December 2017 - Wine-Bugs - list.winehq.org

[Bug 15462] New: Wine blocking on exit when stdout and stderr are piped
by wine-bugs＠winehq.org May 18, 2021

May 18, 2021

http://bugs.winehq.org/show_bug.cgi?id=15462 Summary: Wine blocking on exit when stdout and stderr are piped Product: Wine Version: 1.0.0 Platform: PC OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: chris(a)neko.net.nz Created an attachment (id=16357) --> (http://bugs.winehq.org/attachment.cgi?id=16357) Output of two simultaneous runs of described script The issue occurs when concurrently running shell scripts which call Wine processes whose standard out and standard error are both piped to another program. When two instances of a script such as the following are run concurrently, the first child wine processes started by each of then run together, but in one of the scripts that child doesn't fully exit until the _other_ script has completely run and exited. (The Wine process terminates, but the pipeline doesn't complete and so the script doesn't continue.) A bash test-case to show this behaviour is as follows: for iter in first second ; do echo "[$$] running $iter child wine" `date`; wine sleep-long.exe 2>&1 | tee /dev/null; done; echo "[$$] done " `date` The problem doesn't occur if either the redirection or pipe is removed, if I run the scripts in separate Wine environments using $WINEPREFIX, or if I have another long-lived Wine process which stays running throughout the entire operation. An example run with timing is attached. The test Wine program used above, 'sleep-long', is the following tiny C program (tested with Microsoft cl and Cygwin GCC with mingw): #include <stdio.h> #include <windows.h> int main(){ printf("Started fake compiler, sleeping 8000ms\n"); Sleep(8000); printf("Exiting fake compiler\n"); exit(0); } I've been able to reproduce this problem on several systems and Wine versions (Debian Lenny with distro-packaged Wine 1.0.0-1, Debian Etch with Wine 1.0-rc1-1~bpo40+1 from Debian backports, Ubuntu Dapper with distro-packaged Wine 0.9.9-0ubuntu2) but not on one other (Debian Etch with distro-packaged Wine 0.9.25-2.1) -- 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 15

[Bug 15906] New: EnumFonts returns each font multiple times
by wine-bugs＠winehq.org May 18, 2021

May 18, 2021

http://bugs.winehq.org/show_bug.cgi?id=15906 Summary: EnumFonts returns each font multiple times Product: Wine Version: 1.1.5 Platform: PC-x86-64 URL: http://msdn.microsoft.com/en- us/library/ms533964(VS.85).aspx OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: gdi32 AssignedTo: wine-bugs(a)winehq.org ReportedBy: goeran(a)uddeborg.se Created an attachment (id=17078) --> (http://bugs.winehq.org/attachment.cgi?id=17078) Test program executing EnumFonts() According to the documentation (see URL) the EnumFonts function returns one (random) font of each available typeface. The Wine implementation returns each typeface many times. As a test case I installed the typeface "Signet Roundhand" and ran the attached test program. Under indows it returns "Signet Roundhand" once. Under wine it returns it three times, with different values for lfCharSet and tmCharSet. I also copied the Arial font files from Windows (arial.ttf, ariali.ttf, arialbd.ttf, and arialbi.ttf) to my $HOME/.fonts directory. Under Windows, Arial is returned once with the test program. Under Wine it is returned ten(!) times. -- 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 17

[Bug 14350] New: SetClassLongW() to subclass window EDIT control Problem
by wine-bugs＠winehq.org May 18, 2021

