September 2011 - Wine-devel

Re: d3dx9_26: Add custom ID3DXEffect interface.
by Erich Hoover 29 Sep '11

29 Sep '11

SetRawValue is at the end of the table, so it being extra will not influence any older application. So, you should be able to get the same functionality with way less code by adding that alternative GUID to the QueryInterface in d3dx9_36. Erich Hoover ehoover(a)mines.edu

5 8

Re: [PATCH] Enable X11DRV_DIB_GetImageBits_8 in all cases and optimize X11DRV_DIB_GetNearestIndex to fix bug 2666 and other strange behavior
by Frédéric Delanoy 29 Sep '11

29 Sep '11

Please do not include bug number in commit message summary line. Put something like 'fixes bug 123 in the commit header See http://wiki.winehq.org/SubmittingPatches for more info (e.g. "subject line" and "include a description" sections) Frédéric

2 3

Re: d3dx9_26: Add custom ID3DXEffect interface
by luis.busquets＠ilidium.com 29 Sep '11

29 Sep '11

And what about introducing wined3dx to wrap up all d3dx9, d3dx10, ... functions?

1 0

CreateDC and DRIVER_GetDriverName
by Bruno Jesus 28 Sep '11

28 Sep '11

Hi, I'm seeking advice as I really don't understand anything about wine display drawing (gdi, x11, ...). Every game created with an old win3.11 tool called Klik & Play do not work in wine because a driver called DIB is passed as parameter to CreateDC. Till today I've only seen a driver called DISPLAY being used by all sorts of applications. Looking at the DRIVER_GetDriverName [1] function I can see a list where 3 different drivers are supported. In an attempt to run an old game called Mortuum [2] I can see that it always fail due to this missing DIB driver: err:dc:CreateDCW no driver found for L"DIB" My lack of knowledge made me do a dirty hack in the CreateDCW function forcing the driver DISPLAY to be used instead of DIB and when I started the game it draw it's first screen and then crashed. What should be done to support the DIB driver name now that DIB is implemented in wine? [1] = http://source.winehq.org/source/dlls/gdi32/driver.c#L800 [2] = ftp://ftp.padua.org/pub/msdos/windows/games/local/mortuum.zip Best wishes, Bruno -- universe* god::bigbang (void); //and then it all began...

1 0

Re: [PATCH] d3dx9_36: avoid using fabsf/fmaxf
by Marvin 28 Sep '11

28 Sep '11

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 http://testbot.winehq.org/JobDetails.pl?Key=14617 Your paranoid android. === WINEBUILD (build) === Patch failed to apply

