June 2004 - Wine-Devel - list.winehq.org

Re: [USER] 'Fix' SetWindowsHookEx
by Alexandre Julliard June 20, 2004

June 20, 2004

Lionel Ulmer <lionel.ulmer(a)free.fr> writes: > I have actually no idea what exactly is what Windows is doing there, but > this attached patch seems to do the trick. I have no idea if it's correct > though (as I do not know if what GetModuleHandle returns is really the same > thing as what SetWindowsHookEx expects as an HINSTANCE input). This doesn't look right, it doesn't make sense to create a system-wide hook without an instance, since in that case there is no way to call the function in other processes. This requires more investigation to find out exactly what Windows does in this case. -- Alexandre Julliard julliard(a)winehq.org

2 4

Re: docu update
by Alexandre Julliard June 20, 2004

June 20, 2004

"Ivan" <puoti(a)inwind.it> writes: > This patch has been written thinking that a) it's better to have no doc than an > out of date one and b) most of the stuff that's commented out can be easily > re-written before the next release (I have school holidays until september so I > have lots of time to do this). wineinstall is not out of date, it still works fine, I don't see any reason to remove it from the doc. -- Alexandre Julliard julliard(a)winehq.org

6 20

Device drivers in Wine
by Damjan Jovanovic June 20, 2004

June 20, 2004

Hi Basically, I have a Windows user-space device driver, and I need to get it working under Linux. I note Wine has some "device" functions (FILE_CreateDevice(), and DOSFS_OpenDevice()). How do these work and what do they do? Are they documented anywhere? And is there a way to open and use a Windows DLL from a Linux program? Thanks Damjan __________________________________ Do you Yahoo!? Yahoo! Mail - Helps protect you from nasty viruses. http://promotions.yahoo.com/new_mail

2 3

winemaker problems for creating winelib DLL projects
by Shachar Shemesh June 20, 2004

June 20, 2004

Hi all, I'm trying to create a winelib DLL project from scratch. You can reproduce using the following steps: create a file called "name.c", put the following content in it: int func( int a ) { return a+1; } Create a file called "name.spec". Put the following content in it: @ cdecl func( long ) In this directory, run "winemaker --dll ." Run "make". I would expect, at this stage, to get a file called "name.dll.so". Instead, I get a file called "name.dll.exe.so", with which I cannot link other applications, even if I try to rename it to "name.dll.so". Am I doing something wrong here? Shachar -- Shachar Shemesh Lingnu Open Source Consulting ltd. http://www.lingnu.com/

2 2

Re: docu update
by Ivan June 20, 2004

June 20, 2004

> Why is that? I have no idea, but X11drv doesn't initialize properly. This has been reported by users, I don't have x.org myself. Ivan.

2 1

Attn Christian - Last texops patch
by Ann and Jason Edmeades June 20, 2004

June 20, 2004

Hi Christian - I tried to sent you an email but it bounced... Are you sure that patch is correct (I've only glanced at it though!), as I thought the src1 was set from GetSrvcAndOpFromValue, and this should already handle the ONE_MINUS* type values? Jason

1 0

Re: Fix winedbg stack_backtrace on non i386
by Eric Pouech June 20, 2004

June 20, 2004

emmanuel maillard a écrit : > Hi, > > stack.c fail to build on ppc host due to CONTEXT_SEGMENTS flag, wich is > only define for i386 host. > > > ChangeLog: > Fix compilation error on non-i386 in stack_backtrace. > we shouldn't need the __i386__ tests in the winedbg code, except in the CPU specific files (be_*.c). IMO, we should do something like this: Index: stack.c =================================================================== RCS file: /home/cvs/cvsroot/wine/wine/programs/winedbg/stack.c,v retrieving revision 1.3 diff -u -u -r1.3 stack.c --- stack.c 4 Jun 2004 00:59:16 -0000 1.3 +++ stack.c 20 Jun 2004 07:45:37 -0000 @@ -119,7 +119,11 @@ return; } memset(&ctx, 0, sizeof(ctx)); - ctx.ContextFlags = CONTEXT_CONTROL | CONTEXT_SEGMENTS; + ctx.ContextFlags = CONTEXT_CONTROL +#ifdef CONTEXT_SEGMENTS + | CONTEXT_SEGMENTS +#endif + ; if (SuspendThread(thread->handle) == -1 || !GetThreadContext(thread->handle, &ctx))

1 0

Re: Msg reply
by Johane June 20, 2004

June 20, 2004

1 0

RE: Incoming Msg
by Twickline June 20, 2004

June 20, 2004

1 0

Lotus SmartSuite '97 nonfunctional
by Dan Kegel June 20, 2004

June 20, 2004

In the same yard sale where I bought Lotus SmartSuite '96 (whose Lotus 1-2-3 works under wine-20040615), I also bought the '97 version. I had numerous problems with the '97 version. Its installer failed while installing Word Pro (though it seemed to install Lotus 1-2-3). Lotus 1-2-3 fails with numerous errors when loading a small file or even starting without any file. Something (maybe heavier use of OLE?) changed from the '96 to '97 versions such that Wine is much less happy running the '97 version. If anyone's interested, I can post more details. - Dan -- My technical stuff: http://kegel.com My politics: see http://www.misleader.org for examples of why I'm for regime change

1 0

Re: docu update
by Ivan June 20, 2004

June 20, 2004

> Why is that? Adam Schreiber may know as slackware he build 2 binaries for slack users using xfree86 and x.org, you can reach him at shipssadam -at - clemson.edu Ivan.

