wine/dlls/wined3d/surface.c
2011-05-13 17:05:27 +02:00

7497 lines
266 KiB
C

/*
* Copyright 1997-2000 Marcus Meissner
* Copyright 1998-2000 Lionel Ulmer
* Copyright 2000-2001 TransGaming Technologies Inc.
* Copyright 2002-2005 Jason Edmeades
* Copyright 2002-2003 Raphael Junqueira
* Copyright 2004 Christian Costa
* Copyright 2005 Oliver Stieber
* Copyright 2006-2008 Stefan Dösinger for CodeWeavers
* Copyright 2007-2008 Henri Verbeet
* Copyright 2006-2008 Roderick Colenbrander
* Copyright 2009-2011 Henri Verbeet for CodeWeavers
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library; if not, write to the Free Software
* Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA
*/
#include "config.h"
#include "wine/port.h"
#include "wined3d_private.h"
WINE_DEFAULT_DEBUG_CHANNEL(d3d_surface);
WINE_DECLARE_DEBUG_CHANNEL(d3d);
static HRESULT surface_cpu_blt(struct wined3d_surface *dst_surface, const RECT *dst_rect,
struct wined3d_surface *src_surface, const RECT *src_rect, DWORD flags,
const WINEDDBLTFX *fx, WINED3DTEXTUREFILTERTYPE filter);
static HRESULT surface_cpu_bltfast(struct wined3d_surface *dst_surface, DWORD dst_x, DWORD dst_y,
struct wined3d_surface *src_surface, const RECT *src_rect, DWORD trans);
static HRESULT IWineD3DSurfaceImpl_BltOverride(struct wined3d_surface *dst_surface, const RECT *dst_rect,
struct wined3d_surface *src_surface, const RECT *src_rect, DWORD flags, const WINEDDBLTFX *fx,
WINED3DTEXTUREFILTERTYPE filter);
static void surface_cleanup(struct wined3d_surface *surface)
{
TRACE("surface %p.\n", surface);
if (surface->texture_name || (surface->flags & SFLAG_PBO) || !list_empty(&surface->renderbuffers))
{
const struct wined3d_gl_info *gl_info;
renderbuffer_entry_t *entry, *entry2;
struct wined3d_context *context;
context = context_acquire(surface->resource.device, NULL);
gl_info = context->gl_info;
ENTER_GL();
if (surface->texture_name)
{
TRACE("Deleting texture %u.\n", surface->texture_name);
glDeleteTextures(1, &surface->texture_name);
}
if (surface->flags & SFLAG_PBO)
{
TRACE("Deleting PBO %u.\n", surface->pbo);
GL_EXTCALL(glDeleteBuffersARB(1, &surface->pbo));
}
LIST_FOR_EACH_ENTRY_SAFE(entry, entry2, &surface->renderbuffers, renderbuffer_entry_t, entry)
{
TRACE("Deleting renderbuffer %u.\n", entry->id);
gl_info->fbo_ops.glDeleteRenderbuffers(1, &entry->id);
HeapFree(GetProcessHeap(), 0, entry);
}
LEAVE_GL();
context_release(context);
}
if (surface->flags & SFLAG_DIBSECTION)
{
/* Release the DC. */
SelectObject(surface->hDC, surface->dib.holdbitmap);
DeleteDC(surface->hDC);
/* Release the DIB section. */
DeleteObject(surface->dib.DIBsection);
surface->dib.bitmap_data = NULL;
surface->resource.allocatedMemory = NULL;
}
if (surface->flags & SFLAG_USERPTR)
wined3d_surface_set_mem(surface, NULL);
if (surface->overlay_dest)
list_remove(&surface->overlay_entry);
HeapFree(GetProcessHeap(), 0, surface->palette9);
resource_cleanup(&surface->resource);
}
void surface_set_container(struct wined3d_surface *surface, enum wined3d_container_type type, void *container)
{
TRACE("surface %p, container %p.\n", surface, container);
if (!container && type != WINED3D_CONTAINER_NONE)
ERR("Setting NULL container of type %#x.\n", type);
if (type == WINED3D_CONTAINER_SWAPCHAIN)
{
surface->get_drawable_size = get_drawable_size_swapchain;
}
else
{
switch (wined3d_settings.offscreen_rendering_mode)
{
case ORM_FBO:
surface->get_drawable_size = get_drawable_size_fbo;
break;
case ORM_BACKBUFFER:
surface->get_drawable_size = get_drawable_size_backbuffer;
break;
default:
ERR("Unhandled offscreen rendering mode %#x.\n", wined3d_settings.offscreen_rendering_mode);
return;
}
}
surface->container.type = type;
surface->container.u.base = container;
}
struct blt_info
{
GLenum binding;
GLenum bind_target;
enum tex_types tex_type;
GLfloat coords[4][3];
};
struct float_rect
{
float l;
float t;
float r;
float b;
};
static inline void cube_coords_float(const RECT *r, UINT w, UINT h, struct float_rect *f)
{
f->l = ((r->left * 2.0f) / w) - 1.0f;
f->t = ((r->top * 2.0f) / h) - 1.0f;
f->r = ((r->right * 2.0f) / w) - 1.0f;
f->b = ((r->bottom * 2.0f) / h) - 1.0f;
}
static void surface_get_blt_info(GLenum target, const RECT *rect, GLsizei w, GLsizei h, struct blt_info *info)
{
GLfloat (*coords)[3] = info->coords;
struct float_rect f;
switch (target)
{
default:
FIXME("Unsupported texture target %#x\n", target);
/* Fall back to GL_TEXTURE_2D */
case GL_TEXTURE_2D:
info->binding = GL_TEXTURE_BINDING_2D;
info->bind_target = GL_TEXTURE_2D;
info->tex_type = tex_2d;
coords[0][0] = (float)rect->left / w;
coords[0][1] = (float)rect->top / h;
coords[0][2] = 0.0f;
coords[1][0] = (float)rect->right / w;
coords[1][1] = (float)rect->top / h;
coords[1][2] = 0.0f;
coords[2][0] = (float)rect->left / w;
coords[2][1] = (float)rect->bottom / h;
coords[2][2] = 0.0f;
coords[3][0] = (float)rect->right / w;
coords[3][1] = (float)rect->bottom / h;
coords[3][2] = 0.0f;
break;
case GL_TEXTURE_RECTANGLE_ARB:
info->binding = GL_TEXTURE_BINDING_RECTANGLE_ARB;
info->bind_target = GL_TEXTURE_RECTANGLE_ARB;
info->tex_type = tex_rect;
coords[0][0] = rect->left; coords[0][1] = rect->top; coords[0][2] = 0.0f;
coords[1][0] = rect->right; coords[1][1] = rect->top; coords[1][2] = 0.0f;
coords[2][0] = rect->left; coords[2][1] = rect->bottom; coords[2][2] = 0.0f;
coords[3][0] = rect->right; coords[3][1] = rect->bottom; coords[3][2] = 0.0f;
break;
case GL_TEXTURE_CUBE_MAP_POSITIVE_X:
info->binding = GL_TEXTURE_BINDING_CUBE_MAP_ARB;
info->bind_target = GL_TEXTURE_CUBE_MAP_ARB;
info->tex_type = tex_cube;
cube_coords_float(rect, w, h, &f);
coords[0][0] = 1.0f; coords[0][1] = -f.t; coords[0][2] = -f.l;
coords[1][0] = 1.0f; coords[1][1] = -f.t; coords[1][2] = -f.r;
coords[2][0] = 1.0f; coords[2][1] = -f.b; coords[2][2] = -f.l;
coords[3][0] = 1.0f; coords[3][1] = -f.b; coords[3][2] = -f.r;
break;
case GL_TEXTURE_CUBE_MAP_NEGATIVE_X:
info->binding = GL_TEXTURE_BINDING_CUBE_MAP_ARB;
info->bind_target = GL_TEXTURE_CUBE_MAP_ARB;
info->tex_type = tex_cube;
cube_coords_float(rect, w, h, &f);
coords[0][0] = -1.0f; coords[0][1] = -f.t; coords[0][2] = f.l;
coords[1][0] = -1.0f; coords[1][1] = -f.t; coords[1][2] = f.r;
coords[2][0] = -1.0f; coords[2][1] = -f.b; coords[2][2] = f.l;
