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

Re: [PATCH] user32: Support activation context class redirection in CreateWindow.

by Nikolay Sivov

On 03/29/2015 01:54 AM, Mark Harmstone wrote: > This patch implements the class redirection feature of activation contexts, as > used by version 6 of comctl32. Among other things, this is a necessary step in > fixing the behaviour seen in bug 38124. > > --- > dlls/user32/win.c | 62 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 62 insertions(+) > In principle it looks almost okay, except that I don't think we need it right now, as long as comctl32 doesn't register versioned classes at all. It will just wait time for every window creation requested looking for something that can't be created. Comments on code itself to follow. > + classatom = get_int_atom_value(className); > + if (classatom) > + GlobalGetAtomNameW(classatom, classNameStr, sizeof(classNameStr) / sizeof(*classNameStr)); > + else > + strcpyW(classNameStr, className); You don't need to copy in case of a string. > + if (wcrd->name_len == 0) > + return FALSE; I don't think it can be 0, ever. Name comes from class name from module manifest and assembly version string. > + memcpy(redirClassName, (unsigned char*)askd.lpData + wcrd->name_offset, wcrd->name_len); > + redirClassName[wcrd->name_len / sizeof(WCHAR)] = 0; All strings in context data are null-terminated, so you don't need this complexity. In fact you don't need to copy anything at all, you can just use a (WCHAR*)((BYTE*)askd.lpData + wcrd->name_offset) as a ready-to-use pointer, I think it's a preferred way, given that overhead of this redirection lookup should be minimal. So your helper only needs to return a pointer - original one or computed from redirection data. And atoms should be resolved outside of it. > + WCHAR redirClassName[256]; Is it defined somewhere to be limited to that? > + > + wndPtr = create_window_handle( parent, owner, redirClassName, module, unicode ); > + > + if (GetLastError() == ERROR_INVALID_HANDLE && GetClassInfoW( 0, redirClassName, &wc )) > + wndPtr = create_window_handle( parent, owner, redirClassName, module, unicode ); > + I don't quite get what you're trying to do here? How does same call help if called second time? Also last error check is strange.

March 29, 2015

Re: [PATCH] wined3d: Fix some videocards property detection

by Charles Davis

> On Mar 28, 2015, at 4:21 AM, Stefan Dösinger <stefandoesinger(a)gmail.com> wrote: > then check for software fallbacks. There's a CGL function for > checking that, but I don't remember from the top of my head how it works. You mean: CGLGetParameter(ctx, kCGLCPGPUVertexProcessing, &gpu_vertex); CGLGetParameter(ctx, kCGLCPGPUFragmentProcessing, &gpu_frag); When each one is false, it means that the corresponding stage fell back to software. We may need some kind of Wine-specific WGL extension for this, though. wined3d doesn’t know about CGL contexts. By the way, those are “get” only. You can’t force the driver to do it in hardware by forcing them to true (in case you get that idea). Chip

March 28, 2015

Re: [BISECTED REGRESSION] file dialog does not shown

by Frédéric Delanoy

On Fri, Mar 27, 2015 at 3:55 PM, Andrey Skvortsov <andrej.skvortzov(a)gmail.com> wrote: > Hi, > > original problem is described here: > https://bugs.winehq.org/show_bug.cgi?id=34825 > I have the same problem with running Nanocad 5.1. The file open/import > dialog does not open with wine-1.7.39. I bisected the problem to > commit: > [bbf2cce16041e90d621a1f5889ea1bd760e21a0e] comdlg32: Do not modify > dialog resource directly. > > I reverted this commit on latest wine and that fixed my issue with file open dialog. > > -- > Best regards, > Andrey Skvortsov > > Secure eMail with gnupg: See http://www.gnupg.org/ > PGP Key ID: 0x57A3AEAD I'd suggest you update the bug report to mention the regression commit -- Frédéric Delanoy

