- wine-devel - winehq.org

wine for two, please.
by Athel 30 Nov '07

30 Nov '07

Hi Rob, i was looking a long time the way to do this. Can you help me to done this?, plis I really need it. If I could understand better the functionality I'll do it for myself. Sorry for my english. Regaards Athel > I was going to wait til tomorrow to test this, but couldnt wait. > This works perfectly for my app and couldnt have came at a better time. > > I just created a simple Win32 dispatcher program for each screen and sent > the window creation messages to the HWNDs from FindWindow instead of the > local app and it creates the dialogs on any screen I want. Joy Joy!!!!! > > It worked out perfectly since the threads that create windows have to send > messages to the main loop for it to create them so that the dialogs dont > from close when the thread exits anyway. > > It is refreshing that after I went through all the battles of getting > multihead display, gcc, and wine running on Solaris that FINALLY something > would work so simply. > > Thanks again to the whole Wine crew that made this possible!!!

1 0

Re: widl [1/7]: Output working exception handling macros. [take 3]
by Alexandre Julliard 29 Nov '07

29 Nov '07

Dan Hipschman <dsh(a)linux.ucla.edu> writes: > In this revision, I leave setting permissions on page faults to the > unhandled exception filter and simply ignore page faults on execution > access (read access exceptions should be caught since they translate > into bad stub data, but these don't pose a problem anyway). I also > fixed a bug regarding how the mask field of __widl_except_frame was > set. There's no reason to ignore execute access errors, those are bugs too. You are still not filtering the exception properly though, the condition checking has to be done directly in the handler, you can't longjmp back to the code and then re-raise the exception. That's doubly true for the finally handling. This means you have to generate different handlers for the different exception conditions. -- Alexandre Julliard julliard(a)winehq.org

2 2

Interesting Behavior of OS X
by Steven Edwards 29 Nov '07

29 Nov '07

When tracking down a crash in the kernel32 loader test, Dmitry found a bug in the Mac OS loader when Wine tried to load his dummy PE file. Upon further research I found that the Mac loader seems to have its own undocumented PE loader built in. I did some further testing with a windows binary and got some really interesting results.... #include <stdio.h> #include <dlfcn.h> int main(int argc, char **argv) { void *handle; const char *error; // handle = dlopen("./ldr330e.tmp", RTLD_NOW | RTLD_FIRST ); handle = dlopen("./procexp.exe", RTLD_NOW | RTLD_FIRST ); if (!handle) { fputs (dlerror(), stderr); printf("I guess it didn't load\n"); exit(1); } printf("I guess it loaded\n"); dlclose(handle); } Its trying to load the dlls. I get this output steven-edwardss-imac:temp sedwards$ ./a.out dlopen(./procexp.exe, 258): Library not loaded: WS2_32.dll Referenced from: /Users/sedwards/Library/Application Support/CrossOver/Bottles/winetest/drive_c/windows/temp/procexp.exe Reason: image not foundI guess it didn't load steven-edwardss-imac:temp sedwards$ file procexp.exe procexp.exe: MS-DOS executable PE for MS Windows (GUI) Intel 80386 32-bit steven-edwardss-imac:temp sedwards$ open . steven-edwardss-imac:temp sedwards$ ./a.out dlopen(./procexp.exe, 258): Library not loaded: MPR.dll Referenced from: /Users/sedwards/Library/Application Support/CrossOver/Bottles/winetest/drive_c/windows/temp/procexp.exe Reason: image not foundI guess it didn't load steven-edwardss-imac:temp sedwards$ ./a.out dlopen(./procexp.exe, 258): Library not loaded: MPR.dll So this leads to the question. Whats going on? Is this a hold over from EFI which is PE by default? Why would the OS need to load the EFI files? Maybe just for easy of development and testing? Or is something else going on? Is Apple going to be adding a win32 compatibility layer to OS X? Is having a native PE loader of any use to us? Thanks Steven -- 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

1 2

re: wininet [2/2]: Fix error checking for FtpGetCurrentDirectoryW
by Dan Kegel 29 Nov '07

29 Nov '07

Hi Zac, you wrote * Fixed errors in error checking in FtpGetCurrentDirectoryW in wininet/ftp.c It's probably a good idea to explain what real-world problem this fixes. One good way to do that is to submit a test first that fails, with todo(wine) around the failing part, and then submit a second patch which fixes the problem and removes the todo(wine) from the test. - if (NULL == lpwfs || WH_HFTPSESSION != lpwfs->hdr.htype) + if ( lpwfs == NULL || WH_HFTPSESSION != lpwfs->hdr.htype ) Why did you make this change? It doesn't fix anything, and it removes an old coder's anti-typo-measure of putting the const on the left hand of an equality test. Please remove this hunk. - if (lpwfs->download_in_progress != NULL) + if ( lpszCurrentDirectory == NULL || *lpdwCurrentDirectory == NULL ) You switched to tabs, tsk. Also, you changed whitespace style. Always try to match the file's existing style. - Dan -- Wine for Windows ISVs: http://kegel.com/wine/isv

1 0

re: wininet [1/2]: Add test for FtpGetCurrentDirectoryA
by Dan Kegel 29 Nov '07

29 Nov '07

Nice to see you contributing tests! Three comments: 1. you seem to have ragged indentation. Did you remember to remove all tabs? (Hmm. Our doc on the tab issue is a bit weak: neither http://winehq.org/site/sending_patches http://winehq.org/site/docs/winedev-guide/style-notes really capture the current anti-tab mood.) 2. You have a nice testDir variable, but then you hard-code "pub" somewhere. Use the variable everywhere. 3. Did you make sure the new tests pass on Windows? -- Wine for Windows ISVs: http://kegel.com/wine/isv

1 0

Adding Bugzilla keyword for Copy-protection/Debugger issues
by Ben Hodgetts (Enverex) 29 Nov '07

29 Nov '07

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Just a thought but it may be a good idea to add a keyword to Bugzilla for issues related to debuggers or copy-protection, that would help group them all together as at the moment there seem to be many bugs related to breakages from obscure debugger protections or things like GameGuard. It's much cleaner than the hated meta bug idea and is easier to implement. Just a thought. Ben H. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFHTrfQuqCUopk11kIRAhDnAJ4iGkr6bUk1/FPUFrrdzlsnqC5dDwCfd2r5 WDylDKMPprC22RtEwYG01o4= =fLJB -----END PGP SIGNATURE-----

2 1

Re: [server] set serial port fd back to blocking if necessary
by Alexandre Julliard 29 Nov '07

29 Nov '07

"Damjan Jovanovic" <damjan.jov(a)gmail.com> writes: > This fixes a regression caused by commit > f85437c57fb44d511db97edbaa5b1f8abfe75fd3 (bug 9356) where serial ports > break because they are not made blocking. > > Changelog: > * Set serial port fd back to blocking if necessary This will break timeout handling. There shouldn't be any reason for the fd to be blocking, we should handle EAGAIN properly. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH 1/3] advpack/tests: Rename the wrappers around HeapAlloc() &Co to use the new standard naming.
by James Hawkins 28 Nov '07

28 Nov '07

On Nov 28, 2007 3:39 PM, Michael Stefaniuc <mstefani(a)redhat.de> wrote: > Alexandre, > > i'm unsure about this patch series. Yes this are simple wrappers around > HeapAlloc() that are the same as the "standard" wrappers used in Wine. > But they are used only as callbacks aka different usage. > I don't think this particular series is beneficial. Your script won't be able to use these names because they're not called directly. -- James Hawkins

2 1

d3dx9__xx dlls
by David.Adam＠math.cnrs.fr 28 Nov '07

28 Nov '07

Following Alexandre (and I agree with him :D ), one only needs to export the functions of d3dx9_36 (the latest one) to the older d3dx9_xx dlls. So, we just just need to implement the functions in the d3dx9_36 dll repertory. No wine_d3dx9 is useful. David

5 5

Re: [try2 1/1] msi/tests: automation: Fix pointer check before calling lstrcpyW.
by Alexandre Julliard 28 Nov '07

28 Nov '07

Misha Koshelev <mk144210(a)bcm.edu> writes: > The previous patch was last week I think. This fixes a Valgrind error in the test. This is > correct as vtResult is just hte _expected_ return type, and this way we check the actual return type > (new this version) and then the validity of the pointer (same as old patch) before calling lstrcpyW > so we don't pass NULL to lstrcpyW. It seems to me what you really want to do is check that the call succeeded, instead of relying on the variant being cleared on error. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH] svchost: Implementation of svchost (revised submission 2)
by Alexandre Julliard 28 Nov '07

28 Nov '07

Roy Shea <roy(a)cs.hmc.edu> writes: > This is another revised and standalone version of the svchost patch. > Changes in this revision include: > > - Only using HEAP_ZERO_MEMORY in calls to HeapReAlloc. Calls to > HeapAlloc explicitly set the NULL terminator when required. The realloc is overkill too, you can just allocate one extra char from the start, and set it to zero unconditionally. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: ole32: Fix two comments in ok strings in the WdtpInterfacePointer tests.
by Alexandre Julliard 28 Nov '07

28 Nov '07

Robert Shearman <rob(a)codeweavers.com> writes: > diff --git a/dlls/ole32/tests/usrmarshal.c b/dlls/ole32/tests/usrmarshal.c > index 9959f74..5819ec1 100644 > --- a/dlls/ole32/tests/usrmarshal.c > +++ b/dlls/ole32/tests/usrmarshal.c > @@ -801,9 +801,9 @@ static void test_marshal_WdtpInterfacePointer(void) > wireip = buffer; > if (size >= 28) > { > - ok(*(DWORD *)wireip == 0x44, "wireip + 0x0 should be 0x4c instead of 0x%08x\n", *(DWORD *)wireip); > + ok(*(DWORD *)wireip == 0x44, "wireip + 0x0 should be 0x44 instead of 0x%08x\n", *(DWORD *)wireip); > wireip += sizeof(DWORD); > - ok(*(DWORD *)wireip == 0x44, "wireip + 0x8 should be 0x4c instead of 0x%08x\n", *(DWORD *)wireip); > + ok(*(DWORD *)wireip == 0x44, "wireip + 0x8 should be 0x44 instead of 0x%08x\n", *(DWORD *)wireip); While we are on the subject, why does it say wireip + 0x8 when the pointer is incremented by 4? The rest of the offsets are wrong too, especially the iid comparison. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [1/8] WineD3D: Inform the texture about filtering changes
by H. Verbeet 28 Nov '07

28 Nov '07

On 27/11/2007, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > Don't you need to mark the sampler dirty if it's bound somewhere? Or is it guaranteed it never is at that point?

2 1

Re: [4/8] WineD3D: Do not try to disable unsupported texture units
by H. Verbeet 28 Nov '07

28 Nov '07

On 27/11/2007, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > textures. In this case we build a texture map to read a texture from a > lower texture unit from a higher register combiner. This texture unit > always stays below GL_LIMITS(textures) though. Yes, but you're checking against the sampler, not the texture unit. Shouldn't you be checking against mapped_stage instead?

1 0

Re: [PATCH] svchost: Implementation of svchost (revised submission)
by James Hawkins 27 Nov '07

27 Nov '07

On Nov 27, 2007 3:21 PM, Roy Shea <roy(a)cs.hmc.edu> wrote: > This is a revised and standalone version of svchost patch. Changes in > this revision include: > > - Unicode based with a simple UNICODE to ASCII function used to > convert the lpProcName passed into GetProcAddress > I think you're misunderstanding Wine's policy about not using ansi functions or constant strings. That only applies to external APIs because we don't want to convert from unicode to ansi (lossy). There's nothing wrong with just using: GetProcAddress(library, "ServiceMain"); There aren't many instances where W->A is valid. > - Using HEAP_ZERO_MEMORY flag in calls to HeapAlloc and HeapReAlloc > that are allocating space for a string and the NULL terminator to > set the NULL terminator to zero. > Why HEAP_ZERO_MEMORY when you can just set the last byte to NULL? str[LEN - 1] = '\0'; -- James Hawkins

2 1

Re: d3dx9_xx [patch 2/2] Implementation of the different d3dx9_xx lls
by Maciej Kowalczyk 27 Nov '07

27 Nov '07

Hi! > +@ stdcall D3DXVec2Normalize(ptr ptr) d3dx8.D3DXVec3Normalize Shouldn't it be "d3dx8.D3DXVec2Normalize"? This line is copy-pasted 13 times around the patch. -- Maciej Kowalczyk

1 0

[keyboard] GetAsyncKeyState: request for fixing #5623
by Rafał Miłecki 27 Nov '07

27 Nov '07

Hello, I have two favourite games (Diablo II & Ultima Online) and both of them works great under Wine. Unfortunately this is quite hard to play UO without additional tool called "EasyUO" which doesn't work under Wine. EasyUO generally works fine but it can not detect pressing any key-combination in focued UO window. Possibility of detecting is the main functionality for EUO. This problem is cuased by little bug in support for GetAsyncKeyState function. I am waiting about 8 months now for fixing this bug and I wonder if I can ask sb to look info this bug. Just to make fixing this little faster. Bug is quite well decribed in: http://bugs.winehq.org/show_bug.cgi?id=5623 but if you don't understand this, I can try to explain more detailed. So my request is someone try to fix this bug. Anyone have a free moment for this, please? -- Rafał Miłecki

2 3

Re: urlmon: Callback can be NULL
by Alexandre Julliard 27 Nov '07

27 Nov '07

Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> writes: > Hi, > Changelog: > urlmon: The callback parameter can be null, check pointer before using. This breaks the tests: ../../../tools/runtest -q -P wine -M urlmon.dll -T ../../.. -p urlmon_test.exe.so stream.c && touch stream.ok stream.c:210: Test failed: unexpected call OnDataAvailable stream.c:329: Test failed: IStream_Read failed with error 0x00000001 stream.c:330: Test failed: read data differs from file make: *** [stream.ok] Error 3 -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: d3dx8: Implementation of WINAPI D3DXAssembleShaderFromFileA
by Dmitry Timoshkov 27 Nov '07

27 Nov '07

""Luis C. Busquets Pérez"" <luis.busquets(a)ilidium.com> wrote: > Changed the if clauses to make the implementation more readable. ... > + if ( !pSrcFile ) > + return D3DXERR_INVALIDDATA; > + else > + { > + len = MultiByteToWideChar( CP_ACP, 0, pSrcFile, -1, NULL, 0 ); > + pSrcFileW = HeapAlloc( GetProcessHeap(), 0, len * sizeof(WCHAR) ); > + MultiByteToWideChar( CP_ACP, 0, pSrcFile, -1, pSrcFileW, len ); > + ret=D3DXAssembleShaderFromFileW(pSrcFileW, Flags, ppConstants, ppCompiledShader, ppCompilationErrors); > + HeapFree( GetProcessHeap(), 0, pSrcFileW ); > + return ret; > + } > } This is not more readable than a previous version. There is no need for 'else {}' construct and therefore for additional indentation level at all. -- Dmitry.

4 4

Re: urlmon: Wrap heap functions.
by Michael Stefaniuc 27 Nov '07

27 Nov '07

Hello! (Jacek, this is not directed at you but a general rant). HeapAlloc is an awful API and even its creators wrap it in a couple of DLLs. In Wine HeapAlloc gets wrapped in different places, each using a different name and even a different naming convention. This makes the code harder to understand for people that browse the code (people doing janitorial work) as they have to check every time if the function is yet another wrapper around HeapFree or if the function is doing more than that. And this makes it even harder for static code analyzing tools. I therefor propose to standardize all the internal HeapAlloc wrappers to a common name and form, Something along the lines of: walloc() wzalloc() wrealloc() wzrealloc() wfree() It's short and concise and doesn't conflict with the Win32 API. Comments? bye michael Jacek Caban wrote: > diff --git a/dlls/urlmon/urlmon_main.h b/dlls/urlmon/urlmon_main.h > index 6c2887e..aeb0eb2 100644 > --- a/dlls/urlmon/urlmon_main.h > +++ b/dlls/urlmon/urlmon_main.h > @@ -63,4 +63,24 @@ HRESULT start_binding(LPCWSTR url, IBindCtx *pbc, REFIID riid, void **ppv); > > HRESULT create_binding_protocol(LPCWSTR url, IInternetProtocol **protocol); > > +static inline void *urlmon_alloc(size_t len) > +{ > + return HeapAlloc(GetProcessHeap(), 0, len); > +} > + > +static inline void *urlmon_alloc_zero(size_t len) > +{ > + return HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY, len); > +} > + > +static inline void *urlmon_realloc(void *mem, size_t len) > +{ > + return HeapReAlloc(GetProcessHeap(), 0, mem, len); > +} > + > +static inline BOOL urlmon_free(void *mem) > +{ > + return HeapFree(GetProcessHeap(), 0, mem); > +} > + > #endif /* __WINE_URLMON_MAIN_H */ -- Michael Stefaniuc Tel.: +49-711-96437-199 Consulting Communications Engineer Fax.: +49-711-96437-111 -------------------------------------------------------------------- Reg. Adresse: Red Hat GmbH, Hauptstätter Strasse 58, 70178 Stuttgart Handelsregister: Amtsgericht Stuttgart HRB 153243 Geschäftsführer: Brendan Lane, Charlie Peters, Michael Cunningham, Werner Knoblich

2 2

Re: user32: Fix a couple of typos
by Alexandre Julliard 27 Nov '07

27 Nov '07

Andrew Talbot <andrew.talbot(a)talbotville.com> writes: > Changelog: > user32: Fix a couple of typos. The patch is clearly correct but it causes test failures, so it looks like more work is needed here: ../../../tools/runtest -q -P wine -M user32.dll -T ../../.. -p user32_test.exe.so dde.c && touch dde.ok dde.c:250: Test failed: Unhandled msg: 000080c2 dde.c:243: Test failed: client failed with 1 error(s) make[2]: *** [dde.ok] Error 1 -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: msvcrt heap: fix _aligned_offset_realloc (move correct block of memory); add a comment to explain why. [1/2]
by Alexandre Julliard 27 Nov '07

27 Nov '07

Lionel Debroux <lionel_debroux(a)yahoo.fr> writes: > + Therefore, "size" bytes of actual data have to be moved from the offset > + they were at in the old block (temp + old_padding) to the offset they > + have to be in the new block (temp + new_padding == memblock). > +*/ > if (new_padding != old_padding) > - memmove((char *)memblock + old_padding, (char *)memblock + new_padding, size); > + memmove((char *)memblock, (char *)temp + old_padding, size); You also have to take into account the fact that the size may have changed. -- Alexandre Julliard julliard(a)winehq.org

1 0

Reading Registry Example
by James McKenzie 27 Nov '07

27 Nov '07

All: I would like to work on the eventlog.c and would like pointers to an example of reading a registry example under Wine. Thank you. James

2 1

Re: USER32: win16 applications should not see long filenames returned from dialog control (try 4) (RESEND)
by Alexandre Julliard 27 Nov '07

27 Nov '07

Alex Villacís Lasso <a_villacis(a)palosanto.com> writes: > Changelog: > * Win16 applications should not see long filenames. Change directory > listbox to return short names when called from the 16-bit wrappers. You are still calling GetShortPathName on potentially uninitialized data. You simply can't do this with a wrapper around the 32-bit DlgDirSelect, you have to duplicate the code. But first it would be worth checking if the listbox shouldn't rather be populated with short path names in the first place. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: David Adam : d3dx8: Implement D3DXQuaternionInverse.
by Robert Shearman 27 Nov '07

27 Nov '07

Alexandre Julliard wrote: > /*_________________D3DXQUATERNION________________*/ > > -D3DXQUATERNION* WINAPI D3DXQuaternionMultiply(D3DXQUATERNION * pout, CONST D3DXQUATERNION *pq1, CONST D3DXQUATERNION * pq2) > +D3DXQUATERNION* WINAPI D3DXQuaternionInverse(D3DXQUATERNION *pout, CONST D3DXQUATERNION *pq) > +{ > + D3DXQUATERNION temp; > + FLOAT norm; > + > + norm = D3DXQuaternionLengthSq(pq); > + if ( !norm ) > + { > + pout->x = 0.0f; > + pout->y = 0.0f; > + pout->z = 0.0f; > + pout->w = 0.0f; > + } > + else > + { > + D3DXQuaternionConjugate(&temp, pq); > + pout->x = temp.x / norm; > + pout->y = temp.y / norm; > + pout->z = temp.z / norm; > + pout->w = temp.w / norm; > + } > + return pout; > +} > + > +D3DXQUATERNION* WINAPI D3DXQuaternionMultiply(D3DXQUATERNION *pout, CONST D3DXQUATERNION *pq1, CONST D3DXQUATERNION *pq2) > { > pout->x = pq2->w * pq1->x + pq2->x * pq1->w + pq2->y * pq1->z - pq2->z * pq1->y; > pout->y = pq2->w * pq1->y - pq2->x * pq1->z + pq2->y * pq1->w + pq2->z * pq1->x; This change causes the following warnings for me: math.c: In function ‘D3DXQuaternionInverse’: math.c:695: warning: ‘temp.x’ may be used uninitialised in this function math.c:695: warning: ‘temp.y’ may be used uninitialised in this function math.c:695: warning: ‘temp.z’ may be used uninitialised in this function math.c:695: warning: ‘temp.w’ may be used uninitialised in this function $ gcc --version gcc (GCC) 4.0.3 (Ubuntu 4.0.3-1ubuntu5) Copyright (C) 2006 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. I believe that this is causes by D3DXQuaternionConjugate not writing to temp in one case (pq being NULL) and the compiler not recognising that this case is already covered by the !norm branch above, since D3DXQuaternionLengthSq will return 0.0f if pq is NULL. The attached patch works around the warning. However, there are currently no tests for the case of pq being NULL so I don't know whether it is correct. -- Rob Shearman diff --git a/dlls/d3dx8/math.c b/dlls/d3dx8/math.c index 04072bf..8edd1d3 100644 --- a/dlls/d3dx8/math.c +++ b/dlls/d3dx8/math.c @@ -696,7 +696,7 @@ D3DXQUATERNION* WINAPI D3DXQuaternionInverse(D3DXQUATERNION *pout, CONST D3DXQUA FLOAT norm; norm = D3DXQuaternionLengthSq(pq); - if ( !norm ) + if ( !norm || !pq ) { pout->x = 0.0f; pout->y = 0.0f;

2 1

Re: [dbghelp 1/11] Update some mscvpdb.h definitions and descriptions
by Alexandre Julliard 27 Nov '07

27 Nov '07

"Jason Green" <jave27(a)gmail.com> writes: > The Original Author for the following set of patches is Eric van > Beurden from TransGaming, Inc. When submitting patches written by someone else please include a valid From: line in the patch header with his full name and email. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [dbghelp 2/11] Return TRUE in SymInitializeW if process is already initialized and add a number of TRACEs
by Alexandre Julliard 27 Nov '07

27 Nov '07

"Jason Green" <jave27(a)gmail.com> writes: > @@ -216,11 +216,17 @@ static BOOL WINAPI process_invade_cb(PCSTR name, ULONG base, ULONG size, PVOID u > char tmp[MAX_PATH]; > HANDLE hProcess = (HANDLE)user; > > - if (!GetModuleFileNameExA(hProcess, (HMODULE)base, > - tmp, sizeof(tmp))) > + > + TRACE("found module %s {base = 0x%08x, size = 0x%08x, user = %p}\n", debugstr_a(name), base, size, user); > + > + if (!GetModuleFileNameExA(hProcess, (HMODULE)base, tmp, sizeof(tmp))){ > + WARN("could not find the full path name of the module. Using the base name of %s instead\n", debugstr_a(name)); > lstrcpynA(tmp, name, sizeof(tmp)); > + } > > + TRACE("loading the module %s {base = 0x%08x}\n", debugstr_a(name), base); > SymLoadModule(hProcess, 0, tmp, name, base, size); > + TRACE("done loading the module\n"); > return TRUE; You are adding way too many traces, most of which are not necessary. Here for instance, SymLoadModule already contains traces, so there's no need to add traces around the call. It's fine to do that sort of thing to help following things when you are trying to understand the code, but afterwards they should be removed. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: d3dx8: Implementation of D3DXAssembleShaderFromFileA
by Dmitry Timoshkov 27 Nov '07

27 Nov '07

<luis.busquets(a)ilidium.com> wrote: > else eliminated Did you actually compile it? If you did, did you see a warning you introduced? -- Dmitry.

