wined3d: Introduce D3D7+ normal matrix computation.

dlls/d3d8/tests/visual.c

@@ -283,18 +283,13 @@ static void lighting_test(void)
         unsigned int size;
         DWORD expected;
         const char *message;
-        BOOL todo;
     }
     tests[] =
     {
-        {&mat, nquad, sizeof(nquad[0]), 0x000000ff,
-         "Lit quad with light", FALSE},
-        {&mat_singular, nquad, sizeof(nquad[0]), 0x000000ff,
-         "Lit quad with singular world matrix", FALSE},
-        {&mat_transf, rotatedquad, sizeof(rotatedquad[0]), 0x000000ff,
-         "Lit quad with transformation matrix", FALSE},
-        {&mat_nonaffine, translatedquad, sizeof(translatedquad[0]), 0x00000000,
-         "Lit quad with non-affine matrix", TRUE},
+        {&mat, nquad, sizeof(nquad[0]), 0x000000ff, "Lit quad with light"},
+        {&mat_singular, nquad, sizeof(nquad[0]), 0x000000ff, "Lit quad with singular world matrix"},
+        {&mat_transf, rotatedquad, sizeof(rotatedquad[0]), 0x000000ff, "Lit quad with transformation matrix"},
+        {&mat_nonaffine, translatedquad, sizeof(translatedquad[0]), 0x00000000, "Lit quad with non-affine matrix"},
     };
 
     window = CreateWindowA("static", "d3d8_test", WS_OVERLAPPEDWINDOW | WS_VISIBLE,
@@ -397,10 +392,7 @@ static void lighting_test(void)
         ok(SUCCEEDED(hr), "Failed to end scene, hr %#x.\n", hr);
 
         color = getPixelColor(device, 320, 240);
-        if (tests[i].todo)
-            todo_wine ok(color == tests[i].expected, "%s has color 0x%08x.\n", tests[i].message, color);
-        else
-            ok(color == tests[i].expected, "%s has color 0x%08x.\n", tests[i].message, color);
+        ok(color == tests[i].expected, "%s has color 0x%08x.\n", tests[i].message, color);
     }
 
     refcount = IDirect3DDevice8_Release(device);

dlls/d3d9/tests/visual.c

@@ -367,18 +367,13 @@ static void lighting_test(void)
         unsigned int size;
         DWORD expected;
         const char *message;
-        BOOL todo;
     }
     tests[] =
     {
-        {&mat, nquad, sizeof(nquad[0]), 0x000000ff,
-         "Lit quad with light", FALSE},
-        {&mat_singular, nquad, sizeof(nquad[0]), 0x000000ff,
-         "Lit quad with singular world matrix", FALSE},
-        {&mat_transf, rotatedquad, sizeof(rotatedquad[0]), 0x000000ff,
-         "Lit quad with transformation matrix", FALSE},
-        {&mat_nonaffine, translatedquad, sizeof(translatedquad[0]), 0x00000000,
-         "Lit quad with non-affine matrix", TRUE},
+        {&mat, nquad, sizeof(nquad[0]), 0x000000ff, "Lit quad with light"},
+        {&mat_singular, nquad, sizeof(nquad[0]), 0x000000ff, "Lit quad with singular world matrix"},
+        {&mat_transf, rotatedquad, sizeof(rotatedquad[0]), 0x000000ff, "Lit quad with transformation matrix"},
+        {&mat_nonaffine, translatedquad, sizeof(translatedquad[0]), 0x00000000, "Lit quad with non-affine matrix"},
     };
 
     window = CreateWindowA("static", "d3d9_test", WS_OVERLAPPEDWINDOW | WS_VISIBLE,
@@ -481,10 +476,7 @@ static void lighting_test(void)
         ok(SUCCEEDED(hr), "Failed to end scene, hr %#x.\n", hr);
 
