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

Re: [1/2] include: Add stuff related to WSAPoll()
by Bruno Jesus 21 May '15

21 May '15

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 21 May '15

21 May '15

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 21 May '15

21 May '15

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 21 May '15

21 May '15

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 21 May '15

21 May '15

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 21 May '15

21 May '15

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 20 May '15

20 May '15

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 20 May '15

20 May '15

---- 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 20 May '15

20 May '15

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 20 May '15

20 May '15

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