January 2002 - Wine-Devel - list.winehq.org

Patch for OLEClipbrd_IDataObject_GetData
by Sander van Leeuwen Jan. 26, 2002

Jan. 26, 2002

Hi, Forte Agent was unable to copy text from the clipboard into the message body window because of a bug in this method. I'm not sure how Wine's clipboard implementation works, but handles returned by GetClipboardData are not supposed to be valid after the app calls CloseClipboard (see MS docs). The attached patch makes a copy of the clipboard handle and passes that to the application. Forte Agent frees this handle after it has extracted the clipboard data, so there's no memory leak. Sander

1 0

The cat is out of the bag (Lindows)
by Hetz Ben Hamo Jan. 26, 2002

Jan. 26, 2002

Hi People, In case you didn't see - Lindows "preview" review is on www.newsforge.com... Hetz

1 1

cvs run
by lee johnson Jan. 26, 2002

Jan. 26, 2002

just got wine cvs but it refuses to run even with copying missing shared file from $HOME to say /usr/lib here is error: --------------- [codewind(a)imyourhandiman codewind]$ wine wine: error while loading shared libraries: libwine.so: cannot open shared object file: No such file or directory [codewind(a)imyourhandiman codewind]$ thx lee -====

2 1

Does Windows have one of these?
by Bill Medland Jan. 25, 2002

Jan. 25, 2002

As I look deeper into the DrawText boundary behaviour it becomes clear to me that the Microsoft version must actually keep the & symbols in the string until the very last moment. In particular it looks to me like they actually measure strings that contain them. I can't see how they can do that with GetTextExtent and its relatives; I can't see where to hide a flag as to whether to interpret the & or not. Does anyone know if there are Microsoft undocumented functions or undocumented ways to measure and output a text that includes & symbols with a way of specifying whether they count as prefix characters or not? Personally it seems silly to me and I don't intend doing so myself unless there is a good reason to duplicate all of Microsoft's weird bugs. Bill

1 0

RE: Excessive clipping of text in listview
by Medland, Bill Jan. 25, 2002

Jan. 25, 2002

> -----Original Message----- > From: Dimitrie O. Paun [mailto:dimi(a)cs.toronto.edu] > Sent: Friday, January 25, 2002 9:11 AM > To: Bill Medland > Cc: wine-devel(a)winehq.com > Subject: Re: Excessive clipping of text in listview > > > On Fri, 25 Jan 2002, Bill Medland wrote: > > > Does anyone know how we (again) broke the text in the large > icon display of > > listview? It seems to come and go as we develop. > > <blush>I probably broke it</blush> I know the feeling. (not surprising given the size of the delta; you ought to see what my last big one would have screwed up). > > > The symptom. Our application has a ListView window. As of > yesterday's cvs > > most of the text only displays about 2 or 3 pixels wide, > while a few are > > readable. > > Are you using Unicode or ANSI? Can I get access to the code > to test with? ANSI Do you have access to the Microsoft "rowlist" sample? If so then use that but edit a couple of the labels to be longer than one line. It's the "being longer than one line" that messes it up. > > > I am just backtracking but I'm in a rush (I have to put together > > a demo machine by yesterday) and I'd like to know how far > back to go. > > Tell me, is 1.106 OK? Most likely the patch 1.106 -> 1.107 broke it. It did work on 1/jan so it was probably your delta. > > > -- > Dimi. > Thanks for the rapid response Bill p.s. what is going on in DrawLargeItem with the ellipsification; it looks like it ellipsifies it and then throws the result away. I am currently working on DrawText which should provide everything needed to draw the text properly. I am feeding it to Alexandre in small pieces so that we don't break anything. Microsoft, as far as I can see, don't actually use DrawText; they appear to reimplement all the ellipsification but I don't see why.

