wine-bugs November 2024

wine-bugs@winehq.org

1 participants
780 discussions

[Bug 45208] New: Microsoft Office 2010 installer crashes at around 50 % mark
by wine-bugs＠winehq.org 05 Nov '24

05 Nov '24

https://bugs.winehq.org/show_bug.cgi?id=45208 Bug ID: 45208 Summary: Microsoft Office 2010 installer crashes at around 50 % mark Product: Wine Version: 3.8 Hardware: x86 OS: Linux Status: NEW Severity: normal Priority: P2 Component: msi Assignee: wine-bugs(a)winehq.org Reporter: bunglehead(a)gmail.com Distribution: --- The crash happens in custom action process I think, installer does not recover from that. I'm using 32bit prefix, default Win7 version, and Home & Business product edition. This looks like a regression, I'm going to try to find what broke it. -- 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 49

[Bug 56843] New: wineandroid: fixing output on recent android versions.
by WineHQ Bugzilla 05 Nov '24

05 Nov '24

https://bugs.winehq.org/show_bug.cgi?id=56843 Bug ID: 56843 Summary: wineandroid: fixing output on recent android versions. Product: Wine Version: unspecified Hardware: aarch64 OS: Android Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: twaikyont(a)gmail.com Currently wineandroid relies on Gralloc V1 which is not an option on recent android versions. There is other acceptable solution. You can pass Surface through Binder API so you will get ANativeWindow (or at least Surface which can be converted to ANativeWindow) directly from Activity instead of creating it with broken `alloc_win_data`. Binder API is not available in regular processes, but you can use it in Java. You can start java code from your Wine apk installed to system like ``` export CLASSPATH=<path to apk> export LD_PRELOAD=<path to apk>!/lib/<ABI>/libwine.so # wher ABI is something you can get with `Build.SUPPORTED_ABIS[0]` /system/bin/app_process / your.class.name :0 ``` And JVM will start the `static void main(String[] args);` from `your.class.name`. Getting path to apk will be like ``` /** * Get the apk path of this application. * @param context any context (e.g. an Activity or a Service) * @return full apk file path, or null if an exception happened (it should not happen) */ public static String getApkName(Context context) { String packageName = context.getPackageName(); PackageManager pm = context.getPackageManager(); try { return pm.getApplicationInfo(packageName, 0).publicSourceDir; } catch (Throwable x) { } return null; } ``` `main` function can be fully native so you can simply pass control to native code right after starting process. Of course wine's main executable should be built as a library to let JVM load it via `System.loadLibrary` C version of `main` can not use Java's `args` so it should be converted like ``` JNIEXPORT jboolean JNICALL Java_wine_CmdEntryPoint_start(JNIEnv *env, __unused jclass cls, jobjectArray args) { pthread_t t; // execv's argv array is a bit incompatible with Java's String[], so we do some converting here... int argc = (*env)->GetArrayLength(env, args) + 1; // Leading executable path char** argv = (char**) calloc(argc, sizeof(char*)); argv[0] = (char*) "wine"; for(int i=1; i<argc; i++) { jstring js = (jstring)((*env)->GetObjectArrayElement(env, args, i - 1)); const char *pjc = (*env)->GetStringUTFChars(env, js, JNI_FALSE); argv[i] = (char *) calloc(strlen(pjc) + 1, sizeof(char)); //Extra char for the terminating NULL strcpy((char *) argv[i], pjc); (*env)->ReleaseStringUTFChars(env, js, pjc); } return main(argc, argv); } ``` After this point you will be able to use java code and java Android APIs. Next step is obtaining Binder connection to your activity. It is not very straightforward but it is possible. Full code will be pretty much long so I will simply put links to my project which works on Android 8+. Pay attention to functions sendBroadcast and createContext (createContext is used in static block in the end of file). It requires exposing some private Android APIs, but it is not restricted if you are using app_process this way. https://github.com/termux/termux-x11/blob/master/app/src/main/java/com/term… Make sure you are creating thread with performing `Looper.prepareMainLooper();` and `Looper.loop()` to make sure Binder background requests will work fine in static block (to make sure it will work before JVM invokes `main` function, just for case it is `native`). Next step will be acquiring Surface from Wine's activity. Of course you can create AIDL with single function like `Surface getSurface(int hwnd);` and obtain Surface directly from `alloc_win_data` or simillar function via JNI. But in my cases it did not work. I mean sending request from my process to Activity which should respond with Surface object did not work. But I succeeded passing Surface by sending requst from Activity to my process containing Surface in arguments like `void pushSurface(in Surface surface, int hwnd);`. Binder request from Activity to your process will work in different thread asyncronously so `alloc_win_data` should send surface request to activity and wait for Binder response or timeout. I hope I explained everything with enough details. Thank you for your hard work. -- 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 52386] New: err:module:LdrInitializeThunk "libclang_rt.asan_dynamic-i386.dll" failed to initialize when using clang address sanitizer
by WineHQ Bugzilla 05 Nov '24

