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

Re: user32/tests: Add tests for removing already seen messages from the queue.
by Marvin March 17, 2015

March 17, 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=12229 Your paranoid android. === w7u (32 bit msg) === msg.c:12230: Test failed: unexpected hwnd 0017014A msg.c:12252: Test failed: bad time 309b0/309c0

1 0

An issue with msvcp110 test's design
by YongHao Hu March 16, 2015

March 16, 2015

Hi, all Thanks for Piotr's comment on my GSOC ideas. In order to have a good knowledge of functions from tr2 namespace, I had made a test and written some functions' implementation. However, I had an issue with msvcp test's design when I need to free library of msvcp90 and load msvcp110. 619 static HMODULE msvcr, msvcp; 620 #define SETNOFAIL(x,y) x = (void*)GetProcAddress(msvcp,y) 621 #define SET(x,y) do { SETNOFAIL(x,y); ok(x != NULL, "Export '%s' not found\ n", y); } while(0) In my patch, I use a hmodule variable named msvcp90 to save msvcp90 hmodule and then reuse msvcp to save msvcp110. As far as I am concerned, the name msvcp90 or msvcp90_saver is not good enough. Could you give it a good name? And the other method I came up with is undefining SETNOFAIL and SET and using msvcp110 to define it again, which I think is ugly. By the way, my crosstest on all the testbots[1] show that msvcp110.dll or msvcrt110.dll not installed. It seems testbot doesn't install msvcp110. I will be grateful for any advice you can provide. [1]: https://testbot.winehq.org/JobDetails.pl?Key=12146&log_110=1#k110

5 6

Re: [PATCH] ddraw: Add validation and tests for DDBLT_ROP (v2).
by Marvin March 16, 2015

March 16, 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=12219 Your paranoid android. === build (build) === Patch failed to apply

1 0

Re: [PATCH 2/5] d3d9/tests: Add a test for UpdateTexture.
by Stefan Dösinger March 16, 2015

March 16, 2015

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am 2015-03-10 um 19:29 schrieb Matteo Bruni: > + /* This works on AMD but fails on Nvidia (it draws black). > */ + /* {8, 8, 8, 8, 4, 4, D3DFMT_A8R8G8B8, D3DFMT_R5G6B5}, > */ /* Different format */ What does AMD draw in this case? Does it convert the format from A8R8G8B8 to R5G6B5 like StretchRect would? Or does it just memcpy? In the latter case I'd expect a memory corruption due to the size mismatch. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBAgAGBQJVAA/0AAoJEN0/YqbEcdMwKlcP/RVMf9nx/D4Emx4Viu9U3iGZ P1XV8AvAjKxJ/BNQ58HGCVH+9X3GXLLOJiv3H5CanOt7MxpwT+MX9hADYz9aZZTr lo45Gp5H4i2d9X2fNNNTcoY7y0lYV1R3YCEcSiQlick3o/0KTXPV5a6+m+Z1QDcV hlSjkj0SpZICVpzUwuU/FQuSyKE/ihcyZNGwtfxt+q+QdJDPzDp86QBV+s80snKD Yzsif4pAU2bLzrrM181qsRv0MB8mvfmfT25gPtjrZq5JekRe+WWhZ2uLrqQhkxAH y6A3TeVuIiRY0qwOMe05eEXAEZjLxDswJJDD7fLps3XcYYkpbIBLECjNqHhEFtK1 7DCjb7nQtJBFy+cX4ZBxRZ6X15Rsri6iQfP04e2/V44JWZrO6EHQvEybUw2i57+Y R1yLsf3KzDzpcy9At7gH4sFIvjDfuCROlblsdYmEwacwq+rO8kyRI6DhixPRtWDG beW6eMRUAVVyeQvXVCuLW6uPIpWuEdpmQZ7neug1YFfdqsG4YfR76RgQv6Gkj32V f3G/wGri8fF6PUWavZscdVdcD51LbkUQRyNy8CpRpxJe5AABGaecVj+nirskoKEo qFXcWxSUK5YHQ055WWn0UwQ2NabX62FNTki35tNqbZru/iivVbsEb4kcYeaYsMe9 TZrdk2aXSVrmHbd52LSi =xL1U -----END PGP SIGNATURE-----

