wine-devel

wine-devel@winehq.org

10 participants
84459 discussions

Re: Road to 0.9
by Ulrich Czekalla 19 Mar '02

19 Mar '02

On Tue, 2002-03-19 at 12:57, Dimitrie O. Paun wrote: > On March 19, 2002 12:48 pm, Andreas Mohr wrote: > > > > > The point is not to freeze the protocol completely, it will clearly > > > continue to evolve even after 1.0. The idea is that once the protocol > > > is declared frozen all future changes are done in a way that preserves > > > backwards compatibility. And the mechanisms to do that are mostly in > > > place now, which is what I mean by saying it is finished. > > > > OK, so in other words the wineserver *is* more or less "independent from > > such changes". > > How can that be? What we want is that DLLs compiled with 1.x, work (at the > binary level) with any 1.y server, where x <= y, right? > As long as you don't change the protocol used you can make as many changes as you like to the wineserver. It shouldn't affect dll compatibility. > -- > Dimi. > >

1 0

Re: Road to 0.9
by Michael Cardenas 19 Mar '02

19 Mar '02

I think this will really bring us a long way to having a stable wine. Does anyone besides lawson have a pet application that they use all the time? Any volunteers? James Hatheway wrote: >>C. Testing apps? I have no real idea what you mean. >> > >Sorry, I guess I should have been more clear, I just typed in >word for word my brief shorthand notes :) > >Basically, the idea that we had was the concept of getting users >to volunteer to "own" an app, ie. responsible for yelling when >we break an app. This already sort of happens, but never >in a formal way. We never decided the details, but I envisiage >having one person volunteering to each test their favourite app on every >release and filing bugs. (in a way that is helpful, we'll have docs for that) > > >-James >

2 1

Re: wineconf summary
by Michael Cardenas 19 Mar '02

19 Mar '02

I tried it and it doesn't work on any platform. I'll let lpbn know now. Sylvain Petreolle wrote: >I know, but it's not a normal situation ... >everyone should be able to see the entire data, >particularily those that don't have Internet Explorer >... >Think you don't have kde ? (like pure RedHat users) > > >>>why does the site lpbn.org does not show the >>> >>summary >> >>>of the conference when we use netscape or mozilla >>> >>? >> >>Works with Konqueror... >> >>LLaP >>bero >> > > >___________________________________________________________ >Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français ! >Yahoo! Mail : http://fr.mail.yahoo.com > >

1 0

[Fwd: Re: Bugzilla, was Re: wineconf summary]
by Michael Cardenas 19 Mar '02

19 Mar '02

Dimitrie O. Paun wrote: >On March 19, 2002 08:49 am, Andriy Palamarchuk wrote: > >>>dimitry suggested posting the top 10 or >>>20 bugs to wine-devel every other week, and I >>>suggested putting the in WWN. >>> >>I'm not sure about usefullness of this one. What we >>need are people, responsible for Wine subsystems. >> > >No, this is from another perspective. And that is, getting more people into >Wine devel, like the Abiword folks do. Problem is that right now the barrier >to entry is enormous: the project is huge, and there is no clear list of >_small_ tasks (which require little knowledge of the project). Going through >Bugzilla is not an option -- few people do that. If we extract a list of >10-20 bugs, and post them on WWN, a _lot_ more people will see them, click on >it, etc. It's just a way of getting people into the game, that's all. > I think it's also important to keep the developers focused. If we just work on whatever comes up, or whatever users ask for at any given moment, we won't focus on getting to 0.9 and it won't happen. I think that ideally we should move to a release schedule like KDE does where they work towards releases at specified dates. There needs to be a time where we all try to do clean up and focus on doing grunt work to solidify releases.

