February 2011 - Wine-Devel - list.winehq.org

Re: d3dx9: Implement D3DXFloat16to32Array and D3DXFloat32to16Array.

by Misha Koshelev

On Wed, Feb 9, 2011 at 4:40 AM, Stefan Dösinger <stefandoesinger(a)gmx.at> wrote: > Am Dienstag 08 Februar 2011, 22:45:52 schrieb Misha Koshelev: >> I don't have msvc. Please suggest how I can check this. Thank you. Yours >> Misha > According to http://www.johndcook.com/math_h.html it exists, but has a > different name. So we'd need some wrapper around it in libwine. Maybe one > exists already, I can't check right now. If not I recommend to stick to > isnan() and isinf() > Thanks. Do you have a good way to compare to 0.0f and -0.0f? For some reason, when I took out references and instead used actually float parameters, my comparisons of ((unsigned float) in) == 0x00000000 became true for Inf and NaN as well? Thank you Misha

Feb. 9, 2011

Patch 21026 superseded by patch 71118.

by Max TenEyck Woodbury

No text.

Feb. 9, 2011

Re: Try to Implement my first Stub function - AbortPrinter() - (try 2).

by Andrew Eikum

On 02/09/2011 09:09 AM, Loïc Maury wrote: >> The indentation looks fine to me now. You've got an extra newline >> after the "if(printer->doc)" block and before the "else." The >> formatting on the TRACE statement is still bizarre. Fix the commas and >> put it on just a couple of lines. No reason for one line and a ton of >> whitespace for each parameter. See line 1944 for an example. > Ok, it's correct now ? The content looks fine, but there are still formatting problems in your patch. You're still using commas strangely in your wrapped TRACE statement. Your editor is broken again, and you are now mixing tabs and spaces. A few lines still have whitespace at the end. These are things you should verify yourself before sending it off for someone else to look at. Andrew

Feb. 9, 2011

Re: Try to Implement my first Stub function - AbortPrinter() - (try 2).

by Loïc Maury

Hello Mr.Le Cam On Tue, Feb 8, 2011 at 10:07 PM, Nicolas Le Cam <niko.lecam(a)gmail.com>wrote: > > > 2011/2/8 Loïc Maury <lmaury(a)gmail.com> > > >> Hello Mr.Timoshkov, >> >> Thank you for your reply. >> >>> Loīc Maury<lmaury(a)gmail.com> wrote: >>> >>> >>> After the various comments, I have modified the patch. >>>> >>> First of all set your tab size to 8, and ask your editor to not use tabs >>> at all. >>> >> I have modified my editor, but I don't know if it 's correct now ? >> >>> + TRACE("(%s, %s, %p, %d, %d)\n",debugstr_w(printer->name) >>>> >>>> + ,debugstr_w(printer->printername) >>>> + ,printer->backend_printer >>>> + ,printer->queue->ref >>>> + ,list_count(&printer->queue->jobs)); >>>> >>> TRACE() with the API parameters usually is the very first statement in >>> the API implementation, comma should be placed at the end of the >>> statement, >>> not before. >>> >> Ok, I have remplaced this TRACE(). >> >>> + GetPrinterW(hPrinter, 2, NULL, >>>> 0,&needed); >>>> >>>> + pi2 = HeapAlloc(GetProcessHeap(), 0, >>>> needed); >>>> + GetPrinterW(hPrinter, 2, (LPBYTE)pi2, >>>> needed,&needed); >>>> >>> You need to check the return value of GetPrinterW() and handle the >>> errors. >>> >> Ok >> >>> + if(pi2) >>>> + HeapFree(GetProcessHeap(), 0, pi2); >>>> >>> NULL check before HeapFree() is not needed. >>> >> Ok, I have removed the NULL check. >> >> + TRACE("return %d\n", ret); >>>> >>> This trace is redundant. >>> >> Ok, I have removed this TRACE(). >> >> I have make an other patch. >> >> Thank you >> >> Loīc >> >> >> >> >> > Hi Loïc, > > In addition to Dmitry and Andrew comments, you're adding a lot of trailing > spaces (git apply output : warning: squelched 22 whitespace errors warning: > 27 lines add whitespace errors.), there's also no need for an extra newline > after the end label. > > Now, The patch is correct ? Thank you Loïc > -- > Nicolas Le Cam >

Feb. 9, 2011

Re: Try to Implement my first Stub function - AbortPrinter() - (try 2).