1 0

modified: dlls/d3dx8/d3dx8_main.c modified: include/d3dx8core.h
by luis.busquets＠ilidium.com 27 Nov '07

27 Nov '07

else eliminated. --- dlls/d3dx8/d3dx8_main.c | 18 +++++++++++++----- include/d3dx8core.h | 2 +- 2 files changed, 14 insertions(+), 6 deletions(-)

2 1

Re: ole32: Add tests for WdtpInterfacePointer_* functions.
by Juan Lang 26 Nov '07

26 Nov '07

Hi Rob, + ok(*(DWORD *)wireip == 0x44, "wireip + 0x0 should be 0x4c instead of 0x%08x\n", *(DWORD *)wireip); + wireip += sizeof(DWORD); + ok(*(DWORD *)wireip == 0x44, "wireip + 0x8 should be 0x4c instead of 0x%08x\n", *(DWORD *)wireip); Shouldn't these be "should be 0x44 instead of..."? --Juan

1 0

Re: [PATCH 3/3] advapi32: Add support for using the Mac Keychain services as a backend for the credential functions instead of the registry.
by Juan Lang 26 Nov '07

26 Nov '07

Hi Rob, @@ -232,6 +239,198 @@ static DWORD registry_read_credential(HKEY hkey, PCREDENTIALW credential, return ret; } +static DWORD mac_read_credential_from_item(SecKeychainItemRef item, BOOL require_password, + PCREDENTIALW credential, char *buffer, + DWORD *len) You're missing compilation protection here (and elsewhere) on non-Apple systems. --Juan

2 1

1.0 bugs
by Dan Kegel 26 Nov '07

26 Nov '07

[putting on tastemaster hat] So far, 82 bugs have been nominated to be fixed for 1.0. I checked a bunch of them, and so far most look reasonable. Some even look easy or have patches, e.g. http://bugs.winehq.org/show_bug.cgi?id=5402 http://bugs.winehq.org/show_bug.cgi?id=5884 http://bugs.winehq.org/show_bug.cgi?id=7544 http://bugs.winehq.org/show_bug.cgi?id=7571 http://bugs.winehq.org/show_bug.cgi?id=9104 http://bugs.winehq.org/show_bug.cgi?id=10147 The list looks meaty enough that it's probably worth folks going through and seeing what they can pick off. Stuff that's silly or too hard - just skip it for now, we'll defer those for later. - Dan

1 1

Re: d3dx8: Implementation of D3DXAssembleShaderFromFileA using D3DXAssembleShaderFromFileW
by Stefan Dösinger 26 Nov '07

26 Nov '07

Am Montag, 26. November 2007 11:37:28 schrieb luis.busquets(a)ilidium.com: > --- > dlls/d3dx8/d3dx8_main.c | 22 ++++++++++++++++------ > include/d3dx8core.h | 4 ++-- > 2 files changed, 18 insertions(+), 8 deletions(-) Please watch out with indentions. You are mixing tabs and spaces in D3DXAssembleShaderFromFileA, so the code looks a bit dissorted here.

1 0

Re: Re: urlmon: Wrap heap functions.
by Lionel Debroux 26 Nov '07

26 Nov '07

Hi, > In Wine HeapAlloc gets wrapped in different places, each using a > different name and even a different naming convention. This makes the > code harder to understand for people that browse the code (people doing > janitorial work) as they have to check every time if the function is yet > another wrapper around HeapFree or if the function is doing more than > that. And this makes it even harder for static code analyzing tools. I second Michael on that. A number of spurious entries previously in his list of potential issues detected by Smatch ( http://people.redhat.com/mstefani/wine/smatch/bugs_all.html ) were due to the wrappers around HeapFree that the script wasn't yet aware of. Currently, the list of functions that allocate memory, in the unfree_wine script, contains 33 entries. The list of functions that free memory contains 34 entries. > I therefore propose to standardize all the internal HeapAlloc wrappers > to a common name and form, Something along the lines of: > walloc() > wzalloc() > wrealloc() > wzrealloc() > wfree() > It's short and concise and doesn't conflict with the Win32 API. I like that solution. Bye, Lionel. ____________________________________________________________________________________________ Yahoo! Mail innove : interface hyper pratique, messenger intégré, couleurs - http://mail.yahoo.fr

1 0

Re: widl [1/7]: Output working exception handling macros. [take 2]
by Alexandre Julliard 26 Nov '07

26 Nov '07

Dan Hipschman <dsh(a)linux.ucla.edu> writes: > This is essentially the same as the last attempt but doesn't add a wine- > internal function to kernel32. It just puts the code directly in the > widl-generated exception handler. An alternative is to handle these access- > based page faults for which we can continue in the signal handler (signal_i386, > etc., in ntdll) so they never actually result in an exception being raised. > I'm not sure if anything depends on the current behavior and didn't want to > mess much more with the exception handling code than I had to. You shouldn't have to do anything of the kind, there's no reason that the generated code has to duplicate the unhandled exception filter. What you have to do is to filter properly and not catch exceptions you are not interested in. For cases where Windows would catch the access violation exception, then the unhandled filter won't be called on Windows either, that's the way it's supposed to work. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: d3dx8: Implementation of D3DXAssembleShaderFromFileA using D3DXAssembleShaderFromFileW
by Dmitry Timoshkov 26 Nov '07

26 Nov '07

<luis.busquets(a)ilidium.com> wrote: > + LPWSTR pSrcFileW = NULL; > + DWORD len; > + HRESULT ret; > + if (pSrcFile) > + { > + len = MultiByteToWideChar( CP_ACP, 0, pSrcFile, -1, NULL, 0 ); > + pSrcFileW = HeapAlloc( GetProcessHeap(), 0, len * sizeof(WCHAR) ); > + MultiByteToWideChar( CP_ACP, 0, pSrcFile, -1, pSrcFileW, len ); > + ret=D3DXAssembleShaderFromFileW(pSrcFileW, Flags, ppConstants, ppCompiledShader, ppCompilationErrors); > + HeapFree( GetProcessHeap(), 0, pSrcFileW ); > + return ret; > + } > + else return D3DXERR_INVALIDDATA; > } Usually it's much more readable and helps to avoid not necessary indentation to check parameters at the start of the function and return right away with appropriate error code. Besides you are mixing tabs and spaces. > -HRESULT WINAPI D3DXAssembleShaderFromFileW(LPSTR pSrcFile, DWORD Flags, > +HRESULT WINAPI D3DXAssembleShaderFromFileW(LPWSTR pSrcFile, DWORD Flags, You marked pSrcFile parameter 'const' in D3DXAssembleShaderFromFileA declaration but not in D3DXAssembleShaderFromFileW, any reason for that? -- Dmitry.

1 0

Fontforge and building wine
by Dan Kegel 26 Nov '07

26 Nov '07

It's been mentioned many times, but building Wine without a recent fontforge yields a wine that doesn't work too well. Tonight, for instance, I built wine on an updated dapper system, and forgot; here are the commands I had to use to recover (just in case somebody needs to do this again). The fontforge I chose just happens to be the latest; the --without-freetype-bytecode option is there to keep configure from searching the whole freaking hard drive for the nonexistent freetype source tree. tar -xjvf fontforge_full-20071110.tar.bz2 cd fontforge-20071110/ ./configure --prefix=/usr/local/fontforge --without-freetype-bytecode make sudo make install cd ~/wine-git cd fonts make clean PATH=/usr/local/fontforge/bin:$PATH make - Dan

1 0

Re: shdocvw: Silence common invalid QueryInterface FIXMEs.
by Alistair Leslie-Hughes 25 Nov '07

25 Nov '07

> hres = IUnknown_QueryInterface(unk, &IID_IRunnableObject, > (void**)&runnable); > ok(hres == E_NOINTERFACE, "QueryInterface returned %08x, expected > E_NOINTERFACE\n", hres); > ok(runnable == NULL, "runnable=%p, ezpected NULL\n", runnable); > + > + hres = IUnknown_QueryInterface(unk, &IID_IPerPropertyBrowsing, > (void**)&propbrowse); > + ok(hres == E_NOINTERFACE, "QueryInterface returned %08x, expected > E_NOINTERFACE\n", hres); > + ok(runnable == NULL, "runnable=%p, ezpected NULL\n", runnable); > + > + hres = IUnknown_QueryInterface(unk, &IID_IOleCache, (void**)&cache); > + ok(hres == E_NOINTERFACE, "QueryInterface returned %08x, expected > E_NOINTERFACE\n", hres); > + ok(cache == NULL, "runnable=%p, ezpected NULL\n", runnable); > } Hi Jacek, The tests seem wrong. Is there a reson your printing runnable twice, and not propbrowse and cache. Best Regards Alistair Leslie-Hughes

1 0

Re: bug fix: Prevented strlen(NULL) in PRINTDLG_CreateDevNames()
by Stefan Dösinger 25 Nov '07

25 Nov '07

Am Sonntag, 25. November 2007 12:35:36 schrieb Jorma Laaksonen: > Hello! > > wine-0.9.49 (as well as many earlier versions) seems to crash when > PRINTDLG_CreateDevNames() is for some reason called with DeviceDriverName > equal to NULL. The real cause of this problem might be somewhere else > where the NULL is assigned, but this simple fix prevents NULL being > passed to strlen(). If you suspect that the real problem is somewhere else, it is recommended to write a test case to see how Windows deals with this situation. Windows could return an error if the device driver name is NULL. If Windows crashes as well, your patch would only hide a bug instead of fixing it.

1 0

Re: [Bug 10488] GIT of 16.11 Exeption becaus of VBA while analysing a text
by Zac Brown 25 Nov '07

25 Nov '07

wine-bugs(a)winehq.org wrote: > http://bugs.winehq.org/show_bug.cgi?id=10488 > > > > > > --- Comment #14 from Yolande Haneder <yolande(a)haneder.biz> 2007-11-25 00:30:20 --- > Re: > > err:ole:StdMarshalImpl_ReleaseMarshalData could not map object ID to stub > manager, oxid=3700000038, oid=d > err:ole:CoReleaseMarshalData IMarshal::ReleaseMarshalData failed with error > 0x8001011d > > Which I have been told to be a memory conflict in C > > and > > err:ole:apartment_getclassobject DllGetClassObject returned error 0x80040111 > > Could someone tell me which file in Wine is concerned so that someone my look > at it? > > Thank you > > Looks like dlls/ole32/marshal.c by the dump. I believe Dan Kegel had reported a bug for Wine 1.0 based on the dlls/ole32/tests/marshal.c in which it failed. The relevant bug is as follows: http://bugs.winehq.com/show_bug.cgi?id=10076 -Zac

1 1

Asking Valve for Steam Survey Results
by Scott Ritchie 24 Nov '07

24 Nov '07

Valve regularly runs surveys of steam users, where they profile many different things about the hardware. They're nice enough to publish the results: http://www.steampowered.com/status/survey.html Some of these provide a way to detect Wine users. Everyone using the "Wine waveout" audio driver, for instance, must be a Wine user. Unfortunately, valve lumps all of these into "other" on the survey results page. Worries about sampling bias aside (maybe Wine users are less likely to do the steam survey), if we had the raw data from Valve we could determine the actual percentage of Steam users playing through Wine. Looking at the data we do have, we may be able to make a good guess from examining the video card driver name field. Excluding NVidia, ATI, and Intel leaves just over 4% of respondents using an "other" video driver like the Wine one. If even only a quarter of these "others" are using Wine, that still gives us 1% of the entire 10 million+ Steam user base, translating into hundreds of thousands of users. Does anyone know whom we can contact at Valve for more specific results? Thanks, Scott Ritchie

5 6

AppDB and Bugzilla don't currently work...
by Lionel_Debroux 24 Nov '07

24 Nov '07

Hi, Both AppDB and Bugzilla currently complain about too many connections to the server: (AppDB) Warning: mysql_connect() [function.mysql-connect]: Too many connections in /home/winehq/opt/appdb/include/query.php on line 82 Warning: mysql_select_db(): supplied argument is not a valid MySQL-Link resource in /home/winehq/opt/appdb/include/query.php on line 83 Warning: mysql_real_escape_string() [function.mysql-real-escape-string]: Too many connections in /home/winehq/opt/appdb/include/query.php on line 116 Warning: mysql_real_escape_string() [function.mysql-real-escape-string]: A link to the server could not be established in /home/winehq/opt/appdb/include/query.php on line 116 Warning: mysql_connect() [function.mysql-connect]: Too many connections in /home/winehq/opt/appdb/include/query.php on line 14 Database error, please try again soon. (Bugzilla) Software error: Can't connect to the database. Error: Too many connections Is your database installed and up and running? Do you have the correct username and password selected in localconfig? For help, please send mail to the webmaster (bugs-admin(a)winehq.org), giving this error message and the time and date of the error. Regards, Lionel Debroux.

2 1

xlviewer install cmdline passed from CreateProcessA to msiexec looks wrong
by Darragh Bailey 24 Nov '07

24 Nov '07

OK, long time lurker, only recently decided to try my hand at coding for wine. Picked a bug 9628 and decided to have a bash. Bug occured because a patch to the msiexec command line parsing was reverted due to regressions in the xlviewer installer and other apps. I don't know about the other apps, I can't find a corresponding bug and only gleaned this bit of info from the changelog and in the cvs mailing list that tracks commits. http://www.winehq.org/pipermail/wine-cvs/2007-March/031339.html After some digging I found that problem was that with current git the command line being sent to msiexec when installing "RPG Maker" was missing the msiexec executable. So the msiexec command line parsing would result in "/i" as being the first element of the arguments array to be processed. The main loop, expected that msiexec should always be argv[0], and so will always start from the second element. Then comparing of what used to happen when msiexec used the call to CommandLineToArgvW suggested that use of CommandLineToArgvW was more correct given the main loop design. If the program name is absent as the first argument, msiexec would just get back an empty string as the first element and the arguments to msiexec would start at the second element. Since this appears more correct, I looked at the behaviour of xlviewer (free download from microsoft) installer, to see what the differences were. I had to add some extra traces to the start of the main function in msiexec to get an idea of what was going on. Basically they would just dump out the contents of argv, one per line, so that I could see what it was getting as the command line. Added the following to the top of the main function: for( i=0; i<argc; i++) { WINE_TRACE("bug 9628: argv[%d] = \"%s\"\n", i, argv[i]); } using the newly build wine tree, and just pointing wine at the exe. The command line msiexec would get on installing xlviewer was trace:msiexec:main bug 9628: argv[0] = "c:\windows\system32\msiexec.exe" trace:msiexec:main bug 9628: argv[1] = "/I" trace:msiexec:main bug 9628: argv[2] = "C:\windows\temp\IXP001.TMP\XLVIEW.MSI" trace:msiexec:main bug 9628: argv[3] = "CDCACHE=2" trace:msiexec:main bug 9628: argv[4] = "LAUNCHEDFROMSETUP=1" trace:msiexec:main bug 9628: argv[5] = "SETUPEXEPATH=C:\windows\temp\IXP001.TMP" SETUPEXENAME=SETUP.EXE /lpiwaeo C:\windows\temp\Microsoft" trace:msiexec:main bug 9628: argv[6] = "Office" trace:msiexec:main bug 9628: argv[7] = "Excel" trace:msiexec:main bug 9628: argv[8] = "Viewer" trace:msiexec:main bug 9628: argv[9] = "2003" trace:msiexec:main bug 9628: argv[10] = "Setup(0004)_Task(0001).txt CDCACHE=0 DWSETUPLOGFILE=C:\windows\temp\Microsoft" trace:msiexec:main bug 9628: argv[11] = "Office" trace:msiexec:main bug 9628: argv[12] = "Excel" trace:msiexec:main bug 9628: argv[13] = "Viewer" trace:msiexec:main bug 9628: argv[14] = "2003" trace:msiexec:main bug 9628: argv[15] = "Setup(0004).txt DWMSILOGFILE=C:\windows\temp\Microsoft" trace:msiexec:main bug 9628: argv[16] = "Office" trace:msiexec:main bug 9628: argv[17] = "Excel" trace:msiexec:main bug 9628: argv[18] = "Viewer" trace:msiexec:main bug 9628: argv[19] = "2003" trace:msiexec:main bug 9628: argv[20] = "Setup(0004)_Task(0001).txt" Now this looks wrong to me. It looks like there is a bug in how the command line is being passed to msiexec. The existing code that parses this command line, is in fact working around this bug. Adding +process to the WINEDEBUG variable I got to see what CreateProcessA was calling msiexec with[1]: L"\"c:\\windows\\system32\\msiexec.exe\" /I C:\\windows\\temp\\IXP008.TMP\\XLVIEW.MSI CDCACHE=\"2\" LAUNCHEDFROMSETUP=\"1\" SETUPEXEPATH=\"C:\\windows\\temp\\IXP008.TMP\\\" SETUPEXENAME=\"SETUP.EXE\" /lpiwaeo \"C:\\windows\\temp\\Microsoft Office Excel Viewer 2003 Setup(0014)_Task(0001).txt\" CDCACHE=\"0\" DWSETUPLOGFILE=\"C:\\windows\\temp\\Microsoft Office Excel Viewer 2003 Setup(0014).txt\" DWMSILOGFILE=\"C:\\windows\\temp\\Microsoft Office Excel Viewer 2003 Setup(0014)_Task(0001).txt\"" 1. I also have to increase the debug str length in debug.c otherwise the end of the string would get dropped. Now when I run msiexec with the following command line msiexec /I C:\\windows\\temp\\IXP008.TMP\\XLVIEW.MSI CDCACHE="2" LAUNCHEDFROMSETUP="1" SETUPEXEPATH="C:\\windows\\temp\\IXP008.TMP\\" SETUPEXENAME="SETUP.EXE" /lpiwaeo "C:\\windows\\temp\\Microsoft Office Excel Viewer 2003 Setup(0014)_Task(0001).txt" CDCACHE="0" DWSETUPLOGFILE="C:\\windows\\temp\\Microsoft Office Excel Viewer 2003 Setup(0014).txt" DWMSILOGFILE="C:\\windows\\temp\\Microsoft Office Excel Viewer 2003 Setup(0014)_Task(0001).txt" I see the following appear in the traces from msiexec: trace:msiexec:main bug 9628: argv[0] = "programs/msiexec/msiexec.exe.so" trace:msiexec:main bug 9628: argv[1] = "/I" trace:msiexec:main bug 9628: argv[2] = "C:\windows\temp\IXP002.TMP\XLVIEW.MSI" trace:msiexec:main bug 9628: argv[3] = "CDCACHE=2" trace:msiexec:main bug 9628: argv[4] = "LAUNCHEDFROMSETUP=1" trace:msiexec:main bug 9628: argv[5] = "SETUPEXEPATH=C:\windows\temp\IXP002.TMP\" trace:msiexec:main bug 9628: argv[6] = "SETUPEXENAME=SETUP.EXE" trace:msiexec:main bug 9628: argv[7] = "/lpiwaeo" trace:msiexec:main bug 9628: argv[8] = "C:\windows\temp\Microsoft Office Excel Viewer 2003 Setup(0003)_Task(0001).txt" trace:msiexec:main bug 9628: argv[9] = "CDCACHE=0" trace:msiexec:main bug 9628: argv[10] = "DWSETUPLOGFILE=C:\windows\temp\Microsoft Office Excel Viewer 2003 Setup(0003).txt" trace:msiexec:main bug 9628: argv[11] = "DWMSILOGFILE=C:\windows\temp\Microsoft Office Excel Viewer 2003 Setup(0003)_Task(0001).txt" Which seems bang on with what msiexec should be getting as its commandline arguments. In fact the only way I could pass in the command line to msiexec and get the same output near the top was with the following: msiexec /I C:\\windows\\temp\\IXP002.TMP\\XLVIEW.MSI CDCACHE="2" LAUNCHEDFROMSETUP="1" SETUPEXEPATH="C:\\windows\\temp\\IXP002.TMP\" SETUPEXENAME="SETUP.EXE" /lpiwaeo "C:\\windows\\temp\\Microsoft Office Excel Viewer 2003 Setup$0003$_Task$0001$.txt" CDCACHE="0" DWSETUPLOGFILE="C:\\windows\\temp\\Microsoft Office Excel Viewer 2003 Setup$0003$.txt" DWMSILOGFILE="C:\\windows\\temp\\Microsoft Office Excel Viewer 2003 Setup$0003$_Task$0001$.txt"" Notice the difference between the following: SETUPEXEPATH="C:\\windows\\temp\\IXP008.TMP\\" = correct command line SETUPEXEPATH="C:\\windows\\temp\\IXP002.TMP\" = broken It seems to me that current command line parsing in msiexec is a hack to work around a bug somewhere between CreateProcessA and where the executable actually gets run, when you have a directory path ending in a slash and also enclosed by quotes. It appears that one of the double slashes at the end gets substituted before its time. So basically, I'm wondering if I'm actually on the right track here. Is this a bug in the behaviour of how the command line is sent to the new process? I've yet to log a bug since I'm still trying to understand what exactly is happening and if this is in fact a bug. At the moment I trying to work out how to get more information on what's happening to the commandline in the create_process function, so I've yet to work out why the command line is altered by the time it reaches the msiexec program. Any suggestions are welcome. :) -- Darragh "Nothing is foolproof to a sufficiently talented fool."

