September 2018 - Wine-Bugs - list.winehq.org

[Bug 37834] New: RtlSetCurrentDirectory_U prepends "UNC\" for network paths; the resulting path is invalid
by wine-bugs＠winehq.org Nov. 29, 2018

Nov. 29, 2018

https://bugs.winehq.org/show_bug.cgi?id=37834 Bug ID: 37834 Summary: RtlSetCurrentDirectory_U prepends "UNC\" for network paths; the resulting path is invalid Product: Wine Version: 1.7.33 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: minor Priority: P2 Component: ntdll Assignee: wine-bugs(a)winehq.org Reporter: atakama0512(a)yahoo.com Distribution: --- Created attachment 50382 --> https://bugs.winehq.org/attachment.cgi?id=50382 cmd.exe output showing how UNC\ gets prepended when running cd \\server\share Calling RtlSetCurrentDirectory_U with a network share argument like "\\server\share" sets the current directory to "UNC\server\share". Subsequent calls to GetCurrentDirectory will return the modified path. The resulting path is invalid and unusable: attempting to read the directory or writing a file to it results in ERROR_PATH_NOT_FOUND. The bug can be reproduced using the internal command prompt, as seen in the attachment. >From what I've gathered from the source code, RtlSetCurrentDirectory_U calls RtlDosPathNameToNtPathName_U, which prepends UncPfxW to the supplied path. The last edits to dlls/ntdll/path.c were made in 2003-2004. I stumbled across this behavior while investigating why Total Commander cannot copy files to network shares. The sequence of calls in Total Commander is: SetCurrentDirectory to the network path, then GetCurrentDirectory (which returns the modified path in the form "UNC\server\share") and then it builds the destination parameter by concatenating this path with the destination file name. This gets passed to CopyFileEx, which fails with ERROR_PATH_NOT_FOUND because of the "UNC\" prefix. -- 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 44369] New: cmd's %0 path variables (e.g. %~dp0) wrong inside subroutine call
by wine-bugs＠winehq.org Nov. 29, 2018

Nov. 29, 2018

https://bugs.winehq.org/show_bug.cgi?id=44369 Bug ID: 44369 Summary: cmd's %0 path variables (e.g. %~dp0) wrong inside subroutine call Product: Wine Version: 3.0-rc6 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: cmd Assignee: wine-bugs(a)winehq.org Reporter: ryan.prichard(a)gmail.com Distribution: --- Test program (C:\bin\workaround.cmd): @echo off call :fix exit /b :fix echo [%0] [%~d0] [%~p0] [%~n0] [%~x0] exit /b Expected behavior (tested on XP): C:\>set PATH=C:\bin;%PATH% C:\>workaround.cmd [:fix] [C:] [\bin\] [workaround] [.cmd] C:\>"workaround.cmd" [:fix] [C:] [\bin\] [workaround] [.cmd] C:\>"w"or"k"aroun"d.c"md [:fix] [C:] [\bin\] [workaround] [.cmd] Behavior on wine-3.0-rc6: c:\>set PATH=C:\bin;%PATH% c:\>workaround.cmd [:fix] [c:] [\] [:fix] [] c:\>"workaround.cmd" [:fix] [c:] [\] [:fix] [] c:\>"w"or"k"aroun"d.c"md [:fix] [c:] [\] [:fix] [] Background: The Android NDK generates wrapper batch files that invoke clang.exe. If the wrapper is invoked from the PATH using double-quotes, then %~dp0 is wrong (when using Windows CMD). A workaround I've found is to evaluate %~dp0 in a subroutine in the batch file. This workaround works with Windows CMD, but breaks with Wine CMD. (It expands to the working directory instead.) See https://github.com/android-ndk/ndk/issues/616#issuecomment-358117090. FWIW: Wine CMD doesn't have the same %~dp0 bugginess as Windows CMD that motivated the GitHub issue I linked to. Wine handles all of the example command lines on that GitHub comment consistently (aside from case differences in the drive letter or the filename extension). If Wine tries to match Windows bug-for-bug, then maybe this too should change. -- 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 8

