[Bug 50583] New: Problems when installing Foxit PhantomPDF Business
https://bugs.winehq.org/show_bug.cgi?id=50583 Bug ID: 50583 Summary: Problems when installing Foxit PhantomPDF Business Product: Wine Version: 6.0 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: critical Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: te.regalo.er.burro(a)gmail.com Distribution: --- Created attachment 69259 --> https://bugs.winehq.org/attachment.cgi?id=69259 log, register dump Problems when installing Foxit PhantomPDF Business -- 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.
https://bugs.winehq.org/show_bug.cgi?id=50583 Anastasius Focht <focht(a)gmx.net> changed: What |Removed |Added ---------------------------------------------------------------------------- URL| |https://web.archive.org/web | |/20210129125039/https://cdn | |09.foxitsoftware.com/produc | |t/phantomPDF/desktop/win/10 | |.0/3944C0ADCC878224B13B138F | |985EC3D7/FoxitPhantomPDF100 | |_L10N_Setup_Website.exe Severity|critical |normal Summary|Problems when installing |Foxit PhantomPDF Business |Foxit PhantomPDF Business |v10.0 installer crashes | |with WinVer set to 'Windows | |10' Ever confirmed|0 |1 Keywords| |download, Installer CC| |focht(a)gmx.net Status|UNCONFIRMED |NEW --- Comment #1 from Anastasius Focht <focht(a)gmx.net> --- Hello folks, confirming. Adding stable download link via Internet Archive: https://web.archive.org/web/20210129125039/https://cdn09.foxitsoftware.com/p...
From the attached backtrace:
--- snip --- 000003d4 (D) C:\windows\syswow64\msiexec.exe 0000042c 0 00000498 0 <== 000004c0 0 ... System information: Wine build: wine-6.0 Platform: i386 (WOW64) Version: Windows 10 Host system: Linux Host version: 5.8.0-40-generic --- snip --- Is there any reason for the 'Windows 10' version setting of the WINEPREFIX? Don't do that unless you know the technical implications. With default 'Windows 7' setting the installer runs into a different problem (PDF printer installation fails) which warrants a separate bug report. Have to actually check for dupe first. Anyway, from a cursory look there is an unhandled C++ exception thrown in custom action 'SetStartMenuWindows10T': --- snip --- $ WINEDEBUG=+seh,+relay,+loaddll,+msi wine ./FoxitPhantomPDF100_L10N_Setup_Website.exe >>log.txt 2>&1 ... 0124:trace:msi:ACTION_CustomAction Handling custom action L"SetStartMenuWindows10T" (c41 L"FX_CUSTOM" L"SetStartMenuWindows10T") ... 01e4:Call KERNEL32.LoadLibraryW(00174a58 L"C:\\users\\focht\\Temp\\msi7e70.tmp") ret=100210f6 ... 01e4:Ret PE DLL (proc=017E3329,module=01600000 L"msi7e70.tmp",reason=PROCESS_ATTACH,res=00000000) retval=1 ... 01e4:Ret KERNEL32.LoadLibraryW() retval=01600000 ret=100210f6 01e4:Call KERNEL32.GetProcAddress(01600000,00172680 "SetStartMenuWindows10T") ret=1002110a 01e4:Ret KERNEL32.GetProcAddress() retval=0163b820 ret=1002110a ... 01e4:Call shlwapi.PathFileExistsW(01d57aa0 L"C:\\users\\focht\\Start Menu\\Programs\\Foxit PhantomPDF\\") ret=0163b92e 01e4:Call kernelbase.PathFileExistsW(01d57aa0 L"C:\\users\\focht\\Start Menu\\Programs\\Foxit PhantomPDF\\") ret=7bc3ab94 ... 01e4:Call ntdll.RtlDosPathNameToNtPathName_U(01d57aa0 L"C:\\users\\focht\\Start Menu\\Programs\\Foxit PhantomPDF\\",015ef9c0,00000000,00000000) ret=7b014e4a 01e4:Ret ntdll.RtlDosPathNameToNtPathName_U() retval=00000001 ret=7b014e4a 01e4:Call ntdll.NtQueryAttributesFile(015ef9a8,015ef9c8) ret=7b014e87 01e4:Ret ntdll.NtQueryAttributesFile() retval=00000000 ret=7b014e87 ... 01e4:Ret kernelbase.PathFileExistsW() retval=00000001 ret=7bc3ab94 01e4:Ret shlwapi.PathFileExistsW() retval=00000001 ret=0163b92e 01e4:Call shlwapi.PathFileExistsW(01d57280 L"") ret=0163a89f 01e4:Call kernelbase.PathFileExistsW(01d57280 L"") ret=7bc3ab94 .... 01e4:Ret kernelbase.PathFileExistsW() retval=00000000 ret=7bc3ab94 01e4:Ret shlwapi.PathFileExistsW() retval=00000000 ret=0163a89f ... 01e4:Call KERNEL32.RaiseException(e06d7363,00000001,00000003,015ef738) ret=017e0108 ... 