coords[3][0] = -1.0f; coords[3][1] = -f.b; coords[3][2] = f.r;
break;
case GL_TEXTURE_CUBE_MAP_POSITIVE_Y:
info->binding = GL_TEXTURE_BINDING_CUBE_MAP_ARB;
info->bind_target = GL_TEXTURE_CUBE_MAP_ARB;
info->tex_type = tex_cube;
cube_coords_float(rect, w, h, &f);
coords[0][0] = f.l; coords[0][1] = 1.0f; coords[0][2] = f.t;
coords[1][0] = f.r; coords[1][1] = 1.0f; coords[1][2] = f.t;
coords[2][0] = f.l; coords[2][1] = 1.0f; coords[2][2] = f.b;
coords[3][0] = f.r; coords[3][1] = 1.0f; coords[3][2] = f.b;
break;
case GL_TEXTURE_CUBE_MAP_NEGATIVE_Y:
info->binding = GL_TEXTURE_BINDING_CUBE_MAP_ARB;
info->bind_target = GL_TEXTURE_CUBE_MAP_ARB;
info->tex_type = tex_cube;
cube_coords_float(rect, w, h, &f);
coords[0][0] = f.l; coords[0][1] = -1.0f; coords[0][2] = -f.t;
coords[1][0] = f.r; coords[1][1] = -1.0f; coords[1][2] = -f.t;
coords[2][0] = f.l; coords[2][1] = -1.0f; coords[2][2] = -f.b;
coords[3][0] = f.r; coords[3][1] = -1.0f; coords[3][2] = -f.b;
break;
case GL_TEXTURE_CUBE_MAP_POSITIVE_Z:
info->binding = GL_TEXTURE_BINDING_CUBE_MAP_ARB;
info->bind_target = GL_TEXTURE_CUBE_MAP_ARB;
info->tex_type = tex_cube;
cube_coords_float(rect, w, h, &f);
coords[0][0] = f.l; coords[0][1] = -f.t; coords[0][2] = 1.0f;
coords[1][0] = f.r; coords[1][1] = -f.t; coords[1][2] = 1.0f;
coords[2][0] = f.l; coords[2][1] = -f.b; coords[2][2] = 1.0f;
coords[3][0] = f.r; coords[3][1] = -f.b; coords[3][2] = 1.0f;
break;
case GL_TEXTURE_CUBE_MAP_NEGATIVE_Z:
info->binding = GL_TEXTURE_BINDING_CUBE_MAP_ARB;
info->bind_target = GL_TEXTURE_CUBE_MAP_ARB;
info->tex_type = tex_cube;
cube_coords_float(rect, w, h, &f);
coords[0][0] = -f.l; coords[0][1] = -f.t; coords[0][2] = -1.0f;
coords[1][0] = -f.r; coords[1][1] = -f.t; coords[1][2] = -1.0f;
coords[2][0] = -f.l; coords[2][1] = -f.b; coords[2][2] = -1.0f;
coords[3][0] = -f.r; coords[3][1] = -f.b; coords[3][2] = -1.0f;
break;
}
}
static inline void surface_get_rect(struct wined3d_surface *surface, const RECT *rect_in, RECT *rect_out)
{
if (rect_in)
*rect_out = *rect_in;
else
{
rect_out->left = 0;
rect_out->top = 0;
rect_out->right = surface->resource.width;
rect_out->bottom = surface->resource.height;
}
}
/* GL locking and context activation is done by the caller */
void draw_textured_quad(struct wined3d_surface *src_surface, const RECT *src_rect,
const RECT *dst_rect, WINED3DTEXTUREFILTERTYPE Filter)
{
struct blt_info info;
surface_get_blt_info(src_surface->texture_target, src_rect, src_surface->pow2Width, src_surface->pow2Height, &info);
glEnable(info.bind_target);
checkGLcall("glEnable(bind_target)");
/* Bind the texture */
glBindTexture(info.bind_target, src_surface->texture_name);
checkGLcall("glBindTexture");
/* Filtering for StretchRect */
glTexParameteri(info.bind_target, GL_TEXTURE_MAG_FILTER,
wined3d_gl_mag_filter(magLookup, Filter));
checkGLcall("glTexParameteri");
glTexParameteri(info.bind_target, GL_TEXTURE_MIN_FILTER,
wined3d_gl_min_mip_filter(minMipLookup, Filter, WINED3DTEXF_NONE));
checkGLcall("glTexParameteri");
glTexParameteri(info.bind_target, GL_TEXTURE_WRAP_S, GL_CLAMP);
glTexParameteri(info.bind_target, GL_TEXTURE_WRAP_T, GL_CLAMP);
glTexEnvi(GL_TEXTURE_ENV, GL_TEXTURE_ENV_MODE, GL_REPLACE);
checkGLcall("glTexEnvi");
/* Draw a quad */
glBegin(GL_TRIANGLE_STRIP);
glTexCoord3fv(info.coords[0]);
glVertex2i(dst_rect->left, dst_rect->top);
glTexCoord3fv(info.coords[1]);
glVertex2i(dst_rect->right, dst_rect->top);
glTexCoord3fv(info.coords[2]);
glVertex2i(dst_rect->left, dst_rect->bottom);
glTexCoord3fv(info.coords[3]);
glVertex2i(dst_rect->right, dst_rect->bottom);
glEnd();
/* Unbind the texture */
glBindTexture(info.bind_target, 0);
checkGLcall("glBindTexture(info->bind_target, 0)");
/* We changed the filtering settings on the texture. Inform the
* container about this to get the filters reset properly next draw. */
if (src_surface->container.type == WINED3D_CONTAINER_TEXTURE)
{
struct wined3d_texture *texture = src_surface->container.u.texture;
texture->texture_rgb.states[WINED3DTEXSTA_MAGFILTER] = WINED3DTEXF_POINT;
texture->texture_rgb.states[WINED3DTEXSTA_MINFILTER] = WINED3DTEXF_POINT;
texture->texture_rgb.states[WINED3DTEXSTA_MIPFILTER] = WINED3DTEXF_NONE;
}
}
static HRESULT surface_create_dib_section(struct wined3d_surface *surface)
{
const struct wined3d_format *format = surface->resource.format;
SYSTEM_INFO sysInfo;
BITMAPINFO *b_info;
int extraline = 0;
DWORD *masks;
UINT usage;
HDC dc;
TRACE("surface %p.\n", surface);
if (!(format->flags & WINED3DFMT_FLAG_GETDC))
{
WARN("Cannot use GetDC on a %s surface.\n", debug_d3dformat(format->id));
return WINED3DERR_INVALIDCALL;
}
switch (format->byte_count)
{
case 2:
case 4:
/* Allocate extra space to store the RGB bit masks. */
b_info = HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY, sizeof(BITMAPINFOHEADER) + 3 * sizeof(DWORD));
break;
case 3:
b_info = HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY, sizeof(BITMAPINFOHEADER));
break;
default:
/* Allocate extra space for a palette. */
b_info = HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY,
sizeof(BITMAPINFOHEADER) + sizeof(RGBQUAD) * (1 << (format->byte_count * 8)));
break;
}
if (!b_info)
return E_OUTOFMEMORY;
/* Some applications access the surface in via DWORDs, and do not take
* the necessary care at the end of the surface. So we need at least
* 4 extra bytes at the end of the surface. Check against the page size,
* if the last page used for the surface has at least 4 spare bytes we're
* safe, otherwise add an extra line to the DIB section. */
GetSystemInfo(&sysInfo);
if( ((surface->resource.size + 3) % sysInfo.dwPageSize) < 4)
{
extraline = 1;
TRACE("Adding an extra line to the DIB section.\n");
}
b_info->bmiHeader.biSize = sizeof(BITMAPINFOHEADER);
/* TODO: Is there a nicer way to force a specific alignment? (8 byte for ddraw) */
b_info->bmiHeader.biWidth = wined3d_surface_get_pitch(surface) / format->byte_count;
b_info->bmiHeader.biHeight = 0 - surface->resource.height - extraline;
b_info->bmiHeader.biSizeImage = (surface->resource.height + extraline)
* wined3d_surface_get_pitch(surface);
b_info->bmiHeader.biPlanes = 1;
b_info->bmiHeader.biBitCount = format->byte_count * 8;
b_info->bmiHeader.biXPelsPerMeter = 0;
b_info->bmiHeader.biYPelsPerMeter = 0;
b_info->bmiHeader.biClrUsed = 0;
