec.c:1010: Test failed: ShellExecuteEx(null, "C:\windows\temp\\test_shortcut_exe.lnk", ""C:\windows\temp\\test file.noassoc"", null) failed: rc=31 err=2 shlexec.c:1062: Test failed: ShellExecuteEx(null, "Z:\home\julliard\wine\wine\dlls\shell32\tests\shell32_test.exe.so", "shlexec "C:\windows\temp\wt51f.tmp" Exec", null) returned 31 shlexec.c:1064: Test failed: argvA3 expected 'Exec', got '(null)' shlexec.c:1071: Test succeeded inside todo block: ShellExecute(null, "C:\windows\temp\\test file.noassoc", "shlexec "C:\windows\temp\wt51f.tmp" Exec", null) succeeded: rc=31 shlexec.c:1093: Test failed: ShellExecuteEx(null, "Z:\home\julliard\wine\wine\dlls\shell32\tests\shell32_test.exe.so", "shlexec "C:\windows\temp\wt51f.tmp" abcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxy", null) returned 31 shlexec.c:1095: Test failed: argvA3 expected 'abcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxy', got '(null)' shlexec.c:1102: Test succeeded inside todo block: ShellExecute(null, "C:\windows\temp\\test file.noassoc", "shlexec "C:\windows\temp\wt51f.tmp" abcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxy", null) succeeded: rc=31 make: *** [shlexec.ok] Error 13 -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: msi 2: Handle carriage returns in MsiDatabaseImport
by Alexandre Julliard Feb. 7, 2008

Feb. 7, 2008

"James Hawkins" <truiken(a)gmail.com> writes: > Hi, > > Fixes bug 11479. http://bugs.winehq.org/show_bug.cgi?id=11479 This doesn't work: ../../../tools/runtest -q -P wine -M msi.dll -T ../../.. -p msi_test.exe.so automation.c && touch automation.ok automation.c:276: Test failed: Expected ERROR_SUCCESS, got 1627 wine: Unhandled page fault on read access to 0x00000018 at address 0x6002a3da (thread 0009), starting debugger... Unhandled exception: page fault on read access to 0x00000018 in 32-bit code (0x6002a3da). Register dump: CS:0023 SS:002b DS:002b ES:002b FS:0063 GS:006b EIP:6002a3da ESP:0033f400 EBP:0033f578 EFLAGS:00010206( - 00 - RIP1) EAX:00000018 EBX:60121844 ECX:00000018 EDX:0033f5b8 ESI:602adce0 EDI:0033f52d Stack dump: 0x0033f400: 00000000 4d430002 00132fac 00000004 0x0033f410: 600425f6 0033f44c 6065e20e 602adce0 0x0033f420: 0033f5b8 0033f5b8 00000001 7bc417fa 0x0033f430: 00000000 4d430002 00133004 00000004 0x0033f440: 00000010 0033f47c 0033f464 00000030 0x0033f450: 00134180 00110014 00000020 00000000 Backtrace: =>1 0x6002a3da vsnprintfW+0x3ca(str=0x33f67a, len=0x7fffffff, format=0x6065e20a, valist=0x33f5b8) [/home/julliard/wine/wine/libs/wine/string.c:376] in libwine.so.1 (0x0033f578) 2 0x6002a596 sprintfW+0x36(str=0x33f678, format=0x6065e20a) [/home/julliard/wine/wine/libs/wine/string.c:458] in libwine.so.1 (0x0033f5a8) 3 0x60610808 MSI_DatabaseImport+0x13e8(db=0x130ef8, folder=<is not available>, file=0x1329ca) [/home/julliard/wine/wine/dlls/msi/database.c:603] in msi (0x0033f7f8) 4 0x60610b29 MsiDatabaseImportW+0x129(handle=<register EDI not in topmost frame>, szFolder=0x12fcd0, szFilename=0x1329b0) [/home/julliard/wine/wine/dlls/msi/database.c:777] in msi (0x0033f828) 5 0x60610cce MsiDatabaseImportA+0x18e(handle=0x1, szFolder=0x605b43c0, szFilename=0x605a3db7) [/home/julliard/wine/wine/dlls/msi/database.c:803] in msi (0x0033f868) 6 0x60527631 test_Installer+0x571() [/home/julliard/wine/wine/dlls/msi/tests/automation.c:275] in msi_test (0x0033fbd8) 7 0x6052983b func_automation+0xebb() [/home/julliard/wine/wine/dlls/msi/tests/automation.c:2428] in msi_test (0x0033fe68) 8 0x60592cea run_test+0x14a(name=0x1103ec) [/home/julliard/wine/wine/dlls/msi/tests/../../../include/wine/test.h:389] in msi_test (0x0033fea8) 9 0x60593332 main+0x142(argc=<register ECX not in topmost frame>, argv=0x1103d0) [/home/julliard/wine/wine/dlls/msi/tests/../../../include/wine/test.h:437] in msi_test (0x0033fed8) 10 0x6059340b __wine_spec_exe_entry+0x5b(peb=0x7ffdf000) [/home/julliard/wine/wine/dlls/winecrt0/exe_entry.c:36] in msi_test (0x0033ff08) 11 0x7b8744ee start_process+0xee(arg=0x0) [/home/julliard/wine/wine/dlls/kernel32/process.c:883] in kernel32 (0x0033ffe8) 12 0x600288d7 wine_switch_to_stack+0x17() in libwine.so.1 (0x00000000) 0x6002a3da vsnprintfW+0x3ca [/home/julliard/wine/wine/libs/wine/string.c:376] in libwine.so.1: movzwl 0x0(%ecx),%edx 376 while (*striter) -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: 2/2 start.exe: Handle the process title argument (try3)
by Alexandre Julliard Feb. 7, 2008

