July 2005 - Wine-Devel - list.winehq.org

Re: DINPUT: Detect force feedback joysticks (FF #1)

by Anssi Hannula

Daniel Remenak wrote: > The newer interface has working > iForce (logitech, guillemot, et al) and MS sidewinder drivers written > for it, so that's the one I was targetting. Well, if you mean the Sidewinder Force Feedback 2, that driver seems to be only a stub. I am currently working on a new PID FF driver (which will cover that Sidewinder). Also I'm planning some kernel internal FF interface updates and probably introducing FF interface to /dev/input/jsX also. -- Anssi Hannula

July 29, 2005

Re: Another AppDB screenshot problem...

by Jeremy Newman

Different issue. The issue I fixed was the unable to upload images larger than 512K. This other issue is because move_uploaded_file() is returning false. I did some debug and saw that $_FILES['imageFile']['tmp_name'] was null. Looking at the error code it says: The uploaded file exceeds the upload_max_filesize directive in php.ini. I removed the 'maxupload_max_filesize 5M' from .htaccess and that seemed to fix it. Very odd that that was causing problems. On Thu, 2005-07-28 at 14:51 -0700, James Liggett wrote: > Hi Jeremy, > Still doesn't work. Same error. :( > > James > > On Thu, 2005-07-28 at 13:28 -0500, Jeremy Newman wrote: > > The error we are getting is: > > > > Requested content-length of 1419105 is larger than the configured limit > > of 524288, referer: > > http://appdb.winehq.org/screenshots.php?appId=1383&versionId=1873 > > > > Apparently, Apache 2 in Red Hat has it's PHP4 configured with a limit of > > 512K on uploads. I had to change the "LimitRequestBody" setting in: > > /etc/httpd/conf.d/php.conf > > > > It should be fixed now. > > > > On Wed, 2005-07-27 at 21:35 -0600, Tony Lambregts wrote: > > > James Liggett wrote: > > > > Hi, > > > > Tonight I tried to submit a shot for Palm Desktop 4.1 and it gave me > > > > this error: > > > > > > > > Unable to move screenshot from to data/screenshots/originals/781 > > > > > > > > What's going on? > > > > > > > I can confirm this (I cannot upload a screenshot either) > > > Chris and I are not sure what is going on here. I took a fresh install of the > > > appdb and it works ok here. I think there might be an issue with the tmp > > > directory but Chris says there is space there. > > > > > > > > > Jeremy could you have a look? > > > > > > -- > > > > > > Tony Lambregts > > > > > >

July 28, 2005

Re: KERNEL32: Re: Fix thread tests on WinME

by Felix Nawothnig

Saulius Krasuckas wrote: > Felix, I have reverted part of your patch and tested this. Reversion is > doing fine on WinME -- zero failed tests. Didn't tested on XP or so. > > Is this acceptable? If so, feel free to submit it under own name. Or > correct the code. :-) This is not really correct, SetThreadPriority, GetThreadPriority and GetThreadTimes all return ERROR_CALL_NOT_IMPLEMENTED on WinME. The order of the tests is wrong (first the restrictions are tested, then it's tested if they are actually implemented) - the tests for the former two functions need to be put into the scope below and the latter has to be put below the scope below. IIRC (I looked at it a few days ago). I'll do that tomorrow. Felix

July 28, 2005

Re: KERNEL32: Re: Fix thread tests on WinME

by Saulius Krasuckas

* On Fri, 22 Jul 2005, Saulius Krasuckas wrote: > * Felix Nawothnig wrote: > > > > Tested on Wine and WinXP. I couldn't verify it on WinME since I don't > > have it but since this is quite trivial... *crosses fingers* > > Though three tests are still failing: > > | thread.c:293:not restricted, assuming consistent behaviour > | thread.c:412: Test failed: access restrictions obeyed > | thread.c:414: Test failed: access restrictions obeyed > | thread.c:488: Test failed: access restrictions obeyed Felix, I have reverted part of your patch and tested this. Reversion is doing fine on WinME -- zero failed tests. Didn't tested on XP or so. Is this acceptable? If so, feel free to submit it under own name. Or correct the code. :-) Index: dlls/kernel/tests/thread.c =================================================================== RCS file: /home/wine/wine/dlls/kernel/tests/thread.c,v retrieving revision 1.22 diff -p -u -r1.22 thread.c --- dlls/kernel/tests/thread.c 25 Jul 2005 11:07:54 -0000 1.22 +++ dlls/kernel/tests/thread.c 28 Jul 2005 21:55:41 -0000 @@ -409,9 +409,11 @@ static VOID test_thread_priority(void) obey_ar(SetThreadPriority(access_thread,1)==0); obey_ar(GetThreadPriority(access_thread)==THREAD_PRIORITY_ERROR_RETURN); if (pSetThreadPriorityBoost) - obey_ar(pSetThreadPriorityBoost(access_thread,1)==0); + ok(pSetThreadPriorityBoost(access_thread,1)==0, + "SetThreadPriorityBoost did not obey access restrictions\n"); if (pGetThreadPriorityBoost) - obey_ar(pGetThreadPriorityBoost(access_thread,&disabled)==0); + ok(pGetThreadPriorityBoost(access_thread,&disabled)==0, + "GetThreadPriorityBoost did not obey access restrictions\n"); obey_ar(GetExitCodeThread(access_thread,&exitCode)==0); ok(CloseHandle(access_thread),"Error Closing thread handle\n"); } @@ -485,7 +487,7 @@ static VOID test_GetThreadTimes(void) if(access_thread!=NULL) { error=GetThreadTimes(access_thread,&creationTime,&exitTime, &kernelTime,&userTime); - obey_ar(error==0); + ok(error==0, "GetThreadTimes did not obey access restrictions\n"); ok(CloseHandle(access_thread)!=0,"CloseHandle Failed\n"); } creationTime.dwLowDateTime=99; creationTime.dwHighDateTime=99;