b_info->bmiHeader.biClrImportant = 0;
/* Get the bit masks */
masks = (DWORD *)b_info->bmiColors;
switch (surface->resource.format->id)
{
case WINED3DFMT_B8G8R8_UNORM:
usage = DIB_RGB_COLORS;
b_info->bmiHeader.biCompression = BI_RGB;
break;
case WINED3DFMT_B5G5R5X1_UNORM:
case WINED3DFMT_B5G5R5A1_UNORM:
case WINED3DFMT_B4G4R4A4_UNORM:
case WINED3DFMT_B4G4R4X4_UNORM:
case WINED3DFMT_B2G3R3_UNORM:
case WINED3DFMT_B2G3R3A8_UNORM:
case WINED3DFMT_R10G10B10A2_UNORM:
case WINED3DFMT_R8G8B8A8_UNORM:
case WINED3DFMT_R8G8B8X8_UNORM:
case WINED3DFMT_B10G10R10A2_UNORM:
case WINED3DFMT_B5G6R5_UNORM:
case WINED3DFMT_R16G16B16A16_UNORM:
usage = 0;
b_info->bmiHeader.biCompression = BI_BITFIELDS;
masks[0] = format->red_mask;
masks[1] = format->green_mask;
masks[2] = format->blue_mask;
break;
default:
/* Don't know palette */
b_info->bmiHeader.biCompression = BI_RGB;
usage = 0;
break;
}
if (!(dc = GetDC(0)))
{
HeapFree(GetProcessHeap(), 0, b_info);
return HRESULT_FROM_WIN32(GetLastError());
}
TRACE("Creating a DIB section with size %dx%dx%d, size=%d.\n",
b_info->bmiHeader.biWidth, b_info->bmiHeader.biHeight,
b_info->bmiHeader.biBitCount, b_info->bmiHeader.biSizeImage);
surface->dib.DIBsection = CreateDIBSection(dc, b_info, usage, &surface->dib.bitmap_data, 0, 0);
ReleaseDC(0, dc);
if (!surface->dib.DIBsection)
{
ERR("Failed to create DIB section.\n");
HeapFree(GetProcessHeap(), 0, b_info);
return HRESULT_FROM_WIN32(GetLastError());
}
TRACE("DIBSection at %p.\n", surface->dib.bitmap_data);
/* Copy the existing surface to the dib section. */
if (surface->resource.allocatedMemory)
{
memcpy(surface->dib.bitmap_data, surface->resource.allocatedMemory,
surface->resource.height * wined3d_surface_get_pitch(surface));
}
else
{
/* This is to make maps read the GL texture although memory is allocated. */
surface->flags &= ~SFLAG_INSYSMEM;
}
surface->dib.bitmap_size = b_info->bmiHeader.biSizeImage;
HeapFree(GetProcessHeap(), 0, b_info);
/* Now allocate a DC. */
surface->hDC = CreateCompatibleDC(0);
surface->dib.holdbitmap = SelectObject(surface->hDC, surface->dib.DIBsection);
TRACE("Using wined3d palette %p.\n", surface->palette);
SelectPalette(surface->hDC, surface->palette ? surface->palette->hpal : 0, FALSE);
surface->flags |= SFLAG_DIBSECTION;
HeapFree(GetProcessHeap(), 0, surface->resource.heapMemory);
surface->resource.heapMemory = NULL;
return WINED3D_OK;
}
static void surface_prepare_system_memory(struct wined3d_surface *surface)
{
IWineD3DDeviceImpl *device = surface->resource.device;
const struct wined3d_gl_info *gl_info = &device->adapter->gl_info;
TRACE("surface %p.\n", surface);
/* Performance optimization: Count how often a surface is locked, if it is
* locked regularly do not throw away the system memory copy. This avoids
* the need to download the surface from OpenGL all the time. The surface
* is still downloaded if the OpenGL texture is changed. */
if (!(surface->flags & SFLAG_DYNLOCK))
{
if (++surface->lockCount > MAXLOCKCOUNT)
{
TRACE("Surface is locked regularly, not freeing the system memory copy any more.\n");
surface->flags |= SFLAG_DYNLOCK;
}
}
/* Create a PBO for dynamically locked surfaces but don't do it for
* converted or NPOT surfaces. Also don't create a PBO for systemmem
* surfaces. */
if (gl_info->supported[ARB_PIXEL_BUFFER_OBJECT] && (surface->flags & SFLAG_DYNLOCK)
&& !(surface->flags & (SFLAG_PBO | SFLAG_CONVERTED | SFLAG_NONPOW2))
&& (surface->resource.pool != WINED3DPOOL_SYSTEMMEM))
{
struct wined3d_context *context;
GLenum error;
context = context_acquire(device, NULL);
ENTER_GL();
GL_EXTCALL(glGenBuffersARB(1, &surface->pbo));
error = glGetError();
if (!surface->pbo || error != GL_NO_ERROR)
ERR("Failed to create a PBO with error %s (%#x).\n", debug_glerror(error), error);
TRACE("Binding PBO %u.\n", surface->pbo);
GL_EXTCALL(glBindBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB, surface->pbo));
checkGLcall("glBindBufferARB");
GL_EXTCALL(glBufferDataARB(GL_PIXEL_UNPACK_BUFFER_ARB, surface->resource.size + 4,
surface->resource.allocatedMemory, GL_STREAM_DRAW_ARB));
checkGLcall("glBufferDataARB");
GL_EXTCALL(glBindBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB, 0));
checkGLcall("glBindBufferARB");
/* We don't need the system memory anymore and we can't even use it for PBOs. */
if (!(surface->flags & SFLAG_CLIENT))
{
HeapFree(GetProcessHeap(), 0, surface->resource.heapMemory);
surface->resource.heapMemory = NULL;
}
surface->resource.allocatedMemory = NULL;
surface->flags |= SFLAG_PBO;
LEAVE_GL();
context_release(context);
}
else if (!(surface->resource.allocatedMemory || surface->flags & SFLAG_PBO))
{
/* Whatever surface we have, make sure that there is memory allocated
* for the downloaded copy, or a PBO to map. */
if (!surface->resource.heapMemory)
surface->resource.heapMemory = HeapAlloc(GetProcessHeap(), 0, surface->resource.size + RESOURCE_ALIGNMENT);
surface->resource.allocatedMemory = (BYTE *)(((ULONG_PTR)surface->resource.heapMemory
+ (RESOURCE_ALIGNMENT - 1)) & ~(RESOURCE_ALIGNMENT - 1));
if (surface->flags & SFLAG_INSYSMEM)
ERR("Surface without memory or PBO has SFLAG_INSYSMEM set.\n");
}
}
static void surface_evict_sysmem(struct wined3d_surface *surface)
{
if (surface->flags & SFLAG_DONOTFREE)
return;
HeapFree(GetProcessHeap(), 0, surface->resource.heapMemory);
surface->resource.allocatedMemory = NULL;
surface->resource.heapMemory = NULL;
surface_modify_location(surface, SFLAG_INSYSMEM, FALSE);
}
/* Context activation is done by the caller. */
static void surface_bind_and_dirtify(struct wined3d_surface *surface,
const struct wined3d_gl_info *gl_info, BOOL srgb)
{
IWineD3DDeviceImpl *device = surface->resource.device;
DWORD active_sampler;
GLint active_texture;
/* We don't need a specific texture unit, but after binding the texture
* the current unit is dirty. Read the unit back instead of switching to
* 0, this avoids messing around with the state manager's GL states. The
* current texture unit should always be a valid one.
*
* To be more specific, this is tricky because we can implicitly be
* called from sampler() in state.c. This means we can't touch anything
* other than whatever happens to be the currently active texture, or we
* would risk marking already applied sampler states dirty again.
