wine-bugs

wine-bugs@winehq.org

1 participants
158776 discussions

[Bug 50495] New: the WS_EX_LAYERED set SetWindowLong GWL_EXSTYLE , Window show white rather than transparency.
by WineHQ Bugzilla 14 Jan '21

14 Jan '21

https://bugs.winehq.org/show_bug.cgi?id=50495 Bug ID: 50495 Summary: the WS_EX_LAYERED set SetWindowLong GWL_EXSTYLE , Window show white rather than transparency. Product: WineHQ Bugzilla Version: unspecified Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: major Priority: P2 Component: bugzilla-unknown Assignee: wine-bugs(a)winehq.org Reporter: 490165508(a)qq.com CC: austinenglish(a)gmail.com Distribution: --- Created attachment 69150 --> https://bugs.winehq.org/attachment.cgi?id=69150 This program will create a transparent full screen window. When I use the following code to create a transparent full screen window. ``` HWND hWnd; hWnd = CreateWindowEx(WS_EX_APPWINDOW, "WindowClass", "Demo", WS_VISIBLE| WS_MAXIMIZE, CW_USEDEFAULT, 0, CW_USEDEFAULT, 0, NULL, NULL, hInstance, NULL); SetWindowLong(hWnd, GWL_STYLE, GetWindowLong(hWnd, GWL_STYLE) & (~WS_DLGFRAME)); RECT rect; GetWindowRect(hWnd, &rect); SetWindowPos(hWnd, HWND_TOPMOST, rect.left, rect.top, rect.right - rect.left, rect.bottom - rect.top, SWP_DRAWFRAME); LONG ret = GetWindowLong(hWnd, GWL_EXSTYLE); ret = ret | WS_EX_LAYERED; SetWindowLong(hWnd, GWL_EXSTYLE, ret); ShowWindow(hWnd, SW_SHOWMAXIMIZED); UpdateWindow(hWnd); ``` In Windows this program display window is a full screen transparent window，But running this little program with Wine on Linux displays a full-screen white window. This problem occurred. I tried to modify the SetWindowLong API function implementation, but it didn't change much. I hope the bosses can help me. -- 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 50354] New: Team Fortress 2 crashes during loading
by WineHQ Bugzilla 14 Jan '21

14 Jan '21

https://bugs.winehq.org/show_bug.cgi?id=50354 Bug ID: 50354 Summary: Team Fortress 2 crashes during loading Product: Wine Version: 5.20 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: gdi32 Assignee: wine-bugs(a)winehq.org Reporter: random-nick(a)mail.com Regression SHA1: d7a54394d469480e4a5e7a8b3359f0e3b32515ac Distribution: --- Created attachment 68952 --> https://bugs.winehq.org/attachment.cgi?id=68952 log of launching tf2 before d7a54394d469480e4a5e7a8b3359f0e3b32515ac Team Fortress 2 crashes while it's loading, without any sort of error message from the game itself or a crash dialog. This is a regression, since it stopped working in Wine 5.20. I bisected it and the first bad commit is 7a54394d469480e4a5e7a8b3359f0e3b32515ac (gdi32: Move the realized font handle allocation out of freetype.c.), which is why I set the component to gdi32. Note that tf2 crashes when ran without Steam just after it finishes loading (which might happen on Windows too, not sure), so if you are testing without Steam running you have to look for when the black square that says "Loading" disappears to know this issue isn't happening. -- 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 7

[Bug 50479] New: A Vampyre Story from GOG crashes due to missing resolution
by WineHQ Bugzilla 14 Jan '21

14 Jan '21

https://bugs.winehq.org/show_bug.cgi?id=50479 Bug ID: 50479 Summary: A Vampyre Story from GOG crashes due to missing resolution Product: Wine Version: 6.0-rc6 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-d3d Assignee: wine-bugs(a)winehq.org Reporter: titan.costa(a)gmail.com Distribution: --- Created attachment 69115 --> https://bugs.winehq.org/attachment.cgi?id=69115 log with +d3d,+ddraw The game wants a resolution of 1024x768 but does not find 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.

1 26

[Bug 37620] New: MySims: Unable to pick up building blocks
by wine-bugs＠winehq.org 13 Jan '21

13 Jan '21

https://bugs.winehq.org/show_bug.cgi?id=37620 Bug ID: 37620 Summary: MySims: Unable to pick up building blocks Product: Wine Version: 1.7.31 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: daniel.attard120201(a)icloud.com Distribution: --- Created attachment 50068 --> https://bugs.winehq.org/attachment.cgi?id=50068 MySims Wine log MySims is fully working but when you are going to pick up a block to build your first house, the blocks/items can't be picked up. The cursor just ignores them. -- 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 3

[Bug 50497] New: StarCraft Mass Recall hangs on loading screen
by WineHQ Bugzilla 13 Jan '21