May 18, 2021

http://bugs.winehq.org/show_bug.cgi?id=14350 Summary: SetClassLongW() to subclass window EDIT control Problem Product: Wine Version: 1.0.0 Platform: PC OS/Version: Linux Status: UNCONFIRMED Severity: minor Priority: P2 Component: user32 AssignedTo: wine-bugs(a)winehq.org ReportedBy: hongbo(a)njstar.com CC: hongbo(a)njstar.com Created an attachment (id=14649) --> (http://bugs.winehq.org/attachment.cgi?id=14649) MSVC6 MFC test project with exe in debug dir Summary: SetClassLongW(hEditWnd, GCL_WNDPROC, (DWORD)EditWndProc) behave different on WINE 1.0 from Windows. Any EDIT window created after will be ANSI window on Windows, but Unicode Windows on Wine. How To Repeat: In the attached VC6 MFC ANSI App project, SetClassLongW is called in ------------------------------------------------------------------- static WNDPROC lpfnOldEditWndProc=NULL; LONG FAR PASCAL EditWndProc(HWND hWnd, WORD Message, WORD wParam, LONG lParam) { return CallWindowProcW((WNDPROC)lpfnOldEditWndProc, hWnd, Message, wParam,lParam); } int CDialogTestView::OnCreate(LPCREATESTRUCT lpCreateStruct) { if (CView::OnCreate(lpCreateStruct) == -1) return -1; HWND hEditWnd1 = CreateWindowEx(0,"EDIT", "E1",WS_POPUP,0, 0, 1, 1,NULL,NULL,NULL,NULL); lpfnOldEditWndProc =(WNDPROC)SetClassLongW(hEditWnd1, GCL_WNDPROC, (DWORD)EditWndProc); // TODO: Add your specialized creation code here return 0; } ------------------------------------------------------------------- There is a EDIT control on the About dialogbox, when click 'OK' on window the program report the Edit Box is NOT unicode Window. But Under Wine 1.0, It says 'the Edit Box IS unicode Window'. To Test: run the EXE in debug folder, Help - About DialogTest - click 'OK'. -- 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 18

[Bug 12891] New: ntdll: exception.ok test fails in PC-BSD but not Linux
by wine-bugs＠winehq.org May 18, 2021

May 18, 2021

http://bugs.winehq.org/show_bug.cgi?id=12891 Summary: ntdll: exception.ok test fails in PC-BSD but not Linux Product: Wine Version: 0.9.60 Platform: PC OS/Version: FreeBSD Status: NEW Keywords: testcase Severity: minor Priority: P2 Component: ntdll AssignedTo: wine-bugs(a)winehq.org ReportedBy: austinenglish(a)gmail.com Created an attachment (id=12597) --> (http://bugs.winehq.org/attachment.cgi?id=12597) +relay,+seh,+ntdll in git exception.c:602: Test failed: expected 4 exceptions, got 0 fixme:seh:fpe_handler untested SIMD exception: 0. Might not work correctly *** Error code 1 +relay,+seh,+ntdll log attached -- 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 10

[Bug 36282] New: valgrind shows an unitialized write in ieframe/tests/ie.c
by wine-bugs＠winehq.org May 16, 2021

