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

Call for winetest ing

by Jeremy White

Okay, so I got winetest to do something rational, and now I'd like to see if it will work for someone else. It's easy! Just apply this patch to a current Wine tree: http://www.winehq.org/pipermail/wine-patches/2008-February/049675.html Then configure and make Wine as you normally would. Then, cd programs/winetest ./runtests <your-tag-here> (The tag I used was jwhite-etch64; I think you should identify who you are, and then if you have multiple systems, your system within that). If you have winetricks in your path, it should be completely unattended. (If you don't, you'll need to hang around to click 'go' to the Gecko install dialog). After about 5 minutes, your test results should be run and sent up to test.winehq.org. The cron job runs every 5 minutes, so sometime after that, a reload of: firefox http://test.winehq.org/data/`date +%Y%m%d`/ Should show you your results! Easy as Pi! There is still quite a bit to do. First, I'm waiting for a 'real' winetest expert to wack me with a clue bat. But I'll stumble along until the hammer falls... Second, I want to do a stupidity check and make sure that winetest essentially correlates to a make -k test run. If they are not really identical, then we need to fix that problem... Next, I think we should probably shift the build-id from being the date to being the HEAD sha1, but that's a radical enough change that I felt I needed to start simpler and within the current conventions. But I'd really appreciate it if folks could try this and see if we can gather some data quickly. In theory, we should be able to quickly see the common points of failure and know where to attack first. Cheers, Jeremy

Feb. 5, 2008

Re: Blocked conformance tests

by Reece Dunn

On 05/02/2008, Francois Gouget <fgouget(a)free.fr> wrote: > > When I run winetest on Windows XP I get some dialog boxes asking me: > > Do you want to keep blocking this program? This is because the Windows firewall is blocking the application from connecting to the internet and/or making TCP/IP calls. > According to the title bar, the impacted tests are: > dplayx > rpcrt4 > 'Microsoft DirectPlay Helper' ??? > ws2_32 > > My understanding is that it should only prevent the test from receiving > network connections which I guess is not common. Does anyone know if > this could cause any test failure currently? If the firewall is blocking the network traffic, it should cause those tests to fail. > How does XP determine whether to block the test or not? > > Is the 'authorisation' based on the executable hash? If so there's no > way to pre-authorise the tests for nightly tests. > > Should the winetest mark the executable so it is not hampered by this > check? I don't know if it is possible to automatically (or even globally) allow these tests to run without disabling the firewall. This would need further investigation. Also, could you add a note about the presence of a firewall to the ConformanceTests wiki page. Thanks, - Reece

Feb. 5, 2008

Re: kernel32: Implement ReplaceFileW. [RESEND]

by Alexandre Julliard

Robert Shearman <rob(a)codeweavers.com> writes: > + /* > + * Open the replacement file for reading, writing, and deleting > + * (writing and deleting are needed when finished) > + */ > + if ((hReplacement = CreateFileW(lpReplacementFileName, > + GENERIC_READ | GENERIC_WRITE, > + FILE_SHARE_READ | FILE_SHARE_WRITE | FILE_SHARE_DELETE, > + NULL, OPEN_EXISTING, 0, 0)) == INVALID_HANDLE_VALUE) > + { > + error = GetLastError(); /* Use error returned by CreateFileW */ > + goto fail; > + } Files should be opened with NtOpenFile the same way MoveFile does, to avoid having to convert path names twice. -- Alexandre Julliard julliard(a)winehq.org

Feb. 5, 2008

Re: [10/20] WineD3D: Make Clear aware of the depth stencil sharing

by H. Verbeet

On 05/02/2008, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > Am Dienstag, 5. Februar 2008 09:39:35 schrieb H. Verbeet: > > On 05/02/2008, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > > > > Stefan already knows this, but just for what it's worth, I've got a > > more complete fix in my tree for depth stencil sharing. Unfortunately > > that patch causes a couple of regressions, so it needs more work. > Ya, I sent this patch nevertheless because this bug is causing some problems > for me and I want the fix in finally. > Yes, the patch should go in.

Feb. 5, 2008

Re: [7/20] WineD3D: De-Statify depth blit opengl resources

by H. Verbeet

On 05/02/2008, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > Am Dienstag, 5. Februar 2008 09:34:50 schrieb H. Verbeet: > > On 05/02/2008, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > > > > I was aware of this issue, but technically these resources should be > > stored in the shader backend, not the device. > Well, the textures are per device, and for now the shader backend doesn't have > its own per device storage, thus I put them in the device > The backend itself should be set per device already, but it needs some private storage.