[Bug 45530] New: No$Gba crash when open.
by wine-bugs＠winehq.org Nov. 29, 2018

Nov. 29, 2018

https://bugs.winehq.org/show_bug.cgi?id=45530 Bug ID: 45530 Summary: No$Gba crash when open. Product: Wine Version: 3.13 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: wzhy90(a)gmail.com Distribution: --- Created attachment 61927 --> https://bugs.winehq.org/attachment.cgi?id=61927 no$gba ver.2.9a from http://problemkaputt.de/gba.htm Host is linux mint 19, it works on mint 18.3 but after I do a fresh install to mint 19 it crash. I have try stable or devel version, both crash when open. -- 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 43036] New: SetNamedPipeHandleState returns ERROR_ACCESS_DENIED when setting PIPE_NOWAIT
by wine-bugs＠winehq.org Nov. 29, 2018

Nov. 29, 2018

https://bugs.winehq.org/show_bug.cgi?id=43036 Bug ID: 43036 Summary: SetNamedPipeHandleState returns ERROR_ACCESS_DENIED when setting PIPE_NOWAIT Product: Wine Version: 2.7 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: dan-wine(a)berrange.com Distribution: --- In GNULIB, there is code which tries to create an implementation of the POSIX pipe() method for Windows, using the methods _pipe, _get_osfhandle and SetNamedPipeHandleState. This code works when run on real windows systems (i've tried Win2k8 myself), but fails when run under Wine (I've tested versions 2.5 and 2.7 in Fedora), with ERROR_ACCESS_DENIED from the SetNamedPipeHandleState method Since GNULIB code is fairly complex to follow I created this short demo program: $ cat > demo.c <<EOF #include <sys/unistd.h> #include <stdio.h> #include <windows.h> int nonblock(int fd) { HANDLE h = (HANDLE)_get_osfhandle(fd); char errbuf[1024]; DWORD state; if (GetNamedPipeHandleState (h, &state, NULL, NULL, NULL, NULL, 0) == 0) { FormatMessage(FORMAT_MESSAGE_FROM_SYSTEM, NULL, GetLastError(), MAKELANGID(LANG_NEUTRAL, SUBLANG_DEFAULT), errbuf, 1024, NULL); fprintf(stderr, "Failed GetNamedPipeHandleState (%lu) %s", GetLastError(), errbuf); return -1; } if ((state & PIPE_NOWAIT) != 0) { return 0; } state |= PIPE_NOWAIT; if (SetNamedPipeHandleState (h, &state, NULL, NULL) == 0) { FormatMessage(FORMAT_MESSAGE_FROM_SYSTEM, NULL, GetLastError(), MAKELANGID(LANG_NEUTRAL, SUBLANG_DEFAULT), errbuf, 1024, NULL); fprintf(stderr, "Failed SetNamedPipeHandleState (%lu) %s", GetLastError(), errbuf); return -1; } return 0; } int main(int argc, char **argv) { int fd[2]; if (_pipe(fd, 4096, 0) < 0) { fprintf(stderr, "Failed to create pipe\n"); return -1; } if (nonblock(fd[0]) < 0){ fprintf(stderr, "Could not set non-blocking on fd[0]\n"); return -1; } if (nonblock(fd[1]) < 0){ fprintf(stderr, "Could not set non-blocking on fd[1]\n"); return -1; } fprintf(stderr, "Created non-blocking pipe pair\n"); } EOF $ i686-w64-mingw32-gcc -Wall -mconsole -o pipe.exe pipe.c $ ./pipe.exe fixme:winediag:start_process Wine Staging 2.7 is a testing version containing experimental patches. fixme:winediag:start_process Please mention your exact version when filing bug reports on winehq.org. fixme:sync:GetNamedPipeHandleStateW 0x2c 0x61f8f8 (nil) (nil) (nil) (nil) 0: semi-stub Failed SetNamedPipeHandleState (5) Access denied. Could not set non-blocking on fd[0] Strangely, this only seems to fail on fd[0] - if I let it run on fd[1] it will work. The original GNULIB code I hit the problem on is here: http://git.savannah.gnu.org/cgit/gnulib.git/tree/lib/pipe2.c http://git.savannah.gnu.org/cgit/gnulib.git/tree/lib/nonblocking.c The Fedora Wine 2.7 package I tested on was built with these sources: SHA512 (wine-2.7.tar.xz) = 1e61b9a4aa1f5f42fb27d11d5254a9ba90f348ad9c4d1ddd4b5da47cd7de638290a20accf7447db9c0e4ced4c2144497cdf5fc906a5eac60e923dabb61f65d3a SHA512 (wine-2.7.tar.xz.sign) = b03f4376b10bd8ea66e5e6fc0862a4f948e009862a374677c326744b31c9d9fdcf1efd3b789149fcb6fe617f9c75b1b47d61f884e06e8c0fe16633a99911b667 SHA512 (wine-staging-2.7.tar.gz) = 0abc89af701ae1b95c0eb08e72894c7bc40bdfe792e05b8af9282eab8407bb90b7dfcd4eb3a193a88759ce5d6ea6c2aa9696cac2d744f543c92529bb0d2636ee -- 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 8

