February 2007 - Wine-devel - list.winehq.org

Re: [9/12] WineD3D: When multithreading, call glFinish after drawing
by H. Verbeet 01 Mar '07

01 Mar '07

Would GL_NV_fence be usefull here?

2 1

Re: [6/6] d3d9: Add a mova rounding test
by Alexandre Julliard 01 Mar '07

01 Mar '07

"H. Verbeet" <hverbeet(a)gmail.com> writes: > It should be noted that native d3d rounds to nearest even while Wine > doesn't. Applications are not supposed to depend on that though. > > Changelog: > - Add a mova rounding test The test fails here: ../../../tools/runtest -q -P wine -M d3d9.dll -T ../../.. -p d3d9_test.exe.so visual.c && touch visual.ok visual.c:405: Test failed: Expected color 00ffff00, got 00000000 (for input -2.400000) visual.c:405: Test failed: Expected color 00ffff00, got 00000000 (for input -1.600000) visual.c:405: Test failed: Expected color 0000ffff, got 00000000 (for input -0.400000) visual.c:405: Test failed: Expected color 0000ffff, got 00000000 (for input 0.400000) visual.c:405: Test failed: Expected color 00ff00ff, got 00000000 (for input 1.600000) visual.c:405: Test failed: Expected color 00ff00ff, got 00000000 (for input 2.400000) make: *** [visual.ok] Error 6 -- Alexandre Julliard julliard(a)winehq.org

3 4

Wine's Wiki Defaced
by Francois Gouget 01 Mar '07

01 Mar '07

Someone has defaced Wine's Wiki frontpage. Also I have either forgotten my password or the user accounts don't work anymore. Could someone look into this? -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ War doesn't determine who's right. War determines who's left.

2 2

One school's list of must-have apps
by Dan Kegel 28 Feb '07

28 Feb '07

Here's a school that actually did transition to Linux: http://searchopensource.techtarget.com/originalContent/0,289142,sid39_gci12… They didn't use Wine. Here's what they said about legacy windows apps: "Using Wyse terminals, the district also planned to offer students and staff thin-client desktops running OpenOffice.org. Ericom software – a Citrix alternative -- enabled the terminals to run the district's existing and irreplaceable Microsoft Windows educational applications, including Type to Learn, Reading Counts and Kid Pix." - Dan

1 0

Using ex-user32 controls from native comctl32 6
by Felix Nawothnig 28 Feb '07

28 Feb '07

(CC-ing to Frank Richter because I hope he knows the answer :) Hi. I've been trying to make wine use the ex-user32 controls (listbox, scrollbox, etc.) in comctl32 where Microsoft copied them to in version 6 to get message traces of those controls (I know I could use msg spy tools on Windows but comparing two Wine spy logs is way more straight forward, especially since those tools don't get you stuff like +win). You need a manifest file on Windows to get application to use comctl6 but we don't use that since we've chosen to theme the user controls via subclassing. I've tried to just remove registration of the classes from our user32 and use native comctl32 6 (which otherwise works nice) but no success. So, any idea how Windows makes comctl32 register the user classes? Felix

2 2

msiexec.exe service registration
by Jeremy 28 Feb '07

28 Feb '07

In msiexec.c, the DoRegServer calls OpenSCManager, however, on a fresh installation, ie no ~/.wine, the HKLM\CurrentControlSet\Services does not yet exist, so that OpenSCManager fails. What is the correct way to guarantee that this key exists? Regards, Jeremy

2 1

msi automation patches
by Misha Koshelev 28 Feb '07

28 Feb '07

Ok, so since my patches weren't commited to git yet, I've consolidated the new automation changes I've made in the attached patches. James and Rob, if they look good to you I will submit these version to wine-patches (so my total will once again be four patches, 1 is http://www.winehq.org/pipermail/wine-patches/2007-February/036359.html , 2 and 3 attached, and 4 is http://www.winehq.org/pipermail/wine-patches/2007-February/036362.html ) and wait to submit anything more for automation until all 4 patches are in. Please let me know Rob (I used DispGetParam, thanks a lot, so probably just need to look at my AutomationObject::Invoke and then the individual object Invoke handlers in the 0003 patch below) and James (I guess in regards to comments and formatting). Thanks a lot Misha

1 0

Re: [8/8] WineD3D: Use the source surface's texture for writing the backup
by H. Verbeet 28 Feb '07

28 Feb '07

On 28/02/07, Stefan Dösinger <stefan(a)codeweavers.com> wrote: >The hwstretch blit code creates a new texture each time it is called to back up the back buffer and >releases it afterwards. It is more efficient to keep the texture and release it with the surface. This >patch uses the source surface's texture to do the job. Because the source surface is an active render >target its texture is either only needed for blitting(onscreen target) or will be completely overwritten >on the next render target change(offscreen target). Is this supposed to work with FBOs?

1 0

Re: [4/8] WineD3D: Prepare only fixed function samplers for blit
by H. Verbeet 28 Feb '07

28 Feb '07

On 28/02/07, Stefan Dösinger <stefan(a)codeweavers.com> wrote: >+ /* The blitting code uses (for now) the fixed function pipeline, so make sure to reset all fixed >+ * function texture unit. No need to care for higher samplers >+ */ >+ for(i = GL_LIMITS(texture_stages) - 1; i > 0 ; i--) { You should use GL_LIMITS(textures) here, since that's the number of fixed function texture units.

