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

Question about mmap issues on FreeBSD/NetBSD
by Yorick Hardy July 29, 2004

July 29, 2004

Just a question about the mmap issues experienced on FreeBSD, i.e. http://www.winehq.org/?issue=230#Wine%20on%20FreeBSD I encountered the same on NetBSD. It seems the error message is set from ld.elf_so in NetBSD and ld-elf.so.1 in FreeBSD, when the address space is mapped for the run time linking? Since MAP_NORESERVE is not documented (and not supported?) for mmap on FreeBSD and NetBSD, it would appear that wine and the ld.elf_so together require too much memory for all the mmap calls. Am I correct? Kind regards, -- Yorick Hardy

2 1

Re: Is bugzilla worth keeping?
by Dan Kegel July 29, 2004

July 29, 2004

Mike Hearn wrote: > So are we misleading users by having a bugzilla into thinking that if > they file a bug there, it'll be fixed when it probably won't? Not at all. Just because not enough people are doing bug triage or fixing doesn't mean we should drop bugzilla. Somebody will eventually go through the bug reports, I think. > If we were to simply drop bugzilla, how would it impact the project? It would deprive us of long-term memory. For instance, I filed the bug report http://bugs.winehq.org/show_bug.cgi?id=1404 in January 2003 with a nice test case (but one that's hard to translate into a unit test). Without Bugzilla, we'd have a harder time remembering problems like that. (Unless cxtest really takes off, I suppose...) - 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

Re: Wine and locales
by Filip Navara July 29, 2004

July 29, 2004

1 0

Re: winetests/winrash
by Rolf Kalbermatter July 29, 2004

July 29, 2004

Chris Morgan <cmorgan(a)alum.wpi.edu> wrote: >I wouldn't mind adding a systray icon. I wasn't really sure how that was >supposed to be done. I tried having the service add a systray icon without >much success, I don't believe services can interact with the desktop. That >may mean that a separate winrash monitor may have to be written that >communicates with the service. Services can interact with the desktop but not by default. It is a special option you specify when installing the service. The magic for this is in the dwServiceType parameter of the CreateService() to register a service object in the system. >From MSDN: SERVICE_FILE_SYSTEM_DRIVER File system driver service. SERVICE_KERNEL_DRIVER Driver service. SERVICE_WIN32_OWN_PROCESS Service that runs in its own process. SERVICE_WIN32_SHARE_PROCESS Service that shares a process with one or more other services. For more information, see Service Programs. If you specify either SERVICE_WIN32_OWN_PROCESS or SERVICE_WIN32_SHARE_PROCESS, and the service is running in the context of the LocalSystem account, you can also specify the following type. Type Meaning SERVICE_INTERACTIVE_PROCESS The service can interact with the desktop. Not sure how winrash is currently installed but if it is not through a small Win32 setup app but an InstallShield app, I'm sure it has according parameters to let it install as interactive service Rolf Kalbermatter