March 28, 2015

Re: [PATCH] wined3d: Fix some videocards property detection

by Stefan Dösinger

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am 2015-03-28 um 13:00 schrieb Ken Thomases: > The Mac driver is already setting those, by default. Software > rendering is disabled unless you set a registry setting. In that situation I'd expect Sergey to get a black screen in the affected 3dmark2001 test rather than slow rendering. Sergey, do you have HKCU\Software\Wine\Mac Driver\AllowSoftwareRendering set? -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBAgAGBQJVFpfOAAoJEN0/YqbEcdMwEu8QAIbbxnWHSeuQIPMidiYSDD13 BtbiDGMslKfSRLoMKOsCbeXBoXMLHh4tzYMj3ns2DsmQBiP8LziT6wwdT0WOPpEX g1KUHNhYMG0hdUeX5lgSuTA1pjcXVbgZwL5MgJosalZmdx8LoVc9kfVeqZ2qijBX EMWF1Ld6PalNxCEqnqWsqrzOhGluKEFoObB8/W8ETaeypvXIr+BYwKKMBWsFd7Dn 7rxeZTr07uu85sb+eyfM9twEy7OCqKiAKlHwmoaxVILftDNyShn/27BisvU1ThSa jO6lJUIgUeWNYZCdX4BSHlNLRbM5MkjpmglALLdk1RarG/XpihUvtJggnqlqj+f8 VQeRhr8ZZ6kzALTF7YlOsIAmWt0CgTKsGQ7uyc+7MJsc+s+AR1Lm4iqYZX55hVRN nHwriFqvZSxy41w3kTaXFZPiVgs6u5StpMCp+QduM7wLH8aCyFgXqfb+h3xl+yTe tAz2d1dVG0Nrj9jzXQou1wSrB1Jur1O/7u+MuDcD+71oZrZj7hPZSxu5X46ju5SY oaOrgj5wdSUvZ23yEXzNvA0RCtfEBg7KjJF+5gmJtIkP4F5LgM9v4D+B+/xtEior iYOJ63VbTcc/Yd8U7oplAL+0Kg8QRGSmxVafvK8vMM0jAXsFUgk0yNQAyQExhglZ k/+BCb8R2taOaTqgJre7 =/wRR -----END PGP SIGNATURE-----

March 28, 2015

Re: [PATCH] wined3d: Fix some videocards property detection

by Ken Thomases

On Mar 28, 2015, at 6:51 AM, Stefan Dösinger <stefandoesinger(a)gmail.com> wrote: > There are OpenGL flags winemac.drv can set to disable the software > fallback: > https://developer.apple.com/library/mac/qa/qa1502/_index.html . What > happens when you set them? Do they influence the reported capabilities? The Mac driver is already setting those, by default. Software rendering is disabled unless you set a registry setting. That doesn't mean that fragment processing and vertex processing are done on the GPU, though. Disabling the software renderer is not the same as insisting that nothing significant gets done in software. So, evidently, it doesn't affect how many uniforms are reported. -Ken

March 28, 2015

Re: [PATCH] wined3d: Fix some videocards property detection

by Stefan Dösinger

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, Please keep wine-devel in CC. There are more developers than just myself :-) . Am 2015-03-28 um 12:38 schrieb Sergey Isakov: > I know this crash. It was appeared because this patch was > introduced in the wrong place as I describe in the bug 38277 Too > slow test Advanced Pixel Shaders in 3DMark2001SE > <https://bugs.winehq.org/show_bug.cgi?id=38277> My patch works > fine. It may work fine on the Macs you tested or even all the macs that ever shipped, but it is still ugly to hardcode GPU capabilities that the driver should provide. There are OpenGL flags winemac.drv can set to disable the software fallback: https://developer.apple.com/library/mac/qa/qa1502/_index.html . What happens when you set them? Do they influence the reported capabilities? -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBAgAGBQJVFpW6AAoJEN0/YqbEcdMwaz4P/298Ha4rtgWPYEFF+w6HzAWP f7G9fenhHkgjaudNwjcFXTr1dweYTVwvVll5Ahpu9bRd4iC/hM4jyKf2TVcpcXbM +yIri1UnsgSt63YJ1+XRfvVsV28BYlhqnlZvqKkmNHf/5dxg0iOfz0noDYAcspFZ BjfVQOFgaRmuiS9AyVpgc5WO2+IptKvDa1ExSOSNhOO5ye7T9Nh071QoWCgakqza invrADPRjRm/Ri7JV6CtQeRZCLYdMjeONchhJOHif2KYjk8/J5PNVAUeSSt73ZXQ TPA/VmPAs9mge3QlKgD9NgIab8auPt0tHW3MLLluKbLmwOJSMsOuxx85VC4yxlvK K/h0Q7Ldx8BvmsraLjmNMPmB9FMUNuxmEhjrkdbN90v9FwfsfZs5PGm6V+Rmt4uq gBHQxY2wd6xUt6I4AFH+6s088Ubl7zQRdAsE12LTMnRHF0ctIvHZCVdtdgVG/0hu ZpxrnHgoWm4E9arzslkHOxaqAh1Tln8k9WtpF6PcGNv4xbLWKLNiRBHMF4fK9MQI PtbT7gzL71+38lArzqnCzKb4Ijw1X7PkgA//4ZlkENyCp9SKxWpLQaDq853mhGlQ Q8z6TqAgb1wIwqO/rfiX6Rqj+aWt9Zet4XDhLaLbsifRnGjr4gad9LSC28GzQfet d80ZXwigJshmdtl05+I6 =+1lX -----END PGP SIGNATURE-----

