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

Re: MSHTML: fix blank.htm resources

by Alexandre Julliard

Jacek Caban <jack(a)itma.pwr.wroc.pl> writes: > This patch depends on my patch to wrc. Alexandre, > I discovered why the test didn't fail for me. There is > a bug in Wine that causes loading of the native dll instead > of builtin when you load it as a data file, so I tested the > builtin dll with native resources. That's a feature ;-) We can't support loading builtins as datafiles, so the idea is that it's more likely that the app will be happy with loading the native as datafile than loading the builtin in normal mode. It does make the behavior a bit surprising I agree. -- Alexandre Julliard julliard(a)winehq.org

June 30, 2005

Re: Sony Station

by Oliver Stieber

--- Chuck Hall <chall(a)bcrl.stcloudstate.edu> wrote: > Was there a specific game that you wanted to test > here? > Well, I've just got a copy of Everquest II, but it would be nice to get any of the Sony Station games working. It looks like I can just change a setting in the registry so long as I've run the post-scan file update under windows. > I have not tried to install any game that required > it, but I can if you > want something tested. > > Have Fun! > Chuck Hall > > On Wed, 29 Jun 2005, Oliver Stieber wrote: > > > Hi, > > > > Has anyone managed to get sony's station > > (http://www.station.sony.com/en/) past the file > scan? > > > > > > > > > > > > > ___________________________________________________________ > > > Yahoo! Messenger - NEW crystal clear PC to PC > calling worldwide with voicemail > http://uk.messenger.yahoo.com > > > > -- > > > ___________________________________________________________ How much free photo storage do you get? Store your holiday snaps for FREE with Yahoo! Photos http://uk.photos.yahoo.com

June 30, 2005

Re: Fix ntdll compilation in Darwin

by Phil Krylov

Hi Dmitry, On Thu, 30 Jun 2005 22:27:01 +0900 "Dmitry Timoshkov" <dmitry(a)baikal.ru> wrote: > "Phil Krylov" <phil(a)newstar.rinet.ru> wrote: > > > Index: dlls/ntdll/cdrom.c > > =================================================================== > > RCS file: /home/wine/wine/dlls/ntdll/cdrom.c,v > > retrieving revision 1.59 > > diff -p -u -r1.59 cdrom.c > > --- dlls/ntdll/cdrom.c 29 Jun 2005 19:18:54 -0000 1.59 > > +++ dlls/ntdll/cdrom.c 30 Jun 2005 11:27:41 -0000 > > @@ -75,6 +75,11 @@ > > # include <sys/scsiio.h> > > #endif > > > > +#ifdef __APPLE__ > > +# include <IOKit/IOKitLib.h> > > +# include <IOKit/scsi/SCSICmds_REQUEST_SENSE_Defs.h> > > +#endif > > + > > #define NONAMELESSUNION > > #define NONAMELESSSTRUCT > > #include "ntstatus.h" > > @@ -1418,6 +1423,8 @@ static NTSTATUS CDROM_ScsiPassThroughDir > > > > #ifdef SENSEBUFLEN > > if (pPacket->SenseInfoLength > SENSEBUFLEN) > > +#elif defined( __APPLE__ ) > > + if (pPacket->SenseInfoLength > kSenseDefaultSize) > > #else > > if (pPacket->SenseInfoLength > sizeof(struct request_sense)) > > #endif > > Please do not introduce even more platform dependent #ifdefs into cdrom.c, > currently it already has enough mess with all that #if defined(linux)/(__FreeBSD__)/ > (__NetBSD__). Add proper configure checks for headers and structures your > platform needs. To be sure I get you right: would changing __APPLE__ to HAVE_IOKIT (and adding HAVE_IOKIT detection to configure) be sufficient? -- Ph.

June 30, 2005

Re: [user/tests]: Re: Fix bug that left mouse buttons swapped after tests

by Saulius Krasuckas

* On Thu, 30 Jun 2005, Paul Vriens wrote: > > > Saulius Krasuckas <saulius.krasuckas(a)ieee.org> > > - Break from the loop to restore SM_SWAPBUTTON metrics. ... > > --- dlls/user/tests/sysparams.c 20 Jun 2005 15:36:17 -0000 1.38 > > +++ dlls/user/tests/sysparams.c 30 Jun 2005 07:36:40 -0000 > > @@ -1011,7 +1011,7 @@ static void test_SPI_SETMOUSEBUTTONSWAP( > > rc=SystemParametersInfoA( SPI_SETMOUSEBUTTONSWAP, vals[i], 0, > > SPIF_UPDATEINIFILE | SPIF_SENDCHANGE ); > > if (!test_error_msg(rc!=vals[i],"SPI_{GET,SET}MOUSEBUTTONSWAP")) > > - return; > > + break; > > It's nice to break but the the end-result will be that the test fails on > all windows versions I could my hands on Test fails because of flawed logic at if-statement, which I have no time to debug for. So I chose to have flawed check instead of having mouse-buttons swapped (which isn't so visible from the test results). > and that the previous setting is restored. Not very useful it seams. So if the buttons were standard mapped, then they got swapped during test and then test failed, you think it is OK to leave buttons swapped?