by Loïc Maury

Hello Mr.Eikum, Thank you for your reply. On 08/02/11 18:13, Andrew Eikum wrote: > On 02/08/2011 10:48 AM, Loïc Maury wrote: >>> First of all set your tab size to 8, and ask your editor to not use >>> tabs >>> at all. >> I have modified my editor, but I don't know if it 's correct now ? > > The indentation looks fine to me now. You've got an extra newline > after the "if(printer->doc)" block and before the "else." The > formatting on the TRACE statement is still bizarre. Fix the commas and > put it on just a couple of lines. No reason for one line and a ton of > whitespace for each parameter. See line 1944 for an example. Ok, it's correct now ? > > Otherwise, two little things I noticed. > > First, you must also check the return value for HeapAlloc(). Ok > > Second, you should be using the LIST_FOR_EACH_ENTRY_SAFE macro, which > will eliminate the local cursor and cursor2 variables and make the > code a little more simple. (Unless there's something I'm missing?) Ok, I have modified and removed cursor and cursor2 by job and next_job, and list_remove(cursor) by list_remove(&job->entry). > > Almost done, nice work :) Thank you Loïc > > Andrew >

Feb. 9, 2011

Re: [PATCH 2/5] d3dx9_36: Implemented ID3DXConstantTable_SetIntArray and ID3DXConstantTable_SetInt

by Matteo Bruni

2011/2/9 Travis Athougies <iammisc(a)gmail.com>: > > On Feb 8, 2011, at 4:10 AM, Matteo Bruni <matteo.mystral(a)gmail.com> wrote: > >>> + /* We need the for loop, since we need to convert the integer to >>> a float */ >>> + for (i = 0; i < count && i < desc.RegisterCount; i++) >>> + { >>> + row[0] = (float)n[i]; >> >> The cast is unneeded, and the comment above is too. > > If it's being passed into set*shaderconstantf, shouldn't it need to be cast > to a float? Or am I mishnderstaning you? > > Travis > In that statement you are assigning n[i], an int, to row[0], a float. Automatic type conversion will happen anyway, so no (float) cast is necessary. In general you should avoid casts unless they are really needed.

Feb. 9, 2011

Re: d3dx9_36: make sure return values are used (LLVM/Clang)

by Matteo Bruni

hr = IDirect3DTexture9_GetLevelDesc(texture, 0, &desc); + ok(hr == D3D_OK, "D3DXCreateTexture returned %#x, expected %#x\n", hr, D3D_OK); Copy-paste error, it should be GetLevelDesc in the ok() message. :)

Feb. 9, 2011

Re: user32: Add message tests for SetParent().

by Marvin

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=9096 Your paranoid android. === W7PROX64 (64 bit msg) === Timeout

Feb. 9, 2011

Re: d3dx9_36: Implement a few tests for D3DXCreateBox

by Stefan Dösinger

Am Mittwoch 09 Februar 2011, 07:34:21 schrieb David Adam: > + wnd = CreateWindow("static", "d3dx9_test", 0, 0, 0, 0, 0, NULL, NULL, NULL, NULL); The class "static" doesn't work on Windows 7, I am afraid you'll need a proper window class. And I think a CreateWindow failure should count as test failure, it's a stupid reason to skip the entire test. (The other d3dx9 tests may have the same issue)

Feb. 9, 2011

Re: Wine and Summer of Code 2011

by Stefan Dösinger

Count me in as potential mentor as well Am Dienstag 08 Februar 2011, 21:51:55 schrieb Eric Durbin: > On Tue, Feb 8, 2011 at 12:04 PM, Dan Kegel <dank(a)kegel.com> wrote: > > http://www.google-melange.com/document/show/gsoc_program/google/gsoc2011/ > > faqs#mentor_apply says projects can apply starting on Feb 28. > > > > In preparation, I moved the 2010 projects from > > http://wiki.winehq.org/SummerOfCode > > to > > http://wiki.winehq.org/SummerOfCode/PreviousProjects > > > > I also added a DSS Provider project idea, removed the gstreamer > > idea since that's implemented, and tidied up the dxdiag idea a bit. > > > > Who's interested in mentoring this year? And can y'all think of any > > fresh ideas for the suggested projects list, or ideas that should be > > retired? > > - Dan > > I would be interested in mentoring again this year.