July 28, 2005

Re: Const Function Parameters?

by Joseph Garvin

Has anyone checked if the performance improvement stays if only the pointers are set to const? I'd also suggest running the test multiple times and averaging. Felix Nawothnig wrote: > Andreas Mohr wrote: > >> While I'm not too convinced in this case (1.5% improvement sounds like >> within statistic noise), it should be a good idea to mark things in Wine >> const whenever possible (objdump -x helps here), since it improves >> reliability > > > Huh? He did stuff like... > > -void foo(int i) > +void foo(const int i) > > This will most likely not improve reliability. He also changed some > pointers to const which is wanted and will ofcourse improve reliability. > > Assuming that gcc has a pretty good optimizer the only reason I can > think of (besides a GCC bug :-) why performance increased is that gcc > doesn't have to perform aliasing-analysis and can be sure that the dst > pointer will never point into local parameters (constifying local > variables could speed up things slightly more). > > I'm not sure about the "restrict" semantics but maybe it could be used > instead of constifying "int i" to archieve the same effects? > > Felix >

July 28, 2005

Re: wine/dlls/ntdll loadorder.c

by Alexandre Julliard

Peter Oberndorfer <kumbayo84(a)arcor.de> writes: > It's a bit of both :-) > The dbghelp shipped with WinDbg has a SymSetParentWindow entry point which is > not (yet) available in wine. > So it catches fire when the function is called. > Adding a SymSetParentWindow stub is easy but shouldn't it load the dbghelp.dll > in the app dir? Well no, that's what has changed, we are now respecting the config settings even for files in the app dir, so you have to explicitly configure it as native if that's what you want (but of course fixing the builtin is preferable). The idea is that the config should work the same way no matter where we load the dll from. -- Alexandre Julliard julliard(a)winehq.org

July 28, 2005

Re: Another AppDB screenshot problem...

by James Liggett

Hi Jeremy, Still doesn't work. Same error. :( James On Thu, 2005-07-28 at 13:28 -0500, Jeremy Newman wrote: > The error we are getting is: > > Requested content-length of 1419105 is larger than the configured limit > of 524288, referer: > http://appdb.winehq.org/screenshots.php?appId=1383&versionId=1873 > > Apparently, Apache 2 in Red Hat has it's PHP4 configured with a limit of > 512K on uploads. I had to change the "LimitRequestBody" setting in: > /etc/httpd/conf.d/php.conf > > It should be fixed now. > > On Wed, 2005-07-27 at 21:35 -0600, Tony Lambregts wrote: > > James Liggett wrote: > > > Hi, > > > Tonight I tried to submit a shot for Palm Desktop 4.1 and it gave me > > > this error: > > > > > > Unable to move screenshot from to data/screenshots/originals/781 > > > > > > What's going on? > > > > > I can confirm this (I cannot upload a screenshot either) > > Chris and I are not sure what is going on here. I took a fresh install of the > > appdb and it works ok here. I think there might be an issue with the tmp > > directory but Chris says there is space there. > > > > > > Jeremy could you have a look? > > > > -- > > > > Tony Lambregts > > >

July 28, 2005

Re: wine/dlls/ntdll loadorder.c

by Peter Oberndorfer

On Thursday 28 July 2005 23:14, Alexandre Julliard wrote: > Peter Oberndorfer <kumbayo84(a)arcor.de> writes: > > Same thing happens here with WinDbg 6.5.3.7 which comes with dbghelp.dll > > in app directory. > > If you mean it fails to load dbghelp completely then that's a bug. If > it's loading the builtin dbghelp instead of the native then it's a > feature <g> It's a bit of both :-) The dbghelp shipped with WinDbg has a SymSetParentWindow entry point which is not (yet) available in wine. So it catches fire when the function is called. Adding a SymSetParentWindow stub is easy but shouldn't it load the dbghelp.dll in the app dir? Or does dbghelp get some special treatment because it has ability to read other debug formats?) OllyDbg is another application that comes with newer dbghelp. I don' know if it is important but the dependency tree for windbg is: windbg ->dbgeng.dll(native dll in app dir) ->dbghelp windbg ->dbghelp Greetings Peter

