May 2015 - Wine-Devel - list.winehq.org

Re: [1/2] include: Add stuff related to WSAPoll()
by Bruno Jesus May 21, 2015

May 21, 2015

Please ignore the series, it seems to break ws2_32 due to multiple definitions for struct pollfd. Any ideas what can I do to fix this? Like what #ifdef could be used to prevent it? Best wishes, Bruno On Thu, May 21, 2015 at 3:52 PM, Bruno Jesus <00cpxxx(a)gmail.com> wrote: >

1 1

Re: [1/2] ntdll: Implement RtlUnicodeToUTF8N.
by Nikolay Sivov May 21, 2015

May 21, 2015

Any reason you can't use wine_utf8_mbstowcs/wine_utf8_wcstombs ?

1 0

Re: [PATCH 3/3] riched20/tests: Added tests for ITextFont getters on detached font
by Marvin May 21, 2015

May 21, 2015

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=13520 Your paranoid android. === wvistau64 (64 bit richole) === richole: unhandled exception c0000005 at 000007FEF57FACFA === w2008s64 (64 bit richole) === richole: unhandled exception c0000005 at 000007FEF648ACFA

1 0

Re: [PATCH 1/2] msvcp120: Added _Call_once implementation
by Nikolay Sivov May 21, 2015

May 21, 2015

On 20.05.2015 23:18, Piotr Caban wrote: > --- > dlls/msvcp110/msvcp110.spec | 4 +-- > dlls/msvcp120/msvcp120.spec | 4 +-- > dlls/msvcp120/tests/msvcp120.c | 55 > +++++++++++++++++++++++++++++++++++++ > dlls/msvcp120_app/msvcp120_app.spec | 4 +-- > dlls/msvcp90/misc.c | 35 +++++++++++++++++++++++ > 5 files changed, 96 insertions(+), 6 deletions(-) > > You sure it's not a wrapper around InitOnce* calls from kernel32?

2 1

Re: [PATCH] winex11drv: Fix uninitialized memory access appearing in multiple tests (Valgrind) (resend)
by Nikolay Sivov May 21, 2015

May 21, 2015

On 05/21/2015 08:05 AM, efbiaiinzinz(a)hotmail.com wrote: > From: Indrek Altpere <efbiaiinzinz(a)hotmail.com> > > --- > dlls/winex11.drv/window.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/dlls/winex11.drv/window.c b/dlls/winex11.drv/window.c > index 06e2294..de39a44 100644 > --- a/dlls/winex11.drv/window.c > +++ b/dlls/winex11.drv/window.c > @@ -701,6 +701,7 @@ static void set_size_hints( struct x11drv_win_data *data, DWORD style ) > static void set_mwm_hints( struct x11drv_win_data *data, DWORD style, DWORD ex_style ) > { > MwmHints mwm_hints; > + memset(&mwm_hints, 0, sizeof(mwm_hints)); > > if (data->hwnd == GetDesktopWindow()) > { > This looks too much, isn't it enough to just set input_mode = 0 once, before making a call?

2 1

widl problem in argument ordering
by Marcus Meissner May 21, 2015

May 21, 2015

Hi, Coverity issue "CID 1299032 Explicit null dereferenced" shows I think a problem of the widl generated stub marshaller. The IDL is: interface IAccIdentity : IUnknown { HRESULT GetIdentityString([in] DWORD idchild, [out, size_is(*string_len)] BYTE **str, [out] DWORD *string_len); } the generated stub is: void __RPC_STUB IAccIdentity_GetIdentityString_Stub( IRpcStubBuffer* This, IRpcChannelBuffer *_pRpcChannelBuffer, PRPC_MESSAGE _pRpcMessage, DWORD* _pdwStubPhase) { struct __frame_IAccIdentity_GetIdentityString_Stub __f, * const __frame = &__f; __frame->_This = (IAccIdentity*)((CStdStubBuffer*)This)->pvServerObject; NdrStubInitialize(_pRpcMessage, &__frame->_StubMsg, &Object_StubDesc, _pRpcChannelBuffer); RpcExceptionInit( 0, __finally_IAccIdentity_GetIdentityString_Stub ); __frame->str = 0; __frame->string_len = 0; RpcTryFinally { if ((_pRpcMessage->DataRepresentation & 0xffff) != NDR_LOCAL_DATA_REPRESENTATION) NdrConvert( &__frame->_StubMsg, &__MIDL_ProcFormatString.Format[216]); __frame->_StubMsg.Buffer = (unsigned char *)(((ULONG_PTR)__frame->_StubMsg.Buffer + 3) & ~0x3); if (__frame->_StubMsg.Buffer + sizeof(DWORD) > __frame->_StubMsg.BufferEnd) { RpcRaiseException(RPC_X_BAD_STUB_DATA); } __frame->idchild = *(DWORD *)__frame->_StubMsg.Buffer; __frame->_StubMsg.Buffer += sizeof(DWORD); __frame->str = NdrAllocate(&__frame->_StubMsg, *__frame->string_len * 4); // DEREFERENCED memset(__frame->str, 0, *__frame->string_len * 4); __frame->string_len = &__frame->_W0; // ACTUALLY initialized __frame->_W0 = 0; Not that __frame->string_len is dereferenced before it is set to &__frame->_W0. There seems to be a bit of an ordering problem here? Ciao, Marcus