3 6

RE: Excessive clipping of text in listview
by Medland, Bill Jan. 25, 2002

Jan. 25, 2002

Panic over; the 20020101 version will do for now. > -----Original Message----- > From: Dimitrie O. Paun [mailto:dimi(a)cs.toronto.edu] > Sent: Friday, January 25, 2002 9:11 AM > To: Bill Medland > Cc: wine-devel(a)winehq.com > Subject: Re: Excessive clipping of text in listview > > > On Fri, 25 Jan 2002, Bill Medland wrote: > > > Does anyone know how we (again) broke the text in the large > icon display of > > listview? It seems to come and go as we develop. > > <blush>I probably broke it</blush> > > > The symptom. Our application has a ListView window. As of > yesterday's cvs > > most of the text only displays about 2 or 3 pixels wide, > while a few are > > readable. > > Are you using Unicode or ANSI? Can I get access to the code > to test with? > > > I am just backtracking but I'm in a rush (I have to put together > > a demo machine by yesterday) and I'd like to know how far > back to go. > > Tell me, is 1.106 OK? Most likely the patch 1.106 -> 1.107 broke it. > > > -- > Dimi. >

1 0

Excessive clipping of text in listview
by Bill Medland Jan. 25, 2002

Jan. 25, 2002

Does anyone know how we (again) broke the text in the large icon display of listview? It seems to come and go as we develop. The symptom. Our application has a ListView window. As of yesterday's cvs most of the text only displays about 2 or 3 pixels wide, while a few are readable. I am just backtracking but I'm in a rush (I have to put together a demo machine by yesterday) and I'd like to know how far back to go. Bill

2 1

Re: Patch to improve Microsoft Office 2000 functionality
by Duane Clark Jan. 25, 2002

Jan. 25, 2002

Jeremy White wrote: > ... > by getting it out now, maybe some other folks will help out <grin>. > I'll be happy to sync this to the current CVS, test it out with the apps I have, and chop it up into a series of patches, if no one else wants to take that on.

2 4

C++ Winelib problems
by Peter Hunnisett Jan. 25, 2002

Jan. 25, 2002

Hello all, I've been tooling around with building a C++ application using Winelib and I've come across a couple of problems resulting from the use of global constructors. The problems can be reproduced with a simple C++ application which from a constructor calls a method which is resolved through a dll import. 1) If this program is the main program, wine_dll_set_callback has not yet been invoked so the __wine_dll_register call doesn't resolve the import list. The problem is that the constructors of the program are run immediately after the dll is registered which results in calling a non resolved import. This can be worked around by simply creating a non C++ stub which does a LoadLibrary/GetProcAddress on the real main exe. However I'm wondering if there is a good reason for needing to wait until after the main exe is loaded before being able to resolve imports? I must be missing something from my quick view of the code. 2) The order of running constructor execution and DLL entry point is incorrect. To show the problem assume you have foo.dll which imports bar.dll Present order for builtins: Load foo.dll Start to resolve foo.dll dependencies Start to resolve bar.dll dependencies Finsh resolving bar.dll dependencies Run bar.dll C++ constructors Finish resolving foo.dll dependencies Run foo.dll C++ constructors Run bar.dll DLL entry point Run foo.dll DLL entry point I actually haven't checked if PE images presently have the same problem since I don't know what triggers the ctor execution for a PE image. I assume that it's the running of the DLL entry point, but I don't know for sure. Hopefully I'll find the time at some point this week to whip up a test. The correct order: Load foo.dll Start to resolve foo.dll dependencies Start to resolve bar.dll dependencies Finsh resolving bar.dll dependencies Run bar.dll C++ constructors Run bar.dll DLL entry point Finish resolving foo.dll dependencies Run foo.dll C++ constructors Run foo.dll DLL entry point The problem seems reasonably simple enough to resolve but I'm sure that I'm missing something since my patch doesn't work for the present version of wine :(. I think that the problem is that there's an issue with inter DLL dependencies somewhere as it dies in SYSMETRICS_Init with a bad hdc. I've attached a patch which removes the recursive dep search from MODULE_DllProcessAttach and moves MODULE_DllProcessAttach from LoadLibraryExA into MODULE_LoadLibraryExA. Does this seem like a reasonable approach to take? Any problems with the patch? Ciao, Peter