2 1

Fwd: Re: rsabase.dll
by Michael Jung July 29, 2004

July 29, 2004

I'm sorry, if this mail appears three times on the wine-devel list. I've sent it twice already, but it doesn't seem to get forwarded to the wine-devel list. Is there a problem with the mailing list? Rob, could you give me a short notificaton, if you have received those mails? Thank, Michael ---------- Forwarded Message ---------- Subject: Re: rsabase.dll Date: Tuesday 27 July 2004 10:08 From: Michael Jung <mjung(a)iss.tu-darmstadt.de> To: wine-devel(a)winehq.org Cc: rob(a)codeweavers.com Rob, > I think there are many reasons to favour "real handles": If a long running > process is using rsabase (e.g. a webserver serving https: requests) with > many temporary connections and many keys, resource leakage is an issue. > Furthermore, since rsabase is a cryptographic service provider, it might be > important that a key, which is not referenced anymore is in fact destroyed > (and the > corresponding memory be nulled). I would think that otherwise you will > over time accumulate a lot of keys in your address space, which is probably > a security problem. Now, I'm aware that wine is not in a state today to be > used for security critical applications and first rsabase implementations > will probably have security issues more severe than this. However, sometime > I think rsabase will have to use real handles. Why not right from the > start? I've thought about this statement again and realized that I'm wrong and that I probably misunderstood you. I realized that one doesn't need "real handles" in user mode code at all to implement correct resource management. And I think I understand that one does need them in the kernel (aka wineserver) for objects, which are accessed from multiple threads. However, there are some instances of "real handle" implementations outside wineserver in wine (e.g. dlls/msi/handle.c). Do you (or anybody else on the mailing list) now what for? Some things, which I could imagine would still be easier to implement correctly with "real handles", are the INVALID_HANDLE_VALUE value, checking that a thread does not access a resource via an already released handle and in general checking in a precise manner if a handle value is valid (which, as I understand it, is often done in wine with a magic value at the beginning of the object.) Greetings, Michael -- Michael Jung, MSc Be liberal in what you accept, Integrated Circuits and Systems Lab and conservative in what you send. University of Technology, Darmstadt -- Jonathan Postel, Internet Pioneer phone: +49(6151)16-6692 / email: mjung(a)iss.tu-darmstadt.de

1 0

Is bugzilla worth keeping?
by Mike Hearn July 29, 2004

July 29, 2004

One thing I notice about most other open source projects is that they have many more flamewars than we do. So, I thought I'd start one: It strikes me, looking at the wine-bugs list, that there is a huge disparity between the number of people maintaining it and the number of people filing bugs in it. It seems to be quite rare for communication on bugfixes to take place there, wine-devel is the more usual forum. So are we misleading users by having a bugzilla into thinking that if they file a bug there, it'll be fixed when it probably won't? If so, does it matter? If we were to simply drop bugzilla, how would it impact the project? Thoughts, anybody? thanks -mike

10 10

winetests/winrash
by James Hawkins July 29, 2004

July 29, 2004

What is it exactly that winrash does? Is it running winetests using the native libraries of the Windows operating system that the user is running the tests on, or are the wine libraries being tested through some means? Is there a site where this information is readily available? I have one suggestion: users typically like to see results when they volunteer their computer time to a project such as this, which is understandable, so a future version might benefit from a system tray icon that can be opened up to see the test running and/or display results of the tests that have been run. -- James Hawkins

2 1

How do we handle unicode (utf-8) codepages?
by Shachar Shemesh July 29, 2004

July 29, 2004

What do we do if the user requests a Unicode codepage? I've just had a conversation with a user on IRC where I believe things didn't work for him because LANG was set to es_ES.UTF-8. On one hand, I don't think we can force ANSI applications to work in Unicode. On the other hand, we cannot change the locale - it will mean noone (whether ANSI or not) will not be able to access non-ASCII file names and files. Comments? Shachar -- Shachar Shemesh Lingnu Open Source Consulting ltd. http://www.lingnu.com/

2 1

Re: Re: WineHQ:service.cgi: fix winrash option
by chmorgan＠charter.net July 28, 2004

July 28, 2004

Which errors are you referring to? The most frequent ones lately are the ones where the winetest package isn't available on the sourceforge osdn mirror. We are still asking people to download the tests from this location for particular test releases. I'd imagine the fix is simply to point the downloads at valid mirror for the file. Chris > > From: Ferenc Wagner <wferi(a)afavant.elte.hu> > Date: 2004/07/26 Mon AM 07:57:51 EDT > To: wine-devel(a)winehq.org > CC: <wine-patches(a)winehq.com>, <cmorgan(a)alum.WPI.EDU> > Subject: Re: WineHQ:service.cgi: fix winrash option > > <chmorgan(a)charter.net> writes: > > > What is the service doing with winrash options btw? The > > installer stops, installs and starts the winrash service. > > > > The installer uses /S for silent installs, otherwise it > > will try to popup windows. Are you sure the service isn't > > calling the installer with /S? > > You are right, I was wrong. This patch is rubbish. I can't > really understand the errors on wine-test-results. Have you > got an idea what may have gone wrong? Aren't any functional > winrash instances running? Is something broken in the > service script? The documentation is somewhat scarce so I'm > doing a slow start... > -- > Feri. > >

