February 2011 - Wine-Devel - list.winehq.org

Re: server: Fix compilation on systems with a two-argument sched_setaffinity prototype.

by Alexandre Julliard

Andrew Nguyen <anguyen(a)codeweavers.com> writes: > My motivation was to determine the reason that the reporter of bug 26088 > (http://bugs.winehq.org/show_bug.cgi?id=26088) felt the need to compile > Wine configured with --without-pthread, which doesn't seem to be > something anyone should be doing. CentOS 3 seemed to be a natural > candidate for testing, although later developments revealed that the > reporter's system of concern was actually Debian Sarge. > > Given that there are apparently a non-zero number of users wanting to > compile and run at least stable Wine on antique platforms, a few patches > to accommodate such users doesn't seem to be too much trouble. The > sched_setaffinity compile error was the only real blocker for > compilation on CentOS 3, and after fixing it, I was able to at least > successfully launch winecfg. It's not worth making the code more complicated for this. You should disable it entirely if configure detects a broken prototype. -- Alexandre Julliard julliard(a)winehq.org

Feb. 21, 2011

Re: msxml3: Assign the IBindStatusCallback as soon as it created.

by Jacek Caban

On 2/18/11 10:24 AM, Alistair Leslie-Hughes wrote: > Hi, > This stops a crash in httprequest_get_responseText, when running from > a webpage since "This->bsc" was NULL. > Tests, please. Jacek

Feb. 21, 2011

Re: server: Fix compilation on systems with a two-argument sched_setaffinity prototype.

by Andrew Nguyen

On 02/21/2011 04:43 AM, Michael Stefaniuc wrote: > Hello Andrew! > > Andrew Nguyen wrote: >> This lets Wine compile successfully on CentOS 3. > CentOS 3? Alexandre was bitching about RHEL 5 being "prediluvian" ... > RHEL 3 has a 2.4.21 based kernel, is over 7 years old and EOL. I'm not > sure it is worth spending the effort to make current Wine compile and > run on such an old system. > > bye > michael > The fact that CentOS 3 has a Linux 2.4 kernel is precisely why I tried compiling Wine on the system. My motivation was to determine the reason that the reporter of bug 26088 (http://bugs.winehq.org/show_bug.cgi?id=26088) felt the need to compile Wine configured with --without-pthread, which doesn't seem to be something anyone should be doing. CentOS 3 seemed to be a natural candidate for testing, although later developments revealed that the reporter's system of concern was actually Debian Sarge. Given that there are apparently a non-zero number of users wanting to compile and run at least stable Wine on antique platforms, a few patches to accommodate such users doesn't seem to be too much trouble. The sched_setaffinity compile error was the only real blocker for compilation on CentOS 3, and after fixing it, I was able to at least successfully launch winecfg.

Feb. 21, 2011

Re: mshtml: Implement IHTMLAnchorElement get/put target

by Jacek Caban

On 2/18/11 9:54 AM, Alistair Leslie-Hughes wrote: > static HRESULT WINAPI HTMLAnchorElement_get_target(IHTMLAnchorElement *iface, BSTR *p) > { > HTMLAnchorElement *This = impl_from_IHTMLAnchorElement(iface); > - FIXME("(%p)->(%p)\n", This, p); > - return E_NOTIMPL; > + nsAString target_str; > + nsresult nsres; > + HRESULT hres = S_OK; > + > + TRACE("(%p)->(%p)\n", This, p); > + > + nsAString_Init(&target_str, NULL); > + nsres = nsIDOMHTMLAnchorElement_GetTarget(This->nsanchor,&target_str); > + if(NS_SUCCEEDED(nsres)) { > + const PRUnichar *target; > + > + nsAString_GetData(&target_str,&target); > + *p = *target ? SysAllocString(target) : NULL; > + }else { > + ERR("GetHref failed: %08x\n", nsres); > + hres = E_FAIL; > + } > + > + nsAString_Finish(&target_str); > + return hres; > } Please use use return_nsstr here. Jacek