1 0

Re: WineD3D: Implement detection of ATI cards with Mesa DRI drivers [try 4]
by Roderick Colenbrander 23 Nov '07

23 Nov '07

> This is an updated version of my previous patch for the same purpose. > > The proposed code now uses the WINE_D3DX_CAPABLE macros instead of DRI > module names for guessing the exact card model which is to be returned. > The patch also no longer includes manual addition of an entry to the > ChangeLog. Thanks to Roderick Colenbrander and Stefan Dösinger for the > suggestions. I have tested and confirmed that this patch fixes the bug > #7267 (http://bugs.winehq.org/show_bug.cgi?id=7267 ) to the same extent > as the previous version of the patch did. I think the code should be like this (in pseudo code) this way also other cards will at least be classified: case VENDOR_MESA: if(D3D9_CAPABLE(gl_info) if(strstr(gl_info->gl_renderer, "R300")) card = radeon_9500; else card = generic_nv_card; if(D3D8_CAPABLE(...) if(strstr(gl_info->gl_renderer, "R200")) card = ... 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

2 3

re: Start: Convert start to Unicode. [try7]
by Dan Kegel 23 Nov '07

23 Nov '07

Anatoly, thank you for working to fix "start" to work better for Unicode, I should have done that originally. But you seem to be having trouble getting the patch in, so here's what little advice I can give: If there is any difference between versions of your patch, you should describe it in the patch description. If there isn't, you should call it a resend, rather than a new try. - Dan

2 1

Driver-supported DirectX
by Remco 23 Nov '07

23 Nov '07

I don't know the first thing about driver- and DirectX-programming, so please forgive (and point out) any mistakes. As a reader of this list I'm wondering; there are quite a few problems that come from the fact that DirectX isn't 1:1 translatable to OpenGL. How about talking to some guys from the GPU-driver department about creating a driver-interface that gives you the right hooks. I guess Parallels and Transgaming would also be interested in such a development. I can imagine that the Nouveau-devs and Xorg-radeon-devs would be more than happy to listen. This goes beyond the scope of the Wine project I think. But since Wine has the higher level part of DirectX documented and implemented on top of OpenGL, wouldn't this be the place to start an independent library? Codeweavers has a lot of knowledge about Windows, DirectX and Linux. Only Microsoft itself would be a better choice, but I don't think they really care that much about Linux. ;) This would mean that DirectX would be as native to Linux and OSX (and friends) as it would be for Windows. It would be an actual reliable platform that could be used by game developers. It would de-Windows-ize DirectX. Maybe NVIDIA, ATI and Intel would also be interested. They could sell their expensive next-gen cards to those 5% that don't run Windows if games would actually be released for non-Windows OSes. Or are there really compelling technical reasons to wrap around OpenGL? I can think of the Compiz-issue. Similarly, Microsoft stated that they have to wrap OpenGL around DirectX on Windows, to be able to use both OpenGL and DirectX at the same time (for Aero). But I suspect that this implementation just developed naturally because messing with the drivers would be unthinkable way back when. Remco ____________________________________________________________________________________ Get easy, one-click access to your favorites. Make Yahoo! your homepage. http://www.yahoo.com/r/hs

5 6

Re: Start: Convert start to Unicode. [try5]
by Anatoly Lyutin 23 Nov '07

23 Nov '07

Alexandre Julliard wrote: > No, what you need is one function for errors that can happen in normal > usage (file not found, invalid option etc.) that is Unicode and > localized, and one function for internal errors (LoadString failures > etc.) that simply uses WINE_ERR with an English message. > > I send new version according your requirements to patches-lists. Have I made all correctly? -- Best regards Anatoly Lyutin.

1 0

Re: WineD3D: Implement detection of ATI cards with Mesa DRI drivers [try 3]
by Roderick Colenbrander 23 Nov '07

23 Nov '07

> Presently, if an ATI card uses open-source DRI drivers from the Mesa > project ( http://dri.freedesktop.org/wiki/ATI ), Wine will report it to > Windows applications as being some generic nVIDIA model, based on D3D > capabilities of the card. > > This patch adds detection of the used ATI video card, based on Mesa DRI > module name, which is returned by drivers in GL Renderer string. In my > opinion, returning even an approximate ATI card model should be better, > than the present condition, disguising the ATI card as some equally > generic model from nVIDIA. Also, the patch adds detection of VENDOR_MESA > >from GL Renderer string, as Mesa DRI drivers may have various things as > GL Vendor, but the GL Renderer name always contains the string "Mesa". > This patch resolves the bug #7267, "Lineage 2 complains about outdated > NVIDIA drivers using ATI card with Mesa drivers" - > http://bugs.winehq.org/show_bug.cgi?id=7267 . The idea behind the whole code is that we detect a 'compatible' card based on the OpenGL capabilites. That stuff is done using the D3DX_CAPABLE() macros. The extension I check directly correspond to D3D functionality. In the end the result is further refind by checking the opengl renderer string. The reason why it is all done this way is because we need to detect the card based on OpenGL / GLX information and they don't export the pci id (which is good) but bad for D3D. Retrieving the real card name and so on from /proc or whatever isn't guaranteed to get good results and it also doesn't scale with remote X and if the system uses multiple videocards. It is very tricky. I would really ask you to use the D3DX_CAPABLE() macros as not using them can result in bad problems. E.g. a game assuming a certain d3d functionality is there while it isn't backed by any GL extension. The reason we default to Nvidia cards is that they are very generic and have quite solid drivers, so from an application point of view that is nice. Second games don't really need the exact GPU but they just use the info to get an impression of the features / performance. You can debate whether reporting Nvidia or ATI on totally different drivers matters (I would say no). So at least use the D3DX_CAPABLE macros as those are correct. Roderick -- Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen! Ideal für Modem und ISDN: http://www.gmx.net/de/go/smartsurfer

2 2

Re: [msvcrt] short tty reads don't imply EOF
by Alexandre Julliard 23 Nov '07

23 Nov '07

"Damjan Jovanovic" <damjan.jov(a)gmail.com> writes: > Changelog: > * For ttys, short reads don't mean we're at the end of the file. That's not limited to ttys, other file types can have short reads too. -- Alexandre Julliard julliard(a)winehq.org

2 1

Ansi to Unicode
by luis.busquets＠ilidium.com 22 Nov '07

22 Nov '07

Is there any standard function in WINE to transform an Ansi string into a Unicode string? And the opposite?

3 2

Re: ntdll: Shared manifests should have a less-strict version check performed when loading them as dependencies.
by Alexandre Julliard 22 Nov '07

22 Nov '07

Robert Shearman <rob(a)codeweavers.com> writes: > --- > dlls/ntdll/actctx.c | 56 > +++++++++++++++++++++++++++++++------------------- > 1 files changed, 35 insertions(+), 21 deletions(-) Some test cases would be nice... -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: wineprefixcreate: get xdg-user-dirs environment variables [1/2]
by Alexandre Julliard 22 Nov '07

22 Nov '07

"Lei Zhang" <thestig(a)google.com> writes: > We need to get the "well known" XDG user directories so we can symlink to them. This should be done inside shell32, not in wineprefixcreate. -- Alexandre Julliard julliard(a)winehq.org

3 2

Re: dlls/winedos/int21.c -- type adjustments
by Dmitry Timoshkov 22 Nov '07

22 Nov '07

"Gerald Pfeifer" <gerald(a)pfeifer.com> wrote: > ChangeLog: > Use unsigned long instead of long for return values of SetFilePointer. It wouldn't take more than 30 additional seconds to look up the SetFilePointer prototype in include/ and use real type instead. -- Dmitry.

2 1

Disregard earlier ntdll/tests/env.c patches
by Zac Brown 22 Nov '07

22 Nov '07

Hi, Please disregard any patches prior to the latest patch. It seems my mail server has been a little messed up. The only one to consider is the latest one that the wine-patches list has received. Thank you, Zac Brown -- Zac Brown (817-266-6867) zac(a)zacbrown.org http://zacbrown.org

1 0

Re: d3dx8 [patch 7/7] Implement D3DXMatrixTransformation
by Ivan Gyurdiev 22 Nov '07

22 Nov '07

Any chance the formatting can be improved to include whitespace between blocks and proper indentation?

2 1

re: ntdll: fix uninitialized buffers, found via valgrind
by Dan Kegel 22 Nov '07

22 Nov '07

James wrote: > Why did you move the variable declaration? Because the variable is only used inside the loop. Moving the declaration to the top of the loop is the right scoping. - Dan

1 0

Re: Use GCC's -Wtype-limits
by Alexandre Julliard 22 Nov '07

22 Nov '07

Gerald Pfeifer <gerald(a)pfeifer.com> writes: > If applied, I will commit to address, directly and by engaging others, > the occurrences of such warnings in Wine. I'm not opposed to applying it, but you have to fix the warnings first, as far as possible the default build should not trigger any warnings. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: ntdll: fix uninitialized buffers, found via valgrind
by James Hawkins 22 Nov '07

22 Nov '07

On Nov 21, 2007 6:08 PM, Zac Brown <zac(a)zacbrown.org> wrote: > Hi, > > Fixes valgrind error: > http://kegel.com/wine/valgrind/20071121/vg-ntdll_env.txt > > Changelog: > * Initialize buffers used in test using memset > > -- > Zac Brown <zac AT zacbrown DOT org> > > > dlls/ntdll/tests/env.c | 7 +++++-- > 1 files changed, 5 insertions(+), 2 deletions(-) > > diff --git a/dlls/ntdll/tests/env.c b/dlls/ntdll/tests/env.c > index d822946..90a3bf1 100644 > --- a/dlls/ntdll/tests/env.c > +++ b/dlls/ntdll/tests/env.c > @@ -80,8 +80,7 @@ static void testQuery(void) > {NULL, 0, 0, NULL} > }; > > - WCHAR bn[257]; > - WCHAR bv[257]; > + > UNICODE_STRING name; > UNICODE_STRING value; > const struct test* test; > @@ -89,6 +88,10 @@ static void testQuery(void) > > for (test = tests; test->var; test++) > { > + WCHAR bn[257]; > + WCHAR bv[257]; > + memset(bn, '\0', 257 * sizeof(WCHAR)); > + memset(bv, '\0', 257 * sizeof(WCHAR)); > name.Length = strlen(test->var) * 2; > name.MaximumLength = name.Length + 2; > name.Buffer = bn; > Why did you move the variable declaration? Also, please don't use magic constants; sizeof(bn) *sizeof(WCHAR) is what you want. -- James Hawkins

2 2

Re: WineD3D: Implement detection of ATI cards with Mesa DRI drivers [try 3]
by Stefan Dösinger 22 Nov '07

22 Nov '07

Hi, > --- a/ChangeLog > +++ b/ChangeLog > @@ -1,3 +1,8 @@ > +2007-11-22 Roman Mamedov <roman(a)rm.pp.ru> > + > + * dlls/wined3d/directx.c: > + wined3d: Implement detection of ATI cards with Mesa DRI drivers. > + Sorry, I missed this in the first patch: You don't have to update the changelog file in the patches. The ChangeLog is autogenerated from the git commit logs(ie, the subject in the patch file). I am not sure if Alexandre's scripts filter that out or if he asks to resend the patch.

1 0

Re: WineD3D: Implement detection of ATI cards with Mesa DRI drivers
by Stefan Dösinger 21 Nov '07

21 Nov '07

Am Donnerstag, 22. November 2007 00:15:30 schrieb Roman Mamedov: > Hello. > > Presently, if an ATI card uses open-source DRI drivers from the Mesa > project ( http://dri.freedesktop.org/wiki/ATI ), Wine will report it to > Windows applications as being some generic nVIDIA model, based on D3D > capabilities of the card. The patch looks good to me content wise, but please attach it as an attachment to the mail instead of copypasting it into the mail. Also I am not sure if we should report R200 cards. They have pixel shader 1.x support on Windows, but Wine cannot implement this because the cards are too limited to support GL_ARB_fragment_program(~ ps 2.0), but there is no common extension that maps 1.x pixel shaders, only an ATI specific one. Any volunteers for a GL_ATI_fragment_shader shader backend :-) ? On the other hand, those cards support a few features that older cards did not have, like vertex shaders or volume textures. Also note that Alexandre is on holiday this week, so the patch won't be committed before next week.

1 0

Re: WineD3D: Implement detection of ATI cards with Mesa DRI drivers
by James Hawkins 21 Nov '07

21 Nov '07

On Nov 21, 2007 5:15 PM, Roman Mamedov <roman(a)rm.pp.ru> wrote: > Hello. > > Presently, if an ATI card uses open-source DRI drivers from the Mesa > project ( http://dri.freedesktop.org/wiki/ATI ), Wine will report it to > Windows applications as being some generic nVIDIA model, based on D3D > capabilities of the card. > > This patch adds detection of the used ATI video card, based on Mesa DRI > module name, which is returned by drivers in GL Renderer string. In my > opinion, returning even an approximate ATI card model should be better, > than the present condition, disguising the ATI card as some equally > generic model from nVIDIA. Also, the patch adds detection of VENDOR_MESA > from GL Renderer string, as Mesa DRI drivers may have various things as > GL Vendor, but the GL Renderer name always contains the string "Mesa". > This patch resolves the bug #7267, "Lineage 2 complains about outdated > NVIDIA drivers using ATI card with Mesa drivers" - > http://bugs.winehq.org/show_bug.cgi?id=7267 . > If you send in a fixed patch, you need to add some variation of 'try X' in the subject, where X is the number of tries you're on. On top of this, you need to say in the body what you've changed between submissions. -- James Hawkins

2 1

Re: WineD3D: Implement detection of ATI cards with Mesa DRI drivers
by James Hawkins 21 Nov '07

21 Nov '07

On Nov 21, 2007 4:51 PM, Roman Mamedov <roman(a)rm.pp.ru> wrote: > Sorry, a small mistake. Always check everything twice, and when you > checked twice - check the third time :( > > --- > dlls/wined3d/directx.c | 2 +- > 1 files changed, 1 insertions(+), 1 deletions(-) > > diff --git a/dlls/wined3d/directx.c b/dlls/wined3d/directx.c > index fa844b0..7e09014 100644 > --- a/dlls/wined3d/directx.c > +++ b/dlls/wined3d/directx.c > @@ -1145,7 +1145,7 @@ BOOL IWineD3DImpl_FillGLCaps(WineD3D_GL_Info > *gl_info) { > } else if (strstr(gl_info->gl_renderer, "R300")) { /* > r300_dri - all R3xx and R4xx-based cards */ > gl_info->gl_card = CARD_ATI_RADEON_9500; > vidmem = 128; > - } else if (strstr(gl_info->gl_renderer, "Radeon")) { /* > radeon_dri, for all R3xx-based cards */ > + } else if (strstr(gl_info->gl_renderer, "Radeon")) { /* > radeon_dri - all R100-based cards */ > /* This one should be the last case, so if the word > "Radeon" ever added to Mesa gl_renderer > * string of R200 or R300, this detection code will > keep working properly. > */ > -- > 1.5.3.4 > If you make a mistake, you have to fix the first patch and resend it. -- James Hawkins

1 0

No release this week
by Alexandre Julliard 21 Nov '07

21 Nov '07

Folks, Because of the Thanksgiving weekend, there won't be a release this Friday. 0.9.50 will most likely be on next Friday. -- Alexandre Julliard julliard(a)winehq.org

1 0

Removing Windows and Solaris from the Downloads page
by Scott Ritchie 21 Nov '07

21 Nov '07

It appears that the links for the Windows and Solaris versions on the winehq downloads page are very, very old. Since they're no longer maintained, we should just remove them, yes? Thanks, Scott Ritchie

7 9

Re: d3dx8 [patch 5/7] Implement D3DXQuaternionLn
by Andrew Talbot 21 Nov '07

21 Nov '07

David.Adam(a)math.cnrs.fr wrote: > Hi David, There was no code in this patch. -- Andy.

1 0

something wrong with patches?
by Kirill K. Smirnov 21 Nov '07

21 Nov '07

Hi, Is there anything wrong with the following patches? "winhelp: add basic support for ancient win3.0 help files" http://www.winehq.org/pipermail/wine-patches/2007-November/046886.html and comctl32: Implemented algorithm is a mergesort, not a quicksort http://www.winehq.org/pipermail/wine-patches/2007-November/046943.html Did I miss something important? Regards Kirill

1 0

Re: [6/9] d3d9: add a test for vertex shader input matching
by Alexandre Julliard 21 Nov '07

21 Nov '07

Stefan Dösinger <stefan(a)codeweavers.com> writes: > From 44f176ad4fcbc92ea66d8526670a1ec573ffa9f0 Mon Sep 17 00:00:00 2001 > From: Stefan Doesinger <stefan(a)codeweavers.com> > Date: Mon, 19 Nov 2007 10:53:26 +0100 > Subject: [PATCH] d3d9: add a test for vertex shader input matching This fails with the usual rounding errors: ../../../tools/runtest -q -P wine -M d3d9.dll -T ../../.. -p d3d9_test.exe.so visual.c && touch visual.ok visual.c:4716: Test failed: Input test: Quad 1(2crd) returned color 0x00ffff7f, expected 0x00FFFF80 visual.c:4721: Test failed: Input test: Quad 3(2crd-wrongidx) returned color 0x00ff007f, expected 0x00FF0080 visual.c:4704: Test failed: Input test: Quad 1(2crd) returned color 0x00ffff7f, expected 0x00FFFF80 visual.c:4710: Test failed: Input test: Quad 3(2crd-wrongidx) returned color 0x00ff007f, expected 0x00FF0080 visual.c:4704: Test failed: Input test: Quad 1(2crd) returned color 0x00ffff7f, expected 0x00FFFF80 visual.c:4710: Test failed: Input test: Quad 3(2crd-wrongidx) returned color 0x00ff007f, expected 0x00FF0080 visual.c:4849: Test failed: Input test: Quad 2(color-ubyte) returned color 0x00407ffe, expected 0x014080ff visual.c:4849: Test failed: Input test: Quad 2(color-ubyte) returned color 0x00407ffe, expected 0x014080ff visual.c:4849: Test failed: Input test: Quad 2(color-ubyte) returned color 0x00407ffe, expected 0x014080ff make[2]: *** [visual.ok] Error 9 -- Alexandre Julliard julliard(a)winehq.org

1 0

[0/3] RSA AES provider to rsaenh
by Vijay Kiran Kamuju 20 Nov '07

20 Nov '07

On Nov 20, 2007 4:23 PM, Juan Lang <juan.lang(a)gmail.com> wrote: > > The test fail due to missing implementation, Hence I put them in todo's. > > After the implementation is there, they do not fail. > > I think thats the procedure followed. > > I have put the remove todo's in separate patch, as its large. > > James is correct, patches 2 and 3 should be concatenated. 2 is > already necessarily large, and adding 3 to it doesn't add much to its > size. > --Juan > Damn, I have to reset the git tree. Then apply patches manually and resend them :( Should I wait.................? :( --- VJ

2 1

Re: [0/3] RSA AES provider to rsaenh
by Juan Lang 20 Nov '07

20 Nov '07

> Damn, I have to reset the git tree. > Then apply patches manually and resend them :( This isn't hard. Assuming you have these as the last three patches in your tree: git reset HEAD~2 git commit -a This will rewind your tree two steps without changing your local files, then commit them again. The more manual method, assuming you have the last patches named 2 and 3: git reset --hard HEAD~2 patch -p1 < 2 patch -p1 < 3 git commit -a This latter method isn't necessary in this case, but it sometimes comes in handy when you have some patches intermingled, and you want to commit several of them in a bunch without committing some in between. E.g. if you wanted to merge patches 1 and 3 into a single patch, you could do it this way instead. HTH, --Juan

1 0

Re: [0/3] RSA AES provider to rsaenh
by James Hawkins 20 Nov '07

20 Nov '07

On 11/20/07, Vijay Kiran Kamuju <infyquest(a)gmail.com> wrote: > Hi, > > This is a description of patch series which I am now going to send. > It adds the new Enhanced RSA AES provider to the rsaenh. > The one application which I know uses it is Office 2007. > The first patch contains the tests for the RSA AES provider. > the second patch contains the RSA AES provider implementation > the last patch removes the todo's in the tests. > > The implementation and tests have been verified by Juan Lang. > You can't fix the tests and remove the todo_wines in separate patches. The tests will fail inbetween commits. -- James Hawkins

3 2

Re: bits: Description of patch set (resend 2) (0/26)
by James Hawkins 20 Nov '07

20 Nov '07

On Nov 19, 2007 6:19 PM, Roy Shea <roy(a)cs.hmc.edu> wrote: > Howdy, > > This is another resend of the BITS patch set. This resend changes > svnhost to be ANSI based, merges patches 25 and 26 (old numbering), > and removes extra headers and linked libraries. > Any particular reason for using ansi? Generally we're moving everything towards unicode and away from ansi. Also, the entire patch set is relatively large, and you might want to consider breaking the set up into chunks (svnhost, bits cleanup etc, bits interface impl) and waiting till the first chunk is committed before moving on to the next. That'll make it easier to get the patches committed (assuming they don't get accepted this time) and it'll keep the mail count down for list members. -- James Hawkins

3 2

Re: server: Change the get_dll_info server request to allow retrievingthe image file name of a process.
by Dmitry Timoshkov 20 Nov '07

20 Nov '07

"Robert Shearman" <rob(a)codeweavers.com> wrote: > + if (req->exe_module) > + dll = list_head( &process->dlls ) ? > + LIST_ENTRY(list_head( &process->dlls ), struct process_dll, entry) : NULL; > + else > + dll = find_process_dll( process, req->base_address ); Is it really necessary to introduce the exe_module request parameter? Main executable has its own base address which should work just fine in find_process_dll(). -- Dmitry.

3 4

Re: d3dx8 [patch 1/10] Implement D3DXPlaneFromPointNormal
by David.Adam＠math.cnrs.fr 20 Nov '07

20 Nov '07

Hello, In the beginning, I thought to share the code between d3drm and d3dx8, but unfortunelaty, it was not a good idea, since the structure quaternion is not the same in the both dll (definitively 4 floats, but 4 floats in d3dx8, 1 float and a d3drmvector in d3drm). So, it was not pretty. Moreover, Once I asked to Alexandre what was the best method to implement d3dx9__.dll. He answered to implement the maximum of functions in d3dx8 and to call these functions from d3dx9__.dll. I proposed the idea of wined3dx.dll and he found that it was not a good idea. So, the basis dll is d3dx8 and we consider d3dx9__.dll as a wrapper (at least, for the math functions) Thunderbird was interested in implementing the other functions. Anyway, I am interested in it too. And also, I am interesting in completing d3drm since there still are a few bugs related to this dll. To finish my post: today (or tommorrow) I shall send the last patch for math-functions in d3dx8. The real job will begin then :D :D David

1 0

Re: Unimplemented function KERNEL32.dll.FindFirstVolumeW
by devzero＠web.de 20 Nov '07

20 Nov '07

i`ve tracked this down to a watchdog issue. whenever i do "cat /dev/watchdog" - my system reboots after ~1 minute. i`ve contacted the maintainer of the watchdog subsystem. looks like a bug to me. thanks roland > -----Ursprüngliche Nachricht----- > Von: "Hans Leidekker" <hans(a)it.vu.nl> > Gesendet: 19.11.07 21:59:29 > An: wine-devel(a)winehq.org > Betreff: Re: Unimplemented function KERNEL32.dll.FindFirstVolumeW > > On Monday 19 November 2007 21:39:16 devzero(a)web.de wrote: > > > i retried with ordinary account and the crash didn`t happen there. > > anyway - should i expect such problems with root ? > > Well, a reboot is not expected but it can certainly happen. If your > app is leaking file descriptors for example, given that there are no > limits for root, it will bring your system to a halt. > > -Hans > ______________________________________________________________________________ Jetzt neu! Im riesigen WEB.DE Club SmartDrive Dateien freigeben und mit Freunden teilen! http://www.freemail.web.de/club/smartdrive_ttc.htm/?mc=021134

2 1

Re: kernel32 [1/5]: Export part of UnhandledExceptionHandler as aninternal Wine function.
by Dmitry Timoshkov 20 Nov '07

20 Nov '07

"Dan Hipschman" <dsh(a)linux.ucla.edu> wrote: > This patch exports the part of UnhandledExceptionFilter that sets a page > as either writable or executable in some cases to allow continued execution > as an internal wine function. This is used in the next patch to implement > exception handling in code generated by WIDL. It needs to be able to do > this because if we can continue execution in these cases then we shouldn't > wind down the handler stack. We also shouldn't execute an "except" block > if we can continue normally. There is no need to introduce Wine-specific exception handlers, there is a sample in dlls/ntdll/tests/exception.c how to add an exception handler without Wine-specific hacks. A DIB exception handler in dlls/winex11.drv/dib.c is another one. -- Dmitry.