Feb. 7, 2008

Alexander Nicolaysen Sørnes <alex(a)thehandofagony.com> writes: > + vi.dwOSVersionInfoSize = sizeof(vi); > + GetVersionExW(&vi); > + > + /* On Windows 9x start does not have the process title argument. > + It works such that if the first parameter begins with a double quote ("), > + it is taken as the process title. Thus 'start "notepad" a' will work on > + Windows 9x but not NT variants */ > + if(argc > 1 && (vi.dwMajorVersion >= 5 || (vi.dwMajorVersion == 4 && vi.dwPlatformId == 2))) The right way to check for Win9x is GetVersion() & 0x80000000. > + { > + int cmdcount; > + WCHAR* cmdline = GetCommandLineW(); > + WCHAR** cmdargs = CommandLineToArgvW(cmdline, &cmdcount); > + WCHAR* pos = cmdline; > + > + pos += lstrlenW(cmdargs[0]); This won't work if the program name is quoted. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: systray[3/4]: Better validate icon owner
by Robert Shearman Feb. 7, 2008

Feb. 7, 2008

Kirill K. Smirnov wrote: > @@ -131,15 +130,18 @@ static LRESULT WINAPI adaptor_wndproc(HW > case WM_LBUTTONDBLCLK: > case WM_RBUTTONDBLCLK: > case WM_MBUTTONDBLCLK: > - /* notify the owner hwnd of the message */ > - WINE_TRACE("relaying 0x%x\n", msg); > - ret = PostMessage(icon->owner, icon->callback_message, (WPARAM) icon->id, (LPARAM) msg); > - if (!ret && (GetLastError() == ERROR_INVALID_HANDLE)) > + if (!IsWindow(icon->owner)) > { > WINE_WARN("application window was destroyed without removing " > "notification icon, removing automatically\n"); > delete_icon_directly(icon); > } > + else > + { > + /* notify the owner hwnd of the message */ > + WINE_TRACE("relaying 0x%x\n", msg); > + PostMessage(icon->owner, icon->callback_message, (WPARAM) icon->id, (LPARAM) msg); > + } > break; > > case WM_NCDESTROY: I don't get why you need this change. PostMessage should correctly handle the case where icon->owner has been destroyed and adding a call to IsWindow just introduces a race condition. -- Rob Shearman

2 2

Re: systray[3/4]: Better validate icon owner
by Kirill K. Smirnov Feb. 7, 2008

Feb. 7, 2008

> I don't get why you need this change. PostMessage should correctly > handle the case where icon->owner has been destroyed and adding a call > to IsWindow just introduces a race condition. Just a testcase: 1) run any wine app (to be sure that explorer is running), e.g winecfg 2) run taskmgr 3) select taskmgr process in tasklist and kill it. taskmgr window will dissapear but tray icon not. 4) move mouse cursor over taskmgr tray icon. Before this patch it will not dissapear, but should. The reason is that ERROR_INVALID_HANDLE is not the only value which indicates the absence of owner window (ERROR_INVALID_WINDOW_HANDLE too). I think that using robust and straightforward IsWindow() function is better than verify last error. Is it acceptable to check return value of PostMessage only and do not check last error? And what race condition introduces IsWindow() here? -- Kirill