2 2

Word97 (winver win95) - no luck with wine
by Heiko Nardmann Jan. 25, 2002

Jan. 25, 2002

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi! I have problems starting Word97 using wine-20020122. I use the following command line: env LANG=de_DE LD_LIBRARY_PATH=/opt/wine-20020122/lib:$LD_LIBRARY_PATH /opt/wine-20020122/bin/wine --winver=win95 winword.exe The working directory is: /usr/Windows_NT_Files/Programs/Office/Word Inside this directory there are the following files: - -rwxr----- 1 nardmann users 1108752 Jun 23 2001 MSO7DEU.DLL - -rwxr--r-- 1 nardmann users 3792896 Aug 18 2000 mso97.dll - -rw-r--r-- 1 nardmann users 31232 Jun 29 2001 Normal.dot - -rwxr--r-- 1 nardmann users 5337088 Mai 8 2001 winword.exe - -rwxr--r-- 1 nardmann users 1724416 Aug 18 2000 wwintl32.dll I do not use an installed Windows, only these files! I use Wine on SuSE Linux 7.1 remote using Exceed on Windows NT 4.0. Some time ago I got Word97 running using a CVS version but now I get the following: For language 'german' several language ids were found: de_DE - 0407; de_CH - 0807; de_LU - 1007; de_LI - 1407; Instead of using first in the list, suggest to define your LANG environment variable like this: LANG=de_DE fixme:keyboard:X11DRV_KEYBOARD_DetectLayout Your keyboard layout was not found! Using closest match instead (German keyboard layout without dead keys) for scancode mapping. Please define your layout in windows/x11drv/keyboard.c and submit them to us for inclusion into future Wine releases. See the Wine User Guide, chapter "Keyboard" for more information. wine: Unhandled exception, starting debugger... And nothing happens then ... I could add '--debugmsg +all' and provide gzipped file if necessary. How can I locate the problem? I tried to run Wine at the console of the Linux machine logged in as root instead of user nardmann. Then I see the splash screen and get a message box saying that my registry is corrupted and that I should restart the Word setup program. I cannot click the button in this message box. I am going to check the '--debugmsg +reg' output during the next week and compare it with the contents of the registry in the .wine directory and the registry on my local WinNT machine. Running Wine remote using Exceed but being root gives me the following in the shell: snsrv053:root[Word]#env LANG=de_DE LD_LIBRARY_PATH=/opt/wine-20020122/lib:$LD_LIBRARY_PATH /opt/wine-20020122/bin/wine --winver=nt40 winword.exe fixme:keyboard:X11DRV_KEYBOARD_DetectLayout Your keyboard layout was not found! Using closest match instead (German keyboard layout without dead keys) for scancode mapping. Please define your layout in windows/x11drv/keyboard.c and submit them to us for inclusion into future Wine releases. See the Wine User Guide, chapter "Keyboard" for more information. X Error of failed request: BadMatch (invalid parameter attributes) Major opcode of failed request: 1 (X_CreateWindow) Serial number of failed request: 27 Current serial number in output stream: 37 The display variable is set correctly. How can I debug this bug? - -- Heiko Nardmann (Dipl.-Ing.), h.nardmann(a)secunet.de, Software Development secunet Security Networks AG - Sicherheit in Netzwerken (www.secunet.de) Weidenauer Str. 223-225, D-57076 Siegen Tel. : +49 271 48950-13, Fax : +49 271 48950-50 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE8URippm53PRScYygRAnAZAJ4tAWZzU+QnKaChlXFZtyvJJkLwhQCgsoUG WwRlq+Fb46jaQuZ7s0kHbAI= =ObqT -----END PGP SIGNATURE-----