13 Jan '21

https://bugs.winehq.org/show_bug.cgi?id=50497 Bug ID: 50497 Summary: StarCraft Mass Recall hangs on loading screen Product: Wine Version: 6.0-rc1 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-d3d-util Assignee: wine-bugs(a)winehq.org Reporter: pentsak(a)gmail.com Distribution: --- Created attachment 69151 --> https://bugs.winehq.org/attachment.cgi?id=69151 logs Lutris: 0.5.8.2 Wine: lutris-6.0-rc1-x86_64 StarCraftII Mass Recall https://www.sc2mapster.com/projects/starcraft-mass-recall/files Files: SCMR Assets v8.0 StarCraft Mass Recall v8.0 Enslavers Redux v8.0 SCMR Localization Russian v8.0 SCMR Cinematics English (SCMRcinematics_v7.2) Install StarCraftII via Lutris Install Mass Recall Run it using Lutris like: /opt/WINE_DISK/COD/starcraft-ii/dosdevices/c:/Games/StarCraft II/Support64/SC2Switcher_x64.exe "c:\Games\StarCraft II\Maps\Starcraft Mass Recall\SCMR Campaign Launcher.SC2Map" In log 0180:err:ntdll:RtlpWaitForCriticalSection section 000000006737C0C0 "/home/ubuntu/buildbot/runners/wine/wine-src/dlls/d3dcompiler_43/compiler.c: wpp_mutex" wait timed out in thread 0180, blocked by 0168, retrying (60 sec) Logs 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 1

[Bug 50493] New: Midtown Madness 2 graphical error when not in virtual desktop
by WineHQ Bugzilla 12 Jan '21

12 Jan '21

https://bugs.winehq.org/show_bug.cgi?id=50493 Bug ID: 50493 Summary: Midtown Madness 2 graphical error when not in virtual desktop Product: Wine Version: 6.0-rc6 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-d3d Assignee: wine-bugs(a)winehq.org Reporter: haakobja(a)gmail.com Distribution: --- Created attachment 69143 --> https://bugs.winehq.org/attachment.cgi?id=69143 Midtown Madness 2 running in virtual desktop Hi, I tried running Midtown Madness 2 the other day and I came over a graphical glitch when running full screen. When I ran the game in Virtual Desktop it seems to render correctly. -- 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 48144] New: Dream Aquarium (screensaver) fails to read monitor power state (support for 'SPDRP_DEVICE_POWER_DATA' property missing)
by WineHQ Bugzilla 12 Jan '21

12 Jan '21

https://bugs.winehq.org/show_bug.cgi?id=48144 Bug ID: 48144 Summary: Dream Aquarium (screensaver) fails to read monitor power state (support for 'SPDRP_DEVICE_POWER_DATA' property missing) Product: Wine Version: 4.20 Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: setupapi Assignee: wine-bugs(a)winehq.org Reporter: focht(a)gmx.net Distribution: --- Hello folks, continuation of bug 39210 --- snip --- $ pwd /home/focht/.wine/drive_c/Program Files (x86)/Dream Aquarium $ WINEDEBUG=+seh,+relay,+setupapi wine ./Dream_Aquarium.scr -S >>log.txt 2>&1 ... 0039:Call setupapi.SetupDiGetDeviceRegistryPropertyW(001802d0,007dfdc4,0000001e,00000000,007dfd8c,00000038,00000000) ret=00415839 0039:trace:setupapi:SetupDiGetDeviceRegistryPropertyW devinfo 001802D0, device_data 007DFDC4, prop 30, type 00000000, buffer 007DFD8C, size 56, required 00000000 0039:Ret setupapi.SetupDiGetDeviceRegistryPropertyW() retval=00000000 ret=00415839 0039:Call KERNEL32.GetLastError() ret=00415841 0039:Ret KERNEL32.GetLastError() retval=00000000 ret=00415841 ... --- snip --- Application log files: --- snip --- $ pwd /home/focht/.wine/drive_c/users/focht/Application Data/Dream Aquarium $ cat stdout.txt setVerbose(5,init) CERROR allowed ec:3 allow:3 CERR allowed activetimes= aqua versions: 1.27 1.27 build:5 installDir=C:\Program Files (x86)\Dream Aquarium\ countryCode=-1 countryName=�lK TRIAL has valid license 0 use full version: 0 hasValidLicenseButIsRunningTrialVersion:0 about to check if monitors are asleep --- snip --- --- snip --- $ pwd /home/focht/.wine/drive_c/users/focht/Application Data/Dream Aquarium $ cat runLog.txt start 2737: Tue Nov 19 23:19:57 2019 -S 2737: Monitors Asleep 2737: Closing (0 secs) Tue Nov 19 23:19:57 2019 --- snip --- https://source.winehq.org/git/wine.git/blob/ab92627e19f10d7ad23abb101b21ec7… $ wine --version wine-4.20-46-g288a6625ad Regards -- 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 48450] New: MindManager Pro v7.0 shows XML processing error on startup (needs support for 'charset' property)
by WineHQ Bugzilla 12 Jan '21

