July 2003 - Wine-devel - list.winehq.org

Relocate error
by Fabian Cenedese 10 Sep '03

10 Sep '03

Hi I worked with wine about six months ago. Except some minor errors my app worked quite good in wine. Then I didn't have time anymore. Now I tried again with the newest wine from cvs. But now I can't even start my app anymore. Wine gives me this error: err:virtual:map_image FATAL: Need to relocate module from addr 40000000, but there are no relocation records wine: could not load 'C:\IMD\Bin\Imd.exe' as Win32 binary I looked in the archives and the only messages concerning this error were at least two years old stating that this error has been fixed long ago. I didn't change this link address in my app (VC6), it has been there even before I started with wine at all. But when I change this address to say 0x48000000 then it works. So is there something else on this address? I know six months is a long time and a lot happened this year. But does anybody know if this is a real error? What could have introduced this behaviour as it worked a few months ago? Thanks Fabi

3 5

problem with apps in desktop mode
by Rok Mandeljc 19 Aug '03

19 Aug '03

When I try running InstallShield in windowed mode ("Desktop" = "640x480" as per default in config), window appears for a second and then dissapears printing this message: X Error of failed request: BadMatch (invalid parameter attributes) Major opcode of failed request: 12 (X_ConfigureWindow) Serial number of failed request: 93 Current serial number in output stream: 98 Did latest modifications of x11drv break anything? -- Rok Mandeljc <rok.mandeljc(a)gimb.org> "All that is gold does not glitter, Not all those who wander are lost; The old that is strong does not wither, Deep roots are not reached by the frost. >From the ashes a fire shall be woken, A light from the shadows shall spring; Renewed shall be blade that was broken, The crownless again shall be king." -- J.R.R. Tolkien

4 3

Re: Sync(1): include/winerror.h
by Alexandre Julliard 16 Aug '03

16 Aug '03

Jon Griffiths <jon_p_griffiths(a)yahoo.com> writes: > This patch casts hresult errors into HRESULTs as per the native > headers. The native headers don't have a _HR_() macro, so we shouldn't have one either. Please do normal casts instead. -- Alexandre Julliard julliard(a)winehq.com

2 1

RPM for Red Hat 7.3
by William M. Quarles 14 Aug '03

14 Aug '03

Is it possible that you could continue building and distributing packages for Red Hat 7.3? A lot of people are still using it, and I was able to build your "RH8" source rpm just fine on an RH 7.3 system -- William M. Quarles quarlewm(a)jmu.edu wquarles(a)bucknell.edu walrus(a)bellsouth.net