2 1

Re: [PATCH] rpcrt4: make sure we do not have *data allocated on error (Coverity)
by Michael Stefaniuc May 20, 2015

May 20, 2015

On 05/20/2015 04:50 PM, Marcus Meissner wrote: > 713688 Resource leak > > callers expect *data only allocated when success is returned. > --- > dlls/rpcrt4/rpc_transport.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/dlls/rpcrt4/rpc_transport.c b/dlls/rpcrt4/rpc_transport.c > index 34f434d..4080b14 100644 > --- a/dlls/rpcrt4/rpc_transport.c > +++ b/dlls/rpcrt4/rpc_transport.c > @@ -2359,6 +2359,7 @@ static RPC_STATUS rpcrt4_http_read_http_packet(HINTERNET request, RpcHttpAsyncDa > if (!RPCRT4_IsValidHttpPacket(hdr, *data, data_len)) > { > ERR("invalid http packet\n"); > + if (*data) HeapFree(GetProcessHeap(), 0, *data); Marcus, you don't need the if not NULL check as HeapFree(NULL) is a no-op. bye michael

1 0

Re: [PATCH 2/4] ntdll: If read_directory_getattrlist() finds that the file doesn't exist, return successful "no file" result to stop search.
by Matteo Bruni May 20, 2015

May 20, 2015

---- diff --git a/dlls/ntdll/directory.c b/dlls/ntdll/directory.c index 0e5af8a..42e1842 100644 --- a/dlls/ntdll/directory.c +++ b/dlls/ntdll/directory.c @@ -2185,6 +2185,11 @@ static int read_directory_getattrlist( int fd, IO_STATUS_BLOCK *io, void *buffer } else io->u.Status = STATUS_NO_MORE_FILES; } + else if (errno == ENOENT || errno == ENOTDIR) + { + io->u.Status = STATUS_NO_MORE_FILES; + ret = 0; + } } else ret = -1; ---- Actually that's how my patch originally looked like (http://source.winehq.org/patches/data/111375) but then I realized that would break if the filesystem is case-sensitive, since it would return the file as missing while it might be present with a different name casing. I haven't actually tested it though.

2 3

Re: [PATCH 4/4] ntdll: Fix read_directory_getattrlist() to get the name of a symlink rather than its target, but still detect if the symlink is broken.
by Matteo Bruni May 20, 2015

May 20, 2015

2015-05-20 20:40 GMT+02:00 Ken Thomases <ken(a)codeweavers.com>: > --- > dlls/ntdll/directory.c | 18 ++++++++++++++++-- > 1 file changed, 16 insertions(+), 2 deletions(-) Not sure it matters in practice but it might be better to check for HAVE_SYS_VNODE_H in both the definition of and the call to read_directory_getattrlist().

2 1

Re: [PATCH] kernel32: Implement a few more classes in GetFileInformationByHandleEx
by Marvin May 20, 2015

May 20, 2015

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=13506 Your paranoid android. === wvistau64 (32 bit file) === file.c:444: Test failed: attributes of file "testfile.xxx" are 0x0820 === wvistau64 (64 bit file) === file.c:444: Test failed: attributes of file "testfile.xxx" are 0x0820

