May 2014 - Wine-Devel - list.winehq.org

Re: [2/2] dpnet: Implement IDirectPlay8Client Initialize (try 2)
by Stefan Dösinger May 7, 2014

May 7, 2014

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am 2014-05-07 03:39, schrieb Alistair Leslie-Hughes: > Hi, Just a rebase. This looks ok to me. It depends on the test added by patch 1 though, so it'll have to wait until the test is sorted out. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQIcBAEBAgAGBQJTajllAAoJEN0/YqbEcdMwiTYP/icYQO8oZyCNCRTwoJ3Vdsbl mBbpWrXI2pT0ASEj30+WdJ7a7iJvZXc8tVSvvV8UtbQPLBRGI5Y9asJi9rKpNCLb ksP5vXKmsSxUewGGbX+rtYqRS/hL67NSTai1mb/hxkqGmmJqLbchzM3NP+GX4fvv e9w9p/DpfcOXGCrAtvuKSCzFPDwn60IVIIZVXTCnEs6KR3Vvq+NGnxmV9kwtEjCj idZI0cgPXwR/R0ybJdhFJpU3GHiyNWPAhI6kZcXNCv+kP90A6aDJ8rgd3SvcNEyt oxROFT7PsHUJmC72xixpscGMM5DLDijffVmtooTIgGHsM2iC8Yi4B5s+U77P0sBJ 3LGfYH68AIZtAA187A21WM+WLOtWV5vjihHIHxEO9UFmdI1G88f/9bG0n5YRlzkf Rh4xevGBw7/oks3q08uEMY2EwQdCL/NSx1HWNBtbohBKClZv+373cXEY9+lXPbl2 vA6P0KCUHf/JRM9v2b19UG7vffklCoymroQ8i4TD8qksBHdGq640MUqqyQaMfuu4 ZBrJH3Q+VTe79fcPTG4AyQrEUb1Xy+ohNIY5gN4e8UuvydEpcIId5a0Isu04u9Xr 02tjdraySJaW7ELq5+NoegH30O2Q8aJzPkY7aS2lWe+Kt8lgAUG+KKQXF4fKoeSH x6wbbtc5ojYjXXS91b8S =59Gh -----END PGP SIGNATURE-----

1 0

Re: [1/2] dpnet: Add IDirectPlay8Client tests (try 2)
by Stefan Dösinger May 7, 2014

May 7, 2014

Am 2014-05-07 03:39, schrieb Alistair Leslie-Hughes: > Hi, > Changed test_init_dp to return a boolean. Sorry, I was a bit slow with replying to your previous mail. Is there anything that prevents you from having a helper function that creates a new client object for each test? If you share one client object and the tests modify the state, then the tests end up depending on each other. We had a similar situation in the d3d8/9 visual tests (and still have in ddraw:visual), and it leads to hard to debug test failures. If the client object takes too long to create (I'd say > 0.25 seconds) then that's a reason to keep the client around. Otherwise creating a fresh one for each test and destroying it afterwards will really help you in the long run. Somewhat related: > +static BOOL test_init_dp(void) > ... > + hr = CoInitialize(0); > ... > +static void test_cleanup_dp(void) > ... > + CoUninitialize(); > +} I'd move that to the test's main function. (And no, I don't think there's a need to call CoInitialize / CoUninitialize more than once.) Cheers, Stefan