1 0

C testing framework. ASCII/Unicode portable version
by Andriy Palamarchuk Jan. 24, 2002

Jan. 24, 2002

Attached version of the C testing framework, which is implemented with using TCHAR.H macros, so it is portable between ASCII and Unicode platforms. Also implemented test which can be used to test ASCII and Unicode API. As I found out creating ASCII/Unicode portable code requires much more attention than creating plain ASCII-only code, probably the same level of difficulty as creation Unicode-only code. Andriy Palamarchuk __________________________________________________ Do You Yahoo!? Send FREE video emails in Yahoo! Mail! http://promo.yahoo.com/videomail/

5 33

Re: Microsoft's version of "WINE"
by Srdjan Matovina -- Customer Engineering Jan. 23, 2002

Jan. 23, 2002

They use MainWin from MainSoft Inc. See http://www.mainsoft.com >To: wine-devel(a)winehq.com >From: "Sander Wichers" <wichers(a)usa.net> >Subject: Microsoft's version of "WINE" >X-Priority: 3 >X-MSMail-Priority: Normal >X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 >X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4807.1700 >X-BeenThere: wine-devel(a)winehq.com >X-Mailman-Version: 2.0 >List-Help: <mailto:wine-devel-request(a)winehq.com?subject=help> >List-Post: <mailto:wine-devel(a)winehq.com> >List-Subscribe: <http://www.winehq.com/mailman/listinfo/wine-devel>, <mailto:wine-devel-request(a)winehq.com?subject=subscribe> >List-Id: Wine Developer's List <wine-devel.winehq.com> >List-Unsubscribe: <http://www.winehq.com/mailman/listinfo/wine-devel>, <mailto:wine-devel-request(a)winehq.com?subject=unsubscribe> >List-Archive: <http://www.winehq.com/hypermail/wine-devel/> >Date: Wed, 23 Jan 2002 16:20:35 +0100 > >Hi, > >Maybe you guys already seen this, and it's old news for you... but i'll post >this anyway in case you didn't know. It can be a good thing to learn from. > >I do not own a SUN or a HP-UX system. But I was just interested how >microsoft implemented IE 4 / IE 5 on HP-UX and SUN. So I extracted their >installation files to disk. It seems that microsoft made their own version >of wine. It's amazing that they didn't modified IE to run on unix. Instead >they made the choice to emulate all the windows functionality. And compile >IE on that. Microsoft is ready to run their applications on linux. They have >a complete "system" libary full of ported functionality to SUN and HP-UX... >all the libaries are there, libkernel32.so etc.... It's also nice to see how >they seperated the resources from the binaries. and let the kernel deal with >these resource files. > >Has any one of you looked into these system libaries? Is this information >relevant to anyone? > >bye, > >Sander Wichers > > > > > > ==== Srdjan Matovina Sun Microsystems Inc. SunPS Integration Services/ Americas Customer Engineering srdjan.matovina(a)sun.com Ph.(905) 415-2883 Fax.(905) 477-0217

2 1

Privaleged instruction
by Nog Jan. 23, 2002

Jan. 23, 2002

Hi, Just for the fun of it I tried to run the dos installer of Mechwarrior 2: Mercenaries. It didn't get too far as the installer crashed on an unhandeled Privileged instruction. According to winedbg the instruction is 0x0002:0x10000 push %cs. To see what the opcode of this instruction might be, I added a trace to the INSTR_EmulateInstruction. Looking at the trace I found that the instruction it was having to emulate was 0x00 which looks very wrong to me. But just to check I simply made that function return true if the instruction was 0x00 but this only seemed to make things worse, the exception still occured but the debugger didn't launch. What I would like to know is, is push %cs realy a Privileged instruction (which I think it might be)? Looking into why INSTR_EmulateInstruction doesn't get the right pointer, the offset looked too big to me. Shouldn't it roll over to the next segment after 0xffff? The problem with the way that it is now is that the following macro, which converts the segment offset pair to linear addresses cuts off the upper bits, which looks correct. #define PTR_REAL_TO_LIN(seg,off) \ ((void*)(((unsigned int)(seg) << 4) + LOWORD(off))) What I want to know is why the offset is so high? nog.