1 0

wininet: InternetAttemptConnect() Stub
by Pavel Troller 20 Nov '07

20 Nov '07

Hi! I have a problem running an offline client for ordering photo prints from digital cameras. Even it is an "offline" client, it requires the computer to be online. It seems that in wine, it feels to be offline, thus refusing further operation with a dialog saying something like "Internet connection unavailable" and terminating. The machine is normally connected to the LAN with full net availability. The following is its full console log. patrol@gonio:~/.wine/drive_c/Program Files/droxi/OfflineClient$ wine PhotoOfflineClient.exe fixme:wininet:InternetGetConnectedState always returning LAN connection. fixme:win:SetLayeredWindowAttributes (0x10030,0x00000000,255,2): stub! fixme:win:LockWindowUpdate (0x10028), partial stub! fixme:win:LockWindowUpdate ((nil)), partial stub! fixme:shdocvw:OleObject_Close (0x199510)->(1) fixme:wininet:InternetAttemptConnect Stub I see two network related fixmes - the first one is very frequent but it seems harmless, many programs are able to use the net even they issue such a message. However, the second fixme looks more interesting, it looks like that the program tries to activate the internet connection, which is of course just a stub. Why is it doing so, I don't know. Is this information enough to find a problem (and maybe its fix) or is more debugging needed ? What else to try ?

2 2

Re: mshtml #6: Change FIXME that is usually invalid to TRACE.
by James Hawkins 20 Nov '07

20 Nov '07

On Nov 19, 2007 6:51 PM, Jacek Caban <jacek(a)codeweavers.com> wrote: > --- > dlls/mshtml/navigate.c | 4 ++-- > 1 files changed, 2 insertions(+), 2 deletions(-) > How about a warn instead of a trace? That is unless all the correct interfaces are already available, and any other requested interface shouldn't be available. -- James Hawkins

1 0

Missed Patches
by Alistair Leslie-Hughes 20 Nov '07

20 Nov '07

Hi, Is there any reason the following patches haven't been applied? [1/2] msxml: Implement createTextNode [2/2] msxml: Tests for createTextNode mshtml: Provide support for the IInternetProtocol Interface Best Regards Alistair Leslie-Hughes

2 4

Wrong comparison in ntdll/actctx.c
by nlataill 19 Nov '07

19 Nov '07

Hi, Working on a D3D game, I have found that wine is refusing to load the following manifest file (when Windows does not fail). <?xml version="1.0" encoding="UTF-8" standalone="yes"?>  <assembly xmlns="urn:schemas-microsoft-com:asm.v1" manifestVersion="1.0"> <noInheritable/> <assemblyIdentity type="win32" name="Microsoft.VC80.OpenMP" version=" 8.0.50727.42" processorArchitecture="x86" publicKeyToken="1fc8b3b9a1e18e3b"/> <file name="vcomp.dll" hash="0d009725bd5653af143fdfd9d8abf8adad533d16" hashalg="SHA1"/> </assembly> Adding verbosity to actctx.c, it seems that this game is requesting version 8.0.50608.0, which does match major.minor, but not the whole version number (*major.minor.build.revision)*. trace:actctx:parse_assembly_identity_elem name=L"Microsoft.VC80.OpenMP" version=8.0.50727.42 arch=L"x86" fixme:actctx:parse_assembly_elem VERBOSE: wrong version FOUND: 8/0/50727/42 fixme:actctx:parse_assembly_elem VERBOSE: wrong version EXP: 8/0/50608/0 fixme:actctx:parse_manifest_buffer failed to parse manifest L"C:\\Program Files\\Microsoft.VC80.OpenMP.manifest" fixme:actctx:parse_depend_manifests Could not find dependent assembly L" Microsoft.VC80.OpenMP" According to MSDN, it seems only major.minor is considered for version matching: - http://msdn2.microsoft.com/en-us/library/aa374234.aspx : A version number that changes only in the *build* or *revision* parts indicates that the assembly is backward compatible with prior versions. - http://msdn2.microsoft.com/en-us/library/aa375365.aspx: [...] are manifests used to redirect the version of a side-by-side assembly to another compatible version. The version that the assembly is being redirected to should have the same major.minor values as the original version. I am proposing the following one-liner to ntdll/actctx.c: isa@isa:~/downloads/tmp$ diff -u wine-0.9.49/dlls/ntdll/actctx.c /home/isa/downloads/wine-0.9.49/dlls/ntdll/actctx.c --- wine-0.9.49/dlls/ntdll/actctx.c 2007-11-09 17:56:12.000000000 +0100 +++ /home/isa/downloads/wine-0.9.49/dlls/ntdll/actctx.c 2007-11-19 21:07: 35.000000000 +0100 @@ -1390,8 +1390,8 @@ if (expected_ai) { /* FIXME: more tests */ - if (assembly->type == ASSEMBLY_MANIFEST && - memcmp(&assembly->id.version, &expected_ai->version, sizeof(assembly->id.version))) + if (assembly->type == ASSEMBLY_MANIFEST && + ((assembly->id.version.major != expected_ai->version.major) || (assembly->id.version.minor != expected_ai->version.minor)) ) { FIXME("wrong version\n"); return FALSE; With this patch, the manifest (and related DLL) is loading correctly. This does not prevent my game die shortly after, in D3D code this time. Still work to do ;-) Thanks, Norbert Lataille

3 3

Re: Unimplemented function KERNEL32.dll.FindFirstVolumeW
by devzero＠web.de 19 Nov '07

19 Nov '07

checked if app is leaking fd`s, but lsof didn`t tell anything special. after some investigation i came across the following entry in dmesg: iTCO_wdt: Unexpected close, not stopping watchdog seems that wine is triggering something with this - after unloading iTCO_wdt kernel module, the problem went away. looks that wine is (un-)triggering the watchdog timer and the reboot happens due to this. will further investigate (strace) - maybe i can provide some more input so, it`s probably not a wine issue.... regards roland > -----Ursprüngliche Nachricht----- > Von: "Hans Leidekker" <hans(a)it.vu.nl> > Gesendet: 19.11.07 21:59:29 > An: wine-devel(a)winehq.org > Betreff: Re: Unimplemented function KERNEL32.dll.FindFirstVolumeW > > On Monday 19 November 2007 21:39:16 devzero(a)web.de wrote: > > > i retried with ordinary account and the crash didn`t happen there. > > anyway - should i expect such problems with root ? > > Well, a reboot is not expected but it can certainly happen. If your > app is leaking file descriptors for example, given that there are no > limits for root, it will bring your system to a halt. > > -Hans > __________________________________________________________________________ Erweitern Sie FreeMail zu einem noch leistungsstärkeren E-Mail-Postfach! Mehr Infos unter http://produkte.web.de/club/?mc=021131

1 0

recent cedega contributions to wine
by EA Durbin 19 Nov '07

19 Nov '07

http://www.cedega.com./devreports/ The latest Cedega dev report states it plans to include wine's msi implementation back into it's code base and they have been contributing back to wine. "TransGaming's developers have been contributing some changes back to Wine as well, including in several networking and debugging libraries. " Just out of curiousity, which patches are these? _________________________________________________________________ Your smile counts. The more smiles you share, the more we donate. Join in. www.windowslive.com/smile?ocid=TXT_TAGLM_Wave2_oprsmilewlhmtagline

4 6

Re: credui: Show a balloon tip in the credential dialog if CREDUI_FLAGS_INCORRECT_PASSWORD was specified.
by Juan Lang 19 Nov '07

19 Nov '07

Hi Rob, + IDS_INCORRECTPASSWORDTITLE "Logon unsucessful" The typo's your own work too, eh? --Juan

2 1

Re: d3dx8 [patch 1/10] Implement D3DXPlaneFromPointNormal
by Stefan Dösinger 19 Nov '07

19 Nov '07

Hi, It's not really about this patch, but I am curious what the plans about sharing code between the various d3dx?_* libraries are. I think a big amount of functionality can be shared among d3dx8.dll, d3dx9_*.dll, and propably d3dxof.dll and d3drm.dll. I think this was discussed in the past, but I am not sure if there was any outcome. The solution chosen for d3d was to create an internal library wined3d.dll, and have the other libs call it. I guess something similar should be done for d3dx as well. For non COM interfaces this is pretty easy anyway. We can rename d3dx8 to wined3dx at any time and just load the functions into a new d3d8 thunk using the .spec file, without any C code. So I guess we can leave things as they currently are and move over to a shared lib once non-math functions are implemented, or someone starts d3dx9_*.dll Functions like D3DXCreateTextureFromFileA seem to be a bit harder. Since they take an IDirect3DDevice8 * interface pointer d3dx9 cannot use them as is. Making a shared implementation is also not quite possible because it would need a wined3ddevice then, but only d3d8.dll knows how to get a wined3ddevice from a d3ddevice8. Also, d3dx* would depend on wined3d then, which is propably not what we want. David, do you have any plans regarding that? As a sidenote, I really like your patches! The patches are small but still every patch makes sense for itself, and tests in every patch :-) I will write a HLSL compiler at some point, and propably a shader assembler. This will be needed for a complete d3dx8 and d3dx9 implementation, as well as d3d10. I am not yet sure where to put it, but it could be a standalone lib in libs/, to allow using the compiler at build time, or just include it in wined3dx, if we decide to add such a library.

2 2

Re: Unimplemented function KERNEL32.dll.FindFirstVolumeW
by devzero＠web.de 19 Nov '07

19 Nov '07

Hello Marcus, hello Hans, thanks very much - that was a quick fix ! :) I tried the latest snapshot and now was going able start windirstat. great! i can tell , that it basically works and i`m able to scan a disk. despite some minor quirks like crash of the app when i click trough the cushion treemap, there seems some major issue. at least some very weird major issue: when scannnig trough drive z:\ , my notebook reboots ! This happens reproduceably when windirstat is at ~100.000 entries i had absolutely no stability issues with my notebook so far, running it for some days now and using it very very intensive (much i/o, firefox/open with many many windows containing flash etc) i did an test afterwards to see if this is caused by some corrupt files or whatever, but some find / -type f -exec stat {} \; >/dev/null ls -laR / >/dev/null completes without an error. there are around 200.000 files on my notebook. wasn`t wine a userspace application with no kernel related stuff ? so it`s weird, that windirstat is crashing my notebook, isn`t it ? maybe somebody is able to reproduce ? regards roland > -----Ursprüngliche Nachricht----- > Von: "Marcus Meissner" <marcus(a)jet.franken.de> > Gesendet: 19.11.07 07:48:04 > An: devzero(a)web.de > CC: Hans Leidekker <hans(a)it.vu.nl>, wine-devel(a)winehq.org > Betreff: Re: Unimplemented function KERNEL32.dll.FindFirstVolumeW > > On Sun, Nov 18, 2007 at 10:14:17PM +0100, devzero(a)web.de wrote: > > Hello Hans, > > > > thanks for that info. > > > > latest version i tried here is > > > > wine-0.9.49-11.3.i586.rpm > > wine-snapshot-0.9.49.20071118-11.1.i586.rpm > > > > but same error. > > maybe that`s not recent enough ? > > > > do i need to enable special option or add some files ? > > > > sorry, i`m quite new to wine and maybe i should read deeper into the docs instead of posting :) > > It is not yet committed to the mainline repository. Once it > is it will show up in "wine-snapshot" of the openSUSE buildservice > of this day. > > Ciao, Marcus > _______________________________________________________________________ Jetzt neu! Schützen Sie Ihren PC mit McAfee und WEB.DE. 3 Monate kostenlos testen. http://www.pc-sicherheit.web.de/startseite/?mc=022220

2 1

Re: svchost: Added base implementation of svchost (1/27)
by Juan Lang 19 Nov '07

19 Nov '07

Hi Roy, +#define WIN32_LEAN_AND_MEAN That's not needed if you're not including windows.h (which you're not.) + if (unicode[i] > 129) You have an off-by-two error here. 7-bit clean means if (unicode[i] > 127). You can also use something like if (unicode[i] & 0x80). Stylistic note, use or ignore at your option: + HeapFree(GetProcessHeap(), 0, service_param_key); + return FALSE; (snip) + HeapFree(GetProcessHeap(), 0, service_param_key); + RegCloseKey(service_hkey); + return FALSE; (snip) + HeapFree(GetProcessHeap(), 0, service_param_key); + HeapFree(GetProcessHeap(), 0, dll_name_short); + RegCloseKey(service_hkey); + return FALSE; (snip) Some judicious use of goto could reduce code for all your error paths. Remember that HeapFree of a NULL pointer is allowed. + dll_service_main = GetRegValue(service_hkey, service_main); + RegCloseKey(service_hkey); + if (!dll_service_main) + { + service_main_a = UnicodeToAscii(service_main); + } + else + { + service_main_a = UnicodeToAscii(dll_service_main); + HeapFree(GetProcessHeap(), 0, dll_service_main); Why not just query dll_service_main and service_main from the registry as an ASCII value to begin with? --Juan

2 2

imagelist test exception?
by Dan Kegel 19 Nov '07

19 Nov '07

comctl32/tests/imagelist.c seems to throw an exception during GetDIBits(), but for some reason, the test continues and succeeds (except in Valgrind, which is why I noticed). Do other folks also see a seh exception in the +seh log on this test? I narrowed the test case down a bit (see patch). Can anyone clue me in on what's going on here?

3 4

Re: Unimplemented function KERNEL32.dll.FindFirstVolumeW
by devzero＠web.de 19 Nov '07

19 Nov '07

Hello Hans, thanks for that info. latest version i tried here is wine-0.9.49-11.3.i586.rpm wine-snapshot-0.9.49.20071118-11.1.i586.rpm but same error. maybe that`s not recent enough ? do i need to enable special option or add some files ? sorry, i`m quite new to wine and maybe i should read deeper into the docs instead of posting :) regards roland > -----Ursprüngliche Nachricht----- > Von: "Hans Leidekker" <hans(a)it.vu.nl> > Gesendet: 18.11.07 21:45:34 > An: wine-devel(a)winehq.org > Betreff: Re: Unimplemented function KERNEL32.dll.FindFirstVolumeW > > On Sunday 18 November 2007 20:38:34 devzero(a)web.de wrote: > > > i`m new to wine. > > I tried windirstat under wine-0.9.44-15 and crossover-office 6.2.0-1 under opensuse10.3 - but it doesn`t work. > > > > i`m getting > > wine: Call from 0x7b8447d0 to unimplemented function KERNEL32.dll.FindFirstVolumeW, aborting > > wine: Unimplemented function KERNEL32.dll.FindFirstVolumeW called at address 0x7b8447d0 (thread 0019), starting debugger... > > Unhandled exception: unimplemented function KERNEL32.dll.FindFirstVolumeW called in 32-bit code (0x7b844852). > > I submitted a stub implementation for this function which makes > windirstat start up fine here. > > -Hans > ______________________________________________________________________________ Jetzt neu! Im riesigen WEB.DE Club SmartDrive Dateien freigeben und mit Freunden teilen! http://www.freemail.web.de/club/smartdrive_ttc.htm/?mc=021134

3 2

Re: msi: Install Entire Feature
by James Hawkins 19 Nov '07

19 Nov '07

On 11/18/07, Daniel Sim <daniel.ssq89(a)gmail.com> wrote: > "Install entire feature" enabled and implemented. > What is this patch for? Is this to fix a bug? + /* FIXME: Why 0x1000? */ Don't add random comments about things you don't understand in the code. +msi_seltree_install_entire_feature(HWND hwnd, HTREEITEM hItem, MSIPACKAGE *package, UINT r) { Please use the style of the rest of the file (bracket below function). + for ( + child = TreeView_GetChild (hwnd, hItem); + child; + child = TreeView_GetNextSibling (hwnd, child) ) Style changed again...and that's just ugly. + msi_seltree_install_entire_feature (hwnd, child, package, r); Why are you doing this recursively? -- James Hawkins

1 0

Unimplemented function KERNEL32.dll.FindFirstVolumeW
by devzero＠web.de 18 Nov '07

18 Nov '07

Hello ! i`m new to wine. I tried windirstat under wine-0.9.44-15 and crossover-office 6.2.0-1 under opensuse10.3 - but it doesn`t work. i`m getting wine: Call from 0x7b8447d0 to unimplemented function KERNEL32.dll.FindFirstVolumeW, aborting wine: Unimplemented function KERNEL32.dll.FindFirstVolumeW called at address 0x7b8447d0 (thread 0019), starting debugger... Unhandled exception: unimplemented function KERNEL32.dll.FindFirstVolumeW called in 32-bit code (0x7b844852). does somebody know when this function will be available ? is this a rarely used function? http://msdn2.microsoft.com/en-us/library/aa364425.aspx tells "Retrieves the name of a volume on a computer. FindFirstVolume is used to begin scanning the volumes of a computer." regards roland _____________________________________________________________________ Der WEB.DE SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen! http://smartsurfer.web.de/?mc=100071&distributionid=000000000066

2 1

re: advapi32/service.c -- remove untriggerable check
by Dan Kegel 18 Nov '07

18 Nov '07

Uwe wrote: >Gerald Pfeifer writes: > >> n is of type DWORD which is unsigned, so n < 0 always will >> evaluate to false. > > Is dropping the check the right answer? Shouldn't the check test for high > values like > 0xff00 and report a possible problem? Indeed. IMHO we don't need patches like this, and Gerald is not thinking hard enough. Simplistic just-remove-the-warning patches are a Bad Thing, and we Don't Want To Encourage Them. Please stop. - Dan -- Wine for Windows ISVs: http://kegel.com/wine/isv

2 1

Re: gdi32: Add stub for RemoveFontMemResourceEx
by Dmitry Timoshkov 18 Nov '07

18 Nov '07

"Klaus Layer" <klaus.layer(a)gmx.de> wrote: > Nokia PC Suite installation crashes due to missing function > RemoveFontMemResourceEx. This patch adds function RemoveFontMemResourceEx > which just calls WineEngRemoveFontMemResourceEx as stub You forgot to create WineEngRemoveFontMemResourceEx implementation for a case when HAVE_FREETYPE is not defined. -- Dmitry.

2 1

Re: advapi32/service.c -- remove untriggerable check
by Uwe Bonnes 18 Nov '07

18 Nov '07

>>>>> "Gerald" == Gerald Pfeifer <gerald(a)pfeifer.com> writes: Gerald> n is of type DWORD which is unsigned, so n < 0 always will Gerald> evaluate to false. Gerald> Gerald Gerald> ChangeLog: Remove untriggerable check. Gerald> Index: dlls/advapi32/service.c Gerald> =================================================================== Gerald> RCS file: /home/wine/wine/dlls/advapi32/service.c,v retrieving Gerald> revision 1.160 diff -u -3 -p -r1.160 service.c --- Gerald> dlls/advapi32/service.c 15 Oct 2007 16:29:59 -0000 1.160 +++ Gerald> dlls/advapi32/service.c 18 Nov 2007 06:01:28 -0000 @@ -2107,9 Gerald> +2107,6 @@ QueryServiceConfigW( SC_HANDLE hService, n -= Gerald> sizeof(WCHAR); } Gerald> - if( n < 0 ) - ERR("Buffer overflow!\n"); - TRACE("Image path = Gerald> %s\n", debugstr_w(lpServiceConfig->lpBinaryPathName) ); Gerald> TRACE("Group = %s\n", Gerald> debugstr_w(lpServiceConfig->lpLoadOrderGroup) ); Gerald> TRACE("Dependencies = %s\n", Gerald> debugstr_w(lpServiceConfig->lpDependencies) ); Is dropping the check the right answer? Shouldn't the check test for high values like > 0xff00 and report a possible problem? -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------

1 0

Valgrind results for Nov 12 & 13
by Dan Kegel 18 Nov '07

18 Nov '07

It seems that valgrind doesn't like Ubuntu 7.10; I had to drop back to my Feisty system to generate good valgrind stack dumps. And the tests didn't hang once! Results for the last two days are at http://kegel.com/wine/valgrind/20071112/ http://kegel.com/wine/valgrind/20071113/ The logs are trimmed a bit more to make diffs less noisy. Diff now shows clearly that there were significant improvements in three tests between those two runs: ole32/compobj, ole32/marshal, and user32/dde Errors still remain in all three tests, though. - Dan

5 7

Re: bits: Added framework for testing the BackgroundCopyJob interface (25/27)
by Detlef Riekenberg 18 Nov '07

18 Nov '07

On Fr, 2007-11-16 at 16:21 -0800, Roy Shea wrote: > +/* Generic test setup */ > +static BOOL setup(void) unused code > +/* Generic test cleanup */ > +static void teardown(void) unused code +START_TEST(job) > +{ > + CoInitialize(NULL); > + CoUninitialize(); > +} Do you want to call "setup()" and "teardown()" here? -- By by ... Detlef

1 0

Re: bits: INF and resource file for qmgr registration (8/27)
by Detlef Riekenberg 18 Nov '07

18 Nov '07

On Fr, 2007-11-16 at 16:16 -0800, Roy Shea wrote: > dlls/qmgr/qmgr.inf | 32 ++++++++++++++++++++++++++++++++ > dlls/qmgr/rsrc.rc | 21 +++++++++++++++++++++ > 2 files changed, 53 insertions(+), 0 deletions(-) The change for Makefile.in to add rsrc.rc is missing here. (found in the next patch) -- By by ... Detlef

1 0

Re: svchost: Added base implementation of svchost (1/27)
by Robert Shearman 18 Nov '07

18 Nov '07

Roy Shea wrote: > +TOPSRCDIR = @top_srcdir@ > +TOPOBJDIR = ../.. > +SRCDIR = @srcdir@ > +VPATH = @srcdir@ > +MODULE = svchost.exe > +APPMODE = -municode > +IMPORTS = advapi32 kernel32 ole32 > I can't see any reason why svchost needs to import ole32. -- Rob Shearman

1 0

Hi, and a regression in bug 10486
by Paweł Różański 18 Nov '07

18 Nov '07

Hi I have found a regression in wine. a strange one. What is a policy about posting such regressions, here on malling list? (Or is regression is worth enough to post it here?) I don't play that particular game on wine and it's "minor" bug - but i have found (via bitsect) that "user32: Cache the window class name on the client side." produce strange effects on directx code. I have faith that regressions are supposed to post here as well. More info in bug #10486 Take care & Hope that helps -- Paweł Różański, a wine user witch occasionally post something on appdb/bugzilla

1 0

Regression testing and cc-ing the author
by Paul Vriens 18 Nov '07

18 Nov '07

Hi, I've opened a bug (which if fixed now btw) and the regression testing page at http://wiki.winehq.org/RegressionTesting said "Also be sure to add the author of the patch to the CC". So this is what I did. Should that line be removed from the wiki or do we generally want the author to be CC-ed? -- Cheers, Paul.

