- Wine-Devel - list.winehq.org

Re: InstallShield and ole question...
by Mike Bond April 26, 2001

April 26, 2001

On Thu, Apr 26, 2001 at 12:55:30PM -0400, Gavriel State wrote: > The fix we put into the TransGaming tree was from Andreas Mohr, and it > fixed some problems related to previous versions of InstallSheild. We > haven't made any progress on v6. > > We'd certainly like to see it get done, but our resources are somewhat > limited at the moment, and we're concentrating on the DirectX work. I understand how that works. > Something that's interesting about the InstallShield objects is this: > They seem to expose a fair amount of their interfaces through OLE > Automation. I don't know for certain, but I have a recollection that > some work was done to support Automation for WordPerfect. I can't > remember whether that work was actually completed, or bypassed through > the use of dde for scripting instead. Hmm, would be interesting to get more details of what became of that work. > If automation does work, then perhaps it's enough to simply run the > InstallShield ikernel.exe and see if the user-side of the installer > works through automation rather than requiring marshalling. Unfortunately I had already tried that, first thing that came to mind once I realized what was going on. -- TTFN MikeB

1 0

dll-separated lzexpand doesn't work
by Ove Kaaven April 26, 2001

April 26, 2001

At least not on debian unstable... I've tracked the problem down to: lzexpand_main.c, function LZCopy, line 525: xread=(_readfun)LZRead; assembles as: movl LZRead(a)GOT(%ebx),%edx movl %edx,-1012(%ebp) so it reads out the address of LZRead out of the GOT. The final result in liblz32.so becomes: 5eb7: 8b 93 44 00 00 00 mov 0x44(%ebx),%edx 5ebd: 89 95 0c fc ff ff mov %edx,0xfffffc0c(%ebp) but because of the -symbolic linker switch, there's no such GOT entry. CRASH! The GOT entry for LZRead should be at 00007d64 in the dynamic relocation table, as far as I can tell from objdump and peeking at the GOT, but in the relocation table there's only this... ... 00007d60 R_386_GLOB_DAT __dso_handle 00007d68 R_386_GLOB_DAT __deregister_frame_info ... What can be done about this? Submit bug reports on ld?

1 0

Using config.status to recreate current setup
by Matthew Cline April 26, 2001

April 26, 2001

According to the comments at the start of config.status, I can run that file to recreate the current setup. However, if I use it, it appears to delete the makefile in wine/dlls/wintrust, causing "make depend" to fail; if I just copy the configure command that config.status has and run that on the command line, everything works fine. Am I doing something wrong? Thanks in advance. -- Matthew Cline | Suppose you were an idiot. And suppose that matt(a)nightrealms.com | you were a member of Congress. But I repeat | myself. -- Mark Twain

2 1

Re: Trouble with VirtualQuery()
by gerard patel April 25, 2001

April 25, 2001

At 02:07 PM 25/04/2001 +0200, you wrote: >Do you mean >http://support.microsoft.com/support/kb/articles/Q171/7/93.ASP >? Yes Gerard

1 0

Re: New breakages
by gerard patel April 25, 2001

April 25, 2001

At 09:30 AM 25/04/2001 +0200, I wrote: >There may be other problems, I have not the time to search for them all just now. Also several apps are not erasing parts of the screen : among others the word viewer 97 (freeware downloadable). A few other strange crashes. Gerard

1 0

Re: PATCH: drop relay stuff
by Andreas Mohr April 25, 2001

April 25, 2001

On Tue, Apr 24, 2001 at 09:33:16PM +0200, Marcus Meissner wrote: > Hi, > > Changelog: > Drop SYSLEVEL checks from relay debugging, since they break debugging > builtin GDI dlls. Hmm, do we really want to do this ? (I know it's already been applied) Shouldn't we create a different SYSLEVEL_CheckNotLevel_unenforced() function instead which simply doesn't call DebugBreak() ? That way we don't lose the important information about possible syslevel violations. Syslevel violations are an *ongoing* problem. They are not simply "solved". OTOH I could take Alexandre's cvs commit as his firm promise to go through every future patch line by line to check for syslevel mishaps/omissions ;-) Oh yeah, and in case you didn't notice: it's me again complaining about yet another silencing ;-) This patch is not even "too early", it's simply "wrong" IMHO, as it is an ongoing problem (I did mention this before, didn't I ? :). Sorry for me being of such an opposing nature ;-) -- Andreas Mohr Stauferstr. 6, D-71272 Renningen, Germany Tel. +49 7159 800604 http://home.germany.net/100-30936/