2 1

Re: DCOM: Fix Ref Counting in StdMarshalImpl_MarshalInterface
by Robert Shearman July 28, 2004

July 28, 2004

Robert Shearman wrote: > Changelog: > > Mike Hearn <mh(a)codeweavers.com> > Fix ref counting in StdMarshalImpl_MarshalInterface for case where the > stub already exists. > >Index: wine/dlls/ole32/compobj_private.h >=================================================================== >RCS file: /home/wine/wine/dlls/ole32/compobj_private.h,v >retrieving revision 1.14 > Oops. Wrong patch. Rob

1 0

Re: COM RPC: Fix Infinite Loop
by Robert Shearman July 28, 2004

July 28, 2004

Robert Shearman wrote: > Changelog: > Fix for infinite loop that could occur if _invoke_onereq failed. Please don't apply this patch. I'll submit a better and more complete one in a minute. Rob

1 0

Non-working patch for 3GB of memory allocation
by Shachar Shemesh July 28, 2004

July 28, 2004

Ok, this patch would probably not be appropriate even if it did work. The surprising thing, however, is that it doesn't. The idea was simple. We use mmap as usual until mmap starts to fail with no memory. When that happens, we switch to allocating memory with malloc, which allegably knows how to use "brk" in order to allocate memory from the lower 1GB area. The patch is against 20040716. The attached program tries to repeatedly allocate ~100MB, until it fails. It is compiled for advanced server (i.e. - it has 3GB of address space, but that's besides the point). When actually run (through strace), one can see the mmaps being carried out and successful, and then the mmap that fails (no memory). Then malloc is attempted, which tries mmap, then brk, and then mmap again. All fail. This is the point where I stop understanding. Does anyone have any explanation as to why malloc fails? Does wine prevent large scale use of the lower area due to some reason? Any ideas would be appreciated. Shachar ? runconfig ? dlls/ntdll/.virtual.c.swp Index: dlls/ntdll/virtual.c =================================================================== RCS file: /home/wine/wine/dlls/ntdll/virtual.c,v retrieving revision 1.37 diff -u -r1.37 virtual.c --- dlls/ntdll/virtual.c 22 Jun 2004 03:02:35 -0000 1.37 +++ dlls/ntdll/virtual.c 26 Jul 2004 13:53:41 -0000 @@ -78,6 +78,7 @@ /* Per-view flags */ #define VFLAG_SYSTEM 0x01 /* system view (underlying mmap not under our control) */ #define VFLAG_VALLOC 0x02 /* allocated by VirtualAlloc */ +#define VFLAG_MALLOC 0x04 /* allocated using malloc */ /* Conversion from VPROT_* to Win32 flags */ static const BYTE VIRTUAL_Win32Flags[16] = @@ -168,9 +169,13 @@ DPRINTF( "View: %p - %p", addr, addr + view->size - 1 ); if (view->flags & VFLAG_SYSTEM) DPRINTF( " (system)\n" ); - else if (view->flags & VFLAG_VALLOC) - DPRINTF( " (valloc)\n" ); - else if (view->mapping) + else if (view->flags & VFLAG_VALLOC) { + DPRINTF( " (valloc" ); + if (view->flags & VFLAG_MALLOC) + DPRINTF( " using malloc)\n" ); + else + DPRINTF( ")\n" ); + } else if (view->mapping) DPRINTF( " %p\n", view->mapping ); else DPRINTF( " (anonymous)\n"); @@ -348,7 +353,12 @@ */ static void delete_view( struct file_view *view ) /* [in] View */ { - if (!(view->flags & VFLAG_SYSTEM)) unmap_area( view->base, view->size ); + if (!(view->flags & VFLAG_SYSTEM)) { + if( !(view->flags & VFLAG_MALLOC) ) + unmap_area( view->base, view->size ); + else + free( view->base ); + } list_remove( &view->entry ); if (view->mapping) NtClose( view->mapping ); free( view ); @@ -559,6 +569,7 @@ { void *ptr; NTSTATUS status; + int used_malloc=0; if (base) { @@ -602,7 +613,19 @@ { if ((ptr = wine_anon_mmap( NULL, view_size, VIRTUAL_GetUnixProt(vprot), 0 )) == (void *)-1) { - if (errno == ENOMEM) return STATUS_NO_MEMORY; + if (errno == ENOMEM) + { + /* mmap allocation failed. Let's see if we can try malloc, and whether it is going to be more lucky */ + FIXME("About to try malloc\n"); + if( /* Unix protection is not malloc compatible, or */ (ptr = malloc( size ))!=NULL ) + { + used_malloc=1; + goto memory_allocated; + } else { + FIXME("malloc failed\n"); + return STATUS_NO_MEMORY; + } + } return STATUS_INVALID_PARAMETER; } /* if we got something beyond the user limit, unmap it and retry */ @@ -623,8 +646,17 @@ munmap( (char *)ptr + size, view_size - size ); } +memory_allocated: status = create_view( view_ret, ptr, size, vprot ); - if (status != STATUS_SUCCESS) unmap_area( ptr, size ); + if( used_malloc && status==STATUS_SUCCESS ) + (*view_ret)->flags|=VFLAG_MALLOC; + else if (status != STATUS_SUCCESS) + { + if( !used_malloc ) + unmap_area( ptr, size ); + else + free( ptr ); + } return status; }

