- Wine-Devel - list.winehq.org

PATCH: assembler _invoke
by Marcus Meissner Jan. 5, 2006

Jan. 5, 2006

On Wed, Jan 04, 2006 at 09:14:18PM -0800, Dan Kegel wrote: > This would definitely be better handled in assembly... Here is the first try. It leaves the old version for non-intel and non-gcc compilers. Ciao, Marcus Index: typelib.c =================================================================== RCS file: /home/wine/wine/dlls/oleaut32/typelib.c,v retrieving revision 1.200 diff -u -r1.200 typelib.c --- typelib.c 3 Jan 2006 12:42:34 -0000 1.200 +++ typelib.c 5 Jan 2006 09:07:25 -0000 @@ -4843,7 +4843,33 @@ switch (callconv) { case CC_STDCALL: +#if defined(__i386__) && defined(__GNUC__) + __asm__( + /* store all stuff into registers now, since we modify the stack */ + " mov %1, %%ecx\n" + " mov %2, %%edx\n" + " mov %3, %%esi\n" + " test %%ecx, %%ecx\n" + " jz 1f\n" /* no args, nothing to push */ + /* + * Arguments are pushed from right to left. + * We use intel memory move operations. + */ + " std\n" /* direction: backward */ + "2: lodsl\n" /* moves *esi -> eax, esi-= 4 */ + " pushl %%eax\n" + " dec %%ecx\n" + " jnz 2b\n" + "1: cld\n" /* direction: forward (default) */ + " call *%%edx\n" + " mov %%eax, %0\n" + : "=g" (res) + : "g" (nrargs), "g" (func), "g" (args + (nrargs-1)) + : "memory", "cc", "ax", "si", "cx", "dx" + ); + break; +#else switch (nrargs) { case 0: res = func(); @@ -4905,6 +4931,7 @@ break; } break; +#endif default: FIXME("unsupported calling convention %d\n",callconv); res = -1;

1 0

New, flex-related warnings
by Gerald Pfeifer Jan. 5, 2006

Jan. 5, 2006

Using flex 2.5.4 and GCC 3.4, I started getting the following new warnings for my Wine builds about two days ago (which I never got before): ./ppl.l:176:1: warning: "YY_NO_UNPUT" redefined ./ppl.l:129:1: warning: this is the location of the previous definition ./parser.l:57:1: warning: "YY_NO_UNPUT" redefined warning: this is the location of the previous definition ./parser.l:58:1: warning: "YY_NO_TOP_STATE" redefined ./parser.l:23:1: warning: this is the location of the previous definition ./parser.l:116:1: warning: "YY_NO_UNPUT" redefined ./parser.l:100:1: warning: this is the location of the previous definition ./parser.l:117:1: warning: "YY_NO_TOP_STATE" redefined ./parser.l:101:1: warning: this is the location of the previous definition ./debug.l:73:1: warning: "YY_NO_UNPUT" redefined ./debug.l:22:1: warning: this is the location of the previous definition ./macro.lex.l:42:1: warning: "YY_NO_UNPUT" redefined ./macro.lex.l:22:1: warning: this is the location of the previous definition I'm afraid, this is caused by the following patchset: date: 2006-01-03 12:43:14 +0000; author: julliard; state: Exp; lines: +4 -2 Alexandre Julliard <julliard(a)winehq.org> Added some flex options to avoid compiler warnings. Moved options from the Makefile into the source. Gerald

1 0

Licensing question
by Dominic Wise Jan. 5, 2006

Jan. 5, 2006

I have a question regarding the use of portions of Wine in a commercial application. Sorry if this is not the right place to post but I am not sure who I can directly address this to. The application my employer develops is a financial application designed to work on Win 2K and Win XP, but we have a need for a Win32 function that is only supported in XP (TzSpecificLocalTimeToSystemTime). We could write an implementation of this function ourselves for Win 2K but I have noticed that there is a full implementation in Wine. Is it permissible to use the source for this function in our application? If so, what provisions do we need to make with regard to recognition of Wine and supply of source code to our customers ? I am currently trying to demonstrate the benefits of Open Source technology to my employer (I have already replaced one proprietary third-party component in our application with a better Open Source implementation) and this could really help to move this process along. Longer-term, I want to try and get the application running on Wine, but that's another story. Regards, Dominic

