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

Re: [PATCH 9/11] qcap: add the SmartTee filter automatically as necessary, and test this

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=13651 Your paranoid android. === w2008s64 (32 bit smartteefilter) === smartteefilter.c:1926: Test failed: couldn't create NullRenderer, hr=0x80040154 smartteefilter.c:1926: Test failed: couldn't create NullRenderer, hr=0x80040154 smartteefilter.c:1985: Test failed: couldn't create NullRenderer, hr=0x80040154 smartteefilter.c:1985: Test failed: couldn't create NullRenderer, hr=0x80040154 === w2008s64 (64 bit smartteefilter) === smartteefilter.c:1926: Test failed: couldn't create NullRenderer, hr=0x80040154 smartteefilter.c:1926: Test failed: couldn't create NullRenderer, hr=0x80040154 smartteefilter.c:1985: Test failed: couldn't create NullRenderer, hr=0x80040154 smartteefilter.c:1985: Test failed: couldn't create NullRenderer, hr=0x80040154

May 28, 2015

Re: [PATCH 5/5] wined3d: Always enable GL_PROGRAM_POINT_SIZE in the GLSL backend.

by Matteo Bruni

2015-05-28 23:43 GMT+02:00 Henri Verbeet <hverbeet(a)gmail.com>: > On 28 May 2015 at 23:23, Matteo Bruni <mbruni(a)codeweavers.com> wrote: >> @@ -6987,6 +7006,13 @@ static void shader_glsl_select(void *shader_priv, struct wined3d_context *contex >> GLuint program_id = 0, prev_id = 0; >> GLenum old_vertex_color_clamp, current_vertex_color_clamp; >> >> + if (!ctx_data->backend_init_done) >> + { >> + gl_info->gl_ops.gl.p_glEnable(GL_PROGRAM_POINT_SIZE); >> + checkGLcall("GL_PROGRAM_POINT_SIZE"); >> + ctx_data->backend_init_done = TRUE; >> + } >> + > Is this shared state between contexts? Otherwise you need to do this > once for every context. Also, shader_glsl_select() is kind of a hot > path, it may be better to just add an entry to struct > wined3d_shader_backend_ops for this kind of initialization. It's not shared but it should already be executed for each context (the flag is stored in struct wined3d_context). It would be surprising to me if that additional flag check has any performance impact since it is next to ctx_data->glsl_program which is accessed anyway just a bit later. It's probably cleaner to add a separate function though so I'm going to try that.

May 28, 2015

Re: [PATCH 5/5] wined3d: Always enable GL_PROGRAM_POINT_SIZE in the GLSL backend.

by Henri Verbeet

On 28 May 2015 at 23:23, Matteo Bruni <mbruni(a)codeweavers.com> wrote: > @@ -6987,6 +7006,13 @@ static void shader_glsl_select(void *shader_priv, struct wined3d_context *contex > GLuint program_id = 0, prev_id = 0; > GLenum old_vertex_color_clamp, current_vertex_color_clamp; > > + if (!ctx_data->backend_init_done) > + { > + gl_info->gl_ops.gl.p_glEnable(GL_PROGRAM_POINT_SIZE); > + checkGLcall("GL_PROGRAM_POINT_SIZE"); > + ctx_data->backend_init_done = TRUE; > + } > + Is this shared state between contexts? Otherwise you need to do this once for every context. Also, shader_glsl_select() is kind of a hot path, it may be better to just add an entry to struct wined3d_shader_backend_ops for this kind of initialization.

May 28, 2015

Re: [PATCH 8/11] qcap: get the SmartTee filter to a minimally functional level

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=13650 Your paranoid android. === w2000pro (32 bit smartteefilter) === The test timed out

May 28, 2015

Re: [PATCH 7/11] qcap: implement the SmartTee allocator functions

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=13649 Your paranoid android. === w2000pro (32 bit smartteefilter) === smartteefilter.c:1366: Test failed: timeout: threads did not receive sample in time === wxppro (32 bit smartteefilter) === smartteefilter.c:1366: Test failed: timeout: threads did not receive sample in time === w7u (32 bit smartteefilter) === smartteefilter.c:1366: Test failed: timeout: threads did not receive sample in time === w8 (32 bit smartteefilter) === smartteefilter.c:1366: Test failed: timeout: threads did not receive sample in time