May 16, 2021

https://bugs.winehq.org/show_bug.cgi?id=36282 Bug ID: 36282 Summary: valgrind shows an unitialized write in ieframe/tests/ie.c Product: Wine Version: 1.3.13 Hardware: x86 OS: Linux Status: NEW Keywords: download, source, testcase Severity: minor Priority: P2 Component: ieframe Assignee: wine-bugs(a)winehq.org Reporter: austinenglish(a)gmail.com ==16468== Thread 6: ==16468== Syscall param write(buf) points to uninitialised byte(s) ==16468== at 0x4EA8262B: ??? (in /usr/lib/libpthread-2.18.so) ==16468== by 0x7BC46BB8: NtWriteFile (file.c:1073) ==16468== by 0x7B83E0AA: WriteFile (file.c:563) ==16468== by 0x5A01BB8: rpcrt4_conn_np_write (rpc_transport.c:444) ==16468== by 0x59F6B27: rpcrt4_conn_write (rpc_binding.h:189) ==16468== by 0x59F8AA9: RPCRT4_SendWithAuth (rpc_message.c:849) ==16468== by 0x59F9BA0: RPCRT4_Send (rpc_message.c:1245) ==16468== by 0x59FCD02: process_request_packet (rpc_server.c:464) ==16468== by 0x59FCF82: RPCRT4_process_packet (rpc_server.c:519) ==16468== by 0x59FD115: RPCRT4_worker_thread (rpc_server.c:541) ==16468== by 0x7BC909EA: worker_thread_proc (threadpool.c:110) ==16468== by 0x7BC87087: ??? (signal_i386.c:2571) ==16468== by 0x7BC870D0: call_thread_func (signal_i386.c:2630) ==16468== by 0x7BC87065: ??? (signal_i386.c:2571) ==16468== by 0x7BC8E4B0: start_thread (thread.c:428) ==16468== by 0x4EA7BD89: start_thread (in /usr/lib/libpthread-2.18.so) ==16468== by 0x4E95CA0D: clone (in /usr/lib/libc-2.18.so) ==16468== Address 0x47f9cf4 is 44 bytes inside a recently re-allocated block of size 92 alloc'd ==16468== at 0x7BC4C735: notify_alloc (heap.c:255) ==16468== by 0x7BC50F79: RtlAllocateHeap (heap.c:1716) ==16468== by 0x59F88E9: RPCRT4_SendWithAuth (rpc_message.c:808) ==16468== by 0x59F9BA0: RPCRT4_Send (rpc_message.c:1245) ==16468== by 0x59FCD02: process_request_packet (rpc_server.c:464) ==16468== by 0x59FCF82: RPCRT4_process_packet (rpc_server.c:519) ==16468== by 0x59FD115: RPCRT4_worker_thread (rpc_server.c:541) ==16468== by 0x7BC909EA: worker_thread_proc (threadpool.c:110) ==16468== by 0x7BC87087: ??? (signal_i386.c:2571) ==16468== by 0x7BC870D0: call_thread_func (signal_i386.c:2630) ==16468== by 0x7BC87065: ??? (signal_i386.c:2571) ==16468== by 0x7BC8E4B0: start_thread (thread.c:428) ==16468== by 0x4EA7BD89: start_thread (in /usr/lib/libpthread-2.18.so) ==16468== by 0x4E95CA0D: clone (in /usr/lib/libc-2.18.so) ==16468== Uninitialised value was created by a client request ==16468== at 0x7BC4C52C: mark_block_uninitialized (heap.c:209) ==16468== by 0x7BC4C697: initialize_block (heap.c:240) ==16468== by 0x7BC50FA3: RtlAllocateHeap (heap.c:1717) ==16468== by 0x501E635: IMalloc_fnAlloc (ifs.c:186) ==16468== by 0x501EF50: CoTaskMemAlloc (objidl.h:1236) ==16468== by 0x505C935: RemUnknown_RemQueryInterface (stubmanager.c:690) ==16468== by 0x50661C6: IRemUnknown_RemQueryInterface_Stub (dcom_p.c:369) ==16468== by 0x59CA5BF: CStdStubBuffer_Invoke (cstub.c:477) ==16468== by 0x503D280: RPC_ExecuteCall (objidl.h:9086) ==16468== by 0x4FFC1ED: apartment_wndproc (compobj.c:1460) ==16468== by 0x548C509: ??? (winproc.c:173) ==16468== by 0x548C67E: call_window_proc (winproc.c:244) ==16468== by 0x548E817: WINPROC_call_window (winproc.c:900) ==16468== by 0x544FBE4: DispatchMessageW (message.c:4019) ==16468== by 0x4793F60: IEWinMain (iexplore.c:1061) ==16468== by 0x472CA34: wWinMain (main.c:84) ==16468== by 0x472CD21: wmain (exe_wmain.c:51) ==16468== by 0x472CBFC: __wine_spec_exe_wentry (exe_wentry.c:36) ==16468== by 0x7B864247: ??? (process.c:1045) ==16468== by 0x7B864394: start_process (process.c:1097) ==16468== -- 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 28831] New: secur32/tests/schannel.ok: occasional uninitialised memory read in _gnutls_recv_int() in test_communication()?
by wine-bugs＠winehq.org May 16, 2021

May 16, 2021

http://bugs.winehq.org/show_bug.cgi?id=28831 Bug #: 28831 Summary: secur32/tests/schannel.ok: occasional uninitialised memory read in _gnutls_recv_int() in test_communication()? Product: Wine Version: 1.3.30 Platform: x86 OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: secur32 AssignedTo: wine-bugs(a)winehq.org ReportedBy: dank(a)kegel.com Classification: Unclassified I don't see this on every run, but sometimes Valgrind complains: Conditional jump or move depends on uninitialised value(s) at _gnutls_recv_int (in /usr/lib/i386-linux-gnu/libgnutls.so.26.16.14) by ??? (in /usr/lib/i386-linux-gnu/libgnutls.so.26.16.14) by ??? (in /usr/lib/i386-linux-gnu/libgnutls.so.26.16.14) by ??? (in /usr/lib/i386-linux-gnu/libgnutls.so.26.16.14) by schan_InitializeSecurityContextW (schannel.c:728) by schan_InitializeSecurityContextA (schannel.c:775) by InitializeSecurityContextA (wrapper.c:282) by test_communication (schannel.c:644) by func_schannel (schannel.c:773) by run_test (test.h:556) by main (test.h:624) Uninitialised value was created by a client request by mark_block_uninitialized (heap.c:208) by initialize_block (heap.c:239) by RtlAllocateHeap (heap.c:1703) by init_buffers (schannel.c:496) by test_communication (schannel.c:633) by func_schannel (schannel.c:773) by run_test (test.h:556) by main (test.h:624) -- 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 4