2 1

RE: runtest: new option -M for testing current module as builtin
by Patrik Stridvall Jan. 23, 2002

Jan. 23, 2002

> On Wed, 23 Jan 2002, Patrik Stridvall wrote: > [...] > > BEGIN { > > require wine; > > &wine::module_native("wininet"); > > &wine::module_builtin("imm32", "msacm32"); > > } > > I am not sure any test should require the native version > of any dll. > > We cannot ship native dlls with Wine for obvious reasons, so requiring > the native dll would force Wine developpers to have a Windows > partition > installation around, and one configured for use in the tests. > Otherwise > they would not be able to run the tests, would they? > > I don't think that's good. Of course it shouldn't require it. It should simply skip parts of the test if the native version is not available. if(&wine::is_native("wininet")) { # Do native wininet tests } > Choosing to use a native dll is more up to the developper > to debug a > test. And then modifying the configuration file, as we usually do to > debug applications, should be simple and familiar enough. > > But maybe I misunderstood something. Well it allows us to automatically test a non-developer Windows installation directly under Linux without requiring him to reboot. Even developers make find it useful to quickly be able to modify the loadorder without having to modify his configuration. Sometimes what you want to test is quite subtle, like you have a function A in module B that calls function C in module D. Perhaps you want to test the robustness of Wines implementation of function A by testing it both with the native and the builtin version of D. If you have many such tests it would be a lot of unnessary manual work to switch D between native and builtin while testing B as builtin. Then perhaps somebody else have some use I haven't even though of. The main use will be I think to force tests to as the default test the builtin version since if you type "make test" in a directory you really expect to test the files in THAT directory, not the native version even if your ~/.wine/config specifies that.

1 0

Re: runtest: new option -M for testing current module as builtin
by Francois Gouget Jan. 23, 2002

Jan. 23, 2002

On Wed, 23 Jan 2002, Patrik Stridvall wrote: [...] > BEGIN { > require wine; > &wine::module_native("wininet"); > &wine::module_builtin("imm32", "msacm32"); > } I am not sure any test should require the native version of any dll. We cannot ship native dlls with Wine for obvious reasons, so requiring the native dll would force Wine developpers to have a Windows partition installation around, and one configured for use in the tests. Otherwise they would not be able to run the tests, would they? I don't think that's good. Choosing to use a native dll is more up to the developper to debug a test. And then modifying the configuration file, as we usually do to debug applications, should be simple and familiar enough. But maybe I misunderstood something. -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ May your Tongue stick to the Roof of your Mouth with the Force of a Thousand Caramels.

1 0

Re: wine/windows dialog.c
by Uwe Bonnes Jan. 23, 2002

Jan. 23, 2002

>>>>> "Alexandre" == Alexandre Julliard <julliard(a)winehq.com> writes: Alexandre> ChangeSet ID: 1011730184150795614346862 CVSROOT: Alexandre> /opt/cvs-commit Module name: wine Changes by: Alexandre> julliard(a)wine2. 02/01/22 15:09:44 Alexandre> Modified files: windows : dialog.c Alexandre> Log message: Removed no longer used average width Alexandre> calculation. Well, for the encarta 99 setup this gives the same bad result as before. I have attached a gif of the dialog in question. The manual calculation with the corrections in my resent patch gave better results. Bye -- 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

