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

Re: test.winehq.org

by Reece Dunn

On 31/01/2008, Stefan Leichter <Stefan.Leichter(a)camline.com> wrote: > Hello, > > i just noticed that the test results for advapi32:service tests are not > available for the last build (http://test.winehq.org/data/200801301937/) It > looks like the parser did not create the files. > > Can someone familar with this stuff have a look? I noticed during the run on my Vista machine that the passed, failed, todo line for one of the tests outputted to the console. I don't know for which test it was though. Also, the security tests (the previous one) have two output lines, e.g.: security: 18 tests executed (0 marked as todo, 0 failures), 0 skipped. security: 944 tests executed (0 marked as todo, 0 failures), 1 skipped. - Reece

Jan. 31, 2008

test.winehq.org

by Stefan Leichter

Hello, i just noticed that the test results for advapi32:service tests are not available for the last build (http://test.winehq.org/data/200801301937/) It looks like the parser did not create the files. Can someone familar with this stuff have a look? Thanks Stefan

Jan. 31, 2008

Review comparing Wine and Vista for running games

by Dan Kegel

http://wastingtimewithmikeandari.wordpress.com/2008/01/31/linux-has-better-… It'd be nice if we ran those games properly, but still, at least we didn't lock the system.

Jan. 31, 2008

Re: qmgr: Implement DLL server registration.

by Dan Hipschman

On Thu, Jan 31, 2008 at 09:52:17PM +0100, Roderick Colenbrander wrote: > > In case of your dll I guess it should be hardcoded. Shipping extra > files puts an extra burden on packagers and it likely isn't what > windows is doing. The .INF is compiled into the DLL as a resource. Packagers shipping binaries won't have to do anything different. It simply makes editting the registry keys much easier and less error prone. Also, Roy obviously put a lot of work into doing it the nice way, so I feel bad ripping it out and hardcoding everything which just makes the code uglier anyway. The .INF is not loaded from a path, it's loaded from the HINSTANCE of the DLL, which is provided to the DLL in DllMain. I admit this isn't exactly the way Windows does it because the registry entries appear to be in the initialized data section of the Windows DLL, but then again, we don't have to maintain the Windows DLL.

Jan. 31, 2008

Re: qmgr: Implement DLL server registration.