Feb. 9, 2011

Re: d3dx9: Implement D3DXFloat16to32Array and D3DXFloat32to16Array.

by Stefan Dösinger

Am Dienstag 08 Februar 2011, 22:45:52 schrieb Misha Koshelev: > I don't have msvc. Please suggest how I can check this. Thank you. Yours > Misha According to http://www.johndcook.com/math_h.html it exists, but has a different name. So we'd need some wrapper around it in libwine. Maybe one exists already, I can't check right now. If not I recommend to stick to isnan() and isinf()

Feb. 9, 2011

error message for ddraw

by David Adam

When I test the following test rc = IDirectDraw_SetCooperativeLevel(lpDD, NULL, DDSCL_FULLSCREEN | DDSCL_EXCLUSIVE | DDSCL_CREATEDEVICEWINDOW); ok(rc==DDERR_INVALIDPARAMS, "Expected DDERR_INVALIDPARAMS, received %x\n", rc); in my vista box, I receive the following failing message ddrawmodes: test failed: expected DDERR_INVALIDPARAMS, received 8876035a. But I don't know what is this message. Does anyone know what means this message? Thanks in advance A+ David

Feb. 9, 2011

Re: user32: Add a test for reparenting a WS_POPUP window to a WS_CHILD parent. Take 2.

by Dmitry Timoshkov

Alexandre Julliard <julliard(a)winehq.org> wrote: > > This test is supposed to show the root of the problem with the test > > application attached to the bug 13951. It appears that the check performed > > by server/window.c,is_top_level_window() is not correct (which leads to > > a SetWindowPos() failure to set foreground window), and probably should be > > replaced by simple (win->style & (WS_CHILD | WS_POPUP)) != WS_CHILD. > > Maybe, but the test doesn't really demonstrate that. That was just a guess, I hope that doesn't invalidate the test. If there is a need for additional tests please let me know and I'll add them later. -- Dmitry.

Feb. 9, 2011

Re: mshtml/tests: make sure return values are used (LLVM/Clang)

by Marvin

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=9073 Your paranoid android. === WXPPROSP3 (32 bit dom) === No test summary line found

Feb. 9, 2011

Re: secur32/tests: make sure a return value is used (LLVM/Clang)

by Marvin

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=9068 Your paranoid android. === WNT4WSSP6 (32 bit ntlm) === ntlm.c:1231: Test failed: AcquireCredentialsHande() returned SEC_E_INVALID_TOKEN ntlm.c:423: Test failed: expected SEC_E_BUFFER_TOO_SMALL, got SEC_E_INSUFFICIENT_MEMORY ntlm.c:433: Test failed: expected SEC_E_INTERNAL_ERROR or SEC_I_CONTINUE_NEEDED, got SEC_E_NO_IMPERSONATION ntlm.c:445: Test failed: expected SEC_E_BUFFER_TOO_SMALL, got SEC_E_INSUFFICIENT_MEMORY ntlm.c:457: Test failed: expected SEC_E_BUFFER_TOO_SMALL, got SEC_E_INSUFFICIENT_MEMORY ntlm.c:885: Test failed: cbSecurityTrailer should be 16 but is 0 ntlm.c:423: Test failed: expected SEC_E_BUFFER_TOO_SMALL, got SEC_E_INSUFFICIENT_MEMORY ntlm.c:433: Test failed: expected SEC_E_INTERNAL_ERROR or SEC_I_CONTINUE_NEEDED, got SEC_E_NO_IMPERSONATION ntlm.c:445: Test failed: expected SEC_E_BUFFER_TOO_SMALL, got SEC_E_INSUFFICIENT_MEMORY ntlm.c:457: Test failed: expected SEC_E_BUFFER_TOO_SMALL, got SEC_E_INSUFFICIENT_MEMORY ntlm.c:885: Test failed: cbSecurityTrailer should be 16 but is 0 ntlm.c:423: Test failed: expected SEC_E_BUFFER_TOO_SMALL, got SEC_E_INSUFFICIENT_MEMORY ntlm.c:433: Test failed: expected SEC_E_INTERNAL_ERROR or SEC_I_CONTINUE_NEEDED, got SEC_E_NO_IMPERSONATION ntlm.c:445: Test failed: expected SEC_E_BUFFER_TOO_SMALL, got SEC_E_INSUFFICIENT_MEMORY ntlm.c:457: Test failed: expected SEC_E_BUFFER_TOO_SMALL, got SEC_E_INSUFFICIENT_MEMORY ntlm.c:885: Test failed: cbSecurityTrailer should be 16 but is 0 ntlm.c:423: Test failed: expected SEC_E_BUFFER_TOO_SMALL, got SEC_E_INSUFFICIENT_MEMORY ntlm.c:433: Test failed: expected SEC_E_INTERNAL_ERROR or SEC_I_CONTINUE_NEEDED, got SEC_E_NO_IMPERSONATION ntlm.c:445: Test failed: expected SEC_E_BUFFER_TOO_SMALL, got SEC_E_INSUFFICIENT_MEMORY ntlm.c:457: Test failed: expected SEC_E_BUFFER_TOO_SMALL, got SEC_E_INSUFFICIENT_MEMORY ntlm.c:885: Test failed: cbSecurityTrailer should be 16 but is 0 ntlm.c:423: Test failed: expected SEC_E_BUFFER_TOO_SMALL, got SEC_E_INSUFFICIENT_MEMORY ntlm.c:433: Test failed: expected SEC_E_INTERNAL_ERROR or SEC_I_CONTINUE_NEEDED, got SEC_E_NO_IMPERSONATION ntlm.c:445: Test failed: expected SEC_E_BUFFER_TOO_SMALL, got SEC_E_INSUFFICIENT_MEMORY ntlm.c:457: Test failed: expected SEC_E_BUFFER_TOO_SMALL, got SEC_E_INSUFFICIENT_MEMORY ntlm.c:1031: Test failed: MakeSignature returned SEC_E_INVALID_TOKEN, not SEC_E_OK. ntlm.c:1061: Test failed: Crypt trailer not as expected. ntlm.c:1075: Test failed: Crypt message not as expected. ntlm.c:1097: Test failed: DecryptMessage returned SEC_E_MESSAGE_ALTERED, not SEC_E_OK. ntlm.c:1099: Test failed: Failed to decrypt message correctly. ntlm.c:1127: Test failed: MakeSignature returned SEC_E_INVALID_TOKEN, not SEC_E_OK. ntlm.c:1135: Test failed: VerifySignature returned SEC_E_INVALID_TOKEN, not SEC_E_OK ntlm.c:1139: Test failed: EncryptMessage returned SEC_E_INVALID_TOKEN, not SEC_E_OK. ntlm.c:1157: Test failed: Crypt message not as expected. ntlm.c:1176: Test failed: DecryptMessage returned SEC_E_INVALID_TOKEN, not SEC_E_OK.