1 0

Re: dplayx: Initial msgtid variable
by Stefan Dösinger May 7, 2014

May 7, 2014

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am 2014-05-07 04:08, schrieb Alistair Leslie-Hughes: > + obj->msgtid = 0; > obj->ref = 0; > obj->refA = 0; > obj->ref2 = 0; IMO it would be better to use HEAP_ZERO_MEMORY in the HeapAlloc call a few lines above. This problem will keep coming back when more members are added. Cheers, Stefan -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQIcBAEBAgAGBQJTajXtAAoJEN0/YqbEcdMwF6kP/iJveqvV58YLeyd2bpJKOjim H2dF+Cll8hAayzziokzOhgf6BoMikhnwqQHtXSgZKuGjQl7oF6sy2qCr5wvfl13V bC9Me4DbXE9PEwT2ArEDiGkejg0G7KQ8qFVo6vFNIaRBEt0mgH0bFUntGPtZZgbV 9qT3KgoNGqOw4iI8rzh4ATabu9NizkI+b2UxMY+9lEf/RSur9pnX/XQhjRIvzZKS YT4exS5bdjZXcSN9aEPax9f+VmYKRyVBvo6O5PfdkqQNarFMc5bw6Utkab/0VbQb j5IM5dRdevTsqx1ZdqzAxlDCHEtRlQWR07A0unsg6JuCX8Rwn+lAh78NTozHWwIV qG31CoHV37WJggleZc2Rk5oSatlAVARwX/f2qJ9GA3fv5muJkpThc8HcEvTN1e/V zSmWBLljyF0I6a6a4P8k2Ch9ZmDnmDvIaGaJshnETt+ypgK2T2QI35pb2bh/pGmC ltWvEwRpbATmftl62Bu4511Afo8AGcXEa83SvfukPmHaAtiJ2TNI9/c1vCP/Pvke TeARP1gtvSh3Q7pwcXPIaAry/RLEYmMPJDAoKRkc+FM2O/4gNiUDUBIDeEpKJaB1 BM2HE0/TCrrIEkfE1RI9JseX79uylagiStAIUgKkqMfJxrXA50VX0bLnQq/qtayD vRfIAuYYb3hvpNll1gIT =rA1r -----END PGP SIGNATURE-----

1 0

Re: [PATCH 2/2] gdi32: Report unsubstituted font face names.
by Alexandre Julliard May 7, 2014

May 7, 2014

Alex Henrie <alexhenrie24(a)gmail.com> writes: > @@ -5765,6 +5765,11 @@ static BOOL enum_face_charsets(const Family *family, Face *face, struct enum_cha > else > strcpyW(elf.elfFullName, family->FamilyName); > } > + /* Make sure that MS Shell Dlg says "MS Shell Dlg" and not "Tahoma" */ > + if (unsubstituted_face_name) > + { > + lstrcpynW(elf.elfLogFont.lfFaceName, unsubstituted_face_name, LF_FACESIZE); > + } So what happens to the full name in that case? -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: comctl32: Make Treeview Notify the parent on a return key press
by Nikolay Sivov May 7, 2014

May 7, 2014

Quick testing doesn't show this notification, so I'm not convinced it's a right change. Please add a test.