2 1

Re: [PATCH] dinput: Skip old mouse movement evenets.
by Alexandre Julliard Feb. 7, 2008

Feb. 7, 2008

Vitaliy Margolen <wine-patches(a)kievinfo.com> writes: > @@ -280,6 +281,9 @@ static void dinput_mouse_hook( LPDIRECTINPUTDEVICE8A iface, WPARAM wparam, LPARA > { > POINT pt, pt1; > > + /* Skip old movement events */ > + if (This->last_event_time > hook->time) break; That's not an appropriate way to compare timestamps. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: x11drv: implement X11DRV_AlphaRender for non bitmap sources
by Alexandre Julliard Feb. 7, 2008

Feb. 7, 2008

Aric Stewart <aric(a)codeweavers.com> writes: > @@ -1569,12 +1570,14 @@ BOOL X11DRV_AlphaBlend(X11DRV_PDEVICE *devDst, INT xDst, INT yDst, INT widthDst, > > if (!devSrc->bitmap || GetObjectW( devSrc->bitmap->hbitmap, sizeof(dib), &dib ) != sizeof(dib)) > { > - FIXME("not a dibsection\n"); > - return FALSE; > + TRACE("not a dibsection\n"); > + bitmap = FALSE; It can be an invalid handle too, you can't assume that it's a valid bitmap just because it isn't a DIB. Also naming the boolean 'bitmap' for the DIB section case is very confusing, both are bitmaps. > > - if (xSrc < 0 || ySrc < 0 || widthSrc < 0 || heightSrc < 0 || xSrc + widthSrc > dib.dsBmih.biWidth > - || ySrc + heightSrc > abs(dib.dsBmih.biHeight)) > + if (xSrc < 0 || ySrc < 0 || widthSrc < 0 || heightSrc < 0 || (bitmap && xSrc + widthSrc > dib.dsBmih.biWidth) > + || (bitmap && ySrc + heightSrc > abs(dib.dsBmih.biHeight))) Testing the flag all over the place is ugly, you should separate the code completely, have one block for DIBs and one block for DDBs, with a single if at the top. -- Alexandre Julliard julliard(a)winehq.org

1 0

regression in worms2
by luis.busquets＠ilidium.com Feb. 7, 2008

Feb. 7, 2008

The same bug happens in Patrician3.exe If you apply the attached patch if should be ok. Please let me know, if that worked also for you.

3 2

Re: systray[2/4]: correctly handle icon addition/deletion
by Kirill K. Smirnov Feb. 6, 2008

Feb. 6, 2008

В сообщении от Thursday 07 February 2008 02:58:42 вы написали: > Kirill K. Smirnov wrote: > > @@ -71,6 +71,12 @@ struct icon > > static struct tray tray; > > static BOOL hide_systray; > > > > +static BOOL add_icon(NOTIFYICONDATAW *nid); > > +static BOOL modify_icon(NOTIFYICONDATAW *nid); > > +static BOOL delete_icon(const NOTIFYICONDATAW *nid); > > +static BOOL delete_icon_directly(struct icon *icon); > > +static BOOL display_icon(struct icon *icon, BOOL hide); > > + > > /* adaptor code */ > > > > #define ICON_SIZE GetSystemMetrics(SM_CXSMICON) > > I don't see any reason why you need to add forward declarations for all > of these functions. Hmmmm..... It may be an artifact from my development when I need to call them from places before they defined.... Anyway, is it really bad to have them declared here? If so I'll remove it. -- Kirill

2 1

Re: systray[2/4]: correctly handle icon addition/deletion
by Robert Shearman Feb. 6, 2008

Feb. 6, 2008

Kirill K. Smirnov wrote: > @@ -71,6 +71,12 @@ struct icon > static struct tray tray; > static BOOL hide_systray; > > +static BOOL add_icon(NOTIFYICONDATAW *nid); > +static BOOL modify_icon(NOTIFYICONDATAW *nid); > +static BOOL delete_icon(const NOTIFYICONDATAW *nid); > +static BOOL delete_icon_directly(struct icon *icon); > +static BOOL display_icon(struct icon *icon, BOOL hide); > + > /* adaptor code */ > > #define ICON_SIZE GetSystemMetrics(SM_CXSMICON) > I don't see any reason why you need to add forward declarations for all of these functions. -- Rob Shearman

