June 2008 - Wine-Devel - list.winehq.org

Re: Time to revert bad commits?

by Roderick Colenbrander

> I'm afraid we are out of time to fix bugs that introduces major d3d > regressions. I've yet to see a single patch that addressed those issues. > > It's time to start reverting bad commits and testing the results. > > The bugs I'm talking about: > Bug 10580 > Bug 11584 > Bug 12794 > Bug 13101 > > Vitaliy. > Regarding 13101 that bug is an Nvidia driver bug. The code itself is really correct and there are more cases in which the issue gets triggered (not just multisampling). I need more testers (I added comments to the bug) to reproduce it on different systems, games and driver versions so that I can create a good bug report to submit to Nvidia. Roderick -- GMX startet ShortView.de. Hier findest Du Leute mit Deinen Interessen! Jetzt dabei sein: http://www.shortview.de/?mc=sv_ext_mf(a)gmx

June 2, 2008

Re: [1/3] ddraw/tests: Detach surface if attaching worked by surprise (or refrast)

by Michael Karcher

Am Sonntag, den 01.06.2008, 18:19 -0500 schrieb James Hawkins: > On Sun, Jun 1, 2008 at 5:47 PM, Michael Karcher > <wine(a)mkarcher.dialup.fu-berlin.de> wrote: > > --- > > dlls/ddraw/tests/dsurface.c | 3 +++ > > 1 files changed, 3 insertions(+), 0 deletions(-) > > > Patch already sent: > http://winehq.org/pipermail/wine-patches/2008-May/055424.html Yes, right. Sorry for the noise. Regards, Michael Karcher

June 2, 2008

Re: [2/3] ddraw/tests: Add another refrast exception

by Michael Karcher

Am Sonntag, den 01.06.2008, 18:23 -0500 schrieb James Hawkins: > On Sun, Jun 1, 2008 at 5:48 PM, Michael Karcher > <wine(a)mkarcher.dialup.fu-berlin.de> wrote: > > --- > > dlls/ddraw/tests/dsurface.c | 5 ++++- > > 1 files changed, 4 insertions(+), 1 deletions(-) > > > I've gone through all the latest test results, and this failure isn't > happening anywhere that I've seen. Please link to where you're seeing > this. You are right. I also don't see it on the web. I have seen it on my own system, in an virtual machine (kvm/qemu). It only happens on 16 bpp. On 24 bpp it fails in failure modes also seen on the test page, because refrast doesn't support 24 bpp. (I get the typical "12 failures, 2 skipped" failure mode on 24 bpp)[1]. Probably no one runs the tests at 16 bpp. kvm/qemu's virtual card (Cirrus Logic 5446) does not support 32 bpp. If you declare 16 bpp as generally unsupported, please ignore this patch. If you declare virtual machines as unsupported, I can try that on a real machine (with Win'98 instead of XP, though). I really have a real hardware board of the card emulated by qemu, so I can try on that very hardware (it runs on vga.dll, with a small addition for modesetting and banking). Regards Michael Karcher [1]: http://test.winehq.org/data/dae64e67fc9b1bf1b1950b86f7d58184a4b4c942/xp_XPp… http://test.winehq.org/data/dae64e67fc9b1bf1b1950b86f7d58184a4b4c942/xp_win…

June 2, 2008

Time to revert bad commits?

by Vitaliy Margolen

I'm afraid we are out of time to fix bugs that introduces major d3d regressions. I've yet to see a single patch that addressed those issues. It's time to start reverting bad commits and testing the results. The bugs I'm talking about: Bug 10580 Bug 11584 Bug 12794 Bug 13101 Vitaliy.

June 2, 2008

Re: Fixing crashes in Tests (OS version check)

by Juan Lang

> The thing is that this evolves with time. Back in 2000 nobody would have > called SnmpUtilOidToA with a NULL parameter as it would have meant > crashing on a lot of the Windows install base. Nowadays every ISV > expects you to have at least Windows XP and wouldn't care one bit if > that call means crashing the application on NT4 or lower. > > So you could say that nowadays applications are likely to expect > SnmpUtilOidToA(NULL) not to crash which means keeping these tests. > However I don't think our conformance tests should crash on older > systems and that's what makes things tricky... I agree with you that resolving issues like this can be subtle, since there will always be specific examples when a general principle does not apply. They don't invalidate the general principle unless they outnumber it or outweigh it in some way. So, in my opinion, the general principle is this: if the behavior is very different on different Windows versions, and if the behavior on some Windows version is to crash, it seems reasonable to remove the test, as: 1) we want our tests to reflect Windows behavior that applications count on, and different behavior (especially crashing) often implies the applications don't count on a particular behavior, and 2) having regression tests crash is undesirable, as many other valid tests will not be run on a platform. This doesn't mean that we can't make Wine not crash given the bogus input, it just means we may not be able to have a regression test for this behavior. And, of course, there may be specific instances in which we wish to test potentially crashy behavior, but I'd argue that they better be really important to override points 1) and 2). And coming back to this test: I still think it's easier just to remove the test than to come up with a hacky way of avoiding it on a crashy system. --Juan

June 2, 2008

Re: Winetest grrr...

by Juan Lang

> here's a workaround for 'extracting' the test executables > from Paul Millar's exe. Start winetest, wait until the first test > starts, and then kill winetest.exe (or close the winetest window). Then > you can find the test binaries in c:\Windows\Temp\wct (or possibly > c:\Documents and Settings\your account\Local Settings\Temp or something > like that). Then you can run just the test you want. Thanks, I'll do that! --Juan

June 2, 2008

Re: [2/3] ddraw/tests: Add another refrast exception

by James Hawkins

On Sun, Jun 1, 2008 at 5:48 PM, Michael Karcher <wine(a)mkarcher.dialup.fu-berlin.de> wrote: > --- > dlls/ddraw/tests/dsurface.c | 5 ++++- > 1 files changed, 4 insertions(+), 1 deletions(-) > I've gone through all the latest test results, and this failure isn't happening anywhere that I've seen. Please link to where you're seeing this. -- James Hawkins

June 1, 2008

Re: [1/3] ddraw/tests: Detach surface if attaching worked by surprise (or refrast)

by James Hawkins

On Sun, Jun 1, 2008 at 5:47 PM, Michael Karcher <wine(a)mkarcher.dialup.fu-berlin.de> wrote: > --- > dlls/ddraw/tests/dsurface.c | 3 +++ > 1 files changed, 3 insertions(+), 0 deletions(-) > Patch already sent: http://winehq.org/pipermail/wine-patches/2008-May/055424.html -- James Hawkins

June 1, 2008

Re: How long does it take you to compile wine?

by Evil Jay

real 5m22.625s user 17m16.993s sys 1m1.336s This is on Kubuntu 8.04 w/GCC 4.2.3. The system uses a Q6600 overclocked to 2.9Ghz, 4GB of DDR2 RAM(a)1066Mhz, and a 9600GT video card. I had Amarok and some other stuff going in the background, but nothing that would make a huge diff. -J

June 1, 2008

iexplore/gecko problems with recent builds

by Steven Edwards

Hi, I did not find anything about this in bugzilla. Is anyone else having problems using builtin IE with gecko? I keep getting this error with the tip err:mshtml:before_async_open GetWineURL returned NULL -- Steven Edwards "There is one thing stronger than all the armies in the world, and that is an idea whose time has come." - Victor Hugo

June 1, 2008