1 0

Re: [1/2] dpnet: Add IDirectPlay8Client tests
by Stefan Dösinger May 7, 2014

May 7, 2014

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am 2014-05-05 08:46, schrieb Alistair Leslie-Hughes: > +static IDirectPlay8Client* client = NULL; > ... > + hr = CoCreateInstance(&CLSID_DirectPlay8Client, NULL, CLSCTX_INPROC_SERVER, &IID_IDirectPlay8Client, (void **)&client); > + ok(hr == S_OK, "CoCreateInstance failed with 0x%x\n", hr); > ... > + hr = IDirectPlay8Client_Close(client, 0); > + ok(hr == S_OK, "IDirectPlay8Client_Close failed with %x\n", hr); Please avoid such dependencies between different tests. If creating and releasing the client is something that every test has to do you can add helper functions for this task. > + for (i=0;i<items;i++) > + { > + trace("Found Service Provider: %s\n", wine_dbgstr_w(serv_prov_info->pwszName)); > + trace("Found guid: %s\n", wine_dbgstr_guid(&serv_prov_info->guid)); > + > + serv_prov_info++; > + } > + > + serv_prov_info -= items; /* set pointer back */ > + ok(HeapFree(GetProcessHeap(), 0, serv_prov_info), "Failed freeing server provider info\n"); You could use serv_prov_info as an array instead of doing the pointer arithmentic twice. E.g. trace("Found guid: %s\n", wine_dbgstr_guid(&serv_prov_info[i].guid)); > + hr = IDirectPlay8Client_EnumHosts(client, &appdesc, host, local, NULL, 0, INFINITE, 0, INFINITE, NULL, &async, 0); > + todo_wine ok(hr == DPNSUCCESS_PENDING, "IDirectPlay8Client_EnumServiceProviders failed with 0x%08x\n", hr); > + todo_wine ok(async, "No Handle returned\n"); > + > + hr = IDirectPlay8Client_CancelAsyncOperation(client, async, 0); > + ok(hr == S_OK, "IDirectPlay8Client_CancelAsyncOperation failed with 0x%08x\n", hr); Are you sure that this always returns DPNSUCCESS_PENDING? Couldn't it also return a result instantly? Overall I don't see the point in this test at the moment. As it is now it doesn't say much. This is just an opinion though, I'll leave the decision up to you. > + todo_wine ok(caps.dwMaxEnumPayloadSize == 983, "expected 983, got %d\n", caps.dwMaxEnumPayloadSize); Out of curiosity, do you have any idea where this value comes from? -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQIcBAEBAgAGBQJTaTX9AAoJEN0/YqbEcdMwB6AP/jd1T0OLGPXO3pgF9boIHp6i yQTDCgf7Wq8zKJuxDZtOzfGzVbYqVeN8TDx9c3oD3RSru9bBIJ9Bjeu3hKqvL2Qo 4JRJZyQn665RymbKo/S9ivXBHmd9cuh4tfdTxHU7n0nVXICCNDniBxfAwr2HkkeB iyvMe8Ga3SPVCL+33XZAniPPFnfnEYkaqU42CTUUK2hlYyzb4bScihgdrUqoX36k RtvK2UKxOBT5E9nd0VIS/b8fF4uXFIzD7M6oEuzLObXL6XGsRNl2mtMpZ0ztlFpy PIF1AbFyK4tQA3obsTSWUVJVlQFLCurvqg2klLogewdyVy4wE4h0/oOCRCPhuVbR baJf0by5ZC4Os4YIZIzN0OOy0N3A+bMQI1VdvQTuMPwF+dYk268p0t77SN6F2gfa ts9MyQxYrhrao3+oI4Hw2ouYUxIYj+WxNSVCb01TjUrMVOMAeD1Ksttbk1g5ADdZ +DbV9CWuOhISjXh5OT8ToH0FxrImrby3FDyX/UFf2KPzwzKoX2kNyjtSHguPIOko 3jOnuVh8HsD7jX0sQqMMkdwsESdfTCwIHXLhbE4C08yxsu8ByNhFHme3oX4EcyPf oHRUA+w8Yr8rr+wSU1fpPEb79ZnMn5xfR7xZQKzX5af0ICta6k3UjtSeAuE6P1xX HFJrTsoKGibhnm6z2wZH =sF7u -----END PGP SIGNATURE-----

2 1

Re: [PATCH 2/2] gdi32: Report unsubstituted font face names.
by Marvin May 7, 2014

May 7, 2014

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=6860 Your paranoid android. === wvistau64 (32 bit font) === The test timed out

1 0

Testbot returning 500 - Internal server error message
by Bruno Jesus May 7, 2014

May 7, 2014

Hi all. While trying to upload a test in the command line parameter input the server returned a 500 error. Then I went back to the index and it's working but clicking on any test key will result in the error message as well. This may be an intermittent failure and I'm reporting it just to check because I don't recall any scheduled maintenance emails. Example: https://newtestbot.winehq.org/JobDetails.pl?Key=6858 Best regards, Bruno

2 1

Re: setupapi: SetupDiGetClassDevsExW should ignore DIGCF_ALLCLASSES if class set
by Nikolay Sivov May 6, 2014

May 6, 2014

Is it possible to add a test for that?