[Bug 37849] New: Wine causing color temperature setting(redshift) not working for a short time
by wine-bugs＠winehq.org May 14, 2021

May 14, 2021

https://bugs.winehq.org/show_bug.cgi?id=37849 Bug ID: 37849 Summary: Wine causing color temperature setting(redshift) not working for a short time Product: Wine Version: 1.7.32 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: anonymous_temp_mail(a)yahoo.co.jp Distribution: --- [Bug] While using redshift(a display color temperature) and entered the sunset time, if then use wine to start a GUI program, the current display color temperature will be reset for several seconds and return back. [Reproducer] 1. Start redshift and ensure it entered sunset color. Better use -l LAT:LON to fake that you are in somewhere sun is already set. Default redshift will use randr backend to reset the color temp, so no extra setting is needed to reproduce. And default redshift will use 5500K for daytime and 3500K for nighttime, so the color temp change will be quite obvious. 2. Start wine GUI program I'm using Startcraft2 an galgame called "月に寄り添う乙女の作法2", both of them cause the display color temperature reset for several seconds. [Version] wine: 1.7.33 redshift: 1.9.1 -- 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

[Bug 33031] New: Microsoft Windows Driver Development Kit (win7 ddk/wdk) fails
by wine-bugs＠winehq.org May 13, 2021

May 13, 2021

http://bugs.winehq.org/show_bug.cgi?id=33031 Bug #: 33031 Summary: Microsoft Windows Driver Development Kit (win7 ddk/wdk) fails Product: Wine Version: 1.3.28 Platform: x86 URL: https://www.microsoft.com/en-us/download/details.aspx? id=11800 OS/Version: Linux Status: NEW Keywords: download, Installer Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs(a)winehq.org ReportedBy: austinenglish(a)gmail.com Classification: Unclassified Prerequisite: winetricks mfc42 Mount the iso, run kitsetup.exe. I chose to install 'Build Environments' (a subset of 'Full Development Environment'). This fails: wine: Unhandled page fault on read access to 0x84d5c3a5 at address 0x100228d (thread 0027), starting debugger... ... Backtrace: =>0 0x0100228d in eula (+0x228d) (0x0033f708) 1 0x0100236b in eula (+0x236a) (0x0033f790) 2 0x5f8019d1 in mfc42u (+0x19d0) (0x0033f7b0) 3 0x5f80195a in mfc42u (+0x1959) (0x0033f810) 4 0x5f8018e2 in mfc42u (+0x18e1) (0x0033f82c) 5 0x5f8018a1 in mfc42u (+0x18a0) (0x0033f858) the backtrace makes me suspect riched20, trying native gives: wine: Call from 0x7b83f2a9 to unimplemented function msls31.dll.LssbFDonePresSubline, aborting which may be a future problem, but belongs in a separate bug. austin(a)aw25 ~ $ sha1sum GRMWDK_EN_7600_1.ISO du -hde6abdb8eb4e08942add4aa270c763ed4e3d8242 GRMWDK_EN_7600_1.ISO austin(a)aw25 ~ $ du -h GRMWDK_EN_7600_1.ISO 620M GRMWDK_EN_7600_1.ISO austin(a)aw25 ~ $ wine --version wine-1.5.24-74-g5069ad7 luckily, I only wanted the headers, and running: $ msiexec /i wdk/headers.msi installs the headers fine. -- 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 9

[Bug 42675] New: [Overwatch] Phantom mouse input / view pulled up to ceiling
by wine-bugs＠winehq.org May 12, 2021