1 0

Re: systray[1/4]: rewrite tooltips support
by Kirill K. Smirnov Feb. 6, 2008

Feb. 6, 2008

> Please explain better what you are doing, "rewrite tooltips support" > doesn't say what is changing or why the change is needed (and "rewrite" > is an exaggeration, you are mostly just moving code around). The main issue this patch fixes is an "attempt to set/modify a tooltip to a nonexistent window". This may happen when application modifies tip text while icon is hidden (e.g. when adding initially disabled icon with tooltip). [verbose] The change I did: 1) Store tip text in the icon structure. 2) Move tooltip creation code to the separate function. 3) Immediately bind tooltip to icon window at creation. We can bind empty tooltip text with icon window and further just update text when needed. 4) set_tooltip function was renamed to update_tooltip_text and simplified. Since tooltip support code is relatively small, this was enough for me to exaggeratedly proclaim tooltip support be rewritten ;-) [/verbose] So this ChangeLog is better: Avoid modifying non-existent tooltip windows of non-existent icon windows when application updates tip of hidden icon. -- Kirill

1 0

Re: systray[1/4]: rewrite tooltips support
by Alexandre Julliard Feb. 6, 2008

Feb. 6, 2008

"Kirill K. Smirnov" <lich(a)math.spbu.ru> writes: > diff --git a/programs/explorer/systray.c b/programs/explorer/systray.c Please explain better what you are doing, "rewrite tooltips support" doesn't say what is changing or why the change is needed (and "rewrite" is an exaggeration, you are mostly just moving code around). -- Alexandre Julliard julliard(a)winehq.org

1 0

Building tests with Visual Studio 2005 Express
by Dan Kegel Feb. 6, 2008

Feb. 6, 2008

After giving up for the moment on building with Visual C 6, I tried Visual Studio 2005. First hitch: Boy, does it spew out lots of warnings about deprecated functions! http://forums.microsoft.com/MSDN/ShowPost.aspx?PostID=87401&SiteID=1 talks about this a bit; defining _CRT_SECURE_NO_DEPRECATE and _CRT_NONSTDC_NO_DEPRECATE suppresses the warnings. Fix: You can't set them via corewin_express.vsprops because our project is converted from an older visual studio version, so you have to add those two defines to all the .dsp files, e.g. perl -p -i -e 's,/D "WIN32",/D "WIN32" /D "_CRT_NONSTDC_NO_DEPRECATE" /D "_CRT_SECURE_NO_DEPRECATE",' `find . -iname '*.dsp'` Second hitch: For some reason, the imported projects lose their list of .lib files to link against. Fix: As suggested at http://msdn2.microsoft.com/en-us/express/aa700755.aspx add user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib to the "Additional Dependencies" setting in the file corewin_express.vsprops. Third hitch: error C2065: 'PFORM_INFO_2A': undeclared identifier See http://msdn2.microsoft.com/en-us/library/aa383745(VS.85).aspx and http://blogs.msdn.com/oldnewthing/archive/2007/04/11/2079137.aspx winspool.h only defines the new symbol if NTDDI_VERSION is NTDDI_LONGHORN or higher. NTDDI_LONGHORN is 0x06000000. Fix: add /D _WIN32_WINNT=0x0600 to that list of changes to all the .dsp files in hitch #1 above. Fourth hitch: urlmon_test.exe refers to CLSID_IdentityUnmarshal, but that symbol is not defined. ?? Fifth hitch: generated.c refers to CS_STUB_INFO, which is not defined in Microsoft's rpcndr.h. ?? Sixth hitch: it can't find server.h Seventh hitch: can't find tmarshal.h Fix: Not sure, but maybe these files need to be generated by midl, and msvcmaker doesn't know about that yet? See http://www.mail-archive.com/wine-devel(a)winehq.org/msg37442.html Eighth hitch: riched20/tests/editor.c fails to compile because in GETTEXTEX, the field is named lpUsedDefChar, not lpUsedDefaultChar Fix: fix our richedit.h? ... Seventy-seventh hitch: gdiplus/tests/brush.c includes Microsoft's gdiplus.h, but that's not safe to include from C, so you get an error on the namespace. Golly. I guess I won't be building the tests with any flavor of microsoft C any time soon! - Dan