01e4:trace:seh:dispatch_exception code=e06d7363 flags=1 addr=7B00FF28 ip=7b00ff28 tid=01e4 01e4:trace:seh:dispatch_exception info[0]=19930520 01e4:trace:seh:dispatch_exception info[1]=015ef77c 01e4:trace:seh:dispatch_exception info[2]=01901fe0 01e4:trace:seh:dispatch_exception eax=015ef684 ebx=00000000 ecx=015ef738 edx=0000000c esi=00000003 edi=015ef6f0 01e4:trace:seh:dispatch_exception ebp=015ef6d8 esp=015ef684 cs=7bc50023 ds=15e002b es=7bc3002b fs=15e0063 gs=006b flags=00000216 01e4:trace:seh:call_vectored_handlers calling handler at 7B00F270 code=e06d7363 flags=1 01e4:trace:seh:call_vectored_handlers handler at 7B00F270 returned 0 01e4:trace:seh:call_stack_handlers calling handler at 01831E46 code=e06d7363 flags=1 ... wine: Unhandled exception 0xe06d7363 in thread 1e4 at address 7B00FF28 (thread 01e4), starting debugger... --- snip --- Not sure where the empty path for the second 'shlwapi.PathFileExistsW' call comes from. Needs some debugging. $ sha1sum FoxitPhantomPDF100_L10N_Setup_Website.exe 7c22cfb60e8e23d6106ced96e3979c5e053fd60a FoxitPhantomPDF100_L10N_Setup_Website.exe $ du -sh FoxitPhantomPDF100_L10N_Setup_Website.exe 670M FoxitPhantomPDF100_L10N_Setup_Website.exe $ wine --version wine-6.0-293-gf72ef20e88f 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.
https://bugs.winehq.org/show_bug.cgi?id=50583 Anastasius Focht <focht(a)gmx.net> changed: What |Removed |Added ---------------------------------------------------------------------------- Summary|Foxit PhantomPDF Business |Foxit PhantomPDF Business |v10.0 installer crashes |v10.0 installer crashes in |with WinVer set to 'Windows |MSI custom action |10' |'SetStartMenuWindows10T' | |with WinVer set to 'Windows | |10' Component|-unknown |msi --- Comment #2 from Anastasius Focht <focht(a)gmx.net> --- Hello folks, I've debugged the custom action multiple times and came to conclusion the C++ exception can't be avoided. The path parameter for the first 'shlwapi.PathFileExistsW' call comes from the custom action data (copy). The path parameter for second 'shlwapi.PathFileExistsW' call is always empty (local var). There is no code path that can initialize it from a non-empty value. Might be a compiler bug or just crappy coding. --- snip --- ... 0163B903 | sub eax,dword ptr ss:[ebp-210] | 0163B909 | sar eax,1 | 0163B90B | sub esi,eax | 0163B90D | dec esi | 0163B90E | push esi | 0163B90F | lea eax,dword ptr ss:[ebp-218] | 0163B915 | push eax | 0163B916 | lea ecx,dword ptr ss:[ebp-210] | 0163B91C | call msi5fa6.160D580 | 0163B921 | mov esi,dword ptr ss:[ebp-218] | 0163B927 | push esi | L"C:\\users\\focht\\..." 0163B928 | call dword ptr ds:[<&PathFileExistsW>] | 0163B92E | mov ebx,dword ptr ss:[ebp-214] | 0163B934 | cmp eax,1 | 0163B937 | jne msi5fa6.163B971 | 0163B939 | push ecx | 0163B93A | lea eax,dword ptr ds:[ebx-10] | arg_4 = L"" 0163B93D | mov dword ptr ss:[ebp-214],esp | 0163B943 | mov edi,esp | 0163B945 | push eax | arg_4 0163B946 | call msi5fa6.16050F0 | 0163B94B | add eax,10 | 0163B94E | mov dword ptr ds:[edi],eax | 0163B950 | lea ecx,dword ptr ds:[esi-10] | 0163B953 | mov dword ptr ss:[ebp-214],esp | 0163B959 | mov edi,esp | 0163B95B | push ecx | arg_0 0163B95C | call msi5fa6.16050F0 | 0163B961 | add eax,10 | 0163B964 | add esp,4 | 0163B967 | mov dword ptr ds:[edi],eax | arg_0 = L"C:\\users\\..." 0163B969 | call msi5fa6.163A860 | ... 0163A860 | push ebp | func takes two arguments 0163A861 | mov ebp,esp | 0163A863 | push FFFFFFFF | 0163A865 | push msi5fa6.1830C67 | 0163A86A | mov eax,dword ptr fs:[0] | 0163A870 | push eax | 0163A871 | sub esp,270 | 0163A877 | mov eax,dword ptr ds:[1932340] | 0163A87C | xor eax,ebp | 0163A87E | mov dword ptr ss:[ebp-10],eax | 0163A881 | push ebx | 0163A882 | push esi | 0163A883 | push edi | 0163A884 | push eax | 0163A885 | lea eax,dword ptr ss:[ebp-C] | 0163A888 | mov dword ptr fs:[0],eax | 0163A88E | mov eax,dword ptr ss:[ebp+C] | arg_4 = pszPath = L"" 0163A891 | push eax | 0163A892 | mov dword ptr ss:[ebp-4],1 | 0163A899 | call dword ptr ds:[<&PathFileExistsW>] | problem! 0163A89F | test eax,eax | 0163A8A1 | jne msi5fa6.163A8AF | ... --- snip --- I'm a bit stumped here and curious if this installer really works on Windows 10 without crashing the 32-bit custom action server process. Could someone create a verbose MSI log and attach it here? https://docs.microsoft.com/en-us/troubleshoot/windows-client/application-man... Tidbit: https://answers.microsoft.com/en-us/windows/forum/all/windows-installer-stop... Two search engine hits on custom action 'SetStartMenuWindows10T' name. 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.