June 30, 2005

Question about deadlock in winmm

by Vitaliy Margolen

Lately I'm seeing a lot of deadlocks in wine's sound system. Most notably winealsa. But the code in question affects all (in my case oss since it's working better for me then others). I'm not sure what exactly happens (deadlock because of race condition or else) but code bellow deadlocks with other heap functions (like GetHeap). Removing HeapValidate didn't cause any problems for as of yet. And I haven't seen a single deadlock here at all. Is there are reasons we perform this heap validation here? Could someone comment on this? If someone interested I could reproduce backtrace of the deadlock. -- Best regards, Vitaliy Index: dlls/winmm/lolvldrv.c =================================================================== RCS file: /home/wine/wine/dlls/winmm/lolvldrv.c,v retrieving revision 1.68 diff -u -p -r1.68 lolvldrv.c --- dlls/winmm/lolvldrv.c 22 Jun 2005 11:59:43 -0000 1.68 +++ dlls/winmm/lolvldrv.c 29 Jun 2005 14:54:42 -0000 @@ -441,7 +441,7 @@ LPWINE_MLD MMDRV_Get(HANDLE _hndl, UINT hndl = hndl & ~0x8000; if (hndl < sizeof(MM_MLDrvs) / sizeof(MM_MLDrvs[0])) { mld = MM_MLDrvs[hndl]; - if (!mld || !HeapValidate(GetProcessHeap(), 0, mld) || mld->type != type) + if (!mld || mld->type != type) mld = NULL; } hndl = hndl | 0x8000;

June 30, 2005

Re: Fix ntdll compilation in Darwin

by Dmitry Timoshkov

"Phil Krylov" <phil(a)newstar.rinet.ru> wrote: >> > #ifdef SENSEBUFLEN >> > if (pPacket->SenseInfoLength > SENSEBUFLEN) >> > +#elif defined( __APPLE__ ) >> > + if (pPacket->SenseInfoLength > kSenseDefaultSize) >> > #else >> > if (pPacket->SenseInfoLength > sizeof(struct request_sense)) >> > #endif >> >> Please do not introduce even more platform dependent #ifdefs into cdrom.c, >> currently it already has enough mess with all that #if defined(linux)/(__FreeBSD__)/ >> (__NetBSD__). Add proper configure checks for headers and structures your >> platform needs. > > To be sure I get you right: would changing __APPLE__ to HAVE_IOKIT (and > adding HAVE_IOKIT detection to configure) be sufficient? I'd assume it's enough, yes. kSenseDefaultSize requires another test I think. -- Dmitry.

June 30, 2005

Re: Fix ntdll compilation in Darwin

by Dmitry Timoshkov

"Phil Krylov" <phil(a)newstar.rinet.ru> wrote: > Index: dlls/ntdll/cdrom.c > =================================================================== > RCS file: /home/wine/wine/dlls/ntdll/cdrom.c,v > retrieving revision 1.59 > diff -p -u -r1.59 cdrom.c > --- dlls/ntdll/cdrom.c 29 Jun 2005 19:18:54 -0000 1.59 > +++ dlls/ntdll/cdrom.c 30 Jun 2005 11:27:41 -0000 > @@ -75,6 +75,11 @@ > # include <sys/scsiio.h> > #endif > > +#ifdef __APPLE__ > +# include <IOKit/IOKitLib.h> > +# include <IOKit/scsi/SCSICmds_REQUEST_SENSE_Defs.h> > +#endif > + > #define NONAMELESSUNION > #define NONAMELESSSTRUCT > #include "ntstatus.h" > @@ -1418,6 +1423,8 @@ static NTSTATUS CDROM_ScsiPassThroughDir > > #ifdef SENSEBUFLEN > if (pPacket->SenseInfoLength > SENSEBUFLEN) > +#elif defined( __APPLE__ ) > + if (pPacket->SenseInfoLength > kSenseDefaultSize) > #else > if (pPacket->SenseInfoLength > sizeof(struct request_sense)) > #endif Please do not introduce even more platform dependent #ifdefs into cdrom.c, currently it already has enough mess with all that #if defined(linux)/(__FreeBSD__)/ (__NetBSD__). Add proper configure checks for headers and structures your platform needs. -- Dmitry.

June 30, 2005

Re: unixfs registered by default now in cvs

by Michael Jung

On Thursday 30 June 2005 15:01, Dimi Paun wrote: > This is fantastic! Way to go Michael, you've nailed an important > integration issue. BTW, if it is registered by default, why do we > need to 'regsvr32 shell32'? Thanks. If you start without a .wine directory, you won't need the 'regsvr32 shell32'. So for a fresh install, you will end up with unixfs. But if you already have a '.wine' directory and want unixfs to be registered, you'll have to do it by hand. Bye, -- Michael Jung mjung(a)iss.tu-darmstadt.de

June 30, 2005

The wine-patches archives

by Tom Wickline

Looks like ""Downloadable version" is stuck at 9 MB last time I checked we were at a new record of 10.2 MB ......... Newman to the rescue ? Cheers, Tom

June 30, 2005

Re: unixfs registered by default now in cvs

by Dimi Paun

On Thu, 2005-06-30 at 14:05 +0200, Michael Jung wrote: > With the current CVS version, the unixfs shell namespace extension is > now registered by default at the desktop. So if you do a 'regsvr32 > shell32' you will see the unix filesystem in the file dialogs. This is fantastic! Way to go Michael, you've nailed an important integration issue. BTW, if it is registered by default, why do we need to 'regsvr32 shell32'? -- Dimi Paun <dimi(a)lattica.com> Lattica, Inc.

June 30, 2005