5 6

Re: added wgl_bind_to_texture_depth_nv to ConvertAttribWGLtoGLX
by Chris Robinson Feb. 6, 2008

Feb. 6, 2008

On Wednesday 06 February 2008 11:44:58 am you wrote: > Chris Robinson wrote: > > I don't think it's correct to add this since it's not supported (the > > extension string isn't exported and it can't be easilly emulated). If an > > app is trying to set that attribute without the extension being > > advertised, I'd say the app is broken. If you're intending on adding > > support, you may want to familiarize yourself with how wine emulates > > render_to_texture and look over the extension: > > http://opengl.org/registry/specs/NV/render_depth_texture.txt > > "I don't think it's correct to add this since it's not supported" > The > > case WGL_BIND_TO_TEXTURE_RGB_ARB: > case WGL_BIND_TO_TEXTURE_RGBA_ARB: > case WGL_BIND_TO_TEXTURE_RECTANGLE_FLOAT_R_NV: > are not supported either. The WGL_BIND_TO_TEXTURE_DEPTH_NV came up as a > fixme since it isn't in the switch. The formats are supported in X11DRV_wglGetPixelFormatAttribivARB, X11DRV_wglCreatePbufferARB and the other render-to-texture functions, and the render-to-texture extension is advertised.. There's just nothing to do in that function for them since the hint will always be accepted. But I don't see the point in supporting a pbuffer creation hint that an app shouldn't be using without the extension being advertised (actually, I think pbuffer creation is supposed to fail if an unsupported attribute is specified). Which app is trying to use it? > Again, just trying to learn, not annoy > people with tiny fixes. No annoyance to me, just explaining why I don't think it's something to handle. :)

1 0

Stupid new winetricks: added a trivial GUI
by Dan Kegel Feb. 6, 2008

Feb. 6, 2008

Some users freak at the thought of a commandline, so I threw together a trivial GUI using whatever gui-from-shell tool is appropriate for the current desktop. Unfortunately, that means the default action is to put up a gui, not show help. You have to say winetricks -h or the like to get help. It's ugly, so shoot me :-) As always, it's available at http://kegel.com/wine/winetricks and the svn repository is http://code.google.com/p/winezeug/

2 2

Re: RegOverridePredefKey stub - Take 2
by Dmitry Timoshkov Feb. 6, 2008

Feb. 6, 2008

"Jens Nestler" <nessi(a)nessi-online.de> wrote: > Stub implementation for RegOverridePredefKey. > The added prototype in the winreg.h is now in alphabetical order. Please review yuor patches before sending, this one contains not related changes. -- Dmitry.

1 0

Re: [5/8] WineD3D: Add infrastructure for faking driver versions
by Alexandre Julliard Feb. 6, 2008

Feb. 6, 2008

Stefan Dösinger <stefan(a)codeweavers.com> writes: > +/* Certain applications(Steam) complain if we report an outdated driver version. In general, > + * reporting a driver version is moot because we are not the Windows driver, and we have different > + * bugs, features, etc. > + * > + * Some drivers(ATI, Nvidia Linux drivers) have a shared codebase with the Windows driver, so we're > + * reading the Linux driver version and report it to the app. Some drivers(Apple drivers, Mesa) don't > + * have the version match, so we overwrite the OpenGL driver version with a hardcoded one. Obviously > + * the hardcoded driver version is outdated once a new Windows driver is released, so we'll have to > + * keep them in sync manually. > + * > + * If a card is not found in this table, the gl driver version is reported > + */ > +struct driver_version_information { > + WORD vendor; /* reported PCI card vendor ID */ > + WORD card; /* reported PCI card device ID */ > + BOOL loword_matches_apple; /* Are Mac and Win driver versions matching? */ > + BOOL loword_matches_linux; /* Are Lin and Win driver versions matching? */ The world is not limited to Mac and Linux, this sort of thing can't scale. -- Alexandre Julliard julliard(a)winehq.org

3 6

Re: uxtheme: fix the system tests so they pass on Vista. [try 2] [RESEND]
by Alexandre Julliard Feb. 6, 2008

