November 2014 - Wine-Bugs - list.winehq.org

[Bug 34088] New: Symantec Antivirus 10.1 (64-bit) installer very slow
by wine-bugs＠winehq.org Feb. 23, 2015

Feb. 23, 2015

http://bugs.winehq.org/show_bug.cgi?id=34088 Bug #: 34088 Summary: Symantec Antivirus 10.1 (64-bit) installer very slow Product: Wine Version: 1.6-rc4 Platform: x86-64 OS/Version: Linux Status: NEW Keywords: Installer, performance Severity: trivial Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: kennybobs(a)o2.co.uk Classification: Unclassified Created attachment 45324 --> http://bugs.winehq.org/attachment.cgi?id=45324 wine-1.6-rc4 console output Towards the end of the installation for Symantec Antivirus 10.1 for Win64 the installer appears to hang. Left long enough it does complete. Lots of these: err:ntdll:RtlpWaitForCriticalSection section 0x1320060 "heap.c: HEAP.critSection" wait timed out in thread 0072, blocked by 0071, retrying (60 sec) err:ntdll:RtlpWaitForCriticalSection section 0x1320060 "heap.c: HEAP.critSection" wait timed out in thread 0075, blocked by 0071, retrying (60 sec) err:ntdll:RtlpWaitForCriticalSection section 0x1320060 "heap.c: HEAP.critSection" wait timed out in thread 0076, blocked by 0071, retrying (60 sec) err:ntdll:RtlpWaitForCriticalSection section 0x1320060 "heap.c: HEAP.critSection" wait timed out in thread 0077, blocked by 0071, retrying (60 sec) err:ntdll:RtlpWaitForCriticalSection section 0x1320060 "heap.c: HEAP.critSection" wait timed out in thread 0078, blocked by 0071, retrying (60 sec) Running with a +heap trace causes an assertion failure and the installer crashes. Running with an +msi,+msiexec,+msidb trace and the installer completes as normal but much faster. Workarounds: Bug 22009, Bug 34082 and Bug 34083. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 2

[Bug 34082] New: Symantec Antivirus 10.x installer fails in custom action AddServiceGroup.18376EA1_A7B1_41ED_A75C_CD6FAB0434AA
by wine-bugs＠winehq.org Feb. 23, 2015

Feb. 23, 2015

http://bugs.winehq.org/show_bug.cgi?id=34082 Bug #: 34082 Summary: Symantec Antivirus 10.x installer fails in custom action AddServiceGroup.18376EA1_A7B1_41ED_A75C_CD6FAB0434AA Product: Wine Version: 1.6-rc4 Platform: x86-64 OS/Version: Linux Status: NEW Keywords: Installer Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: kennybobs(a)o2.co.uk Classification: Unclassified Created attachment 45311 --> http://bugs.winehq.org/attachment.cgi?id=45311 wine-1.6-rc4 console output Working around Bug 22009 the installer for Symantec Antivirus 10.1 (32-bit) fails in a custom action AddServiceGroup.18376EA1_A7B1_41ED_A75C_CD6FAB0434AA. fixme:service:svcctl_ChangeServiceConfig2W SERVICE_CONFIG_FAILURE_ACTIONS not implemented: period 86400 msg (null) cmd (null) err:msi:ITERATE_Actions Execution halted, action L"AddServiceGroup.18376EA1_A7B1_41ED_A75C_CD6FAB0434AA" returned 1603 Removing the custom action from the .msi allows the installer to continue. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 4

[Bug 34071] New: Create an application for verifying embedded digital signatures
by wine-bugs＠winehq.org Feb. 23, 2015

Feb. 23, 2015

http://bugs.winehq.org/show_bug.cgi?id=34071 Bug #: 34071 Summary: Create an application for verifying embedded digital signatures Product: Wine Version: 1.6-rc5 Platform: x86 OS/Version: Linux Status: UNCONFIRMED Severity: enhancement Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: t.artem(a)mailcity.com Classification: Unclassified Currently there's no way to verify a digital signature of Windows applications (EXE and MSI) that you download from the Internet. That would be great to have a way to see if the file is indeed digitally signed (so there's a lesser chance that it's malware). -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 2

[Bug 33642] New: Microsoft SQL Server Management Studio Express 2005: restarting SQL server fails randomly
by wine-bugs＠winehq.org Feb. 23, 2015