2 1

Wine talk at Linuxwochen Österreich
by Stefan Dösinger May 6, 2014

May 6, 2014

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, I will give a talk about Wine gaming at Linuxwochen Österreich later this week: When: Friday, May 9th, 15:00 Where: FH Technikum Wien, HS 2.01, Vienna See http://www.linuxwochen.at/ and https://cfp.linuxwochen.at/en/LWW14/public/events/154 for more details Cheers, Stefan -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQIcBAEBAgAGBQJTaQLIAAoJEN0/YqbEcdMwscQP/2OuAs1Z2MSrYNlXH4CJngDQ 0rqb4aCf32Yy5iVJ6XRbIkfPxrA+CpeX1fwCPpx1rGjJM9djQ1RnbQDb1dafCbYg 4qHJeMooZ9zbqB/6rUV9q2xk8E3Bnhi5RptBnwtEljUWCOTbCbTtevr5BjLbM/Dk Si4S2gFSkGCHyNFupxaIFGOseew3zLNSkFvqV4Pr/eX+j0GIT78t6CyqtFB/Vpvs nMBptteZSlwmvJ5Z9bX7ruUus9+Az2Q17UV4c0tN+x5xsLD6QpoJWocFyO/7SYy9 HTa6QWwDF+AWOiCkP3TdPz1U/bNcEzSEGpDcFaQNLABGFhVm3rpgsqoBwJln2BDH Ig0+pG7jQSaYm1qK6Y5Sx58rVTDXizWmr2QfGkqh3ZmaOBBXTqudtPbZ5lrCmk50 xtbwqymo/vLXaU5FbgY83VvDip3oCBwny+E8R+O/rL20h5bkLNnn5doU3Kt2GTKU c3eohbg/qAlBzxG0UrtSQdiNYeItABaFYvhAkg11g8y0yjiZYdiGHMz4cgUMaVoe qDR0DLLEFGqTqYvhs5+fiK09EjABSpTVSnuGwsX0zRphgLZqsBWvg/O7G3pCwZ4Y MamqhvKBXhS6sQVjyDujuqNvSon3IEcomPz/BazbzZjuqfMU8IdP2lk7CizfvBvM dX/IsNZR6MKxhDVHtfEB =F/a5 -----END PGP SIGNATURE-----

1 0

Re: setupapi: SetupDiGetClassDevsExW should ignore DIGCF_ALLCLASSES if class set
by Henri Verbeet May 6, 2014

May 6, 2014

On 6 May 2014 16:05, Mark Harmstone <hellas(a)burntcomma.com> wrote: > + if (flags & DIGCF_ALLCLASSES && class) > + flags -= DIGCF_ALLCLASSES; This works, of course, but the usual way to do that kind of thing would be "if (class) flags &= ~DIGCF_ALLCLASSES;"

1 0

Re: [PATCH] gdi32: Report unsubstituted font face names.
by Alexandre Julliard May 6, 2014

May 6, 2014

Alex Henrie <alexhenrie24(a)gmail.com> writes: > @@ -5805,12 +5810,13 @@ static BOOL freetype_EnumFonts( PHYSDEV dev, LPLOGFONTW plf, FONTENUMPROCW proc, > GDI_CheckNotLock(); > EnterCriticalSection( &freetype_cs ); > if(plf->lfFaceName[0]) { > - FontSubst *psub; > - psub = get_font_subst(&font_subst_list, plf->lfFaceName, plf->lfCharSet); > + FontSubst *psub = get_font_subst(&font_subst_list, plf->lfFaceName, plf->lfCharSet); > + WCHAR *unsubstitutedFaceName = NULL; > > if(psub) { > TRACE("substituting %s -> %s\n", debugstr_w(plf->lfFaceName), > debugstr_w(psub->to.name)); > + unsubstitutedFaceName = plf->lfFaceName; > lf = *plf; > strcpyW(lf.lfFaceName, psub->to.name); > plf = &lf; This could be seriously simplified then. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH 3/5] server: Add socket-side support for the interface change notification object (resend).
by Alexandre Julliard May 6, 2014

May 6, 2014

"Erich E. Hoover" <erich.e.hoover(a)gmail.com> writes: > This patch adds an async queue, list object, and interface change > object to the socket object in order to support interface change > notification objects (fully implemented in part 4). A significant > deviation from previous versions of this patch is that errors are > passed up from each routine so that it is easier to ensure that all > possible errors are properly reported. I believe that this is the > server style comment AJ had on IRC, but some clarification would be > appreciated. No, that's not how error handling works in the server. The functions that fail set last error and then you propagate that. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH 2/5] server: Add delayed processing for socket-specific ioctl() (resend).
by Alexandre Julliard May 6, 2014