4 4

Re: Removing Windows and Solaris from the Downloads page
by Chris McKillroy 17 Nov '07

17 Nov '07

I compiled wine on Solaris using a .spec provided by the people of http://pkgbuild.sourceforge.net/spec-files-extra/ Maybe contact them ? My version runs quite nice as far as i can see. Cheers

1 0

Re: Add tests for RSA_AES provider in rsaenh
by Reece Dunn 17 Nov '07

17 Nov '07

On 16/11/2007, Vijay Kiran Kamuju <infyquest(a)gmail.com> wrote: > ChangeLog > -------------- > Add tests for RSA_AES provider in rsaenh > > Note: > ------- > Some tests might fail as there is no RSA_AES provider in wine. > These will be fixed once the implementation is in place, which will be > sent next. The recommended way to do this is mark the parts of the tests that fail in Wine with todo_wine. - Reece

3 3

Re: ReadProcessMemory question
by Robert Shearman 17 Nov '07

17 Nov '07

Dave Cartright wrote: > I am trying to read the memory of another process with > ReadProcessmemory and I can only get it to work when I commit the > entire region I want to read with VirtualAllocEx - but as you probably > know when I do this it initializes the entire memory region to 0s. > So my question is how do I get ReadProcessmemory to read from the region?? It sounds to me like you're trying to read the previous contents of areas of memory. However, you don't seem to realise either that virtual memory is not physical memory or that uncommitted virtual memory is not backed by a page of memory in swap or physical memory, so calling ReadProcessMemory on it doesn't make sense as there is no data there to be read! Furthermore, when a page of virtual memory is deallocated then it free to be allocated again by any process in the system, and as you say it is initialised to all 0's as a security measure and you can't get that same page back again. -- Rob Shearman

1 0

re: ReadProcessMemory question
by Dan Kegel 17 Nov '07

17 Nov '07

Dave wrote: > I am trying to read the memory of another process with > ReadProcessmemory and I can only get it to work when > I commit the entire region I want to read with VirtualAllocEx - > but as you probably know when I do this it > initializes the entire memory region to 0s. > So my question is how do I get ReadProcessmemory to read from the region?? What are you trying to do? A little more context might help. FWIW, ReadProcessMemory has some drawbacks as an IPC mechanism: http://blogs.msdn.com/oldnewthing/archive/2006/01/17/513779.aspx Here are a few examples of how to use ReadProcessMemory: http://www.gamedev.net/community/forums/topic.asp?topic_id=328279 http://www.codeproject.com/threads/int64_memsteal.asp

1 0

ReadProcessMemory Question
by Dave Cartright 17 Nov '07

17 Nov '07

I am trying to read the memory of another process with ReadProcessMemory but I can only get it to work when I use VirtualAllocEx to commit the region of memory - problem is (as you very well know) that VirtualAllocEx initializes the region with 0s. So my question is how do I get ReadProcessMemory to work without using VirtualAllocEx?? --------------------------------- Be a better pen pal. Text or chat with friends inside Yahoo! Mail. See how.

1 0

#winehackers invite only?
by Stefan Dösinger 17 Nov '07

17 Nov '07

Hello, Did anyone change #winehackers to require an invite? What was the reason for this? Thanks, Stefan

2 2

Re: [Bug 10464] Regression when installing Skype
by James Hawkins 17 Nov '07

17 Nov '07

On 11/16/07, wine-bugs(a)winehq.org <wine-bugs(a)winehq.org> wrote: > http://bugs.winehq.org/show_bug.cgi?id=10464 > > --- Comment #5 from Austin English <austinenglish(a)gmail.com> 2007-11-16 18:33:31 --- > James, you can disable bugzilla from e-mailing you for CC's and/or disable all > e-mail. > > Austin, it's not about that. Disabling all emails is against the point. I'm subscribed to wine-bugs, so I don't need to be cc'ed. This is the position of several other developers, so I don't see where the confusion is coming from. If anything, ask the developer whose patch created a regression to assign the bug to himself. I'm going to stop now before I start ranting. -- James Hawkins

1 0

Re: USER32: win16 applications should not see long filenames returned from dialog control (try 4)
by Alex Villacís Lasso 16 Nov '07

16 Nov '07

Alex Villacís Lasso escribió: > I hope this version addresses the fact DlgDirSelect does not modify the > last error code, and therefore cannot be used to distinguish between a > valid and an invalid file. > > Changelog: > * Applications should not see long filenames returned from dialog control > > ------------------------------------------------------------------------ > > This patch has not been applied, and has not been commented on either. Is there something wrong with it? -- perl -e '$x=2.4;print sprintf("%.0f + %.0f = %.0f\n",$x,$x,$x+$x);'

1 0

pglXMakeCurrent causes wine to crash on Mac os x
by Tobias Platen 16 Nov '07

16 Nov '07

Hello if i run wine on Mac os x 10.4.10 with Apples X11 (opengl linked staticly) i get the following runtime errors; X Error of failed request: BadValue (integer parameter out of range for operation) Major opcode of failed request: 129 (Apple-DRI) Minor opcode of failed request: 2 () Value in failed request: 0x57 Serial number of failed request: 16 Current serial number in output stream: 16 I have found out that commenting out the bad code will not cause that X Error wine crashes in a access violation (SIGSEGV) if (ctx) { // pglXMakeCurrent(gdi_display, win, ctx); } else { ERR(" couldn't initialize OpenGL, expect problems\n"); wine_tsx11_unlock(); return FALSE; } Does anybody know how to get opengl working on the mac.

1 0

Re: ntdll: send_debug_event: don't send uninitialized bytes
by Robert Shearman 16 Nov '07

16 Nov '07

Dan Kegel wrote: > +/* Returns size of initialized part of a particular EXCEPTION_RECORD */ > +static int exception_record_size( const EXCEPTION_RECORD *rec ) > +{ > + return sizeof(*rec) > + - (EXCEPTION_MAXIMUM_PARAMETERS-rec->NumberParameters) * sizeof(ULONG_PTR); > +} FIELD_OFFSET exists for this exact purpose. -- Rob Shearman

2 2

winetest_interactive : how does it work?
by Paul Vriens 16 Nov '07

16 Nov '07

Hi, I just ran winetest (by double clicking on the executable) on my WinXP box. I was missing some advpack popups that were there before. Francois sent in a patch to only do some particular tests when winetest_interactive is true. I would assume that clicking on a executable is interactive, not? AFAIK winetest_interactive is set to true|1 if the WINETEST_INTERACTIVE variable is set to 1 (include/wine/test.h). In programs/winetest/main.c we however only set WINETEST_INTERACTIVE to 0: 692 interactive_no[] = "WINETEST_INTERACTIVE=0", shouldn't WINETEST_INTERACTIVE be set according to the interactive variable in that same file? -- Cheers, Paul.

2 1

Re: Start: Convert start to Unicode. [try5]
by Alexandre Julliard 16 Nov '07

16 Nov '07

Anatoly Lyutin <vostok(a)etersoft.ru> writes: > @@ -202,9 +246,9 @@ int main(int argc, char *argv[]) > DWORD waitcode; > waitcode = WaitForSingleObject(sei.hProcess, INFINITE); > if (waitcode) > - fatal_error("WaitForSingleObject", GetLastError()); > + fatal_error(WaitForSingleObjectW, GetLastError()); > if (!GetExitCodeProcess(sei.hProcess, &exitcode)) > - fatal_error("GetExitCodeProcess", GetLastError()); > + fatal_error(GetExitCodeProcessW, GetLastError()); This is way overkill, there's no reason to convert errors like these to Unicode. WINE_ERR would work just as well. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH] svchost: Added base implementation of svchost (1/12)
by Alexandre Julliard 16 Nov '07

16 Nov '07

Roy Shea <roy(a)cs.hmc.edu> writes: > +/* Allocate and initializes an ANSI version of the Unicode string */ > +static LPSTR UnicodeToAnsi(const WCHAR *unicode) > +{ > + int size; > + LPSTR ansi; > + > + size = WideCharToMultiByte(CP_UTF8, 0, unicode, -1, NULL, 0, NULL, NULL); > + ansi = HeapAlloc(GetProcessHeap(), 0, size); > + WideCharToMultiByte(CP_UTF8, 0, unicode, -1, ansi, size, NULL, NULL); > + return ansi; > +} UTF-8 is not the same thing as ANSI, this function is misleading. For procedure names what you probably want is a special-purpose function that rejects anything that isn't 7-bit ASCII. > + ret = RegOpenKeyExW(HKEY_LOCAL_MACHINE, service_param_key, 0, > + KEY_READ, &service_hkey); > + if (ret != ERROR_SUCCESS) > + { > + WINE_ERR("cannot open key %s, err=%d\n", > + wine_dbgstr_w(service_param_key), ret); > + HeapFree(GetProcessHeap(), 0, service_param_key); > + return FALSE; > + } > + > + /* Find DLL associate with service from key */ > + dll_name_short = GetRegValue(service_hkey, service_dll); > + if (!dll_name_short) > + { > + WINE_ERR("cannot find registry value %s for service %s\n", > + wine_dbgstr_w(service_dll), wine_dbgstr_w(service_name)); > + HeapFree(GetProcessHeap(), 0, service_param_key); > + return FALSE; > + } You are leaking registry handles pretty much everywhere. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: server: Add a call for retrieving the image file name of a process.
by Alexandre Julliard 16 Nov '07

16 Nov '07

Robert Shearman <rob(a)codeweavers.com> writes: > --- > dlls/ntdll/process.c | 20 +++++++++++++++++++- > dlls/ntdll/tests/info.c | 36 ++++++++++++++++++++++++++++++++++++ > server/process.c | 31 +++++++++++++++++++++++++++++++ > server/protocol.def | 9 +++++++++ > 4 files changed, 95 insertions(+), 1 deletions(-) You should use the existing get_dll_info request for that. -- Alexandre Julliard julliard(a)winehq.org

2 2

Re: [1/6] try 2: msvcrt: Implement strcpy_s
by Detlef Riekenberg 16 Nov '07

16 Nov '07

On Do, 2007-11-15 at 17:35 +0100, Stefan Dösinger wrote: > + memset(dest, 'X', sizeof(dest)); > + ret = pstrcpy_s(dest, sizeof(dest), small); > + ok(ret == 0, "Copying a string into a big enough destination > returned %d, expected 0\n", ret); > + ok(dest[0] == 's' && dest[1] == 'm' && dest[2] == 'a' && dest[3] > == 'l' && > + dest[4] == 'l' && dest[5] == '\0'&& dest[6] == 'X' && dest[7] > == 'X', Is it really needed to check the memory after the end of the string? Does an application depend on this? --- cut --- memset(dest, 'X', sizeof(dest)-1); /* never read past the buffer, when we check the result */ dest[sizeof(dest)-1] = '\0'; ret = pstrcpy_s(dest, sizeof(dest), small); ok( !ret && !lstrcmpA(dest, small), "got %d and '%s' (expected '0' and '%s')\n", ret, dest, small); -- cut -- For my tests with buffers, i found it useful to test with: - buffer present, but size is 0 - buffer one byte smaller as the string (no space left for the '\0') - buffer large enough for the string, but no space left for the '\0' - buffer exact as needed - buffer one byte larger as needed - empty source string - NULL source - NULL target This allow you to verify, how the terminating '\0' is handled. Did you test with invalid pointers (crash / handled graceful)? -- By by ... Detlef

2 2

re: dlls/msi/string.c -- fix use of signed versus unsigned (RESEND)
by Dan Kegel 16 Nov '07

16 Nov '07

Gerald Pfeifer wrote: > How do you propose to address the following? > string.c: In function 'msi_addstring': > string.c:208: warning: comparison between signed and unsigned For starters: diff --git a/configure.ac b/configure.ac index 3d62f45..c97aa8e 100644 --- a/configure.ac +++ b/configure.ac @@ -1134,6 +1134,7 @@ then WINE_TRY_CFLAGS([-fno-strict-aliasing]) WINE_TRY_CFLAGS([-Wdeclaration-after-statement]) WINE_TRY_CFLAGS([-Wwrite-strings]) + WINE_TRY_CFLAGS([-Wno-sign-compare]) dnl Check for noisy string.h saved_CFLAGS="$CFLAGS" Then, after we've fixed all the real bugs, maybe consider going back and turning on the signed/unsigned comparison warning.

1 0

Re: [2/6] try 2: msvcrt: implement strcat_s
by Dmitry Timoshkov 16 Nov '07

16 Nov '07

"Stefan Dösinger" <stefan(a)codeweavers.com> wrote: > +@ cdecl strcat_s(str str) MSVCRT_strcat_s ... > +int CDECL MSVCRT_strcat_s( char* dst, MSVCRT_size_t elem, const char* src ) The spec file entry still has wrong number of parameters. -- Dmitry.

1 0

Re: [3/6] try 2: msvcrt: implement _mbsnbcpy_s
by Detlef Riekenberg 15 Nov '07

15 Nov '07

On Do, 2007-11-15 at 17:37 +0100, Stefan Dösinger wrote: > + const unsigned char small[] = "small"; > + int ret; > + memset(dest, 'X', sizeof(dest)); > + ret = p_mbsnbcpy_s(dest, sizeof(dest), small, sizeof(small)); > + ok(ret == 0, "_mbsnbcpy_s: Copying a string into a big enough > destination returned %d, expected 0\n", ret); > + ok(dest[0] == 's' && dest[1] == 'm' && dest[2] == 'a' && dest[3] > == 'l' && > + dest[4] == 'l' && dest[5] == '\0'&& dest[6] == 'X' && dest[7] > == 'X you can use memcmp, when small is defined different: + const unsigned char small[] = "small\0XX"; I also suggest to make the buffer larger as you tell the api and always terminate the buffer unsigned char dest[10]; dest[sizeof(dest)-1] = '\0'; -- By by ... Detlef

1 0

Re: [PATCH] bits: Stub version of qmgr with svchost and proxy (5/12)
by Detlef Riekenberg 15 Nov '07

15 Nov '07

On Do, 2007-11-15 at 08:22 -0800, Roy Shea wrote: > This adds svchost, proxying, and a stub version of the qmgr.dll > process. The > functional code in this patch is boilerplate required to run a DLL as > a > service via svchost. > --- > dlls/qmgr/Makefile.in | 6 ++- > dlls/qmgr/factory.c | 85 +++++++++++++++++++++++ > dlls/qmgr/qmgr.c | 8 ++- > dlls/qmgr/qmgr.inf | 32 +++++++++ > dlls/qmgr/qmgr.spec | 1 + > dlls/qmgr/qmgr_main.c | 98 +++++++++++++++++++++++++-- > dlls/qmgr/qmgr_private.h | 14 ++++- > dlls/qmgr/qmgr_service.c | 156 > +++++++++++++++++++++++++++++++++++++++++++ > dlls/qmgr/rsrc.rc | 21 ++++++ > dlls/qmgrprxy/Makefile.in | 25 +++++++ > dlls/qmgrprxy/qmgrprxy.idl | 20 ++++++ > dlls/qmgrprxy/qmgrprxy.spec | 5 ++ > 12 files changed, 461 insertions(+), 10 deletions(-) It would be better, when you can reduce the Patchsize, but I have no knowledge about the code. (must everything from factory.c, qmgr_main.c and qmgr_service.c in the same Patch?) qmgrprxy looks really splitable from the rest as a seperate Patch. > -0,0 +1,5 @@ > +@ stdcall -private DllCanUnloadNow() > +@ stdcall -private DllGetClassObject(ptr ptr ptr) > +@ stdcall -private DllRegisterServer() > +@ stdcall -private DllUnregisterServer() > +@ stdcall GetProxyDllInfo(ptr ptr) I'm unable to find the code for this functions. Is everything build from qmgrprxy.idl ? -- By by ... Detlef

1 0

Re: [PATCH] bits: Implemented BITS_IBackgroundCopyManager_Release and Test (12/12)
by Detlef Riekenberg 15 Nov '07

15 Nov '07

On Do, 2007-11-15 at 08:24 -0800, Roy Shea wrote: > dlls/qmgr/qmgr.c | 19 ++++++++++++++- > dlls/qmgr/qmgr.spec | 4 +- > dlls/qmgr/tests/Makefile.in | 17 ++++++++++++++ > dlls/qmgr/tests/qmgr.c | 50 > +++++++++++++++++++++++++++++++++++++++++++ > 4 files changed, 86 insertions(+), 4 deletions(-) The initial test can always be a seperate Patch. -- By by ... Detlef

1 0

Re: [PATCH] bits: Added BITS IDL and standard header file (3/12)
by Detlef Riekenberg 15 Nov '07

15 Nov '07

On Do, 2007-11-15 at 08:21 -0800, Roy Shea wrote: > +++ b/include/bitsmsg.h > @@ -0,0 +1,83 @@ > +/* > + * Return values that may be generated by BITS as described at: > + * http://msdn2.microsoft.com/en-us/library/aa362823.aspx We want to remove all links to MSDN from wine (the URL changes to often) and include/bitsmsg.h fit perfect in a seperate Patch. -- By by ... Detlef

1 0

Re: [PATCH] bits: Add implementation of BITS_IClassFactory_CreateInstance (9/12)
by Detlef Riekenberg 15 Nov '07

15 Nov '07

On Do, 2007-11-15 at 08:24 -0800, Roy Shea wrote: > Implementation of BITS_IClassFactory_CreateInstance ond the support > function > BackgroundCopyManagerConstructor. > --- > dlls/Makefile.in | 5 +++++ > + qmgrprxy \ > + qmgrprxy/libqmgrprxy.$(IMPLIBEXT) \ This changes are unrelated for the actual Patch You should never send any autogenerated code. (<wine>/tools/make_makefiles does the changes during commit) -- By by ... Detlef

1 0

Re: [3/4] msxml: Implement removeAttribute
by Mikołaj Zalewski 15 Nov '07

15 Nov '07

A problem with this patch is that the user may have an IXMLDOMNode * reference to the attribute node being removed. The xmlFreeProp will free the xmlnode so accessing the IXMLDOMNode could result in memory corruption. Under MSXML it's possible to access such node and even reconnect it e.g. to a different documents (at least I seem to remember such a thing for element nodes). I don't think supporting it is possible with our current reference counting only for the whole document. I was thinking about also keeping reference counts for each connected component and each node. The connected component count could be used to determine if the node can be freed in such a situation and the node count could be used to update the connected component counts if the tree gets split. But so far I haven't written any code (Another possibility is to keep in each node the refcount of the whole subtree starting in this node. This is more elegant but it would take a linear time with respect to the tree height to update the refcount). Mikołaj Zalewski

2 1

Re: Patch: ntoskrnl.exe - Added PsCreateSystemThread
by Stefan Dösinger 15 Nov '07

15 Nov '07

Am Donnerstag, 20. September 2007 03:49:40 schrieb Carroll Vance: > I have tested this with a driver I made and it seemed to > work fine. I don't know much about ntoskrnl.exe, but if you have a test driver, you may want to include it in the patch. Other libraries have a set of regression tests in dlls/<lib>/tests . ntoskrnl.exe doesn't have any yet, but I don't know if that is a policy or just because no one started it yet.

2 1

Re: dlls/msi/string.c -- fix use of signed versus unsigned (RESEND)
by James Hawkins 15 Nov '07

15 Nov '07

On 11/14/07, Gerald Pfeifer <gerald(a)pfeifer.com> wrote: > Last one for today, promised. ;-) > > Gerald > > ---------- Forwarded message ---------- > From: Gerald Pfeifer <gerald(a)pfeifer.com> > To: <wine-patches(a)winehq.org> > Date: Sat, 3 Nov 2007 20:02:18 +0100 (CET) > Subject: dlls/msi/string.c -- fix use of signed versus unsigned > > This was a bit more tricky, but I hope I got it right. Basically the > problem here was that we used a parameter/variable n of type UINT both > as a true unsigned (the return of the hash function) and as a true signed > variable (comparing against -1). > > This removes the following three compiler diagnostics that I've been > seeing. In other words, this is also addressing very real issues! > > string.c: In function 'msi_addstring': > string.c:208: warning: comparison of unsigned expression < 0 is always false > string.c: In function 'msi_addstringW': > string.c:260: warning: comparison of unsigned expression < 0 is always false > string.c: In function 'msi_string2idW': > string.c:400: warning: comparison between signed and unsigned > > Gerald > > ChangeLog: > Fix use of signed versus unsigned variables. > This change is unnecessarily complicated. Also, the only actual bug of this type that was in string.c (comparison..always false) has been fixed: commit 1c1cf26997cdf491642b2086ad817a79ced99a2e Author: James Hawkins <truiken(a)gmail.com> Date: Mon Nov 5 04:49:07 2007 -0500 msi: Explicitly check the returned value against -1 as the variable is unsigned. All these patches to silence extraneous warnings makes me wonder whether you really understand the code or if you're just trying to get rid of warnings... -- James Hawkins

2 3

Question: How to link kernel32.dll.so to get loaded at a fixed address (0x7bf0000) each time in a Solaris environment
by Chris McKillroy 15 Nov '07

15 Nov '07

Hello everybody, I recently run into the following problem: I tried to use a DLL injection into a process started by a controller Process using an external code hooking dll (http://www.madshi.net/madCodeHookDescription.htm) to achieve some sort of datalink. In a windows environment the solution works stable and perfectly. (Controller Process NWNX4 as provided by nwnx4.org, child process nwn2server.exe, as provided by Obsidian - all legal and intended use). Under Solaris10 (wine version 0.9.49) the child process regularly crashed upon creation, respectivley when the dll was injected. After a lengthy debugging session it appeared that kernel32.dll.so always gets loaded to a varying address by the runtime linker, which seems to crash the whole procedure. I succeeded making a kernel32.dll.so which gets loaded to a constant address in Solaris by using the tool "crle" on Solarice, but I couldn't control the address - it appeared to be somewhat arbitrary which base address was chosen. The crash problem remained. I would like to try if the procedure would work if I could get a kernel32.dll.so which always gets loaded to the same address 0x7bf0000, as in standard windows. After a posting to the Solaris Linker mailing list we got an answer (http://mail.opensolaris.org/pipermail/tools-linking/2007-November/000945.ht…) which was designed to work with the SUN toolchain. My question now is: How would I have to change things (makefile or whatsever) to get kernel32.dll.so linked in a way to achive that it will be loaded to address 0x7bf0000each time under Solaris. Thanks Chris

2 1

re: Question: How to link kernel32.dll.so to get loaded at a fixed
by Dan Kegel 15 Nov '07

15 Nov '07

Chris McKillroy asked: > I tried to use a DLL injection ... > (http://www.madshi.net/madCodeHookDescription.htm) > Under Solaris10 (wine version 0.9.49) the child process regularly > crashed... [because] kernel32.dll.so > always gets loaded to a varying address by the runtime linker Yep. Known problem with Wine. > After a posting to the Solaris Linker mailing list we got an answer > (http://mail.opensolaris.org/pipermail/tools-linking/2007-November/000945.ht…) > ... > My question now is: > How would I have to change things (makefile or whatsever) to get > kernel32.dll.so linked in a way to achive that it will be loaded to > address 0x7bf0000 each time under Solaris. Good question. It's probably pretty involved; you may have to wait for an expert to try it and see what it breaks. Alexandre suggested it might not work recently (see http://bugs.winehq.org/show_bug.cgi?id=10273#c17 ), but perhaps he's being overly pessimistic. BTW if that idea works for Solaris, it might work for Linux, too. - Dan

2 1

Re: Start: Convert start to Unicode. [try4]
by Dmitry Timoshkov 15 Nov '07

15 Nov '07

"Anatoly Lyutin" <vostok(a)etersoft.ru> wrote: > Changelog: > Convert start.exe to Unicode. Please carefully check buffer sizes passed to Win32 APIs. -- Dmitry.