Feb. 6, 2008

"Reece Dunn" <msclrhd(a)googlemail.com> writes: > The Vista return code is what wine is returning, so remove the todo > block as well. Not for me: ../../../tools/runtest -q -P wine -M uxtheme.dll -T ../../.. -p uxtheme_test.exe.so system.c && touch system.ok system.c:420: Test failed: Expected E_POINTER or S_OK, got 0x80070490 make: *** [system.ok] Error 1 -- Alexandre Julliard julliard(a)winehq.org

1 0

winedbg - using pdb from Visual Studio 2005
by Michel Jacques Feb. 6, 2008

Feb. 6, 2008

Hello, I am using PDBs generated with 2005 with wine version 0.9.54. I get errors related to my PDBs and I am wondering if I need to change the pdb format? Anything special to do when using 2005? Here are the errors I get, each time one of my pdb gets loaded: 0009:trace:dbghelp:SymFindFileInPathW (hProcess = 0x30, searchPath = (null), full_path = L"C:\\master\\x86\\debug\\bin\\mydll.pdb", id = 0x34c160, two = 0x00000002, three = 0x00000000, flags = 0x00000008, buffer = 0x34b718, cb = 0x6020fb50, user = 0x34bd30) 0009:err:dbghelp_msc:pdb_process_types -Unknown type info version 20040203 0009:err:dbghelp_msc:pe_load_debug_directory Got a page fault while loading symbols 0009:trace:dbghelp:pcs_callback 0x111998 2 0x34c378 0009:trace:dbghelp:pcs_callback 0x111998 1 0x34c378 0009:trace:dbghelp:pe_load_stabs failed to load the STABS debug info Thanks, Michel Jacques Découvrez les styles qui font sensation sur Yahoo! Québec Avatars. http://cf.avatars.yahoo.com/

4 5

Re: [winetest] Add a new utility to simply running + submitting tests via winetest when run from the Wine tree.
by Alexandre Julliard Feb. 6, 2008

Feb. 6, 2008

Jeremy White <jwhite(a)winehq.org> writes: > +date +%Y%m%d >build.id > +gcc --version >build.nfo > +uname -a >>build.nfo > +/lib/libc.so.6 >>build.nfo > +echo N/A >tests.url > +make dist > +make > +export WINEPREFIX=`mktemp -d` > +wineprefixcreate > + > +type winetricks >/dev/null 2>&1 > +if [ $? -eq 0 ] ; then > + winetricks gecko > +fi > + > +./winetest-dist -c -t "$1" None of that should be necessary. The make dist stuff is supposed to be for people who want to distribute a winetest binary. If what you want is simply run the tests then you should just have to run winetest; if some things don't work in that mode they should be fixed. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: Bug in patch winex11.drv: Don't create a win_data structure for the desktop window, except in the process that owns it.
by Vitaliy Margolen Feb. 6, 2008

Feb. 6, 2008

What bug # is this? Please note this is not bugzilla, you will be expected to debug this problem yourself and provide technical information. Vitaliy Luis C. Busquets Pérez wrote: > There is a bug in the patch: > > c19af910ebafa9b1928b0f67bee68f6b0b237e8e is first bad commit > commit c19af910ebafa9b1928b0f67bee68f6b0b237e8e > Author: Alexandre Julliard <julliard(a)winehq.org> > Date: Thu Jan 17 19:53:59 2008 +0100 > > winex11.drv: Don't create a win_data structure for the desktop > window, except in the process that owns it. > > Handle desktop size changes by sending a message to the desktop owner. > > :040000 040000 b7c030353fe92121bb999481724a3689429b2e8a > 80fd2c5f4ace2707abb01648bd61663a6c9d7626 M dlls > > > It affects to Patrician3.exe. When initiating, it hangs and does not > allow to play the movie. Actually, the programm changes the screen > resolution and hangs. I understand that the problem is that it does not > send the appropiate message for binkplayer.exe to play the movie. > >

3 2

New winetricks: msxml6, riched30
by Dan Kegel Feb. 6, 2008

Feb. 6, 2008

Johan Dahlin sent in a patch to add msxml6, and said "It's enough to make msi2xml.exe to at least start-up (before failing in msi later)." Tom Wickline some time ago sent in a suggestion for adding riched30. I didn't know any apps it helped, but apparantly Novell thinks it helps Groupwise a bit. So it's in now. Caution, it doesn't update riched32, no idea what happens if apps try to use it after this... As always, it's available at http://kegel.com/wine/winetricks and the svn repository is http://code.google.com/p/winezeug/