2 6

delayed unloading of .so in WINE
by Santosh Siddheshwar July 28, 2004

July 28, 2004

Hi, Is there any way to specify in config file or any other place, a way to prevent/delay a shared object from being unloaded till the application using it terminates? In dlopen there is a flag RTLD_NODELETE which keeps the loaded shared object in memory even after dlclose is called. The shared object is unloaded only when the app exits. Do we have some way to do this in WINE? Or does any feature in Windows necessitate implementing this kind of behaviour in WINE? I can think of VC++ linker flags /DELAYLOAD and /DELAY:unload which might need the use of such mechanism. Regards Santosh Siddheshwar ********************************************************************* Disclaimer: The information in this e-mail and any attachments is confidential / privileged. It is intended solely for the addressee or addressees. If you are not the addressee indicated in this message, you may not copy or deliver this message to anyone. In such case, you should destroy this message and kindly notify the sender by reply email. Please advise immediately if you or your employer does not consent to Internet email for messages of this kind. *********************************************************************

2 1

generation date/version of the (online) wine documentation
by Jeroen Janssen July 28, 2004

July 28, 2004

Hi, I have a few questions: * is it possible to add the date/version in the online wine documentation at http://www.winehq.com/site/documentation (for the user, devel docs) * when exactly are these pages generated? (nightly/daily/?) best regards, Jeroen

2 1

handling IUnknown/IRemUnknown for (outproc) objects
by Jeroen Janssen July 28, 2004

July 28, 2004

Hello, After last weeks discussion on the "unhandled interface" message in CFProxy_QueryInterface I was getting, we concluded (if I'm correct) that this is part of the "IRemUnknown" handling that is not (yet) implemented correctly in the wine CVS tree. I'm trying to figure out exactly where in the (OLE) code, this should be handled. If someone could point me in the right direction I would appreciate it very much. I think it would also be helpfull if there was some "fixme" logging so other people can also see if they hit upon the same problem (since it took a while to finally figure out the problem). --- Jeroen

3 3

Re: change the nt service pack version from 6 to 6a, correct win2k3 name
by Ivan Leo Puoti July 28, 2004

July 28, 2004

I've found this hex value in user32.dll 5e07c is this good or bad? Ivan.