by Roderick Colenbrander

If possible you should do the registration in the same way as windows as we try to be like windows in every way. Depending on the dll the settings are either hardcoded in a dll (that's what most dlls do) or when it are default windows dlls the keys are stored in wine.inf. In case of your dll I guess it should be hardcoded. Shipping extra files puts an extra burden on packagers and it likely isn't what windows is doing. Regards, Roderick > I like the way Roy used a .INF file for the DLL server registration. > Looking > at some of the other DLL's code, this seems comparably like a very clean > idea. > > ChangeLog: > > Add the server registration code to qmgr. > > From: Roy Shea <roy at cs.hmc.edu> > Date: Thu Dec 20 18:38:31 CST 2007 > > --- > dlls/qmgr/Makefile.in | 4 ++- > dlls/qmgr/qmgr.h | 11 ++++++ > dlls/qmgr/qmgr.inf | 31 ++++++++++++++++ > dlls/qmgr/qmgr.spec | 4 +- > dlls/qmgr/qmgr_main.c | 92 > +++++++++++++++++++++++++++++++++++++++++++++++-- > dlls/qmgr/rsrc.rc | 20 +++++++++++ > 6 files changed, 156 insertions(+), 6 deletions(-) > create mode 100644 dlls/qmgr/qmgr.inf > create mode 100644 dlls/qmgr/rsrc.rc > > diff --git a/dlls/qmgr/Makefile.in b/dlls/qmgr/Makefile.in > index a19eea6..7159607 100644 > --- a/dlls/qmgr/Makefile.in > +++ b/dlls/qmgr/Makefile.in > @@ -3,7 +3,7 @@ TOPOBJDIR = ../.. > SRCDIR = @srcdir@ > VPATH = @srcdir@ > MODULE = qmgr.dll > -IMPORTS = kernel32 > +IMPORTS = advpack kernel32 > EXTRALIBS = -luuid > > C_SRCS = \ > @@ -11,6 +11,8 @@ C_SRCS = \ > qmgr.c \ > qmgr_main.c > > +RC_SRCS = rsrc.rc > + > IDL_I_SRCS = qmgr_local.idl > > @MAKE_DLL_RULES@ > diff --git a/dlls/qmgr/qmgr.h b/dlls/qmgr/qmgr.h > index 9b708c8..3ad314d 100644 > --- a/dlls/qmgr/qmgr.h > +++ b/dlls/qmgr/qmgr.h > @@ -27,6 +27,8 @@ > #define COBJMACROS > #include "bits.h" > > +#include <string.h> > + > /* Background copy manager vtbl and related data */ > typedef struct > { > @@ -43,4 +45,13 @@ extern ClassFactoryImpl BITS_ClassFactory; > > HRESULT BackgroundCopyManagerConstructor(IUnknown *pUnkOuter, LPVOID > *ppObj); > > +/* Little helper functions */ > +static inline char * > +qmgr_strdup(const char *s) > +{ > + size_t n = strlen(s) + 1; > + char *d = HeapAlloc(GetProcessHeap(), 0, n); > + return d ? memcpy(d, s, n) : NULL; > +} > + > #endif /* __QMGR_H__ */ > diff --git a/dlls/qmgr/qmgr.inf b/dlls/qmgr/qmgr.inf > new file mode 100644 > index 0000000..0550f01 > --- /dev/null > +++ b/dlls/qmgr/qmgr.inf > @@ -0,0 +1,31 @@ > +[version] > +Signature="$CHICAGO$" > + > + > +[Strings] > +Services="System\CurrentControlSet\Services" > + > + > +[RegisterDll] > +AddReg = Qmgr.Reg > + > + > +[UnregisterDll] > +DelReg = Qmgr.Reg > + > + > +[Qmgr.Reg] > + > +HKCR,"AppID\BITS","AppID",,"%CLSID_BackgroundCopyQMgr%" > +HKCR,"AppID\%CLSID_BackgroundCopyQMgr%","LocalService",,"BITS" > +HKCR,"CLSID\%CLSID_BackgroundCopyManager%","AppID",,"%CLSID_BackgroundCopyQMgr%" > + > +HKLM,"Software\Microsoft\Windows > NT\CurrentVersion\SvcHost","netsvcs",0x00010000,"BITS" > + > +HKLM,"%Services%\BITS","Type",0x00010001,00000110 > +HKLM,"%Services%\BITS","Start",0x00010001,00000003 > +HKLM,"%Services%\BITS","ErrorControl",0x00010001,00000001 > +HKLM,"%Services%\BITS","Description",,"BITS Service" > +HKLM,"%Services%\BITS","DispalyName",,"BITS" > +HKLM,"%Services%\BITS","ImagePath",0x00020000,"svchost.exe -k netsvcs" > +HKLM,"%Services%\BITS\Parameters","ServiceDll",0x00020000,"qmgr.dll" > diff --git a/dlls/qmgr/qmgr.spec b/dlls/qmgr/qmgr.spec > index 1a63b32..0f9b073 100644 > --- a/dlls/qmgr/qmgr.spec > +++ b/dlls/qmgr/qmgr.spec > @@ -1,2 +1,2 @@ > -@ stub DllRegisterServer > -@ stub DllUnregisterServer > +@ stdcall -private DllRegisterServer() > +@ stdcall -private DllUnregisterServer() > diff --git a/dlls/qmgr/qmgr_main.c b/dlls/qmgr/qmgr_main.c > index d00ac2d..2bb5258 100644 > --- a/dlls/qmgr/qmgr_main.c > +++ b/dlls/qmgr/qmgr_main.c > @@ -21,15 +21,27 @@ > * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, > USA > */ > > -#include <stdarg.h> > +#include <stdio.h> > > -#include "windef.h" > -#include "winbase.h" > +#include "objbase.h" > +#include "winuser.h" > +#include "winreg.h" > +#include "advpub.h" > + > +#include "bits.h" > +#include "qmgr.h" > +#include "initguid.h" > > #include "wine/debug.h" > > WINE_DEFAULT_DEBUG_CHANNEL(qmgr); > > +/* Handle to the base address of this DLL */ > +static HINSTANCE hInst; > + > +/* Other GUIDs used by this module */ > +DEFINE_GUID(CLSID_BackgroundCopyQMgr, 0x69AD4AEE, 0x51BE, 0x439b, > 0xA9,0x2C, 0x86,0xAE,0x49,0x0E,0x8B,0x30); > + > /* Entry point for DLL */ > BOOL WINAPI DllMain(HINSTANCE hinstDLL, DWORD fdwReason, LPVOID > lpvReserved) > { > @@ -41,6 +53,7 @@ BOOL WINAPI DllMain(HINSTANCE hinstDLL, DWORD fdwReason, > LPVOID lpvReserved) > return FALSE; /* prefer native version */ > case DLL_PROCESS_ATTACH: > DisableThreadLibraryCalls(hinstDLL); > + hInst = hinstDLL; > break; > case DLL_PROCESS_DETACH: > break; > @@ -48,3 +61,76 @@ BOOL WINAPI DllMain(HINSTANCE hinstDLL, DWORD > fdwReason, LPVOID lpvReserved) > > return TRUE; > } > + > +static HRESULT init_register_strtable(STRTABLEA *strtable) > +{ > + #define CLSID_EXPANSION_ENTRY(id) { "CLSID_" #id, &CLSID_ ## id } > + static const struct { > + const char *name; > + const CLSID *clsid; > + } expns[] = { > + CLSID_EXPANSION_ENTRY(BackgroundCopyQMgr), > + CLSID_EXPANSION_ENTRY(BackgroundCopyManager) > + }; > + #undef CLSID_EXPANSION_ENTRY > + static STRENTRYA pse[sizeof expns / sizeof expns[0]]; > + int i; > + > + strtable->cEntries = sizeof pse / sizeof pse[0]; > + strtable->pse = pse; > + for (i = 0; i < strtable->cEntries; i++) { > + static const char dummy_sample[] = > "{12345678-1234-1234-1234-123456789012}"; > + const CLSID *clsid = expns[i].clsid; > + pse[i].pszName = qmgr_strdup(expns[i].name); > + pse[i].pszValue = HeapAlloc(GetProcessHeap(), 0, sizeof > dummy_sample); > + if (!pse[i].pszName || !pse[i].pszValue) > + return E_OUTOFMEMORY; > + sprintf(pse[i].pszValue, > "{%08X-%04X-%04X-%02X%02X-%02X%02X%02X%02X%02X%02X}", > + clsid->Data1, clsid->Data2, clsid->Data3, > clsid->Data4[0], > + clsid->Data4[1], clsid->Data4[2], clsid->Data4[3], > clsid->Data4[4], > + clsid->Data4[5], clsid->Data4[6], clsid->Data4[7]); > + } > + > + return S_OK; > +} > + > +static void cleanup_register_strtable(STRTABLEA *strtable) > +{ > + int i; > + for (i = 0; i < strtable->cEntries; i++) { > + HeapFree(GetProcessHeap(), 0, strtable->pse[i].pszName); > + HeapFree(GetProcessHeap(), 0, strtable->pse[i].pszValue); > + if (!strtable->pse[i].pszName || !strtable->pse[i].pszValue) > + return; > + } > +} > + > +/* Use an INF file to register or unregister the DLL */ > +static HRESULT register_server(BOOL do_register) > +{ > + HRESULT hr; > + STRTABLEA strtable; > + > + TRACE("(%x)\n", do_register); > + > + hr = init_register_strtable(&strtable); > + if (SUCCEEDED(hr)) > + hr = RegInstallA(hInst, do_register ? "RegisterDll" : > "UnregisterDll", > + &strtable); > + cleanup_register_strtable(&strtable); > + > + if (FAILED(hr)) > + ERR("failed: %08x\n", hr); > + > + return hr; > +} > + > +HRESULT WINAPI DllRegisterServer() > +{ > + return register_server(TRUE); > +} > + > +HRESULT WINAPI DllUnregisterServer() > +{ > + return register_server(FALSE); > +} > diff --git a/dlls/qmgr/rsrc.rc b/dlls/qmgr/rsrc.rc > new file mode 100644 > index 0000000..dfc64d0 > --- /dev/null > +++ b/dlls/qmgr/rsrc.rc > @@ -0,0 +1,20 @@ > +/* > + * Copyright 2007 Google (Roy Shea) > + * > + * This library is free software; you can redistribute it and/or > + * modify it under the terms of the GNU Lesser General Public > + * License as published by the Free Software Foundation; either > + * version 2.1 of the License, or (at your option) any later version. > + * > + * This library is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU > + * Lesser General Public License for more details. > + * > + * You should have received a copy of the GNU Lesser General Public > + * License along with this library; if not, write to the Free Software > + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, > USA > + */ > + > +/* @makedep: qmgr.inf */ > +REGINST REGINST qmgr.inf > -- Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen! Ideal für Modem und ISDN: http://www.gmx.net/de/go/smartsurfer