1 0

RE: Linuxism in dlls/msvcrt/msvcrt.spec?
by Medland, Bill 19 Mar '02

19 Mar '02

> -----Original Message----- > From: Gerald Pfeifer [mailto:pfeifer@dbai.tuwien.ac.at] > Sent: Monday, March 18, 2002 8:31 AM > To: wine-patches(a)winehq.com > Cc: Martin Faxer > Subject: Linuxism in dlls/msvcrt/msvcrt.spec? > > > I believe there is a Linuxism in dlls/msvcrt/msvcrt.spec which expects > that _tolower and _toupper exist as functions. > > The following patch by Martin Faxér > <gmh003532(a)brfmasthugget.se> addresses > this; if this patch seems unacceptable, we'd appreciate a > note and a short > explanation why the current code is fine and the patch is not fine. > > Thanks, > Gerald Well actually I think both are wrong; I am just doing the patch. (Actually I don't understand where the _toupper is coming from) The problem is that _toupper is NOT the same as toupper; _toupper does not test that the character is lowercase; it just converts it. The patch is to hook into the existing MSVCRT__toupper and MSVCRT__tolower. Bill

1 0

debugging longman dictionary
by Rizsanyi Zsolt 19 Mar '02

19 Mar '02

Hi! I'm trying to debug what goes wrong with the Longman Dictionary of Contemporary English. I have succeeded installing it under wine on a fake windows installation. (I had some problems, and posted them on wine-user, but I have succeded) Now I get an unhandled exception (page fault on read access). But I cant find the cause of the problem. Maybe somebody could hint me how to proceed. I'm attaching the output of wine --debugmsg +relay LDOCE32.EXE Also I have set up a breakpoint, and stepped through the last 10 instructions before the page fault, but I dont understand what that code is trying to do (it is somewhat strange). Also it may be usefull if you could put in a faq or some other documentation that the next output (which is very frequent), is just a check running debugger. This check is usually done in sharewares to check whether the program is debugged. This is the characteristic output: fixme:win32:DEVICE_Open Unknown/unsupported VxD SICE. Try --winver nt40 or win31 ! fixme:win32:DEVICE_Open Unknown/unsupported VxD SIWVID. Try --winver nt40 or win31 ! fixme:win32:DEVICE_Open Unknown/unsupported VxD NTICE. Try --winver nt40 or win31 ! At first I tried to define these VxD-s, but when I did so the program instracted me to stop the debugger and restart. Thanks for your attention, and I hope you can help me :) Zsolt Rizsanyi

2 1

Status of Wrc port to mingw
by Steven Edwards 18 Mar '02

18 Mar '02

I'm currently trying to port wrc to mingw. Everthing builds fine except when running "bison -bppy -ppp -d -t ./ppy.y" I get "conflicts: 2 shift/reduce". I have started bug 474 in bugzilla if anyone wants to help with this. Thanks Steven _________________________________________________________ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com

2 2

Re: PATCH: files/smb.c
by Francois Gouget 18 Mar '02

18 Mar '02

On Sat, 16 Mar 2002, Gerald Pfeifer wrote: > This patch is required to make the newly contributed files/smb.c compile > on FreeBSD 4.x (and Solaris 2.x, far as I can see). You should test HAVE_NETINET_IN_SYSTM_H and HAVE_NETINET_IN_H. looks good otherwise. > ChangeLog: > Add missing #includes for FreeBSD and Solaris. > > Index: smb.c > =================================================================== > RCS file: /home/wine/wine/files/smb.c,v > retrieving revision 1.1 > diff -u -3 -p -r1.1 smb.c > --- smb.c 12 Mar 2002 19:24:04 -0000 1.1 > +++ smb.c 16 Mar 2002 12:48:20 -0000 > @@ -53,6 +53,8 @@ > #include <utime.h> > #include <sys/socket.h> > #include <sys/types.h> > +#include <netinet/in_systm.h> > +#include <netinet/in.h> > #include <netinet/ip.h> > #include <arpa/inet.h> -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ It really galls me that most of the computer power in the world is wasted on screen savers. Chris Caldwell from the GIMPS project http://www.mersenne.org/prime.htm

1 0

PE loader integrated into BSD
by Dan Kegel 18 Mar '02

18 Mar '02

At Wineconf tonight, people were talking about whether integrating bits of wine into the kernel was useful, and how much should be moved in. (Hello task ornaments!) Anyway, I mentioned that somebody had integrated a PE loader into BSD. Here's the page: http://chiharu.hauN.ORG/peace/ I think this is pretty darn cool, and hope somebody does something similar for Linux. - Dan

3 5

Re: Enhancements for the C testing framework
by Dimitrie O. Paun 18 Mar '02

18 Mar '02

Hi Francois, I still think we should rename the START_TEST macro to DEF_TEST or DEFINE_TEST, 'cause that's what we're actually doing. -- Dimi.