May 28, 2015

Re: winex11: Fix ctrl-<symbol> to generate codes below 0x20 where necessary.

by Huw Davies

On 28 May 2015, at 12:04, Marvin wrote: > === wvistau64 (32 bit input) === > input.c:731: Test failed: 0 (a4/0): 10 from 01 -> 00 unexpected > input.c:731: Test failed: 0 (a4/0): 11 from 01 -> 00 unexpected > input.c:731: Test failed: 0 (a4/0): a0 from 01 -> 00 unexpected > input.c:731: Test failed: 0 (a4/0): a2 from 01 -> 00 unexpected These are unrelated to my tests. Huw.

May 28, 2015

Re: Save the Date: WineConf on September 19, 20 in Vienna

by Stefan Dösinger

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Again, The hotel reserved rooms for us, so if you use the regular website booking you'll find that the hotel is empty. I have attached a PDF form to the wiki entry that will allow you to book rooms. You'll have to send them an E-Mail with your request. Cheers, Stefan Am 2015-05-22 um 16:33 schrieb Stefan Dösinger: > Hi, > > I am happy to announce that we will have a WineConf again this > year: It will be on Saturday September 19th and Sunday September > 20th 2015 in Vienna, in the InterCityHotel Wien. > > I have added some basic information to > http://wiki.winehq.org/WineConf2015 , and more detailed information > will follow soon. Please help by suggesting talk topics and adding > them to the list. > > I will write another mail to the list once we have an RSVP page set > up. > > Cheers, Stefan > -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBAgAGBQJVZvhXAAoJEN0/YqbEcdMwgj8QAIdRERGnQv1RarWn/k0Bpx80 8DiHL7EkajdTjx0uiy6rwbCQRF394vengpPZx0aCBw5NgBdzXIKp/00zcvHtJpuh BeS6RdrqT2VlCEG92B3p11wN3ivrtRVRNLsCEbFz5ADf3wYiw/L1MblGQatlRQtj b7BR0dkygpKkJr3nUVcSvSv0WQwMt7hYlAEjXyKio4lMvNhWsWvK/QDVsBx5RQRX mg2/UivIwa9LAS/Ef3W+jsaY8wPc+57n05uD+nQtpzIpCNtKsM7K9XwSHq05Pax6 S0tW0qj7AEPjod+B1WsIPC66o0uTTsDDZMswOpa4DOhMQkp6d2n+bdPWu/8Hg1dR yzunOIGP296/H4NnJWRgtZbK8tuzh6H9IRr+wwEsI0zVxRP0u3IqygfD1t1yBsoL v6LhVORIDHBwayfkTvnKrGQYJf2uvSMpkfJ3lLgyixq/Ul26dU6D0YzRVwUeGqnH kMMBCY0LlSEqNpSn/akWgErY7Pf0Jz4977fyuyfrWEyLhQi+HLVhiDE7YiNH5G2i Ho5beL6mo0x67+t7b0fDwpcnlRTDfgZc6855i9CPimPbNmeh2OG/cUnPPW0Y6Ml2 pQ2DfEVr+fewOIlSLBWyKIvEFoXo7s95oZwcgxwzy55kZxlW484ldrZz6wRU77sx I3YbyrJiVvCfO+Ntn8LI =n6KV -----END PGP SIGNATURE-----

May 28, 2015

Re: winex11: Fix ctrl-<symbol> to generate codes below 0x20 where necessary.

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=13631 Your paranoid android. === wvistau64 (32 bit input) === input.c:731: Test failed: 0 (a4/0): 10 from 01 -> 00 unexpected input.c:731: Test failed: 0 (a4/0): 11 from 01 -> 00 unexpected input.c:731: Test failed: 0 (a4/0): a0 from 01 -> 00 unexpected input.c:731: Test failed: 0 (a4/0): a2 from 01 -> 00 unexpected