2 7

Re: [PATCH 5/5] ddraw/tests: Add tests for DDBLT_ROP.
by Matteo Bruni March 16, 2015

March 16, 2015

2015-03-16 14:48 GMT+01:00 Stefan Dösinger <stefan(a)codeweavers.com>: > --- > dlls/ddraw/surface.c | 16 ++++++++ > dlls/ddraw/tests/ddraw7.c | 98 +++++++++++++++++++++++++++++++++++++++-------- > 2 files changed, 98 insertions(+), 16 deletions(-) This patch also changes the implementation while the title suggests only additions to the tests. This might come as a surprise a few months down the line in the (admittedly unlikely) case this causes regressions or something similar, so I'd either change the name of the patch or split the implementation change from the test part. Again, it's not that important in this specific case so I don't care much.

2 1

Re: [PATCH] wine.inf: Add some knownfolders to the registry
by Alexandre Julliard March 16, 2015

March 16, 2015

Andrew Eikum <aeikum(a)codeweavers.com> writes: > This time with a cleaned up win8 testbot. > > dlls/shell32/tests/shellpath.c | 212 ++++++++++++++++++++--------------------- > loader/wine.inf.in | 24 +++++ > 2 files changed, 129 insertions(+), 107 deletions(-) Shouldn't this be generated from the data we already have in shell32? -- Alexandre Julliard julliard(a)winehq.org

2 2

Re: [PATCH] dsound: Apply filters before sound is multipled to speakers.
by Andrew Eikum March 16, 2015

March 16, 2015

On Sun, Mar 15, 2015 at 05:51:32PM +0000, Mark Harmstone wrote: > - for (i = 0; i < count; i++) > - for (channel = 0; channel < dsb->mix_channels; channel++) > - dsb->put(dsb, i * ostride, channel, get_current_sample(dsb, > - dsb->sec_mixpos + i * istride, channel)); > + float* buf; > + > + buf = HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY, count * dsb->mix_channels * sizeof(float)); Is it necessary to zero this buffer? I think the whole thing gets overwritten in the following lines. I'm a bit concerned that splitting the processing out into another buffer will affect performance. Similarly in the resampling case below, you're adding another loop over every sample. We already have a performance issue introduced by the higher quality resampler: https://bugs.winehq.org/show_bug.cgi?id=30639 Could we take the slower path in each case only if we have filters to apply? BTW do you have an application that uses DMOs? I'd be curious to take a look and add it to my list of dsound test applications. Andrew

1 0

Re: d3drm: Add tests for querying between IDirect3DRM interfaces
by Stefan Dösinger March 16, 2015

March 16, 2015

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am 2015-03-15 um 11:45 schrieb Aaryaman Vasishta: > + hr = IUnknown_QueryInterface(base_iface, tests[i].iid, (void > **)&iface1); It looks like your mail client line-wrapped the patch. Unfortunately it won't apply this way :-( . > + if (!GetProcAddress(GetModuleHandleA("d3drm.dll"), "Direct3DRMCreate")) > + { > + win_skip("Direct3DRMCreate not available, skipping test.\n"); > + return; > + } You should not need this. If Direct3DRMCreate is not available InitFunctionPtrs() will return false and your test is never run. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBAgAGBQJVBWQHAAoJEN0/YqbEcdMwq/IP/20pRxWk3tYXzbZmKILCNH65 JRTlm/fvqtO/Q+Q1C57WkOMiusE2w4i25/AQNsaIYOW/e8nATIGni3PPRlK6v9gT /JzzntJLD0I4mEOBxrZvnQhQ3iZ91YMla+Sg326b3PkwONEfkWhORP4YGrkt75is 5YNCKVZG/msPT7IxUIbZji7K8McDEX/OsLLFEFDr3mg4i5z7WMxAJVBIxOGuN8re 2N2TKGfcWLEIaE5Hpyh+r9xYbS8x7JyIN5md5EdQrnBR/9KMn18VFLWX3u+4jfm3 o8uGOjhhOUqjjtzYomRe317l3ujtqjora6xNZ5upGrcxwBWSZKDB5gk233o/62W4 FJuHGeHnCaUgctBUYQstFHr9FsBFt3KkgFqhds3nnSz4MF3/I46+X7KprCiAIaMZ q/YNasm8lGGAdNZp5DkViq9K06umOG7iz+21XY+dP9hRrTVqq11eEqEmveqRHU2+ GuOiNcKeo45Sj+4yl+r78fD3JNN2Ltki3O/z/KVjfE2aE+e5cHKn1oGAEFUaGzDl zHd2KcI6W6e6eDy3qgo8TezfiQYogQzTriyVd9Kyxi9CxKAVnZOrskKoFj2tT3M4 PEn2qX13OtcBcIaO83QJdHtwxg1g3yQ771uIPK0MHU1zjSHAz2sHf+amn6/gF8N+ MT9amN40cQrxGXpaplzm =d/6Z -----END PGP SIGNATURE-----