1 0

Re: DCOM: Document wrong behaviour for IRunningObjectTable
by Robert Shearman July 28, 2004

July 28, 2004

Robert Shearman wrote: > Changelog: > Document wrong behaviour for IRunningObjectTable. Sorry, the changelog for this patch should have been as follows: Mike Hearn <mh(a)codeweavers.com> Robert Shearman <rob(a)codeweavers.com> Document wrong behaviour for IRunningObjectTable.

1 0

Re: change the nt service pack version from 6 to 6a, correct win2k3 name
by Andreas Mohr July 28, 2004

July 28, 2004

Hi, On Wed, Jul 28, 2004 at 11:39:30AM +0200, Ivan Leo Puoti wrote: > I'm not 100% sure this patch is correct, please let me know if it isn't. > BTW at line 342 of dlls/ntdll/version.c I read /* FIXME: Not sure, should be > verified with a Win2K3 dll */ > how can this be verified? Easy, take several W2K3 system DLLs and kick them through winedump (or pedump.exe) to verify whether their RtlImageNtHeader(hmod)->OptionalHeader.MajorOperatingSystemVersion value (i.e. their MajorOperatingSystemVersion value) is 7 like our Wine code assumes here. Otherwise you'll have to find out other characteristic DLL version number differences of these DLLs compared to WinXP to be able to properly return the WIN2K3 version for these AND ONLY these... Andreas Mohr

1 0

icons contribuition
by Marcio Esper July 28, 2004

July 28, 2004

Hello, My name is Marcio Esper. The Ivan tell me than I can send to you using this e-mail some icons. >We could do with some better icons, email wine-devel(a)winehq.com if you >want to contribute I made a new icon to be used with wine. 128x128, 64x64, 48x48, 32x32, 16x16. I hope you like it. The icons are made in Kpovmodeler on Mandrake linux 10 AMD64 and you can use and spread it without any restrition just geve me credits puting my name in the readmes and or were you think a good place to talk about it. ;p If we whant more models and sizes, let me know. Best regards, Marcio Esper

3 5

RelayExclude
by Diego 'Flameeyes' Pettenò July 28, 2004

July 28, 2004

Hi, I've a question about RelayExclude setting in [Debug] section of ~/.wine/config : I was trying to debug the load of ClrMamePro watching at the relay output, but there's a lot of call of ntdll.RtlEnterCriticalSection ntdll.RtlLeaveCriticalSection and so on. I already added them to RelayExclude last year when I debugged another app, but seems like now this setting is ignored. I also tried to add the complete ntdll.RtlEnterCriticalSection name, but nothing changed. Any suggestion? TIA, -- Diego "Flameeyes" Pettenò dgp85(a)users.sourceforge.net - http://flameeyes.web.ctonet.it/

4 3

Re: Make wine honor LC_CTYPE and LC_MESSAGES
by Dmitry Timoshkov July 28, 2004

July 28, 2004

"Shachar Shemesh" <shachar(a)lingnu.com> wrote: > http://support.microsoft.com/default.aspx?scid=kb;en-us;258047 says you > are wrong. My personal experience says you are wrong too, but that's > besides the point. You are misinterpreting that MSDN article, especially part about user locale. Since there is no official API to change user locale, what they call user locale settings are registry overrides for locale data. I already mentioned that on my english win2k with locale in Control Panel set to russian both GetSystemDefaultLCID and GetUserDefaultLCID return 0419 (russian). That dismisses MSDN and your statements. Thread locale is completely unrelated thing. -- Dmitry.

2 1

AmigaOS4
by Manuel Fink July 27, 2004

July 27, 2004

Hello I has here a small question. That appeared Amiga OS4 Prerelease for some time and it seems a good and stable system to be. will it give in the near future WINE also for the AmigaOS4? Greeting Manuel