March 28, 2015

Re: [PATCH] krnl386.exe16: Fix NULL pointer dereferencing

by Stefan Dösinger

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am 2015-03-27 um 19:16 schrieb Sergey Isakov: > This if statement body will perform only if driver_request is NULL > pointer. Next statements will cause crash. Bug 38159 NULL pointer > dereferencing <https://bugs.winehq.org/show_bug.cgi?id=38159> I think in this case the segfault is intentional. The caller passes a stupid parameter. Also I'm not sure if it actually segfaults. In vm86 mode memory is mapped to this page, so NULL may be a valid address. The best solution would be to write a test and see what happens on an ancient DOS, but we don't have a test infrastructure for DOS and Win16. The next best idea might be to just remove the entire block. It's funny though since (according to git) the patch that added this check already added the "/* FIXME - to be deleted ?? */" comment. At some point we may as well remove the entire DOS code since we just call dosbox these days and 64 bit CPUs don't allow vm86 mode anyway. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBAgAGBQJVFpBqAAoJEN0/YqbEcdMwVUQP/itrU95T4mlo2vo7f4J3k7fd Wx4goKV1/s5iPT+56T3q+MsqYp0gcNgmiwmVrf+HSNl94jXo1ysWfsuD1LeAAPGi O6FJBpszJdjw+fjm0whKidM3L+QYOEqGb/8AP5ZuXQ2zE0+iDA2cGo5oFrP2LpmM ZR7t3uYTIqk4VYWuZz4rlfSaHu552LpT/zEIYghteL/ANTqkad+oVleOFLB6Nbb/ j1PK7QKl6D+n6Kd2KSwlI9jj3XADf14i96J2knNh7yyLVZAsjNTe/3F9eoI98N5M cI22fGkW6jp70HoS1yMpJG66jqrhamAi6QBzMUI2cfAhHtjngKj36X3MRZMXIiK7 4E1g/Z7hcJ3ZtbrfzHnUdUXCQMXHszA93U69QRERTnjatDB+VNTZ67AaPGMXBiE9 2i5KlGg5TXjZPJwPqetHuGZ18uUXwCXfbC5+VPXP5AqbhxltE6IPfbKs95Bo90gh EvZsooU7pqNJxNAfscDIKliDsIcFV3TATqEyr/LCeczKNbE1DOKQ/afBrGyivabe f/YzcLLHfx5x8Lxx5bsmn2+3xR+LVfppH/rCSC5YA2ZwU6gB8NeuDAHh/BvsV7E1 i9IkN43sbQ4xZQFckBsyhdcpJd0kCCp3n1kmCfCWx3c2uIwLNvzzwT9HM8mY36my qcBfCcz1CCOpKpaayCSL =78sc -----END PGP SIGNATURE-----

March 28, 2015

Re: [PATCH] user32: Use function parameter if it set to be not stub

by Stefan Dösinger

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am 2015-03-27 um 21:52 schrieb Sergey Isakov: > { + if (!lpDisplayDevice) { + return FALSE; + } This is probably correct, but it needs a test. Currently this function crashes if this parameter is NULL. I can imagine that EnumDisplayDevices(NULL, 1, NULL, 0) returns TRUE if display 1 exists and FALSE otherwise. > + + if (i) { FIXME("(%s,%d,%p,0x%08x), > stub!\n",debugstr_w(lpDevice),i,lpDisplayDevice,dwFlags); - - if > (i) return FALSE; + } This is an unrelated change and matters more for what the patch description says it does. Though if I understand this right we want this check to be "if (i || lpDevice)" because the application can also pass in a device name. > > + if (lpDevice) { + strcpyW(lpDisplayDevice->DeviceName, > lpDevice); + } else { memcpy(lpDisplayDevice->DeviceName, > primary_device_name, sizeof(primary_device_name)); + } This is yet another unrelated change. It looks OK, but I don't think it really matters. Copying the input display name will at best paper over the fact that our implementation of this function does not support multiple display devices. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBAgAGBQJVFo0tAAoJEN0/YqbEcdMw+eQP/1SnlfkRjPZ/8PW8+m3iOYHC OTLWOE7Io94FzLfFQO487pCbsqgd9199UWp3IolfUe/y2vS4LtosxQQh1tLG7j7A rsypwNjV8OwSHvXiL5BH/4QvZTAqoctW484E7z7IHzMXMqa2Oyrj5K3QXHVes6O5 9xEB1UuSkB19Z0jIQqHPR4Hc/o8gYuCAER05BiTqOZcBSsrVhfOOZ+FKR7krWqdP o3VkOmUSWhX+9ziTsKmWGsBEZk0r81RMG2cbT/eShAtBBZs/+YtUfYiNY2q+tCoF fMiEfiMU7e1TqroBJAw37gYju+qt3xon/RuyOQD13a3hwWSDzMeoE9PAf9TxCK4Q OinEV45U8RiDkczviuOQluojfmJSXmVCC2BjPPQ2ZtmGXyuRXLrlRZ4KsCE/xhyV p2+Za2U/cvfqy0ug5isnnFGu/lKi1RCeGXsbHX6ZF5amTEfuklQpWolacrGilxZ2 kYiBtV93A/IJzUPaj2d97JbEFG2nMP6pT+KTSFBJIY2bzzzhnX1Fltd8kt3Vp/rv MOUPs/peC9sv17wDc5LREPAbA7LOBuN4M3M+5O2glO/F96gt8FSsPYyZy9sMRXEv eEuQgGbFbm9KKCWBK53mm78aAFiulqEd3S4vFXnvNU5sFxUW/BO3TxDZJhPEAK2v BeJJ0ZjH01mfQ8lhDQph =RL+u -----END PGP SIGNATURE-----