12 Jan '21

https://bugs.winehq.org/show_bug.cgi?id=48450 Bug ID: 48450 Summary: MindManager Pro v7.0 shows XML processing error on startup (needs support for 'charset' property) Product: Wine Version: 5.0-rc5 Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: msxml3 Assignee: wine-bugs(a)winehq.org Reporter: focht(a)gmx.net Distribution: --- Hello folks, continuation of bug 44594 ("MindManager Pro v7.0 shows XML processing error on startup (needs support for 'exhaustive-errors', 'external-general-entities' and 'external-parameter-entities' features in MSXML SAX reader)"). Prerequisite: 'winetricks -q dotnet20' (if Wine was built without Wine-Mono support). --- snip --- $ pwd /home/focht/.wine/drive_c/Program Files (x86)/Mindjet/MindManager 7 $ WINEDEBUG=+seh,+relay,+msxml wine ./MindManager.exe >>log.txt 2>&1 ... 003c:trace:msxml:isaxxmlreader_putFeature (0x3cbcc08)->(L"exhaustive-errors" 0) ... 003c:trace:msxml:isaxxmlreader_putFeature (0x3cbcc08)->(L"http://xml.org/sax/features/external-general-entities" ffffffff) ... 003c:fixme:msxml:isaxxmlreader_putFeature (0x3cbcc08)->(L"http://xml.org/sax/features/external-general-entities" ffffffff) stub ... 003c:trace:msxml:isaxxmlreader_putFeature (0x3cbcc08)->(L"http://xml.org/sax/features/external-parameter-entities" ffffffff) ... 003c:fixme:msxml:isaxxmlreader_putFeature (0x3cbcc08)->(L"http://xml.org/sax/features/external-parameter-entities" ffffffff) stub ... 003c:trace:msxml:isaxxmlreader_putFeature (0x3cbcc08)->(L"schema-validation" 0) ... 003c:trace:msxml:internal_putProperty (0x3cbcc08)->(L"charset" 0x32dc20 {VT_BSTR: L"UTF-8"}) ... 003c:fixme:msxml:internal_putProperty (0x3cbcc08)->(L"charset":0x32dc20 {VT_BSTR: L"UTF-8"}): unsupported property ... 003c:Call KERNEL32.GetLastError() ret=022d84c5 003c:Ret KERNEL32.GetLastError() retval=000036b7 ret=022d84c5 003c:Call KERNEL32.FormatMessageW(00001300,00000000,000036b7,00000400,00000000,00000000,00000000) ret=022d84f3 003c:Ret KERNEL32.FormatMessageW() retval=00000000 ret=022d84f3 ... 003c:Call msvcr80.vswprintf_s(03ae09a0,0000000b,023068ec L"0x%08X",0032dc5c) ret=78306c82 003c:Ret msvcr80.vswprintf_s() retval=0000000a ret=78306c82 003c:Call ole32.GetErrorInfo(00000000,0032dca0) ret=022d8557 003c:Ret ole32.GetErrorInfo() retval=00000001 ret=022d8557 003c:Call msvcr80._CxxThrowException(0032dd40,0230ce1c) ret=022e5cc7 003c:Call KERNEL32.RaiseException(e06d7363,00000001,00000003,0032dc64) ret=7d926935 ... 003c:trace:seh:raise_exception code=e06d7363 flags=1 addr=0x7b00dab1 ip=7b00dab1 tid=003c 003c:trace:seh:raise_exception info[0]=19930520 003c:trace:seh:raise_exception info[1]=0032dd40 003c:trace:seh:raise_exception info[2]=0230ce1c 003c:trace:seh:raise_exception eax=0032dbb4 ebx=0032ea90 ecx=0032dc64 edx=0032dbb4 esi=00000003 edi=0032dc20 003c:trace:seh:raise_exception ebp=0032dc08 esp=0032dbb4 cs=320023 ds=ffff002b es=002b fs=f7b90063 gs=006b flags=00000216 ... 003c:Call user32.SetWindowTextW(0001033a,03ae0448 L"XML processing error") ret=7833abe2 --- snip --- Wine source: https://source.winehq.org/git/wine.git/blob/78259fceb3c591f91471ad8bd98d843… --- snip --- 2807 static HRESULT internal_putProperty( 2808 saxreader* This, 2809 const WCHAR *prop, 2810 VARIANT value, 2811 BOOL vbInterface) 2812 { 2813 VARIANT *v; 2814 2815 TRACE("(%p)->(%s %s)\n", This, debugstr_w(prop), debugstr_variant(&value)); 2816 2817 if (This->isParsing) return E_FAIL; 2818 2819 v = V_VT(&value) == (VT_VARIANT|VT_BYREF) ? V_VARIANTREF(&value) : &value; 2820 if(!memcmp(prop, PropertyDeclHandlerW, sizeof(PropertyDeclHandlerW))) 2821 return saxreader_put_handler_from_variant(This, SAXDeclHandler, v, vbInterface); 2822 2823 if(!memcmp(prop, PropertyLexicalHandlerW, sizeof(PropertyLexicalHandlerW))) 2824 return saxreader_put_handler_from_variant(This, SAXLexicalHandler, v, vbInterface); 2825 2826 if(!memcmp(prop, PropertyMaxXMLSizeW, sizeof(PropertyMaxXMLSizeW))) 2827 { 2828 if (V_VT(v) == VT_I4 && V_I4(v) == 0) return S_OK; 2829 FIXME("(%p)->(%s): max-xml-size unsupported\n", This, debugstr_variant(v)); 2830 return E_NOTIMPL; 2831 } 2832 2833 if(!memcmp(prop, PropertyMaxElementDepthW, sizeof(PropertyMaxElementDepthW))) 2834 { 2835 if (V_VT(v) == VT_I4 && V_I4(v) == 0) return S_OK; 2836 FIXME("(%p)->(%s): max-element-depth unsupported\n", This, debugstr_variant(v)); 2837 return E_NOTIMPL; 2838 } 2839 2840 FIXME("(%p)->(%s:%s): unsupported property\n", This, debugstr_w(prop), debugstr_variant(v)); 2841 2842 if(!memcmp(prop, PropertyCharsetW, sizeof(PropertyCharsetW))) 2843 return E_NOTIMPL; 2844 ... 2862 2863 return E_INVALIDARG; 2864 } --- snip --- $ sha1sum MM70-E-429_Pro.exe a546bdbad175260764b0a11f250d02950b93e940 MM70-E-429_Pro.exe $ du -sh MM70-E-429_Pro.exe 77M MM70-E-429_Pro.exe $ wine --version wine-5.0-rc5 Regards -- 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 48534] New: 32-bit MSXML 6.0 package fails to install in 64-bit WINEPREFIX 'package is not supported on the current processor type.'
by WineHQ Bugzilla 12 Jan '21