5 6

Re: PATCH: winspool.spec adjustments
by Detlef Riekenberg Jan. 4, 2006

Jan. 4, 2006

Am Mittwoch, den 04.01.2006, 18:40 +0100 schrieb Marcus Meissner: -noname is wrong here: > +202 stub -noname SetDefaultPrinterA(a)4 > +204 stub -noname SetDefaultPrinterW(a)4 winspool.drv from w2k export this Functions by name. (The Ordinal-Numbers are correct) -- By By ... ... Detlef

1 0

Re: RpcRT4: Implement NdrClientCall2 and NdrServerCall2
by Roderick Colenbrander Jan. 4, 2006

Jan. 4, 2006

+ * Copyright 2003-5 Robert Shearman (for CodeWeavers) * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public It is 2006 now ;) Roderick

1 0

Re: CRYPTUI: Add an empty cryptui implementation.
by Juan Lang Jan. 4, 2006

Jan. 4, 2006

> May not want this empty dll in WineHQ, but it makes SmartFTP > start up I don't see any reason not to include it. It is always available, as far as I know, when crypt32.dll is available. Programs that need a real one probably also need native crypt32.dll for now, so people already have to mess with their dll overrides for apps that might get broken by its presence. --Juan __________________________________________ Yahoo! DSL Something to write home about. Just $16.99/mo. or less. dsl.yahoo.com

1 0

Re: PATCH: winspool ordinal export
by Robert Shearman Jan. 4, 2006

Jan. 4, 2006

Hi Marcus, Thanks for submitting a patch for this. I also have an application that depends on one of the exports to have a specific ordinal (203 - GetDefaultPrinterW). Marcus Meissner wrote: >-100 stub @ >+100 stub EnumPrinterPropertySheets(a)16 > > This should be --noname. >+# 101 ClusterSplOpen(a)20 >+# 102 ClusterSplClose(a)4 >+# 103 ClusterSplIsAlive(a)4 >+201 stdcall GetDefaultPrinterA(ptr ptr) >+# 202 _SetDefaultPrinterA(a)4 >+203 stdcall GetDefaultPrinterW(ptr ptr) >+# 204 _SetDefaultPrinterW(a)4 >+# 205 _SplReadPrinter(a)12 >+# 206 _AddPerMachineConnectionA(a)16 >+# 207 _AddPerMachineConnectionW(a)16 >+# 208 _DeletePerMachineConnectionA(a)8 >+# 209 _DeletePerMachineConnectionW(a)8 >+# 210 _EnumPerMachineConnectionsA(a)20 >+# 211 _EnumPerMachineConnectionsW(a)20 >+# 212 _LoadPrinterDriver(a)4 >+# 213 _RefCntLoadDriver(a)16 >+# 214 _RefCntUnloadDriver(a)8 >+# 215 _ForceUnloadDriver(a)4 >+# 216 _PublishPrinterA(a)24 >+# 217 _PublishPrinterW(a)24 >+# 218 _CallCommonPropertySheetUI(a)16 >+# 219 _PrintUIQueueCreate(a)16 >+# 220 _PrintUIPrinterPropPages(a)16 >+# 221 _PrintUIDocumentDefaults(a)16 >+# 222 _SendRecvBidiData(a)16 >+# 223 _RouterFreeBidiResponseContainer(a)4 >+# 224 _ExternalConnectToLd64In32Server(a)4 >+# 226 _PrintUIWebPnpEntry(a)4 >+# 227 _PrintUIWebPnpPostEntry(a)16 >+# 228 _PrintUICreateInstance(a)8 >+# 229 _PrintUIDocumentPropertiesWrap(a)28 >+# 230 _PrintUIPrinterSetup(a)24 >+# 231 _PrintUIServerPropPages(a)16 >+# 232 _AddDriverCatalog(a)16 >+ > > We should add this ordinals, not just comment them out. Why? Because they are exported by ordinal only and so if an application (or more likely a Microsoft DLL) tries to use one of them it will end up with the wrong function, rather than just failing to import it and most likely will result in stack corruption. All of these should also be exported with --noname as they are ordinal-only exports. > @ stub ADVANCEDSETUPDIALOG > @ stdcall AbortPrinter(long) > @ stdcall AddFormA(long long ptr) >@@ -89,8 +124,6 @@ > @ stdcall FindFirstPrinterChangeNotification(long long long ptr) > @ stdcall FindNextPrinterChangeNotification(long ptr ptr ptr) > @ stdcall FreePrinterNotifyInfo(ptr) >-@ stdcall GetDefaultPrinterA(ptr ptr) >-@ stdcall GetDefaultPrinterW(ptr ptr) > @ stdcall GetFormA(long str long ptr long ptr) > @ stdcall GetFormW(long wstr long ptr long ptr) > @ stdcall GetJobA(long long long ptr long ptr) > -- Rob Shearman