March 28, 2015

Re: [PATCH] wined3d: Fix some videocards property detection

by Stefan Dösinger

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, This patch has a few unrelated changes. The core change is that it re-introduces a hack we once had in Wine and removed because it broke newer mac GPUs. The CrossOver version of the hack works a bit better than the unconditional Wine version, but it's still ugly. Am 2015-03-28 um 06:15 schrieb Sergey Isakov: > + /* Non-Apple GL vendors report uniforms correctly */ > + if (!match_apple(gl_info, gl_renderer, gl_vendor, card_vendor, device)) return FALSE; > + /* All dx9 cards are overreported on OSX */ > + if (!match_dx10_capable(gl_info, gl_renderer, gl_vendor, card_vendor, device)) return TRUE; > + /* Nvidia and Intel DX10 cards support > 256 uniforms */ > + if (card_vendor != HW_VENDOR_AMD) return FALSE; It is terrible that OSX lies about the GPU capability and then forces applications to have a crystal ball that tells them what the GPU can actually do. - From a pragmatic point of view I don't think OSX will ever change in this regard. It was a design decision by Apple to do this, and they stopped updating drivers for the old hardware a long time ago. I am open to re-introducing the hack if we can find a way to detect the bug more directly. E.g. try to declare a shader with 257 uniforms, read them with indirect addressing so the driver can't optimize the access and then check for software fallbacks. There's a CGL function for checking that, but I don't remember from the top of my head how it works. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBAgAGBQJVFoC/AAoJEN0/YqbEcdMwcUIP/Rw24kSROYMK2Xvfe9vyEZ9S NQd9RPHRQRNiXuXRfAmAUKm/PF17O7SUMKLAwjg90HGZySdDOH4MCb7rwaXryKYU qV1ba75KR/JOi7db0r/WrZjf+mQ2+z1hES0SEg9jSYHxIcSNRhWvjB3H3rId7Och TAEvsiT897rE7U/2GQmnw9popFPOPRm0Y2YY4f1XecF1XSdmBokpdmpcDzoRTxbX 8q11cWlPyqi3u6xiyX7SwEwoh7v/zpNd4c1yuL6kytSfV1w+sD4fMC4BitIvfd2X OltqvQz5MfexglNlLJaeE1yplhrEKE2NHHdoIXhOhg7TdOeHYJfvrzgpjLWlUVnC 6BWsSkgge3OO3Pk7Fhi7Lht0EPhds+iq+7NCLwDV051Qi89kBlyJOY33bMMjryez wHo+YEZEvk0IkkVQGkAjJfZzLFtxjMzIxcixijkJc3/aYwuF0ZyDVqFe3hj4gPTw Ei1IThyWC4hdVqIuduD4t+5VH3GwU6JiYE6a16uHVTfoTSboN9te1WzV+lnhRFMf P54xaJ+98sJrpq+670WU68jHzM/BmXXojyF+lxv8t0noXmHLJy7S4yskzDhowMMG myjbDbeb/fB2O/oSdeOEXkskl2+XkotROM9qwF7a/yyaGe1/f47Y16gmrPyeCFD5 0Z68wLM5//XliOz91+hs =1uP/ -----END PGP SIGNATURE-----

March 28, 2015

Re: [PATCH] vbscript: Fixed Hex

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 https://testbot.winehq.org/JobDetails.pl?Key=12482 Your paranoid android. === w2000pro (32 bit) === No test summary line found === wxppro (32 bit) === No test summary line found === w2003std (32 bit) === No test summary line found === wvistau64 (32 bit) === No test summary line found === w2008s64 (32 bit) === No test summary line found === w7u (32 bit) === No test summary line found === w8 (32 bit) === No test summary line found === wvistau64 (64 bit) === No test summary line found === w2008s64 (64 bit) === No test summary line found

March 28, 2015