1 0

regression in worms2
by Olaf Leidinger Feb. 6, 2008

Feb. 6, 2008

Hello! I noticed that the game "worms 2" broke between wine-0.9.53 and wine-0.9.54. A git-bisect revealed the guilty patches. The game includes a frontend.exe which starts game.exe and controls the settings of the game (the map and teams etc.). Running game.exe directly works but you can't alter the settings. When running game.exe via frontend the game doesn't show up but hangs after changing resolution. Before the following commit the game starts and is playable: 2921f5c156b368b4ce6e615eb61dad5c5f423bf8 is first bad commit commit 2921f5c156b368b4ce6e615eb61dad5c5f423bf8 Author: Alexandre Julliard <julliard(a)winehq.org> Date: Tue Jan 22 20:06:35 2008 +0100 server: Invalidate the correct region when custom valid rects are specified. :040000 040000 1385d5899b5b6affd5da84328e1bc15985406976 5c8756013a67c0402f3bb5fb9f3ee7b4a50e124a M server And this one makes worms2 hang after exiting the game: c19af910ebafa9b1928b0f67bee68f6b0b237e8e is first bad commit commit c19af910ebafa9b1928b0f67bee68f6b0b237e8e Author: Alexandre Julliard <julliard(a)winehq.org> Date: Thu Jan 17 19:53:59 2008 +0100 winex11.drv: Don't create a win_data structure for the desktop window, except in the process that owns it. Handle desktop size changes by sending a message to the desktop owner. :040000 040000 b7c030353fe92121bb999481724a3689429b2e8a 80fd2c5f4ace2707abb01648bd61663a6c9d7626 M dlls Thanks for your attention! Ciao, Olaf

3 7

Re: RegOverridePredefKey stub
by Dmitry Timoshkov Feb. 5, 2008

Feb. 5, 2008

"Jens Nestler" <nessi(a)nessi-online.de> wrote: > diff -urN wine-0.9.54_org/include/winreg.h wine-0.9.54/include/winreg.h > --- wine-0.9.54_org/include/winreg.h 2008-01-25 17:05:38.000000000 +0100 > +++ wine-0.9.54/include/winreg.h 2008-01-28 22:07:55.000000000 +0100 > @@ -131,6 +131,7 @@ > WINADVAPI LSTATUS WINAPI RegLoadKeyA(HKEY,LPCSTR,LPCSTR); > WINADVAPI LSTATUS WINAPI RegLoadKeyW(HKEY,LPCWSTR,LPCWSTR); > #define RegLoadKey WINELIB_NAME_AW(RegLoadKey) > +WINADVAPI LSTATUS WINAPI RegOverridePredefKey(HKEY,HKEY); > WINADVAPI LSTATUS WINAPI RegLoadMUIStringA(HKEY,LPCSTR,LPSTR,DWORD,LPDWORD,DWORD,LPCSTR); > WINADVAPI LSTATUS WINAPI RegLoadMUIStringW(HKEY,LPCWSTR,LPWSTR,DWORD,LPDWORD,DWORD,LPCWSTR); > #define RegLoadMUIString WINELIB_NAME_AW(RegLoadMUIString) Isn't there something that suggests that the entries are alphabetically sorted? -- Dmitry.

