November 2011 - Wine-Devel - list.winehq.org

Re: [PATCH] kernel32: implement GetLogicalProcessorInformation

by GOUJON Alexandre

On 11/25/2011 08:35 PM, Claudio Fontana wrote: > That's great! When it's done it will be one less out of tree patch to > maintain, always a win. > > Ciao, > > C You're free to keep your patches on your own/private tree but if you're discussing here, I guess you wanted your stuff being committed in the main/official tree. People here take time to read patches and to give useful advices to make patches accepted. Open Source philosophy is "I know how to do that better, let's build together something great", not "I'll code this part myself and keep it private". Of course, it's my two cents but it's sad people react like you.

Nov. 26, 2011

Re: [PATCH 1/5] wined3d: Don't apply fixups to converted surfaces.

by Stefan Dösinger

Am Freitag, 25. November 2011, 20:23:01 schrieb Henri Verbeet: > > I don't think this is correct. E.g. signed formats without > > GL_NV_texture_shader have load time and read time fixups and both have to > > be applied. > > That's silly. If that's really the case that should be fixed. I don't see why it is silly. d3d sysmem format -> gl texture format and gl texture format -> rgba conversions are two separate things. What's unfortunate is that in the case of P8 both can do the same job and the selection code is a mess. For signed surfaces the upload conversion maps [-1.0;1.0] to [0.0;1.0] to load it into a unsigned rgba surface. When we read the surface in the shader we have to reverse that. You may be able to avoid the upload conversion by a more tricky shader conversion, but that misses the point. (And yeah, SNORM<->UNORM blits are broken for other reasons.) > P8 -> RGBA. I'd say always store the index in the alpha component. We can do that since P8 textures are disabled, there's no other use for the alpha value. Also the primary_render_target_is_p8 is stupid. This way the additional shader conversion is redundant, but produces the correct result. Not perfect, but works for Wine 1.4. After Wine 1.4 the code should be changed to use the load time P8 conversion only if shader conversion isn't available, and only for the final blit to the screen(and software blits otherwise). Also, do you have a game that needs P8->RGBA blits? In my testing with ddraw, P8->RGBA blits don't do what you expect. They ignore the palette, and replicate the index to all channels(ie, 0xa4 -> 0xa4a4a4a4).

Nov. 25, 2011

Re: [PATCH] kernel32: implement GetLogicalProcessorInformation

by Claudio Fontana

On Fri, Nov 25, 2011 at 8:01 PM, Charles Davis <cdavis(a)mymail.mines.edu> wrote: > > On Nov 25, 2011, at 8:17 AM, Claudio Fontana wrote: > >> On Fri, Nov 25, 2011 at 4:07 PM, Claudio Fontana >> <claudio.fontana(a)gmail.com> wrote: >>> I have resent the patch as a "try 2" (there was a try 1.5 to fix test >>> issues under windows older than XP SP3). >>> >>> The patch contains changes based on your feedback. >> >> Which is completely broken, since the U() macro you suggested to use >> is a wine/test.h feature only, >> and completely broke the patch. > You are right. The way most files (other than tests) in Wine deal with that is that they unconditionally define NONAMELESSUNION and NONAMELESSSTRUCT. So you wouldn't use U(buffer[i]) (except in the tests, where you can do that), you'd use buffer[i].u . >> So I fall back to my try 1.5 for my tree. > OK, you do that. I'm pretty sure AJ won't take your patch anyway until it's done "right" (i.e. like I said, in ntdll). I guess I'll take that up, since I have nothing better to do ;). That's great! When it's done it will be one less out of tree patch to maintain, always a win. Ciao, C

Nov. 25, 2011

Re: [PATCH 1/5] wined3d: Don't apply fixups to converted surfaces.

by Henri Verbeet

On 25 November 2011 19:50, Stefan Dösinger <stefandoesinger(a)gmx.at> wrote: > Am Donnerstag, 24. November 2011, 22:52:32 schrieb Henri Verbeet: >> + if (surface->flags & SFLAG_CONVERTED) >> + { >> + ENTER_GL(); >> + glEnable(textype); >> + checkGLcall("glEnable(textype)"); >> + LEAVE_GL(); >> + return WINED3D_OK; >> + } > I don't think this is correct. E.g. signed formats without > GL_NV_texture_shader have load time and read time fixups and both have to be > applied. > That's silly. If that's really the case that should be fixed. Note that this case would already be broken with the current code though, arbfp_blit_set() only handles P8 and the various YUV fixups. More generally, I have some doubts about whether the way converted surfaces currently work is really what we want. > What exactly are you trying to fix? I assume it is something about P8 blits, > but P8->P8 blits don't enter the ARB blitting code(due to the dest fixup > check), and I don't know of any app that uses P8->RGB or SNORM->UNORM blits. > P8 -> RGBA.