May 28, 2015

Re: dpnet: Implement IDirectPlay8Address GetComponentByIndex (try 10)

by Stefan Dösinger

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, I'm sorry for the late review. I completely forgot about this patch (also in the private mail you sent me). It looks good to me, except one issue noted below that I missed earlier. Am 2015-05-23 um 13:22 schrieb Alistair Leslie-Hughes: > index 7e10136..bad84ef 100644 > --- a/dlls/dpnet/address.c > +++ b/dlls/dpnet/address.c > @@ -405,9 +405,58 @@ static HRESULT WINAPI IDirectPlay8AddressImpl_GetComponentByIndex(IDirectPlay8Ad > const DWORD dwComponentID, WCHAR *pwszName, DWORD *pdwNameLen, void *pvBuffer, > DWORD *pdwBufferSize, DWORD *pdwDataType) > { > - IDirectPlay8AddressImpl *This = impl_from_IDirectPlay8Address(iface); > - TRACE("(%p): stub\n", This); > - return DPN_OK; > + IDirectPlay8AddressImpl *This = impl_from_IDirectPlay8Address(iface); > + struct component *entry; > + int namesize; > + > + TRACE("(%p)->(%p %p %p %p)\n", This, pwszName, pvBuffer, pdwBufferSize, pdwDataType); You're not tracing dwComponentID and pdwNameLen here. In particular dwComponentID seems important, as it is the thing that the method call retrieves. > + if(!pdwNameLen || !pdwBufferSize || !pdwDataType) > + return DPNERR_INVALIDPOINTER; > + > + if(dwComponentID > This->comp_count) > + return DPNERR_DOESNOTEXIST; As I've noted in other patches it would be a good idea to write a WARN in error cases, especially in a COM method where +relay can't be used to check return values. You can also print the available and required size in case of DPNERR_BUFFERTOSMALL. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBAgAGBQJVZvMGAAoJEN0/YqbEcdMww2sP+gJRvmzLAZ4oXx3FJFSPIvpn JsbdhoKeZkmdsTCPQjvfPVAUKj1qSQcm1w9rRFBthZhkVWNgXGDTbsbTmtVRiATz qqoX+MjKKQeguWmgnEiQY0k/NW15kAvUkP/vT1qQfKOyqadXJs1a6gli0zfqHSqO uviq2P5mQ82vvIYbv5rS8tAGBTh/FHd1802tSPcMGXQqrqqjW3UH/rV4DymP33FF IsSpPBt2niCz7jSt6QGg/8R9Q2xK8D2sOOe1CcLvtRzrh7Ws+j0p1B8Y73aX8047 s7Pp/oUgfmcEUGheeD+c1YlAQOW6Xp9urjiCc9Pqp+sJBv9fJ5p40r9A6wXrP/Jb Zkei91cF+XcwuhZysQX/qJCwff2VyqG55cDdS1R7DTva/DSamhFB6jYrRAjotgrT g4ulnkbVR4khDzJRVk4K9rxJ2OPwoUkPVpspjrKzuT7oEm2U31PVHPUQa8qeIWzr b1vLyDnG/Zz+B27yHc9sLfwZm5olWz2XTDht7kZKGMU2FRy1Q2wK0wZ6Q6ekB/f5 jnnQep7DmgsMe4EN175k5L5e3yW/SISk0U2O/VtukKMbZQ/hoIhHw8ieiaAo16MM S6nvCRbHxRCe9vB/0zKt5+5SAzWrTTsdw03qdeFg82PSGi6I/F7RrrQnY76GtYKB gtjHMs5QPp0FZRriR0T0 =MUbC -----END PGP SIGNATURE-----

May 28, 2015

Re: [PATCH] kernel32: Return ERROR_CALL_NOT_IMPLEMENTED in GetUserPreferredUILanguages.

by Alexandre Julliard

Mark Harmstone <hellas(a)burntcomma.com> writes: > On 27/05/15 14:12, Alexandre Julliard wrote: >> Not that it really matters, but I don't see why you'd need to change >> Wine in order to be able to write a test. > > How am I supposed to know if GetUserPreferredUILanguages fails because it's not implemented, > or fails for some other reason? That doesn't matter. You are testing the Windows behavior, so unless there's a Windows version that returns ERROR_CALL_NOT_IMPLEMENTED, you should never need to check for it. Once your test succeeds on Windows, you add todo_wine in places where Wine differs. Whether that's because the function is unimplemented, or because it's implemented but broken, doesn't make any difference as far as the test is concerned. -- Alexandre Julliard julliard(a)winehq.org

May 28, 2015