12 Jan '21

https://bugs.winehq.org/show_bug.cgi?id=48534 Bug ID: 48534 Summary: 32-bit MSXML 6.0 package fails to install in 64-bit WINEPREFIX 'package is not supported on the current processor type.' Product: Wine Version: 5.0 Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: msi Assignee: wine-bugs(a)winehq.org Reporter: focht(a)gmx.net Distribution: --- Hello folks, extracted/continuation from bug 44618 The MSXML 6.0 package is part of MindManager Pro v7.0 installer (app is 32-bit). It fails in default 64-bit WINEPREFIX as the MSXML6 redistributable package is only for 32-bit. --- snip --- $ WINEDEBUG=+msgbox wine ./MM70-E-429_Pro.exe ... 0077:trace:msgbox:MSGBOX_OnInit L"This MSXML6.0 package is not supported on the current processor type." 0077:err:msi:ITERATE_Actions Execution halted, action L"LaunchConditions" returned 1603 ... 0039:err:msi:ITERATE_Actions Execution halted, action L"RunMSXML6Setup" returned 1627 ... 0039:err:msi:ITERATE_Actions Execution halted, action L"ExecuteAction" returned 1627 --- snip --- MSXML6 package 'LaunchCondition' is indeed: --- snip --- (NOT VersionNT64) [WrongPackage] --- snip --- Direct link from Microsoft: https://download.microsoft.com/download/e/a/f/eafb8ee7-667d-4e30-bb39-4694b… $ sha1sum msxml6_x86.msi 5125220e985b33c946bbf9f60e2b222c7570bfa2 msxml6_x86.msi $ du -sh msxml6_x86.msi 1.5M msxml6_x86.msi $ wine --version wine-5.0-201-gcf6546fb3b Regards -- 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 48861] New: x64dbg fails to set debuggee command line via File -> Change Command Line menu
by WineHQ Bugzilla 12 Jan '21