Microsoft's version of "WINE"
by Sander Wichers Jan. 23, 2002

Jan. 23, 2002

Hi, Maybe you guys already seen this, and it's old news for you... but i'll post this anyway in case you didn't know. It can be a good thing to learn from. I do not own a SUN or a HP-UX system. But I was just interested how microsoft implemented IE 4 / IE 5 on HP-UX and SUN. So I extracted their installation files to disk. It seems that microsoft made their own version of wine. It's amazing that they didn't modified IE to run on unix. Instead they made the choice to emulate all the windows functionality. And compile IE on that. Microsoft is ready to run their applications on linux. They have a complete "system" libary full of ported functionality to SUN and HP-UX... all the libaries are there, libkernel32.so etc.... It's also nice to see how they seperated the resources from the binaries. and let the kernel deal with these resource files. Has any one of you looked into these system libaries? Is this information relevant to anyone? bye, Sander Wichers

1 0

Verbose tests ?
by Sylvain Petreolle Jan. 23, 2002

Jan. 23, 2002

Hi, running programs/winetest/runtest -v -v -v -v -v tests/wine.pl or programs/winetest/runtest -vvvvv only reports : tests/wine.pl: 22 tests executed, 2 marked as todo, 0 failures. Is this the normal behavior or have I made something wrong ? ___________________________________________________________ Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! Yahoo! Courrier : http://courrier.yahoo.fr

2 1

Open Letter to DOJ re Microsoft Settlement
by Dan Kegel Jan. 22, 2002

Jan. 22, 2002

The proposed settlement in the Microsoft antitrust trial will probably be accepted by the judge unless a lot of people complain soon, even though it lacks any useful enforcement provisions and fails to prohibit many of Microsoft's anticompetitive practices. I've drafted an open letter to the Department of Justice to explain why the proposed settlement should not be accepted; it's at http://www.kegel.com/remedy/letter.html I'd be honored if you would read it, and consider signing it with me, or submitting your own comment to the DOJ. Either way, the federal deadline for public comment is Monday the 28th. For more information, see http://www.kegel.com/remedy/ See also Jeremy White's page http://www.codeweavers.com/~jwhite/tunney.html Sincerely, Dan Kegel

2 2

sparc solaris build
by Srdjan Matovina -- Customer Engineering Jan. 22, 2002

Jan. 22, 2002

Hi, I've built wine-20011226 on sparc with the only one issue left open: The resource compiler wrc core dumps when hit the bitmap (BINRES) part of .rc file. There is an alignment problem inside convert_bitmap() in newstruc.c file. The alignment problem came out from BITMAPINFOHEADER struct. But this was just the beginning! I've commented out body of convert_bitmap() just to see what will happen when run libtest/hello test case. After running wineserver, the libtest/hello core with the same problem: Program terminated by signal BUS (invalid address alignment). This time at different place: RtlSecondsSince1970ToTime from ntdll.so (time.c) struct FILETIME. Obviously I'll have an alignment problem with every structure in general. Is there any guideline for this kind of problem? Thanks Srdjan

2 1

Re: [PATCH] Overlapped recv(), WSARecv()
by Martin Wilck Jan. 22, 2002

Jan. 22, 2002

As you may see from the different email address, I am currently off work - I'll look deeper into your traces next week. Many thanks for generating them, anyway. >From a first glance, it seems that the app doesn't do overlapped recv(). (lpOverlapped & completion_func are always NULL), so there should actually be no difference (well I'm now using recvmsg() instead of recv(), and the overhead for 16 bit Winsock apps is definitely larger, but I cannot see why that should slow everything down that much). Perhaps I screwed up the blocking semantics of non-overlapped IO. Have you generated the traces under exactly the same conditions? Can you explain what exactly you did? Martin

3 7

Article about WINE on DesktopLinux.com
by Roland Jan. 22, 2002

Jan. 22, 2002

Hello, I found a very interesting article about WINE there: http://www.desktoplinux.com/articles/AT5502857506.html I have to fully agree with that author, I think he is a real genius! :))) Roland