         color = getPixelColor(device, 320, 240);
-        if (tests[i].todo)
-            todo_wine ok(color == tests[i].expected, "%s has color 0x%08x.\n", tests[i].message, color);
-        else
-            ok(color == tests[i].expected, "%s has color 0x%08x.\n", tests[i].message, color);
+        ok(color == tests[i].expected, "%s has color 0x%08x.\n", tests[i].message, color);
     }
 
     hr = IDirect3DDevice9_SetTransform(device, D3DTS_WORLD, &mat);

dlls/ddraw/ddraw.c

@@ -4832,7 +4832,7 @@ static const struct wined3d_device_parent_ops ddraw_wined3d_device_parent_ops =
     device_parent_create_swapchain,
 };
 
-HRESULT ddraw_init(struct ddraw *ddraw, enum wined3d_device_type device_type)
+HRESULT ddraw_init(struct ddraw *ddraw, DWORD flags, enum wined3d_device_type device_type)
 {
     WINED3DCAPS caps;
     HRESULT hr;
@@ -4849,9 +4849,11 @@ HRESULT ddraw_init(struct ddraw *ddraw, enum wined3d_device_type device_type)
     ddraw->numIfaces = 1;
     ddraw->ref7 = 1;
 
-    if (!(ddraw->wined3d = wined3d_create(DDRAW_WINED3D_FLAGS)))
+    flags |= DDRAW_WINED3D_FLAGS;
+    if (!(ddraw->wined3d = wined3d_create(flags)))
     {
-        if (!(ddraw->wined3d = wined3d_create(DDRAW_WINED3D_FLAGS | WINED3D_NO3D)))
+        flags |= WINED3D_NO3D;
+        if (!(ddraw->wined3d = wined3d_create(flags)))
         {
             WARN("Failed to create a wined3d object.\n");
             return E_FAIL;

dlls/ddraw/ddraw_private.h

@@ -122,7 +122,7 @@ struct ddraw
 
 #define DDRAW_WINDOW_CLASS_NAME "DirectDrawDeviceWnd"
 
-HRESULT ddraw_init(struct ddraw *ddraw, enum wined3d_device_type device_type) DECLSPEC_HIDDEN;
+HRESULT ddraw_init(struct ddraw *ddraw, DWORD flags, enum wined3d_device_type device_type) DECLSPEC_HIDDEN;
 void ddraw_d3dcaps1_from_7(D3DDEVICEDESC *caps1, D3DDEVICEDESC7 *caps7) DECLSPEC_HIDDEN;
 void ddraw_destroy_swapchain(struct ddraw *ddraw) DECLSPEC_HIDDEN;
 HRESULT ddraw_get_d3dcaps(const struct ddraw *ddraw, D3DDEVICEDESC7 *caps) DECLSPEC_HIDDEN;

dlls/ddraw/main.c

@@ -230,15 +230,13 @@ DDRAW_Create(const GUID *guid,
     enum wined3d_device_type device_type;
     struct ddraw *ddraw;
     HRESULT hr;
+    DWORD flags = 0;
 
     TRACE("driver_guid %s, ddraw %p, outer_unknown %p, interface_iid %s.\n",
             debugstr_guid(guid), DD, UnkOuter, debugstr_guid(iid));
 
     *DD = NULL;
 
-    /* We don't care about this guids. Well, there's no special guid anyway
-     * OK, we could
-     */
     if (guid == (GUID *) DDCREATE_EMULATIONONLY)
     {
         /* Use the reference device id. This doesn't actually change anything,
@@ -260,6 +258,9 @@ DDRAW_Create(const GUID *guid,
     if (UnkOuter != NULL)
         return CLASS_E_NOAGGREGATION;
 
+    if (!IsEqualGUID(iid, &IID_IDirectDraw7))
+        flags = WINED3D_LEGACY_FFP_LIGHTING;
+
     /* DirectDraw creation comes here */
     ddraw = HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY, sizeof(*ddraw));
     if (!ddraw)
@@ -268,7 +269,7 @@ DDRAW_Create(const GUID *guid,
         return E_OUTOFMEMORY;
     }
 
