- Wine-Bugs - list.winehq.org

[Bug 56565] New: Command and Conquer 3 Tiberium Wars installer crashes
by WineHQ Bugzilla April 15, 2024

April 15, 2024

https://bugs.winehq.org/show_bug.cgi?id=56565 Bug ID: 56565 Summary: Command and Conquer 3 Tiberium Wars installer crashes Product: Wine Version: 9.6 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: hibbsncc1701(a)gmail.com Distribution: --- Split from bug #53561. This bug is about a crash / hang in the installer for the retail release of Command and Conquer 3 Tiberium Wars when used with the recommended installation type. (The advanced installation type is unusable under Wine, see the original bug #53561.) After clicking on the next button on the installation type screen, installer appears to try and copy files. This is supposed to have a progress indicator that fills up, but it gets drawn in wine as completely full, with status "Copying new files". The UI has stopped updating and placing another window (Wine or from the host's WM) over it will cause graphical corruption. After about a minute on my machine, the UI window disappears entirely. The msiexec process however is still running in the background, and can be seen on the task bar if wine is running with a virtual desktop. It is also still visible in wine's taskmgr.exe. Where it shows two msiexec processes and one unnamed 32bit process. (Wine's taskmgr shows "*32" for the process name.) After this the process does not exit unless killed. Doing so kills both msiexec processes and the unnamed 32bit process. Looking through the registry after killing the installer, we find that it did create the needed registry keys expected by the game, in addition to installing the games files at the expected default location. The game does run in this state. So it appears that the installation was successful, despite the bug. Running the installer with WINEDEBUG="+msi" causes the installer to run correctly. I.e. It doesn't crash / hang indefinitely. Although, the UI still stops updating while the files are copied, it will recover in this case and allow the installer to run to completion. With the user able to exit the installer normally. WINEDEBUG="+msi" log attached. -- 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 3

[Bug 56560] New: There is a memory leak defect at line 1068 in the file /wine/dlls/winex11.drv/clipboard.c.
by WineHQ Bugzilla April 15, 2024

April 15, 2024

https://bugs.winehq.org/show_bug.cgi?id=56560 Bug ID: 56560 Summary: There is a memory leak defect at line 1068 in the file /wine/dlls/winex11.drv/clipboard.c. Product: Wine Version: 9.6 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: major Priority: P2 Component: kernel32 Assignee: wine-bugs(a)winehq.org Reporter: lumingying(a)bupt.edu.cn Distribution: --- Trigger Path for the Defect: 1.A pointer variable named "path" is defined at line 1056 in the file /wine/dlls/winex11.drv/clipboard.c. This pointer allocates a dynamic memory area through the function "get_dos_file_name" at line 1060, as shown in the following diagram: https://github.com/LuMingYinDetect/wine_defects/blob/main/wine_1.png 2.The function "get_dos_file_name" allocates a dynamic memory area and returns it. Specifically, at line 587, a pointer variable named "ret" is defined. At line 589, memory space is allocated for the "ret" variable using the function "malloc," and at line 602, this variable is returned, as shown in the following diagram: https://github.com/LuMingYinDetect/wine_defects/blob/main/wine_2.png 3.After the execution of the "get_dos_file_name" function, when the conditional statement at line 1064 returns false, it indicates that the dynamic memory area pointed to by the "path" variable is successfully allocated. However, when the conditional statement at line 1068 returns true, the program exits without releasing the memory. In this process, the dynamic memory area pointed to by the "path" pointer is not freed, resulting in a memory leak defect, as shown in the following diagram: https://github.com/LuMingYinDetect/wine_defects/blob/main/wine_3.png -- 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 56564] New: There is a memory leak defect at line 2212 in the file /wine/dlls/comctl32/imagelist.c.
by WineHQ Bugzilla April 15, 2024

April 15, 2024

https://bugs.winehq.org/show_bug.cgi?id=56564 Bug ID: 56564 Summary: There is a memory leak defect at line 2212 in the file /wine/dlls/comctl32/imagelist.c. Product: Wine Version: 9.6 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: major Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: lumingying(a)bupt.edu.cn Distribution: --- In the file /wine/dlls/comctl32/imagelist.c, a variable named himl is defined at line 2190. This pointer allocates a dynamic memory area through the function ImageList_Create at line 2205. When the if statement at line 2209 returns true, the program will return at line 2212. During this process, the dynamic memory area pointed to by himl is neither used nor released, resulting in a memory leak, as illustrated in the diagram below: https://github.com/LuMingYinDetect/wine_defects/blob/main/wine_7.png -- 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 56563] New: There is a memory leak defect at line 1106 in the file /wine/libs/vkd3d/libs/vkd3d-shader/dxil.c.
by WineHQ Bugzilla April 15, 2024