Feb. 21, 2011

Re: msxml3: Add IObjectWithSite support to IXMLHttpRequest

by Jacek Caban

On 2/17/11 10:47 AM, Alistair Leslie-Hughes wrote: > Hi, > > > Changelog: > msxml3: Add IObjectWithSite support to IXMLHttpRequest > More tests, please. Jacek

Feb. 21, 2011

Valgrind --malloc-fill=A5 --free-fill=DE crashes Wine

by Joerg-Cyril.Hoehle＠t-systems.com

Hi, I used to run Valgrind with the additional options --malloc-fill=A5 --free-fill=DE around Wine 1.2 times. Now I retried it and it crashes Wine 1.3.13 and 1.3.14 (it fails to load kernel32.dll). It turns out that malloc-fill is harmless, -free-fill causes havoc currently. Am I the only one using or seeing this (Ubuntu Lucid)? Thanks for your help, Jörg Höhle

Feb. 21, 2011

Re: server: Fix compilation on systems with a two-argument sched_setaffinity prototype.

by Michael Stefaniuc

Hello Andrew! Andrew Nguyen wrote: > This lets Wine compile successfully on CentOS 3. CentOS 3? Alexandre was bitching about RHEL 5 being "prediluvian" ... RHEL 3 has a 2.4.21 based kernel, is over 7 years old and EOL. I'm not sure it is worth spending the effort to make current Wine compile and run on such an old system. bye michael

Feb. 21, 2011

Re: GSOC idea to implement the explorer

by Andrew Green

I had assumed that the browser part of explorer simply hosted another control or collection of controls. Though explorer is more than a browser. It also acts like a desktop shell. Though I understand this is probably the lower priority on the wine developers list. As integrating into Linux desktop is probably the preferred and most used mode. I thought that sharing reactos explorer and working on making a complete explorer(including extensions) would be the goal. Though now I'm saying it I feel rather silly, as this really is the least important part of wine.

Feb. 20, 2011

Re: d3dx9: Implement D3DXFloat32To16Array and D3DXFloat16To32Array. (try 2)

by Rico Schüller

Hi, I have a small hint to the D3DXFloat32To16Array function. It seems that some values are handled in some other way. A small test: D3DXFLOAT16 res; union p { float f; DWORD d; } x; x.d = 0x33000800; D3DXFloat32To16Array(&res, &x.f, 1); ok(res.value == 1, "Failed %f %x\n", x.f, res.value); This fails with the current implementation. It's not the only value, there are a couple of them which are of by one. This could be easily seen by loading both d3dx36.dll (wine and native) and comparing the values directly. Cheers Rico Am 18.02.2011 02:18, schrieb Misha Koshelev: > Fixes bug 25769. Extensively vetted by Stefan Dosinger. Passes winetestbot: > https://testbot.winehq.org/JobDetails.pl?Key=9366 > > Thank you > Misha > --- > dlls/d3dx9_36/d3dx9_36.spec | 4 +- > dlls/d3dx9_36/math.c | 105 +++++++++++++++++++++++++++++++++++++++++++ > dlls/d3dx9_36/tests/math.c | 69 ++++++++++++++++++++++++++++ > include/d3dx9math.h | 18 +++++++ > include/d3dx9math.inl | 31 +++++++++++++ > 5 files changed, 225 insertions(+), 2 deletions(-) > > >

