May 2016 - Wine-devel - list.winehq.org

Re: include: Fix a typo in IDirect3D9Ex method
by Henri Verbeet 22 May '16

22 May '16

On 22 May 2016 at 14:30, Andrey Gusev <andrey.goosev(a)gmail.com> wrote: > - STDMETHOD(GetAdapterLUID)(THIS_ UINT Adatper, LUID *pLUID) PURE; > + STDMETHOD(GetAdapterLUID)(THIS_ UINT Adapter, LUID *pLUID) PURE; If you're touching it anyway, "Adapter" => "adapter" and "pLUID" => "luid".

1 0

Re: [PATCH v2 09/12] d3d10core/tests: Add a geometry shader test.
by Henri Verbeet 22 May '16

22 May '16

On 20 May 2016 at 19:56, Matteo Bruni <mbruni(a)codeweavers.com> wrote: > +#if 0 > +struct gs_data > +{ > + float4 pos : SV_POSITION; > + float4 color : COLOR; > +}; > + > +[maxvertexcount(4)] > +void main(point struct gs_data vin[1], inout TriangleStream<gs_data> vout) > +{ > + float offset = 0.2 * vin[0].pos.w; > + gs_data v; > + > + v.color = vin[0].color; > + > + v.pos = float4(vin[0].pos.x - offset, vin[0].pos.y - offset, vin[0].pos.z, 1.0); > + vout.Append(v); > + v.pos = float4(vin[0].pos.x - offset, vin[0].pos.y + offset, vin[0].pos.z, 1.0); > + vout.Append(v); > + v.pos = float4(vin[0].pos.x + offset, vin[0].pos.y - offset, vin[0].pos.z, 1.0); > + vout.Append(v); > + v.pos = float4(vin[0].pos.x + offset, vin[0].pos.y + offset, vin[0].pos.z, 1.0); > + vout.Append(v); > +} > +#endif > + static const DWORD gs_code[] = > + { > + 0x43425844, 0x5c14d638, 0x1089a334, 0x4103d097, 0x218e3173, 0x00000001, 0x00000428, 0x00000005, > + 0x00000034, 0x0000008c, 0x000000e0, 0x00000134, 0x000003ac, 0x46454452, 0x00000050, 0x00000000, > + 0x00000000, 0x00000000, 0x0000001c, 0x47530400, 0x00000100, 0x0000001c, 0x7263694d, 0x666f736f, > + 0x52282074, 0x4c482029, 0x53204c53, 0x65646168, 0x6f432072, 0x6c69706d, 0x31207265, 0x2e302e30, > + 0x31303031, 0x36312e31, 0x00343833, 0x4e475349, 0x0000004c, 0x00000002, 0x00000008, 0x00000038, > + 0x00000000, 0x00000001, 0x00000003, 0x00000000, 0x00000f0f, 0x00000044, 0x00000000, 0x00000000, > + 0x00000003, 0x00000001, 0x00000f0f, 0x505f5653, 0x5449534f, 0x004e4f49, 0x4f4c4f43, 0xabab0052, > + 0x4e47534f, 0x0000004c, 0x00000002, 0x00000008, 0x00000038, 0x00000000, 0x00000001, 0x00000003, > + 0x00000000, 0x0000000f, 0x00000044, 0x00000000, 0x00000000, 0x00000003, 0x00000001, 0x0000000f, > + 0x505f5653, 0x5449534f, 0x004e4f49, 0x4f4c4f43, 0xabab0052, 0x52444853, 0x00000270, 0x00020040, > + 0x0000009c, 0x05000061, 0x002010f2, 0x00000001, 0x00000000, 0x00000001, 0x0400005f, 0x002010f2, > + 0x00000001, 0x00000001, 0x02000068, 0x00000001, 0x0100085d, 0x0100285c, 0x04000067, 0x001020f2, > + 0x00000000, 0x00000001, 0x03000065, 0x001020f2, 0x00000001, 0x0200005e, 0x00000004, 0x0f000032, > + 0x00100032, 0x00000000, 0x80201ff6, 0x00000041, 0x00000000, 0x00000000, 0x00004002, 0x3e4ccccd, > + 0x3e4ccccd, 0x00000000, 0x00000000, 0x00201046, 0x00000000, 0x00000000, 0x05000036, 0x00102032, > + 0x00000000, 0x00100046, 0x00000000, 0x06000036, 0x00102042, 0x00000000, 0x0020102a, 0x00000000, > + 0x00000000, 0x05000036, 0x00102082, 0x00000000, 0x00004001, 0x3f800000, 0x06000036, 0x001020f2, > + 0x00000001, 0x00201e46, 0x00000000, 0x00000001, 0x01000013, 0x05000036, 0x00102012, 0x00000000, > + 0x0010000a, 0x00000000, 0x0e000032, 0x00100052, 0x00000000, 0x00201ff6, 0x00000000, 0x00000000, > + 0x00004002, 0x3e4ccccd, 0x00000000, 0x3e4ccccd, 0x00000000, 0x00201106, 0x00000000, 0x00000000, > + 0x05000036, 0x00102022, 0x00000000, 0x0010002a, 0x00000000, 0x06000036, 0x00102042, 0x00000000, > + 0x0020102a, 0x00000000, 0x00000000, 0x05000036, 0x00102082, 0x00000000, 0x00004001, 0x3f800000, > + 0x06000036, 0x001020f2, 0x00000001, 0x00201e46, 0x00000000, 0x00000001, 0x01000013, 0x05000036, > + 0x00102012, 0x00000000, 0x0010000a, 0x00000000, 0x05000036, 0x00102022, 0x00000000, 0x0010001a, > + 0x00000000, 0x06000036, 0x00102042, 0x00000000, 0x0020102a, 0x00000000, 0x00000000, 0x05000036, > + 0x00102082, 0x00000000, 0x00004001, 0x3f800000, 0x06000036, 0x001020f2, 0x00000001, 0x00201e46, > + 0x00000000, 0x00000001, 0x01000013, 0x05000036, 0x00102032, 0x00000000, 0x00100086, 0x00000000, > + 0x06000036, 0x00102042, 0x00000000, 0x0020102a, 0x00000000, 0x00000000, 0x05000036, 0x00102082, > + 0x00000000, 0x00004001, 0x3f800000, 0x06000036, 0x001020f2, 0x00000001, 0x00201e46, 0x00000000, > + 0x00000001, 0x01000013, 0x0100003e, 0x54415453, 0x00000074, 0x00000019, 0x00000001, 0x00000000, > + 0x00000004, 0x00000002, 0x00000000, 0x00000000, 0x00000001, 0x00000000, 0x00000000, 0x00000000, > + 0x00000000, 0x00000000, 0x00000004, 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000000, > + 0x00000000, 0x00000000, 0x00000000, 0x00000000, 0x00000001, 0x00000005, 0x00000004, 0x00000000, > + 0x00000000, 0x00000000, > + }; That looks like it still contains debug info.