April 15, 2024

https://bugs.winehq.org/show_bug.cgi?id=56563 Bug ID: 56563 Summary: There is a memory leak defect at line 1106 in the file /wine/libs/vkd3d/libs/vkd3d-shader/dxil.c. Product: Wine Version: unspecified Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: major Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: lumingying(a)bupt.edu.cn Distribution: --- In the file /wine/libs/vkd3d/libs/vkd3d-shader/dxil.c, a pointer variable named global_abbrev is defined at line 1085. This variable allocates a dynamic memory area through the function vkd3d_malloc at line 1091. When the if statement at line 1103 returns true, the function will return at line 1106. During this process, the dynamic memory area pointed to by global_abbrev is neither assigned to the function parameter sm6 as shown in line 1115, nor released, thus resulting in a memory leak defect, as illustrated in the diagram below: https://github.com/LuMingYinDetect/wine_defects/blob/main/wine_6.png -- 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 56561] New: There is a memory leak defect at line 494 in the file /wine/dlls/dmloader/container.c.
by WineHQ Bugzilla April 15, 2024

April 15, 2024

https://bugs.winehq.org/show_bug.cgi?id=56561 Bug ID: 56561 Summary: There is a memory leak defect at line 494 in the file /wine/dlls/dmloader/container.c. Product: Wine Version: 9.6 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: major Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: lumingying(a)bupt.edu.cn Distribution: --- Trigger Path for the Defect: 1.A variable named "pNewEntry" is defined at line 389 in the file /wine/dlls/dmloader/container.c. This variable allocates a dynamic memory area using the calloc function at line 391. When the switch-case block at line 492 is satisfied, the program returns at line 494. In this process, the dynamic memory area pointed to by the "pNewEntry" pointer is neither used nor freed like at line 547, resulting in a memory leak defect, as illustrated in the diagram below: https://github.com/LuMingYinDetect/wine_defects/blob/main/wine_4.png -- 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 53561] New: Command and Conquer 3 Kane's Wrath Installer fails during component selection.
by WineHQ Bugzilla April 15, 2024

April 15, 2024

https://bugs.winehq.org/show_bug.cgi?id=53561 Bug ID: 53561 Summary: Command and Conquer 3 Kane's Wrath Installer fails during component selection. Product: Wine Version: 7.15 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: hibbsncc1701(a)gmail.com Distribution: --- Created attachment 72925 --> https://bugs.winehq.org/attachment.cgi?id=72925 +msi log of CNC3 Kane's Wrath install attempt. When installing the game in Advanced install type mode (to change the destination install directory) the installer fails to continue past the component selection screen, displaying the error: "At least one language pack must be selected" The default component selection includes the language pack, and even deselecting and re-selecting it does not fix the error. Making the installer unable to proceed. This can be worked around by using the "Recommended" install type, but that means the user must manually change the game's installation directory after the game is installed. It should also be noted that changing the install directory causes the installer to display "_BrowseProperty" as the selected path after closing the folder selection dialog. The original game Command and Conquer 3 also has the same installer bug. An +msi log is attached. -- 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 56562] New: There is a memory leak defect at line 733 in the file /wine/dlls/user32/dde_client.c.
by WineHQ Bugzilla April 15, 2024

April 15, 2024

https://bugs.winehq.org/show_bug.cgi?id=56562 Bug ID: 56562 Summary: There is a memory leak defect at line 733 in the file /wine/dlls/user32/dde_client.c. Product: Wine Version: 9.6 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: lumingying(a)bupt.edu.cn Distribution: --- In the file /wine/dlls/user32/dde_client.c, a pointer variable named pXAct is defined at line 702. This variable allocates a dynamic memory area through the function WDML_AllocTransaction at line 712. When the if statement at line 713 returns false, it means that the dynamic memory area pointed to by pXAct has been successfully allocated. When the if statement at line 730 returns true, the program will return at line 733. During this process, the dynamic memory area pointed to by pXAct is not used, resulting in a memory leak, as shown in the following diagram: https://github.com/LuMingYinDetect/wine_defects/blob/main/wine_5.png -- 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 0

[Bug 34003] New: File save problem with Visual Cadd 6.0
by wine-bugs＠winehq.org April 14, 2024