3 4

DRIVE_GetUnixCwd and drives other than the current one
by Troy Rollo 12 Aug '03

12 Aug '03

It appears that Wine is storing and using a "current directory" for each DOS drive. According to some testing we have just done on Windows 2000 and Windows 98, Windows only stores a current directory for the current drive. "But CMD.EXE seems to remember drives, and their children too", comes the reply. Well, sort of. The documentation for CreateProcess talks about environment variables used to hold the current directory for each drive, with an environment variable with a name of the form "=C:" having the value, say "C:\WINDOWS". That documentation suggests you must make sure all these environment variables are set if you construct a new environment block, and that you need to get the values of them using "GetFullPathName". Shock, horror, the Windows documentation is wrong. What actually happens is that CMD.EXE and COMMAND.COM independently set those environment variables when you change directories in those shells. SetCurrentDirectory has no effect on those environment variables - either in the current process or in child processes. It is the responsibility of the application to set them if it wants the children to have access to its current directory on drives other than its current drive. However, SetCurrentDirectory("D:") (assuming a starting point of drive "C") will read those environment variables to determine which directory of drive "D" to go to. You can even 'SetEnvironmentVariable("=D:", "\\foo")' to make the next 'SetCurrentDirectory("D:")' end up in "D:\foo". GetFullPathName also picks up the environment variable if the path requested is not on the current drive (although it also checks that the path is valid, so 'GetFullPathName("A:\\", ...)' will rattle drive A if drive A is empty). If there's no environment variable, both GetFullPathName and SetCurrentDirectory assume "x:\". If the drive in question is the current drive, SetCurrentDirectory and GetFullPathName both ignore the environment variables and use the value from GetCurrentDirectory. So, for example: // Start with no "=D:" variable in environment SetCurrentDirectory("d:\\bar"); SetCurrentDirectory("c:"); SetCurrentDirectory("d:"); // Current directory now d:\ SetCurrentDirectory("d:\\bar"); SetCurrentDirectory("c:"); SetEnvironmentVariable("=D:", "\\foo"); SetCurrentDirectory("d:"); // Current directory now "d:\\foo"; SetEnvironmentVariable("=D:", "\\bar"); SetCurrentDirectory("d:"); // Current directory still d:\\foo The Borland C++ run-time libraries have code in their chdir functions that sets these variables, but it's inside a "#if 0" section, presumably reflecting an attempt to be compatible with MSC. I'm not aware of anything that relies on this perverse behaviour, so whether Wine needs to be adjusted to mimic it may be an open question. It is probably of more interest to people who want to pass an environment block to CreateProcess. If you're doing that, ignore what the documentation says about including "=C:" & etc environment variables in the environment block - simply start with the environment block returned by GetEnvironmentBlock and make the desired changes, ensuring the block is sorted according to instructions I gave previously on one of the Cygwin mailing lists.

2 1

can't redirect stdout?!
by Dan Kegel 12 Aug '03

12 Aug '03

OK, so with the normal X11 driver, you can no longer redirect stdout from commandline apps under Wine; the output just vanishes if you try to send it to a file. That's rather unacceptable dain bramage, if you ask me. - Dan -- Dan Kegel http://www.kegel.com http://counter.li.org/cgi-bin/runscript/display-person.cgi?user=78045

4 5

Is wine a debugger?
by Dan Timis 12 Aug '03

12 Aug '03

I am trying to load a security Windows dll that does not run under a debugger. When I load the dll I get a SIGTRAP, and then wine breaks in its debugger. My understanding is that if I run wine as opposed to winedbg I am not running under a debugger. But, if there is a crash or some other exception wine will start a debugger. I sent some memory dumps to the developer of the security dll, and he said that it behaves like it was running under a debugger. He's question was something like does wine have an always "on" debugger? I would appreciate any help. Thanks, Dan Timis Muse Research, Inc.

3 2

Re: WaitForMultipleObjects loops forever if passed an array of size zero
by Alexandre Julliard 05 Aug '03

05 Aug '03

Troy Rollo <wine(a)troy.rollo.name> writes: > @@ -412,7 +412,7 @@ > { > int ret, cookie; > > - if (count > MAXIMUM_WAIT_OBJECTS) return STATUS_INVALID_PARAMETER_1; > + if (count > MAXIMUM_WAIT_OBJECTS || count <= 0) return > STATUS_INVALID_PARAMETER_1; This is wrong, 0 is a valid count, and shouldn't cause an infinite loop. You'll need to debug this some more. -- Alexandre Julliard julliard(a)winehq.com

6 12

Profiling wine
by Mike Hearn 04 Aug '03

04 Aug '03

Hi, Is there any way to get measurements of where Wine is spending most of its time? I have a problem where typing text into the WebBrowser control is incredibly slow, but don't really have a clue where to start looking. A profiling tool would be handy about now...... thanks -mike

6 6

Re: Small FAQ fixes
by Tom 03 Aug '03

03 Aug '03

Francois Gouget wrote: >Changelog: > > * documentation/faq.sgml > > Small spelling/punctuation fixes. > Small rephrasings. > > > > Hi, Small question.. If I take Mikes updates and these updates and apply them againt the faq.template in /lostwages can we just update the faq there ? And whenever a large amount of changes have been made just build a .sgml of it. Or just make sure we update both at the same time ? Because the way were going I have sent patches to update the .template and you guy's have sent patches againt this one and before long there going to be two very different faq's. Comments ? Tom

2 2