1 0

Re: [2/6] msvcrt: Implement strcat_s
by Dmitry Timoshkov 15 Nov '07

15 Nov '07

"Stefan Dösinger" <stefan(a)codeweavers.com> wrote: > +@ cdecl strcat_s(str str) MSVCRT_strcat_s ... > +int CDECL MSVCRT_strcat_s( char* dst, size_t elem, const char* src ) > +{ > + unsigned int i, j; ... > +static int (*pstrcat_s)(char *dst, unsigned int len, const char *src); This one has the same objections as 2 previous patches. -- Dmitry.

1 0

Re: [3/6] msvcrt: implement _mbsnbcpy_s
by Dmitry Timoshkov 15 Nov '07

15 Nov '07

"Stefan Dösinger" <stefan(a)codeweavers.com> wrote: > +int CDECL _mbsnbcpy_s(unsigned char* dst, MSVCRT_size_t size, const unsigned char* src, MSVCRT_size_t n) > +{ > + unsigned int pos = 0; Same problem as with strcpy_s patch, and in addition MSVCRT_size_t is used. Perhaps strcpy_s implementation should use MSVCRT_size_t as well. > +@ cdecl _mbsnbcpy_s(ptr str long) The spec entry has wrong number of parameters. -- Dmitry.

1 0

Re: [1/6] msvcrt: Implement strcpy_s
by Dmitry Timoshkov 15 Nov '07

15 Nov '07

"Stefan Dösinger" <stefan(a)codeweavers.com> wrote: > +int CDECL MSVCRT_strcpy_s( char* dst, size_t elem, const char* src ) > +{ > + unsigned int i; ... > + for(i = 0; i < elem; i++) Since 'i' is used to enumerate [0 < elem] and elem has type size_t, 'i' should have type size_t as well. > +static int (*pstrcpy_s)(char *dst, unsigned int len, const char *src); But here 'len' has type 'unsigned int' for some reason. -- Dmitry.

1 0

Re: taskmgr/perfpage.c simplification (RESEND)
by Alexandre Julliard 15 Nov '07

15 Nov '07

Gerald Pfeifer <gerald(a)pfeifer.com> writes: > Another simplifcation I'd like to ping if you don't mind. > > ---------- Forwarded message ---------- > From: Gerald Pfeifer <gerald(a)pfeifer.com> > To: <wine-patches(a)winehq.org> > Date: Thu, 1 Nov 2007 16:14:28 +0100 (CET) > Subject: taskmgr/perfpage.c simplification > > CpuUsage and CpuKernelUsage are of type ULONG, so it does not make > sense checking whether they are less than 0. The checks should most likely done in PerfDataGetProcessorUsage before converting to ULONG. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: Patch to make Wine prettier
by Alexander Nicolaysen Sørnes 15 Nov '07

15 Nov '07

On Sunday 11 November 2007 14:56:21 Fred . wrote: > Applications that run under Wine looks so legacy like Windows 95. > > This patch makes the software applications look a bit more modern and > prettier by changing the colors from the Windows 95-look to the > "Windows Classic"-look, the one in Windows ME/2000/2003. > > I have not yet tested the patch. Because I was unable to compile Wine > on my computer. > > The file should goto; > /wine/dlls/user32/resources/display.rc Great idea, but the patch does not change anything. Alexander N. Sørnes

1 0

server: avoid sending uninitialized bytes in break_thread()
by Dan Kegel 15 Nov '07

15 Nov '07

In break_thread(), generate_debug_event() seems to send the entire ExceptionInformation array regardless of how many entries are used. The smart thing to do would be to send only the used elements, but I couldn't figure out how to do that. So here's a quick fix that just zeroes that array before calling generate_debug_event(). Is there a better way? Found by running the Word Viewer 97 under valgrind: 000d: wait_debug_event() = 0 { pid=000a, tid=000b, wait=(nil), event={exception,{code=80000003,flags=0,rec=(nil),addr=0x40007f0,params={},first=1} } Syscall param writev(vector[1]) points to uninitialised byte(s) at 0x40007F2: (within /lib/ld-2.5.so) by 0x806AF20: call_req_handler (request.c:249) by 0x8056075: main_loop (fd.c:523) Address 0x451AAB8 is 24 bytes inside a block of size 88 alloc'd by 0x806A167: set_reply_data_size (request.c:200) by 0x80510A6: req_wait_debug_event (request.h:96) by 0x806AFC7: call_req_handler (request.c:286) by 0x8056075: main_loop (fd.c:523) BTW, here's how I ran word viewer under valgrind: wordview.sh: WINEDIR=$HOME/wine-git; VALGRIND_CMD="/usr/local/valgrind-3.2.3-wine/bin/valgrind --trace-children=yes --gen-suppressions=all --suppressions=$WINEDIR/tools/valgrind-suppressions --leak-check=full --num-callers=20 " WINE=$WINEDIR/wine wget http://downloads.pcworld.com/pub/new/business/word_processing/wordview.exe cp wordview.exe ~/.wine/drive_c/wvinstall.exe mv ~/.wine/dosdevices/z: ~/.wine/dosdevices/oldz cd ~/.wine/drive_c $VALGRIND_CMD $WINE wvinstall.exe /q cd ../.. mv ~/.wine/dosdevices/oldz ~/.wine/dosdevices/z:

1 1

Re: server: Make timer id allocation algorithm conform to the Windows one
by Uwe Bonnes 15 Nov '07

15 Nov '07

>>>>> "Dmitry" == Dmitry Timoshkov <dmitry(a)codeweavers.com> writes: Dmitry> Hello, this patch should fix the problem reported in the bug Dmitry> 10343. My test which calls SetTimer in an infinite loop shows Dmitry> that XP starts to allocate timer ids at 0x7fff and goes Dmitry> backwards to 0x101, then restarts from 0x7fff. Isn't this worth a test suite case? -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------

2 1

AppDB down =(
by John Klehm 15 Nov '07

15 Nov '07

Erm yeah the appdb is down... too many connections...sad times. John

3 2

Re: [4/4] msxml: Test for removeAttribute (try 2)
by Alistair Leslie-Hughes 14 Nov '07

14 Nov '07

Hi, Changelog: msxml: Test for removeAttribute Best Regards Alistair Leslie-Hughes

2 1

Re: [4/8] WineD3D: Depth stencil fixes
by H. Verbeet 14 Nov '07

14 Nov '07

On 13/11/2007, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > static void depth_copy(IWineD3DDevice *iface) { > IWineD3DDeviceImpl *This = (IWineD3DDeviceImpl *)iface; > - IWineD3DSurfaceImpl *depth_stencil = (IWineD3DSurfaceImpl *)This->depthStencilBuffer; > + IWineD3DSurfaceImpl *depth_stencil = (IWineD3DSurfaceImpl *)This->auto_depth_stencil_buffer; > > /* Only copy the depth buffer if there is one. */ I think this one should be changed to use stencilBufferTarget instead, although not in this patch.

2 2

Re: USER32: Patch 3 of 6: DlgDirList converts path specification to uppercase.
by Alexandre Julliard 14 Nov '07

14 Nov '07

Alex Villacís Lasso <a_villacis(a)palosanto.com> writes: > Observed while preparing conformance tests. The filepath (if valid) gets > converted to uppercase on return. > > Changelog: > * DlgDirList must convert path specification to uppercase. You should do that after the SetCurrentDirectory, uppercasing it first means we are much more likely to have to do a slow case-insensitive search on the whole path. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [4/4] msxml: Test for removeAttribute
by Robert Shearman 14 Nov '07

14 Nov '07

Alistair Leslie-Hughes wrote: > @@ -589,6 +589,10 @@ static void test_domnode( void ) > ok( !lstrcmpW(V_BSTR(&var), szstr1), "wrong attr value\n"); > VariantClear( &var ); > SysFreeString( str ); > + > + > + str = SysAllocString( szvr ); > + SysFreeString( str ); > > r = IXMLDOMElement_get_attributes( element, &map ); > ok( r == S_OK, "get_attributes returned wrong code\n"); > @@ -1077,7 +1081,7 @@ static void test_create(void) > ok( r == S_OK, "returns %08x\n", r ); > r = IXMLDOMNamedNodeMap_get_length( attr_map, &num ); > ok( r == S_OK, "returns %08x\n", r ); > - ok( num == 1, "num %ld\n", num ); > + ok( num == 1, "num %ld\n", num ); > IXMLDOMNamedNodeMap_Release( attr_map ); > VariantClear(&var); > r = IXMLDOMElement_getAttribute( element, name, &var ); > These two changes don't appear to be relevant to the test you're adding. -- Rob Shearman

1 0

Re: msxml: Implement cloneNode and correct insertBefore
by James Hawkins 14 Nov '07

14 Nov '07

On 11/13/07, Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> wrote: > > Hi James, > Thanks for the pointers. > > I resend the patches shortly. Is there an easy way to split a Diff up? via > git? More to the point changes to a single file (without have to backup it > up every change) > > I just get frustrated with try to keep lots of diffs around. > mailing list etiquette: make sure to bottom-post on this ML and CC wine-devel so others can share in the information exchange. To answer your question, you have to get into the git mindset. That is, you make one fix, commit the fix, write the next fix, etc. It's really a development change. You should know what to fix before you go in to fix it, not hack around until an app works. -- James Hawkins

3 2

Re: msxml: Implement Element removeAttribute
by Steven Edwards 14 Nov '07

14 Nov '07

On Nov 12, 2007 3:57 AM, Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> wrote: > > + > + TRACE("(%p)->%s\n", iface, debugstr_w(p)); > + // TODO Replace if Attribute has a Default value. Also another thing. No C++ style comments in Wine source code. Thanks Steven -- 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

1 0

Re: msxml: Implement cloneNode and correct insertBefore
by James Hawkins 14 Nov '07

14 Nov '07

On 11/12/07, Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> wrote: > Hi, > > Changelog: > InsertBefore is allowed to have outNewChild as NULL > Implement cloneNode > One fix per patch please. Also, please add tests for these fixes. > Best Regards > Alistair Leslie-Hughes > > diff --git a/dlls/msxml3/node.c b/dlls/msxml3/node.c > index 61d5ab2..86d7614 100644 > --- a/dlls/msxml3/node.c > +++ b/dlls/msxml3/node.c > @@ -373,7 +373,7 @@ static HRESULT WINAPI xmlnode_insertBefore( > > TRACE("(%p)->(%p,var,%p)\n",This,newChild,outNewChild); > > - if (!(newChild && outNewChild)) > + if (!newChild) > return E_INVALIDARG; > > switch(V_VT(&refChild)) > @@ -412,9 +412,13 @@ static HRESULT WINAPI xmlnode_insertBefore( > xmlAddChild(This->node, new_child_node); > } > > - IXMLDOMNode_Release(new); > IXMLDOMNode_AddRef(newChild); > - *outNewChild = newChild; > + IXMLDOMNode_Release(new); > + if(outNewChild) > + { > + *outNewChild = newChild; > + } > + > TRACE("ret S_OK\n"); > return S_OK; > } > @@ -517,8 +521,36 @@ static HRESULT WINAPI xmlnode_cloneNode( > VARIANT_BOOL deep, > IXMLDOMNode** cloneRoot) > { > - FIXME("\n"); > - return E_NOTIMPL; > + xmlnode *This = impl_from_IXMLDOMNode( iface ); > + xmlNodePtr pClone = NULL; > + IXMLDOMNode *pNode = NULL; > + > + TRACE("%p (%d)\n", This, deep); > + > + if(!cloneRoot) > + return E_INVALIDARG; > + > + pClone = xmlCopyNode(This->node, deep ? 1 : 0); > + if(pClone) > + { > + pClone->doc = This->node->doc; > + > + pNode = create_node(pClone); > + if(!pNode) > + { > + ERR("Copy failed\n"); > + return E_FAIL; > + } > + > + *cloneRoot = pNode; > + } > + else > + { > + ERR("Copy failed\n"); > + return E_FAIL; > + } > + > + return S_OK; > } > > static HRESULT WINAPI xmlnode_get_nodeTypeString( > @@ -832,7 +864,7 @@ static ULONG WINAPI Internal_Release( > if ( ref == 0 ) > { > if( This->node ) > - xmldoc_release( This->node->doc ); > + xmldoc_release( This->node->doc ); > Please don't make random whitespace changes. -- James Hawkins

1 0

svchost Patch Question
by Roy Shea 13 Nov '07

13 Nov '07

Howdy, I'm looking for advice on how to clean up the implementation of svchost. The latest version is available at: http://www.winehq.org/pipermail/wine-patches/2007-November/046467.html If the stopper for this patch is the lack of a DLL using it, I can resend svchost later this week when I have a stub implementation of BITS using svchost ready. Thanks for any pointers! -Roy

2 1

xrun handling in winealsa.drv
by Timofei V. Bondarenko 13 Nov '07

13 Nov '07

Hi. I've noticed the winealsa does not handle alsa XRUNs. In the case of XRUN application hangs. That's happened because return value of snd_pcm_avail_update() is not checked. For example: wavein.c 175 DWORD bytes; 180 /* read all the fragments accumulated so far */ 181 frames = snd_pcm_avail_update(wwi->pcm); 182 bytes = snd_pcm_frames_to_bytes(wwi->pcm, frames); waveout.c 476 DWORD availInQ; 479 availInQ = snd_pcm_avail_update(wwo->pcm); 482 if (availInQ > 0) { The snd_pcm_avail_update() returns a _signed_ type and it really does return negative number -EPIPE in the case of XRUN. More over, any return value of less than period_size may indicate XRUN state too. The snd_pcm_state() may help then. Th dsoutput.c and dscapture.c do exactly the same wrong thigs. BTW, using DWORD type causes problems. Following fragment is from wavein.c too: 177 DWORD read; 181 frames = snd_pcm_avail_update(wwi->pcm); 182 bytes = snd_pcm_frames_to_bytes(wwi->pcm, frames); 193 read = wwi->read(wwi->pcm, lpWaveHdr->lpData + 194 bytesRead = snd_pcm_frames_to_bytes(wwi->pcm, 197 if (bytesRead != (DWORD) -1) This is obviously will not work as expected. I hope someone will fix these issues :) Regards, Tim.

3 5

Grumble. Lost the Valgrind magic.
by Dan Kegel 13 Nov '07

13 Nov '07

I had to reinstall my main valgrind testing machine (I foolishly ran the vmware server install script, and it nuked my wireless), and since then, valgrind isn't giving me stack traces. I guess something's wrong in my recipe, but I'm not sure what, the vg-3.2.3 patch is being applied. Sigh. Is anybody else having luck building a valgrind that gives good stack traces for wine?

2 1

Re: rpcrt4: Implement RpcCancelThread for the ncacn_ip_tcp protocol sequence.
by Hans Leidekker 12 Nov '07

12 Nov '07

On Monday 12 November 2007 21:10:19 Robert Shearman wrote: > +static CRITICAL_SECTION_DEBUG threaddata_cs_debug = > +{ > + 0, 0, &uuid_cs, I smell a copy/paste artifact ;) -Hans

1 0

Too many WM_MOUSEACTIVATE
by Peter Åstrand 12 Nov '07

12 Nov '07

I have lot's of problems with WM_MOUSEACTIVATE messages in "Magna Cura". The effekt is that certain popup dialog doesn't work: After selecting something, say a date, the whole "form" is cleared. After much debugging, I've found out that WM_MOUSEACTIVATE is the problem. If I explicitly stop WM_MOUSEACTIVATE from propagating from dialogs to parents, then things works OK. I do this by addding: case WM_MOUSEACTIVATE: return MA_ACTIVATE; in DefDlgProc{A,W}. I wonder how many things that will brake with a change like this.... Just a long shot, but has anyone else seen anything similiar to this? (It turns out that there are multiple differences compared to Windows; one with two many messages generated when the popup is created, another with too few messages when changing tab in a SysTabControl32 (but zero messages works fine as well.) Rgds, --- Peter Åstrand ThinLinc Chief Developer Cendio AB http://www.cendio.se Wallenbergs gata 4 583 30 Linköping Phone: +46-13-21 46 00 ThinLinc User Group 2007 Höstkonferens --- 15-16 november, Linköping. Program och anmälan: http://www.thinlincusergroup.se/aktiviteter

1 0

Problem with wine + Dreamweaver + VNC
by Francisco José Gallego Durán 12 Nov '07

12 Nov '07

Hello, I am trying to run Dreamweaver.exe and Photoshop.exe in a remote gnome-session through vnc4server. When I try to run it I get the following message: Xlib: extension "XFree86-DRI" missing on display ": 3.0". X Error of failed request: BadValue (integer parameter out of range for operation) Major opcode of failed request: 148 (GLX) Minor opcode of failed request: 3 (X_GLXCreateContext) Value in failed request: 0x24 Serial number of failed request: 11 Current serial number in output stream: 12 Might it be a kind of a bug? I am running wine 0.9.49, Debian etch with latest updates, and vnc4server. Thank you for your help. Francisco Gallego.

1 0

Re: wininet: update test to pass on newer versions of windows
by Alexandre Julliard 12 Nov '07

12 Nov '07

"Lei Zhang" <thestig(a)google.com> writes: > + > + /* sz is 14 on XP SP1 and beyond, 30 on XP and before */ > + todo_wine ok( sz == 14, "sz wrong\n"); The tests have to pass on all Windows versions. If some return 30 and some return 14 the test must accept both. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: widl [2/9]: Add an option to use Wine exception handlers.
by Alexandre Julliard 12 Nov '07

12 Nov '07

Dan Hipschman <dsh(a)linux.ucla.edu> writes: > This patch adds a --wine-seh option to widl to use Wine's SEH instead of > the Windows one. This should not be a widl option. widl should generate a single version of the C code that works on all compilers, using appropriate macros and #ifdefs. Also you shouldn't use the wine/exception.h macros, since you are generating the code you can output the code directly, in a much more straightforward way than the convoluted __TRY macros. -- Alexandre Julliard julliard(a)winehq.org

1 0

How can we handle undcomented shell functionsAW, that MS declared now (WSTR version only)
by Detlef Riekenberg 12 Nov '07

12 Nov '07

Hi There are a some undocumented Functions in shell32, which have one entrypoint that use STR in win9x, but WSTR in NT: - dlls/shell32/undocshell.h - http://www.geocities.com/SiliconValley/4942/paths.html Some of them are now declared without A/W suffix (include/shobj.h from SDK 2003), but since win9x is no longer supported by MS, it's only for the Wide version. Examples: PathProcessCommend, PathResolve, PathQualify Question for include/shlobj.h: What is the correct way to declare such functions? In undocshell.h, we use functionAW with LPVOID for the first two, but for PathQualify, we use: // BOOL WINAPI PathQualifyA(LPCSTR path); BOOL WINAPI PathQualifyW(LPCWSTR path); #define PathQualify WINELIB_NAME_AW(PathQualify) BOOL WINAPI PathQualifyAW(LPCVOID path); // And PathResolve in exported by name (see shell32.spec) ############################################################# Next question for shell32.spec: Are we fixed to "ptr" for such functions? ("str", "wstr" and "long" are all used and need to be fixed) Is it possible / useful to add support for "awstr" (or a similar name)? What about support for UNICODE-Strings ("ustr")? -- By by ... Detlef

2 1

Re: cmdlgtst: Rename to dialogtest
by Dmitry Timoshkov 12 Nov '07

12 Nov '07

"Detlef Riekenberg" <wine.dev(a)web.de> wrote: > Windows has a lot more Dialogs in shell32, user32, kernel32 and various > other dlls. > > This is only a rename, the only code that changed is the name of the > debug-channel. IMHO it's better to remove programs/cmdlgtst altogether, it doesn't really test anything, and this is not a place for tests anyway. -- Dmitry.

2 1

Patch Follow up
by Alistair Leslie-Hughes 12 Nov '07

12 Nov '07

Hi, Are there any reason they following patches where missed? Patch 1 of 2 - Return E_INVALIDARG on bad parameter Patch 2 of 2 - Enable Test for Bad Argument / Fix Memory leak clusapi: Corrected value in GetNodeClusterState Best regards Alistair Leslie-Hughes

2 1

Valgrind results for Oct 9: still lots of hangs
by Dan Kegel 12 Nov '07

12 Nov '07

Results for today's git (plus my dde patch) are online at http://kegel.com/wine/valgrind/20071109/ ole32/tests/marshall.c has always been hanging when run under valgrind. In the last two days, the following additional tests hang under valgrind: mshtml/tests/dom.c mshtml/tests/htmldoc.c ole32/tests/marshal.c shdocvw/tests/webbrowser.c I can reproduce one hang with cd dlls/mshtml/tests make testclean wineserver -k RUNTEST_USE_VALGRIND=1 make test (with the wine-vg-utils.patch I provide. Make by itself does not behave well under valgrind, so you can't just do "valgrind make test"). Sadly, I don't have time to do a regression test, but it's pretty easy to repeat my results. See http://wiki.winehq.org/Wine_and_Valgrind It'd be nice if somebody else verified that these hangs occur for them too, that they didn't occur with wine-0.9.48, and maybe did a regression test to find out when they started... - Dan

3 4

Re: OLEDLG: use wide strings
by Dmitry Timoshkov 12 Nov '07

12 Nov '07

"Vitaly Lipatov" <lav(a)etersoft.ru> wrote: > Changelog: > - fix buffers size for registry functions > - use unicode registry functions > - do not requires ntdll anymore (remove RtlMultiByteToUnicodeN using) This patch is extremely buggy, it passes wrong buffer sizes to various APIs in many places. -- Dmitry.