1 0

[Fwd: Re: wineconf summary]
by Michael Cardenas 18 Mar '02

18 Mar '02

Dimi, sorry about the spelling! I'll correct that. >Also, we need to: > 1) post the pics > 2) pick the best group picture, and annotate it >I'll bet you the annotated picture will be the most popular thing. > >When all is done, we need to make an announcement on Slashdot. > I'll send my pics tonight to andi, since I left my digital camera at home.

1 0

Re: Heap allocation problem
by Duane Clark 18 Mar '02

18 Mar '02

Duane Clark wrote: > Uwe Bonnes wrote: >>Just a shot in the dark: Did you perhaps install on a real WinNT machine and >>now run wine in --winver win95 mode or vice versa? >>http://support.microsoft.com/default.aspx?scid=kb;EN-US;q125713 >>points out some possible problems. > > > I'm running a dual boot Win98/RH7.2 system, and booted Win98 to install > the app. I don't ever supply a winver, but have Tweak.Layout set for > Win98. The app ran fine (except for the toolbars) prior to the addition > of anon_mmap_aligned() this past October (I had not run it for awhile, > but needed it now again). It looks like I never mentioned the circumstances where this error occurs. Most of the app comes up and runs, but there is a large area where messages are printed that uses a richedit control. Since the wine richedit does not work, I use a Win98 native richedit. What happens along with the heap error, is that the richedit control is not drawn. Instead the desktop remains showing through a large area in the center of the app. With my quicky "fix", the entire app including the native richedit works fine. However this is not specific to richedit. I have another test app I put together that uses richedit, and it works fine without the heap "fix".

1 0

TurboTax
by Kevin O'Connor 18 Mar '02

18 Mar '02

> I've been following Wine development for the last 6 months, mostly watching > for improvements for the 2 Windows applications I actually care about -- > Quicken and TurboTax. [...] > TurboTax was basically unusable until recently -- but when I returned from > a short vacation and updated my wine sources, all the show-stopping issues > with TurboTax had been resolved. Amazing. Hi Jeff, After seeing your note, I synched up my cvs and attempted to run turbo tax as well. After a brief failure due to a missing NdrDllUnregisterProxy function (which I fixed using the patch below), I too was impressed to see turbo tax running. > The one issue in both tools that I still need to check on is both > consistently ran out of GDI space, which I've hacked around with some > tweaks to gdiobj.c. Unfortunately, I quickly ran into this problem as well. I am not familiar enough with wine to fix this error by myself - would you be willing to push your hack along so I can take a look at this also? Thanks, -Kevin --- dlls/rpcrt4/rpcrt4.spec 19 Feb 2002 18:37:26 -0000 1.13 +++ dlls/rpcrt4/rpcrt4.spec 18 Mar 2002 03:34:28 -0000 @@ -26,7 +26,7 @@ @ stdcall NdrDllCanUnloadNow(ptr) NdrDllCanUnloadNow @ stdcall NdrDllGetClassObject(ptr ptr ptr ptr ptr ptr) NdrDllGetClassObject @ stdcall NdrDllRegisterProxy(long ptr ptr) NdrDllRegisterProxy -@ stub NdrDllUnregisterProxy +@ stdcall NdrDllUnregisterProxy(long ptr ptr) NdrDllUnregisterProxy @ stub RpcAbortAsyncCall @ stub RpcAsyncAbortCall --- dlls/rpcrt4/rpcrt4_main.c 9 Mar 2002 23:39:12 -0000 1.21 +++ dlls/rpcrt4/rpcrt4_main.c 18 Mar 2002 03:34:28 -0000 @@ -363,7 +363,20 @@ HMODULE hDll, /* [in] */ const ProxyFileInfo **pProxyFileList, /* [in] */ const CLSID *pclsid /* [in] */ -) +) +{ + FIXME("(%x,%p,%s), stub!\n",hDll,pProxyFileList,debugstr_guid(pclsid)); + return S_OK; +} + +/*********************************************************************** + * NdrDllUnregisterProxy (RPCRT4.@) + */ +HRESULT WINAPI NdrDllUnregisterProxy( + HMODULE hDll, /* [in] */ + const ProxyFileInfo **pProxyFileList, /* [in] */ + const CLSID *pclsid /* [in] */ +) { FIXME("(%x,%p,%s), stub!\n",hDll,pProxyFileList,debugstr_guid(pclsid)); return S_OK; -- ------------------------------------------------------------------------ | Kevin O'Connor "BTW, IMHO we need a FAQ for | | kevin(a)koconnor.net 'IMHO', 'FAQ', 'BTW', etc. !" | ------------------------------------------------------------------------