Feb. 5, 2008

Re: War on conformance test suite failures!

by Stefan Dösinger

Am Dienstag, 5. Februar 2008 02:18:11 schrieb Chris Robinson: > On Monday 04 February 2008 04:06:10 pm Stefan Dösinger wrote: > > The bit that interests me is the sRGB test result. It shows that the sRGB > > correction is applied after the fog, which would allow us to use > > GLX_EXT_framebuffer_sRGB for performance improvements if we can find out > > in which cases games have to expect that. Currently I only know that Half > > Life 2 should fail on this system. > > This might be the reason: > http://msdn2.microsoft.com/en-us/library/bb172583(VS.85).aspx > > D3DPMISCCAPS_POSTBLENDSRGBCONVERT - > Device supports conversion to sRGB after blending. > This flag is available in Direct3D 9Ex only. Blending as in D3DRS_ALPHABLENDENABLE is not used by the game in this case. However, in another cap flag Microsoft called fog blending. So maybe sRGB isn't applied at all with fog enabled on dx9 cards. I'll extend the test case to test that.

Feb. 5, 2008

Re: [7/20] WineD3D: De-Statify depth blit opengl resources

by Stefan Dösinger

Am Dienstag, 5. Februar 2008 09:34:50 schrieb H. Verbeet: > On 05/02/2008, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > > I was aware of this issue, but technically these resources should be > stored in the shader backend, not the device. Well, the textures are per device, and for now the shader backend doesn't have its own per device storage, thus I put them in the device

Feb. 5, 2008

Re: [10/20] WineD3D: Make Clear aware of the depth stencil sharing

by Stefan Dösinger

Am Dienstag, 5. Februar 2008 09:39:35 schrieb H. Verbeet: > On 05/02/2008, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > > Stefan already knows this, but just for what it's worth, I've got a > more complete fix in my tree for depth stencil sharing. Unfortunately > that patch causes a couple of regressions, so it needs more work. Ya, I sent this patch nevertheless because this bug is causing some problems for me and I want the fix in finally.

Feb. 5, 2008

Re: War on conformance test suite failures!

by Stefan Dösinger

Am Dienstag, 5. Februar 2008 07:59:58 schrieb Reece Dunn: > On 05/02/2008, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > > Am Dienstag, 5. Februar 2008 00:50:55 schrieb Dan Kegel: > > > Great progress was made at Wineconf last year on > > > getting our test suite closer to always passing on wine, but > > > http://test.winehq.org/data/200801301937/#Wine shows > > > that, according to the results collected from people who run > > > programs/winetest on Wine, twelve DLLs' tests are currently failing. > > > (Let alone the test failures on Windows!) > > > http://bugs.winehq.org/show_bug.cgi?id=9916 also > > > talks about the fact that "make test" shows failures. > > > > This is only half on-topic, but the d3d9:visual test ran on this machine > > has some interesting results: > > > > Vista: Vista-HomePrem-NoUAC > > > > Do we have any track who reported them? > > That one is mine, along with Vista-HomePrem-UAC. They are both on the > same high-spec recent HP laptop (HP Pavilion DV95000) machine with an > NVidia GeForce 8600M GS graphics card. Do you have Half Life 2? If yes, can you start a new game and take a few screenshots from Doktor Kleiner's lab? If you don't have the game I'll try to find a place to reproduce the fog bug at a place that can be reached in the demo.

Feb. 5, 2008

Re: running "make test" results in an error

by Alexandre Julliard

Stefan Dösinger <stefan(a)codeweavers.com> writes: > Am Montag, 4. Februar 2008 18:41:23 schrieb Christopher: >> err:xrandr:X11DRV_XRandR_SetCurrentMode Resolution change not successful -- >> perhaps display has changed? device.c:753: Test failed: Screen width is >> 1680 >> device.c:754: Test failed: Screen height is 1050 > That is causing the tests to fail. I've never seen that specific behavior > before, but I think that winex11 doesn't return an error if the resolution > change fails. This is because some users have "broken" X11 setups which only > support one resolution. That way games run in a corner of the screen instead > of bailing out. The usual cause for this failure is an Xrandr bug, the timestamp checking is broken WRT wraparound. If you wait a few days (or restart your X server) it will work again. -- Alexandre Julliard julliard(a)winehq.org

Feb. 5, 2008