05 Nov '24

https://bugs.winehq.org/show_bug.cgi?id=52386 Bug ID: 52386 Summary: err:module:LdrInitializeThunk "libclang_rt.asan_dynamic-i386.dll" failed to initialize when using clang address sanitizer Product: Wine Version: 5.0.5 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: ralf.habacker(a)freenet.de Distribution: --- Created attachment 71603 --> https://bugs.winehq.org/attachment.cgi?id=71603 testcase binary This test case $ cat use-after-free.cc int main(int argc, char **argv) { int *array = new int[100]; delete [] array; return array[argc]; // BOOM } compiled with $ i686-w64-mingw32-clang use-after-free.cc -fsanitize=address -g -o use-after-free-i386.exe makes wine fail with the error message in the title. The compiler was downloaded from https://github.com/mstorsjo/llvm-mingw/releases (https://github.com/mstorsjo/llvm-mingw/releases/download/20211002/llvm-ming…) -- 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 11

[Bug 57389] New: x86_64-linux-android30.no member named 'mc_context' in 'struct unw_context_t'
by WineHQ Bugzilla 05 Nov '24

05 Nov '24

https://bugs.winehq.org/show_bug.cgi?id=57389 Bug ID: 57389 Summary: x86_64-linux-android30.no member named 'mc_context' in 'struct unw_context_t' Product: Wine Version: 9.20 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: euloanty(a)live.com Distribution: --- here is the error message -- 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 26888] New: Wine limits file names to 255 bytes even on NTFS
by wine-bugs＠winehq.org 04 Nov '24

04 Nov '24

http://bugs.winehq.org/show_bug.cgi?id=26888 Summary: Wine limits file names to 255 bytes even on NTFS Product: Wine Version: unspecified Platform: x86 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: jjudin+wine(a)iki.fi Wine limits file names to 255 bytes even when underlying file system permits longer file names (like NTFS) when creating a file. This is an issue with programs that can create file names that are something like 140 characters but if one character takes 2 or 3 bytes, this 255 byte filename limit will be evident when same files that can be created in Windows can't be created with Wine. NTFS supports 255 UTF-16 code units that can easily result in longer file names than 255 bytes, especially for users of Japanese systems. You can test this by creating a NTFS file system and some files with long names on it: truncate -s 2g /tmp/ntfsfs losetup /dev/loop7 /tmp/ntfsfs mkfs.ntfs -f /dev/loop7 mkdir /tmp/ntfsfs-mnt mount -t ntfs-3g /dev/loop7 /tmp/ntfsfs-mnt # Do this or whatever enables regular user to access this file system. chmod 777 /tmp/ntfsfs-mnt Then you can try to create long files outside of Wine: # 257 UTF-8 bytes, 129 characters echo foo > /tmp/ntfsfs-mnt/ääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääa # 325 UTF-8 bytes, 109 characters echo foo > /tmp/ntfsfs-mnt/亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜a And inside Wine: wine cmd # 257 UTF-8 bytes, 129 characters echo foo > /tmp/ntfsfs-mnt/ääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääääb # 325 UTF-8 bytes, 109 characters echo foo > /tmp/ntfsfs-mnt/亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜亜b Remember to have UTF-8 support enabled in your system. -- 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 19

[Bug 56739] New: Alchemy is rotated 90°
by WineHQ Bugzilla 03 Nov '24

03 Nov '24

https://bugs.winehq.org/show_bug.cgi?id=56739 Bug ID: 56739 Summary: Alchemy is rotated 90° Product: Wine Version: 9.8 Hardware: x86-64 URL: https://alchemy.software.informer.com/download/ OS: Linux Status: NEW Keywords: download, regression Severity: minor Priority: P2 Component: winex11.drv Assignee: wine-bugs(a)winehq.org Reporter: imwellcushtymelike(a)gmail.com CC: zzhang(a)codeweavers.com Regression SHA1: 4a330b29212402b9700828be82939112bd11a786 Distribution: Ubuntu Created attachment 76518 --> https://bugs.winehq.org/attachment.cgi?id=76518 Screenshot The game Alchemy is rotated 90°. A bisect points to: 4a330b29212402b9700828be82939112bd11a786 is the first bad commit commit 4a330b29212402b9700828be82939112bd11a786 Author: Zhiyi Zhang Date: Tue Sep 22 15:02:24 2020 +0800 winex11.drv: Support display orientations for XRandR 1.4 display settings handler. Running in an emulated desktop works around this. https://alchemy.software.informer.com/download/ $ sha1sum alchemysetup-en.exe 2fa2bd427e79b7fb585d090891ae368b0cea0156 alchemysetup-en.exe $ xrandr --version xrandr program version 1.5.1 Server reports RandR version 1.6 -- 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 57343] New: Unhandled exception: page fault on read access to 0x0000000000000020 in 64-bit code (0x00000140055d76).
by WineHQ Bugzilla 03 Nov '24