1 0

Re: docu update
by Ivan June 20, 2004

June 20, 2004

> Maybe, but for example Mandrake users won't get wine listed in the "remove > software" utility, and that will be confusing > Solving that problem is a desktop issue, I've done a bit of work on it but nothing has got into mainstream yet. OK, but in the mean time I still don't see how one binary can fit all needs, if a binary build for X.org wo'nt run on XFree86 and vice-versa > I don't see why it'd slow down after we freeze libwine, it's not that > many functions... Not all users care, some people still use linux 1.0 > So the user has to upgrade ... and ? use CVS even if he didn't/doesn't want to, that isn't a bug issue but it also isn't a nice thing. > So it could have been anything, in other words. I'd need to see hard > figures in a controlled environment. I can send you a copy of Linux Magazine, I should still have a copy somewhere. You'll have to translate it with google or something because it's in Italian. Also, form sutopackage.org > Nobody should produce any .package files for their own projects until we've hit version 1 I guess we won't be using it too soon anyway.

1 0

IconsDir
by Vitaly Lipatov June 20, 2004

June 20, 2004

Does anyone knows why we have not in registry Software\\Wine\\Wine\\Config\\Wine\IconsDir entry? winemenubuilder can't extract any icons without this dir. -- Lav Vitaly Lipatov Russia, Saint-Petersburg GNU! ALT Linux Team! LaTeX! LyX!

1 0

Re: docu update
by Ivan June 19, 2004

June 19, 2004

About building one binary of wine for all distros, it's interesting to notice that one the same distro you need different binaries depending on if you're using xfree86 or x.org. Ivan.

2 1

Re: docu update
by Ivan June 19, 2004

June 19, 2004

> Anyway, Wine doesn't have any menu > integration. And why shouldn't we have the wine programs in the menu? > Compiling it as i586 or i686 should be fine, very few people use > processors old enough to barf on it Sure. > That's what apbuild solves. It lets you compile for glibc 2.2 on a 2.3 > system, amongst other things. And what if the user wants a glibc 2.3 build? > Yes, we still do that, but most people use the Loki installer - in fact > I'm pretty sure it's the most popular by a long way. Maybe, but for example Mandrake users won't get wine listed in the "remove software" utility, and that will be confusing > 1.0 implies API stability, not that it's suddenly good enough that > people won't want to upgrade. I presume by then wine won't change so quickly. And the average user (the average pc user, not the average linux geek) is lazy and often doesn't upgrade if he doesn't have to > That's entirely wrong, it works fine. What did you think the recent > high-gig mmap patches were for? they won't help wine-20040615 any more. > Also some distros are built for i386, others for i586 and others for i686, how can you > have one binary for all, force them to all use i386 (That will be about 30% > slower that a i586 build)? > Just build them for i586 or i686, like I said, that should work just > fine. I really doubt anybody is running Wine on a system that can't cope > with an i586 compiled binary. 30% sounds extremely high anyway, I'd want > to see hard figures on that. That was a benchmark done by comparing Red Hat and Mandrake, it was a few months ago. > Sure, it may be possible to build one binary and run > it anywhere, but in most cases it wouldn't run as well (and as fast) as a build > done natively on the distro. Actually, yes it is. Go read the docs on autopackage.org to find out how. > would be somewhat suboptimal as we sacrifice speed for flexibility and distro neutrality. Do we NEED to sacrifice such things? Also we can't use it until version 1.0, it will be reasonable to look at this then. Ivan.

2 1

Re: docu update
by Ivan June 19, 2004

June 19, 2004

> (a) and (b) can be solved by WineHQ providing its own distro-neutral, run > anywhere binary packages. This isn't easy to do, and it means you have to A) Leave out some features (mime types, menus, that most distros handle in a different way) or B) package menu/mime stuff for all distros, meaning the user downloads a big binary but will only need part of it. Also, it means we have it compiled for only one architecture. But not all distros use the same one. For example, all software on Mandrake linux is built for i586, and I don't see why that should change. Also how can you have one binary work on platforms that are so different? For example, some users will still use (and want to use) systems that use glibc 2.2, so what do you do, ship 2 binaries for the different versions of glibc? Also, lots of users like to install software using the native packaging system of their distro, infect CodeWeavers now provides native Debs and RPMs of crossover, or at least it did at some point (Not sure if this is still true). Now, if wine was shipped on a CD (Could be done for 1.0 that won't change for some time), we could just have an installer that detected the distro, and installed the appropriate files, but if users have to download (Remember many users still use 56k modems, and that will stay true for quite some time) they only want to have what they really need. I think we won't be able to have a single binary that works on all distros, they are all just to different, for example wine currently doesn't work on FC2 at all, but works perfectly on other distros. Also some distros are built for i386, others for i586 and others for i686, how can you have one binary for all, force them to all use i386 (That will be about 30% slower that a i586 build)? Sure, it may be possible to build one binary and run it anywhere, but in most cases it wouldn't run as well (and as fast) as a build done natively on the distro. Ivan.

