WineHQ
Bug Tracking Database – Bug 33886

 Bugzilla

 

Last modified: 2014-01-03 13:09:33 UTC  

Anno 1602: Unhandled exception: divide by zero in 32-bit code

Bug 33886 - Anno 1602: Unhandled exception: divide by zero in 32-bit code
Anno 1602: Unhandled exception: divide by zero in 32-bit code
Status: CLOSED FIXED
AppDB: Show Apps affected by this bug
Product: Wine
Classification: Unclassified
Component: -unknown
1.5.31
x86-64 Linux
: P2 normal
: ---
Assigned To: Mr. Bugs
:
Depends on:
Blocks:
  Show dependency tree
 
Reported: 2013-06-25 17:12 UTC by bart
Modified: 2014-01-03 13:09 UTC (History)
1 user (show)

See Also:
Regression SHA1:
Fixed by SHA1:
Distribution: ---
Staged patchset:


Attachments
warn + all output of program execution (128.37 KB, text/plain)
2013-06-25 17:12 UTC, bart
Details
Terminal log + crash in demo version - wine-1.6-rc3-38-g9c49367 (11.24 KB, text/plain)
2013-06-25 18:13 UTC, Frédéric Delanoy
Details
warn + all output of program execution when invoked directly (pf error) (68.18 KB, text/plain)
2013-06-26 01:57 UTC, bart
Details
program output of direct invokation without WINEDEBUG (5.09 KB, text/plain)
2013-06-26 04:39 UTC, bart
Details
program output of autorun invokation (divide by zero exception) without WINEDEBUG (14.18 KB, text/plain)
2013-06-26 04:43 UTC, bart
Details

Note You need to log in before you can comment on or make changes to this bug.
Description bart 2013-06-25 17:12:06 UTC
Created attachment 44971 [details]
warn + all output of program execution

I'm trying to fool around a bit with ancient games (Anno 1602), but it's not really working.  The application loads, shows its video, the promotion screen, and its main menu.
when launching the game, however, the program breaks down during the "loading files" stage with a divide by zero in 32-bit code error.
the backtrace shows the error occured in the main file (1602.exe) and not one of its components

in attachment: the warn+all report of the program execution.  My pc language is set to NL, so bash generates some NL error statements.  I tried to translate most of them, I'm sorry if I forgot one.

kernel version: 3.2.28
fglrx version 13.1 (legacy)
xorg-server 1.12
glibc 2.16
Comment 1 Frédéric Delanoy 2013-06-25 17:53:56 UTC
(In reply to comment #0)
> My pc language is set to NL, so bash generates some NL error statements.
> I tried to translate most of them, I'm sorry if I forgot one.

You can use something like
 
 export LANG="C"

in your terminal before launching wine
Comment 2 Frédéric Delanoy 2013-06-25 18:05:04 UTC
Confirmed in wine-1.6-rc3-38-g9c49367
Comment 3 Frédéric Delanoy 2013-06-25 18:13:08 UTC
Created attachment 44973 [details]
Terminal log + crash in demo version - wine-1.6-rc3-38-g9c49367

Note I don't get a "division by zero" message, but the game crashes
Comment 4 bart 2013-06-25 18:19:10 UTC
(In reply to comment #3)
> Created attachment 44973 [details]
> Terminal log + crash in demo version - wine-1.6-rc3-38-g9c49367
> 
> Note I don't get a "division by zero" message, but the game crashes

that's what happens to me too, when I launch the program directly! 
- if I run the program directly from the hard drive (wine "C:\ANNO1602\1602.exe"), I also get a page fault, and don't even get to the introduction video.  do you get there?
when I run the program via the CD-ROM image (wine "E:\Autostart.exe"), the behaviour is as described above.
I thought Autostart.exe specified some arguments to 1602.exe which I didn't know, so that's why I didn't mention it in the first case
Comment 5 Frédéric Delanoy 2013-06-25 18:39:57 UTC
(In reply to comment #4)
> (In reply to comment #3)
> > Created attachment 44973 [details]
> > Terminal log + crash in demo version - wine-1.6-rc3-38-g9c49367
> > 
> > Note I don't get a "division by zero" message, but the game crashes
> 
> that's what happens to me too, when I launch the program directly! 
> - if I run the program directly from the hard drive (wine
> "C:\ANNO1602\1602.exe"), I also get a page fault, and don't even get to the
> introduction video.  do you get there?
> when I run the program via the CD-ROM image (wine "E:\Autostart.exe"), the
> behaviour is as described above.
> I thought Autostart.exe specified some arguments to 1602.exe which I didn't
> know, so that's why I didn't mention it in the first case

I only have a splash screen for half a second then a crash (no intro video).