1 0

Re: msvcr120: Add erf functions (try 2)
by Alexandre Julliard May 20, 2015

May 20, 2015

Daniel Lehman <dlehman(a)esri.com> writes: > @@ -73,6 +73,32 @@ int CDECL MSVCRT__set_SSE2_enable(int flag) > return sse2_enabled; > } > > +#if _MSVCR_VER>=120 > +/********************************************************************* > + * erff (MSVCRT.@) > + */ > +float CDECL MSVCRT_erff( float num ) > +{ > + return erff(num); > +} > + > +/********************************************************************* > + * erf (MSVCRT.@) > + */ > +double CDECL MSVCRT_erf( double num ) > +{ > + return erf(num); > +} Please add configure checks. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH 7/7] riched20: Implement GetFont() for selection range
by Marvin May 20, 2015

May 20, 2015

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=13500 Your paranoid android. === build (build) === Patch failed to apply

1 0

Re: [PATCH 6/7] riched20: Implement GetLanguageID()
by Marvin May 20, 2015

May 20, 2015

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=13499 Your paranoid android. === build (build) === Patch failed to apply

1 0

Re: [PATCH 1/7] riched20: Implement GetSize()
by Marvin May 20, 2015

May 20, 2015

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=13498 Your paranoid android. === build (build) === Patch failed to apply

1 0

Re: [PATCH 3/4] wined3d: Check resource types separately in check_fbo_compat.
by Henri Verbeet May 19, 2015

May 19, 2015

On 16 May 2015 at 15:29, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > +static void delete_fbo_attachment(const struct wined3d_gl_info *gl_info, > + enum wined3d_gl_resource_type d3d_type) > +{ > + switch (d3d_type) > + { > + case WINED3D_GL_RES_TYPE_TEX_1D: > + gl_info->fbo_ops.glFramebufferTexture1D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, GL_TEXTURE_1D, 0, 0); > + break; > + > + case WINED3D_GL_RES_TYPE_TEX_2D: > + case WINED3D_GL_RES_TYPE_TEX_RECT: > + gl_info->fbo_ops.glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, > + wined3d_gl_type_to_enum(d3d_type), 0, 0); > + break; > + > + case WINED3D_GL_RES_TYPE_TEX_3D: > + gl_info->fbo_ops.glFramebufferTexture3D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, GL_TEXTURE_3D, 0, 0, 0); > + break; > + > + case WINED3D_GL_RES_TYPE_TEX_CUBE: > + gl_info->fbo_ops.glFramebufferTexture2D(GL_FRAMEBUFFER, GL_COLOR_ATTACHMENT0, > + GL_TEXTURE_CUBE_MAP_POSITIVE_X_ARB, 0, 0); > + break; > + > + case WINED3D_GL_RES_TYPE_BUFFER: > + case WINED3D_GL_RES_TYPE_COUNT: > + break; > + } > +} > + This is kind of pointless. Any of the FramebufferTexture*() calls with a 0 texture will detach the current attachment. You shouldn't need to detach the current attachment before attaching a new one either, but I suppose the existing code already does that.

2 1

Re: [PATCH 3/5] riched20: Implement GetSize() (resend)
by Marvin May 19, 2015

May 19, 2015

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=13468 Your paranoid android. === w7u (32 bit richole) === richole.c:1287: Test failed: got weight 400

1 0

Re: [PATCH 2/5] riched20: Implement GetWeight() (resend)
by Marvin May 19, 2015

May 19, 2015

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=13467 Your paranoid android. === w7u (32 bit richole) === richole.c:1283: Test failed: after 400

1 0

Re: [PATCH 1/5] riched20: Implement GetBold()/GetItalic()
by Marvin May 19, 2015

May 19, 2015

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=13451 Your paranoid android. === build (build) === Patch failed to apply

2 1

Re: kernel32/tests: Add tests for delete behaviour of CopyFileEx.
by Marvin May 19, 2015

May 19, 2015

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=13465 Your paranoid android. === wvistau64 (32 bit file) === file.c:444: Test failed: attributes of file "testfile.xxx" are 0x0820 === wvistau64 (64 bit file) === file.c:444: Test failed: attributes of file "testfile.xxx" are 0x0820

1 0

Re: dmusic: Remove "keep in sync with oneself" comments
by Michael Stefaniuc May 18, 2015