1 0

Re: [PATCH v3 01/12] wined3d: Alpha test emulation for core profile contexts.
by Henri Verbeet 22 May '16

22 May '16

On 20 May 2016 at 19:55, Matteo Bruni <mbruni(a)codeweavers.com> wrote: > +static void shader_glsl_generate_alpha_test(struct wined3d_string_buffer *buffer, > + const struct wined3d_gl_info *gl_info, enum wined3d_cmp_func alpha_func) > +{ > + /* alpha_func is the PASS condition, not the DISCARD condition. Instead of > + * flipping all the operators here, just negate the comparison below. */ > + static const char * const comparison_operator[] = > + { > + "", /* WINED3D_CMP_NEVER */ > + "<", /* WINED3D_CMP_LESS */ > + "==", /* WINED3D_CMP_EQUAL */ > + "<=", /* WINED3D_CMP_LESSEQUAL */ > + ">", /* WINED3D_CMP_GREATER */ > + "!=", /* WINED3D_CMP_NOTEQUAL */ > + ">=", /* WINED3D_CMP_GREATEREQUAL */ > + "" /* WINED3D_CMP_ALWAYS */ > + }; > + > + if (alpha_func == WINED3D_CMP_ALWAYS) > + return; > + > + if (alpha_func != WINED3D_CMP_NEVER) > + shader_addline(buffer, "if (!(%s[0].a %s alpha_test_ref))\n", > + get_fragment_output(gl_info), comparison_operator[alpha_func - WINED3D_CMP_NEVER]); > + shader_addline(buffer, " discard;\n"); > +} I'm calling this good enough for a start. In case of WINED3D_CMP_EQUAL and WINED3D_CMP_NOTEQUAL in particular there may come a point where we'll have to do a range check like for color keys though.