2 1

Yum repository for wine
by Leonardo Quijano Vincenzi June 19, 2004

June 19, 2004

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, I can't help but notice that there's no Yum repository for Wine. Or at least I haven't found it yet ;). So, would it be too hard to setup one? In case anyone needs instructions, there is a tutorial here: http://fedoranews.org/alex/tutorial/yum/ Leonardo -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFA1JEnnuk7PwJlZIURArRtAKC0tlERDAHam+J5TbOTclwSZS/aXgCgsQl0 3IwAd4mFH4JLFx4aOcEy7pk= =Cwf0 -----END PGP SIGNATURE-----

2 1

Investigating a crash at startup
by Phil Krylov June 19, 2004

June 19, 2004

Hello, I have a program here (IBM Translation Manager v.6.0.4) which crashes wine at startup. The program loads a DLL, and Wine starts importing this DLL's imports from kernel32.dll. It crashes on the first import in dlls/ntdll/loader.c:453, which says: thunk_list->u1.Function = (PDWORD)find_named_export( imp_mod, exports, exp_size, pe_name->Name, pe_name->Hint ); This causes an exception because thunk_list happens to be located in .rdata section of the DLL, which is readonly. Currently I have hacked this around by making all PE sections read-write in map_image() in dlls/ntdll/virtual.c, and the program works. If a Wine guru would like to investigate this problem, I can give any support (e.g., ssh access to my computer with the program). If not - I will be using this hacked version of Wine. -- Best regards, Phil mailto:phil(a)newstar.rinet.ru

2 4

Fax Message Received
by Twickline June 19, 2004

June 19, 2004

1 0

Re: Thanks :)
by Twickline June 19, 2004

June 19, 2004

1 0

Forum notify
by Johane June 19, 2004

June 19, 2004

1 0

Where does richedit20.dll fit in the to-do list?
by Dan Kegel June 19, 2004

June 19, 2004

I keep running into apps that won't run without richedit20.dll, yet I can't see any mention of that DLL in the wine 0.9 or wine 1.0 tasklists. Surely it's important enough to merit at least a mention in the wine 1.0 tasklist? - Dan -- My technical stuff: http://kegel.com My politics: see http://www.misleader.org for examples of why I'm for regime change

3 3

font problems with "Hello" application
by Matt Beaumont June 19, 2004

June 19, 2004

There is a font that displays _extremely_ poorly in the tooltips for the "Hello" photo-sharing application (<http://www.hello.com/>, harmless(?) registration required for download). I'm running the 6/15 wine source package, installed with "make install". Hello wants riched20.dll before it will do anything useful; this can be acquired at <http://www.dll-files.com/dllindex/dll-files.shtml?riched20>. The problem can be reproduced by bringing up a tooltip for one of the checkboxes on the login screen; it shows a completely unreadable string. I've been looking at the output from the font and tooltip WINEDEBUG channels, and haven't found anything screamingly obvious. Does anybody have a suggestion for other channels to look at, or places to look in the source? Thanks, Matt

2 3

Re: docu update
by Ivan June 19, 2004

June 19, 2004

> I want to be able to say to a user who reported a bug "please try latest CVS > and confirm that it is fixed" without having to tell them about > configure or LD_LIBRARY_PATH or whatever If they've reported a bug, and you've told them to try the CVS, then it's not a configuration problem, so configure has been run and LD_LIBRARY_PATH is set correctly. Apart from this, all this sort of stuff can be done in the build system anyway. Also removing wineinstall is on the todo, and it does nothing that can't be handled (more transparently) elsewhere. Ivan.

1 0

Re: docu update
by Ivan June 18, 2004

June 18, 2004

> wineinstall is not out of date, it still works fine, I don't see any > reason to remove it from the doc. I see no reason to leave it, it's now as if it did anything essential for wine. Usually things are first removed from wine, and then from the documentation, this causes an afoul lot of confusion, because people tend to really trust the docs. So it would be a good idea to remove the docs about wineinstall before we remove wineinstall, and as wine can be built and run perfectly well without it I don't see why we should keep it in the docs. Also the documentation about the config will have to be re-written, as soon everything will be done with winecfg of regedit, and as that will take a non trivial amount of time, it's a good idea to start well in advance, so the docs aren't behind wine when the switch is done. Ivan.

1 0