1 0

Re: Decorationless managed windows again
by Alexandre Julliard Jan. 21, 2002

Jan. 21, 2002

Ove Kaaven <ovehk(a)ping.uio.no> writes: > The logic in this patch is now somewhat revised according to Alexandre's > comments. I did not make this configurable, though, thinking that if the > user's WM really is too old to support them, then he probably wouldn't > mind getting more top-level windows managed anyway... Top-level windows maybe not, but menus and tooltips with wm decorations are pretty annoying... so I applied it, but without the is_window_managed changes, we will need a better algorithm there. -- Alexandre Julliard julliard(a)winehq.com

2 4

Re: Some fixes for dumb X errors :-)
by Alexandre Julliard Jan. 21, 2002

Jan. 21, 2002

Lionel Ulmer <lionel.ulmer(a)free.fr> writes: > PS for Alexandre: we once discussed how we could 'streamline' this X error > handling (ie to prevent doing the same kind of code at three different > locations in the Wine code). Did you progress on this issue or not ? Yes, here's what I came up with. Basically you have to wrap the call that might fail with X11DRV_expect_error/X11DRV_check_error. Would this work for what you need to do? Index: dlls/x11drv/x11drv_main.c =================================================================== RCS file: /opt/cvs-commit/wine/dlls/x11drv/x11drv_main.c,v retrieving revision 1.47 diff -u -r1.47 x11drv_main.c --- dlls/x11drv/x11drv_main.c 2001/11/30 23:15:32 1.47 +++ dlls/x11drv/x11drv_main.c 2002/01/21 18:58:48 @@ -63,6 +63,12 @@ static char *desktop_geometry; static XVisualInfo *desktop_vi; +static x11drv_error_callback err_callback; /* current callback for error */ +static Display *err_callback_display; /* display callback is set for */ +static void *err_callback_arg; /* error callback argument */ +static int err_callback_result; /* error callback result */ +static int (*old_error_handler)( Display *, XErrorEvent * ); + #define IS_OPTION_TRUE(ch) \ ((ch) == 'y' || (ch) == 'Y' || (ch) == 't' || (ch) == 'T' || (ch) == '1') #define IS_OPTION_FALSE(ch) \ @@ -97,12 +103,57 @@ } #endif /* NO_REENTRANT_X11 */ + +/*********************************************************************** + * X11DRV_expect_error + * + * Setup a callback function that will be called on an X error. The + * callback must return non-zero if the error is the one it expected. + * This function acquires the x11 lock; X11DRV_check_error must be + * called in all cases to release it. + */ +void X11DRV_expect_error( Display *display, x11drv_error_callback callback, void *arg ) +{ + wine_tsx11_lock(); + XSync( display, False ); + err_callback = callback; + err_callback_display = display; + err_callback_arg = arg; + err_callback_result = 0; +} + + +/*********************************************************************** + * X11DRV_check_error + * + * Check if an expected X11 error occurred; return non-zero if yes. + * Also release the x11 lock obtained in X11DRV_expect_error. + */ +int X11DRV_check_error(void) +{ + int ret = err_callback_result; + XSync( err_callback_display, False ); + err_callback = NULL; + wine_tsx11_unlock(); + return ret; +} + + /*********************************************************************** * error_handler */ -static int error_handler(Display *display, XErrorEvent *error_evt) +static int error_handler( Display *display, XErrorEvent *error_evt ) { - DebugBreak(); /* force an entry in the debugger */ + if (err_callback && display == err_callback_display) + { + if ((err_callback_result = err_callback( display, error_evt, err_callback_arg ))) + { + TRACE( "got expected error\n" ); + return 0; + } + } + if (synchronous) DebugBreak(); /* force an entry in the debugger */ + old_error_handler( display, error_evt ); return 0; } @@ -297,6 +348,7 @@ screen = DefaultScreenOfDisplay( display ); visual = DefaultVisual( display, DefaultScreen(display) ); root_window = DefaultRootWindow( display ); + old_error_handler = XSetErrorHandler( error_handler ); /* Initialize screen depth */ @@ -328,11 +380,7 @@ */ TSXOpenIM( display, NULL, NULL, NULL); - if (synchronous) - { - XSetErrorHandler( error_handler ); - XSynchronize( display, True ); - } + if (synchronous) XSynchronize( display, True ); screen_width = WidthOfScreen( screen ); screen_height = HeightOfScreen( screen ); Index: include/x11drv.h =================================================================== RCS file: /opt/cvs-commit/wine/include/x11drv.h,v retrieving revision 1.95 diff -u -r1.95 x11drv.h --- include/x11drv.h 2002/01/04 18:27:45 1.95 +++ include/x11drv.h 2002/01/21 18:59:08 @@ -391,7 +391,9 @@ extern void X11DRV_SetFocus( HWND hwnd ); extern Cursor X11DRV_GetCursor( Display *display, struct tagCURSORICONINFO *ptr ); -extern void X11DRV_expect_error( unsigned char request, unsigned char error, XID id ); +typedef int (*x11drv_error_callback)( Display *display, XErrorEvent *event, void *arg ); + +extern void X11DRV_expect_error( Display *display, x11drv_error_callback callback, void *arg ); extern int X11DRV_check_error(void); extern void X11DRV_register_window( Display *display, HWND hwnd, struct x11drv_win_data *data ); extern void X11DRV_set_iconic_state( WND *win ); -- Alexandre Julliard julliard(a)winehq.com