-    hr = ddraw_init(ddraw, device_type);
+    hr = ddraw_init(ddraw, flags, device_type);
     if (FAILED(hr))
     {
         WARN("Failed to initialize ddraw object, hr %#x.\n", hr);

dlls/ddraw/tests/ddraw7.c

@@ -3495,14 +3495,13 @@ static void test_lighting(void)
         void *quad;
         DWORD expected;
         const char *message;
-        BOOL todo;
     }
     tests[] =
     {
-        {&mat, nquad, 0x000000ff, "Lit quad with light", FALSE},
-        {&mat_singular, nquad, 0x000000ff, "Lit quad with singular world matrix", FALSE},
-        {&mat_transf, rotatedquad, 0x000000ff, "Lit quad with transformation matrix", FALSE},
-        {&mat_nonaffine, translatedquad, 0x00000000, "Lit quad with non-affine matrix", TRUE},
+        {&mat, nquad, 0x000000ff, "Lit quad with light"},
+        {&mat_singular, nquad, 0x000000ff, "Lit quad with singular world matrix"},
+        {&mat_transf, rotatedquad, 0x000000ff, "Lit quad with transformation matrix"},
+        {&mat_nonaffine, translatedquad, 0x00000000, "Lit quad with non-affine matrix"},
     };
 
     HWND window;
@@ -3609,10 +3608,7 @@ static void test_lighting(void)
         ok(SUCCEEDED(hr), "Failed to end scene, hr %#x.\n", hr);
 
         color = get_surface_color(rt, 320, 240);
-        if (tests[i].todo)
-            todo_wine ok(color == tests[i].expected, "%s has color 0x%08x.\n", tests[i].message, color);
-        else
-            ok(color == tests[i].expected, "%s has color 0x%08x.\n", tests[i].message, color);
+        ok(color == tests[i].expected, "%s has color 0x%08x.\n", tests[i].message, color);
     }
 
     IDirectDrawSurface7_Release(rt);

dlls/wined3d/glsl_shader.c

@@ -816,6 +816,218 @@ static BOOL invert_matrix_3d(struct wined3d_matrix *out, const struct wined3d_ma
     return TRUE;
 }
 
