February 2023 - Wine-bugs - list.winehq.org

[Bug 53926] New: New typelib marshaller depends on IID_IDispatch support from target interface
by WineHQ Bugzilla 20 Apr '23

20 Apr '23

https://bugs.winehq.org/show_bug.cgi?id=53926 Bug ID: 53926 Summary: New typelib marshaller depends on IID_IDispatch support from target interface Product: Wine Version: 7.21 Hardware: x86-64 OS: Linux Status: NEW Keywords: regression, testcase Severity: normal Priority: P2 Component: oleaut32 Assignee: wine-bugs(a)winehq.org Reporter: dmitry(a)baikal.ru CC: z.figura12(a)gmail.com Regression SHA1: 077b4391d442e927a2a59b5afb244355b0634aaa Distribution: --- Created attachment 73481 --> https://bugs.winehq.org/attachment.cgi?id=73481 test (source + binary) New typelib marshaller depends on IID_IDispatch support from target interface, however applications created with Delphi/.Net don't always follow this implication. This is a regression, and it starts from commit 077b4391d442e927a2a59b5afb244355b0634aaa Author: Zebediah Figura <z.figura12(a)gmail.com> Date: Fri Nov 9 18:02:00 2018 -0600 oleaut32: Rip out the old typelib marshaller. Signed-off-by: Zebediah Figura <z.figura12(a)gmail.com> Signed-off-by: Marcus Meissner <marcus(a)jet.franken.de> Signed-off-by: Huw Davies <huw(a)codeweavers.com> Signed-off-by: Alexandre Julliard <julliard(a)winehq.org> Old typelib marshaller didn't have a requirement that an interface a proxy/ stub being created for must support IID_IDispatch. Attached testcase replicates what I can deduce from the logs, and runs without failures under wine-3.0.5, but fails in several places under newer versions of Wine. The test produces 2 extra IUnknown::QueryInterface() calls under wine-3.0.5 during ::CreateStub() invocation, however it's pretty simple to fix following a FIXME in oleaut32 and stub manager in ole32, after that the test output from wine-3.0.5 matches what I observe under windows 10. Although the test uses IConnectionPoint::Advise() to reproduce the bug (as the application that I have here does) it seems that any client<=>server remote call reliably replicates the problem, and source of the bug leads to CreateStub/CreateProxy limitations. -- 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 54234] New: vbscript fails to compile when colon follows Else in If...Else
by WineHQ Bugzilla 20 Apr '23

20 Apr '23

https://bugs.winehq.org/show_bug.cgi?id=54234 Bug ID: 54234 Summary: vbscript fails to compile when colon follows Else in If...Else Product: Wine Version: 7.21 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: vbscript Assignee: wine-bugs(a)winehq.org Reporter: jsm174(a)gmail.com Distribution: --- I just ran into a script where there is a ':' separator on a new line after an ELSE: if IsEmpty(BallData(idx).id) and aBall.VelY < -12 then 'if tip hit with no collected data, do vel correction anyway if PSlope(aBall.x, FlipperStart, 0, FlipperEnd, 1) > 1.1 then 'adjust plz VelCoef = LinearEnvelope(5, VelocityIn, VelocityOut) if partialflipcoef < 1 then VelCoef = PSlope(partialflipcoef, 0, 1, 1, VelCoef) if Enabled then aBall.Velx = aBall.Velx*VelCoef'VelCoef if Enabled then aBall.Vely = aBall.Vely*VelCoef'VelCoef end if Else : VelCoef = LinearEnvelope(BallPos, VelocityIn, VelocityOut) if Enabled then aBall.Velx = aBall.Velx*VelCoef if Enabled then aBall.Vely = aBall.Vely*VelCoef end if I was able to fix this by changing: Else_opt : /* empty */ { $$ = NULL; } | tELSE NL StatementsNl_opt { $$ = $3; } | tELSE StatementsNl_opt { $$ = $2; } to Else_opt : /* empty */ { $$ = NULL; } | tELSE StSep_opt StatementsNl_opt { $$ = $3; } | tELSE StatementsNl_opt { $$ = $2; } -- 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 54449] New: nethack crashes
by WineHQ Bugzilla 20 Apr '23

20 Apr '23

https://bugs.winehq.org/show_bug.cgi?id=54449 Bug ID: 54449 Summary: nethack crashes Product: Wine Version: 8.0 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: temp82(a)luukku.com Distribution: --- Created attachment 74000 --> https://bugs.winehq.org/attachment.cgi?id=74000 crash log wine 8.0 https://www.nethack.org/download/3.6.6/nethack-366-win-x86-1_25_2021.zip -- 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 9

[Bug 53274] New: adsldp:ldap - test_ParseDisplayName() sometimes fails to connect to the server
by WineHQ Bugzilla 20 Apr '23

