March 2014 - Wine-devel - list.winehq.org

Re: [PATCH 2/4] vbscript: Use VariantCopyInd in invoke_variant_prop.
by Marvin 08 Mar '14

08 Mar '14

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://newtestbot.winehq.org/JobDetails.pl?Key=5509 Your paranoid android. === w2000pro (32 bit) === No test summary line found === wxppro (32 bit) === No test summary line found === wvista (32 bit) === No test summary line found === wvistau64 (32 bit) === No test summary line found === w2008s64 (32 bit) === No test summary line found === w7pro64 (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 === w7pro64 (64 bit) === No test summary line found

1 0

Re: [PATCH 1/4] vbscipt: Use VariantCopyInd in exec_script.
by Marvin 08 Mar '14

08 Mar '14

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://newtestbot.winehq.org/JobDetails.pl?Key=5508 Your paranoid android. === w2000pro (32 bit) === No test summary line found === wxppro (32 bit) === No test summary line found === wvista (32 bit) === No test summary line found === wvistau64 (32 bit) === No test summary line found === w2008s64 (32 bit) === No test summary line found === w7pro64 (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 === w7pro64 (64 bit) === No test summary line found

1 0

[RFC 4/11] Linux FF: Linux effect parameter: periodic effect's phase
by Elias Vanderstuyft 07 Mar '14

07 Mar '14

a) Periodic phase is not implemented conform most efficient implementations of linux FF-devices: phase should be related to time, not to angle: - HW interface example: In /drivers/input/joystick/iforce/iforce-ff.c : make_period_modifier(...): "iforce_send_packet(iforce, FF_CMD_PERIOD, data)" where data[4] = HI(phase); make_core(...): "iforce_send_packet(iforce, FF_CMD_EFFECT, data)" where data[4] = HI(duration); => Conclusion: phase likely to be related to time, not to angle - SW kernel interface example, which is going to be used for all memless FF devices in the future: (http://permalink.gmane.org/gmane.linux.kernel/1620326) in ff-memless-next.c:149 : mlnxeff->playback_time = effect->u.periodic.phase - SW interface example: SDL2-2.0.1/src/haptic/linux/SDL_syshaptic.c:640 There is also written on /include/uapi/linux/input.h:1058 : "@phase: 'horizontal' shift" where 'horizontal' most likely means time dimension, not phase of the waveform. => To be sure, I asked this at http://www.mail-archive.com/linux-input(a)vger.kernel.org/msg08459.html, but didn't receive any answer yet. So change the following in dinput/effect_linuxinput.c:259 : tsp->lOffset = (This->effect.u.periodic.offset / 33) * 10; tsp->dwPhase = (This->effect.u.periodic.phase / 33) * 36; tsp->dwPeriod = (This->effect.u.periodic.period * 1000); to : tsp->lOffset = (This->effect.u.periodic.offset / 33) * 10; if (!This->effect.u.periodic.period) { /* Avoid division by zero period, set tsp->dwPhase arbitrarily to zero. */ tsp->dwPhase = 0; } else tsp->dwPhase = (((int)This->effect.u.periodic.phase * 36000) / This->effect.u.periodic.period) % 36000; tsp->dwPeriod = (This->effect.u.periodic.period * 1000); and change the following in dinput/effect_linuxinput.c:486 : This->effect.u.periodic.offset = (tsp->lOffset / 10) * 32; This->effect.u.periodic.phase = (tsp->dwPhase / 9) * 8; /* == (/ 36 * 32) */ This->effect.u.periodic.period = tsp->dwPeriod / 1000; to : This->effect.u.periodic.offset = (tsp->lOffset / 10) * 32; This->effect.u.periodic.period = tsp->dwPeriod / 1000; This->effect.u.periodic.phase = ((int)(tsp->dwPhase % 36000) * This->effect.u.periodic.period) / 36000; Elias

2 2

[RFC 3/11] Linux FF: Linux effect parameter: start delay
by Elias Vanderstuyft 07 Mar '14

