http://bugs.winehq.org/show_bug.cgi?id=12874
Summary: The Master Genealogist 4.0d - Unhandled page fault on write access Product: Wine Version: 0.9.60 Platform: PC URL: http://appdb.winehq.org/objectManager.php?sClass=version &iId=11482 OS/Version: Linux Status: UNCONFIRMED Severity: blocker Priority: P5 Component: dos AssignedTo: wine-bugs@winehq.org ReportedBy: saturn_systems@yahoo.com
Created an attachment (id=12587) --> (http://bugs.winehq.org/attachment.cgi?id=12587) Unhandled page fault on write access for tmg4.0d
When attempting to run TMG4.0d as windows 3.1 wine gives a Unhandled page fault on write access
http://bugs.winehq.org/show_bug.cgi?id=12874
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Severity|blocker |normal Keywords| |win16 Priority|P5 |P2
--- Comment #1 from Austin English austinenglish@gmail.com 2008-04-29 20:21:09 --- Not a blocker.
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #2 from nathan.n saturn_systems@yahoo.com 2008-04-30 20:56:27 --- sorry slip of the mouse!
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #3 from Austin English austinenglish@gmail.com 2008-10-30 02:18:56 --- Is this still an issue in current (1.1.7 or newer) wine?
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #4 from nathan.n saturn_systems@yahoo.com 2008-11-01 00:23:47 --- Just tried in wine 1.17 and the result looks almost the same.
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #5 from nathan.n saturn_systems@yahoo.com 2008-11-01 00:25:55 --- Created an attachment (id=17019) --> (http://bugs.winehq.org/attachment.cgi?id=17019) Unhandled page fault on write access for tmg4.0d wine1.17
http://bugs.winehq.org/show_bug.cgi?id=12874
nathan.n saturn_systems@yahoo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Version|0.9.60 |1.1.7
http://bugs.winehq.org/show_bug.cgi?id=12874
Lei Zhang thestig@google.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Version|1.1.7 |0.9.60
--- Comment #6 from Lei Zhang thestig@google.com 2008-11-05 13:42:01 --- Please don't change the original reported version.
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #7 from nathan.n saturn_systems@yahoo.com 2008-12-15 23:22:45 --- still an issue in wine 1.1.10
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #8 from nathan.n saturn_systems@yahoo.com 2009-01-03 15:51:36 --- Created an attachment (id=18439) --> (http://bugs.winehq.org/attachment.cgi?id=18439) Crashed TMG4 Menu shown in emulated desktop in wine 1.1.12
In emulated desktop, the crashed TMG4 Menu is shown. With normal desktop TMG4 crashes after something flashes up quick and disappears instead of showing the TMG4 menu like in previous Wine's.
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #9 from nathan.n saturn_systems@yahoo.com 2009-01-03 15:53:42 --- Created an attachment (id=18440) --> (http://bugs.winehq.org/attachment.cgi?id=18440) Terminal outputs with emulated desktop & without & with ulimit set as recommend
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #10 from nathan.n saturn_systems@yahoo.com 2009-01-03 15:57:11 --- This is what the TMG4 looks like when it works
http://appdb.winehq.org/appimage.php?iId=16974
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #11 from nathan.n saturn_systems@yahoo.com 2009-01-03 17:07:47 --- Seems to be an issue with bug #16754 "Some applications lead to crash of wineserver in Wine 1.1.12"
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #12 from Jeff Zaroyko jeffz@jeffz.name 2009-01-03 23:52:08 --- (In reply to comment #11)
Seems to be an issue with bug #16754 "Some applications lead to crash of wineserver in Wine 1.1.12"
That's fixed now, try again with latest git.
http://bugs.winehq.org/show_bug.cgi?id=12874
nathan.n saturn_systems@yahoo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Attachment #18440|0 |1 is obsolete| |
--- Comment #13 from nathan.n saturn_systems@yahoo.com 2009-01-19 00:04:34 --- Created an attachment (id=18819) --> (http://bugs.winehq.org/attachment.cgi?id=18819) Unhandled page fault on write access for tmg4.0d wine1.1.13
(In reply to comment #12)
(In reply to comment #11)
Seems to be an issue with bug #16754 "Some applications lead to crash of wineserver in Wine 1.1.12"
That's fixed now, try again with latest git.
Thanks, back to the original issue in wine 1.1.13.
Can I do anything else?
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #14 from nathan.n saturn_systems@yahoo.com 2009-05-10 18:19:26 --- Still current in wine 1.1.21
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #15 from Austin English austinenglish@gmail.com 2009-11-19 12:52:30 --- This is your friendly reminder that there has been no bug activity for 6 months. Is this still an issue in current (1.1.33 or newer) wine?
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #16 from nathan.n saturn_systems@yahoo.com 2009-11-19 22:27:46 --- (In reply to comment #15)
This is your friendly reminder that there has been no bug activity for 6 months. Is this still an issue in current (1.1.33 or newer) wine?
Thanks Austin,
I can confirm this happens in the current in wine 1.1.33
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #17 from Austin English austinenglish@gmail.com 2010-05-19 16:45:21 --- This is your friendly reminder that there has been no bug activity for 6 months. Is this still an issue in current (1.1.44 or newer) wine?
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #18 from nathan.n saturn_systems@yahoo.com 2010-07-14 02:55:08 --- Still current in wine 1.2-rc7
Installation works without issue now.