1 0

Re: [PATCH] d3d11: Honor render state FrontCounterClockwise.
by Józef Kucia 22 May '16

22 May '16

On Sun, May 22, 2016 at 1:32 PM, Johannes Brandstätter <jbrandst(a)2ds.eu> wrote: > This fixes geometry flickering in Unigine's Valley and Heaven bechmarks. > The user interface would at least show some parts, but using this patch > it completly vanishes as well as the credits image upon closing. > > Signed-off-by: Johannes Brandstätter <jbrandst(a)2ds.eu> > --- > dlls/d3d11/device.c | 5 +++-- > 1 file changed, 3 insertions(+), 2 deletions(-) > > diff --git a/dlls/d3d11/device.c b/dlls/d3d11/device.c > index 7fb28ac..d54f81f 100644 > --- a/dlls/d3d11/device.c > +++ b/dlls/d3d11/device.c > @@ -808,10 +808,11 @@ static void STDMETHODCALLTYPE d3d11_immediate_context_RSSetState(ID3D11DeviceCon > > desc = &device->rasterizer_state->desc; > wined3d_device_set_render_state(device->wined3d_device, WINED3D_RS_FILLMODE, desc->FillMode); > - wined3d_device_set_render_state(device->wined3d_device, WINED3D_RS_CULLMODE, desc->CullMode); > /* glFrontFace() */ > if (desc->FrontCounterClockwise) > - FIXME("Ignoring FrontCounterClockwise %#x.\n", desc->FrontCounterClockwise); > + wined3d_device_set_render_state(device->wined3d_device, WINED3D_RS_CULLMODE, WINED3D_CULL_CCW); > + else > + wined3d_device_set_render_state(device->wined3d_device, WINED3D_RS_CULLMODE, desc->CullMode); > /* OpenGL style depth bias. */ > if (desc->DepthBias || desc->SlopeScaledDepthBias) > FIXME("Ignoring depth bias.\n"); > -- > 2.8.1 > > > This patch doesn't make sense. The FrontCounterClockwise field needs to be mapped to glFrontFace(). The best what you can get using the current wined3d API is to invert the cull mode when FrontCounterClockwise is true. if (desc->FrontCounterClockwise) { switch (desc->CullMode) { default: case D3D11_CULL_NONE: cullmode = WINED3D_CULL_NONE; break; case D3D11_CULL_FRONT: cullmode = WINED3D_CULL_CCW; break; case D3D11_CULL_BACK: cullmode = WINED3D_CULL_CW; break; } This still is not a proper solution because the front face property can be read in shaders, see the SV_IsFrontFace system-value semantic. The proper solution is to introduce a rasterizer state object in wined3d API. You could probably start with a rasterizer state object containing a single property for FrontCounterClockwise.