[Bug 45529] New: Custom color scheme applied but wine not respecting current text color of scheme on window columns and statusbar.
by wine-bugs＠winehq.org Nov. 29, 2018

Nov. 29, 2018

https://bugs.winehq.org/show_bug.cgi?id=45529 Bug ID: 45529 Summary: Custom color scheme applied but wine not respecting current text color of scheme on window columns and statusbar. Product: Wine Version: 3.13 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: trivial Priority: P2 Component: uxtheme Assignee: wine-bugs(a)winehq.org Reporter: mt_talpur(a)yahoo.com Distribution: --- Created attachment 61923 --> https://bugs.winehq.org/attachment.cgi?id=61923 shows current issue with color scheme applied. When changing the color scheme via registry it does change overall looks of window elements but the text of window columns and statusbar remains black no matter what color scheme is applied though it should respect the currently used text color On windows this does not happen. -- 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 14

[Bug 42577] New: A regression with Far manager
by wine-bugs＠winehq.org Nov. 29, 2018

Nov. 29, 2018

https://bugs.winehq.org/show_bug.cgi?id=42577 Bug ID: 42577 Summary: A regression with Far manager Product: Wine Version: 2.2 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: ozerski(a)list.ru Distribution: --- A regression: Far manager (both 32-bit and 64-bit) running under wine (using wineconsole) crashs due any attemt to change drive using menu or Alt+Fn -- 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 20

[Bug 43072] New: Sonos Controller Program Error with Wine 2.O
by wine-bugs＠winehq.org Nov. 29, 2018

Nov. 29, 2018

https://bugs.winehq.org/show_bug.cgi?id=43072 Bug ID: 43072 Summary: Sonos Controller Program Error with Wine 2.O Product: Wine Version: 2.0 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: critical Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: john.peters(a)gmx.fr Distribution: --- Created attachment 58255 --> https://bugs.winehq.org/attachment.cgi?id=58255 Created via Program Error window Sonos controller worked with Wine 1.8 with Ubuntu 16.04. I had this bug problem but solved it by going from 1.9 back to 1.8. My wine version has been updated to 2.0 and I now have the old problem back. I can't remember how I downgraded to 1.8. I get the program error window almost immediately on starting SONOS. Can you tell me if the problem exists elsewhere and how I can go back to Wine 1.8? Thanks -- 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 38561] New: Painkiller crashes when loading a save
by wine-bugs＠winehq.org Nov. 27, 2018