+static void swap_rows(float **a, float **b)
+{
+    float *tmp = *a;
+
+    *a = *b;
+    *b = tmp;
+}
+
+static BOOL invert_matrix(struct wined3d_matrix *out, struct wined3d_matrix *m)
+{
+    float wtmp[4][8];
+    float m0, m1, m2, m3, s;
+    float *r0, *r1, *r2, *r3;
+
+    r0 = wtmp[0];
+    r1 = wtmp[1];
+    r2 = wtmp[2];
+    r3 = wtmp[3];
+
+    r0[0] = m->_11;
+    r0[1] = m->_12;
+    r0[2] = m->_13;
+    r0[3] = m->_14;
+    r0[4] = 1.0f;
+    r0[5] = r0[6] = r0[7] = 0.0f;
+
+    r1[0] = m->_21;
+    r1[1] = m->_22;
+    r1[2] = m->_23;
+    r1[3] = m->_24;
+    r1[5] = 1.0f;
+    r1[4] = r1[6] = r1[7] = 0.0f;
+
+    r2[0] = m->_31;
+    r2[1] = m->_32;
+    r2[2] = m->_33;
+    r2[3] = m->_34;
+    r2[6] = 1.0f;
+    r2[4] = r2[5] = r2[7] = 0.0f;
+
+    r3[0] = m->_41;
+    r3[1] = m->_42;
+    r3[2] = m->_43;
+    r3[3] = m->_44;
+    r3[7] = 1.0f;
+    r3[4] = r3[5] = r3[6] = 0.0f;
+
+    /* Choose pivot - or die. */
+    if (fabsf(r3[0]) > fabsf(r2[0]))
+        swap_rows(&r3, &r2);
+    if (fabsf(r2[0]) > fabsf(r1[0]))
+        swap_rows(&r2, &r1);
+    if (fabsf(r1[0]) > fabsf(r0[0]))
+        swap_rows(&r1, &r0);
+    if (r0[0] == 0.0f)
+        return FALSE;
+
+    /* Eliminate first variable. */
+    m1 = r1[0] / r0[0]; m2 = r2[0] / r0[0]; m3 = r3[0] / r0[0];
+    s = r0[1]; r1[1] -= m1 * s; r2[1] -= m2 * s; r3[1] -= m3 * s;
+    s = r0[2]; r1[2] -= m1 * s; r2[2] -= m2 * s; r3[2] -= m3 * s;
+    s = r0[3]; r1[3] -= m1 * s; r2[3] -= m2 * s; r3[3] -= m3 * s;
+    s = r0[4];
+    if (s != 0.0f)
+    {
+        r1[4] -= m1 * s;
+        r2[4] -= m2 * s;
+        r3[4] -= m3 * s;
+    }
+    s = r0[5];
+    if (s != 0.0f)
+    {
+        r1[5] -= m1 * s;
+        r2[5] -= m2 * s;
+        r3[5] -= m3 * s;
+    }
+    s = r0[6];
+    if (s != 0.0f)
+    {
+        r1[6] -= m1 * s;
+        r2[6] -= m2 * s;
+        r3[6] -= m3 * s;
+    }
+    s = r0[7];
+    if (s != 0.0f)
+    {
+        r1[7] -= m1 * s;
+        r2[7] -= m2 * s;
+        r3[7] -= m3 * s;
+    }
+
+    /* Choose pivot - or die. */
+    if (fabsf(r3[1]) > fabsf(r2[1]))
+        swap_rows(&r3, &r2);
+    if (fabsf(r2[1]) > fabsf(r1[1]))
+        swap_rows(&r2, &r1);
+    if (r1[1] == 0.0f)
+        return FALSE;
+
+    /* Eliminate second variable. */
+    m2 = r2[1] / r1[1]; m3 = r3[1] / r1[1];
+    r2[2] -= m2 * r1[2]; r3[2] -= m3 * r1[2];
+    r2[3] -= m2 * r1[3]; r3[3] -= m3 * r1[3];
+    s = r1[4];
+    if (s != 0.0f)
+    {
+        r2[4] -= m2 * s;
+        r3[4] -= m3 * s;
+    }
+    s = r1[5];
+    if (s != 0.0f)
+    {
+        r2[5] -= m2 * s;
+        r3[5] -= m3 * s;
+    }
+    s = r1[6];
+    if (s != 0.0f)
+    {
+        r2[6] -= m2 * s;
+        r3[6] -= m3 * s;
+    }
+    s = r1[7];
+    if (s != 0.0f)
+    {
+        r2[7] -= m2 * s;
+        r3[7] -= m3 * s;
+    }
+
+    /* Choose pivot - or die. */
+    if (fabsf(r3[2]) > fabsf(r2[2]))
+        swap_rows(&r3, &r2);
+    if (r2[2] == 0.0f)
+        return FALSE;
+
+    /* Eliminate third variable. */
+    m3 = r3[2] / r2[2];
+    r3[3] -= m3 * r2[3];
+    r3[4] -= m3 * r2[4];
+    r3[5] -= m3 * r2[5];
+    r3[6] -= m3 * r2[6];
