http://bugs.winehq.org/show_bug.cgi?id=31689
Bug #: 31689 Summary: Recent devel versions of WINE report "Not enough memory" on open or save document attempts in MS Office 2003 Product: Wine Version: 1.5.12 Platform: x86-64 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: kernel32 AssignedTo: wine-bugs@winehq.org ReportedBy: olelukoie@mail.ru Classification: Unclassified
When using recent versions of wine (beginning from version 1.5.10 at least) with MS Office 2003 on Linux I encounter the problem with opening and saving documents.
The problem is that on any attempt to open or save any document in Word 2003 or Excel 2003 I see an error message stating "Not enough memory" and suggesting to save document. (I've tested only Word and Excel but the same problem could appear in othe Office 2003 apps).
I've made git bisec and it gave me the following:
4d7f510bef0891004bf5f1570148eeaf9ef73ee0 is the first bad commit commit 4d7f510bef0891004bf5f1570148eeaf9ef73ee0 Author: Alexandre Goujon ale.goujon@gmail.com Date: Sat Jul 28 16:14:15 2012 +0200
kernel32: Reimplement GetLogicalDrives using Nt{Open, Query}DirectoryObject.
:040000 040000 9beeb146ebfe6c1ed40103803560b9c06c706f46 e2eebf8851f2cc03521d45a030e8104ada32adc7 M dlls
I've tried to revert this commit and it fixed the problem.
I use Mageia 2 x86_64 Linux.
http://bugs.winehq.org/show_bug.cgi?id=31689
Henri Verbeet hverbeet@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |regression Regression SHA1| |4d7f510bef0891004bf5f157014 | |8eeaf9ef73ee0
http://bugs.winehq.org/show_bug.cgi?id=31689
--- Comment #1 from Rosanne DiMesio dimesio@earthlink.net 2012-09-12 14:00:41 CDT --- I can't reproduce this in 1.5.12 in either Word or Excel 2003.
Is this in a clean wineprefix (no overrides, nothing else installed)?
http://bugs.winehq.org/show_bug.cgi?id=31689
--- Comment #2 from olelukoie olelukoie@mail.ru 2012-09-12 14:12:23 CDT --- Hmm. It's not a clean prefix but it's a prefix with only MS Office installed and it was created using wine 1.4.1. It doesn't have any overrides.
http://bugs.winehq.org/show_bug.cgi?id=31689
Dan Kegel dank@kegel.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |dank@kegel.com
--- Comment #3 from Dan Kegel dank@kegel.com 2012-09-12 19:15:40 CDT --- Does it happen even with a totally fresh empty document?
Also, if you reinstall into a clean wineprefix with current wine, does it still happen?
http://bugs.winehq.org/show_bug.cgi?id=31689
--- Comment #4 from olelukoie olelukoie@mail.ru 2012-09-13 12:56:41 CDT --- Well, I've made some additional testing and found that this bug is a "derivative" bug. The real problem is a udisks2 support that seem to not correctly work with udisks2 1.93.0 that comes with Mageia 2.
Originally the failure occurs in udisks2_add_device function (dlls/mountmgr.sys/dbus.c file, line 450) when trying to get value of "IdUUID" message for the first block device (type 'a' of this message is treated as a container but wine uses dbus_message_iter_get_basic() to get its value and it fails) so all work with disks becomes broken. It's strange that this breakage do not affect older implementation of GetLogicalDrives function in kernel32.dll and only breaks the new one.
After I've removed all traces of udisks2 from my system the bug disappeared.
Are there any specific version requirements to correctly support udisks2? If yes then is it possible to make udisks2 support switchable through configure script to enable backports for systems with older versions of udisks2?
http://bugs.winehq.org/show_bug.cgi?id=31689
--- Comment #5 from Alexandre Julliard julliard@winehq.org 2012-09-13 13:19:59 CDT --- Does it work if you use udisks2_string_from_array to parse that uuid string, or is it returning something even stranger?
http://bugs.winehq.org/show_bug.cgi?id=31689
--- Comment #6 from olelukoie olelukoie@mail.ru 2012-09-13 13:43:29 CDT --- (In reply to comment #5)
Does it work if you use udisks2_string_from_array to parse that uuid string, or is it returning something even stranger?
I do not understand what you mean. The call to p_dbus_message_iter_get_basic( &variant, &uuid_str ) at line 450 of dbus.c causes an error somewhere inside dbus (in console it writes something like "type 97 is not a basic type") and doesn't return. But wine itself continues to work. If I try to run winecfg and go to Devices tab I see "unable to connect to mount manager" text and no controls on the tab.
To determine that "type 97" (= type 'a') is treated as a container I've added some additional calls to dbus_type_is_*(&variant) before this call and printed the results (* means 'basic', 'container' and 'fixed'). dbus_type_is_basic and dbus_type_is_fixed returned 0 and dbus_type_is_container returned 1.
http://bugs.winehq.org/show_bug.cgi?id=31689
--- Comment #7 from Alexandre Julliard julliard@winehq.org 2012-09-13 13:59:03 CDT --- I mean if you replace that p_dbus_message_iter_get_basic call by something like:
uuid_str = udisks2_string_from_array(&variant);
http://bugs.winehq.org/show_bug.cgi?id=31689
--- Comment #8 from olelukoie olelukoie@mail.ru 2012-09-13 14:02:13 CDT --- O, well I'll try and report.
http://bugs.winehq.org/show_bug.cgi?id=31689
--- Comment #9 from olelukoie olelukoie@mail.ru 2012-09-13 14:16:32 CDT --- Seems like this change fixed the problem (both with MS Office and winecfg) :)
http://bugs.winehq.org/show_bug.cgi?id=31689
--- Comment #10 from Alexandre Julliard julliard@winehq.org 2012-09-13 14:31:07 CDT --- Created attachment 41663 --> http://bugs.winehq.org/attachment.cgi?id=41663 Handle uuid as array
Here's a patch that should handle both formats, does this work for you too?
http://bugs.winehq.org/show_bug.cgi?id=31689
--- Comment #11 from olelukoie olelukoie@mail.ru 2012-09-13 15:20:30 CDT --- Yes it works for me.
http://bugs.winehq.org/show_bug.cgi?id=31689
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Fixed by SHA1| |ed8d665f8c04ee70fdc8bb76967 | |d2ad3526bc691 Status|UNCONFIRMED |RESOLVED Resolution| |FIXED
--- Comment #12 from Alexandre Julliard julliard@winehq.org 2012-09-14 10:58:13 CDT --- Thanks, fixed by ed8d665f8c04ee70fdc8bb76967d2ad3526bc691.
http://bugs.winehq.org/show_bug.cgi?id=31689
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #13 from Alexandre Julliard julliard@winehq.org 2012-09-14 13:33:54 CDT --- Closing bugs fixed in 1.5.13.