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

Re: [PATCH 1/2] winemac: Check the window's display link after adding it as a child of another window, which may order it on screen.
by Huw Davies Nov. 10, 2015

Nov. 10, 2015

On 10 Nov 2015, at 02:31, Ken Thomases wrote: > This fixes a problem where child windows ("owned" windows in Windows parlance) > would never display their contents on OS X 10.8 or earlier. Beginning with > 10.9, Cocoa calls -windowDidChangeOcclusionState: when the window becomes > visible, which is why they display on that version and later. > --- > Reported by Huw Davies. Works great, thanks! Huw.

1 0

Re: [PATCH 1/2] winemac: Use CVDisplayLink to limit window redrawing to the display refresh rate. (try 2)
by Huw Davies Nov. 10, 2015

Nov. 10, 2015

Thanks Ken, Just confirming that this series now builds for me on 10.7 Huw.

3 3

How to change the ACCESS_MASK of a thread?
by Bruno Jesus Nov. 9, 2015

Nov. 9, 2015

Hi all. I have a simple question that I could not find an answer after looking for a lot. All I need is to remove the THREAD_SUSPEND_RESUME flag from the ACCESS_MASK to add it again later. info.GrantedAccess = 0; NtQueryObject(GetCurrentThread(), ObjectBasicInformation, &info, sizeof(info), NULL); if (info.GrantedAccess & THREAD_SUSPEND_RESUME) { changed = TRUE; info.GrantedAccess &= ~THREAD_SUSPEND_RESUME; //magic here } This is about fixing bug 3930 where SuspendThread is called from inside a callback deadlocking the game. If I can prevent the SuspendThread from running it will work. Best wishes, Bruno

1 0

Re: [PATCH] msvcp120: implement _Cnd_* functions (try 2)
by Piotr Caban Nov. 9, 2015

Nov. 9, 2015

On 11/09/15 16:38, Daniel Lehman wrote: > +static int (__cdecl*p__Mtx_init)(_Mtx_t*); This function has different header. > + if (0) /* causes crash in later_Cnd_* calls on Windows */ > + p__Cnd_destroy(&cm.cnd); This is not what I meant. I'm attaching a patch (generated on top of yours with the test I had in mind). It only shows the problem, the test needs more work. > + for(i = 0; i < NUM_THREADS; i++) > + p__Thrd_create(&threads[i], cnd_bcast_thread, (void*)&cm); ... > + p__Mtx_lock(&mtx); > + cm.go = 1; > + p__Mtx_unlock(&mtx); Please add some kind of synchronization to make sure the thread is waiting on condition variable before setting cm.go to 1. > + cm.go = 0; > + p__Thrd_create(&threads[0], cnd_signal_thread, (void*)&cm); > + p__Mtx_lock(&mtx); > + while(!cm.go) { > + r = p__Cnd_wait(&cnd, &mtx); > + ok(!r, "failed to wait\n"); > + } Here's a similar synchronization problem. In most cases p__Cnd_wait will not be called. > + diff = p__Xtime_diff_to_millis2(&after, &before); > + ok(r == 2, "should have timed out\n"); > + ok(diff > 2000 - TIMEDELTA, "got %d\n", diff); This test failed once on my computer (in ~10 runs, diff == 1800). Thanks, Piotr

1 0

Re: [PATCH v2 3/3] msvcp140: Add MSVC 2015 C++ support DLL
by Piotr Caban Nov. 9, 2015

Nov. 9, 2015

On 11/06/15 13:51, Martin Storsjo wrote: > static void init_cxx_funcs(void) > { > @@ -77,6 +125,11 @@ static void init_cxx_funcs(void) > > if (!hmod) FIXME( "%s not loaded\n", MSVCRT_NAME(_MSVCP_VER) ); > > +#if _MSVCP_VER >= 140 > + MSVCRT_operator_new = operator_new; > + MSVCRT_operator_delete = operator_delete; > + MSVCRT_set_new_handler = (void*)GetProcAddress(hmod, "_set_new_handler"); > +#else > if (sizeof(void *) > sizeof(int)) /* 64-bit has different names */ > { > MSVCRT_operator_new = (void*)GetProcAddress(hmod, "??2(a)YAPEAX_K@Z"); > @@ -113,6 +166,7 @@ static void init_cxx_funcs(void) > #endif > #endif /* _MSVCP_VER >= 110 */ > } > +#endif > } Please also initialize critical_section related functions. Thanks, Piotr