April 14, 2024

http://bugs.winehq.org/show_bug.cgi?id=34003 Bug #: 34003 Summary: File save problem with Visual Cadd 6.0 Product: Wine Version: 1.5.31 Platform: x86-64 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: purevw(a)wtxs.net Classification: Unclassified Created attachment 45182 --> http://bugs.winehq.org/attachment.cgi?id=45182 A binary crash report generated by Visual Cadd 6.0 When trying to do a normal file save (with thumbnail written into the file) in Visual Cadd 6.0 the program crashes. The crash only occurs when running under wine, as opposed to running in Windows. I found a workaround that seems to point to the problem. I found that if I clear the check-box for "Show Preview" in the file save dialog box, the file saves normally. Tri-tools.com has been very cooperative in offering help, rather than simply saying "We don't support Linux" as so many developers do when asked for help. Here is a paste of their latest response to information that I had given them: >> If a report file is generated, it will be stored in your Temp folder. The >>file will start with VC and have an extension of .vccrd. >> >>The crash info that you already submitted was sufficient to find the cause, >>though. Wine is failing to correctly create the bitmap and VCADD does not >>handle the failure properly. Since the bitmap is not being created anyhow, >>turning off Preview is the best solution for now. We'll work on a fix for the >>crash, but we won't be able to create the preview unless Wine cooperates. I'll attache the created error report, but it appears to be a binary file. I am able to open it with Kdevelop 4 but have no idea what to do with the information. Perhaps it will give more detailed information as to what is happening. -- 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.

2 22

[Bug 56558] New: cabinet:FCIAddFile: Missing support for LZX compression
by WineHQ Bugzilla April 14, 2024

April 14, 2024

https://bugs.winehq.org/show_bug.cgi?id=56558 Bug ID: 56558 Summary: cabinet:FCIAddFile: Missing support for LZX compression Product: Wine Version: 9.6 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: cabinet Assignee: wine-bugs(a)winehq.org Reporter: winehq(a)jkroepke.de Distribution: --- The native windows implementation of cabinet:FCIAddFile supports LZX compression in addition to MSZip. I'm trying to run https://wixtoolset.org/docs/about/ under WINE to create msi installer, however the creation fails with default settings 0144:fixme:cabinet:FCIAddFile compression 1203 not supported, defaulting to none 0144:fixme:cabinet:FCIAddFile compression 1203 not supported, defaulting to none 0144:fixme:cabinet:FCIAddFile compression 1203 not supported, defaulting to none 0144:fixme:cabinet:FCIAddFile compression 1203 not supported, defaulting to none 0144:fixme:cabinet:FCIAddFile compression 1203 not supported, defaulting to none 0144:fixme:kernelbase:AppPolicyGetProcessTerminationMethod FFFFFFFA, 0062FEB8 wix.exe : error WIX0001: WixToolset.Core.Native.WixNativeException: wixnative.exe failed with error code: -2147467259 - Invalid handle. Output: #cab1.cab #cab1.cab filPltju7VHTe0axpRG.LFcHosyVd8 Error 0x80070006: failed to compress cabinet: C:\users\wix\Temp\sadphplt.t0n\#cab1.cab ---> System.ComponentModel.Win32Exception (6): Invalid handle. --- End of inner exception stack trace --- at WixToolset.Core.Native.WixNativeExe.Run() in D:\a\wix\wix\src\wix\WixToolset.Core.Native\WixNativeExe.cs:line 81 at WixToolset.Core.Native.Cabinet.Compress(IEnumerable`1 files, CompressionLevel compressionLevel, Int32 maxSize, Int32 maxThresh) in D:\a\wix\wix\src\wix\WixToolset.Core.Native\Cabinet.cs:line 60 at WixToolset.Core.WindowsInstaller.Bind.CabinetBuilder.CreateCabinet(CabinetWorkItem cabinetWorkItem) in D:\a\wix\wix\src\wix\WixToolset.Core.WindowsInstaller\Bind\CabinetBuilder.cs:line 192 at WixToolset.Core.WindowsInstaller.Bind.CabinetBuilder.ProcessWorkItems() in D:\a\wix\wix\src\wix\WixToolset.Core.WindowsInstaller\Bind\CabinetBuilder.cs:line 123 Changing the compression to mszip on the wixtoolset project resolve the issue. -- 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 0

[Bug 24655] New: BiDi support broken for writing Arabic
by wine-bugs＠winehq.org April 13, 2024

April 13, 2024

http://bugs.winehq.org/show_bug.cgi?id=24655 Summary: BiDi support broken for writing Arabic Product: Wine Version: 1.3.4 Platform: x86 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: usp10 AssignedTo: wine-bugs(a)winehq.org ReportedBy: fahad.alsaidi(a)gmail.com The next step to improve Arabic support in Wine after fixing bug 24608 is fixing writing Arabic. Let me explain more, Now when you open Arabic text, Notepad will display it good enough to read it although it doesn't support full RTL. The problem is when you try to write something in Arabic. The new text will not be displayed correctly. It is displayed like that wine doesn't support Arabic at all. The interesting thing is when you select the text, the text will be displayed right. -- 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.

2 14