*
* TODO: Track the current active texture per GL context instead of using
* glGet(). */
ENTER_GL();
glGetIntegerv(GL_ACTIVE_TEXTURE, &active_texture);
LEAVE_GL();
active_sampler = device->rev_tex_unit_map[active_texture - GL_TEXTURE0_ARB];
if (active_sampler != WINED3D_UNMAPPED_STAGE)
{
IWineD3DDeviceImpl_MarkStateDirty(device, STATE_SAMPLER(active_sampler));
}
surface_bind(surface, gl_info, srgb);
}
static void surface_force_reload(struct wined3d_surface *surface)
{
surface->flags &= ~(SFLAG_ALLOCATED | SFLAG_SRGBALLOCATED);
}
static void surface_release_client_storage(struct wined3d_surface *surface)
{
struct wined3d_context *context = context_acquire(surface->resource.device, NULL);
ENTER_GL();
glPixelStorei(GL_UNPACK_CLIENT_STORAGE_APPLE, GL_FALSE);
if (surface->texture_name)
{
surface_bind_and_dirtify(surface, context->gl_info, FALSE);
glTexImage2D(surface->texture_target, surface->texture_level,
GL_RGB, 1, 1, 0, GL_RGB, GL_UNSIGNED_BYTE, NULL);
}
if (surface->texture_name_srgb)
{
surface_bind_and_dirtify(surface, context->gl_info, TRUE);
glTexImage2D(surface->texture_target, surface->texture_level,
GL_RGB, 1, 1, 0, GL_RGB, GL_UNSIGNED_BYTE, NULL);
}
glPixelStorei(GL_UNPACK_CLIENT_STORAGE_APPLE, GL_TRUE);
LEAVE_GL();
context_release(context);
surface_modify_location(surface, SFLAG_INSRGBTEX, FALSE);
surface_modify_location(surface, SFLAG_INTEXTURE, FALSE);
surface_force_reload(surface);
}
static HRESULT surface_private_setup(struct wined3d_surface *surface)
{
/* TODO: Check against the maximum texture sizes supported by the video card. */
const struct wined3d_gl_info *gl_info = &surface->resource.device->adapter->gl_info;
unsigned int pow2Width, pow2Height;
TRACE("surface %p.\n", surface);
surface->texture_name = 0;
surface->texture_target = GL_TEXTURE_2D;
/* Non-power2 support */
if (gl_info->supported[ARB_TEXTURE_NON_POWER_OF_TWO] || gl_info->supported[WINED3D_GL_NORMALIZED_TEXRECT])
{
pow2Width = surface->resource.width;
pow2Height = surface->resource.height;
}
else
{
/* Find the nearest pow2 match */
pow2Width = pow2Height = 1;
while (pow2Width < surface->resource.width)
pow2Width <<= 1;
while (pow2Height < surface->resource.height)
pow2Height <<= 1;
}
surface->pow2Width = pow2Width;
surface->pow2Height = pow2Height;
if (pow2Width > surface->resource.width || pow2Height > surface->resource.height)
{
/* TODO: Add support for non power two compressed textures. */
if (surface->resource.format->flags & WINED3DFMT_FLAG_COMPRESSED)
{
FIXME("(%p) Compressed non-power-two textures are not supported w(%d) h(%d)\n",
surface, surface->resource.width, surface->resource.height);
return WINED3DERR_NOTAVAILABLE;
}
}
if (pow2Width != surface->resource.width
|| pow2Height != surface->resource.height)
{
surface->flags |= SFLAG_NONPOW2;
}
if ((surface->pow2Width > gl_info->limits.texture_size || surface->pow2Height > gl_info->limits.texture_size)
&& !(surface->resource.usage & (WINED3DUSAGE_RENDERTARGET | WINED3DUSAGE_DEPTHSTENCIL)))
{
/* One of three options:
* 1: Do the same as we do with NPOT and scale the texture, (any
* texture ops would require the texture to be scaled which is
* potentially slow)
* 2: Set the texture to the maximum size (bad idea).
* 3: WARN and return WINED3DERR_NOTAVAILABLE;
* 4: Create the surface, but allow it to be used only for DirectDraw
* Blts. Some apps (e.g. Swat 3) create textures with a Height of
* 16 and a Width > 3000 and blt 16x16 letter areas from them to
* the render target. */
if (surface->resource.pool == WINED3DPOOL_DEFAULT || surface->resource.pool == WINED3DPOOL_MANAGED)
{
WARN("Unable to allocate a surface which exceeds the maximum OpenGL texture size.\n");
return WINED3DERR_NOTAVAILABLE;
}
/* We should never use this surface in combination with OpenGL! */
TRACE("Creating an oversized surface: %ux%u.\n",
surface->pow2Width, surface->pow2Height);
}
else
{
/* Don't use ARB_TEXTURE_RECTANGLE in case the surface format is P8
* and EXT_PALETTED_TEXTURE is used in combination with texture
* uploads (RTL_READTEX/RTL_TEXTEX). The reason is that
* EXT_PALETTED_TEXTURE doesn't work in combination with
* ARB_TEXTURE_RECTANGLE. */
if (surface->flags & SFLAG_NONPOW2 && gl_info->supported[ARB_TEXTURE_RECTANGLE]
&& !(surface->resource.format->id == WINED3DFMT_P8_UINT
&& gl_info->supported[EXT_PALETTED_TEXTURE]
&& wined3d_settings.rendertargetlock_mode == RTL_READTEX))
{
surface->texture_target = GL_TEXTURE_RECTANGLE_ARB;
surface->pow2Width = surface->resource.width;
surface->pow2Height = surface->resource.height;
surface->flags &= ~(SFLAG_NONPOW2 | SFLAG_NORMCOORD);
}
}
switch (wined3d_settings.offscreen_rendering_mode)
{
case ORM_FBO:
surface->get_drawable_size = get_drawable_size_fbo;
break;
case ORM_BACKBUFFER:
surface->get_drawable_size = get_drawable_size_backbuffer;
break;
default:
ERR("Unhandled offscreen rendering mode %#x.\n", wined3d_settings.offscreen_rendering_mode);
return WINED3DERR_INVALIDCALL;
}
surface->flags |= SFLAG_INSYSMEM;
return WINED3D_OK;
}
static void surface_realize_palette(struct wined3d_surface *surface)
{
struct wined3d_palette *palette = surface->palette;
TRACE("surface %p.\n", surface);
if (!palette) return;
if (surface->resource.format->id == WINED3DFMT_P8_UINT
|| surface->resource.format->id == WINED3DFMT_P8_UINT_A8_UNORM)
{
if (surface->resource.usage & WINED3DUSAGE_RENDERTARGET)
{
/* Make sure the texture is up to date. This call doesn't do
* anything if the texture is already up to date. */
surface_load_location(surface, SFLAG_INTEXTURE, NULL);
/* We want to force a palette refresh, so mark the drawable as not being up to date */
if (!surface_is_offscreen(surface))
surface_modify_location(surface, SFLAG_INDRAWABLE, FALSE);
}
else
{
if (!(surface->flags & SFLAG_INSYSMEM))
{
TRACE("Palette changed with surface that does not have an up to date system memory copy.\n");
surface_load_location(surface, SFLAG_INSYSMEM, NULL);
}
surface_modify_location(surface, SFLAG_INSYSMEM, TRUE);
}
}
if (surface->flags & SFLAG_DIBSECTION)
{
RGBQUAD col[256];
unsigned int i;
TRACE("Updating the DC's palette.\n");
for (i = 0; i < 256; ++i)
{
col[i].rgbRed = palette->palents[i].peRed;
col[i].rgbGreen = palette->palents[i].peGreen;
col[i].rgbBlue = palette->palents[i].peBlue;
col[i].rgbReserved = 0;
}
SetDIBColorTable(surface->hDC, 0, 256, col);
}
/* Propagate the changes to the drawable when we have a palette. */
if (surface->resource.usage & WINED3DUSAGE_RENDERTARGET)
surface_load_location(surface, SFLAG_INDRAWABLE, NULL);
}
static HRESULT surface_draw_overlay(struct wined3d_surface *surface)
{
HRESULT hr;
/* If there's no destination surface there is nothing to do. */
if (!surface->overlay_dest)
return WINED3D_OK;
/* Blt calls ModifyLocation on the dest surface, which in turn calls
* DrawOverlay to update the overlay. Prevent an endless recursion. */
if (surface->overlay_dest->flags & SFLAG_INOVERLAYDRAW)
return WINED3D_OK;
surface->overlay_dest->flags |= SFLAG_INOVERLAYDRAW;
hr = wined3d_surface_blt(surface->overlay_dest, &surface->overlay_destrect, surface,
&surface->overlay_srcrect, WINEDDBLT_WAIT, NULL, WINED3DTEXF_LINEAR);
surface->overlay_dest->flags &= ~SFLAG_INOVERLAYDRAW;
return hr;
}
static void surface_preload(struct wined3d_surface *surface)
{
TRACE("surface %p.\n", surface);