May 18, 2015

Hello Frederic, why would you want to do this? Keeping the files identical just allows to copy it as is. Or easily verify it by just checking the SHA1 sum. On 05/18/2015 02:47 AM, Frédéric Delanoy wrote: > --- > dlls/dmusic/dmobject.c | 1 - > dlls/dmusic/dmobject.h | 1 - > 2 files changed, 2 deletions(-) > > diff --git a/dlls/dmusic/dmobject.c b/dlls/dmusic/dmobject.c > index cabb6f7..8e9f58e 100644 > --- a/dlls/dmusic/dmobject.c > +++ b/dlls/dmusic/dmobject.c > @@ -1,6 +1,5 @@ > /* > * Base IDirectMusicObject Implementation > - * Keep in sync with the master in dlls/dmusic/dmobject.c > * > * Copyright (C) 2003-2004 Rok Mandeljc > * Copyright (C) 2014 Michael Stefaniuc > diff --git a/dlls/dmusic/dmobject.h b/dlls/dmusic/dmobject.h > index ad6bf6d..ffc683c 100644 > --- a/dlls/dmusic/dmobject.h > +++ b/dlls/dmusic/dmobject.h > @@ -1,6 +1,5 @@ > /* > * Base IDirectMusicObject Implementation > - * Keep in sync with the master in dlls/dmusic/dmobject.h > * > * Copyright (C) 2014 Michael Stefaniuc > * > bye michael

1 0

Re: [PATCH 3/5] riched20: Implement GetSize()
by Marvin May 18, 2015

May 18, 2015

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=13453 Your paranoid android. === build (build) === Patch failed to apply

1 0

Re: [PATCH 2/5] riched20: Implement GetWeight()
by Marvin May 18, 2015

May 18, 2015

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=13452 Your paranoid android. === build (build) === Patch failed to apply

1 0

Why does ntdll/thread.c use pthread_attr_setstack
by Mate Antunovic May 18, 2015

May 18, 2015

I have noticed that RtlCreateUserThread in dlls/ntdll.thread.c uses the following call pthread_attr_setstack( &attr, teb->DeallocationStack, (char *)teb->Tib.StackBase - (char *)teb->DeallocationStack ); I have the source code from release 1.7.41. I am working on a complex project that traces the execution of programs running in Wine (won't go into detail here). I have noticed problems in RtlCreateUserThread when services.exe is launched in my environment. Basically the call to pthread_create causes a SIGSEGV. Let me make it clear that under ordinary conditions, Wine works fine. Playing around with things, I note that if I change the call above to be pthread_attr_setstacksize(&attr, 0x10000), then I don't get the same problem. My question is, why does the exact position of the stack need to be set in this way?

1 0

Re: [PATCH] oledlg: Make object name menu item localized
by Alex Henrie May 18, 2015

May 18, 2015

+ WCHAR objectW[50]; /* should be enough*/ In the past I was told to always allocate memory in multiplies of 8 bytes,[1] and I personally have started to prefer powers of 2 in particular. Alexandre seems to prefer allocating at least 64 characters for short strings like this, judging from how he amended a proposed patch of mine several years ago.[2] + if (!LoadStringW(OLEDLG_hInstance, IDS_VERBMENU_OBJECT, objectW, sizeof(objectW)/sizeof(WCHAR))) { + static const WCHAR enobjectW[] = {'&','O','b','j','e','c','t',0}; + strcpyW(objectW, enobjectW); + } Is this really necessary? How could LoadString fail? I recently submitted a very similar patch [3] and didn't have to provide a special fallback. Just to be clear, I do really appreciate you taking the time to make sure that Wine is properly localized :-) I just don't want to see the codebase made unnecessarily complicated or inefficient. -Alex [1] https://www.winehq.org/pipermail/wine-devel/2011-December/093516.html [2] https://source.winehq.org/git/wine.git/commitdiff/6e5bba1b60f97fce34415ea30… [3] https://www.winehq.org/pipermail/wine-patches/2015-May/139132.html

2 1

Re: [PATCH 1/2] kernel32: Implement GetSystemTimePreciseAsFileTime
by Andrew Eikum May 17, 2015

May 17, 2015

On Thu, May 07, 2015 at 11:02:34PM +0300, Martin Storsjo wrote: > +/*********************************************************************** > + * GetSystemTimePreciseAsFileTime (KERNEL32.@) > + * > + * Get the current time in utc format, with <1 us precision. > + * > + * RETURNS > + * Nothing. > + */ > +VOID WINAPI GetSystemTimePreciseAsFileTime( > + LPFILETIME time) /* [out] Destination for the current utc time */ > +{ > + GetSystemTimeAsFileTime(time); > +} I don't know why this one's not getting in. I have a couple small guesses. Personally, I'm opposed to including a bunch of documentation in Wine. I don't think Wine is the right place to document the Windows API. I don't know if others feel this way, or if this is enough for rejection. You're also using VOID, LPJUNK, etc, which I think we're trying to avoid these days. Andrew

3 3