Feb. 9, 2011

Re: [PATCH 2/5] d3dx9_36: Implemented ID3DXConstantTable_SetIntArray and ID3DXConstantTable_SetInt

by Travis Athougies

On Feb 8, 2011, at 4:10 AM, Matteo Bruni <matteo.mystral(a)gmail.com> wrote: >> + /* We need the for loop, since we need to convert the >> integer to a float */ >> + for (i = 0; i < count && i < desc.RegisterCount; i++) >> + { >> + row[0] = (float)n[i]; > > The cast is unneeded, and the comment above is too. If it's being passed into set*shaderconstantf, shouldn't it need to be cast to a float? Or am I mishnderstaning you? Travis

Feb. 9, 2011

Re: comctl32/tests: Use a function instead of a macro for CheckSize test

by Vitaliy Margolen

On 02/08/2011 04:45 PM, EG Galano wrote: > This fixes the warnings: > comctl32/tests/tab.c:525: warning: logical `&&' with non-zero constant > will always evaluate as true > comctl32/tests/tab.c:545: warning: logical `&&' with non-zero constant > will always evaluate as true > comctl32/tests/tab.c:568: warning: logical `&&' with non-zero constant > will always evaluate as true > comctl32/tests/tab.c:1015: warning: logical `&&' with non-zero > constant will always evaluate as true > > for me on Debian gcc 4.3.2 Time to file bug for the compiler. Yes macro looks bad and that's not the way I would have done it now, but there is not a single place where "non-zero constant" is used directly with logical AND. Also what Nikolay said. Vitaliy.

Feb. 9, 2011

Re: d3dx9: Implement D3DXFloat16to32Array and D3DXFloat32to16Array.

by Charles Davis

On 2/8/11 2:45 PM, Misha Koshelev wrote: > I don't have msvc. Please suggest how I can check this. By looking on MSDN. All the library functions that Visual C++ supports are documented there. Chip

Feb. 9, 2011

Re: comctl32/tests: Use a function instead of a macro for CheckSize test

by Nikolay Sivov

On 2/9/2011 02:45, EG Galano wrote: Main problem with that - you will lost ok() context in test output, line numbers will be always the same. Look like it's resolved in other places - a thin macro to pass a line number to a function and ok_(, <linenumber>) test inside.