surface_internal_preload(surface, SRGB_ANY);
}
static void surface_map(struct wined3d_surface *surface, const RECT *rect, DWORD flags)
{
IWineD3DDeviceImpl *device = surface->resource.device;
const RECT *pass_rect = rect;
TRACE("surface %p, rect %s, flags %#x.\n",
surface, wine_dbgstr_rect(rect), flags);
if (flags & WINED3DLOCK_DISCARD)
{
TRACE("WINED3DLOCK_DISCARD flag passed, marking SYSMEM as up to date.\n");
surface_prepare_system_memory(surface);
surface_modify_location(surface, SFLAG_INSYSMEM, TRUE);
}
else
{
/* surface_load_location() does not check if the rectangle specifies
* the full surface. Most callers don't need that, so do it here. */
if (rect && !rect->top && !rect->left
&& rect->right == surface->resource.width
&& rect->bottom == surface->resource.height)
pass_rect = NULL;
if (!(wined3d_settings.rendertargetlock_mode == RTL_DISABLE
&& ((surface->container.type == WINED3D_CONTAINER_SWAPCHAIN) || surface == device->render_targets[0])))
surface_load_location(surface, SFLAG_INSYSMEM, pass_rect);
}
if (surface->flags & SFLAG_PBO)
{
const struct wined3d_gl_info *gl_info;
struct wined3d_context *context;
context = context_acquire(device, NULL);
gl_info = context->gl_info;
ENTER_GL();
GL_EXTCALL(glBindBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB, surface->pbo));
checkGLcall("glBindBufferARB");
/* This shouldn't happen but could occur if some other function
* didn't handle the PBO properly. */
if (surface->resource.allocatedMemory)
ERR("The surface already has PBO memory allocated.\n");
surface->resource.allocatedMemory = GL_EXTCALL(glMapBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB, GL_READ_WRITE_ARB));
checkGLcall("glMapBufferARB");
/* Make sure the PBO isn't set anymore in order not to break non-PBO
* calls. */
GL_EXTCALL(glBindBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB, 0));
checkGLcall("glBindBufferARB");
LEAVE_GL();
context_release(context);
}
if (!(flags & (WINED3DLOCK_NO_DIRTY_UPDATE | WINED3DLOCK_READONLY)))
{
if (!rect)
surface_add_dirty_rect(surface, NULL);
else
{
WINED3DBOX b;
b.Left = rect->left;
b.Top = rect->top;
b.Right = rect->right;
b.Bottom = rect->bottom;
b.Front = 0;
b.Back = 1;
surface_add_dirty_rect(surface, &b);
}
}
}
static void surface_unmap(struct wined3d_surface *surface)
{
IWineD3DDeviceImpl *device = surface->resource.device;
BOOL fullsurface;
TRACE("surface %p.\n", surface);
memset(&surface->lockedRect, 0, sizeof(surface->lockedRect));
if (surface->flags & SFLAG_PBO)
{
const struct wined3d_gl_info *gl_info;
struct wined3d_context *context;
TRACE("Freeing PBO memory.\n");
context = context_acquire(device, NULL);
gl_info = context->gl_info;
ENTER_GL();
GL_EXTCALL(glBindBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB, surface->pbo));
GL_EXTCALL(glUnmapBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB));
GL_EXTCALL(glBindBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB, 0));
checkGLcall("glUnmapBufferARB");
LEAVE_GL();
context_release(context);
surface->resource.allocatedMemory = NULL;
}
TRACE("dirtyfied %u.\n", surface->flags & (SFLAG_INDRAWABLE | SFLAG_INTEXTURE) ? 0 : 1);
if (surface->flags & (SFLAG_INDRAWABLE | SFLAG_INTEXTURE))
{
TRACE("Not dirtified, nothing to do.\n");
goto done;
}
if (surface->container.type == WINED3D_CONTAINER_SWAPCHAIN
|| (device->render_targets && surface == device->render_targets[0]))
{
if (wined3d_settings.rendertargetlock_mode == RTL_DISABLE)
{
static BOOL warned = FALSE;
if (!warned)
{
ERR("The application tries to write to the render target, but render target locking is disabled.\n");
warned = TRUE;
}
goto done;
}
if (!surface->dirtyRect.left && !surface->dirtyRect.top
&& surface->dirtyRect.right == surface->resource.width
&& surface->dirtyRect.bottom == surface->resource.height)
{
fullsurface = TRUE;
}
else
{
/* TODO: Proper partial rectangle tracking. */
fullsurface = FALSE;
surface->flags |= SFLAG_INSYSMEM;
}
surface_load_location(surface, SFLAG_INDRAWABLE, fullsurface ? NULL : &surface->dirtyRect);
/* Partial rectangle tracking is not commonly implemented, it is only
* done for render targets. INSYSMEM was set before to tell
* surface_load_location() where to read the rectangle from.
* Indrawable is set because all modifications from the partial
* sysmem copy are written back to the drawable, thus the surface is
* merged again in the drawable. The sysmem copy is not fully up to
* date because only a subrectangle was read in Map(). */
if (!fullsurface)
{
surface_modify_location(surface, SFLAG_INDRAWABLE, TRUE);
surface_evict_sysmem(surface);
}
surface->dirtyRect.left = surface->resource.width;
surface->dirtyRect.top = surface->resource.height;
surface->dirtyRect.right = 0;
surface->dirtyRect.bottom = 0;
}
else if (surface->resource.format->flags & (WINED3DFMT_FLAG_DEPTH | WINED3DFMT_FLAG_STENCIL))
{
FIXME("Depth / stencil buffer locking is not implemented.\n");
}
done:
/* Overlays have to be redrawn manually after changes with the GL implementation */
if (surface->overlay_dest)
surface->surface_ops->surface_draw_overlay(surface);
}
static HRESULT surface_getdc(struct wined3d_surface *surface)
{
WINED3DLOCKED_RECT lock;
HRESULT hr;
TRACE("surface %p.\n", surface);
/* Create a DIB section if there isn't a dc yet. */
if (!surface->hDC)
{
if (surface->flags & SFLAG_CLIENT)
{
surface_load_location(surface, SFLAG_INSYSMEM, NULL);
surface_release_client_storage(surface);
}
hr = surface_create_dib_section(surface);
if (FAILED(hr))
return WINED3DERR_INVALIDCALL;
/* Use the DIB section from now on if we are not using a PBO. */
if (!(surface->flags & SFLAG_PBO))
surface->resource.allocatedMemory = surface->dib.bitmap_data;
}
/* Map the surface. */
hr = wined3d_surface_map(surface, &lock, NULL, 0);
if (FAILED(hr))
ERR("Map failed, hr %#x.\n", hr);
/* Sync the DIB with the PBO. This can't be done earlier because Map()
* activates the allocatedMemory. */
if (surface->flags & SFLAG_PBO)
memcpy(surface->dib.bitmap_data, surface->resource.allocatedMemory, surface->dib.bitmap_size);
return hr;
}
static HRESULT surface_flip(struct wined3d_surface *surface, struct wined3d_surface *override)
{
TRACE("surface %p, override %p.\n", surface, override);
/* Flipping is only supported on render targets and overlays. */
if (!(surface->resource.usage & (WINED3DUSAGE_RENDERTARGET | WINED3DUSAGE_OVERLAY)))
{
WARN("Tried to flip a non-render target, non-overlay surface.\n");
return WINEDDERR_NOTFLIPPABLE;
}
if (surface->resource.usage & WINED3DUSAGE_OVERLAY)
{
flip_surface(surface, override);
/* Update the overlay if it is visible */
if (surface->overlay_dest)
return surface->surface_ops->surface_draw_overlay(surface);
else
return WINED3D_OK;
}
return WINED3D_OK;
}
static BOOL surface_is_full_rect(struct wined3d_surface *surface, const RECT *r)
{
if ((r->left && r->right) || abs(r->right - r->left) != surface->resource.width)
return FALSE;
if ((r->top && r->bottom) || abs(r->bottom - r->top) != surface->resource.height)
return FALSE;
return TRUE;
}
static void wined3d_surface_depth_blt_fbo(IWineD3DDeviceImpl *device, struct wined3d_surface *src_surface,
const RECT *src_rect, struct wined3d_surface *dst_surface, const RECT *dst_rect)
{
const struct wined3d_gl_info *gl_info;
struct wined3d_context *context;
DWORD src_mask, dst_mask;
GLbitfield gl_mask;