2 1

Heap allocation problem
by Duane Clark 18 Mar '02

18 Mar '02

Something else to look at when people get back from wineconf. One of my apps (Actel Designer) was generating the error message from HEAP_CreateSystemHeap(): system heap base address 65430000 not available This results from the call to MapViewOfFileEx() with the address SYSTEM_HEAP_BASE, which is 65430000. The comment in the header of MapViewOfFileEx() for this parameter, addr, says it is a "Suggested starting address for mapped view". That then calls anon_mmap_aligned(). In this particular application, the address 6543xxxx appears to have already been allocated elsewhere, so anon_mmap_aligned() appears to have allocated a valid block, but at a different address. Then at the bottom of the routine, the test else if (ptr != base) determines that these two addresses do not match and then unmaps the allocated block, leading to the heap allocation failure. Oddly enough, back in October, when this problem first appeared (with the introduction of anon_mmap_aligned()), HEAP_CreateSystemHeap() contained the comment "pre-defined address not available, use any one", but later on the "use any one" part was removed. So I tried commenting out the unmapping in anon_mmap_aligned(), and returning the allocated pointer, and my program is back to working fine. But I am not anywhere near understanding what the right fix is.

2 1

correct way to get windows version
by michael cardenas 18 Mar '02

18 Mar '02

What's the correct way to determine the current windows version wine is using from advapi32.dll? I looked through the list archives and in the documentation and I found VERSION_GetVersion(), but if I use that advapi32 won't compile. I'm preparing my patches for aol7, but I know that one of them is specific to windows95, and it won't be accepted if I don't put the version check in.

2 1

starcraft don't work well , please fix the problem
by centi_riccardo＠libero.it 17 Mar '02

17 Mar '02

Starcraft (blizzard game) don't work well on wine . I see a black screen and nothing other . It's a problem only of the lastest wine releases ........

1 0

WineConf streaming
by Eric Pouech 16 Mar '02

16 Mar '02

if someone reads his mail, I'm tryin to follow up the conf with the lpbn streaming but the sound is awful (R/N ratio horrible) so, if you could pass the info (and try to fix it) TIA A+

5 7

starcraft don't work in the last wine releases
by centi_riccardo＠libero.it 16 Mar '02

16 Mar '02

Starcraft don't work well in the last wine releases . I see a black screen while the game is running in backgroung (I can hear music and sounds work well ) . I have a slackware 8.0 Xfree 4.2.0 kernel 2.4.18 (slackware - current) . Please somebody send a patch to resolve this problem .........