Feb. 20, 2011

Re: GSOC idea to implement the explorer

by Nikolay Sivov

On 2/20/2011 18:58, Roderick Colenbrander wrote: > On Sun, Feb 20, 2011 at 4:05 AM, Andrew Green<greeniekin(a)gmail.com> wrote: >> I was just looking at the Google summer of code(GSOC) ideas and >> noticed "implement the explorer". This grabbed my attention as I have >> recently been working on a windows xp explorer clone. >> Also it mentions a reference to ReactOS explorer written in c++. Well >> there is a second explorer which is planned to be more true to the xp >> implementation and written in c. Currently it's called explorer new as >> it is not yet complete and is just a shell. >> I was wondering if it was possible to share this code. As it would >> probably be beneficial in the long term. >> It already has support for shell extensions. When running under >> windows I can add any extension such as Quick launch, language bar and >> so on. >> Here is a pic of my working copy(trunk doesn't have this theming yet. >> Though I should finish it soon) >> http://img156.imageshack.us/f/yaytheme.png/ >> >> This is just an idea. I know it's very incomplete. Though it is showing promise. >> >> Here is a link to the "GSOC implement the explorer" page I mentioned >> http://wiki.winehq.org/SummerOfCode#head-18241c95194e4e1c597bd2923cf296084f… >> >> Any comments and ideas are much appreciated. >> > As of Vista, Explorer itself seems to be a small application which > hosts the IExplorerBrowser control. Last year it was decided to > implement this control properly (someone worked on it for GSoc, but > not sure about its status and whether the code is in). This control > can be used in the file open/save dialogs and in explorer.exe itself. > I would research what happened to this code (and where it is, perhaps > it has to be merged in). David Hedberg was working on this and it's merged already in shell32, it implements some new interfaces and kind of treeview support for shellview (that was originally list based only). > Roderick > > >

Feb. 20, 2011

Re: GSOC idea to implement the explorer

by Roderick Colenbrander

On Sun, Feb 20, 2011 at 4:05 AM, Andrew Green <greeniekin(a)gmail.com> wrote: > I was just looking at the Google summer of code(GSOC) ideas and > noticed "implement the explorer". This grabbed my attention as I have > recently been working on a windows xp explorer clone. > Also it mentions a reference to ReactOS explorer written in c++. Well > there is a second explorer which is planned to be more true to the xp > implementation and written in c. Currently it's called explorer new as > it is not yet complete and is just a shell. > I was wondering if it was possible to share this code. As it would > probably be beneficial in the long term. > It already has support for shell extensions. When running under > windows I can add any extension such as Quick launch, language bar and > so on. > Here is a pic of my working copy(trunk doesn't have this theming yet. > Though I should finish it soon) > http://img156.imageshack.us/f/yaytheme.png/ > > This is just an idea. I know it's very incomplete. Though it is showing promise. > > Here is a link to the "GSOC implement the explorer" page I mentioned > http://wiki.winehq.org/SummerOfCode#head-18241c95194e4e1c597bd2923cf296084f… > > Any comments and ideas are much appreciated. > As of Vista, Explorer itself seems to be a small application which hosts the IExplorerBrowser control. Last year it was decided to implement this control properly (someone worked on it for GSoc, but not sure about its status and whether the code is in). This control can be used in the file open/save dialogs and in explorer.exe itself. I would research what happened to this code (and where it is, perhaps it has to be merged in). Roderick

Feb. 20, 2011

GSOC idea to implement the explorer

by Andrew Green

I was just looking at the Google summer of code(GSOC) ideas and noticed "implement the explorer". This grabbed my attention as I have recently been working on a windows xp explorer clone. Also it mentions a reference to ReactOS explorer written in c++. Well there is a second explorer which is planned to be more true to the xp implementation and written in c. Currently it's called explorer new as it is not yet complete and is just a shell. I was wondering if it was possible to share this code. As it would probably be beneficial in the long term. It already has support for shell extensions. When running under windows I can add any extension such as Quick launch, language bar and so on. Here is a pic of my working copy(trunk doesn't have this theming yet. Though I should finish it soon) http://img156.imageshack.us/f/yaytheme.png/ This is just an idea. I know it's very incomplete. Though it is showing promise. Here is a link to the "GSOC implement the explorer" page I mentioned http://wiki.winehq.org/SummerOfCode#head-18241c95194e4e1c597bd2923cf296084f… Any comments and ideas are much appreciated.