TRACE("device %p, src_surface %p, src_rect %s, dst_surface %p, dst_rect %s.\n",
device, src_surface, wine_dbgstr_rect(src_rect),
dst_surface, wine_dbgstr_rect(dst_rect));
src_mask = src_surface->resource.format->flags & (WINED3DFMT_FLAG_DEPTH | WINED3DFMT_FLAG_STENCIL);
dst_mask = dst_surface->resource.format->flags & (WINED3DFMT_FLAG_DEPTH | WINED3DFMT_FLAG_STENCIL);
if (src_mask != dst_mask)
{
ERR("Incompatible formats %s and %s.\n",
debug_d3dformat(src_surface->resource.format->id),
debug_d3dformat(dst_surface->resource.format->id));
return;
}
if (!src_mask)
{
ERR("Not a depth / stencil format: %s.\n",
debug_d3dformat(src_surface->resource.format->id));
return;
}
gl_mask = 0;
if (src_mask & WINED3DFMT_FLAG_DEPTH)
gl_mask |= GL_DEPTH_BUFFER_BIT;
if (src_mask & WINED3DFMT_FLAG_STENCIL)
gl_mask |= GL_STENCIL_BUFFER_BIT;
/* Make sure the locations are up-to-date. Loading the destination
* surface isn't required if the entire surface is overwritten. */
surface_load_location(src_surface, SFLAG_INTEXTURE, NULL);
if (!surface_is_full_rect(dst_surface, dst_rect))
surface_load_location(dst_surface, SFLAG_INTEXTURE, NULL);
context = context_acquire(device, NULL);
if (!context->valid)
{
context_release(context);
WARN("Invalid context, skipping blit.\n");
return;
}
gl_info = context->gl_info;
ENTER_GL();
context_apply_fbo_state_blit(context, GL_READ_FRAMEBUFFER, NULL, src_surface, SFLAG_INTEXTURE);
glReadBuffer(GL_NONE);
checkGLcall("glReadBuffer()");
context_check_fbo_status(context, GL_READ_FRAMEBUFFER);
context_apply_fbo_state_blit(context, GL_DRAW_FRAMEBUFFER, NULL, dst_surface, SFLAG_INTEXTURE);
context_set_draw_buffer(context, GL_NONE);
context_check_fbo_status(context, GL_DRAW_FRAMEBUFFER);
if (gl_mask & GL_DEPTH_BUFFER_BIT)
{
glDepthMask(GL_TRUE);
IWineD3DDeviceImpl_MarkStateDirty(device, STATE_RENDER(WINED3DRS_ZWRITEENABLE));
}
if (gl_mask & GL_STENCIL_BUFFER_BIT)
{
if (context->gl_info->supported[EXT_STENCIL_TWO_SIDE])
{
glDisable(GL_STENCIL_TEST_TWO_SIDE_EXT);
IWineD3DDeviceImpl_MarkStateDirty(device, STATE_RENDER(WINED3DRS_TWOSIDEDSTENCILMODE));
}
glStencilMask(~0U);
IWineD3DDeviceImpl_MarkStateDirty(device, STATE_RENDER(WINED3DRS_STENCILWRITEMASK));
}
glDisable(GL_SCISSOR_TEST);
IWineD3DDeviceImpl_MarkStateDirty(device, STATE_RENDER(WINED3DRS_SCISSORTESTENABLE));
gl_info->fbo_ops.glBlitFramebuffer(src_rect->left, src_rect->top, src_rect->right, src_rect->bottom,
dst_rect->left, dst_rect->top, dst_rect->right, dst_rect->bottom, gl_mask, GL_NEAREST);
checkGLcall("glBlitFramebuffer()");
LEAVE_GL();
if (wined3d_settings.strict_draw_ordering)
wglFlush(); /* Flush to ensure ordering across contexts. */
context_release(context);
}
static BOOL fbo_blit_supported(const struct wined3d_gl_info *gl_info, enum wined3d_blit_op blit_op,
const RECT *src_rect, DWORD src_usage, WINED3DPOOL src_pool, const struct wined3d_format *src_format,
const RECT *dst_rect, DWORD dst_usage, WINED3DPOOL dst_pool, const struct wined3d_format *dst_format)
{
if ((wined3d_settings.offscreen_rendering_mode != ORM_FBO) || !gl_info->fbo_ops.glBlitFramebuffer)
return FALSE;
/* Source and/or destination need to be on the GL side */
if (src_pool == WINED3DPOOL_SYSTEMMEM || dst_pool == WINED3DPOOL_SYSTEMMEM)
return FALSE;
switch (blit_op)
{
case WINED3D_BLIT_OP_COLOR_BLIT:
if (!((src_format->flags & WINED3DFMT_FLAG_FBO_ATTACHABLE) || (src_usage & WINED3DUSAGE_RENDERTARGET)))
return FALSE;
if (!((dst_format->flags & WINED3DFMT_FLAG_FBO_ATTACHABLE) || (dst_usage & WINED3DUSAGE_RENDERTARGET)))
return FALSE;
break;
case WINED3D_BLIT_OP_DEPTH_BLIT:
if (!(src_format->flags & (WINED3DFMT_FLAG_DEPTH | WINED3DFMT_FLAG_STENCIL)))
return FALSE;
if (!(dst_format->flags & (WINED3DFMT_FLAG_DEPTH | WINED3DFMT_FLAG_STENCIL)))
return FALSE;
break;
default:
return FALSE;
}
if (!(src_format->id == dst_format->id
|| (is_identity_fixup(src_format->color_fixup)
&& is_identity_fixup(dst_format->color_fixup))))
return FALSE;
return TRUE;
}
static BOOL surface_convert_depth_to_float(struct wined3d_surface *surface, DWORD depth, float *float_depth)
{
const struct wined3d_format *format = surface->resource.format;
switch (format->id)
{
case WINED3DFMT_S1_UINT_D15_UNORM:
*float_depth = depth / (float)0x00007fff;
break;
case WINED3DFMT_D16_UNORM:
*float_depth = depth / (float)0x0000ffff;
break;
case WINED3DFMT_D24_UNORM_S8_UINT:
case WINED3DFMT_X8D24_UNORM:
*float_depth = depth / (float)0x00ffffff;
break;
case WINED3DFMT_D32_UNORM:
*float_depth = depth / (float)0xffffffff;
break;
default:
ERR("Unhandled conversion from %s to floating point.\n", debug_d3dformat(format->id));
return FALSE;
}
return TRUE;
}
/* Do not call while under the GL lock. */
static HRESULT wined3d_surface_depth_fill(struct wined3d_surface *surface, const RECT *rect, float depth)
{
const struct wined3d_resource *resource = &surface->resource;
IWineD3DDeviceImpl *device = resource->device;
const struct blit_shader *blitter;
blitter = wined3d_select_blitter(&device->adapter->gl_info, WINED3D_BLIT_OP_DEPTH_FILL,
NULL, 0, 0, NULL, rect, resource->usage, resource->pool, resource->format);
if (!blitter)
{
FIXME("No blitter is capable of performing the requested depth fill operation.\n");
return WINED3DERR_INVALIDCALL;
}
return blitter->depth_fill(device, surface, rect, depth);
}
static HRESULT wined3d_surface_depth_blt(struct wined3d_surface *src_surface, const RECT *src_rect,
struct wined3d_surface *dst_surface, const RECT *dst_rect)
{
IWineD3DDeviceImpl *device = src_surface->resource.device;
if (!fbo_blit_supported(&device->adapter->gl_info, WINED3D_BLIT_OP_DEPTH_BLIT,
src_rect, src_surface->resource.usage, src_surface->resource.pool, src_surface->resource.format,
dst_rect, dst_surface->resource.usage, dst_surface->resource.pool, dst_surface->resource.format))
return WINED3DERR_INVALIDCALL;
wined3d_surface_depth_blt_fbo(device, src_surface, src_rect, dst_surface, dst_rect);
surface_modify_ds_location(dst_surface, SFLAG_DS_OFFSCREEN,
dst_surface->ds_current_size.cx, dst_surface->ds_current_size.cy);
surface_modify_location(dst_surface, SFLAG_INDRAWABLE, TRUE);
return WINED3D_OK;
}
/* Do not call while under the GL lock. */
static HRESULT surface_blt(struct wined3d_surface *dst_surface, const RECT *dst_rect_in,
struct wined3d_surface *src_surface, const RECT *src_rect_in, DWORD flags,
const WINEDDBLTFX *fx, WINED3DTEXTUREFILTERTYPE filter)
{
IWineD3DDeviceImpl *device = dst_surface->resource.device;
DWORD src_ds_flags, dst_ds_flags;
TRACE("dst_surface %p, dst_rect %s, src_surface %p, src_rect %s, flags %#x, fx %p, filter %s.\n",
dst_surface, wine_dbgstr_rect(dst_rect_in), src_surface, wine_dbgstr_rect(src_rect_in),
flags, fx, debug_d3dtexturefiltertype(filter));
TRACE("Usage is %s.\n", debug_d3dusage(dst_surface->resource.usage));
if ((dst_surface->flags & SFLAG_LOCKED) || (src_surface && (src_surface->flags & SFLAG_LOCKED)))
{
WARN(" Surface is busy, returning DDERR_SURFACEBUSY\n");
return WINEDDERR_SURFACEBUSY;
}
dst_ds_flags = dst_surface->resource.format->flags & (WINED3DFMT_FLAG_DEPTH | WINED3DFMT_FLAG_STENCIL);
if (src_surface)
src_ds_flags = src_surface->resource.format->flags & (WINED3DFMT_FLAG_DEPTH | WINED3DFMT_FLAG_STENCIL);
else
src_ds_flags = 0;
if (src_ds_flags || dst_ds_flags)
{
if (flags & WINEDDBLT_DEPTHFILL)
{
float depth;