1 0

Re: [PATCH] msvcp120: implement _Cnd_* functions
by Piotr Caban Nov. 9, 2015

Nov. 9, 2015

On 11/07/15 00:58, Daniel Lehman wrote: > +void __cdecl _Cnd_destroy(_Cnd_t *cnd) > +{ > + MSVCRT_operator_delete(*cnd); > +} What happens when _Cnd_destroy is called while someone waits on condition variable? > +int __cdecl _Cnd_timedwait(_Cnd_t *cnd, _Mtx_t *mtx, const xtime *xt) > +{ > + CONDITION_VARIABLE *cv = &(*cnd)->cv; > + LARGE_INTEGER timeout; > + NTSTATUS status; > + > + InterlockedExchangeAdd( (LONG *)&cv->Ptr, 1 ); > + _Mtx_unlock(mtx); > + > + timeout.QuadPart = (ULONGLONG)_Xtime_diff_to_millis(xt) * -10000; > + status = NtWaitForKeyedEvent(keyed_event, &cv->Ptr, FALSE, &timeout); > + if (status) > + { > + if (!interlocked_dec_if_nonzero( (LONG *)&cv->Ptr )) > + NtWaitForKeyedEvent( keyed_event, &cv->Ptr, FALSE, NULL ); You need to update status here. Thanks, Piotr

2 1

Re: [1/5] gdi32: Fix the advance width in synthetic bold glyphs when using scaling factors. (try 2)
by Alexandre Julliard Nov. 9, 2015

Nov. 9, 2015

Akihiro Sagawa <sagawa.aki(a)gmail.com> writes: > Try 2: > - Rename the vector normalization function name and returns > calculated value. > - Avoid division by zero. > Thanks Huw and Henri. > > Signed-off-by: Akihiro Sagawa <sagawa.aki(a)gmail.com> > --- > dlls/gdi32/freetype.c | 50 ++++++++++++++++++++++++++++++++++--------------- > dlls/gdi32/tests/font.c | 10 ++++++++++ > 2 files changed, 45 insertions(+), 15 deletions(-) It breaks the tests here: ../../../tools/runtest -q -P wine -T ../../.. -M gdi32.dll -p gdi32_test.exe.so font && touch font.ok font.c:389: Test failed: cx 221 != 243 font.c:389: Test failed: cx 221 != 243 -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH] odbccp32: Implement SQLGetPrivateProfileStringW/SQLGetPrivateProfileString (try 3)
by Alexandre Julliard Nov. 9, 2015

Nov. 9, 2015

Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> writes: > Corrected formatting. > Use strlen where approperate. > Improved tests to show that we cannot write directly to the buffer passed in. It doesn't work here: ../../../tools/runtest -q -P wine -T ../../.. -M odbccp32.dll -p odbccp32_test.exe.so misc && touch misc.ok misc.c:273: Test failed: incorrect string 'testing' make[1]: *** [misc.ok] Error 1 -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH 5/5] d3d9/tests: Improve the shade mode test.
by Henri Verbeet Nov. 9, 2015

Nov. 9, 2015

On 8 November 2015 at 20:25, Matteo Bruni <mbruni(a)codeweavers.com> wrote: > dlls/d3d9/tests/visual.c | 275 ++++++++++++++++++++++++++++++++--------------- > 1 file changed, 191 insertions(+), 84 deletions(-) > I think most of this test would work for d3d8 and ddraw as well. Could you please port it? > + {&novs, &ps_3, D3DPT_TRIANGLESTRIP, D3DSHADE_FLAT, FALSE, 0x00ff0000, 0x0000ff00}, > + {&novs, &ps_3, D3DPT_TRIANGLESTRIP, D3DSHADE_GOURAUD, FALSE, 0x000dca28, 0x000d45c7}, These fail here on Windows: visual.c:8338: Test failed: Test 14 shading has color0 00ffffff, expected 00ff0000. visual.c:8340: Test failed: Test 14 shading has color1 00ffffff, expected 0000ff00. visual.c:8338: Test failed: Test 15 shading has color0 00ffffff, expected 000dca28. visual.c:8340: Test failed: Test 15 shading has color1 00ffffff, expected 000d45c7. IIRC SM3 vertex and pixel shaders need to be used together, or results are undefined / implementation defined.

2 1

Re: [PATCH 3/9] d3d11/tests: Port test_texture() from d3d10core.
by Henri Verbeet Nov. 9, 2015

Nov. 9, 2015

On 6 November 2015 at 18:38, Józef Kucia <jkucia(a)codeweavers.com> wrote: > + for (i = 0; i < 4; ++i) > + { > + for (j = 0; j < 4; ++j) > + { > + color = get_texture_color(backbuffer, 80 + j * 160, 60 + i * 120); > + ok(compare_color(color, bitmap_data[j + i * 4], 1), > + "Got unexpected color 0x%08x at (%u, %u), expected 0x%08x.\n", > + color, j, i, bitmap_data[j + i * 4]); > + } > + } It's not a terribly high priority, but Matteo recently introduced get_rt_readback()/release_surface_readback() in the d3d8/9 tests. We'll want to introduce that interface in the d3d10/11 tests as well.

1 0