1 0

Visual FoxPro and "GDI Regressions"
by Rick Romero 15 Mar '02

15 Mar '02

- Previously posted to wine-users (but with LESS info, worded worse :) As much as I think I won't be able to accomplish anything, I keep trying, and I've come across something interesting.. I'm testing my regression by doing the following: - Start the app - Press CTRL-V to paste text into the Command Window (which is the active 'MDI' window - but not really MDI - by default) - Switch to another virtual desktop - Switch back to Foxpro Desktop (That makes the Command Window 'look' active, by becoming blue instead of grey) - Click on the text that has been pasted. All builds after 5/10/01 will have the problem. The text will highlight as I move the mouse, as if I was still holding the mouse button down. This behavior is the same on scrollbars also. I normally run FoxPro with: wine /mnt/windows/DevStudio/vfp/vfp.exe Trying to discern some useful information, I've been playing with debugmsg. I've noticed if I do: wine --debugmsg +scroll,+event,+toolbar,+gdi,+heap /mnt/windows/DevStu dio/vfp/vfp.exe Clicking works as it should.. Scrollbars take two clicks before the ButtonRelease isn't acknoledged, but the text highlighting doesn't occur. I can click and move around normally. If I attempt to highlight text, (holding down the button) the ButtonRelease then is ignored as before.. If I cut out some of the debug info: wine --debugmsg +scroll,+event,+toolbar /mnt/windows/DevStu dio/vfp/vfp.exe We're back to normal (non-working normal). It seems like there is some timing issue that the (almost extreme) slowness from printing the debugging info is overcoming..(it's just going to stdout) Is there a click/release timer that could be increased? I've posted a small application at http://www.havokmon.com/stuff/vfp5test.zip . It uncompresses to a setup.exe and files, which includes the VFP5 Runtime, and a test.exe application that just opens a MDI window. The above test works on this application. http://www.havokmon.com/stuff/vfp5test.txt has more information on setup and installation with Wine (because it's not 100%). Hopefully everyone won't be too WineConf'd out next week :) Thanks, Rick

1 0

Borland C 5 - 16bits
by lists＠vmn.com.br 14 Mar '02

14 Mar '02

Hi, I am trying to run a point of sale app with the following structure: i) 16 bit ii) written in Borland C 5 iii)uses DDE iv) The app is broken into some 4 or 5 EXE and DLL files that communicate with the windows POST MESSAGE function. Two apps communicate through DDE. Does anyone have experience with applications written with Borland C in wine? It seems that 16 bit Apps written in Borland C are not working as in win98. The thing is I have tried about 30 apps. Everything written in Visual C (16 and 32 bits) works. Almost all apps written in Borland 32 bits, including OWL and BWCC apps work. This is a quite impressive showing for wine. 16 bit Apps written in Borland C seem to perform poorly. Running the apps with --debugmsg warn+all shows consistent problems with GDI: use of invalid pointers HEAP: adressing attempts of invalid addresses Furthermore, it seems that the apps are not correctly sharing the processor. For the program to run it is necessary to click in all the windows and therefore allocate processor time by giving focus. Does anyone have experience with Borland C 16 bit apps. Is your experience consistent with this report? I certainly would appreciate any pointers. Thanks, Miguel Feitosa p.S : Eric -> I have changed the Apps to not use DDE. I am going to do a couple more days of testing. Then I will roll back the app to the original state and generate some traces for DDE. ------------------------------------------------- This mail sent through IMP: webmail.vmn.com.br

1 0

aRts audio driver support
by Chris Morgan 14 Mar '02

14 Mar '02

*dlls/winmm/winearts, arts.c, arts.h, audio.c, mixer.c, mmaux.c, winearts.drv.spec, Makefile.in: Chris Morgan <cmorgan(a)alum.wpi.edu> Added aRts driver to wine *configure, configure.ac, dlls/Makefile.in, include/config.h.in, dlls/dsound/dsound_main.c, documentation/samples/config: Chris Morgan <cmorgan(a)alum.wpi.edu> Changes for aRts driver, added commented out entry to WinMM section of sample config file, added configure time checks for artsc support. One question I had was what to do about EXTRADEFS entry in dlls/winmm/winearts/Makefile.in? Is this ok or does it have to be a configure detected thing? Any comments about this patch are more than welcome. Thanks, Chris

2 1

fix for crash in snoop
by michael cardenas 14 Mar '02

14 Mar '02

I can submit my previous patch for snoop also. It removes functionality but fixes the crash. Should I submit it? ------------------------------------------------------- Changelog: - this patch should get rid of the remaining crashes in snoop. It fixes SNOOP_GetProcAddress. ------------------------------------------------------- COPYRIGHT: All original work contained in this email is released under the terms of the old Wine license (LICENSE.OLD in the distribution). This means that any new functions I have created can be incorporated into a non LGPL Wine fork, for example. Any future patches I submit can be considered to be release under the old Wine license.