Feb. 23, 2015

http://bugs.winehq.org/show_bug.cgi?id=33642 Bug #: 33642 Summary: Microsoft SQL Server Management Studio Express 2005: restarting SQL server fails randomly Product: Wine Version: 1.5.30 Platform: x86 URL: http://www.microsoft.com/download/en/details.aspx?id=8 961 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: wmi&wbemprox AssignedTo: wine-bugs(a)winehq.org ReportedBy: djelinski1(a)gmail.com Classification: Unclassified Now that Bug #33607 is out of the way, it is possible to restart SQL Server service. The application first stops the service, then checks if it is stopped, and if it is, tries to start it. Starting the service quite often throws an exception, and stack trace reported by the application points to wbemprox. >From the initial investigation it seems to me that calling StartService temporarily invalidates all existing instances of the service object. Here one thread tries to get SystemName property from the object while another thread tries to restart the service, and getting property throws an exception. Relevant pieces of +wbemprox,+seh,+tid trace: 0057:trace:wbemprox:wbem_services_GetObject 0x46cad18, L"Win32_Service.Name=\"MSSQLSERVER\"", 0x00000000, (nil), 0xe45dab0, (nil) ... 0057:trace:wbemprox:create_class_object returning iface 0x472eda0 0057:trace:wbemprox:class_object_Get ((nil)) 0x472eda0, L"__PATH", 00000000, 0xe45dc28, 0xe45dd0c, 0xe45dd08 0057:fixme:ole:Context_QueryInterface interface not implemented {51372ae0-cae7-11cf-be81-00aa00a2fa25} 0057:trace:wbemprox:class_object_Get ((nil)) 0x472eda0, L"State", 00000000, 0xe45dc74, 0x1691a54, 0x1691a58 ... 0057:trace:wbemprox:class_object_Get ((nil)) 0x472eda0, L"StartMode", 00000000, 0xe45dd14, 0x1691ec8, 0x1691ecc ... 002f:trace:wbemprox:class_object_Get ((nil)) 0x472eda0, L"SystemName", 00000000, 0x32e380, 0x1696178, 0x169617c 002f:trace:wbemprox:class_object_Get ((nil)) 0x472eda0, L"Name", 00000000, 0x32e37c, 0x16963d8, 0x16963dc ... 0057:trace:wbemprox:wbem_services_ExecMethod 0x46cad18, L"Win32_Service.Name=\"MSSQLSERVER\"", L"StartService", 00000000, (nil), (nil), 0xe45da98, (nil) 0057:trace:wbemprox:grab_table returning 0x7d0b6f90 0057:trace:wbemprox:parse_query wql_parse returned 0 002f:trace:wbemprox:class_object_Get ((nil)) 0x472eda0, L"SystemName", 00000000, 0x32e37c, 0x16965cc, 0x16965d0 002f:trace:seh:raise_exception code=c0000005 flags=0 addr=0x7d0a82c0 ip=7d0a82c0 tid=002f -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 5

[Bug 33558] New: WinEventHook receive no event notify about OBJID_CURSOR
by wine-bugs＠winehq.org Feb. 23, 2015

Feb. 23, 2015

http://bugs.winehq.org/show_bug.cgi?id=33558 Bug #: 33558 Summary: WinEventHook receive no event notify about OBJID_CURSOR Product: Wine Version: 1.5.29 Platform: x86 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: user32 AssignedTo: wine-bugs(a)winehq.org ReportedBy: orzhvs(a)gmail.com Classification: Unclassified Created attachment 44418 --> http://bugs.winehq.org/attachment.cgi?id=44418 WINEDEBUG=+relay,+user32,+hook LANG='zh_CN.utf-8' wine /usr/local/vboxvm/vmshare/magnify/magnify.exe 2>&1 |grep -iE Win.*Event > out.txt There's a method to track mouse move by setting a WinEvent hook: call SetWinEventHook with eventMin=eventMax=EVENT_OBJECT_LOCATIONCHANGE(=0x800B) and provide a WinEventProc WinEventProc will be notified with event=EVENT_OBJECT_LOCATIONCHANGE in case idObject=OBJID_CURSOR we know that mouse moved, then get cursor object by AccessibleObjectFromEvent and further information like the location of the cursor. But the fact is, no event with idObject=OBJID_CURSOR is sent in wine. This is used in windows's zoom-in utility, it does not update frames with mouse move in wine. To reproduce : 1. copy 'magnify.exe' 'mag_hook.dll' 'mfc42u.dll' from system32 directory of windows, I used them from xpsp3. 2. run magnify.exe in wine 3. move mouse in and out the dialog, you will find the zoomed-in view updates almost only when mouse is inside the dialog. Note that we see frames update when mouse is inside the dialog, it's triggered from WM_MOUESMOVE, there's no business of WinEvent, but when mouse is outside there should be. I saw several SetWinEvent calls successfully in log but no event notify. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 1

[Bug 33396] New: GetSystemMetrics always reports native resolution with --with-xinerama.
by wine-bugs＠winehq.org Feb. 23, 2015

