wine-bugs April 2023

wine-bugs@winehq.org

1 participants
3455 discussions

[Bug 54457] New: vbscript memory leaks in interp_redim_preserve
by WineHQ Bugzilla 20 Apr '23

20 Apr '23

https://bugs.winehq.org/show_bug.cgi?id=54457 Bug ID: 54457 Summary: vbscript memory leaks in interp_redim_preserve 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: --- While porting Visual Pinball to cross platform, I started using XCode Instruments to find some outstanding memory leaks. Given the following code: Dim X() Redim preserve X(20) Redim preserve X(30) X(20) = 5 Redim preserve X(40) I noticed a few memory leaks which were not happening with interp_redim. 1) In `interp_redim`, after SafeArrayCreate, there is a free(bounds); In `interp_redim_preserve`, there is no free(bounds); after SafeArrayCreate when array == NULL || array->cDims = 0 2) The bounds is not free after `SafeArrayRedim` Reworking the function as follows fixes the memory leaks: static HRESULT interp_redim_preserve(exec_ctx_t *ctx) { BSTR identifier = ctx->instr->arg1.bstr; const unsigned dim_cnt = ctx->instr->arg2.uint; VARIANT *v; unsigned i; SAFEARRAYBOUND *bounds; SAFEARRAY *array; ref_t ref; HRESULT hres; TRACE("%s %u\n", debugstr_w(identifier), dim_cnt); hres = lookup_identifier(ctx, identifier, VBDISP_LET, &ref); if(FAILED(hres)) { FIXME("lookup %s failed: %08lx\n", debugstr_w(identifier), hres); return hres; } if(ref.type != REF_VAR) { FIXME("got ref.type = %d\n", ref.type); return E_FAIL; } v = ref.u.v; if(V_VT(v) == (VT_VARIANT|VT_BYREF)) { v = V_VARIANTREF(v); } if(!(V_VT(v) & VT_ARRAY)) { FIXME("ReDim Preserve not valid on type %d\n", V_VT(v)); return E_FAIL; } hres = array_bounds_from_stack(ctx, dim_cnt, &bounds); if(FAILED(hres)) return hres; array = V_ARRAY(v); if(array == NULL || array->cDims == 0) { /* can initially allocate the array */ array = SafeArrayCreate(VT_VARIANT, dim_cnt, bounds); free(bounds); if(!array) return E_OUTOFMEMORY; VariantClear(v); V_VT(v) = VT_ARRAY|VT_VARIANT; V_ARRAY(v) = array; return S_OK; } else if(array->cDims != dim_cnt) { /* can't otherwise change the number of dimensions */ TRACE("Can't resize %s, cDims %d != %d\n", debugstr_w(identifier), array->cDims, dim_cnt); return MAKE_VBSERROR(VBSE_OUT_OF_BOUNDS); } else { /* can resize the last dimensions (if others match */ for(i = 0; i+1 < dim_cnt; ++i) { if(array->rgsabound[array->cDims - 1 - i].cElements != bounds[i].cElements) { TRACE("Can't resize %s, bound[%d] %ld != %ld\n", debugstr_w(identifier), i, array->rgsabound[i].cElements, bounds[i].cElements); return MAKE_VBSERROR(VBSE_OUT_OF_BOUNDS); } } hres = SafeArrayRedim(array, &bounds[dim_cnt-1]); free(bounds); return hres; } } -- 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 54357] New: Spurious fixme message when calling ScrollWindow()
by WineHQ Bugzilla 20 Apr '23

20 Apr '23

https://bugs.winehq.org/show_bug.cgi?id=54357 Bug ID: 54357 Summary: Spurious fixme message when calling ScrollWindow() Product: Wine Version: 8.0-rc3 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: trivial Priority: P2 Component: user32 Assignee: wine-bugs(a)winehq.org Reporter: vz-wine(a)zeitlins.org Distribution: --- `ScrollWindow()` calls `NtUserScrollWindowEx()` with `SW_NODCCACHE` in the flags, but this function complains if it's passed anything other than `SW_SCROLLCHILDREN | SW_INVALIDATE | SW_ERASE`. This seems to be due to a relatively recent b2db69efcf4 (win32u: Move NtUserScrollWindowEx implementation from user32., 2022-06-08) as it replaced the call to the internal `scroll_window()` with a call to `NtUserScrollWindowEx()` with this unrecognized flag. -- 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 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 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 54690] New: ldp.exe crashes when attempting to connect to an invalid host
by WineHQ Bugzilla 20 Apr '23

20 Apr '23

https://bugs.winehq.org/show_bug.cgi?id=54690 Bug ID: 54690 Summary: ldp.exe crashes when attempting to connect to an invalid host Product: Wine Version: 8.3 Hardware: x86 OS: Linux Status: NEW Severity: normal Priority: P2 Component: wldap32 Assignee: wine-bugs(a)winehq.org Reporter: alexhenrie24(a)gmail.com Distribution: --- Created attachment 74201 --> https://bugs.winehq.org/attachment.cgi?id=74201 WINEDEBUG=+wldap32 ldp.exe is Microsoft's GUI LDAP client, which was part of the Windows XP Service Pack 2 Support Tools and is still available on Windows 10 by installing the Windows feature "Active Directory Lightweight Directory Services". Steps to reproduce: 1. Create a 32-bit Wine bottle and set the Windows version to XP. 2. Run `winetricks mfc42`. 3. Run `wine WindowsXP-KB838079-SupportTools-ENU.exe`. 4. Run `wine 'C:\Program Files\Support Tools\ldp.exe'`. 5. Click Connection > Connect, leave "Server" blank, and click OK. At this point, ldp.exe crashes. However, the program does not crash if a valid LDAP server is entered. $ sha256sum WindowsXP-KB838079-SupportTools-ENU.exe 7927e87af616d2fb8d4ead0db0103eb845a4e6651b20a5bffea9eebc3035c24d -- 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 54617] New: KakaoTalk IM text edit window leaves artifacts when the text overflows and scroll bar appears
by WineHQ Bugzilla 20 Apr '23

20 Apr '23

https://bugs.winehq.org/show_bug.cgi?id=54617 Bug ID: 54617 Summary: KakaoTalk IM text edit window leaves artifacts when the text overflows and scroll bar appears Product: Wine Version: 8.3 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 Distribution: --- KakaoTalk IM text edit window leaves artifacts when the text overflows and scroll bar appears. -- 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 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 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 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

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 April 2023