1 0

Re: [PATCH] d3dx9_36: avoid using fabsf/fmaxf
by Austin English 28 Sep '11

28 Sep '11

On Wed, Sep 28, 2011 at 01:51, Natanael Copa <ncopa(a)alpinelinux.org> wrote: > This fixes compiling on Alpine Linux and other uClibc based distros. > > This is related bug 28247 and commit 4cddd5ffdf0c568445365c6c62e7788d7dbbca7d > --- > dlls/d3dx9_36/tests/mesh.c | 14 +++++++------- > 1 files changed, 7 insertions(+), 7 deletions(-) > > diff --git a/dlls/d3dx9_36/tests/mesh.c b/dlls/d3dx9_36/tests/mesh.c > index 6a78e64..c09607f 100644 > --- a/dlls/d3dx9_36/tests/mesh.c > +++ b/dlls/d3dx9_36/tests/mesh.c > @@ -6160,7 +6160,7 @@ static void check_vertex_components(int line, int mesh_number, int vertex_number > { > FLOAT *got = (FLOAT*)(got_ptr + decl_ptr->Offset); > FLOAT *exp = (FLOAT*)(exp_ptr + decl_ptr->Offset); > - FLOAT diff = fabsf(*got - *exp); > + FLOAT diff = abs(*got - *exp); > ok_(__FILE__,line)(diff <= FLT_EPSILON, "Mesh %d: Got %f for vertex %d %s, expected %f.\n", > mesh_number, *got, vertex_number, usage_strings[decl_ptr->Usage], *exp); > break; > @@ -6169,7 +6169,7 @@ static void check_vertex_components(int line, int mesh_number, int vertex_number > { > D3DXVECTOR2 *got = (D3DXVECTOR2*)(got_ptr + decl_ptr->Offset); > D3DXVECTOR2 *exp = (D3DXVECTOR2*)(exp_ptr + decl_ptr->Offset); > - FLOAT diff = fmaxf(fabsf(got->x - exp->x), fabsf(got->y - exp->y)); > + FLOAT diff = max(abs(got->x - exp->x), abs(got->y - exp->y)); > ok_(__FILE__,line)(diff <= FLT_EPSILON, "Mesh %d: Got (%f, %f) for vertex %d %s, expected (%f, %f).\n", > mesh_number, got->x, got->y, vertex_number, usage_strings[decl_ptr->Usage], exp->x, exp->y); > break; > @@ -6178,8 +6178,8 @@ static void check_vertex_components(int line, int mesh_number, int vertex_number > { > D3DXVECTOR3 *got = (D3DXVECTOR3*)(got_ptr + decl_ptr->Offset); > D3DXVECTOR3 *exp = (D3DXVECTOR3*)(exp_ptr + decl_ptr->Offset); > - FLOAT diff = fmaxf(fabsf(got->x - exp->x), fabsf(got->y - exp->y)); > - diff = fmaxf(diff, fabsf(got->z - exp->z)); > + FLOAT diff = max(abs(got->x - exp->x), abs(got->y - exp->y)); > + diff = max(diff, abs(got->z - exp->z)); > ok_(__FILE__,line)(diff <= FLT_EPSILON, "Mesh %d: Got (%f, %f, %f) for vertex %d %s, expected (%f, %f, %f).\n", > mesh_number, got->x, got->y, got->z, vertex_number, usage_strings[decl_ptr->Usage], exp->x, exp->y, exp->z); > break; > @@ -6188,9 +6188,9 @@ static void check_vertex_components(int line, int mesh_number, int vertex_number > { > D3DXVECTOR4 *got = (D3DXVECTOR4*)(got_ptr + decl_ptr->Offset); > D3DXVECTOR4 *exp = (D3DXVECTOR4*)(exp_ptr + decl_ptr->Offset); > - FLOAT diff = fmaxf(fabsf(got->x - exp->x), fabsf(got->y - exp->y)); > - diff = fmaxf(diff, fabsf(got->z - exp->z)); > - diff = fmaxf(diff, fabsf(got->w - exp->w)); > + FLOAT diff = max(abs(got->x - exp->x), abs(got->y - exp->y)); > + diff = max(diff, abs(got->z - exp->z)); > + diff = max(diff, abs(got->w - exp->w)); > ok_(__FILE__,line)(diff <= FLT_EPSILON, "Mesh %d: Got (%f, %f, %f, %f) for vertex %d %s, expected (%f, %f, %f, %f).\n", > mesh_number, got->x, got->y, got->z, got->w, vertex_number, usage_strings[decl_ptr->Usage], exp->x, exp->y, exp->z, got->w); > break; > -- > 1.7.6.1 fmaxf was already removed in http://source.winehq.org/git/wine.git/commitdiff/49064869595a8d0c97a79055eb… -- -Austin

1 0

Re: [2/3] cmd: Avoid checking whether we're in console mode for every read (try 2)
by Eric Pouech 28 Sep '11

28 Sep '11

Le 28/09/2011 01:01, Frédéric Delanoy a écrit : > --- > programs/cmd/batch.c | 9 ++++----- > programs/cmd/builtins.c | 4 ++-- > programs/cmd/wcmd.h | 5 +++-- > programs/cmd/wcmdmain.c | 21 +++++++++++++-------- > 4 files changed, 22 insertions(+), 17 deletions(-) actually, console handles have their 2 lower bit set you could use it to even simplify further the code A+ -- Eric Pouech "The problem with designing something completely foolproof is to underestimate the ingenuity of a complete idiot." (Douglas Adams)

1 0

Re: [PATCH 1/2] mmdevapi: Don't fail if dwChannelMask is not set correctly
by Marvin 28 Sep '11