1 0

MUI setup - preliminary conclusions
by Shachar Shemesh July 27, 2004

July 27, 2004

Ok, here is the preliminary conclusions from Windows XP MUI: There are two setups in the system. One is the "language for the default user and new user accounts". It affects the language people get when they create a new user, and the language the login dialog is in. To change that one, you need to run the MUI setup program. A restart is needed to change this. The second is an extra dialog in the Regional Settings dialog under the "languages" tab, and it selects the language the current user will have. You need to log off and log on in order to change this one. At the moment, my system is set up with English, Hebrew and Italian installed. I'm trying to figure out what exactly is passed to FindResourceEx when FindResource is called. So far, it seems that "LANG_NEUTRAL, SUBLANG_NEUTRAL" is used, but it's not definite yet. One thing of interest, however, is that when LANG_NEUTRAL, SUBLANG_DEFAULT is passed, the lookup is for the language used in the second case (the user UI language), and when LANG_NEUTRAL, SUBLANG_SYS_DEFAULT, the first one is used. It seems to me that there is no need for wine to support two separate languages here. Does anyone disagree with this? Given the above settings, my current test program currently tests the order between English(US), English(UK), English(Neutral), Italian(Italy), Italian(Neutral), Hebrew(Default), Hebrew(Neutral), Neutral and Neutral(Default). Are there any other languages I should test for order by? Shachar -- Shachar Shemesh Lingnu Open Source Consulting ltd. http://www.lingnu.com/

1 0

Re: DCOM: Pass -Embedding switch to EXE servers, more tracing
by Jeroen Janssen July 27, 2004

July 27, 2004

James Hawkins wrote: >WCHAR is a wide (unicode) character. sizeof(embedding)/sizeof(WCHAR) >will return the length of the strength in characters and not the size >of it if I'm not mistaken. > > Yes, and I think you will want to have the length in characters when declaring a (C) array in the example code below: With static const WCHAR embedding[] = { ' ', '-','E','m','b','e','d','d','i','n','g',0 }; You will want WCHAR command[MAX_PATH+(sizeof(embedding)/sizeof(WCHAR)+1]; instead of WCHAR command[MAX_PATH+sizeof(embedding)+1]; Because if sizeof(WCHAR) == 2, then command will be strlen(embedding) to large. >On Mon, 26 Jul 2004 18:24:48 +0200, Jeroen Janssen <japj(a)xs4all.nl> wrote: > > >>Mike Hearn wrote: >> >> >>>ChangeLog: >>>Pass -Embedding switch to EXE servers, more tracing >>> >>> >>>diff -u -p -r1.15 rpc.c >>>--- dlls/ole32/rpc.c 15 Jul 2004 22:07:44 -0000 1.15 >>>+++ dlls/ole32/rpc.c 21 Jul 2004 22:43:59 -0000 >>>@@ -471,8 +471,10 @@ create_server(REFCLSID rclsid) { >>> char buf[200]; >>> HRESULT hres = E_UNEXPECTED; >>> char xclsid[80]; >>>- WCHAR dllName[MAX_PATH+1]; >>>- DWORD dllNameLen = sizeof(dllName); >>>+ WCHAR exe[MAX_PATH+1]; >>>+ DWORD exelen = sizeof(exe); >>>+ static const WCHAR embedding[] = { ' ', '-','E','m','b','e','d','d','i','n','g',0 }; >>>+ WCHAR command[MAX_PATH+sizeof(embedding)+1]; >>> >>> >>On a side note, what exactly is a WCHAR? >>Is it a 'char' or a 'unicode char' (2bytes)? >> >>This in relation to sizeof(embedding) instead of >>sizeof(embedding)/sizeof(WCHAR)? >>--- >>Jeroen >> >> >> >> > > > >

2 2

Re: Make wine honor LC_CTYPE and LC_MESSAGES
by Shachar Shemesh July 27, 2004

July 27, 2004

Following a discussion with Alexandre on IRC, here is a revised patch. This one doesn't touch the LC_MESSAGE, as these do not affect the user locale. Further work will be necessary on supporting MUI like language checking. Shachar Shemesh wrote: > Changelog: > Shachar Shemesh <winecode(a)shemesh.biz> > dlls/kernel/locale.c > > * System default locale in Windows determines the ANSI encoding > (LC_CTYPE on Unix) -- Shachar Shemesh Lingnu Open Source Consulting ltd. http://www.lingnu.com/

5 18