1 0

re: Licensing question
by Dan Kegel Jan. 4, 2006

Jan. 4, 2006

dominic.wise(a)ukonline.co.uk wrote: > The application my employer develops is a financial > application designed to work on Win 2K and Win XP, but we > have a need for a Win32 function that is only supported in > XP (TzSpecificLocalTimeToSystemTime). We could write an > implementation of this function ourselves for Win 2K but > I have noticed that there is a full implementation in Wine. > > Is it permissible to use the source for this function in > our application? That function, along with all of Wine, is LGPL'd. Thus to use it in a proprietary app, you just have to put it in its own DLL. > If so, what provisions do we need to make > with regard to recognition of Wine and supply of source > code to our customers ? You need to provide the source code for the LGPL'd DLL you ship. Have a look at the terms of the LGPL for details: http://www.gnu.org/copyleft/lesser.txt Good luck with your project! - Dan -- Wine for Windows ISVs: http://kegel.com/wine/isv

2 1

safedisc
by Ivan Leo Puoti Jan. 4, 2006

Jan. 4, 2006

Raphael Junqueira asked on bugzilla what the safedisc status is. Currently it works fine, and I believe what we have is more or less ready for CVS. However Vitaly told me Alexandre didn't like the object manager Vitaly wrote, mostly he didn't like permanent objects, that drivers depend on. I haven't talked to Alexandre about this but hopefully some reasonable solution can be found so we can get Vitaly's OM into wineserver (I mean my original implementation that uses pointers as handles also works, but things look better with a real OM). So let's try and trigger some community discussion, we are talking about the guts of wine after all. Vitaly, what's the OM status currently? Alexandre, what didn't you like about it? Ivan.

4 3

Re: PATCH: winspool ordinal export
by Dmitry Timoshkov Jan. 4, 2006

Jan. 4, 2006

"Marcus Meissner" <marcus(a)jet.franken.de> wrote: > SecureCRT from bug 4225 imports GetDefaultPrinterA() by > using the ordinal 201. > > Not sure if this fix is correct or if this is even the > correct approach. But it helps SecureCRT. The other > winspool functions are imported by name. This time I ran 'dumpbin /exports' on winspool.lib provided with PSDK and got interesting result: only some APIs are exported by ordinal, here they are sorted by ordinal, I skipped all APIs without ordinals: 100 _EnumPrinterPropertySheets(a)16 101 _ClusterSplOpen(a)20 102 _ClusterSplClose(a)4 103 _ClusterSplIsAlive(a)4 201 _GetDefaultPrinterA(a)8 202 _SetDefaultPrinterA(a)4 203 _GetDefaultPrinterW(a)8 204 _SetDefaultPrinterW(a)4 205 _SplReadPrinter(a)12 206 _AddPerMachineConnectionA(a)16 207 _AddPerMachineConnectionW(a)16 208 _DeletePerMachineConnectionA(a)8 209 _DeletePerMachineConnectionW(a)8 210 _EnumPerMachineConnectionsA(a)20 211 _EnumPerMachineConnectionsW(a)20 212 _LoadPrinterDriver(a)4 213 _RefCntLoadDriver(a)16 214 _RefCntUnloadDriver(a)8 215 _ForceUnloadDriver(a)4 216 _PublishPrinterA(a)24 217 _PublishPrinterW(a)24 218 _CallCommonPropertySheetUI(a)16 219 _PrintUIQueueCreate(a)16 220 _PrintUIPrinterPropPages(a)16 221 _PrintUIDocumentDefaults(a)16 222 _SendRecvBidiData(a)16 223 _RouterFreeBidiResponseContainer(a)4 224 _ExternalConnectToLd64In32Server(a)4 226 _PrintUIWebPnpEntry(a)4 227 _PrintUIWebPnpPostEntry(a)16 228 _PrintUICreateInstance(a)8 229 _PrintUIDocumentPropertiesWrap(a)28 230 _PrintUIPrinterSetup(a)24 231 _PrintUIServerPropPages(a)16 232 _AddDriverCatalog(a)16 -- Dmitry.

1 0