1 0

%fs, %gs on AROS hosted
by Staf Verhaegen 11 Nov '07

11 Nov '07

Hello wine developers, I am an AROS developer (http://www.aros.org); that is an open source implementation of the old AmigaOS. It has also the possiblity to run hosted on other OSes and is thus very similar to wine. I know that this group is for wine posts but I think I could reuse much of the knowledge gained by the wine developers for AROS. Recently there has been discussion about the use of the %fs and %gs segment for AROS purposes and thus also for the hosted version. As wine also has to handle these registers, I am interested to know how wine handles these registers. Or doesn't wine have to take these registers into account ? I did download the 0.9.48 source code of wine and I only did find some places where a proper saving and restoring of these registers was done. Is there something I am overlooking ? Currently wine seems to run on the following platforms: Linux version 2.0.36 or above FreeBSD 6.2 or later Solaris x86 2.5 or later NetBSD-current Mac OS X 10.4 or later I found almost no OS specific code for each of the architectures in the source. Are there special precautions to be taken in one of these OSes to be able to use the segment register in user land code ? My last question is related to an AROS version hosted on Windows. Would Windows put any restrictions on the use of the segment registers for this purpose. Is there a difference between 32bit and 64bit ? thanks, Staf.

2 5

What the hell is up with searching google for wine-devel and wine-patches posts?
by Steven Edwards 10 Nov '07

10 Nov '07

Hi, I guess this issue is more directed at Dan and the rest of the google guys we have lurking here but has anyone else noticed how hard it is to do a google search for discussions on wine-devel or look for patches that have been submitted? I used to just be able to search as follows wine-patches Steven Edwards and see every patch I ever submitted to wine. Ditto for most threads on wine-devel. This is really getting to be a pain for me because I need to be able to search for patches that might not have been merged in but still have merit for some of my customers. Thanks -- 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

3 3

Linux/Windows app connection, how?
by Joshua Scholar 10 Nov '07

10 Nov '07

I really want to have some way to communicate between a Linux program and a Windows program running under WINE. The connection doesn't need to be high speed, a stream is fine - I'm just sending some unicode text. I'm writing both programs myself, so I can implement this in the easiest way, though if there was some way to create and call a Windows COM object from Linux, that would be the most direct solution... But reading what I can about WINE, I'm guessing that the only easy thing to do is to write a Windows server application with winegcc and either connect to it through a socket - or to have it spawn a Linux application that it has a stream to. But I need some sample code to know how to do this, since I don't have too much time to experiment. Just to make it clear why I'm doing this - I have some wonderful TTS voices (Neospeech's SAPI5 Japanese voices) installed in WINE and I need to be able to invoke them from a Linux program. Thanks in advance for any pointers, Joshua Scholar

6 5

alsa->pulseaudio->alsa directsound hang
by Al Tobey 10 Nov '07

10 Nov '07

I spent some time this evening playing around with trying to get wine sound output to go through pulseaudio on my fc8/rawhide (x86_64) box. I tried wine 0.9.43 (built), 0.9.47 (fc8 rpm), and Crossover 6.2.0. I was testing with Steam/TF2, but found that the same issue comes up with a install of Winamp Lite using its default DirectSound output plugin. If I configure nothing and let wine grab the soundcard (I have an Audigy, so it allows multiple streams), it works fine. If I configure it to use pulseaudio through the alsa plugin, it hangs. I see the stream show up in the pulseaudio volume control, but nothing plays and winamp's meter and progress bar are inactive. The "Test Sound" button in winecfg works and shows up in pavucontrol as a mono stream. I've tried numerous variations on emulation and bitrates with no change. I didn't have any luck getting the ESD support (which would go straight to PA) to even show up in winecfg, even after forcing it in regedit. I probably missed something obvious. I'll do some more debugging tomorrow, but I thought I'd throw my config out there for others to experiment with and ridicule. There isn't much out on Google about this kind of config yet, but I expect there will be more requests as the popularity of pulseaudio goes up, especially with the release of fc8. I think the latest Ubuntu has it enabled by default, too. Did anybody actually write a patch to directly support pulseaudio or was the earlier discussion devoid of code? My registry mods and .asoundrc for using the alsa/pulseaudio plugin are attached. Getting pulseaudio is left as an exercise to the reader ;) Thanks, -Al

3 4

Re: USER32: Patch 1 of 6: Fix returned value of LB_DIR
by James Hawkins 10 Nov '07

10 Nov '07

On 11/10/07, Alex Villacís Lasso <a_villacis(a)palosanto.com> wrote: > Tests reveal that (on an unsorted listbox), LB_DIR returns the value of the > highest index inserted, as if it had sorted the list of files before starting > the insertion. This patch attempts to replicate this behavior. > > Changelog: > * Make LB_DIR return an index value as if it had inserted filenames in sorted > order. > Tests? Tests that aren't in the test suite that should be to verify the patch is correct? -- James Hawkins

1 1

CryptAcquireContext Failure, default/Type 024 requested
by Rob Seger 09 Nov '07

09 Nov '07

Wine currently supports types 001 and 012, but I have a new game attempting to use type 024 (Software\\Microsoft\\Cryptography\\Defaults \\Provider Types\\Type 024). I've googled, as well as searched MSDN, for a reference as to what the crypt provider type 024 even is with no success. Does anyone know? Looking through the advapi32.dll code, this looks like it might be a serious undertaking to implement. Is that estimation accurate? Does anyone have any tips or pointers about implementing an entire CSP in wine? I appreciate any help, keep up the good work all! Rob

3 7

Re: [14/15] WineD3D: Get rid of the conditionals in shader_glsl_compare
by Mirek Slugeň 09 Nov '07

09 Nov '07

2 2

winetest.exe broken
by Stefan Leichter 09 Nov '07

09 Nov '07

Hello, as you all may have noticed winetest.exe does not longer build (see http://quisquiliae.physics.gla.ac.uk/crossbuild-logs/20071109-11:05.txt ). The patch: 65d865d96313d96889bf8cbe88c931b5e1e9bb74 is first bad commit commit 65d865d96313d96889bf8cbe88c931b5e1e9bb74 Author: Alexandre Julliard <julliard(a)winehq.org> Date: Thu Nov 8 12:21:10 2007 +0100 libwine: Refuse to use a WINEPREFIX dir owned by a different user. :040000 040000 1bf2b5da793741040732f222ce4a90ebd3f54b74 ee36bfdcf5d65ec8bb1c8ee478e74e0af8679039 M libs broke the build. Can someone please fix this. Thanks Stefan

3 2

Re: [6/15] WineD3D: Replace the position fixup mul-add-add with a MAD
by Mirek Slugeň 09 Nov '07

09 Nov '07

3 3

notepad crashes in VNC @ 16bpp
by Francois Gouget 09 Nov '07

09 Nov '07

So if I start a VNC server with the following command: vncserver :1 -depth 16 -geometry 1024x768 Then notepad crashes on startup while initializing OpenGL. More precisely the crash happens in X11DRV_WineGL_InitOpenglInfo() on line 304 of winex11.drv/opengl.c. The reason for the crash is an OpenGL/X11 assertion: glxcmds.c:343: CreateContext: Assertion "mode != ((void *)0)" failed This seems like a VNC/OpenGL/X11 bug because Wine works fine if VNC is started in 15bpp or 24bpp. Still, the problem is that as this is part of the winex11 initialization it impacts all applications (including the DIB conformance tests). Does anyone see a way to avoid the crash? (maybe detect VNC before that call, but how?) Note: I'm using VNC 4.1.1+X4.3.0-2. Here's the xdpyinfo output in that mode: name of display: :1.0 version number: 11.0 vendor string: The XFree86 Project, Inc vendor release number: 40300000 XFree86 version: 4.3.0 maximum request size: 4194300 bytes motion buffer size: 256 bitmap unit, bit order, padding: 32, LSBFirst, 32 image byte order: LSBFirst number of supported pixmap formats: 2 supported pixmap formats: depth 1, bits_per_pixel 1, scanline_pad 32 depth 16, bits_per_pixel 16, scanline_pad 32 keycode range: minimum 8, maximum 255 focus: window 0x400013, revert to Parent number of extensions: 23 BIG-REQUESTS DEC-XTRAP DOUBLE-BUFFER Extended-Visual-Information FontCache GLX LBX MIT-SCREEN-SAVER MIT-SHM MIT-SUNDRY-NONSTANDARD RECORD SECURITY SGI-GLX SHAPE SYNC TOG-CUP VNC-EXTENSION X-Resource XC-APPGROUP XC-MISC XFree86-Bigfont XTEST XVideo default screen number: 0 number of screens: 1 screen #0: dimensions: 1024x768 pixels (260x195 millimeters) resolution: 100x100 dots per inch depths (2): 1, 16 root window id: 0x2e depth of root window: 16 planes number of colormaps: minimum 1, maximum 1 default colormap: 0x21 default number of colormap cells: 64 preallocated pixels: black 0, white 65535 options: backing-store YES, save-unders YES largest cursor: 1024x768 current input event mask: 0xd83f3f KeyPressMask KeyReleaseMask ButtonPressMask ButtonReleaseMask EnterWindowMask LeaveWindowMask Button1MotionMask Button2MotionMask Button3MotionMask Button4MotionMask Button5MotionMask ButtonMotionMask SubstructureNotifyMask SubstructureRedirectMask PropertyChangeMask ColormapChangeMask number of visuals: 8 default visual id: 0x24 visual: visual id: 0x24 class: TrueColor depth: 16 planes available colormap entries: 64 per subfield red, green, blue masks: 0xf800, 0x7e0, 0x1f significant bits in color specification: 8 bits visual: visual id: 0x25 class: TrueColor depth: 16 planes available colormap entries: 64 per subfield red, green, blue masks: 0x3f, 0x7c0, 0xf800 significant bits in color specification: 8 bits visual: visual id: 0x26 class: TrueColor depth: 16 planes available colormap entries: 64 per subfield red, green, blue masks: 0x3f, 0x7c0, 0xf800 significant bits in color specification: 8 bits visual: visual id: 0x27 class: TrueColor depth: 16 planes available colormap entries: 64 per subfield red, green, blue masks: 0x3f, 0x7c0, 0xf800 significant bits in color specification: 8 bits visual: visual id: 0x28 class: DirectColor depth: 16 planes available colormap entries: 64 per subfield red, green, blue masks: 0x3f, 0x7c0, 0xf800 significant bits in color specification: 8 bits visual: visual id: 0x29 class: DirectColor depth: 16 planes available colormap entries: 64 per subfield red, green, blue masks: 0x3f, 0x7c0, 0xf800 significant bits in color specification: 8 bits visual: visual id: 0x2a class: DirectColor depth: 16 planes available colormap entries: 64 per subfield red, green, blue masks: 0x3f, 0x7c0, 0xf800 significant bits in color specification: 8 bits visual: visual id: 0x2b class: DirectColor depth: 16 planes available colormap entries: 64 per subfield red, green, blue masks: 0x3f, 0x7c0, 0xf800 significant bits in color specification: 8 bits -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ 1 + e ^ ( i * pi ) = 0

1 0

Re: wininet: make a test pass on all versions of windows
by Alexandre Julliard 09 Nov '07

09 Nov '07

"Lei Zhang" <thestig(a)google.com> writes: > + > + { > + OSVERSIONINFOEX foo; > + BOOL r; > + memset(&foo, 0, sizeof(foo)); > + foo.dwOSVersionInfoSize = sizeof(foo); > + r = GetVersionEx((OSVERSIONINFO *) &foo); > + if (!r) > + { > + skip("could not get windows version\n"); > + return; > + } > + > + is_sp1 = -1; > + if (foo.dwMajorVersion > 5) > + is_sp1 = 1; > + else if (foo.dwMajorVersion == 5) > + { > + if (foo.dwMinorVersion > 1) > + is_sp1 = 1; > + else if (foo.dwMinorVersion == 1) > + { > + if (foo.wServicePackMajor > 1) > + is_sp1 = 1; > + else if (foo.wServicePackMajor == 1) > + is_sp1 = 0; > + } > + } Please don't add version checks in tests. -- Alexandre Julliard julliard(a)winehq.org

2 2

None
by Colin M Inglis 09 Nov '07

09 Nov '07

Hi! I'm trying to install Wine for a client, a Mac house, who run some software only supported under Windows or Classic. With the recent introduction of Intel Macs (that no longer support Classic) into the business, the Windows version is, unfortunately, their only upgrade path. I have C development experience and am familiar with compiling C applications under unix, although that was some years ago and processes seem to have changed somewhat. In particular, I am unfamiliar with the 'configure' executable described in your installation manual, which is raising an error. Could you please look briefly at the attached config.log? I would be interested to know if this is an issue to which I can attend, or whether it is a deficiency in the configuration process itself. I would also appreciate a workaround, if possible. Thanks in anticipation of your assistance. Regards, ////////// // // / / Colin M Inglis BInfoTech&Comm //////////// // // / / Software Application Development /// //\//\\ infinite 15 Jasmine St, Strathpine, Q, Australia // // // // Snailmail PO Box 2188 Strathpine Q 4500 /// \\//\// interface email manager(a)infinitei.com.au ///////// // // / / web http://www.infinitei.com.au ////// // // / / Ph/Fax +61 7 3881 3494 ABN: 40 030 880 342 Mobile +61 414 388 134 "infinite interface programmes programmes users use"

5 4

Valgrind results for Oct 8: lots of hangs?!
by Dan Kegel 09 Nov '07

09 Nov '07

Data online at http://kegel.com/wine/valgrind/20071108/ For some reason, today I had to do 'wineserver -k' about five times (yesterday and the day before only once). And one of those times was during a d3d9 test which took over / hung the screen; not even ctl-alt-backspace helped, I actually had to reboot (I didn't try ssh'ing in). Running the tests was like pulling teeth. That said, here are the changes in the list of bad tests; - means test didn't show any valgrind errors today but did yesterday, + means it did show valgrind errors today but not yesterday. -vg-d3d9_visual.txt (might be one I had to reboot for) -vg-iphlpapi_iphlpapi.txt -vg-mlang_mlang.txt +vg-mshtml_dom.txt +vg-mshtml_htmldoc.txt +vg-shdocvw_webbrowser.txt -vg-wininet_ftp.txt -- Wine for Windows ISVs: http://kegel.com/wine/isv

2 1

re: #winehq admin troubles
by Dan Kegel 09 Nov '07

09 Nov '07

Luke Bratch wrote: > After seeing Vitamin deal with people for a long time > now, I can say I totally agree with how he does > things. Technically, he's spot on. It's his bedside manner that is broken. We really don't want him on the front line of support, somebody else should do that. But once the newbies are filtered out, he's great. - Dan -- Wine for Windows ISVs: http://kegel.com/wine/isv

5 5

Re: wininet: check the return value from HeapAlloc in a test
by Alexandre Julliard 09 Nov '07

09 Nov '07

"Lei Zhang" <thestig(a)google.com> writes: > We should skip a test if HeapAlloc() fails. There's no need to clutter up the tests with allocations checks, if we are out of memory pretty much everything in the test is going to fail anyway. -- Alexandre Julliard julliard(a)winehq.org

1 0

valgrind results updated
by Dan Kegel 09 Nov '07

09 Nov '07

I've updated http://kegel.com/wine/valgrind/ Today's results are at http://kegel.com/wine/valgrind/20071103/ The only difference from yesterday is I've added suppressions for the most common memory leaks, on the theory that it's easier to see the module-specific leaks that way. (I haven't looked to see if they're false positives, so I named them todo* to remind us to look at them again later.) As always, there are leaks and uninitialized memory references aplenty! Step right up, no waiting...

2 2

Successful Virtualization startup looking for engineers with system level Windows and Linux experience
by Richard Offer 09 Nov '07

09 Nov '07

We are a well-funded software startup in the virtualization space that is looking to hire several engineers with system level experience in Windows and Linux - ideally in combination. I can't tell you what you'd be doing but you can be re-assured that since we are long-term open source supporters any modifications to Wine or WineLib that you make will, of course, be contributed back to the community. We are particularly interested in experiences in the areas of * Registry modification, manipulation and creation * Windows internals or system level user-space. Device drivers aren't very interesting to us. * Reverse engineering skills * Installable file systems * Windows server technologies such as IIS, AD, 2003, Longhorn etc * Virtualization technologies such as VMware, AppStream, Softricity and Xen. If you or anyone you know have an interest and are eligible for working an a small but rapidly growing company in Silicon Valley on things that are novel and technically challenging, please send a current resume to engr-jobs(a)fastscale.com - including WSE in the subject line of any emails. For more information on the company, our technology and shipping products please visit us at http://www.fastscale.com/ Yours truly, Richard Offer

1 0

Re: wininet: skip strcmp() in a test if the returned len is wrong
by Dmitry Timoshkov 09 Nov '07

09 Nov '07

"Lei Zhang" <thestig(a)google.com> wrote: > We'll skip doing strcmp() on a potentially uninitialized buffer if the > returned length is wrong. > - todo_wine ok(!strcmp(useragent,buffer),"Got wrong user agent string %s instead of %s\n",buffer,useragent); > - todo_wine ok(len == strlen(useragent),"Got wrong user agent length %d instead of %d\n",len,lstrlenA(useragent)); > + len_ok = (len == strlen(useragent)); > + todo_wine ok(len_ok,"Got wrong user agent length %d instead of %d\n",len,lstrlenA(useragent)); > + if (len_ok) > + todo_wine ok(!strcmp(useragent,buffer),"Got wrong user agent string %s instead of %s\n",buffer,useragent); Probably a better solution would be to actually check the return value of InternetQueryOption before touching the returned buffers. -- Dmitry.

1 0

valgrind/urlmon & http/variable question
by Misha Koshelev 09 Nov '07

09 Nov '07

> Hi Dan et al., > > So I just wanted to let you know that I have looked somewhat at one (the only as far as I saw) of the valgrind results related to HttpProtocol in urlmon, and basically it involves passing a pointer to a PROTOCOLDATA from a local variable inside a function to the ProtocolSink call which ends up > sending a message with this pointer. When this pointer is processed later, since it is a pointer to a local variable which (presumably) exists only on the stack, it is not really valid. > > Now this is quite easy to "fix" with a five-line patch that HeapAlloc's the proper structures. However, I was trying to figure out how native does this to copy it, and kind of ran into a dilemma. Namely, doing a simple patch like shown below (this probably won't really be a diff as I copy/pasted it > into WebMail which will probably mess it up, but it should give a general idea) and then doing a WINEDLLOVERRIDES with native urlmon and wininet and +relay WINEDEBUG I can see that the the PROTOCOLDATA pointers are pointing to stuff like 0x7d7ad790 and 0x7d7ad820. Now, much to my suprise, these > addresses, nor addresses in the proximate range (approximate) to that address don't seem to be part of any other relay calls, including any calls to any sort of allocation functions. Now this leaves me a little stumped ,as I can only see a few ways to properly implement these PROTOCOLDATA pointers > (without a memory leak): > > 1. A static PROTOCOLDATA variable (local or global). In this case, the address I see for native should never change I believe, but also this would not be correct as potentially you could have more than one HttpProtocol using the same structure. > > 2. Dynamic allocation of the PROTOCOLDATA structure. This seems like the best option to me, but native either does not seem to be using it or allocates it in some way in which the address or even partial versions of the address does not appear in any allocation functions (or really anywhere) in the > +relay log. Any ideas on what allocation functions I could be missing here? > > 3. Have one PROTOCOLDATA structure per IInternetProtocol implementation of http. This seems reasonable too, except I am also printing out the address of the IInternetProtocol structure and clearly PROTOCOLDATA is not in an address range anywhere near this structure, and it is never allocated so it > can't be a pointer within that structure. > > So anyhow I am a little stuck on the "proper" fix for this. Any suggestiosn would be appreciated. > > Thanks, > > Misha Hmm... so it looks like upon investigation the address ranges for PROTOCOLDATA with antive dlls are actually suspiciously similar to those I see in the current local variable implementation... which makes me think that the implementation is dissimilar (and thus leaky) in the downstream ProtocolSink implementation in binding.c, which will be a little harder to test... Misha

1 0

Problem with client manually starting services
by Roy Shea 08 Nov '07

08 Nov '07

Howdy All, I'm developing missing services in Wine and am running into problems with how Wine starts services. In Windows a service can be started using "net start <service name>", assuming that the service is properly added to the registry. In Wine it appears that the user must manually start the service before calling "net start <service name>", or the service fails to start. This is blocking me from developing an svchost wrapped DLL. I guess its not surprising that the work around of first manually starting a service fails for a DLL with no main function :-) Dan Kegel mentioned a recollection of a known bug in Wine along these lines where, as a work around, clients are manually starting servers. Anyone know more details on this problem? Any pointers to old emails or knowledge of this problem would be a big help I start digging through WINEDEBUG logs. Thanks! -Roy

2 4

Re: [1/6] WineD3D: Only glBindAttribLocation used attributes
by Mirek Slugeň 08 Nov '07

08 Nov '07

2 6

Re: WineD3D: flush gl buffers are blt to drawable [attempt2]
by Stefan Dösinger 08 Nov '07

08 Nov '07

Am Donnerstag, 8. November 2007 23:40:18 schrieb Roderick Colenbrander: > Hi, > > This patch makes sure the gl buffer is flushed after blitting to the > screen. This fixes bug 9826 where traces of old frames appeared due to > this. You really shouldn't do this unconditionally, it is only needed if two contexts are drawing to the same drawable. A suggestion would be to mark the first use of a 2nd thread in a device in FindContext, and call the glFlush only if happened. The same glFlush is needed in all other places that draw something to the drawable: drawprim, clear and BltOverride come to my mind.