Feb. 23, 2015

http://bugs.winehq.org/show_bug.cgi?id=33396 Bug #: 33396 Summary: GetSystemMetrics always reports native resolution with --with-xinerama. Product: Wine Version: 1.5.28 Platform: x86 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: winex11.drv AssignedTo: wine-bugs(a)winehq.org ReportedBy: CFSworks(a)gmail.com Classification: Unclassified As the summary says, GetSystemMetrics(SM_C{X,Y}SCREEN) always reports the native resolution, even if an application changes the resolution, unless I configure Wine with the --without-xinerama option. This can be seen when I run D3D9 device tests, leading to test failures (see attachments). I am on the NVIDIA proprietary driver version 313.18, which may be relevant as this driver does have the "broken" XRandR 1.2. I cannot test with Nouveau because it doesn't support my GPU. Even if this is the proprietary driver's fault, I'm still opening this bug on the WineHQ Bugzilla to track its resolution here, and provide information to the NVIDIA developers if we do have to report it to them. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 6

[Bug 33339] New: AION: regularly crash during game
by wine-bugs＠winehq.org Feb. 23, 2015

Feb. 23, 2015

http://bugs.winehq.org/show_bug.cgi?id=33339 Bug #: 33339 Summary: AION: regularly crash during game Product: Wine Version: 1.5.27 Platform: x86-64 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: legluondunet(a)free.fr Classification: Unclassified Hello, AION regularly crash during game, I joined you two log files. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 15

[Bug 21573] New: Need heap overrun detection at beginning of buffers, too
by wine-bugs＠winehq.org Feb. 23, 2015

Feb. 23, 2015

http://bugs.winehq.org/show_bug.cgi?id=21573 Summary: Need heap overrun detection at beginning of buffers, too Product: Wine Version: 1.1.37 Platform: x86 OS/Version: Linux Status: NEW Severity: enhancement Priority: P2 Component: ntdll AssignedTo: wine-bugs(a)winehq.org ReportedBy: dank(a)kegel.com On Linux, valgrind gives nice warnings when you access before the first byte. It'd be nice if valgrinding Windows apps under Wine gave the same warnings. For instance, the following program should give three warnings under valgrind+wine just as it does under valgrind: #include <stdio.h> #include <stdlib.h> int badness_before_n_after ( char* p ) { return p[-1] + p[10]; } int main ( void ) { char* p = malloc(10); int who_knows = badness_before_n_after(p); // expect 2 x invalid address yelpage who_knows += p[5]; if (who_knows == 42) printf("It's 42 (!)\n"); else printf("It's not 42 (dull but unsurprising)\n"); free(p); return 0; } -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 4

[Bug 16264] New: Builtin Tahoma unreadable when small
by wine-bugs＠winehq.org Feb. 23, 2015

Feb. 23, 2015

http://bugs.winehq.org/show_bug.cgi?id=16264 Summary: Builtin Tahoma unreadable when small Product: Wine Version: 1.1.9 Platform: Other URL: http://www.linuxtechs.net/kruch/tb/forum/viewtopic.php?t =2152&postdays=0&postorder=asc&start=15 OS/Version: other Status: NEW Keywords: download, Installer Severity: normal Priority: P2 Component: fonts AssignedTo: wine-bugs(a)winehq.org ReportedBy: dank(a)kegel.com A user said that ConvWPT ran in Wine, but gave a very complicated recipe, so I tried simplifying it. Along the way, I noticed that ConvWPT's installer used teeny tiny nearly unreadable fonts; installing Tahoma fixed that (now they're just small, but readable). Recipe to reproduce: 1) Visit http://www.linuxtechs.net/kruch/tb/forum/viewtopic.php?t=2152 scroll down to message "Setup Package Part 1 / 4", and download and unpack each of the four files Setup.part1.rar, Setup.part2.rar, ... Setup.part4.rar Ignore warnings from rar about duplicate files. 2) Run the resulting setup.exe The setup program uses small Tahoma text on some screens, and that's very nearly unreadable unless you install native tahoma, e.g. with winetricks tahoma. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 5

[Bug 31385] New: Font in Delphi 3 apps change to "System" after open Report preview.
by wine-bugs＠winehq.org Feb. 23, 2015

Feb. 23, 2015

http://bugs.winehq.org/show_bug.cgi?id=31385 Bug #: 31385 Summary: Font in Delphi 3 apps change to "System" after open Report preview. Product: Wine Version: 1.5.10 Platform: x86 OS/Version: Linux Status: UNCONFIRMED Severity: minor Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: julio_lists(a)mebamutual.com.ar Classification: Unclassified Created attachment 41242 --> http://bugs.winehq.org/attachment.cgi?id=41242 Minimal Delphi app that shows the bug I'm testing an in-house app written en Delphi 3, the only really annoying glitch I have found so far happens after a QuickReport preview windows opens. The Preview and works OK, but then any label drawn in the app uses the System font. The effect is akin to bug #29758, but my logs don't include the error reported there. Instead, I guess Wine is messing somewhat within the WM_GETFONT and/or WM_SETFONT calls, yet my debugging abilities are pretty limited. WM_GETFONT calls aren't returning NULL, but decimal 504, I wonder if that WM_SETFONT takes that as an invalid handle. I'm attaching a minimal (kind of, the binary is pretty big) test case. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 2