Feb. 20, 2011

bluetooth cpl support (Re: USB Device Support - Nokia PC Suite)

by Hin-Tak Leung

--- On Fri, 18/2/11, Mike Yates <eniw(a)fonehelp.co.uk> wrote: > After rebooting, I did get a full > list of "fixme" lines:- > > mike(a)myvmubuntu:~$ wine .wine/drive_c/Program\ > Files/Nokia/Nokia\ PC\ > Suite\ 7/PCSuite.exe <snipped> > err:module:import_dll Library irprops.cpl (which is needed > by > L"C:\\Program Files\\PC Connectivity > Solution\\Transports\\NclMSBTSrv.exe") not found > err:module:LdrInitializeThunk Main exe initialization for > L"C:\\Program Files\\PC Connectivity > Solution\\Transports\\NclMSBTSrv.exe" failed, status > c0000135 <snipped> > (0x14c0d00)->(0x33bcc8) > I/O warning : failed to load external entity > "ConfServer.dtd" > err:msxml:domdoc_validateNode Could not load the external > subset > "ConfServer.dtd" There are really only two important error messages here - I am not sure about the latter, but the first one is simply that part of the Nokia PC suite tries to load "irprops.cpl", which AFAIK is the bluetooth control panel applet. It is not surprising that it fails - that hasn't been written in wine yet. I think both sides are 'barking up the wrong tree' - the initial poster thought PC suite failure is due to lack of sufficient USB support in wine; and some of the wine devs can possibly take a quick look before shouting the poster down. The lack of bluetooth implementation is a valid development discussion? It appears that since Nokia PC suite supports bluetooth connectivity to mobile phones (as well as serial cable, infrared, AFAIK, I own nokia phones for about 10 years), it tries to load some windows OS bluetooth bits and fails. If the poster doesn't actually need bluetooth connectivity, adding a stub minimal implementation of irprops.cpl which does nothing can probably get the software to run? OTOH, I already mentioned both on and off list a few times: xnokii/gnokii does what I need with nokia phones and I have never been drawn to look at running the windows nokia software in wine.

Feb. 19, 2011

Re: new winetricks-alpha 20110219-alpha: added wineprefix selection screen

by Dan Kegel

Roseanne wrote: >> winetricks-alpha version 20110219's gui now starts with a wineprefix >> selection dialog. ... >> When using the gui, games can only be installed into their own >> sandboxed prefix. > > You should make a separate wineprefix the default for Office > 2003 and 2007 as well, though it shouldn't be sandboxed. Right > now you are setting the override for riched20 only for the > main apps, but many of Office's advanced features (e.g., > clip organizer, equation editor) are separate exes and the > override needs to be set for all of them. Installing to a > separate wineprefix and setting the override globally is much > easier. It will also protect Office apps from being messed > up by other overrides a user might set for other apps. You > might also want to disable Wine's crash dialog for Office, > as it prevents Office's error handler from recovering the file > that was being worked on at the time of the crash. Good ideas all. I isolated just games for starters simply because it was an easy place to start, but will get to apps soon. (Incidentally, I did a silent release, 20110219b-alpha, to fix a few install problems that show up on an empty cache.) - Dan

Feb. 19, 2011

No longer possible for users to submit AppDB test reports for the current stable release

by Rosanne DiMesio

With the release of 1.3.14, the current stable release has dropped off the list of versions for which users can submit AppDB test reports, though 1.0.0 and 1.0.1 remain on the list. Can someone please fix this? http://bugs.winehq.org/show_bug.cgi?id=24763 -- Rosanne DiMesio <dimesio(a)earthlink.net>