May 6, 2014

"Erich E. Hoover" <erich.e.hoover(a)gmail.com> writes: > @@ -523,17 +523,32 @@ obj_handle_t sock_ioctl( struct fd *fd, ioctl_code_t code, const async_data_t *a > int blocking, const void *data, data_size_t size ) > { > struct sock *sock = get_fd_user( fd ); > + obj_handle_t wait_handle = 0; > + async_data_t new_data; > + int error; > > assert( sock->obj.ops == &sock_ops ); > > + if (blocking) > + { > + if (!(wait_handle = alloc_wait_event( current->process ))) return 0; > + new_data = *async_data; > + new_data.event = wait_handle; > + async_data = &new_data; > + } > switch(code) > { > case WS_SIO_ADDRESS_LIST_CHANGE: > /* intentional fallthrough, not yet supported */ > default: > - set_error( STATUS_NOT_SUPPORTED ); > - return 0; > + error = STATUS_NOT_SUPPORTED; > + break; > } > + set_error( error ); > + if (error == STATUS_PENDING) > + return wait_handle; > + close_handle( current->process, wait_handle ); > + return 0; You should be doing that only as needed. It doesn't make sense to allocate a handle for unknown ioctls. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH 2/3] winex11.drv: copy and paste files to/from X11 apps (try 2)
by Alexandre Julliard May 6, 2014

May 6, 2014

Damjan Jovanovic <damjan.jov(a)gmail.com> writes: > /************************************************************************** > + * X11DRV_CLIPBOARD_ExportHDROP > + * > + * Export CF_HDROP format to text/uri-list. > + */ > +static HANDLE X11DRV_CLIPBOARD_ExportHDROP(Display *display, Window requestor, Atom aTarget, > + Atom rprop, LPWINE_CLIPDATA lpdata, LPDWORD lpBytes) > +{ > + HDROP hDrop; > + UINT i; > + UINT numFiles; > + UINT totalSize = 0; > + HGLOBAL hClipData = NULL; > + char *text_uri_list = NULL; > + char *next; > + BOOLEAN succeeded = FALSE; > + > + *lpBytes = 0; > + > + if (!X11DRV_CLIPBOARD_RenderFormat(display, lpdata)) > + { > + ERR("Failed to export %04x format\n", lpdata->wFormatID); > + return 0; > + } > + > + hDrop = (HDROP) lpdata->hData; > + numFiles = DragQueryFileW(hDrop, 0xFFFFFFFF, NULL, 0); > + > + for (i = 0; i < numFiles; i++) > + { > + UINT dosFilenameSize; > + WCHAR *dosFilename; > + succeeded = FALSE; > + dosFilenameSize = 1 + DragQueryFileW(hDrop, i, NULL, 0); > + dosFilename = HeapAlloc(GetProcessHeap(), 0, dosFilenameSize*sizeof(WCHAR)); > + if (dosFilename) > + { > + char *unixFilename; > + DragQueryFileW(hDrop, i, dosFilename, dosFilenameSize); > + unixFilename = wine_get_unix_file_name(dosFilename); > + if (unixFilename) > + { > + totalSize += 8 + /* file:/// */ > + 3*(lstrlenA(unixFilename) - 1); /* "%xy" per char except first '/' */ > + succeeded = TRUE; > + HeapFree(GetProcessHeap(), 0, unixFilename); > + } > + HeapFree(GetProcessHeap(), 0, dosFilename); > + } > + if (!succeeded) > + { > + ERR("out of memory\n"); > + goto end; > + } > + } > + totalSize += 2 * numFiles; /* "\r\n" */ > + hClipData = GlobalAlloc(GMEM_MOVEABLE | GMEM_DDESHARE, totalSize); > + if (hClipData == NULL) > + { > + ERR("out of memory\n"); > + succeeded = FALSE; > + goto end; > + } > + text_uri_list = (char*) GlobalLock(hClipData); > + next = text_uri_list; > + for (i = 0; i < numFiles; i++) > + { > + UINT dosFilenameSize; > + WCHAR *dosFilename; > + succeeded = FALSE; > + dosFilenameSize = 1 + DragQueryFileW(hDrop, i, NULL, 0); > + dosFilename = HeapAlloc(GetProcessHeap(), 0, dosFilenameSize*sizeof(WCHAR)); > + if (dosFilename) > + { > + char *unixFilename; > + DragQueryFileW(hDrop, i, dosFilename, dosFilenameSize); > + unixFilename = wine_get_unix_file_name(dosFilename); > + if (unixFilename) > + { > + static const char *hex_table = "0123456789abcdef"; > + UINT u; > + *next++ = 'f'; > + *next++ = 'i'; > + *next++ = 'l'; > + *next++ = 'e'; > + *next++ = ':'; > + *next++ = '/'; > + *next++ = '/'; > + *next++ = '/'; > + /* URL encode everything - unnecessary, but easier/lighter than linking in shlwapi, and can't hurt */ > + for (u = 1; unixFilename[u]; u++) { > + *next++ = '%'; > + *next++ = hex_table[unixFilename[u] >> 4]; > + *next++ = hex_table[unixFilename[u] & 0xf]; > + } > + *next++ = '\r'; > + *next++ = '\n'; > + succeeded = TRUE; > + HeapFree(GetProcessHeap(), 0, unixFilename); > + } > + HeapFree(GetProcessHeap(), 0, dosFilename); > + } > + if (!succeeded) > + { > + ERR("out of memory\n"); > + goto end; > + } > + } > + > +end: > + if (succeeded) > + { > + GlobalUnlock(hClipData); > + *lpBytes = GlobalSize(hClipData); > + return hClipData; > + } > + else > + { > + if (text_uri_list) > + GlobalUnlock(hClipData); > + GlobalFree(hClipData); > + *lpBytes = 0; > + return 0; > + } > +} That's unnecessarily complicated and inefficient. You don't need to retrieve the file name 4 times, or to convert every filename to Unix twice. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: wined3d: Add a registry setting, DisabledExtensions, to disable the use of OpenGL and WGL extensions.
by Henri Verbeet May 6, 2014