But running the main program causes the Unhandled page fault
http://bugs.winehq.org/show_bug.cgi?id=12874
joaopa jeremielapuree@yahoo.fr changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |jeremielapuree@yahoo.fr
--- Comment #19 from joaopa jeremielapuree@yahoo.fr 2010-07-14 05:15:20 --- Does the following demo show the problem http://www.whollygenes.com/files/tmg7setup.exe
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #20 from nathan.n saturn_systems@yahoo.com 2010-07-14 06:30:55 --- (In reply to comment #19)
Does the following demo show the problem http://www.whollygenes.com/files/tmg7setup.exe
No it does not, this bug is for TMG version 4.0d, which is no longer available from the website, if it ever was?
If you are interested in testing TMG4 I can provide the installation files (without the serial number, which is not needed to trigger this bug) email me directly.
http://bugs.winehq.org/show_bug.cgi?id=12874
Andrew Nguyen arethusa26@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |NEW Ever Confirmed|0 |1
--- Comment #21 from Andrew Nguyen arethusa26@gmail.com 2010-07-14 21:52:14 --- For the first crash:
trace:int31:DOSVM_Int31Handler get free memory information trace:int31:DOSVM_Int31Handler resize memory block (0x00bf0000, 2148921344 bytes) fixme:int31:DPMI_xalloc failed to allocate linearly growing memory (2148921344 bytes), using non-linear growing... fixme:int31:DPMI_xalloc failed to allocate any memory of 2148921344 bytes! wine: Unhandled page fault on write access to 0x00000000 at address 0x681f35f6 (thread 001c), starting debugger... Unhandled exception: page fault on write access to 0x00000000 in 32-bit code (0x681f35f6). ... Backtrace: =>0 0x681f35f6 (0x00aedf68) 1 0x685cfe3d DOSVM_Int31Handler+0x228b(context=0xbf0000) [/home/arethusa/wine32/dlls/krnl386.exe16/../../../wine.git/dlls/krnl386.exe16/int31.c:1380] in krnl386.exe16 (0x00aee268) 2 0x685cfe3d DOSVM_Int31Handler+0x228b(context=0xaee2fc) [/home/arethusa/wine32/dlls/krnl386.exe16/../../../wine.git/dlls/krnl386.exe16/int31.c:1380] in krnl386.exe16 (0x00aee298) ...
What happens is that the application tries to resize a memory block, but the DPMI_xrealloc helper calls DPMI_xalloc and assumes success in the block resize case. The application passes an extremely large allocation size (around 2 GB), so DPMI_xalloc fails and DPMI_xrealloc blindly uses the returned NULL pointer. Making xrealloc recognize this condition seems to fix the crash.
For the next problem:
fixme:int31:DPMI_xalloc failed to allocate linearly growing memory (2148921344 bytes), using non-linear growing... fixme:int31:DPMI_xalloc failed to allocate any memory of 2148921344 bytes! fixme:int31:DPMI_xalloc failed to allocate linearly growing memory (2148511744 bytes), using non-linear growing... fixme:int31:DPMI_xalloc failed to allocate any memory of 2148511744 bytes! fixme:int31:DPMI_xalloc failed to allocate linearly growing memory (2148102144 bytes), using non-linear growing... fixme:int31:DPMI_xalloc failed to allocate any memory of 2148102144 bytes! ...
For some reason, the application uses the queried DPMI free memory information to try to allocate all available memory, and it spends a huge amount of time doing this until it crashes. Hacking DPMI_xalloc to reject allocation sizes larger than 512 MB (or some reasonable arbitrary value) gets the application to start. Perhaps the DPMI free memory information needs some hard limits for applications like this one. There are many more issues after this, but that would seem to belong in another bug report.
http://bugs.winehq.org/show_bug.cgi?id=12874
nathan.n saturn_systems@yahoo.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED
--- Comment #22 from nathan.n saturn_systems@yahoo.com 2010-08-05 20:38:04 --- (In reply to comment #21)
For the first crash:
<snip>
Unhandled exception: page fault on write access to 0x00000000 in 32-bit code (0x681f35f6).
<snip>
Making xrealloc recognize this condition seems to fix the crash.
Thanks Andrew, I can confirm that your patch in wine 1.3.0 fixes the "Unhandled page fault on write access"
Should I open a bug report for the second issue you mention? How do I hack DPMI_xalloc to reject allocation sizes larger than 512 MB? I gather you made it to the serial entry screen?
Thank you Nat ----------- Fixed by this commit:
krnl386.exe: Fix a potential leak and NULL dereference in DPMI_xrealloc. Andrew Nguyen [Sun, 18 Jul 2010 21:23:21 +0000 (16:23 -0500)]
http://source.winehq.org/git/wine.git/?a=commitdiff;h=be5d66f5d8df2fd90b6a02... -----------
http://bugs.winehq.org/show_bug.cgi?id=12874
--- Comment #23 from Andrew Nguyen arethusa26@gmail.com 2010-08-05 20:49:43 --- (In reply to comment #22)
Should I open a bug report for the second issue you mention? How do I hack DPMI_xalloc to reject allocation sizes larger than 512 MB? I gather you made it to the serial entry screen?
Yes, please open a new bug report. As for hacking DPMI_xalloc, just add a check that makes DPMI_xalloc bail out when the requested size is above a certain specified value. The Master Genealogist interface didn't appear terribly usable, so I wasn't sure what I was looking at.
http://bugs.winehq.org/show_bug.cgi?id=12874
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #24 from Alexandre Julliard julliard@winehq.org 2010-08-20 12:38:19 --- Closing bugs fixed in 1.3.1.
http://bugs.winehq.org/show_bug.cgi?id=12874
Anastasius Focht focht@gmx.net changed:
What |Removed |Added ---------------------------------------------------------------------------- Fixed by SHA1| |be5d66f5d8df2fd90b6a0274cb1 | |2fe69a0d7385a CC| |focht@gmx.net