Feb. 19, 2011

Re: USB Device Support - Nokia PC Suite

by GOUJON Alexandre

On 02/18/2011 05:08 PM, Mike Yates wrote: > After rebooting, I did get a full list of "fixme" lines:- > > mike(a)myvmubuntu:~$ wine .wine/drive_c/Program\ Files/Nokia/Nokia\ PC\ > Suite\ 7/PCSuite.exe [snip] > Mike, this is a developer's mailing list. With questions about how to > use Wine please refer to Wine user forum: http://forum.winehq.org/ > > Vitaliy Please stop flooding the dev mailing list. It won't make your app working. And developers won't help you here. Please file a bug in Bugzilla as other users. Attach there your compressed log, explain how to reproduce the crash, indicate a URL where this piece of software can be downloaded ... and *wait*. Some 'NEW' bugs are solved months, days or years later, it depends. The USB support in wine is still 'experimental' and may work or not. If you want to help, feel free to send a patch. Or ask a question here if it's related to wine development. But please, stop flooding here. Thanks

Feb. 19, 2011

Re: new winetricks-alpha 20110219-alpha: added wineprefix selection screen

by Rosanne DiMesio

On Fri, 18 Feb 2011 22:48:22 -0800 Dan Kegel <dank(a)kegel.com> wrote: > Another day, another winetricks-alpha. (Well, it *is* alpha.) > > winetricks-alpha version 20110219's gui now starts with a wineprefix > selection dialog. > If you just click OK, it uses the default wineprefix. > When using the gui, games can only be installed into their own > sandboxed prefix. You should make a separate wineprefix the default for Office 2003 and 2007 as well, though it shouldn't be sandboxed. Right now you are setting the override for riched20 only for the main apps, but many of Office's advanced features (e.g., clip organizer, equation editor) are separate exes and the override needs to be set for all of them. Installing to a separate wineprefix and setting the override globally is much easier. It will also protect Office apps from being messed up by other overrides a user might set for other apps. You might also want to disable Wine's crash dialog for Office, as it prevents Office's error handler from recovering the file that was being worked on at the time of the crash. -- Rosanne DiMesio <dimesio(a)earthlink.net>

Feb. 19, 2011

Re: System-wide WoW64 setup

by Jerome Leclanche

Yep, that was the issue. --libdir=/usr/lib32 is needed for the 32bit install on ubuntu. Thanks a lot. This may be worth mentioning on the wiki. J. Leclanche On Sat, Feb 19, 2011 at 1:50 PM, Alexandre Julliard <julliard(a)winehq.org> wrote: > Jerome Leclanche <adys.wh(a)gmail.com> writes: > >> The guide doesn't mention a system-wide setup. Am I doing something wrong here? > > Most likely you are not using the correct libdirs. If your distro > doesn't use lib and lib64 for 32/64-bit respectively, you need to > specify the libdirs in configure. > > -- > Alexandre Julliard > julliard(a)winehq.org >

Feb. 19, 2011

Re: comctl32: Fix read of uninitialized data in LISTVIEW_GetItemExtT when LVIF_TEXT is not set (Valgrind).

by Nikolay Sivov

On 2/19/2011 16:49, Alexander Scott-Johns wrote: > On 19 February 2011 11:42, Nikolay Sivov<bunglehead(a)gmail.com> wrote: >> I think it's enough to add a mask check here: >> --- >> if (bResult&& lpLVItem->pszText != pszText) >> --- >> >> to keep a single call of GetItem. > But isn't > > pszText = lpLVItem->pszText; > > undefined behaviour if lpLVItem->pszText is not initialized? No, it's not undefined as long as you don't dereference it. Valgrind catches condition that depends on this uninitialized value, not assignment of course. By the way, we even have a report for that - http://bugs.winehq.org/show_bug.cgi?id=24290

Feb. 19, 2011

Re: System-wide WoW64 setup

by Alexandre Julliard

Jerome Leclanche <adys.wh(a)gmail.com> writes: > The guide doesn't mention a system-wide setup. Am I doing something wrong here? Most likely you are not using the correct libdirs. If your distro doesn't use lib and lib64 for 32/64-bit respectively, you need to specify the libdirs in configure. -- Alexandre Julliard julliard(a)winehq.org