1 0

Typing Chinese Online On FireFox
by Jiang,Yi 08 Nov '07

08 Nov '07

I developped a site for Typing Chinese Online: http://www.typingChinese.com. It works on FireFox, IE, Mac. On test, I found FireFox is the fastest and used much less memory! Regards to all, Jiang,Yi

1 0

Develop .h for Wine by looking Microsoft Platform SDK
by Pau Garcia i Quiles 08 Nov '07

08 Nov '07

Hello, How legal (if at all) would be to develop a header file by looking reading the corresponding header file from the Microsoft Platform SDK and adapting it to gcc (removing MSVC stuff, etc)? I'm interested in supporting annotations[*] and before I start deducting everything from MSDN documentation, I'd like to know if there is a fast track. [*] More information: http://msdn2.microsoft.com/en-us/library/aa383701.aspx http://msdn2.microsoft.com/en-us/library/ms235402(VS.80).aspx Thank you. -- Pau Garcia i Quiles http://www.elpauer.org (Due to my workload, I may need 10 days to answer)

4 4

valgrind/urlmon & http/variable question
by Koshelev, Misha Vladislavo 08 Nov '07

08 Nov '07

Hi Dan et al., So I just wanted to let you know that I have looked somewhat at one (the only as far as I saw) of the valgrind results related to HttpProtocol in urlmon, and basically it involves passing a pointer to a PROTOCOLDATA from a local variable inside a function to the ProtocolSink call which ends up sending a message with this pointer. When this pointer is processed later, since it is a pointer to a local variable which (presumably) exists only on the stack, it is not really valid. Now this is quite easy to "fix" with a five-line patch that HeapAlloc's the proper structures. However, I was trying to figure out how native does this to copy it, and kind of ran into a dilemma. Namely, doing a simple patch like shown below (this probably won't really be a diff as I copy/pasted it into WebMail which will probably mess it up, but it should give a general idea) and then doing a WINEDLLOVERRIDES with native urlmon and wininet and +relay WINEDEBUG I can see that the the PROTOCOLDATA pointers are pointing to stuff like 0x7d7ad790 and 0x7d7ad820. Now, much to my suprise, these addresses, nor addresses in the proximate range (approximate) to that address don't seem to be part of any other relay calls, including any calls to any sort of allocation functions. Now this leaves me a little stumped ,as I can only see a few ways to properly implement these PROTOCOLDATA pointers (without a memory leak): 1. A static PROTOCOLDATA variable (local or global). In this case, the address I see for native should never change I believe, but also this would not be correct as potentially you could have more than one HttpProtocol using the same structure. 2. Dynamic allocation of the PROTOCOLDATA structure. This seems like the best option to me, but native either does not seem to be using it or allocates it in some way in which the address or even partial versions of the address does not appear in any allocation functions (or really anywhere) in the +relay log. Any ideas on what allocation functions I could be missing here? 3. Have one PROTOCOLDATA structure per IInternetProtocol implementation of http. This seems reasonable too, except I am also printing out the address of the IInternetProtocol structure and clearly PROTOCOLDATA is not in an address range anywhere near this structure, and it is never allocated so it can't be a pointer within that structure. So anyhow I am a little stuck on the "proper" fix for this. Any suggestiosn would be appreciated. Thanks, Misha --- diff --git a/dlls/urlmon/tests/protocol.c b/dlls/urlmon/tests/protocol.c index 00335b3..5b3579b 100644 --- a/dlls/urlmon/tests/protocol.c +++ b/dlls/urlmon/tests/protocol.c @@ -295,7 +295,9 @@ static HRESULT WINAPI ProtocolSink_Switch(IInternetProtocolSink *iface, PROTOCOL } SET_EXPECT(ReportData); + printf("!!! Continue %p -->\n", pProtocolData); hres = IInternetProtocol_Continue(http_protocol, pProtocolData); + printf("!!! Continue <--\n"); ok(hres == S_OK, "Continue failed: %08x\n", hres); CHECK_CALLED(ReportData); @@ -1255,6 +1257,7 @@ static void test_http_protocol_url(LPCWSTR url, BOOL is_first) hres = IClassFactory_CreateInstance(factory, NULL, &IID_IInternetProtocol, (void**)&http_protocol); ok(hres == S_OK, "Could not get IInternetProtocol: %08x\n", hres); + printf("!!! IInternetProtocol: %p %p\n", http_protocol, http_protocol+sizeof(IInternetProtocol)); if(SUCCEEDED(hres)) { BYTE buf[3600]; DWORD cb;

1 0

Re: d3dxof: return correct value for IDirectXFileDataImpl_GetType
by Roderick Colenbrander 08 Nov '07

08 Nov '07

> IDirectXFileDataImpl_GetType should return DXFILEERR_BADVALUE on failure, > not S_FALSE > > This patch fixes bug 10355 > > http://msdn2.microsoft.com/en-us/library/bb174490.aspx > Sure this could fix that bug but it not a real fix. D3dxof.dll is mostly a stub and its real functionality (loading and parsing of .x files) is missing. This GetType function is also related to the parsing of the .x file. In short this file contains data and some 'templates' for describing the data. This function returns the guid of that template. Returning a failure is not really an option in my opion as it won't get the app working. For the time being you should a native d3dxof.dll should be used or the rest of the .x file parsing should be added. Roderick -- Psssst! Schon vom neuen GMX MultiMessenger gehört? Der kann`s mit allen: http://www.gmx.net/de/go/multimessenger

2 1

Bugs in dlls/wsock32/socket.c
by Gerald Pfeifer 08 Nov '07

08 Nov '07

dlls/wsock32/socket.c has the following code: DWORD routeTableSize, numRoutes, ndx, ret; numRoutes = min(routeTableSize - sizeof(MIB_IPFORWARDTABLE), 0) / sizeof(MIB_IPFORWARDROW) + 1; The problem here is that both routeTableSize and sizeof(...) are of an unsigned type, so their difference is also unsigned. Which means that the minimum between that difference and 0 always will be 0, the result of the division will thus always be zero, and the result of the entire expressions is a constant 1. I have a hunch this is not the intendend purpose of this code, is it? Gerald -- Gerald (Jerry) Pfeifer gerald(a)pfeifer.com http://www.pfeifer.com/gerald/

2 2

Re: #winehq admin troubles
by feba thatl 08 Nov '07

08 Nov '07

Tom, you can try to guilt me, but I don't care. There are two sides to this. Either you think that #winehq is a support channel for end users, and should if not cater to them, at least treat them with some respect until they've proved that they deserve none, or you think that #winehq is a channel where people can go for help if some one with an @ isn't in a mood to ban them. If you can honestly say that you fully agree with the latter, and would have no problem getting banned from a random channel when you were seeking help, I don't know what to tell you. If, however, you think the first thing, we're on the same side. I don't think anyone here doesn't want to see #winehq, and the entire project really, become a better place. There might be one or two people exploiting this for personal reasons, but I cannot control that. I posted this in the hopes that #winehq would be made a better place, not for me, but for everyone. Including vitamin and the rest of the ops. I have a question for you, what offends you so much about my comments that you are not only refusing to help me, but also trying to pin me with the blame for your refusal to help others? Is it because you feel that your experience with wine gives you to the right to mistreat users, or because you are offended by what a few other people said in a thread that I started? Do you not want #winehq to be a better place? Or do you not realize the actual point of what I'm saying?

1 0

Re: svchost: Implementation of svchost
by Robert Shearman 08 Nov '07

08 Nov '07

Roy Shea wrote: > Changelog: > * Implementation of svchost > You've got a lot of global functions in this patch that should be static. -- Rob Shearman

1 0

Re: wininet: Fully initialize a variable
by Alexandre Julliard 08 Nov '07

08 Nov '07

"Juan Lang" <juan.lang(a)gmail.com> writes: > static void COOKIE_crackUrlSimple(LPCWSTR lpszUrl, LPWSTR hostName, int hostNameLen, LPWSTR path, int pathLen) > { > - URL_COMPONENTSW UrlComponents; > + URL_COMPONENTSW UrlComponents = { 0 }; > > - UrlComponents.lpszExtraInfo = NULL; > - UrlComponents.lpszPassword = NULL; > - UrlComponents.lpszScheme = NULL; > UrlComponents.lpszUrlPath = path; > - UrlComponents.lpszUserName = NULL; > UrlComponents.lpszHostName = hostName; > UrlComponents.dwHostNameLength = hostNameLen; > UrlComponents.dwUrlPathLength = pathLen; Please avoid that unportable array initialization syntax, and since the code is already explicitly setting the fields you should simply add the ones that are needed and missing. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: wininet: initialize a buffer
by Alexandre Julliard 08 Nov '07

08 Nov '07

"Lei Zhang" <thestig(a)google.com> writes: > Valgrind is complaining we are doing strcmp() on the buffer without > initializing it. You should fix the code to not do the strcmp when we don't have a valid buffer, instead of hiding the bug by clearing memory. -- Alexandre Julliard julliard(a)winehq.org

1 0

Wine LivePC
by T.J. Purtell 07 Nov '07

07 Nov '07

Hey, I've build a Wine LivePC so that people can take a linux operating system capable of running Windows programs on a USB drive. Its great for carrying a portable work environment with you, providing a container for safely running executables you download from the internet, or making an isolated environment for a family member that they can carry with them. http://www.moka5.com/node/1620 Obviously the LivePC is free, but registration on the site is required to download the free engine for using LivePC (Its based off VMware player). It has the latest version of Wine, so using this is an easy way to check if a particular application is well supported in Wine. Cheers, -T.J.

1 0

valgrind results for Oct 7
by Dan Kegel 07 Nov '07

07 Nov '07

This time I filtered out files whose only error was trying to close(-1). http://kegel.com/wine/valgrind/20071107/ Quite a few tests no longer show problems (partly because of the new filtering): comctl32/imagelist crypt32/oid crypt32/store d3d9/visual kernel32/console msi/package oleaut32/vartest rpcrt4/server shell32/shlfileop That leaves 75 tests with problems. - Dan -- Wine for Windows ISVs: http://kegel.com/wine/isv

1 0

Re: advapi32: fix buffer overrun in tests/registry.c:wine_debugstr_wn(), try 3
by Alexandre Julliard 07 Nov '07

07 Nov '07

"Dan Kegel" <dank(a)kegel.com> writes: > Fixes an off-by-one buffer overflow in wine_debugstr_wn() > in which the wchar after the end of the buffer was read. > > This is same as try 2, but is slightly clearer and more correct. > > Found via Valgrind warning: > Conditional jump or move depends on uninitialised value(s) > at 0x45F3219: wine_debugstr_wn (registry.c:151) > by 0x45F34F5: test_hkey_main_Value_W (registry.c:284) > by 0x45F7644: func_registry (registry.c:327) > by 0x460A127: run_test (test.h:387) > > Alexandre, if you don't like this 'un, please tell me why. That function is really supposed to print n chars, not n-1. The terminating null probably needs to be included in the length. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: ntdll: Architecture may be '*' in lookup_manifest_file so don't rely on the first '*' of the format indiacting where the version number is.
by Alexandre Julliard 07 Nov '07

07 Nov '07

Robert Shearman <rob(a)codeweavers.com> writes: > Instead use the underscores that separate the string fields. Of course the assembly name can contain underscores too... I think that the whole idea of checking the version number from the file name is flawed. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [10/14] WineD3D: Implement the varying map
by Allan Tong 07 Nov '07

07 Nov '07

On 11/6/07, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > + /* Don't do any register mapping magic if it is not needed, or if we can't > + * achive anything anyway > + */ > + if(highest_reg_used < (GL_LIMITS(glsl_varyings) / 4) || > + num_regs_used >= (GL_LIMITS(glsl_varyings) / 4) ) { > + if(num_regs_used >= (GL_LIMITS(glsl_varyings) / 4)) { Should it not be "num_regs_used > (GL_LIMITS(glsl_varyings) / 4)"? - Allan

2 2

Mistakes in my yesterday evening patches
by Laurent Vromman 07 Nov '07

07 Nov '07

I have send 4 patches yesterday. The 2 first of them are tagged resend. Please do not take those two first patches into account, they have already been merged in the git tree. My proxy played me a trick, i haven't see that before I sent them again. Only the 2 last ones (3/4 and 4/4) are to be merged by now. Sorry for my mistake, Laurent

1 0

Re: [5/14] D3D9: Use HAL devices in some tests
by Vitaliy Margolen 07 Nov '07

07 Nov '07

Stefan Dösinger wrote: > > ------------------------------------------------------------------------ > > > > @@ -343,7 +343,7 @@ static void test_refcount(void) > d3dpp.EnableAutoDepthStencil = TRUE; > d3dpp.AutoDepthStencilFormat = D3DFMT_D16; > > - hr = IDirect3D9_CreateDevice( pD3d, D3DADAPTER_DEFAULT, D3DDEVTYPE_NULLREF, hwnd, > + hr = IDirect3D9_CreateDevice( pD3d, D3DADAPTER_DEFAULT, D3DDEVTYPE_HAL, hwnd, > D3DCREATE_SOFTWARE_VERTEXPROCESSING, &d3dpp, &pDevice ); > ok(SUCCEEDED(hr), "Failed to create IDirect3D9Device (%s)\n", DXGetErrorString9(hr)); > if (FAILED(hr)) goto cleanup; By doing this change you render all VMs useless to run d3d tests. This is much bigger loss then broken tests on vista. Vitaliy

2 1

Re: [14/14] WineD3D: Release the pixel shader function when destroying the shader
by H. Verbeet 06 Nov '07

06 Nov '07

On 06/11/2007, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > It would probably be more proper to do stuff like this (destroying the constants lists as well) in baseshader.c

2 1

Re: RICHED20: fix test failure in test_WM_PASTE
by Alex Villacís Lasso 06 Nov '07

06 Nov '07

Alex Villacís Lasso escribió: > At last! I finally found the way to make native riched20 actually obey > keystrokes involving Ctrl and some other key. It seems that native riched20 > checks the actual keyboard state during WM_KEYDOWN processing, and won't > process keystrokes if the keyboard state does not match the messages being > sent. However, with keybd_event(), native riched20 can be fooled into > accepting > simulated keystrokes. > > Changelog: > * Native riched20 won't obey simple WM_CHAR or WM_KEYDOWN/WM_KEYUP messages > when processing Ctrl-Key keystrokes. Use keybd_event() instead. Fixes one > test failure in Win98 and WinXP-SP2. > * Add more through tests to test_WM_PASTE, to check that text is actually > being modified in intermediate steps. Fixes false test success in Win98 and > WinXP-SP2. Additional tests pass under Win98, WinXP-SP2 and Wine. > ------------------------------------------------------------------------ > Was there something wrong with this patch? Two other patches of mine for riched20 have been committed, but not this one. As it currently stands, it removes the last test failure and also fixes a false test success in WinXP-SP2. -- perl -e '$x=2.4;print sprintf("%.0f + %.0f = %.0f\n",$x,$x,$x+$x);'

1 0

valgrind results 20071106
by Dan Kegel 06 Nov '07

06 Nov '07

http://kegel.com/wine/valgrind/20071106/ Most obvious change since yesterday is Juan fixed the warnings in setupapi, so that DLL is now clean by this test.

1 0

Monday's valgrind results - ten tests fixed!
by Dan Kegel 06 Nov '07

06 Nov '07

Yesterday's git seems to have fixed all the valgrind warnings in ten tests: cabinet/extract cabinet/fdi crypt32/sip d3d8/device dsound/dsound8 mscms/profile mshtml/dom mshtml/htmldoc netapi32/access shdocvw/webbrowser Results are online at http://kegel.com/wine/valgrind/20071105/ (Sorry, I used an old suppressions file, so this has a lot of leak records in it that should have been suppressed. I'll sort that for today's git results.)

1 0

Re: [1/2] explorer: Store current display mode in the registry on startup
by Alexandre Julliard 06 Nov '07

06 Nov '07

Dmitry Timoshkov <dmitry(a)codeweavers.com> writes: > Changelog: > [1/2] explorer: Store current display mode in the registry on startup. You should either do that only in the case of using the X11 root window, or you need to use a different registry key for each virtual desktop instance. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: USER32: win16 applications should not see long filenames returned from dialog control (try 2)
by Juan Lang 06 Nov '07

06 Nov '07

> What do you think about this version? Better, but still not correct: + ret = DlgDirSelectExA( WIN_Handle32(hwnd), buffer, MAX_PATH, id ); + if (GetLastError() == 0) GetShortPathNameA(buffer, str, len); You shouldn't check GetLastError() for whether DlgDirSelectExA succeeded. Rather, just check ret. You have the same problem in DlgDirSelectComboBoxEx16. --Juan

3 3

Re: #winehq admin troubles
by feba thatl 05 Nov '07

05 Nov '07

> On 11/5/07, Alexander Nicolaysen Sørnes <alex(a)thehandofagony.com> wrote: > On Monday 05 November 2007 11:49:52 Hans Leidekker wrote: > > On Monday 05 November 2007 10:39:32 Kai Blin wrote: > > > Personally, I would like to see more people stepping up to help run > > > #winehq instead of just complaining about the work of one of the very few > > > people who actually are out there helping users. > > > > As I understand this history feba thatl tried to do exactly that: help > > others on our user channel... and got banned by vitamin. So if vitamin > > is really under stress from all these complaining users, he's not helping > > himself by kicking out others who try to help. > > > > -Hans > > > No, he was just telling the user to upgrade Wine, which vitamin knew would > have no effect whatsoever. The only thing that could have benefited from > that advice would have been his self-image. > > > > Alexander N. Sørnes > > > What? How does giving someone a basic bit of advice, usually the first advice you hear (It's even in the channel's topic), do anything to improve anyone's self-image? If I was going to try to feel better about myself, it wouldn't be by sitting on IRC acting like I know all the answers to something I don't. I no more help people for my self-image than Vitamin kicks people for entertainment-- that's not the case, and we shouldn't act like it is. Even still, why should someone who helps others because it makes them feel good be shunned? They can still help. Not to mention, where do you draw the line? People who can only give basic advice and their best guess? People who know more than that, but are off the mark? People shouldn't be expected to think in code and memorize every tidbit under the sun if they just want to try to help people, there's no reason someone should be kicked for making a good faith effort, especially without being told otherwise. I could understand if I had been sitting in the channel, and popped in with a comment that was obviously out of place based on what had already been mentioned, or if I had been asked to stop, or if my advice was obviously and intentionally harmful (telling people to rm -rf / and such), but it wasn't. If people really are expected to know everything there is to know before they try to help, is it really such a wonder that there's only a few solid contributors left, and they're stressed from it? This really has little to do with myself, Vitamin, the other person, or what happened; it's about winehq's problems where it should be concerned with helping people. There are definite civility problems; of course on the side of those that seek help (As many feel like customers instead of part of a community), but also on the side of those that give it. Now, I can obviously understand mistreating people who are just screwing around, or treating people who are trying to help them like crap, but it's treating people that are trying to give or get help the same way that bugs me. I can understand it how it happens, like I said above, I'm not saying I'd be better in such a situation. I have nothing to gain from this being corrected that anyone else seeking help wouldn't, and I'm not asking for any action to be taken against Vitamin for anything, at least yet. I've got nothing against him-- I might have a sour taste in my mouth when I cross him, but I don't think he's a bad person. He knows and cares a lot about something, tries to improve and help people with it, and dislikes people who he perceives as stupid or harmful; even to the point where he becomes a bit too annoyed and starts mislabeling people, or treating people who don't deserve it wrongly. I don't blame him for this, as I've done the same myself. Multiple times. Under different circumstances, I could easily see us being friends. What I want to see changed here isn't treatment of any specific person-- in fact I would probably disagree with any action major enough to matter being taken against Vitamin, as again, from what I've seen he seems to be a valuable asset to Wine and #winehq-- just for some sort of adoption of rules for how to treat other people being established, even if they're as simple as: 1. Users: Be nice to the helpers. We are volunteers, not employees, and have no obligation to help you. 2. Helpers and Admins: Remember Hanlon's razor. P.S. I accidentally only sent this to one email address, and had some trouble digging up the sent message to try to get it out again. Sorry about any trouble this might have.

3 2

#winehq admin troubles
by feba thatl 05 Nov '07

05 Nov '07

Sorry if this mailing list is less than appropriate, but from what I've seen this is where the conversation was last time, and there doesn't seem to be much better. A while ago (Probably one-three months), I went to #winehq to ask about something, and was treated rather rudely by vitamin. I got annoyed by him, and made a somewhat rude comment myself (not as rude as he was being, but still probably inappropriate), and he kicked me. I rejoined the channel, and I don't really remember what happened afterwards, but it can't have been too awful as I don't believe I was banned. I do remember being left with a very sour taste with Wine, as I googled vitamin, and found that this was nowhere near the first time he had been needlessly rude to a user and had even given completely unwarranted bans with seemingly no reprimand. I gradually started going back to winehq for help and questions, with no problems. Last night, I went in to ask if anyone wanted to help me get a program working, and vitamin replied. I groaned to myself, but we had a perfectly civil conversation. At one point he stopped replying, but it's not like I haven't had to walk off myself for various reasons, so I just left quietly. Today, I joined #ubuntu to try to get something working, and found someone else asking for help with Wine. I (amongst others) recommended they go to #winehq. They joined, but commented in #ubuntu that, quote: (09:27:58 PM) <Name removed>: man these #winehq people are not friendly at all I joined #winehq and confirmed my fears that vitamin was being rude to someone else. Vitamin asked what version they were running, they answered with 9.46 (the version in the Ubuntu repos, last I checked), and I recommended they go to winehq.org and follow the instructions on how to add the Wine repos. Vitamin said something along the lines of "that isn't the case here", I said that it's still helpful (to have an up to date version of wine). I then asked if Vitamin had got the program we talked about last night working, and seconds later was kicked. No explanation, the kick line just had "(Vitamin)". Utterly confused, I rejoined, for which I do have a chatlog: (09:32:56 PM) The topic for #winehq is:[long topic removed] (09:33:09 PM) ***vitamin fucking tired all knowning noobs! (09:33:15 PM) <Name removed, same person from #ubuntu>: wine: cannot find '/media/cdrom0/intro.exe' (09:33:15 PM) vitamin: usrl, you better leave (09:33:25 PM) usrl: Why? (09:33:32 PM) mode (+b *!*n=feba(a)*.hsd1.ks.comcast.net ) by vitamin (09:33:32 PM) You have been kicked by vitamin: (vitamin) That's the entire chatlog of my second time joining (I didn't think to save a copy of the first), with nothing removed but a bunch of topic stuff nobody wants to read and the fellow from #ubuntu's nick. Now, I not only have been banned for apparently being a noob (Again, it's a best guess, Vitamin did not give any clue at all in his kick messages, let alone a warning in the channel) in a channel that is supposed to support end users, but a few minutes later, I see this, again in #ubuntu: (09:37:15 PM) <Name removed>: well I got banned from winehq for asking a question This person didn't seem rude at all when I was talking to them. Thankfully, they were able to get their problems resolved in #ubuntu. Given that vitamin seems to have had user abuse problems in the past, what is going on here? I can understand not getting professional service in an IRC channel, but why is someone who is so needlessly abusive (again, kicking people for little more than asking a question; when the channel is for support.) entrusted with admin abilities? From what I read on another complaint, he seems to be in this position because he is a developer, not because he is truly a good fit for being a moderator or even a helper. While he might be great at coding, his people skills definitely suffer. I can understand getting angry over people asking their questions right, but when you're trying to support end users, you have to be able to deal with at least some stupidity-- and even disregarding that, if you're going to not only kick someone, but ban them, give them some idea of what they did wrong. It almost seems to me as if he is banning people not because they did something wrong, but because he had a bad day-- like I said, last night he was fairly civil, even though I'm sure I made my share of stupid questions and comments. I'm not sure if this is the only moderator with these sorts of problems, but something needs to be done. Wine is used by all sorts of people, heck, my mother uses it. When all sorts of people are using a program, all sorts of people are going to be having problems with it. I wouldn't expect my mother, or anyone else's, to be able to phrase their question well and make a support person happy, but I also wouldn't expect someone who is supposed to be supporting users insulting them and banning them (not only ignoring their need for help, but effectively making it impossible for many other people to help them as well). Something needs to change, Wine is something a lot of people switching to non-Windows OS are going to use, and many people could be easily turned off when they find the first place they turn to for help, the place that winehq.org leads them to, not only has very rude people in power, but has no checks or guards to keep said people from ruining a user's experience.

13 14

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

wine-devel