Uwe Bonnes wrote:
Changelog windows/winproc.c: WINPROC_wrapper Add some padding between saved registers and the winproc arguments. Some winprocs may trash the register area.
As Mike Hearn suggested, adding some padding helped with my problem with WebPACK_42wp30_full_installer.exe. I hope, I found the right way to add that padding.
It may work for one application, but it doesn't look like a propper fix. I'd suggest there's probably something else going on that's causing stack corruption.
Mike
"Mike" == Mike McCormack mike@codeweavers.com writes:
Mike> Uwe Bonnes wrote: >> Changelog windows/winproc.c: WINPROC_wrapper Add some padding between >> saved registers and the winproc arguments. Some winprocs may trash >> the register area. >> >> As Mike Hearn suggested, adding some padding helped with my problem >> with WebPACK_42wp30_full_installer.exe. I hope, I found the right way >> to add that padding.
Mike> It may work for one application, but it doesn't look like a Mike> propper fix. I'd suggest there's probably something else going on Mike> that's causing stack corruption.
Mike,
any ideas how to pinpoint that problem? Any debug output you need?
Bye
Uwe Bonnes wrote:
any ideas how to pinpoint that problem? Any debug output you need?
I tried installing WebPACK_42wp30_full_installer.exe, but it seemed to have other problems. Are you installing it on the Wine CVS? Did you install other software into your ~/.wine first ?
Mike
"Mike" == Mike McCormack mike@codeweavers.com writes:
Mike> Uwe Bonnes wrote:
>> any ideas how to pinpoint that problem? Any debug output you need?
Mike> I tried installing WebPACK_42wp30_full_installer.exe, but it Mike> seemed to have other problems. Are you installing it on the Wine Mike> CVS? Did you install other software into your ~/.wine first ?
I have tried on several machines, all with slightly modified trees, but nothing that should influence the behaviour. All machines showed the same error.
The problem happens quite early, when the installation wrapper asks for a location to unzip the file. If the error happens, nothing is displayed. If the wrapper error doesn't happen. a Messagebox "WebPACK" : To begin the installation... is displayed.
Bye
Uwe Bonnes wrote:
The problem happens quite early, when the installation wrapper asks for a location to unzip the file. If the error happens, nothing is displayed. If the wrapper error doesn't happen. a Messagebox "WebPACK" : To begin the installation... is displayed.
Are you talking about the dialog in the attached screenshot? I have no trouble getting past that screen. InstallShield's wizard starts running for me, and starts copying files.
Perhaps the problem is due to a missing registry key or directory? If you'd like I can send you a relay trace to compare.
Mike
"Mike" == Mike McCormack mike@codeweavers.com writes:
Mike> Uwe Bonnes wrote:
>> The problem happens quite early, when the installation wrapper asks >> for a location to unzip the file. If the error happens, nothing is >> displayed. If the wrapper error doesn't happen. a Messagebox >> "WebPACK" : To begin the installation... is displayed.
Mike> Are you talking about the dialog in the attached screenshot? I Mike> have no trouble getting past that screen. InstallShield's wizard Mike> starts running for me, and starts copying files.
Without the patch, I don't get to that dialogue,
Mike> Perhaps the problem is due to a missing registry key or directory? Mike> If you'd like I can send you a relay trace to compare.
Please do so.
Uwe Bonnes wrote:
Mike> Perhaps the problem is due to a missing registry key or directory? Mike> If you'd like I can send you a relay trace to compare.Please do so.
Strangely enough when I try to do so, it crashes :) I'll investigate a bit.
Mike
It may work for one application, but it doesn't look like a propper fix. I'd suggest there's probably something else going on that's causing stack corruption.
Probably it was declared with the wrong calling convention. Windows deals with this transparently but perhaps more transparently than us :)
So the question is: what calling conventions could cause the observed behaviour?
-
Mike Hearn -
Mike McCormack -
Uwe Bonnes