RECT rect;
TRACE("Depth fill.\n");
surface_get_rect(dst_surface, dst_rect_in, &rect);
if (!surface_convert_depth_to_float(dst_surface, fx->u5.dwFillDepth, &depth))
return WINED3DERR_INVALIDCALL;
if (SUCCEEDED(wined3d_surface_depth_fill(dst_surface, &rect, depth)))
return WINED3D_OK;
}
else
{
RECT src_rect, dst_rect;
/* Accessing depth / stencil surfaces is supposed to fail while in
* a scene, except for fills, which seem to work. */
if (device->inScene)
{
WARN("Rejecting depth / stencil access while in scene.\n");
return WINED3DERR_INVALIDCALL;
}
if (src_ds_flags != dst_ds_flags)
{
WARN("Rejecting depth / stencil blit between incompatible formats.\n");
return WINED3DERR_INVALIDCALL;
}
if (src_rect_in && (src_rect_in->top || src_rect_in->left
|| src_rect_in->bottom != src_surface->resource.height
|| src_rect_in->right != src_surface->resource.width))
{
WARN("Rejecting depth / stencil blit with invalid source rect %s.\n",
wine_dbgstr_rect(src_rect_in));
return WINED3DERR_INVALIDCALL;
}
if (dst_rect_in && (dst_rect_in->top || dst_rect_in->left
|| dst_rect_in->bottom != dst_surface->resource.height
|| dst_rect_in->right != dst_surface->resource.width))
{
WARN("Rejecting depth / stencil blit with invalid destination rect %s.\n",
wine_dbgstr_rect(src_rect_in));
return WINED3DERR_INVALIDCALL;
}
if (src_surface->resource.height != dst_surface->resource.height
|| src_surface->resource.width != dst_surface->resource.width)
{
WARN("Rejecting depth / stencil blit with mismatched surface sizes.\n");
return WINED3DERR_INVALIDCALL;
}
surface_get_rect(src_surface, src_rect_in, &src_rect);
surface_get_rect(dst_surface, dst_rect_in, &dst_rect);
if (SUCCEEDED(wined3d_surface_depth_blt(src_surface, &src_rect, dst_surface, &dst_rect)))
return WINED3D_OK;
}
}
/* Special cases for render targets. */
if ((dst_surface->resource.usage & WINED3DUSAGE_RENDERTARGET)
|| (src_surface && (src_surface->resource.usage & WINED3DUSAGE_RENDERTARGET)))
{
if (SUCCEEDED(IWineD3DSurfaceImpl_BltOverride(dst_surface, dst_rect_in,
src_surface, src_rect_in, flags, fx, filter)))
return WINED3D_OK;
}
/* For the rest call the X11 surface implementation. For render targets
* this should be implemented OpenGL accelerated in BltOverride, other
* blits are rather rare. */
return surface_cpu_blt(dst_surface, dst_rect_in, src_surface, src_rect_in, flags, fx, filter);
}
/* Do not call while under the GL lock. */
static HRESULT surface_bltfast(struct wined3d_surface *dst_surface, DWORD dst_x, DWORD dst_y,
struct wined3d_surface *src_surface, const RECT *src_rect_in, DWORD trans)
{
IWineD3DDeviceImpl *device = dst_surface->resource.device;
TRACE("dst_surface %p, dst_x %u, dst_y %u, src_surface %p, src_rect %s, flags %#x.\n",
dst_surface, dst_x, dst_y, src_surface, wine_dbgstr_rect(src_rect_in), trans);
if ((dst_surface->flags & SFLAG_LOCKED) || (src_surface->flags & SFLAG_LOCKED))
{
WARN("Surface is busy, returning WINEDDERR_SURFACEBUSY.\n");
return WINEDDERR_SURFACEBUSY;
}
if (device->inScene && (dst_surface == device->depth_stencil || src_surface == device->depth_stencil))
{
WARN("Attempt to access the depth / stencil surface while in a scene.\n");
return WINED3DERR_INVALIDCALL;
}
/* Special cases for RenderTargets */
if ((dst_surface->resource.usage & WINED3DUSAGE_RENDERTARGET)
|| (src_surface->resource.usage & WINED3DUSAGE_RENDERTARGET))
{
RECT src_rect, dst_rect;
DWORD flags = 0;
surface_get_rect(src_surface, src_rect_in, &src_rect);
dst_rect.left = dst_x;
dst_rect.top = dst_y;
dst_rect.right = dst_x + src_rect.right - src_rect.left;
dst_rect.bottom = dst_y + src_rect.bottom - src_rect.top;
/* Convert BltFast flags into Blt ones because BltOverride is called
* from Blt as well. */
if (trans & WINEDDBLTFAST_SRCCOLORKEY)
flags |= WINEDDBLT_KEYSRC;
if (trans & WINEDDBLTFAST_DESTCOLORKEY)
flags |= WINEDDBLT_KEYDEST;
if (trans & WINEDDBLTFAST_WAIT)
flags |= WINEDDBLT_WAIT;
if (trans & WINEDDBLTFAST_DONOTWAIT)
flags |= WINEDDBLT_DONOTWAIT;
if (SUCCEEDED(IWineD3DSurfaceImpl_BltOverride(dst_surface,
&dst_rect, src_surface, &src_rect, flags, NULL, WINED3DTEXF_POINT)))
return WINED3D_OK;
}
return surface_cpu_bltfast(dst_surface, dst_x, dst_y, src_surface, src_rect_in, trans);
}
static HRESULT surface_set_mem(struct wined3d_surface *surface, void *mem)
{
TRACE("surface %p, mem %p.\n", surface, mem);
if (mem && mem != surface->resource.allocatedMemory)
{
void *release = NULL;
/* Do I have to copy the old surface content? */
if (surface->flags & SFLAG_DIBSECTION)
{
SelectObject(surface->hDC, surface->dib.holdbitmap);
DeleteDC(surface->hDC);
/* Release the DIB section. */
DeleteObject(surface->dib.DIBsection);
surface->dib.bitmap_data = NULL;
surface->resource.allocatedMemory = NULL;
surface->hDC = NULL;
surface->flags &= ~SFLAG_DIBSECTION;
}
else if (!(surface->flags & SFLAG_USERPTR))
{
release = surface->resource.heapMemory;
surface->resource.heapMemory = NULL;
}
surface->resource.allocatedMemory = mem;
surface->flags |= SFLAG_USERPTR;
/* Now the surface memory is most up do date. Invalidate drawable and texture. */
surface_modify_location(surface, SFLAG_INSYSMEM, TRUE);
/* For client textures OpenGL has to be notified. */
if (surface->flags & SFLAG_CLIENT)
surface_release_client_storage(surface);
/* Now free the old memory if any. */
HeapFree(GetProcessHeap(), 0, release);
}
else if (surface->flags & SFLAG_USERPTR)
{
/* Map and GetDC will re-create the dib section and allocated memory. */
surface->resource.allocatedMemory = NULL;
/* HeapMemory should be NULL already. */
if (surface->resource.heapMemory)
ERR("User pointer surface has heap memory allocated.\n");
surface->flags &= ~(SFLAG_USERPTR | SFLAG_INSYSMEM);
if (surface->flags & SFLAG_CLIENT)
surface_release_client_storage(surface);
surface_prepare_system_memory(surface);
surface_modify_location(surface, SFLAG_INSYSMEM, TRUE);
}
return WINED3D_OK;
}
/* Context activation is done by the caller. */
static void surface_remove_pbo(struct wined3d_surface *surface, const struct wined3d_gl_info *gl_info)
{
if (!surface->resource.heapMemory)
{
surface->resource.heapMemory = HeapAlloc(GetProcessHeap(), 0, surface->resource.size + RESOURCE_ALIGNMENT);
surface->resource.allocatedMemory = (BYTE *)(((ULONG_PTR)surface->resource.heapMemory
+ (RESOURCE_ALIGNMENT - 1)) & ~(RESOURCE_ALIGNMENT - 1));
}
ENTER_GL();
GL_EXTCALL(glBindBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB, surface->pbo));
checkGLcall("glBindBufferARB(GL_PIXEL_UNPACK_BUFFER, surface->pbo)");
GL_EXTCALL(glGetBufferSubDataARB(GL_PIXEL_UNPACK_BUFFER_ARB, 0,
surface->resource.size, surface->resource.allocatedMemory));
checkGLcall("glGetBufferSubDataARB");
GL_EXTCALL(glDeleteBuffersARB(1, &surface->pbo));
checkGLcall("glDeleteBuffersARB");
LEAVE_GL();
surface->pbo = 0;
surface->flags &= ~SFLAG_PBO;
}
/* Do not call while under the GL lock. */
static void surface_unload(struct wined3d_resource *resource)
{
struct wined3d_surface *surface = surface_from_resource(resource);
IWineD3DDeviceImpl *device = resource->device;
const struct wined3d_gl_info *gl_info;
renderbuffer_entry_t *entry, *entry2;
struct wined3d_context *context;
TRACE("surface %p.\n", surface);
if (resource->pool == WINED3DPOOL_DEFAULT)
{
/* Default pool resources are supposed to be destroyed before Reset is called.