1 0

Re: [8/12] WineD3D: Create multithreading contexts for swapchains
by Mirek 28 Feb '07

28 Feb '07

Stefan Dösinger napsal(a): > This patch finishes the opengl side of multithreaded direct3d, which was > started with the state management rewrite(well, almost, no offscreen rendering > yet) It does not make Direct3D thread safe, because there is no protection > against race conditions. > > This patch improves multithreaded games a bit. They do not crash any more in a > gl call, but running them is kinda like a lottery, especially on smp systems. > > Vitaly and Andras have reported success with Prince of Persia 3D: Sands of > Time and some other games, and on my laptop Empire Earth is running. Empire > Earth deadlocks regularly on my dual core system at ddraw creation. I can even run Tom Clancy's Rainbow Six Vegas, but there are still problems with offscreen rendering. Mirek > > > ------------------------------------------------------------------------ > > From 49cc9a0aa6e06f559540758d5b0ed2252208a0ad Mon Sep 17 00:00:00 2001 > From: Stefan Doesinger <stefan(a)codeweavers.com> > Date: Sun, 25 Feb 2007 00:18:25 +0100 > Subject: [PATCH] WineD3D: Create multithreading contexts for swapchains > > This patch adds a routine to clone the main context of a swapchain to use it with a different thread on the > same drawable. This will enable multithreaded direct3d rendering, or better enable an attempt to do that. > From the opengl point of view it will work, but because concurrency control is not implemented yet(appart > of ENTER_GL() / LEAVE_GL() ) games may crash randomly, show strange behavior, ... > > The other problem is that this code was only tested with the open source radeon driver yet. While what > wined3d is doing here should be allowed by the spec, some drivers may have some bugs when using multiple > contexts on one drawable. > --- > dlls/wined3d/context.c | 4 ++-- > dlls/wined3d/swapchain.c | 30 ++++++++++++++++++++++++++++++ > dlls/wined3d/wined3d_private.h | 2 ++ > 3 files changed, 34 insertions(+), 2 deletions(-) > > diff --git a/dlls/wined3d/context.c b/dlls/wined3d/context.c > index 3a815b3..8545568 100644 > --- a/dlls/wined3d/context.c > +++ b/dlls/wined3d/context.c > @@ -626,8 +626,8 @@ void ActivateContext(IWineD3DDeviceImpl *This, IWineD3DSurface *target, ContextU > } > > if(!context) { > - /* TODO: Create a new context for the thread */ > - FIXME("Context creation for a new thread not implemented yet\n"); > + /* Create a new context for the thread */ > + context = IWineD3DSwapChainImpl_CreateContextForThread(swapchain); > } > } else { > context = ((IWineD3DSwapChainImpl *) swapchain)->context[0]; > diff --git a/dlls/wined3d/swapchain.c b/dlls/wined3d/swapchain.c > index 2977a05..3b8e3e8 100644 > --- a/dlls/wined3d/swapchain.c > +++ b/dlls/wined3d/swapchain.c > @@ -510,3 +510,33 @@ const IWineD3DSwapChainVtbl IWineD3DSwapChain_Vtbl = > IWineD3DSwapChainImpl_SetGammaRamp, > IWineD3DSwapChainImpl_GetGammaRamp > }; > + > +WineD3DContext *IWineD3DSwapChainImpl_CreateContextForThread(IWineD3DSwapChain *iface) { > + WineD3DContext *ctx; > + IWineD3DSwapChainImpl *This = (IWineD3DSwapChainImpl *) iface; > + WineD3DContext **newArray; > + > + TRACE("Creating a new context for swapchain %p, thread %d\n", This, GetCurrentThreadId()); > + > + ctx = CreateContext(This->wineD3DDevice, (IWineD3DSurfaceImpl *) This->frontBuffer, > + This->context[0]->display, This->win); > + if(!ctx) { > + ERR("Failed to create a new context for the swapchain\n"); > + return NULL; > + } > + > + newArray = HeapAlloc(GetProcessHeap(), 0, sizeof(*newArray) * This->num_contexts + 1); > + if(!newArray) { > + ERR("Out of memory when trying to allocate a new context array\n"); > + DestroyContext(This->wineD3DDevice, ctx); > + return NULL; > + } > + memcpy(newArray, This->context, sizeof(*newArray) * This->num_contexts); > + HeapFree(GetProcessHeap(), 0, This->context); > + newArray[This->num_contexts] = ctx; > + This->context = newArray; > + This->num_contexts++; > + > + TRACE("Returning context %p\n", ctx); > + return ctx; > +} > diff --git a/dlls/wined3d/wined3d_private.h b/dlls/wined3d/wined3d_private.h > index 45e69bf..fa4232c 100644 > --- a/dlls/wined3d/wined3d_private.h > +++ b/dlls/wined3d/wined3d_private.h > @@ -1341,6 +1341,8 @@ typedef struct IWineD3DSwapChainImpl > > extern const IWineD3DSwapChainVtbl IWineD3DSwapChain_Vtbl; > > +WineD3DContext *IWineD3DSwapChainImpl_CreateContextForThread(IWineD3DSwapChain *iface); > + > /***************************************************************************** > * Utility function prototypes > */ > > > ------------------------------------------------------------------------ > >

2 1