Feb. 19, 2011

Re: comctl32: Fix read of uninitialized data in LISTVIEW_GetItemExtT when LVIF_TEXT is not set (Valgrind).

by Alexander Scott-Johns

On 19 February 2011 11:42, Nikolay Sivov <bunglehead(a)gmail.com> wrote: > I think it's enough to add a mask check here: > --- > if (bResult && lpLVItem->pszText != pszText) > --- > > to keep a single call of GetItem. But isn't pszText = lpLVItem->pszText; undefined behaviour if lpLVItem->pszText is not initialized?

Feb. 19, 2011

System-wide WoW64 setup

by Jerome Leclanche

I've recently finished setting up WoW64 with the help of this guide [http://wiki.winehq.org/Wine64] and kevlarman on #winehackers. However, I'm running into a lot of trouble to install it system-wide. Git tree lives at ~/src/wine and my build dirs are ~/src/wine/build and ~/src/wine/build64. A first make install from build64 leads to this (as expected): adys(a)azura:~$ wineboot exec: 52: wine: not found Following it by a make install from build32 I get: adys(a)azura:~$ wineboot /usr/bin/wineserver: error while loading shared libraries: libwine.so.1: wrong ELF class: ELFCLASS32 And if I reinstall from build64 above it, I get: adys(a)azura:~$ wineboot /usr/bin/wine: error while loading shared libraries: libwine.so.1: wrong ELF class: ELFCLASS64 No improvement with a full clean and a reinstall starting by build32: adys(a)azura:~$ wineboot bash: /usr/bin/wineboot: No such file or directory ... adys(a)azura:~$ wineboot /usr/bin/wine: error while loading shared libraries: libwine.so.1: wrong ELF class: ELFCLASS64 The guide doesn't mention a system-wide setup. Am I doing something wrong here? -- J. Leclanche

Feb. 19, 2011

Re: comctl32: Fix read of uninitialized data in LISTVIEW_GetItemExtT when LVIF_TEXT is not set (Valgrind).

by Nikolay Sivov

On 2/19/2011 03:34, Alexander Scott-Johns wrote: I think it's enough to add a mask check here: --- if (bResult && lpLVItem->pszText != pszText) --- to keep a single call of GetItem.

Feb. 19, 2011

new winetricks-alpha 20110219-alpha: added wineprefix selection screen

by Dan Kegel

Another day, another winetricks-alpha. (Well, it *is* alpha.) winetricks-alpha version 20110219's gui now starts with a wineprefix selection dialog. If you just click OK, it uses the default wineprefix. When using the gui, games can only be installed into their own sandboxed prefix. This keeps them from bumping into each other and making them spill their coffee, or worse. (This feature may seem odd, but it's how PlayOnLinux works, and it was specifically requested by Vitaliy Margolen in http://www.winehq.org/pipermail/wine-devel/2011-January/088332.html ) It now has an option to delete the selected wineprefix, and if you do, it also deletes any xdg desktop menuitems that refer to it. Comments welcome. There are some quirks left to work out with the new gui behavior, so caveat downloador. Download url: http://winetricks.org/winetricks-alpha svn repository / bug tracker: http://winetricks.org Changelog: Dan Kegel: In pickprefix menu, sense dosdevices rather than user.reg; makes new wineprefixes show up immediately Deleting a wineprefix also deletes its menu items. Make default wineprefix the, um, default in pickprefix. Games are reachable only via pickprefix menu now. Experimental gui change: outer dialog is now prefix selection. Needs polish and testing still. algodoo_demo: wait for installer to finish in non-silent case Remove dead code for run and uninstall modes/menus.

Feb. 19, 2011

Re: USB Device Support - Nokia PC Suite

by Vitaliy Margolen

Mike, this is a developer's mailing list. With questions about how to use Wine please refer to Wine user forum: http://forum.winehq.org/ Vitaliy

Feb. 18, 2011