Feb. 8, 2011

Re: user32: Add a test for reparenting a WS_POPUP window to a WS_CHILD parent. Take 2.

by Alexandre Julliard

Dmitry Timoshkov <dmitry(a)codeweavers.com> writes: > This test is supposed to show the root of the problem with the test > application attached to the bug 13951. It appears that the check performed > by server/window.c,is_top_level_window() is not correct (which leads to > a SetWindowPos() failure to set foreground window), and probably should be > replaced by simple (win->style & (WS_CHILD | WS_POPUP)) != WS_CHILD. Maybe, but the test doesn't really demonstrate that. -- Alexandre Julliard julliard(a)winehq.org

Feb. 8, 2011

Re: d3dx9: Implement D3DXFloat16to32Array and D3DXFloat32to16Array.

by Misha Koshelev

I don't have msvc. Please suggest how I can check this. Thank you. Yours Misha On Feb 8, 2011 4:01 PM, "Stefan Dösinger" <stefandoesinger(a)gmx.at> wrote: Am Dienstag 08 Februar 2011, 17:52:06 schrieb Misha Koshelev: > switch (fpclassify(in)) Is this portable? Does it work on msvc too?

Feb. 8, 2011

Re: Try to Implement my first Stub function - AbortPrinter() - (try 2).

by Nicolas Le Cam

2011/2/8 Loïc Maury <lmaury(a)gmail.com> > > Hello Mr.Timoshkov, > > Thank you for your reply. > >> Loīc Maury<lmaury(a)gmail.com> wrote: >> >> >> After the various comments, I have modified the patch. >>> >> First of all set your tab size to 8, and ask your editor to not use tabs >> at all. >> > I have modified my editor, but I don't know if it 's correct now ? > >> + TRACE("(%s, %s, %p, %d, %d)\n",debugstr_w(printer->name) >>> >>> + ,debugstr_w(printer->printername) >>> + ,printer->backend_printer >>> + ,printer->queue->ref >>> + ,list_count(&printer->queue->jobs)); >>> >> TRACE() with the API parameters usually is the very first statement in >> the API implementation, comma should be placed at the end of the >> statement, >> not before. >> > Ok, I have remplaced this TRACE(). > >> + GetPrinterW(hPrinter, 2, NULL, 0,&needed); >>> >>> + pi2 = HeapAlloc(GetProcessHeap(), 0, >>> needed); >>> + GetPrinterW(hPrinter, 2, (LPBYTE)pi2, >>> needed,&needed); >>> >> You need to check the return value of GetPrinterW() and handle the errors. >> > Ok > >> + if(pi2) >>> + HeapFree(GetProcessHeap(), 0, pi2); >>> >> NULL check before HeapFree() is not needed. >> > Ok, I have removed the NULL check. > > + TRACE("return %d\n", ret); >>> >> This trace is redundant. >> > Ok, I have removed this TRACE(). > > I have make an other patch. > > Thank you > > Loīc > > > > > Hi Loïc, In addition to Dmitry and Andrew comments, you're adding a lot of trailing spaces (git apply output : warning: squelched 22 whitespace errors warning: 27 lines add whitespace errors.), there's also no need for an extra newline after the end label. -- Nicolas Le Cam

Feb. 8, 2011

Re: d3dx9: Implement D3DXFloat16to32Array and D3DXFloat32to16Array.

by Stefan Dösinger

Am Dienstag 08 Februar 2011, 17:52:06 schrieb Misha Koshelev: > switch (fpclassify(in)) Is this portable? Does it work on msvc too?

Feb. 8, 2011

Re: Wine and Summer of Code 2011

by Eric Durbin

On Tue, Feb 8, 2011 at 12:04 PM, Dan Kegel <dank(a)kegel.com> wrote: > > http://www.google-melange.com/document/show/gsoc_program/google/gsoc2011/fa… > says projects can apply starting on Feb 28. > > In preparation, I moved the 2010 projects from > http://wiki.winehq.org/SummerOfCode > to > http://wiki.winehq.org/SummerOfCode/PreviousProjects > > I also added a DSS Provider project idea, removed the gstreamer > idea since that's implemented, and tidied up the dxdiag idea a bit. > > Who's interested in mentoring this year? And can y'all think of any > fresh ideas for the suggested projects list, or ideas that should be > retired? > - Dan > > > I would be interested in mentoring again this year.

Feb. 8, 2011

Re: Wine and Summer of Code 2011

by Austin English

On Tue, Feb 8, 2011 at 10:17, Juan Lang <juan.lang(a)gmail.com> wrote: > Hi Dan, > >> Who's interested in mentoring this year? And can y'all think of any >> fresh ideas for the suggested projects list, or ideas that should be >> retired? > > Isn't IUri implemented? That idea looks like it could be retired. > > Also, the DOS VGA modes idea: how is that affected by the DOSBox integration? On that same note, how about DOSBox integration? I know Alexandre started it, but my understanding was that he was hoping for someone else to take over, potentially with some work in DosBox itself as well. -- -Austin

Feb. 8, 2011