Jan. 31, 2008

Re: Is http://test.winehq.org down?

by Reece Dunn

On 31/01/2008, Reece Dunn <msclrhd(a)googlemail.com> wrote: > Hi, > > I an getting a network timeout whern connecting to http://test.winehq.org. Ignore... it is working for me again. - Reece

Jan. 31, 2008

Is http://test.winehq.org down?

by Reece Dunn

Hi, I an getting a network timeout whern connecting to http://test.winehq.org. - Reece

Jan. 31, 2008

Re: No new winetest?

by Paul Millar

On Tuesday 29 January 2008 09:26:44 Reece Dunn wrote: > Is there any more news on this? The front-end machine is up and the builds are going ahead. This is mainly due to Andrew Elwell (thanks!). The root-cause of the problem is still unknown; although the machine seems stable, the problem may come back. We're investigating this further. Cheers, Paul.

Jan. 31, 2008

What's needed to make gdb work again?

by Maarten Lankhorst

Hi folks, I've been trying some debugging with kdbg (Graphical kde debugger), but it fails on wine because it can't attach to a process. While it is of course possible to use WINEDEBUG flags, I want to be able to use a full debugger, and not just deduce from logs what happened, but rather be able to see the corresponding lines of source as well. so would it be possible to make gdb work with winedbg again, and perhaps make it understand threads this time? Cheers, Maarten.

Jan. 31, 2008

Re: Wiki Down?

by Dimi Paun

On Wed, 2008-01-30 at 13:28 -0800, Lei Zhang wrote: > The wiki seems to be down again. Yes, it was down for about 2h this evening because we moved it back to the faster server. It should be back with a vengeance now, please let me know if you notice any problems... -- Dimi Paun <dimi(a)lattica.com> Lattica, Inc.

Jan. 31, 2008