03 Nov '24

https://bugs.winehq.org/show_bug.cgi?id=57343 Bug ID: 57343 Summary: Unhandled exception: page fault on read access to 0x0000000000000020 in 64-bit code (0x00000140055d76). Product: Wine Version: 9.17 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: aversa(a)email.arizona.edu Distribution: Slackware Halfway through installing jre-8u431-windows-x64.exe, I get this crash. -- 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 57380] New: winegcc unable to set image-base when building an app
by WineHQ Bugzilla 03 Nov '24

03 Nov '24

https://bugs.winehq.org/show_bug.cgi?id=57380 Bug ID: 57380 Summary: winegcc unable to set image-base when building an app Product: Wine Version: unspecified Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: tools Assignee: wine-bugs(a)winehq.org Reporter: alastair.jones43(a)gmail.com Distribution: --- Using winegcc. The -Wl,--image-base doesn't seem to have any effect on the compiled code, any attempt to use GetModuleHandleExA with GET_MODULE_HANDLE_EX_FLAG_FROM_ADDRESS or GetModuleHandleA with NULL results in a completely different address than was specified. The behavior works as expected using mingw64. -- 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 57376] New: Permission denied vs No such file or directory when opening folders with trailing backslash
by WineHQ Bugzilla 03 Nov '24

03 Nov '24

https://bugs.winehq.org/show_bug.cgi?id=57376 Bug ID: 57376 Summary: Permission denied vs No such file or directory when opening folders with trailing backslash Product: Wine Version: 9.20 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: alex+winehq(a)cogarr.net Distribution: --- Created attachment 77346 --> https://bugs.winehq.org/attachment.cgi?id=77346 A zip file containing a minimal example.c, and log files wine.txt and win.txt There's a subtle difference when calling fopen(3) between Wine configured for win10, and real Windows 10. When attempting to open a folder (e.g. AppData), wine correctly sets errno to 13 (Permission denied), but when opening a folder with a trailing backslash(e.g. AppData\), Windows sets errno to 2 (No such file or directory) while Wine sets it to 13 (Permission denied). The attached example.c shows this behavior. I compile it with mingw32-x86_64-gcc and run it through cmd.exe with my current directory set to C:\users\user on Wine and get wine.txt, then run it on Windows 10, and get win.txt (COPY a.exe C:\users\user && C: && cd users\user && a.exe > win[e].txt) This behavior exists on current Wine 9.20 and has existed since at least Wine 9.0 . I have tried the winecfg versions: win10, win11, win8.1, vista, and all seem to work the same. My actual use case is running the Luarocks package manager(https://github.com/luarocks/luarocks) compiled for Windows on Wine for testing. Luarocks uses this behavior to check if something is a folder (https://github.com/luarocks/luarocks/blob/master/src/luarocks/fs/win32.lua#…). -- 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 57385] New: EA_Desktop - IGOProxy32 - Unhandled page fault
by WineHQ Bugzilla 03 Nov '24

03 Nov '24

https://bugs.winehq.org/show_bug.cgi?id=57385 Bug ID: 57385 Summary: EA_Desktop - IGOProxy32 - Unhandled page fault Product: Wine Version: 9.20 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: arusanu.bu(a)gmail.com Distribution: --- Created attachment 77354 --> https://bugs.winehq.org/attachment.cgi?id=77354 backtrace IGOProxy32 When "EA Desktop" starts after login, the wine debugger pops up with a backtrace of crashed IGOProxy32.exe. While this doesn't give me any trouble installing or running my EA games, I suspect that this crash hinders the installation of an EA game (Dragon Age Inquisition) that I got through Epic Games. Every time I try to install the EA game from the Epic Game application, IGOProxy32 crashes, and the game installation shows as defunct. I've tried various wine versions starting with 9.0, and IGOProxy32 always crashes. Attached is the backtrace of the IGOProxy32 crash. -- 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

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

wine-bugs November 2024