May 12, 2021

https://bugs.winehq.org/show_bug.cgi?id=42675 Bug ID: 42675 Summary: [Overwatch] Phantom mouse input / view pulled up to ceiling Product: Wine Version: unspecified Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: pimmesselink(a)gmail.com Distribution: --- I am using PlayOnLinux on Ubuntu 16.10 and compiled wine with Overwatch patches from source, first awesie's then gamax92's. It seems to run fine but every time I'm given control of the character and not touching the mouse, the viewpoint is pulled up to the ceiling, as if there is some phantom mouse input. See https://www.youtube.com/watch?v=UMiKAl1tR4s. If I then pull the mouse towards me fast enough I can overcome phantom mouse input and pull the view down. Until I stop moving the mouse and the view is pulled up again. Moving the mouse left or right works fine. Things I have tried: If I change the input settings to inverted, the phantom mouse input pulls the view to the floor. My standard mouse is a Roccat Kone XTD. I have tried to change the DPI and polling rate. I have also tried without the drivers installed or plugging the mouse into different USB ports. And I have tried replacing it with an old Razer Diamondback. All had the same result. I have tried the Wine config options "Automatically capture the mouse in full-screen windows", "Emulate a virtual desktop" and "mouse warp override default/enable/disable/force". Within the wine prefix I have ran Xpadder and XInputTest and neither found any gamepads or joysticks connected (I have neither). My keyboard is a standard, non USB keyboard. Apart from my mouse the only other peripherals are a Logitech webcam and a Sound Blaster Tactic(3D) Alpha headset. Unplugging them does not make a difference. My PC is a desktop, not a laptop. Raw mouse input works fine in Natural Selection 2 natively under linux. My /dev/input consists of by-id event0 event10 event12 event14 event16 event2 event4 event6 event8 mice by-path event1 event11 event13 event15 event17 event3 event5 event7 event9 mouse0 In Overwatch I use Borderless Windowed and alt-tabbing out and back in does not help. If I use Fullscreen and alt-tab out the 3D view becomes very small in the top left corner and I can't alt-tab back to it. Not sure if it is relevant, but I am using a GeForce 460 GTX with the latest nvidia-378 drivers. Motherboard is a Asus P8Z77-V Pro. For completeness, this is how I compiled Wine: cd /home/<user>/Documents/Programming/wine64-build make clean make distclean export CFLAGS="-mtune=native -O3 -pipe -fstack-protector-strong" export CPPFLAGS="-mtune=native -O3 -pipe -fstack-protector-strong" ../wine-overwatch/configure --enable-win64 --prefix=/home/<user>/.PlayOnLinux/wine/linux-amd64/2.3-overwatch/ make -j4 sudo lxc-start -n my32bitbox sudo lxc-attach -n my32bitbox cd /home/<user>/Documents/Programming/wine32-tools make clean make distclean export CFLAGS="-mtune=native -O3 -pipe -fstack-protector-strong" export CPPFLAGS="-mtune=native -O3 -pipe -fstack-protector-strong" ../wine-overwatch/configure make -j4 cd ../wine32-build make clean make distclean ../wine-overwatch/configure --with-win64=../wine64-build --with-wine-tools=../wine32-tools --prefix=/home/<user>/.PlayOnLinux/wine/linux-x86/2.3-overwatch/ make -j4 make install shutdown -h now sudo rm -rf /home/<user>/.PlayOnLinux/wine/linux-x86/2.3-overwatch/* cd /home/<user>/Documents/Programming/wine32-build sudo make install cd /home/<user>/Documents/Programming/wine64-build sudo make install cd /home/<user>/.PlayOnLinux/wine/linux-amd64/2.3-overwatch/ ln -s ../../linux-x86/2.3-overwatch/lib lib -- 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 45

[Bug 15956] New: "make test" doesn't always leave screen in same video mode as it started in
by wine-bugs＠winehq.org May 12, 2021

May 12, 2021

http://bugs.winehq.org/show_bug.cgi?id=15956 Summary: "make test" doesn't always leave screen in same video mode as it started in Product: Wine Version: 1.1.7 Platform: Other OS/Version: other Status: NEW Severity: normal Priority: P2 Component: testcases AssignedTo: wine-bugs(a)winehq.org ReportedBy: dank(a)kegel.com Some tests don't properly reset the video mode, even if they don't crash. I'm gathering a list of culprits and will post it here. -- 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 4