1 0

Re: Segmentation fault (core dumped) with cvs update
by Tony Lambregts 14 Mar '02

14 Mar '02

> I have upgraded my internet access recently and have decided to stay > current with the cvs instead of waiting for each new snapshot. I however > have run into a snag > > I downloaded the latest snapshot from ibiblio.org (Wine-200202.28.tar.gz) > and the latest cvsdirs from ftp.winehq.com/pub/wine > (wine-cvsdirs-20020228.tar.gz) > > After following the steps to get a CVS sandbox I now have a local copy > of the wine located at /usr/local/src/wine. > > When I cd to this directory an try to update my copy i get the following: > > [root@tony wine]# cvs u -v > > Concurrent Versions System (CVS) 1.10.5 (client/server) > > Copyright (c) 1989-1998 Brian Berliner, david d `zoo' zuhn, > Jeff Polk, and other authors > > CVS may be copied only under the terms of the GNU General Public License, > a copy of which can be found with the CVS distribution kit. > > Specify the --help option for further information about CVS > > [root@tony wine]# cvs update -PAd > Segmentation fault (core dumped) > > Any help or suggestions would be appreciated > > Tony Lambregts > Ok just to follow up. I upgraded my version of cvs and life is good. Tony Lambregts

1 0

Re: [Fwd: Re: Oh yeah, the docu...]
by Tony Lambregts 14 Mar '02

14 Mar '02

> On Wed, Feb 20, 2002 at 09:40:43AM -0700, Tony Lambregts wrote: >> Tony Lambregts wrote: [Snip] >>> Ok so its just the online doc thats not up to date. Thanks. >>> >>> http://www.winehq.com/Docs/wine-user/configuring.shtml >>> >>> also looks like the the following are also not up to date. >>> >>> http://www.winehq.com/Docs/winedoc-sgml.tgz >>> http://www.winehq.com/Docs/winedoc-pdf.tgz VERY OLD >>> http://www.winehq.com/Docs/winedoc-ps.tgz VERY OLD >>> >>> and I would suspect that these too. (did not actually download) >>> >>> http://www.winehq.com/Docs/wine-set-html.tgz >>> http://www.winehq.com/Docs/winedoc-html.tgz >> [snip] >>> Since my last post on usenet I did download these and they are >>> out of >> date too. IMHO www.winehq.com/Docs should be updated at each >> snapshot. > Yeah, like I was really dying for finally seeing this being done. > > This time I yelled yet again, and Jeremy White muttered some Holy > Promise to make it finally work. (which, on hindsight, is a > relatively cheap promise due to Codeweavers having to go through > the trouble of revamping the whole winehq.com domain anyway due to > Corel's unfortunate demise ;-) > > Anyway, I'll be *really* glad once this finally gets done in a nice > solid timely cronjob. > > -- Andreas Mohr Stauferstr. 6, D-71272 > Renningen, Germany Tel. +49 7159 800604 > http://home.nexgo.de/andi.mohr/ > After checking winehq.com today I see that the online documentation is still not current. I also downloaded the sgml, pdf, and both htmls. They are still out of date. Now I realize that the staff at codeweavers are busy people but I would really like to see these up to date. Could I have some indication as to when this will done. Tony Lambregts Ps: I miss the network news service very much and want it to come back.

1 0

Has CVS access changed?
by Duane Clark 13 Mar '02

13 Mar '02

I just noticed that the patch I submitted is against an old version of CVS, even though I did the "cvs diff" today (3/13). Did I miss a change to the CVS configuration, perhaps in the move to the new server? I currently have .cvspass with :pserver:cvs@cvs.winehq.com:/home/wine Ah<Z

1 0

Jump to page:

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