July 28, 2005

Re: winecfg in 640x480

by Jesse Allen

On 7/28/05, Joe Baker <joebaker(a)dcresearch.com> wrote: > > On 7/28/05, Oliver Stieber <oliver_stieber(a)yahoo.co.uk> wrote: > > > I know this isn't going to be too big an issue with many people but it > would be nice if winecfg > was usable with a 640x480 desktop, at the moment > it's 20 or so pixels to tall and you can't see > the ok/cancel buttons. > > > > > I agree, although you can hit enter to close out the window and save > changes. > Well, in KDE use the Alt key and the left mouse to click and drag the window > from anywhere in the window to move it higher. I think that blackbox > supports this style of Window movement as well if you're on a lighter > resource computer as your resolution suggests. > > -Joe Baker > > > That's managed, not desktop. Right? =) If it's not managed and in the desktop window @ 640x480 then it's a little hard to reach the OK with the mouse cursor. I actually use TWM, and it allows the same thing, at least according to my twmrc.

July 28, 2005

Re: Frequent and annoying Wine bug

by Francois Gouget

Hi, Robbert Xerox wrote: > Hi , i know this is not a wine-bug channel, but this > bug is affecting quite some apps, and seems to affect > more and more ... > At least five apps from wine-bug mailinglist throw up > an exception which either reads: "Assertion Failed > !bogus context in Local_Unwind()" or "in > Exception_Handler". I have seen a bug pretty much like this in an application and here's what Mike Hearn had to say about it: > There were actually two bugs here: > > * Despite what the unit tests indicate, IShellFolder::BindToObject > with a NULL pidl can sometimes work. I guess it depends on the exact > class implementation. In this case I fixed it with an application > specific hack, we just return the desktop folder (it should probably > return an addreffed This). > > * The app apparently relies on an old Win98 specific bug where > SFGAO_FILESYSTEM is set on My Computer. This bug is documented here: > > http://msdn.microsoft.com/msdnmag/issues/05/06/CAtWork/default.aspx > > at the bottom. But it's not present in Windows NT, as our unit tests > actually prove (we should pay more attention to 98/NT differences!!): > > http://test.winehq.com/data/200506231000/98_PaulVriensW98SEFull/shell32:shl… > > It can be fixed with the patch to wine.inf, if you then do "make > install" in the tools directory and re-run wineprefixcreate. I've attached the patch so you can try it. That's not code I can comment on but maybe Mike will when he's back from his not-quite-vacations. -- Francois Gouget fgouget(a)codeweavers.com Index: dlls/shell32/shlfolder.c =================================================================== RCS file: /var/cvs/wine/dlls/shell32/shlfolder.c,v retrieving revision 1.104 diff -u -p -r1.104 shlfolder.c --- dlls/shell32/shlfolder.c 20 Jul 2005 10:29:05 -0000 1.104 +++ dlls/shell32/shlfolder.c 28 Jul 2005 21:33:11 -0000 @@ -272,6 +272,15 @@ HRESULT SHELL32_BindToChild (LPCITEMIDLI HRESULT hr; LPITEMIDLIST pidlChild; + TRACE("pidlRoot=%p, pathRoot=%s, pidlComplete=%p, riid=%s, ppvOut=%p\n", + pidlRoot, pathRoot, pidlComplete, debugstr_guid(riid), ppvOut); + + if (!pidlComplete) + { + SHELL32_CoCreateInitSF(_ILCreateDesktop(), pathRoot, NULL, &CLSID_ShellDesktop, riid, ppvOut); + return S_OK; + } + if (!pidlRoot || !ppvOut || !pidlComplete || !pidlComplete->mkid.cb) return E_INVALIDARG; Index: tools/wine.inf =================================================================== RCS file: /var/cvs/wine/tools/wine.inf,v retrieving revision 1.37 diff -u -p -r1.37 wine.inf --- tools/wine.inf 27 Jul 2005 15:42:40 -0000 1.37 +++ tools/wine.inf 28 Jul 2005 21:30:35 -0000 @@ -100,6 +100,8 @@ HKCR,TypeLib\{00020430-0000-0000-C000-00 HKCR,TypeLib\{00020430-0000-0000-C000-000000000046}\2.0,,,"OLE Automation" HKCR,TypeLib\{00020430-0000-0000-C000-000000000046}\2.0\0\win32,,,"stdole2.tlb" HKCR,TypeLib\{00020430-0000-0000-C000-000000000046}\2.0\FLAGS,,,"0" +; FIXME: this is to emulate a Windows 98 bug which is gone in NT: very few apps need it +HKCR,CLSID\{20d04fe0-3aea-1069-a2d8-08002b30309d}\ShellFolder,"Attributes",0x10001,f0000154 [ControlClass] HKLM,System\CurrentControlSet\Control\Class\{4d36e978-e325-11ce-bfc1-08002be10318},,,"Ports (COM & LPT)"

July 28, 2005