Is there an autorun.inf on the CD-ROM root?
Comment 6 bart 2013-06-26 01:57:03 UTC
Created attachment 44979 [details]
warn + all output of program execution when invoked directly (pf error)
Comment 7 bart 2013-06-26 01:59:04 UTC
(In reply to comment #5)
> (In reply to comment #4)
> I only have a splash screen for half a second then a crash (no intro video).
> 
> Is there an autorun.inf on the CD-ROM root?

yes, it specifies to start Autorun.exe (the exec mentioned in my previous post was autostart - that was a mistake, it should have been autorun.exe).
I provided my version of the page fault error too, but I'm not really sure wether everything should be kept in the same ticket.  these might actually be 3 different program malfunctions
Comment 8 Frédéric Delanoy 2013-06-26 03:59:12 UTC
(In reply to comment #7)
> (In reply to comment #5)
> > (In reply to comment #4)
> > I only have a splash screen for half a second then a crash (no intro video).
> > 
> > Is there an autorun.inf on the CD-ROM root?
> 
> yes, it specifies to start Autorun.exe (the exec mentioned in my previous post
> was autostart - that was a mistake, it should have been autorun.exe).

I just wanted to know if there was a specific parameter passed to the app.

> I provided my version of the page fault error too

You should install wine development packages and recreate the log. 
Backtraces are mostly useless without debugging symbols.
Also, don't add specific WINEDEBUG like warn+all unless asked: it just creates noise. Just run the app normally.
 
> but I'm not really sure whether everything should be kept in the same ticket.  > these might actually be 3 different program malfunctions

True. Just regenerate your terminal log and let's see if we have similar backtraces. We can always create other bugs if not.
Comment 9 bart 2013-06-26 04:39:06 UTC
Created attachment 44982 [details]
program output of direct invokation without WINEDEBUG
Comment 10 bart 2013-06-26 04:43:13 UTC
Created attachment 44984 [details]
program output of autorun invokation (divide by zero exception) without WINEDEBUG

there are no arguments provided to autorun.exe
Comment 11 bart 2013-06-26 05:34:59 UTC
my idea is (but this is just an idea) that there's something wrong with the path file handling:
-> when running autostart.exe, the program loads itself from cd-rom image, loads a basic amount of code from cd-rom image for the program start (the cdrom has a 1602.exe on-board, and keeps running until we reach the point where it actually needs to prepare itself for a real "run", eg: no more running from cdrom (loading files ...)
-> when invoking the program directly, it tries to load everything locally, fails horribly, and generates a pf error due to missing dlls  / ...

this might fit, but I'm not a wine expert
Comment 12 Frédéric Delanoy 2013-06-26 06:17:49 UTC
(In reply to comment #9)
> Created attachment 44982 [details]
> program output of direct invokation without WINEDEBUG

Please install wine development packages to get useful backtraces (e.g. wine1.6-dbg in ubuntu wine ppa)
Comment 13 Frédéric Delanoy 2013-06-26 06:42:51 UTC
Created bug 33890 for the demo version problem (different crash)
Comment 14 bart 2013-06-26 07:11:33 UTC
I work with gentoo; I have all debugging software installed; also, all emul-* packages have debugging symbols installed.

(In reply to comment #12)
> (In reply to comment #9)
> > Created attachment 44982 [details]
> > program output of direct invokation without WINEDEBUG
> 
> Please install wine development packages to get useful backtraces (e.g.
> wine1.6-dbg in ubuntu wine ppa)

I invoked both programs with winedbg, but this is extremely weird:

the direct invokation of the program still fails (PF error), no different output here
the autorun invocation partly runs: the introduction video hangs, but you can skip it by pressing "ESC", it then loads the game, and actually runs it ...
Comment 15 Frédéric Delanoy 2013-09-03 04:32:36 UTC
Now that bug 33890 is fixed, could you please retry with latest git (or with wine 1.7.2 when it comes out?)
Comment 16 bart 2013-12-28 00:35:53 UTC
hey, sorry for my extreme late reply, but it works in 1.7.8.
Also the videos are playing perfectly.
The only thing that does not yet work (but this is another bug):
when I run the executable from the c drive, it does not show the play / load options: the display is empty. I can select easy, medium, hard, ... but not the type of game.  this doesn't happen when launching it from the cdrom. let's mark this as a sidenote, not that important: it works, after all :) thanks guys!
Comment 17 Frédéric Delanoy 2013-12-28 03:39:26 UTC
(In reply to comment #16)
> hey, sorry for my extreme late reply, but it works in 1.7.8.
> Also the videos are playing perfectly.
> The only thing that does not yet work (but this is another bug):
> when I run the executable from the c drive, it does not show the play / load
> options: the display is empty. I can select easy, medium, hard, ... but not
> the type of game.  this doesn't happen when launching it from the cdrom.
> let's mark this as a sidenote, not that important: it works, after all :)
> thanks guys!

Feel free to create a bug for this.
Comment 18 Alexandre Julliard 2014-01-03 13:09:33 UTC
Closing bugs fixed in 1.7.10.


Privacy Policy
If you have a privacy inquiry regarding this site, please write to [email protected]

Hosted By CodeWeavers