1 0

Re: [PATCH resend] wined3d: Don't accept WINED3DFMT_NULL
by Aaryaman Vasishta 21 May '16

21 May '16

Hi, Thanks for the contribution! I noticed that you're missing something: On Sat, May 21, 2016 at 4:14 PM, Maxqia <contrib(a)maxqia.com> wrote: > This fixes https://bugs.winehq.org/show_bug.cgi?id=40624 > --- > dlls/wined3d/context.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > You need to sign-off on this patch. FWIW, Use git commit --amend -s to add a signed-off by field to your patch and resend it. Cheers, Aaryaman

1 0

Re: [PATCH] jscript: Date.parse: make "UTC"/"GMT" optional before the timezone offset
by Marvin 21 May '16

21 May '16

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://testbot.winehq.org/JobDetails.pl?Key=23146 Your paranoid android. === wxppro (32 bit) === Found no trace of the test summary line or of a crash === w2003std (32 bit) === Found no trace of the test summary line or of a crash === wvistau64 (32 bit) === Found no trace of the test summary line or of a crash === w2008s64 (32 bit) === Found no trace of the test summary line or of a crash === w7u (32 bit) === Found no trace of the test summary line or of a crash === w7pro64 (32 bit) === Found no trace of the test summary line or of a crash === w8 (32 bit) === Found no trace of the test summary line or of a crash === w864 (32 bit) === Found no trace of the test summary line or of a crash === w1064 (32 bit) === Found no trace of the test summary line or of a crash === wvistau64 (64 bit) === Found no trace of the test summary line or of a crash === w2008s64 (64 bit) === Found no trace of the test summary line or of a crash === w7pro64 (64 bit) === Found no trace of the test summary line or of a crash === w864 (64 bit) === Found no trace of the test summary line or of a crash === w1064 (64 bit) === Found no trace of the test summary line or of a crash

1 0

Re: [PATCH 1/6] wined3d: Add wined3d creation flag for selecting unbound resource color.
by Henri Verbeet 20 May '16

20 May '16

On 20 May 2016 at 11:47, Józef Kucia <jkucia(a)codeweavers.com> wrote: > diff --git a/include/wine/wined3d.h b/include/wine/wined3d.h > index 8429bea..79ac440 100644 > --- a/include/wine/wined3d.h > +++ b/include/wine/wined3d.h > @@ -1237,6 +1237,7 @@ enum wined3d_display_rotation > #define WINED3D_PIXEL_CENTER_INTEGER 0x00000080 > #define WINED3D_LEGACY_FFP_LIGHTING 0x00000100 > #define WINED3D_SRGB_READ_WRITE_CONTROL 0x00000200 > +#define WINED3D_UNBOUND_RESOURCE_COLOR_ZERO 0x00000400 > We generally try to make the "reasonable" behaviours the defaults, and only have flags for extra functionality or more quirky behaviours. In this case it's perhaps arguable what the default should be, but all zeroes seems like the more reasonable default to me.

1 0

Re: dnsapi: add DnsGetCacheDataTable stub
by Sebastian Lackner 20 May '16

20 May '16

On 20.05.2016 06:27, Austin English wrote: > +/****************************************************************************** > + * DnsGetCacheDataTable [DNSAPI.@] > + * > + */ > +DNS_GET_CACHE_DATA_TABLE WINAPI DnsGetCacheDataTable( DNSCACHEENTRY *cache ) I'm pretty sure it should not return a function pointer here. Also, you might want to move the struct to header file so that it can be used in tests later.

1 0

Re: ntdll: add WinSqmEndSession/WinSqmStartSession stubs
by Michael Müller 20 May '16

20 May '16

Am 20.05.2016 um 06:15 schrieb Austin English: > +NTSTATUS WINAPI WinSqmStartSession(GUID *sessionguid, DWORD sessionid, DWORD unknown1) > +{ Your usage example indicates that the function returns a HANDLE and not a NTSTATUS value. Returning INVALID_HANDLE_VALUE would therefore be the most suitable value for a stub. The linked code also handles this case. > +@ stdcall WinSqmEndSession(ptr) The parameter is a handle, so you should mark it as long instead of ptr. Btw, the remaining functions seems to be sorted alphabetically, so you also might want to correctly insert this function.

1 0

Re: [PATCH] mshtml/tests: Use todo_wine_if() in tests.
by Jacek Caban 20 May '16

20 May '16

Hi Hadrien, Please resend with your sign off. Other than that, the patch looks good. Thanks, Jacek

1 0