20 Apr '23

https://bugs.winehq.org/show_bug.cgi?id=53274 Bug ID: 53274 Summary: adsldp:ldap - test_ParseDisplayName() sometimes fails to connect to the server Product: Wine Version: unspecified Hardware: x86-64 OS: Windows Status: NEW Severity: normal Priority: P2 Component: activeds Assignee: wine-bugs(a)winehq.org Reporter: fgouget(a)codeweavers.com adsldp:ldap - test_ParseDisplayName() sometimes fails to connect to the server: ldap.c:194: Test failed: 13: got 0x8007203a, expected 0 https://test.winehq.org/data/patterns.html#adsldp:ldap 0x8007203a == HRESULT_FROM_WIN32(ERROR_DS_SERVER_DOWN) There are two known cases where this error happened: * 2022-05-26 debian11 wow64 * 2022-06-23 debiant wow32 This may indeed correspond to times when the server was not accessible. However it is strange that this impacted Wine both times. -- 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 54030] New: Snagit needs Win32_Volume class ( 'select deviceid from win32_volume where driveletter =C:')
by WineHQ Bugzilla 20 Apr '23

20 Apr '23

https://bugs.winehq.org/show_bug.cgi?id=54030 Bug ID: 54030 Summary: Snagit needs Win32_Volume class ( 'select deviceid from win32_volume where driveletter =C:') Product: Wine Version: 7.22 Hardware: x86-64 URL: https://download.techsmith.com/snagit/releases/snagit. exe OS: Linux Status: NEW Keywords: dotnet, download Severity: normal Priority: P2 Component: wmi&wbemprox Assignee: wine-bugs(a)winehq.org Reporter: xerox.xerox2000x(a)gmail.com Distribution: Debian Created attachment 73598 --> https://bugs.winehq.org/attachment.cgi?id=73598 +wbemprox log (prerequisite: winetricks -q dotnet48 d3dcompiler_47 arial) The program stops instantly with: 03e4:fixme:wmic:wmain command line switch L"/namespace:'\\\\root\\virtualization\\v2'" not supported so I disabled wmic (google says it is deprecated in win10) but this might really deserve a seperate bugreport. After that there's a crash, which seems to be related to: 014c:trace:wbemprox:wbem_services_ExecQuery 0000000003CD4FB0, L"WQL", L"select deviceid from win32_volume where driveletter =\"C:\"", 0x30, 0000000000000000, 000000000011E1B0 After fixing up some hacky patch with bogus values (i don't know how to implement this properly) it starts into the loginscreen, so this seems enough to get it running. There`s also a query for L"SELECT * FROM WmiMonitorBasicDisplayParams" that seems missing, but this doesn't seem to affect the startup of the program. Attached log (+wbemprox) (with bogus hack applied) -- 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 54318] New: Rich Edit inserts newly composed text at wrong position when system IME composition ends while a selection is active
by WineHQ Bugzilla 20 Apr '23

20 Apr '23

https://bugs.winehq.org/show_bug.cgi?id=54318 Bug ID: 54318 Summary: Rich Edit inserts newly composed text at wrong position when system IME composition ends while a selection is active Product: Wine Version: 8.0-rc4 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: richedit Assignee: wine-bugs(a)winehq.org Reporter: jinoh.kang.kr(a)gmail.com CC: huw(a)codeweavers.com Distribution: --- When the user attempts to perform select-and-replace with newly composed text from system IME (fcitx5) as the replacement, Rich Edit inserts the new text at the _ending_ character position of the current selection, instead of the _starting_ character position. Steps to reproduce: 1. Activate fcitx5 (Hangul). 2. Enter the following text: "[ABCXYZ]" 3. Select the following text: "ABC" 4. Compose a character. Expected behaviour: The composed character should replace "ABC". Actual behaviour: The text "ABC" is erased, but the newly composed character is inserted after "Z", before "]". -- 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 54289] New: RtlCopyContext buffer overflow
by WineHQ Bugzilla 20 Apr '23

20 Apr '23

https://bugs.winehq.org/show_bug.cgi?id=54289 Bug ID: 54289 Summary: RtlCopyContext buffer overflow Product: Wine Version: 7.22 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: ntdll Assignee: wine-bugs(a)winehq.org Reporter: novae.harpist06(a)icloud.com Distribution: --- Created attachment 73857 --> https://bugs.winehq.org/attachment.cgi?id=73857 Reproducer See repro (attached) Windows output: src ContextFlags: 10005f dst ContextFlags: 10001f RtlCopyContext returned 00000000 Wine output: src ContextFlags: 10005f dst ContextFlags: 10001f src ContextFlags: 10005f dst ContextFlags: 10001f src ContextFlags: 10005f dst ContextFlags: 10001f src ContextFlags: 10005f dst ContextFlags: 10001f ... more of this src ContextFlags: 10005f dst ContextFlags: 10001f src ContextFlags: 10005f dst ContextFlags: 10001f src ContextFlags: 10005f dst ContextFlags: 10001f 010c:err:virtual:virtual_setup_exception stack overflow 1808 bytes addr 0x170025cbc stack 0x208f0 (0x20000-0x21000-0x120000) -- 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 54364] New: RtlGenRandom fails on systems with more than 128 cores
by WineHQ Bugzilla 20 Apr '23