May 6, 2014

On 6 May 2014 10:44, Ken Thomases <ken(a)codeweavers.com> wrote: > I'm inclined to say we don't really need this, but I'll leave that up to Alexandre. > + /* Second pass at disabling extensions. This (re)disables extensions that may > + be implied by the GL version or other extensions. */ Is this really a problem in practice? > + if (wined3d_settings.disabled_extensions) > + { > + parse_extension_string(gl_info, wined3d_settings.disabled_extensions, gl_extension_map, > + sizeof(gl_extension_map) / sizeof(*gl_extension_map), FALSE); > + if (WGL_Extensions) > + parse_extension_string(gl_info, wined3d_settings.disabled_extensions, wgl_extension_map, > + sizeof(wgl_extension_map) / sizeof(*wgl_extension_map), FALSE); > + } It may be nicer to just make parse_extension_string() handle a NULL "extensions" pointer, and skip all the conditionals here. > - gl_vendor = wined3d_guess_gl_vendor(gl_info, gl_vendor_str, gl_renderer_str); > - card_vendor = wined3d_guess_card_vendor(gl_vendor_str, gl_renderer_str); > - TRACE("Found GL_VENDOR (%s)->(0x%04x/0x%04x).\n", debugstr_a(gl_vendor_str), gl_vendor, card_vendor); > - > device = wined3d_guess_card(gl_info, gl_renderer_str, &gl_vendor, &card_vendor); > TRACE("Found (fake) card: 0x%x (vendor id), 0x%x (device id).\n", card_vendor, device); I guess this was broken up because of the extension checks in wined3d_guess_gl_vendor(). Is that really justified? Conceptually, if we really want the DisabledExtensions functionality in wined3d, I think it should behave like filtering the extension strings, as opposed to being some kind of generic way to disable specific wined3d code paths, but not others that depend on the same extensions.

1 0

Re: winex11: Silence a noisy winediag ERR in xrandr12_init_modes
by Henri Verbeet May 6, 2014

May 6, 2014

On 5 May 2014 22:53, André Hentschel <nerv(a)dawncrow.de> wrote: > - ERR_(winediag)("Broken NVIDIA RandR detected, falling back to RandR 1.0. " > - "Please consider using the Nouveau driver instead.\n"); > + static int once; > + if (!once++) ERR_(winediag)("Broken NVIDIA RandR detected, falling back to RandR 1.0. " > + "Please consider using the Nouveau driver instead.\n"); Does this actually help? We really shouldn't be calling X11DRV_XRandR_Init() much more often than once per process.