12 Jan '21

https://bugs.winehq.org/show_bug.cgi?id=48861 Bug ID: 48861 Summary: x64dbg fails to set debuggee command line via File -> Change Command Line menu Product: Wine Version: 5.5 Hardware: x86-64 OS: Linux Status: NEW Severity: normal Priority: P2 Component: kernelbase Assignee: wine-bugs(a)winehq.org Reporter: focht(a)gmx.net Distribution: --- Hello folks, https://github.com/x64dbg/x64dbg/wiki/Frequently-Asked-Questions --- quote --- Q: Help, how do I pass arguments to the program I want to debug? A: Use the File -> Change Command Line option: --- quote --- This doesn't work under Wine. One has to start the debuggee using an undocumented way. https://github.com/x64dbg/x64dbg/issues/210 The documentation task https://github.com/x64dbg/docs/issues/26 is still open since 2016. Examples: --- snip -- # starts x64dbg with 64-bit debuggee and args $ wine ./x96dbg.exe "c:\\windows\\system32\\cmd.exe" /k echo "%PROCESSOR_ARCHITECTURE%" # starts x32dbg with 32-bit debuggee and args $ wine ./x96dbg.exe "c:\\windows\\syswow64\\cmd.exe" /k echo "%PROCESSOR_ARCHITECTURE%" --- snip --- Anyway, the problem here is how Wine implements 'kernel32.GetCommandLine{A,W}' functions. Starting with x64dbg sources... https://github.com/x64dbg/x64dbg/blob/d2c62e69c2cd4d42cc4509612ad3cfd34e914… --- snip --- bool dbgsetcmdline(const char* cmd_line, cmdline_error_t* cmd_line_error) { // Make sure cmd_line_error is a valid pointer cmdline_error_t cmd_line_error_aux; if(cmd_line_error == NULL) cmd_line_error = &cmd_line_error_aux; // Get the command line address if(!getcommandlineaddr(&cmd_line_error->addr, cmd_line_error)) return false; auto command_line_addr = cmd_line_error->addr; // Convert the string to UTF-16 auto command_linewstr = StringUtils::Utf8ToUtf16(cmd_line); if(command_linewstr.length() >= 32766) //32766 is maximum character count for a null-terminated UNICODE_STRING command_linewstr.resize(32766); // Convert the UTF-16 string to ANSI auto command_linestr = Utf16ToAnsi(command_linewstr.c_str()); // Fill the UNICODE_STRING to be set in the debuggee UNICODE_STRING new_command_line; new_command_line.Length = USHORT(command_linewstr.length() * sizeof(WCHAR)); //max value: 32766 * 2 = 65532 new_command_line.MaximumLength = new_command_line.Length + sizeof(WCHAR); //max value: 65532 + 2 = 65534 new_command_line.Buffer = PWSTR(command_linewstr.c_str()); //allow cast from const because the UNICODE_STRING will not be used locally // Allocate remote memory for both the UNICODE_STRING.Buffer and the (null terminated) ANSI buffer duint mem = MemAllocRemote(0, new_command_line.MaximumLength + command_linestr.size()); if(!mem) { cmd_line_error->type = CMDL_ERR_ALLOC_UNICODEANSI_COMMANDLINE; return false; } // Write the UNICODE_STRING.Buffer to the debuggee (UNICODE_STRING.Length is used because the remote memory is zeroed) if(!MemWrite(mem, new_command_line.Buffer, new_command_line.Length)) { cmd_line_error->addr = mem; cmd_line_error->type = CMDL_ERR_WRITE_UNICODE_COMMANDLINE; return false; } // Write the (null-terminated) ANSI buffer to the debuggee if(!MemWrite(mem + new_command_line.MaximumLength, command_linestr.data(), command_linestr.size())) { cmd_line_error->addr = mem + new_command_line.MaximumLength; cmd_line_error->type = CMDL_ERR_WRITE_ANSI_COMMANDLINE; return false; } // Change the pointers to the command line if(!fixgetcommandlinesbase(mem, mem + new_command_line.MaximumLength, cmd_line_error)) return false; // Put the remote buffer address in the UNICODE_STRING and write it to the PEB new_command_line.Buffer = PWSTR(mem); if(!MemWrite(command_line_addr, &new_command_line, sizeof(new_command_line))) { cmd_line_error->addr = command_line_addr; cmd_line_error->type = CMDL_ERR_WRITE_PEBUNICODE_COMMANDLINE; return false; } // Copy command line copyCommandLine(cmd_line); return true; } --- snip --- Corresponding Wine trace log snippet: --- snip --- ... 0009:Call KERNEL32.VirtualAllocEx(0000024c,00000000,00000087,00003000,00000040) ret=01dd445c 0009:Call ntdll.NtAllocateVirtualMemory(0000024c,0032ce3c,00000000,0032ce58,00003000,00000040) ret=7b022f18 0009: queue_apc( handle=024c, call={APC_VIRTUAL_ALLOC,addr==00000000,size=00000087,zero_bits_64=0,op_type=3000,prot=40} ) 0045: *wakeup* signaled=256 0009: queue_apc() = 0 { handle=02e8, self=0 } 0045: select( flags=2, cookie=7ffdb10c, timeout=infinite, prev_apc=0000, result={APC_VIRTUAL_QUERY,status=0,base=00330000,alloc_base=00330000,size=00001000,state=1,prot=2,alloc_prot=80,alloc_type=100}, data={WAIT,handles={0048}} ) 0045: select() = KERNEL_APC { timeout=infinite, call={APC_VIRTUAL_ALLOC,addr==00000000,size=00000087,zero_bits_64=0,op_type=3000,prot=40}, apc_handle=004c } 0009: select( flags=2, cookie=0032c9ac, timeout=infinite, prev_apc=0000, result={}, data={WAIT_ALL,handles={02e8}} ) 0009: select() = PENDING { timeout=infinite, call={APC_NONE}, apc_handle=0000 } 0045: select( flags=2, cookie=7ffdb10c, timeout=infinite, prev_apc=004c, result={APC_VIRTUAL_ALLOC,status=0,addr=00380000,size=00001000}, data={WAIT,handles={0048}} ) 0009: *wakeup* signaled=0 0045: select() = PENDING { timeout=infinite, call={APC_NONE}, apc_handle=0000 } 0009: get_apc_result( handle=02e8 ) 0009: get_apc_result() = 0 { result={APC_VIRTUAL_ALLOC,status=0,addr=00380000,size=00001000} } 0009:Ret ntdll.NtAllocateVirtualMemory() retval=00000000 ret=7b022f18 0009:Ret KERNEL32.VirtualAllocEx() retval=00380000 ret=01dd445c 0009:Call KERNEL32.WriteProcessMemory(0000024c,00380000,05ab7250,00000058,0032cea8) ret=0211b7b8 0009:Call ntdll.NtWriteVirtualMemory(0000024c,00380000,05ab7250,00000058,0032cea8) ret=7b023199 0009: write_process_memory( handle=024c, addr=00380000, data={22,00,43,00,3a,00,5c,00,77,00,69,00,6e,00,64,00,6f,00,77,00,73,00,5c,00,73,00,79,00,73,00,77,00,6f,00,77,00,36,00,34,00,5c,00,77,00,69,00,6e,00,65,00,6d,00,69,00,6e,00,65,00,2e,00,65,00,78,00,65,00,22,00,20,00,2d,00,74,00,65,00,73,00,74,00,61,00,72,00,67,00,31,00} ) 0045: *signal* signal=19 0009: write_process_memory() = 0 0009:Ret ntdll.NtWriteVirtualMemory() retval=00000000 ret=7b023199 0009:Ret KERNEL32.WriteProcessMemory() retval=00000001 ret=0211b7b8 0009:Call KERNEL32.WriteProcessMemory(0000024c,0038005a,05a989f8,0000002d,0032cea8) ret=0211b7b8 0009:Call ntdll.NtWriteVirtualMemory(0000024c,0038005a,05a989f8,0000002d,0032cea8) ret=7b023199 0009: write_process_memory( handle=024c, addr=0038005a, data={22,43,3a,5c,77,69,6e,64,6f,77,73,5c,73,79,73,77,6f,77,36,34,5c,77,69,6e,65,6d,69,6e,65,2e,65,78,65,22,20,2d,74,65,73,74,61,72,67,31,00} ) 0045: *signal* signal=19 0009: write_process_memory() = 0 0009:Ret ntdll.NtWriteVirtualMemory() retval=00000000 ret=7b023199 0009:Ret KERNEL32.WriteProcessMemory() retval=00000001 ret=0211b7b8 --- snip --- Both, ANSI and UNICODE buffers are allocated and filled in debuggee process space. https://github.com/x64dbg/x64dbg/blob/d2c62e69c2cd4d42cc4509612ad3cfd34e914… --- snip --- static bool fixgetcommandlinesbase(duint new_command_line_unicode, duint new_command_line_ascii, cmdline_error_t* cmd_line_error) { duint getcommandline; if(!valfromstring("kernelBase:GetCommandLineA", &getcommandline)) { if(!valfromstring("kernel32:GetCommandLineA", &getcommandline)) { cmd_line_error->type = CMDL_ERR_GET_GETCOMMANDLINE; return false; } } if(!patchcmdline(getcommandline, new_command_line_ascii, cmd_line_error)) return false; if(!valfromstring("kernelbase:GetCommandLineW", &getcommandline)) { if(!valfromstring("kernel32:GetCommandLineW", &getcommandline)) { cmd_line_error->type = CMDL_ERR_GET_GETCOMMANDLINE; return false; } } if(!patchcmdline(getcommandline, new_command_line_unicode, cmd_line_error)) return false; return true; } --- snip --- Corresponding Wine trace log snippet: --- snip --- ... 0009:Call KERNEL32.LoadLibraryExW(05ab78d8 L"Z:\\home\\focht\\projects\\wine\\mainline-install-x86_64\\lib\\wine\\kernelbase.dll",00000000,00000001) ret=01e0dbdf ... 0009:Ret KERNEL32.LoadLibraryExW() retval=7b000000 ret=01e0dbdf ... 0009:Call KERNEL32.GetProcAddress(7b000000,05ab78bb "GetCommandLineA") ret=01e0b046 0009:Ret KERNEL32.GetProcAddress() retval=7b0036ec ret=01e0b046 0009:Call KERNEL32.FreeLibrary(7b000000) ret=01e0ddf9 0009:Ret KERNEL32.FreeLibrary() retval=00000001 ret=01e0ddf9 ... 0009:Call KERNEL32.ReadProcessMemory(0000024c,7b0036ec,0032ce2c,00000064,0032ce08) ret=0211b6bc 0009:Call ntdll.NtReadVirtualMemory(0000024c,7b0036ec,0032ce2c,00000064,0032ce08) ret=7b022cf9 0009: read_process_memory( handle=024c, addr=7b0036ec ) 0045: *signal* signal=19 0009: read_process_memory() = 0 { data={8b,ff,55,8b,ec,5d,68,36,01,0c,00,b8,00,50,07,7b,50,ff,50,04,c2,00,00,90,90,90,90,90,90,90,90,90,8b,ff,55,8b,ec,5d,68,37,01,0c,00,b8,00,50,07,7b,50,ff,50,04,c2,00,00,90,90,90,90,90,90,90,90,90,8b,ff,55,8b,ec,5d,68,38,01,0a,00,b8,00,50,07,7b,50,ff,50,04,c2,08,00,90,90,90,90,90,90,90,90,90,8b,ff,55,8b} } 0009:Ret ntdll.NtReadVirtualMemory() retval=00000000 ret=7b022cf9 0009:Ret KERNEL32.ReadProcessMemory() retval=00000001 ret=0211b6bc ... --- snip --- x32dbg reads the opcodes of 32-bit 'kernelbase.GetCommandLineA' and is not happy about it. --- snip --- static bool patchcmdline(duint getcommandline, duint new_command_line, cmdline_error_t* cmd_line_error) { duint command_line_stored = 0; unsigned char data[100]; cmd_line_error->addr = getcommandline; if(!MemRead(cmd_line_error->addr, & data, sizeof(data))) { cmd_line_error->type = CMDL_ERR_READ_GETCOMMANDLINEBASE; return false; } #ifdef _WIN64 /* 00007FFC5B91E3C8 | 48 8B 05 19 1D 0E 00 | mov rax,qword ptr ds:[7FFC5BA000E8] 00007FFC5B91E3CF | C3 | ret | This is a relative offset then to get the symbol: next instruction of getmodulehandle (+7 bytes) + offset to symbol (the last 4 bytes of the instruction) */ if(data[0] != 0x48 || data[1] != 0x8B || data[2] != 0x05 || data[7] != 0xC3) { cmd_line_error->type = CMDL_ERR_CHECK_GETCOMMANDLINESTORED; return false; } DWORD offset = * ((DWORD*) & data[3]); command_line_stored = getcommandline + 7 + offset; #else //x86 /* 750FE9CA | A1 CC DB 1A 75 | mov eax,dword ptr ds:[751ADBCC] | 750FE9CF | C3 | ret | */ if(data[0] != 0xA1 || data[5] != 0xC3) { cmd_line_error->type = CMDL_ERR_CHECK_GETCOMMANDLINESTORED; return false; } command_line_stored = * ((duint*) & data[1]); #endif //update the pointer in the debuggee if(!MemWrite(command_line_stored, &new_command_line, sizeof(new_command_line))) { cmd_line_error->addr = command_line_stored; cmd_line_error->type = CMDL_ERR_WRITE_GETCOMMANDLINESTORED; return false; } return true; } --- snip --- Wine source: https://source.winehq.org/git/wine.git/blob/9bcb1f5195d1e65e0e7afb288d36fee… --- snip --- 1008 static STARTUPINFOW startup_infoW; 1009 static char *command_lineA; 1010 static WCHAR *command_lineW; --- snip --- https://source.winehq.org/git/wine.git/blob/9bcb1f5195d1e65e0e7afb288d36fee… --- snip --- 1043 /*********************************************************************** 1044 * GetCommandLineA (kernelbase.@) 1045 */ 1046 LPSTR WINAPI DECLSPEC_HOTPATCH GetCommandLineA(void) 1047 { 1048 return command_lineA; 1049 } 1050 1051 1052 /*********************************************************************** 1053 * GetCommandLineW (kernelbase.@) 1054 */ 1055 LPWSTR WINAPI DECLSPEC_HOTPATCH GetCommandLineW(void) 1056 { 1057 return NtCurrentTeb()->Peb->ProcessParameters->CommandLine.Buffer; 1058 } --- snip --- Depending on which compiler (GCC or LLVM/MingW) is used to build Wine, the following code is generated: === GCC === --- snip -- $ gcc -v Using built-in specs. COLLECT_GCC=/usr/bin/gcc COLLECT_LTO_WRAPPER=/usr/libexec/gcc/x86_64-redhat-linux/8/lto-wrapper OFFLOAD_TARGET_NAMES=nvptx-none OFFLOAD_TARGET_DEFAULT=1 Target: x86_64-redhat-linux Configured with: ../configure --enable-bootstrap --enable-languages=c,c++,fortran,objc,obj-c++,ada,go,lto --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --with-bugurl=http://bugzilla.redhat.com/bugzilla --enable-shared --enable-threads=posix --enable-checking=release --enable-multilib --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-gnu-unique-object --enable-linker-build-id --with-gcc-major-version-only --with-linker-hash-style=gnu --enable-plugin --enable-initfini-array --with-isl --enable-libmpx --enable-offload-targets=nvptx-none --without-cuda-driver --enable-gnu-indirect-function --enable-cet --with-tune=generic --with-arch_32=i686 --build=x86_64-redhat-linux Thread model: posix gcc version 8.3.1 20190223 (Red Hat 8.3.1-2) (GCC) --- snip --- --- snip --- $ gdb mainline-install-x86_64/lib64/wine/kernelbase.dll.so -batch -ex 'disassemble GetCommandLineA' Dump of assembler code for function GetCommandLineA: 0x000000007b05fd18 <+0>: push %rbp 0x000000007b05fd19 <+1>: mov 0x738a0(%rip),%rax # 0x7b0d35c0 <command_lineA> 0x000000007b05fd20 <+8>: mov %rsp,%rbp 0x000000007b05fd23 <+11>: and $0xfffffffffffffff0,%rsp 0x000000007b05fd27 <+15>: leaveq 0x000000007b05fd28 <+16>: retq End of assembler dump. --- snip --- --- snip --- $ gdb mainline-install-x86_64/lib64/wine/kernelbase.dll.so -batch -ex 'disassemble GetCommandLineW' Dump of assembler code for function GetCommandLineW: 0x000000007b05fd58 <+0>: push %rbp 0x000000007b05fd59 <+1>: mov %gs:0x30,%rax 0x000000007b05fd62 <+10>: mov 0x60(%rax),%rax 0x000000007b05fd66 <+14>: mov %rsp,%rbp 0x000000007b05fd69 <+17>: and $0xfffffffffffffff0,%rsp 0x000000007b05fd6d <+21>: mov 0x20(%rax),%rax 0x000000007b05fd71 <+25>: mov 0x78(%rax),%rax 0x000000007b05fd75 <+29>: leaveq 0x000000007b05fd76 <+30>: retq End of assembler dump. --- snip --- NOTE: offset-0x4 to work around dwarf symbol processing bug. --- snip --- $ gdb mainline-install-x86_64/lib/wine/kernelbase.dll.so -batch -ex 'disassemble GetCommandLineA-0x4' Dump of assembler code for function GetCommandLineA: 0x7b060620 <+0>: mov %edi,%edi 0x7b060622 <+2>: push %ebp 0x7b060623 <+3>: mov %esp,%ebp 0x7b060625 <+0>: pop %ebp 0x7b060626 <+1>: mov 0x7b0d4a20,%eax 0x7b06062b <+6>: ret End of assembler dump. --- snip --- --- snip --- $ gdb mainline-install-x86_64/lib/wine/kernelbase.dll.so -batch -ex 'disassemble GetCommandLineW-0x4' Dump of assembler code for function GetCommandLineW: 0x7b060640 <+0>: mov %edi,%edi 0x7b060642 <+2>: push %ebp 0x7b060643 <+3>: mov %esp,%ebp 0x7b060645 <+0>: mov %fs:0x18,%eax 0x7b06064b <+6>: mov 0x30(%eax),%eax 0x7b06064e <+9>: pop %ebp 0x7b06064f <+10>: mov 0x10(%eax),%eax 0x7b060652 <+13>: mov 0x44(%eax),%eax 0x7b060655 <+16>: ret End of assembler dump. --- snip --- === LLVM MinGW === --- snip --- $ clang -v clang version 10.0.0 (https://github.com/llvm/llvm-project.git c49194969430f0ee817498a7000a979a7a0ded03) Target: x86_64-unknown-linux-gnu Thread model: posix InstalledDir: /home/focht/projects/llvm-mingw-20191230-ubuntu-16.04/bin Found candidate GCC installation: /usr/lib/gcc/x86_64-redhat-linux/8 Selected GCC installation: /usr/lib/gcc/x86_64-redhat-linux/8 Candidate multilib: .;@m64 Candidate multilib: 