Nov. 25, 2011

Re: [PATCH] kernel32: implement GetLogicalProcessorInformation

by Charles Davis

On Nov 25, 2011, at 8:17 AM, Claudio Fontana wrote: > On Fri, Nov 25, 2011 at 4:07 PM, Claudio Fontana > <claudio.fontana(a)gmail.com> wrote: >> I have resent the patch as a "try 2" (there was a try 1.5 to fix test >> issues under windows older than XP SP3). >> >> The patch contains changes based on your feedback. > > Which is completely broken, since the U() macro you suggested to use > is a wine/test.h feature only, > and completely broke the patch. You are right. The way most files (other than tests) in Wine deal with that is that they unconditionally define NONAMELESSUNION and NONAMELESSSTRUCT. So you wouldn't use U(buffer[i]) (except in the tests, where you can do that), you'd use buffer[i].u . > So I fall back to my try 1.5 for my tree. OK, you do that. I'm pretty sure AJ won't take your patch anyway until it's done "right" (i.e. like I said, in ntdll). I guess I'll take that up, since I have nothing better to do ;). Chip

Nov. 25, 2011

Re: [PATCH 1/5] wined3d: Don't apply fixups to converted surfaces.

by Stefan Dösinger

Am Donnerstag, 24. November 2011, 22:52:32 schrieb Henri Verbeet: > + if (surface->flags & SFLAG_CONVERTED) > + { > + ENTER_GL(); > + glEnable(textype); > + checkGLcall("glEnable(textype)"); > + LEAVE_GL(); > + return WINED3D_OK; > + } I don't think this is correct. E.g. signed formats without GL_NV_texture_shader have load time and read time fixups and both have to be applied. What exactly are you trying to fix? I assume it is something about P8 blits, but P8->P8 blits don't enter the ARB blitting code(due to the dest fixup check), and I don't know of any app that uses P8->RGB or SNORM->UNORM blits. Stefan

Nov. 25, 2011

Re: dinput: Stub IDirectInputJoyConfig8 interface. (try 2)

by Vitaliy Margolen

On 11/25/2011 04:14 AM, Jacek Caban wrote: > Hi Vitaliy, > > On 11/25/11 04:17, Vitaliy Margolen wrote: >> Fix QueryInterface, simplify use of AddRef, and replace >> "LPDIRECTINPUTJOYCONFIG8" >> with "IDirectInputJoyConfig8 *". >> --- >> dlls/dinput/dinput_main.c | 7 ++ >> dlls/dinput/dinput_private.h | 2 + >> dlls/dinput/joystick.c | 196 ++++++++++++++++++++++++++++++++++ >> include/dinputd.h | 243 ++++++++++++++++++++++++++++++++++++++++++ >> 4 files changed, 448 insertions(+), 0 deletions(-) >> create mode 100644 include/dinputd.h >> > > + if (IsEqualGUID(&IID_IDirectInputJoyConfig8, riid )) > + { > + return create_joy_config( riid, ppobj ); > + } > > > This has the same problem as we discussed about msxml3 lately: > > http://www.winehq.org/pipermail/wine-devel/2011-November/093279.html > > Are you sure native is broken the same way? You were right. It is a part of IDirectInput not a standalone class. Sent a fixed patch. Vitaliy.

Nov. 25, 2011

Re: [1/2] kernel32: Add a test for protections accepted by CreateFileMapping, make it pass under Wine.

by Dmitry Timoshkov

Alexandre Julliard <julliard(a)winehq.org> wrote: > > @@ -350,9 +350,6 @@ HANDLE WINAPI CreateFileMappingW( HANDLE hFile, LPSECURITY_ATTRIBUTES sa, > > > > switch(protect) > > { > > - case 0: > > - protect = PAGE_READONLY; /* Win9x compatibility */ > > - /* fall through */ > > Did you try your test on Win9x? No, i have no a win9x machine around. -- Dmitry.

Nov. 25, 2011

Re: [1/2] kernel32: Add a test for protections accepted by CreateFileMapping, make it pass under Wine.

by Alexandre Julliard

Dmitry Timoshkov <dmitry(a)baikal.ru> writes: > Alexandre Julliard <julliard(a)winehq.org> wrote: > >> > @@ -350,9 +350,6 @@ HANDLE WINAPI CreateFileMappingW( HANDLE hFile, LPSECURITY_ATTRIBUTES sa, >> > >> > switch(protect) >> > { >> > - case 0: >> > - protect = PAGE_READONLY; /* Win9x compatibility */ >> > - /* fall through */ >> >> Did you try your test on Win9x? > > No, i have no a win9x machine around. Hmmm, we should keep one available on the testbot for cases like this. -- Alexandre Julliard julliard(a)winehq.org

Nov. 25, 2011

Re: [1/2] kernel32: Add a test for protections accepted by CreateFileMapping, make it pass under Wine.

by Alexandre Julliard

Dmitry Timoshkov <dmitry(a)baikal.ru> writes: > @@ -350,9 +350,6 @@ HANDLE WINAPI CreateFileMappingW( HANDLE hFile, LPSECURITY_ATTRIBUTES sa, > > switch(protect) > { > - case 0: > - protect = PAGE_READONLY; /* Win9x compatibility */ > - /* fall through */ Did you try your test on Win9x? -- Alexandre Julliard julliard(a)winehq.org

Nov. 25, 2011