07 Mar '14

a) Keep in mind that startDelay is not supported in DX5 or lower: In the SetParameters(...) case, we're sure This->effect.replay.delay is set to zero, because the HEAP_ZERO_MEMORY flag was set when allocating the LinuxInputEffectImpl 'This' struct in linuxinput_create_effect(...). So change dinput/effect_linuxinput.c:223 : if (dwFlags & DIEP_STARTDELAY) { peff->dwStartDelay = This->effect.replay.delay * 1000; } to : if ((dwFlags & DIEP_STARTDELAY) && (peff->dwSize > sizeof(DIEFFECT_DX5))) { peff->dwStartDelay = This->effect.replay.delay * 1000; } and change dinput/effect_linuxinput.c:463 : if (dwFlags & DIEP_STARTDELAY) This->effect.replay.delay = peff->dwStartDelay / 1000; to : if ((dwFlags & DIEP_STARTDELAY) && (peff->dwSize > sizeof(DIEFFECT_DX5))) This->effect.replay.delay = peff->dwStartDelay / 1000; Elias

2 2

Re: oleaut32: Resolve TypeLib dependencies with GUID
by guillaume.charifi＠sfr.fr 07 Mar '14

07 Mar '14

Hello, I have no certainty, and not even a test, but the Windows version of Skype (which you know is released by Microsoft) tries to load a TypeLib named "uc.tlb", which cannot be find in path, and even in the registry. In fact, this library is into Skype.exe resources section, but it can only be found using its GUID (only the GUID is present in the library, not the name "uc.tlb". In fact it is a unnamed library). So there seems that two possibilities remains : - All the Typelibs of the executable are probed at launch (I don't think so, but...) - The Windows' oleaut32.dll scans the TypeLibs embedded into the executable and tries to match them not only by their name, but by their GUID to. Thanks for your reading, Guillaume ======================================== Message du : 07/03/2014 16:18 De : "Jacek Caban " <jacek(a)codeweavers.com> A : wine-devel(a)winehq.org, guillaume.charifi(a)sfr.fr Copie à : Sujet : Re: oleaut32: Resolve TypeLib dependencies with GUID On 03/07/14 15:53, guillaume.charifi(a)sfr.fr wrote: > Bug: http://bugs.winehq.org/show_bug.cgi?id=34184 Please send a test convincing that that's what Windows does. We already have a typelib in oleaur32 tests, so this should be a matter of loading it by GUID without having it registered. Thanks, Jacek

1 0

Re: oleaut32: Resolve TypeLib dependencies with GUID
by Jacek Caban 07 Mar '14

07 Mar '14

On 03/07/14 15:53, guillaume.charifi(a)sfr.fr wrote: > Bug: http://bugs.winehq.org/show_bug.cgi?id=34184 Please send a test convincing that that's what Windows does. We already have a typelib in oleaur32 tests, so this should be a matter of loading it by GUID without having it registered. Thanks, Jacek

1 0

Re: [PATCH 2/2] vbscript: Directly access 'this' object properties in interpreter.
by Marvin 07 Mar '14

07 Mar '14

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://newtestbot.winehq.org/JobDetails.pl?Key=5490 Your paranoid android. === w2000pro (32 bit) === No test summary line found === wxppro (32 bit) === No test summary line found === wvista (32 bit) === No test summary line found === wvistau64 (32 bit) === No test summary line found === w2008s64 (32 bit) === No test summary line found === w7pro64 (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 === w7pro64 (64 bit) === No test summary line found

1 0

Re: [PATCH 1/2] vbscipt: Added beginning support for arrays as class members.
by Marvin 07 Mar '14

07 Mar '14

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://newtestbot.winehq.org/JobDetails.pl?Key=5489 Your paranoid android. === w2000pro (32 bit) === No test summary line found === wxppro (32 bit) === No test summary line found === wvista (32 bit) === No test summary line found === wvistau64 (32 bit) === No test summary line found === w2008s64 (32 bit) === No test summary line found === w7pro64 (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 === w7pro64 (64 bit) === No test summary line found

1 0

Re: ws2_32/tests: Mark some test results on Windows 8 as broken.
by Marvin 07 Mar '14

07 Mar '14

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://newtestbot.winehq.org/JobDetails.pl?Key=5487 Your paranoid android. === wvistau64 (32 bit sock) === sock: unhandled exception c0000005 at 00000000 === w2008s64 (32 bit sock) === sock: unhandled exception c0000005 at 00000000 === wvistau64 (64 bit sock) === sock: unhandled exception c0000005 at 0000000000000000

1 0

Re: [PATCH 2/5] d3d8/tests: Use create_device() in more tests.
by Marvin 07 Mar '14

07 Mar '14

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://newtestbot.winehq.org/JobDetails.pl?Key=5480 Your paranoid android. === w8 (32 bit device) === device.c:2396: Test failed: Expected foreground window 0005019E, got 00010048. device.c:3038: Test failed: Expected wndproc 0x40f890, got 0x40f890.

1 0