1 0

Re: [PATCH 5/5] ws2_32: Add an interactive test for interface change notifications (resend).
by Marvin May 6, 2014

May 6, 2014

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=6836 Your paranoid android. === wvistau64 (32 bit sock) === sock: unhandled exception c0000005 at 00000000

1 0

Working on ATL, more real world test case needed
by Qian Hong May 5, 2014

May 5, 2014

Hi all, I'm working on atl80/atl100 recently, particularly on AtlAxCreateControl{,Lic}{,Ex}, AtlAxAttachControl, AtlIPersistStreamInit_{Load,Save}, AtlIPersistPropertyBag_{Load,Save} The motivation to work on AtlAxCreateControl* and AtlAxAttachControl* is for AliWangwang [1]; the motivation to work on AtlIPersistPropertyBag_Load is for QQ[2]. I've already have solution for those apps, however, I have some trouble on them: 1. I have more and more test case for ATL, and found more and more difference behaviors between native AtlAxCreateControl/AtlAxAttachControl and buitin's, some of native behavior looks unreasonable to me so I doubt if we want to follow that. It would be nice to test more real world application before deciding what to do next. 2. I found the (seems) undocumented ATL_PROPMAP_ENTRY struct is different between atl80 and atl100 from my black box testing result. I know QQ use atl80.AtlIPersistPropertyBag_Load which receive a ATL_PROPMAP_ENTRY input parameter, but I don't have any real world app to test the atl100 version or any other version. I'm interesting to know more apps depends on any version (atl/atl80/atl90/atl100/atl110) of AtlIPersistStreamInit_*/AtlIPersistPropertyBag_* to see how they work. I really appreciate a lot if anyone can help report ATL related bugs and CC me, maybe someone remembers some Apps have bugs on atl* but haven't report to bugzilla yet - I don't know any good way to find real world test case, that's why I'm here :) I'd really be happy to get some surprise smart way to do it! In short, any ATL related bugs are welcome, particularly on AtlAxCreateControl*/AtlAxAttachControl/AtlIPersistPropertyBag*/AtlIPersistStreamInit* if you know that's the culprit, but never mind for other ATL related bugs. Note: if an app bugs can be workaround by winetricks -q vcrun2005 / winetricks -q vcrun2008 / winetricks -q 2010, even with msvc*.dll override set to builtin, only leave native atl*.dll, then it is possible a ATL bug. Here are a few known ATL bugs: http://bugs.winehq.org/buglist.cgi?cmdtype=runnamed&list_id=147507&namedcmd… Thank you a lot for any help! [1] http://bugs.winehq.org/show_bug.cgi?id=33724 [2] http://bugs.winehq.org/show_bug.cgi?id=33822 -- Regards, Qian Hong - http://www.winehq.org

2 4

Re: [PATCH 2/5] wined3d: Validate and handle query size parameter. (try 2)
by Henri Verbeet May 5, 2014

May 5, 2014

On 2 May 2014 16:30, Matteo Bruni <mbruni(a)codeweavers.com> wrote: > +static void fill_query_data(void *out, unsigned int out_size, const void *result, unsigned int result_size) > +{ > + unsigned int size = min(out_size, result_size); > + > + memcpy(out, result, size); > +} The extra "size" variable is perhaps a bit superfluous. Not sure if I care.

1 0

Re: [PATCH 1/5] d3d9/tests: Test GetData behavior with various size values. (try 2)
by Henri Verbeet May 5, 2014

May 5, 2014

On 2 May 2014 16:30, Matteo Bruni <mbruni(a)codeweavers.com> wrote: > + memset(&data, 0xff, sizeof(data)); This is a bit of an odd way to write "data = ~0u;". > + ok(*(WORD *)&data == 1 || broken(*(WORD *)&data != 1), > + "Occlusion query returned an unexpected result (0x%.8x).\n", data); > + todo_wine ok(*((WORD *)&data + 1) == 0xffff, > + "data was modified outside of the expected size (0x%.8x).\n", data); I think this is a bit awkward, perhaps it makes more sense the declare "data" as something like "WORD data[4];". Also, can this really ever return 0 for the query data? Would it help to draw a different primitive type, instead of a tiny point? (Because, well, point rasterization is complicated.)