20 Apr '23

https://bugs.winehq.org/show_bug.cgi?id=54364 Bug ID: 54364 Summary: RtlGenRandom fails on systems with more than 128 cores Product: Wine Version: 8.0-rc5 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: advapi32 Assignee: wine-bugs(a)winehq.org Reporter: tjeerd.bakker38(a)gmail.com Distribution: --- WinHttpSendRequest() fails with ERROR_OUTOFMEMORY when the WINHTTP_OPTION_UPGRADE_TO_WEB_SOCKET is set on systems with 256 cores. I've tracked it down to dlls/advapi32/crypt.c. RtlGenRandom is implemented by calling NtQuerySystemInformation(SystemInterruptInformation, ...) with a buffer sufficient for the data from at most 128 CPUs. The buffer size is determined by the #define in line 2393: #define MAX_CPUS 128 Changing this line to #define MAX_CPUS 256 make the problem go away for me, but the problem will probably return when someone builds a system with even more cores. I would suggest implementing RtlGenRandom() with a better source of randomness instead. -- 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 54581] New: SpeedCommander: SHELL32.dll.Shell_GetCachedImageIndexW
by WineHQ Bugzilla 20 Apr '23

20 Apr '23

https://bugs.winehq.org/show_bug.cgi?id=54581 Bug ID: 54581 Summary: SpeedCommander: SHELL32.dll.Shell_GetCachedImageIndexW Product: Wine Version: 8.0 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: peter.eszlari(a)gmail.com Distribution: --- https://www.speedproject.de https://www.sp-download.de/sc20/sc20_x64.exe During installation I the following message box: "SpeedCommander.exe has encountered a problem in module ntdll.dll and needs to close." On the command line: wine: Call from 7BC2BF40 to unimplemented function SHELL32.dll.Shell_GetCachedImageIndexW, aborting -- 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 36891] New: Polygon() draws different edges than in Windows
by wine-bugs＠winehq.org 20 Apr '23

20 Apr '23

https://bugs.winehq.org/show_bug.cgi?id=36891 Bug ID: 36891 Summary: Polygon() draws different edges than in Windows Product: Wine Version: 1.7.22 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: minor Priority: P2 Component: gdi32 Assignee: wine-bugs(a)winehq.org Reporter: sashoalm(a)gmail.com The Polygon() function does not draw in the same way in Wine as in Windows. While it looks very similar visually, there are some pixels at the edges that are different. Since the goal of Wine is to replicate Windows behavior perfectly, this deviation would be a bug. The difference is difficult to spot, but it can be noticed if you rapidly switch between the images, or if you compare them pixel by pixel with a program. I have made a test program that draws an inclined rectangle using XFORM and Polygon() into a memory DC, which I then save to a bitmap. The resulting bitmaps when ran under Windows and Wine are slightly different. Image under Windows: https://drive.google.com/file/d/0B9PGUhmmnsm1Q00waFJLSGtEeEU/edit?usp=shari… Image under Wine: https://drive.google.com/file/d/0B9PGUhmmnsm1X1ZVR1UtbDBPMHc/edit?usp=shari… This is the relevant code I used to draw the image. Note that it's the XFORM matrix that makes the rectangle be inclined (it would be parallel to the coordinate system without a world-transform): XFORM xform; memset(&xform, 0, sizeof(xform)); double angle = 15*(PI/180); // 15 degrees incline. xform.eM11 = cos(angle); xform.eM12 = sin(angle); xform.eM21 = -sin(angle); xform.eM22 = cos(angle); xform.eDx = 0; xform.eDy = 0; SetGraphicsMode(hdc, GM_ADVANCED); SetWorldTransform(hdc, &xform); HBRUSH hbr = CreateHatchBrush(HS_BDIAGONAL, RGB(255, 0, 0)); SelectObject(hdc, hbr); POINT p[4]={100,0,200,0,200,100,100,100}; Polygon(hdc, p, 4); I can provide the full test project if requested. Note: I have tested on both Windows 7 and Windows XP, and they produce exactly the same images, only Wine produces a different image. -- 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