Nov. 27, 2018

https://bugs.winehq.org/show_bug.cgi?id=38561 Bug ID: 38561 Summary: Painkiller crashes when loading a save Product: Wine Version: 1.7.41 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: fremenzone(a)poczta.onet.pl Distribution: --- Created attachment 51440 --> https://bugs.winehq.org/attachment.cgi?id=51440 Crash log Loading a saved game in Painkiller: Black Edition bought on GOG.com does not work. Loading freezes at some point (always the same) and pressing a key results in a crash. Wine version: wine-1.7.41-197-gf7e0927. Attaching log. -- 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 36763] New: Rogue Squadron 3D 1.3: Crashes with game resolutions above 640x480
by wine-bugs＠winehq.org Nov. 26, 2018

Nov. 26, 2018

http://bugs.winehq.org/show_bug.cgi?id=36763 Bug ID: 36763 Summary: Rogue Squadron 3D 1.3: Crashes with game resolutions above 640x480 Product: Wine Version: 1.7.20 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: dev(a)ovocean.com Created attachment 48821 --> http://bugs.winehq.org/attachment.cgi?id=48821 Console output. Crash after line 19 If choosing a resolution above 640x480 in the game settings menu, the game crashes at the start of missions, when the set resolution gets used. System used: Nvidia GeForce GTX 660M with Nvidia drivers v331.38 Xubuntu 14.04 -- 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 19

[Bug 36113] New: d3d9/device test has a leak
by wine-bugs＠winehq.org Nov. 26, 2018

Nov. 26, 2018

https://bugs.winehq.org/show_bug.cgi?id=36113 Bug ID: 36113 Summary: d3d9/device test has a leak Product: Wine Version: 1.7.17 Hardware: x86 OS: Linux Status: NEW Keywords: download, source, testcase Severity: minor Priority: P2 Component: directx-d3d Assignee: wine-bugs(a)winehq.org Reporter: austinenglish(a)gmail.com ==17962== 34 bytes in 2 blocks are definitely lost in loss record 309 of 1,416 ==17962== at 0x4006B11: malloc (in /usr/lib/valgrind/vgpreload_memcheck-x86-linux.so) ==17962== by 0x4E8E2B27: strdup (in /usr/lib/libc-2.18.so) ==17962== by 0x67C5806: ??? (in /usr/lib/dri/i965_dri.so) ==17962== by 0x67CE8F0: ??? (in /usr/lib/dri/i965_dri.so) ==17962== by 0x67D7587: ??? (in /usr/lib/dri/i965_dri.so) ==17962== by 0x6661EC8: ??? (in /usr/lib/dri/i965_dri.so) ==17962== by 0x63BDC15: shared_dispatch_stub_482 (in /usr/lib/libglapi.so.0.0.0) ==17962== by 0x4F5BAD5: shader_glsl_compile (glsl_shader.c:330) ==17962== by 0x4F68B3F: shader_glsl_generate_vshader (glsl_shader.c:4588) ==17962== by 0x4F693E4: find_glsl_vshader (glsl_shader.c:4767) ==17962== by 0x4F6C88A: set_glsl_shader_program (glsl_shader.c:5793) ==17962== by 0x4F6E096: shader_glsl_select (glsl_shader.c:6149) ==17962== by 0x4F2EF12: context_apply_draw_state (context.c:2997) ==17962== by 0x4F581D6: draw_primitive (drawprim.c:664) ==17962== by 0x4F2F685: wined3d_cs_exec_draw (cs.c:291) ==17962== by 0x4F30A33: wined3d_cs_st_submit (cs.c:893) ==17962== by 0x4F2F6F2: wined3d_cs_emit_draw (cs.c:308) ==17962== by 0x4F3EFC3: wined3d_device_draw_primitive (device.c:3265) ==17962== by 0x4973D29: d3d9_device_DrawPrimitive (device.c:2101) ==17962== by 0x4C27D7C: test_null_stream (device.c:1952) ==17962== -- 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