1 0

Re: [PATCH] setupapi: Implement a binary compatible string table
by Marvin May 4, 2014

May 4, 2014

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=6812 Your paranoid android. === wvistau64 (64 bit stringtable) === stringtable.c:351: Test failed: got 128 stringtable.c:352: Test failed: got 0, thread lcid 1401 === w2008s64 (64 bit stringtable) === stringtable.c:351: Test failed: got 124 stringtable.c:352: Test failed: got 0, thread lcid 409

1 0

Use WINE API from ELF application.
by John Found May 4, 2014

May 4, 2014

Is it possible to use the WINE functionality from inside regular ELF application. For example starting and debugging Windows application. Creating windows, etc. -- asm32.info <johnfound(a)asm32.info>

2 1

Re: Failing tests on testbots because of missing focus
by Francois Gouget May 4, 2014

May 4, 2014

On Sat, 3 May 2014, Bernhard Übelacker wrote: [...] > In my opinion the reason is that these tests get executed from an > application which does not have the current input focus. On the WineTestBot the tests are being run from a console window which is iconized. On the fg-* VMs the mechanism is similar, though it's possible the command window is no iconized. [...] > Also because the newtb* machines showed this error only sometimes (at > least before the testbot changes lately) probably some other tests are > doing something to change focus situation, or are running long enough > and windows is changing its focus policy there. Some tests cause Windows to be opened and stay around. In particular I remember that WineTest.exe typically leaves an Explorer window behind, most likely triggered by a shell32 or shlwapi test. The rpcrt4 tests and others also leave firewall windows behind, asking whether to unblock various ports (these tests don't actually need the ports to be unblocked and thus are not impacted by the questions being unanswered [1]). So based on your description it's quite likely that one of these windows is the one with the focus. [...] > Because of this affecting more tests (and probably this fix could also affect > tests running after this one), I am not sure if this is the right way to fix it: > > - should the test rely on being started from an application which > had focus? (e.g. change testbot to start the tests by an cmd window > which has guaranteed focus) Tests that need focus for one of their window should set up the focus themselves [2]. More generally tests should not rely on anyone else setting things up for them as that would make them too brittle and hard for developers to run 'correctly'. [1] I assume these tests bind their socket to any address and that's why they get these firewall questions. But in fact they only get connections from localhost which is unaffected by the firewall so it's possible they could avoid these questions by only binding to localhost. It's not clear that it's worth the trouble though. [2] I'll leave checking the proposed code to do so to other, more capable, hands (or eyes, braaaains, as the case may be). -- Francois Gouget <fgouget(a)codeweavers.com>

1 0

Re: [PATCH] vbscript: Fixed CBool(try 3)
by Piotr Caban May 3, 2014

May 3, 2014

On 05/02/14 06:22, Shuai Meng wrote: > testbot:https://newtestbot.winehq.org/JobDetails.pl?Key=6731 > > Changelog: > added more tests on object; > changed the implemention of some tests that caused "Err.number = 0" > rearranged the order of tests > --- > dlls/vbscript/global.c | 41 ++++++++++------------------------- > dlls/vbscript/tests/api.vbs | 53 > +++++++++++++++++++++++++++++++++++++++++++++ > 2 files changed, 65 insertions(+), 29 deletions(-) > The tests are failing on wine on polish locale: LC_ALL=pl_PL make test ../../../tools/runtest -q -P wine -T ../../.. -M vbscript.dll -p vbscript_test.exe.so run && touch run.ok fixme:thread:GetThreadUILanguage : stub, returning default language. run.c:2102: Tests skipped: Skipping some tests in non-English locale fixme:vbscript:parse_script parser failed around L"<!--" fixme:vbscript:do_icall L"y" not found fixme:vbscript:DispatchEx_InvokeEx no letter/setter run.c:1917: Test failed: expected global_success_d run.c:1918: Test failed: expected global_success_i run.c:1920: Test failed: parse_script failed: 800a000d fixme:vbscript:ClassFactory_QueryInterface (0x7e2462ec)->({342d1ea0-ae25-11d1-89c5-006008c3fbfc} 0x33f7b0) fixme:vbscript:RegExp2_QueryInterface (0x130a88)->({fc4801a3-2ba9-11cf-a229-00aa003d7352} 0x33f74c) fixme:vbscript:interp_enumnext uninitialized fixme:vbscript:interp_newenum Unsupported for {VT_EMPTY} fixme:vbscript:interp_enumnext uninitialized fixme:vbscript:show_msgbox failed: 80004005 fixme:vbscript:show_msgbox blocked make: *** [run.ok] B³±d 3

2 6