32;@m32 Selected multilib: .;@m64 --- snip --- --- snip --- $ gdb mainline-install-x86_64/lib/wine/kernelbase.dll -batch -ex 'disassemble GetCommandLineA' Dump of assembler code for function _GetCommandLineA@0: 0x7b0313e0 <+0>: push %ebp 0x7b0313e1 <+1>: mov %esp,%ebp 0x7b0313e3 <+3>: mov 0x7b076bf8,%eax 0x7b0313e8 <+8>: pop %ebp 0x7b0313e9 <+9>: ret End of assembler dump. --- snip --- --- snip --- $ gdb mainline-install-x86_64/lib/wine/kernelbase.dll -batch -ex 'disassemble GetCommandLineW' Dump of assembler code for function _GetCommandLineW@0: 0x7b0313f0 <+0>: push %ebp 0x7b0313f1 <+1>: mov %esp,%ebp 0x7b0313f3 <+3>: mov %fs:0x18,%eax 0x7b0313f9 <+9>: mov 0x30(%eax),%eax 0x7b0313fc <+12>: mov 0x10(%eax),%eax 0x7b0313ff <+15>: mov 0x44(%eax),%eax 0x7b031402 <+18>: pop %ebp 0x7b031403 <+19>: ret End of assembler dump. --- snip --- --- snip --- $ gdb mainline-install-x86_64/lib64/wine/kernelbase.dll -batch -ex 'disassemble GetCommandLineA' Dump of assembler code for function GetCommandLineA: 0x000000007b0389b0 <+0>: mov 0x4de49(%rip),%rax # 0x7b086800 0x000000007b0389b7 <+7>: retq End of assembler dump. --- snip --- --- snip --- $ gdb mainline-install-x86_64/lib64/wine/kernelbase.dll -batch -ex 'disassemble GetCommandLineW' Dump of assembler code for function GetCommandLineW: 0x000000007b0389c0 <+0>: mov %gs:0x30,%rax 0x000000007b0389c9 <+9>: mov 0x60(%rax),%rax 0x000000007b0389cd <+13>: mov 0x20(%rax),%rax 0x000000007b0389d1 <+17>: mov 0x78(%rax),%rax 0x000000007b0389d5 <+21>: retq End of assembler dump. --- snip --- ===== General problems: (1) asymmetric design of kernelbase GetCommandLineA/W commmand buffer handling The ANSI variant returns the command line buffer address directly while the UNICODE variant goes through the PEB. The symmetric design was changed in 2003 with https://source.winehq.org/git/wine.git/commitdiff/b53b5bcb50e75263e8bbffed3… ("- fixed a couple of bugs in ntdll...") to be asymmetrical. A big squash-like commit, doing major refactoring/various bugfixing = several things in one go. No mentioning of this specific design change. Part of Wine 0.9 release. (2) HOTPATCH entry Commit https://source.winehq.org/git/wine.git/commitdiff/911e50849a9bc05ab11d896a3… ("kernel32: Move process startup information functions to kernelbase.") added it. Part of Wine 4.16 release. I couldn't find a bug report about hooking problems of GetCommandLineA/W which requires those to be hotpatch-compatible entries. I guess this was added for no special reason :| (3) compiler choice/optimization settings, influencing generated code --- Fixing above problems is not that easy by just returning 'command_lineW' in 'W' variant and removing DECLSPEC_HOTPATCH because of (3). Generated code: --- snip --- $ gdb mainline-install-x86_64/lib/wine/kernelbase.dll -batch -ex 'disassemble GetCommandLineA' Dump of assembler code for function _GetCommandLineA@0: 0x7b0313e0 <+0>: push %ebp 0x7b0313e1 <+1>: mov %esp,%ebp 0x7b0313e3 <+3>: mov 0x7b076bf8,%eax 0x7b0313e8 <+8>: pop %ebp 0x7b0313e9 <+9>: ret End of assembler dump. --- snip --- --- snip --- $ gdb mainline-install-x86_64/lib/wine/kernelbase.dll -batch -ex 'disassemble GetCommandLineW' Dump of assembler code for function _GetCommandLineW@0: 0x7b0313f0 <+0>: push %ebp 0x7b0313f1 <+1>: mov %esp,%ebp 0x7b0313f3 <+3>: mov 0x7b076bf4,%eax 0x7b0313f8 <+8>: pop %ebp 0x7b0313f9 <+9>: ret End of assembler dump. --- snip --- --- snip --- $ gdb mainline-install-x86_64/lib64/wine/kernelbase.dll -batch -ex 'disassemble GetCommandLineA' Dump of assembler code for function GetCommandLineA: 0x000000007b0389b0 <+0>: mov 0x4de49(%rip),%rax # 0x7b086800 0x000000007b0389b7 <+7>: retq End of assembler dump. --- snip --- --- snip --- $ gdb mainline-install-x86_64/lib64/wine/kernelbase.dll -batch -ex 'disassemble GetCommandLineW' Dump of assembler code for function GetCommandLineW: 0x000000007b0389c0 <+0>: mov 0x4de31(%rip),%rax # 0x7b0867f8 0x000000007b0389c7 <+7>: retq End of assembler dump. --- snip --- The 64-bit variant looks good, but 32-bit is no-go. To have full control over the generated code, small __i386__ and __x86_64__ assembly implementations for A and W variant each are sufficient. NOTE: the assembly for __x86_64__ needs to use RIP-relative addressing (PIE) -> movq command_lineW(%rip), %rax * encoding: [0x48,0x8b,0x05,A,A,A,A] * kind: reloc_riprel_4byte_movq_load With the small assembly functions in place, x64dbg and x32dbg were able to set the debuggee command line via menu. If assembly is not an option, consider changing A/W variant to be symmetrical to make this bug report somewhat useful in the end (but still WONTFIX). $ sha1sum snapshot_2020-02-11_01-07.zip 03fa18ffb1cbdd46a6b3c9b7e19332e7e2024edc snapshot_2020-02-11_01-07.zip $ du -sh snapshot_2020-02-11_01-07.zip 31M snapshot_2020-02-11_01-07.zip $ wine --version wine-5.5 Regards -- 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

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