+    r3[7] -= m3 * r2[7];
+
+    /* Last check. */
+    if (r3[3] == 0.0f)
+        return FALSE;
+
+    /* Back substitute row 3. */
+    s = 1.0f / r3[3];
+    r3[4] *= s;
+    r3[5] *= s;
+    r3[6] *= s;
+    r3[7] *= s;
+
+    /* Back substitute row 2. */
+    m2 = r2[3];
+    s = 1.0f / r2[2];
+    r2[4] = s * (r2[4] - r3[4] * m2);
+    r2[5] = s * (r2[5] - r3[5] * m2);
+    r2[6] = s * (r2[6] - r3[6] * m2);
+    r2[7] = s * (r2[7] - r3[7] * m2);
+    m1 = r1[3];
+    r1[4] -= r3[4] * m1;
+    r1[5] -= r3[5] * m1;
+    r1[6] -= r3[6] * m1;
+    r1[7] -= r3[7] * m1;
+    m0 = r0[3];
+    r0[4] -= r3[4] * m0;
+    r0[5] -= r3[5] * m0;
+    r0[6] -= r3[6] * m0;
+    r0[7] -= r3[7] * m0;
+
+    /* Back substitute row 1. */
+    m1 = r1[2];
+    s = 1.0f / r1[1];
+    r1[4] = s * (r1[4] - r2[4] * m1);
+    r1[5] = s * (r1[5] - r2[5] * m1);
+    r1[6] = s * (r1[6] - r2[6] * m1);
+    r1[7] = s * (r1[7] - r2[7] * m1);
+    m0 = r0[2];
+    r0[4] -= r2[4] * m0;
+    r0[5] -= r2[5] * m0;
+    r0[6] -= r2[6] * m0;
+    r0[7] -= r2[7] * m0;
+
+    /* Back substitute row 0. */
+    m0 = r0[1];
+    s = 1.0f / r0[0];
+    r0[4] = s * (r0[4] - r1[4] * m0);
+    r0[5] = s * (r0[5] - r1[5] * m0);
+    r0[6] = s * (r0[6] - r1[6] * m0);
+    r0[7] = s * (r0[7] - r1[7] * m0);
+
+    out->_11 = r0[4];
+    out->_12 = r0[5];
+    out->_13 = r0[6];
+    out->_14 = r0[7];
+    out->_21 = r1[4];
+    out->_22 = r1[5];
+    out->_23 = r1[6];
+    out->_24 = r1[7];
+    out->_31 = r2[4];
+    out->_32 = r2[5];
+    out->_33 = r2[6];
+    out->_34 = r2[7];
+    out->_41 = r3[4];
+    out->_42 = r3[5];
+    out->_43 = r3[6];
+    out->_44 = r3[7];
+
+    return TRUE;
+}
+
 static void shader_glsl_ffp_vertex_normalmatrix_uniform(const struct wined3d_context *context,
         const struct wined3d_state *state, struct glsl_shader_prog_link *prog)
 {
@@ -824,11 +1036,11 @@ static void shader_glsl_ffp_vertex_normalmatrix_uniform(const struct wined3d_con
     struct wined3d_matrix mv;
     unsigned int i, j;
 
-    /* gl_NormalMatrix is defined in the OpenGL spec as "transpose of the
-     * inverse of the upper leftmost 3x3 of gl_ModelViewMatrix" and that
-     * seems to be correct for D3D too. */
     get_modelview_matrix(context, state, &mv);
-    invert_matrix_3d(&mv, &mv);
+    if (context->swapchain->device->wined3d->flags & WINED3D_LEGACY_FFP_LIGHTING)
+        invert_matrix_3d(&mv, &mv);
+    else
+        invert_matrix(&mv, &mv);
     /* Tests show that singular modelview matrices are used unchanged as normal
      * matrices on D3D3 and older. There seems to be no clearly consistent
      * behavior on newer D3D versions so always follow older ddraw behavior. */

include/wine/wined3d.h

@@ -1248,6 +1248,7 @@ enum wined3d_display_rotation
 #define WINED3D_FOCUS_MESSAGES                                  0x00000020
 #define WINED3D_HANDLE_RESTORE                                  0x00000040
 #define WINED3D_PIXEL_CENTER_INTEGER                            0x00000080
+#define WINED3D_LEGACY_FFP_LIGHTING                             0x00000100
 
 #define WINED3D_RESZ_CODE                                       0x7fa05000