https://bugs.winehq.org/show_bug.cgi?id=50583 --- Comment #3 from Anastasius Focht <focht(a)gmx.net> --- Hello folks, I've created bug 50589 ("Foxit PhantomPDF Business v10.0 installer fails, reports 'Printer installation failed, please check the print service.'") for the installation failure in default 'Windows 7' WINEPREFIX. 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.
http://bugs.winehq.org/show_bug.cgi?id=50583 Alistair Leslie-Hughes <leslie_alistair@hotmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |andreas.franz@arcor.de --- Comment #4 from Alistair Leslie-Hughes <leslie_alistair@hotmail.com> --- *** Bug 58623 has been marked as a duplicate of this bug. *** -- 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.
http://bugs.winehq.org/show_bug.cgi?id=50583 --- Comment #5 from Dmitry Timoshkov <dmitry@baikal.ru> --- (In reply to Anastasius Focht from comment #2)
I've debugged the custom action multiple times and came to conclusion the C++ exception can't be avoided.
Probably __TRY/__EXCEPT around the custom DLL procedure call should be replaced either by vectored or unhandled exception handler because __TRY/__EXCEPT macros in Wine won't catch the exceptions generated by RaiseException(). -- 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.
http://bugs.winehq.org/show_bug.cgi?id=50583 --- Comment #6 from Alexandre Julliard <julliard@winehq.org> --- (In reply to Dmitry Timoshkov from comment #5)
(In reply to Anastasius Focht from comment #2)
I've debugged the custom action multiple times and came to conclusion the C++ exception can't be avoided.
Probably __TRY/__EXCEPT around the custom DLL procedure call should be replaced either by vectored or unhandled exception handler because __TRY/__EXCEPT macros in Wine won't catch the exceptions generated by RaiseException().
They will catch it, but of course __EXCEPT_PAGE_FAULT is not going to handle C++ exceptions. -- 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.
http://bugs.winehq.org/show_bug.cgi?id=50583 --- Comment #7 from Dmitry Timoshkov <dmitry@baikal.ru> --- (In reply to Alexandre Julliard from comment #6)
(In reply to Dmitry Timoshkov from comment #5)
(In reply to Anastasius Focht from comment #2)
I've debugged the custom action multiple times and came to conclusion the C++ exception can't be avoided.
Probably __TRY/__EXCEPT around the custom DLL procedure call should be replaced either by vectored or unhandled exception handler because __TRY/__EXCEPT macros in Wine won't catch the exceptions generated by RaiseException().
They will catch it, but of course __EXCEPT_PAGE_FAULT is not going to handle C++ exceptions.
Thanks for the correction Alexandre, then __EXCEPT_ALL could be a more appropriate filter. -- 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.
http://bugs.winehq.org/show_bug.cgi?id=50583 Dmitry Timoshkov <dmitry@baikal.ru> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Fixed by SHA1| |d91a8a7cf64e7ed91541da53c49 | |d85f79618ddb2 Resolution|--- |FIXED --- Comment #8 from Dmitry Timoshkov <dmitry@baikal.ru> --- This should be fixed by d91a8a7cf64e7ed91541da53c49d85f79618ddb2. -- 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.
http://bugs.winehq.org/show_bug.cgi?id=50583 Alexandre Julliard <julliard@winehq.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED --- Comment #9 from Alexandre Julliard <julliard@winehq.org> --- Closing bugs fixed in 11.11. -- 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.
participants (2)
-
WineHQ Bugzilla -
WineHQ Bugzilla