2 3

Re: Trouble with VirtualQuery()
by Uwe Bonnes April 25, 2001

April 25, 2001

>>>>> "Gerald" == <<gerard.patel(a)asi.fr>@s5.asi.fr> writes: Gerard > ... Gerard > Possibly I'll send a new patch if I find where is the flag Gerard > in the exe to handle the 3 gig memory space but as I am not Gerard > actively searching for it, you are welcome to do it yourself Gerard > :-) Do you mean http://support.microsoft.com/support/kb/articles/Q171/7/93.ASP ? -- Uwe Bonnes bon(a)elektron.ikp.physik.tu-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------

1 0

New breakages
by gerard patel April 25, 2001

April 25, 2001

Unhandled exception: page fault on read access to 0x00000000 in 32-bit code (0x40a39b1b). In 32-bit mode. Symbol h_errno is invalid Symbol hack_digit is invalid 0x40a39b1b (X11DRV_WND_SetGravity+0x1f [wnd.c:414]): cmpl $0,0x0(%edx) 414 if (data->window && data->bit_gravity != value ) Wine-dbg>bt bt Backtrace: =>0 0x40a39b1b (X11DRV_WND_SetGravity+0x1f(wnd=0x40367b80, value=0x1) [wnd.c:414]) (ebp=40c56b8c) 1 0x40a149dc (X11DRV_SetWindowPos+0x678(winpos=0x40c56c54) [winpos.c:1076]) (ebp=40c56c3c) 2 0x406acb69 (SetWindowPos+0x51(hwnd=0x1a20, hwndInsertAfter=0x0, x=0xdd, y=0xcf, cx=0x172, cy=0xef, flags=0x80000015) [winpos.c:2176]) (ebp=40c56c70) 3 0x40a3481e (EVENT_ConfigureNotify+0x1a6(hWnd=0x1a20, event=0x40c56d90) [event.c:983]) (ebp=40c56ccc) 4 0x40a33b0f (EVENT_ProcessEvent+0x49b(event=0x40c56d90) [event.c:438]) (ebp=40c56d70) 5 0x40a335ec (EVENT_ProcessAllEvents+0x68(arg=0x0) [event.c:209]) (ebp=40c56df0) 6 0x400c9b7f (SERVICE_Loop+0xbf(dummy=0x0) [services.c:76]) (ebp=40c56f1c) Hacked away by --- wnd.c.orig Wed Apr 25 01:43:41 2001 +++ wnd.c Wed Apr 25 09:28:52 2001 @@ -411,7 +411,7 @@ { X11DRV_WND_DATA *data = wnd->pDriverData; - if (data->window && data->bit_gravity != value ) + if (data && data->window && data->bit_gravity != value ) { XSetWindowAttributes win_attr; win_attr.bit_gravity = value; Another problem : now in the file open window, the tooltips for the toolbar buttons are appearing as managed windows (in managed mode of course) There may be other problems, I have not the time to search for them all just now. Gerard

1 0

Re: GDI syslevel held during psdrv init ... bad
by Alexandre Julliard April 24, 2001

April 24, 2001

Marcus Meissner <marcus(a)jet.franken.de> writes: > Now the reason is, that pCreateDC is always called with the GDI syslevel (3) > held, so _any_ call to relaying here will dump. > > How do we fix this? We could probably get rid of the check. Syslevel locking is reasonably stable now, and dll separation will cause more and more false alarms here. -- Alexandre Julliard julliard(a)winehq.com

1 0

Re: fault handling - oof
by Alexandre Julliard April 24, 2001

April 24, 2001

lawson_whitney(a)juno.com writes: > I got what looked like a recursive page fault somewhere in > startup/fixup_imports until I reverted all 3. A fixup_imports crash sounds a bit like what other people reported seeing with the shared sections fix. Did you run two instances of the application by any chance? -- Alexandre Julliard julliard(a)winehq.com

2 1