2 4

Building tests with MSVC 6 broken?
by Dan Kegel Feb. 5, 2008

Feb. 5, 2008

Call me an oldtimer, but I decided to try building winetest with MSVC 6 today. I installed cygwin's git and perl, grabbed wine from git, and used msvcmaker as described at http://winehq.org/site/docs/winedev-guide/testing-windows I downloaded and installed the Platform SDK from the first google hit, the Windows Server 2003 SP1 Platform SDK, http://www.microsoft.com/downloads/details.aspx?FamilyId=A55B6B43-E24F-4EA3… (although the latest is Windows Server 2003 R2 Platform SDK, I thought the slightly older one would be ok). Oddly, Microsoft's Platform SDK page lists the SP1 SDK as the latest. Pffft. I then used Tools/ Options / Directories to tell msdev about include and lib in both msvc6 and the platform SDK. Then I tried building wineruntests. First hitch: error C2065: 'RRF_RT_REG_DWORD' : undeclared identifier Fix: Move Platform SDK higher in the priority list in Tools / Options / Directories. msdev will then use the newer headers, which have that symbol. Second hitch: Linking... uuid.lib(cguid_i.obj) : fatal error LNK1103: debugging information corrupt; recompile module Error executing link.exe. Fix: From http://blogs.msdn.com/stephen_griffin/archive/2006/08/03/building-mfcmapi-w… In the platform SDK's lib directory, rename uuid.lib to uuid.lib.bad. msdev will then use the older library. (Microsoft intentionally removed VC6 support from the newer one, whoops.) Third hitch: http://www.winehq.org/pipermail/wine-patches/2007-September/043988.html introduced a type LSTATUS into winreg.h. This isn't present in the 2003 SP1 platform sdk, nor in the newer 2003 R2 platform SDK, but as long as only our winreg.h uses it, that's fine. Sadly, that patch uses it elsewhere, so compilation breaks. A google search for LSTATUS at microsoft.com finds nothing. Fix: change the LSTATUS in comctl32/tests/mru.c to LONG (gee, maybe we should move the definition of LSTATUS from winreg.h to advapi32/registry.c, too). (But see below.) Fourth hitch: comdlg32\tests\printdlg.c: error C2064: 'LPPRINTDLGEXW' : undeclared identifier Turns out commdlg.h in the platform sdk only defines LPPRINTDLGEXW if STDMETHOD is declared. STDMETHOD is declared in basetyps.h, but including that file makes commdlg.h not even compile. Bleah. Fix: replace the #include "windef.h" ... "winuser.h" with #include "windows.h" in comdlg32/tests/printdlg.h. Fifth hitch: crypt32/tests/cert.c: error C2081: 'HCRYPTPROV_OR_NCRYPT_KEY_HANDLE': name in formal parameter list illegal Turns out that identifier was introduced in Vista, and added to Wine here: http://www.winehq.org/pipermail/wine-patches/2007-August/042250.html Fix: Uninstall the platform SDK and install the Windows SDK! http://www.microsoft.com/downloads/details.aspx?familyid=c2b1e300-f358-4523… Sixth hitch: c:\program files\microsoft sdks\windows\v6.0\include\specstrings.h: fatal error C1083: cannot open include file 'sal.h' I think sal.h comes with Visual Studio 8 and higher. Crap. Thus it looks like you can't build wine's tests with Visual C 6 and the Microsoft headers anymore. So I tried building with the Wine headers. That blows up with "can't find stdio.h". So I'm stuck. What's the recommended way to build the wine test suite with Visual C 6, if any? - Dan

3 5

re: [Patch for Bug 8551] MoveFileWithProgressW unconditional fails for directories with flag MOVEFILE_REPLACE_EXISTING
by Dan Kegel Feb. 5, 2008

Feb. 5, 2008

James McKenzie wrote: > Please submit a git diff not a UNIX diff. Now, hang on there. Jens submitted a perfectly good unified diff patch taken from the right directory. There is no reason to turn up our noses at it. It will apply using patch in exactly the same way as a diff generated by git. Jens, you're fine, I wouldn't bother rediffing. You might however resend your patch next week when Alexandre the maintainer is back from vacation! - Dan

2 1

Re: added wgl_bind_to_texture_depth_nv to ConvertAttribWGLtoGLX
by Chris Robinson Feb. 5, 2008

Feb. 5, 2008

On Tuesday 05 February 2008 02:09:09 pm Christopher Harvey wrote: > @@ -758,6 +758,7 @@ static int ConvertAttribWGLtoGLX(const int* iWGLAttr, > int* oGLXAttr, Wine_GLPBuf nvfloatattrib = iWGLAttr[++cur]; > TRACE("pAttr[%d] = WGL_FLOAT_COMPONENTS_NV: %x\n", cur, > nvfloatattrib); break ; > + case WGL_BIND_TO_TEXTURE_DEPTH_NV: > case WGL_BIND_TO_TEXTURE_RGB_ARB: > case WGL_BIND_TO_TEXTURE_RGBA_ARB: > case WGL_BIND_TO_TEXTURE_RECTANGLE_FLOAT_R_NV: I don't think it's correct to add this since it's not supported (the extension string isn't exported and it can't be easilly emulated). If an app is trying to set that attribute without the extension being advertised, I'd say the app is broken. If you're intending on adding support, you may want to familiarize yourself with how wine emulates render_to_texture and look over the extension: http://opengl.org/registry/specs/NV/render_depth_texture.txt

1 0