3 2

Re: vcomp: Add _vcomp_enter_critsect, _vcomp_leave_critsect, omp_init_lock stubs
by Michael Stefaniuc March 16, 2015

March 16, 2015

Hello Andrey, On 03/14/2015 03:00 PM, Andrey Gusev wrote: > Based on an old patch from Dan Kegel with the addition. > Fixes https://bugs.winehq.org/show_bug.cgi?id=36457 > > vcomp-Add-_vcomp_enter_critsect-_vcomp_leave_critsec.txt you can't just stub out locking functions. It is far superior to have a program crash straight ahead than to silently corrupt data. That's the reason the initial submission by Dan was rejected. I don't find any documentation on a quick search for this functions. But they look to me just like wrappers around the standard EnterCriticalSection / LeaveCriticalSection functions. You could try that out and see if your application is happy with that. bye michael > > > From b4c7617b344ee0dcdb058030b19ede5101787824 Mon Sep 17 00:00:00 2001 > Message-Id: <b4c7617b344ee0dcdb058030b19ede5101787824.1426341234.git.andrey.goosev(a)gma…> > From: Andrey Gusev <andrey.goosev(a)gmail.com> > Date: Sat, 14 Mar 2015 15:53:29 +0200 > Subject: [PATCH] vcomp: Add _vcomp_enter_critsect, _vcomp_leave_critsect, > omp_init_lock stubs > > --- > dlls/vcomp/main.c | 15 +++++++++++++++ > dlls/vcomp/vcomp.spec | 6 +++--- > 2 files changed, 18 insertions(+), 3 deletions(-) > > diff --git a/dlls/vcomp/main.c b/dlls/vcomp/main.c > index ab2a372..3c43a4e 100644 > --- a/dlls/vcomp/main.c > +++ b/dlls/vcomp/main.c > @@ -71,6 +71,11 @@ double CDECL omp_get_wtime(void) > return GetTickCount() / 1000.0; > } > > +void CDECL omp_init_lock(int lock) > +{ > + TRACE("(%d): stub\n", lock); > +} > + > void CDECL omp_set_dynamic(int val) > { > TRACE("(%d): stub\n", val);

1 0

Re: [PATCH 1/2] riched20: Implement IOleWindow interface.
by Nikolay Sivov March 16, 2015

March 16, 2015

On 16.03.2015 10:36, Jactry Zeng wrote: > + create_interfaces(&w, &reOle, &txtDoc, NULL); > + hres = IRichEditOle_GetClientSite(reOle, &clientSite); > + ok(hres == S_OK, "IRichEditOle_QueryInterface: 0x%08x\n", hres); > + refcount1 = get_refcount((IUnknown *)clientSite); > + todo_wine ok(refcount1 == 1, "got wrong ref count: %d\n", refcount1); > + > + hres = IOleClientSite_QueryInterface(clientSite, &IID_IOleClientSite, (void **)&clientSite1); > + ok(hres == S_OK, "IOleClientSite_QueryInterface: 0x%08x\n", hres); > + ok(clientSite == clientSite1, "Should not return a new pointer.\n"); > + refcount1 = get_refcount((IUnknown *)clientSite); > + todo_wine ok(refcount1 == 2, "got wrong ref count: %d\n", refcount1); Hi, Jactry. Do we have a test somewhere showing that IRichEditOle cannot be queried directly for IOleClientSite and IOleWindow?

2 2