1 0

Some musings about DGA2 support in Wine...
by Lionel Ulmer Jan. 21, 2002

Jan. 21, 2002

Hi all, After being prodded by people on #WineHQ to look at the current DGA2 problem in Wine (it it works fine for graphics, but does not accept any user input at all and needs to either kill Wine remotely or do the three finger salute to the X server), I found the following things : - the DGA2 code (in dlls/x11drv/dga2.c) used the gdi_display connection to open the framebuffer and setting-up all the input handling. But it was the thread->display connection that was used to process the events. This is a no-go as when DGA2 is entered, there is only one connection that receives the event, the one which has grabbed the frame buffer (thanks to Ove to remind me of the 'one connection per thread' thingy). - so I removed all the mentions to 'gdi_display' in dga2.c and replaced them by the current thread display and still no go (although I checked that the event processing in WaitForMultipleEvents and subsequent process_events was done on the correct display). Still no go. So I investigated a bit with a little test program : = I did first try setting up DGA2 in a separate display and then doing all the 'standard' processing on another display (that is Wine does with DGA2 set-up in the init function to test if DGA2 is available and then really use it in another thread). As my test application still was working fine, this was not the problem... = then, I saw in the logs some calls to X11DRV functions after we have entered DGA2 (some due to the DDraw stuff creating DIB sections and so on, some due to the application I think). After adding the same stuff in my test program (ie I created a window just after entering DGA2), I found that it perturbed the event processing (I still had motion events, but neither key presses nor button presses). - so, finally, just to try, I backed off my changes in dga2.c (ie I used again the gdi_display for all the set-up) and had the process_events function in windows/x11drv/event.c use gdi_display instead of data->display when DGA was used (by testing the DGAUsed variable). And now it seems to work just fine.... All this leads now to the 10 � (an Euro sign for all the people not having an iso8859-15 font :-) ) question : how to fix this 'properly' and not with the hack I just did ? The only somewhat 'clean' solution I found would be to create a display connection specifically used for DGA2 (thus we would be sure that no X calls would 'pollute' the link). And in process_events, if DGA2 is used, use this connection instead of the thread connection (we already have some DGA specific variables there, what would be one more :-)) ). After, the question would be how we are sure that it's the 'good' WaitForMultipleEvents that receives the events (if the application is multi-threaded and has multiple threads calling this function). Anyway, if people want just to play some games using DGA2, they can always use my hacky solutions I just described up there :-) Lionel -- Lionel Ulmer - http://www.bbrox.org/

2 2