28 Sep '11

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 http://testbot.winehq.org/JobDetails.pl?Key=14610 Your paranoid android. === WVISTAADM (32 bit render) === render.c:232: Test failed: IsFormatSupported(dwChannelMask = 0xffff) call returns 00000000 === W2K8SE (32 bit render) === render.c:232: Test failed: IsFormatSupported(dwChannelMask = 0xffff) call returns 00000000 === W7PRO (32 bit render) === render.c:232: Test failed: IsFormatSupported(dwChannelMask = 0xffff) call returns 00000000 === W7PROX64 (32 bit render) === render.c:232: Test failed: IsFormatSupported(dwChannelMask = 0xffff) call returns 00000000 === TEST64_W7SP1 (32 bit render) === render.c:232: Test failed: IsFormatSupported(dwChannelMask = 0xffff) call returns 00000000 === W7PROX64 (64 bit render) === render.c:232: Test failed: IsFormatSupported(dwChannelMask = 0xffff) call returns 00000000 === TEST64_W7SP1 (64 bit render) === render.c:232: Test failed: IsFormatSupported(dwChannelMask = 0xffff) call returns 00000000

1 0

Re: browseui/tests: Skip some tests if IEnumACString is not supported.
by Marvin 28 Sep '11

28 Sep '11

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 http://testbot.winehq.org/JobDetails.pl?Key=14608 Your paranoid android. === WNT4WSSP6 (32 bit autocomplete) === autocomplete.c:250: Test failed: CoCreateInstance(&CLSID_ACLMulti, NULL, CLSCTX_INPROC, &IID_IEnumString, (LPVOID *)&obj) failed: 80040154 === WXPPROSP3 (32 bit autocomplete) === autocomplete.c:261: Test failed: QueryInterface(IID_IEnumACString) failed: 0 === W2K3R2SESP2 (32 bit autocomplete) === autocomplete.c:261: Test failed: QueryInterface(IID_IEnumACString) failed: 0 === WVISTAADM (32 bit autocomplete) === autocomplete.c:261: Test failed: QueryInterface(IID_IEnumACString) failed: 0 === W2K8SE (32 bit autocomplete) === autocomplete.c:261: Test failed: QueryInterface(IID_IEnumACString) failed: 0 === W7PRO (32 bit autocomplete) === autocomplete.c:261: Test failed: QueryInterface(IID_IEnumACString) failed: 0 === W7PROX64 (32 bit autocomplete) === autocomplete.c:261: Test failed: QueryInterface(IID_IEnumACString) failed: 0 === TEST64_W7SP1 (32 bit autocomplete) === autocomplete.c:261: Test failed: QueryInterface(IID_IEnumACString) failed: 0 === W7PROX64 (64 bit autocomplete) === autocomplete.c:261: Test failed: QueryInterface(IID_IEnumACString) failed: 0 === TEST64_W7SP1 (64 bit autocomplete) === autocomplete.c:261: Test failed: QueryInterface(IID_IEnumACString) failed: 0

1 0

Re: [1/3] cmd: Avoid reading char by char from console (try 2)
by Alexandre Julliard 28 Sep '11

28 Sep '11

Frédéric Delanoy <frederic.delanoy(a)gmail.com> writes: > @@ -181,13 +181,21 @@ WCHAR *WCMD_parameter (WCHAR *s, int n, WCHAR **where, WCHAR **end) { > * the LF (or CRLF) from the line. > */ > > -WCHAR *WCMD_fgets (WCHAR *s, int noChars, HANDLE h) { > - > - DWORD bytes; > - BOOL status; > +WCHAR *WCMD_fgets (WCHAR *s, int noChars, HANDLE h) > +{ > + DWORD bytes, charsRead, dummy; > + BOOL status, is_console_handle; > WCHAR *p; > > p = s; > + is_console_handle = !!GetConsoleMode(h, &dummy); > + if (is_console_handle) { > + status = ReadConsoleW(h, s, noChars, &charsRead, NULL); ReadConsole will fail if it's not a console, the check is redundant. -- Alexandre Julliard julliard(a)winehq.org

2 2