* Implicit resources stay however. So this means we have an implicit render target
* or depth stencil. The content may be destroyed, but we still have to tear down
* opengl resources, so we cannot leave early.
*
* Put the surfaces into sysmem, and reset the content. The D3D content is undefined,
* but we can't set the sysmem INDRAWABLE because when we're rendering the swapchain
* or the depth stencil into an FBO the texture or render buffer will be removed
* and all flags get lost
*/
surface_init_sysmem(surface);
}
else
{
/* Load the surface into system memory */
surface_load_location(surface, SFLAG_INSYSMEM, NULL);
surface_modify_location(surface, SFLAG_INDRAWABLE, FALSE);
}
surface_modify_location(surface, SFLAG_INTEXTURE, FALSE);
surface_modify_location(surface, SFLAG_INSRGBTEX, FALSE);
surface->flags &= ~(SFLAG_ALLOCATED | SFLAG_SRGBALLOCATED);
context = context_acquire(device, NULL);
gl_info = context->gl_info;
/* Destroy PBOs, but load them into real sysmem before */
if (surface->flags & SFLAG_PBO)
surface_remove_pbo(surface, gl_info);
/* Destroy fbo render buffers. This is needed for implicit render targets, for
* all application-created targets the application has to release the surface
* before calling _Reset
*/
LIST_FOR_EACH_ENTRY_SAFE(entry, entry2, &surface->renderbuffers, renderbuffer_entry_t, entry)
{
ENTER_GL();
gl_info->fbo_ops.glDeleteRenderbuffers(1, &entry->id);
LEAVE_GL();
list_remove(&entry->entry);
HeapFree(GetProcessHeap(), 0, entry);
}
list_init(&surface->renderbuffers);
surface->current_renderbuffer = NULL;
/* If we're in a texture, the texture name belongs to the texture.
* Otherwise, destroy it. */
if (surface->container.type != WINED3D_CONTAINER_TEXTURE)
{
ENTER_GL();
glDeleteTextures(1, &surface->texture_name);
surface->texture_name = 0;
glDeleteTextures(1, &surface->texture_name_srgb);
surface->texture_name_srgb = 0;
LEAVE_GL();
}
context_release(context);
resource_unload(resource);
}
static const struct wined3d_resource_ops surface_resource_ops =
{
surface_unload,
};
static const struct wined3d_surface_ops surface_ops =
{
surface_private_setup,
surface_cleanup,
surface_realize_palette,
surface_draw_overlay,
surface_preload,
surface_map,
surface_unmap,
surface_getdc,
surface_flip,
surface_blt,
surface_bltfast,
surface_set_mem,
};
/*****************************************************************************
* Initializes the GDI surface, aka creates the DIB section we render to
* The DIB section creation is done by calling GetDC, which will create the
* section and releasing the dc to allow the app to use it. The dib section
* will stay until the surface is released
*
* GDI surfaces do not need to be a power of 2 in size, so the pow2 sizes
* are set to the real sizes to save memory. The NONPOW2 flag is unset to
* avoid confusion in the shared surface code.
*
* Returns:
* WINED3D_OK on success
* The return values of called methods on failure
*
*****************************************************************************/
static HRESULT gdi_surface_private_setup(struct wined3d_surface *surface)
{
HRESULT hr;
TRACE("surface %p.\n", surface);
if (surface->resource.usage & WINED3DUSAGE_OVERLAY)
{
ERR("Overlays not yet supported by GDI surfaces.\n");
return WINED3DERR_INVALIDCALL;
}
/* Sysmem textures have memory already allocated - release it,
* this avoids an unnecessary memcpy. */
hr = surface_create_dib_section(surface);
if (SUCCEEDED(hr))
{
HeapFree(GetProcessHeap(), 0, surface->resource.heapMemory);
surface->resource.heapMemory = NULL;
surface->resource.allocatedMemory = surface->dib.bitmap_data;
}
/* We don't mind the nonpow2 stuff in GDI. */
surface->pow2Width = surface->resource.width;
surface->pow2Height = surface->resource.height;
return WINED3D_OK;
}
static void surface_gdi_cleanup(struct wined3d_surface *surface)
{
TRACE("surface %p.\n", surface);
if (surface->flags & SFLAG_DIBSECTION)
{
/* Release the DC. */
SelectObject(surface->hDC, surface->dib.holdbitmap);
DeleteDC(surface->hDC);
/* Release the DIB section. */
DeleteObject(surface->dib.DIBsection);
surface->dib.bitmap_data = NULL;
surface->resource.allocatedMemory = NULL;
}
if (surface->flags & SFLAG_USERPTR)
wined3d_surface_set_mem(surface, NULL);
if (surface->overlay_dest)
list_remove(&surface->overlay_entry);
HeapFree(GetProcessHeap(), 0, surface->palette9);
resource_cleanup(&surface->resource);
}
static void gdi_surface_realize_palette(struct wined3d_surface *surface)
{
struct wined3d_palette *palette = surface->palette;
TRACE("surface %p.\n", surface);
if (!palette) return;
if (surface->flags & SFLAG_DIBSECTION)
{
RGBQUAD col[256];
unsigned int i;
TRACE("Updating the DC's palette.\n");
for (i = 0; i < 256; ++i)
{
col[i].rgbRed = palette->palents[i].peRed;
col[i].rgbGreen = palette->palents[i].peGreen;
col[i].rgbBlue = palette->palents[i].peBlue;
col[i].rgbReserved = 0;
}
SetDIBColorTable(surface->hDC, 0, 256, col);
}
/* Update the image because of the palette change. Some games like e.g.
* Red Alert call SetEntries a lot to implement fading. */
/* Tell the swapchain to update the screen. */
if (surface->container.type == WINED3D_CONTAINER_SWAPCHAIN)
{
struct wined3d_swapchain *swapchain = surface->container.u.swapchain;
if (surface == swapchain->front_buffer)
{
x11_copy_to_screen(swapchain, NULL);
}
}
}
static HRESULT gdi_surface_draw_overlay(struct wined3d_surface *surface)
{
FIXME("GDI surfaces can't draw overlays yet.\n");
return E_FAIL;
}
static void gdi_surface_preload(struct wined3d_surface *surface)
{
TRACE("surface %p.\n", surface);
ERR("Preloading GDI surfaces is not supported.\n");
}
static void gdi_surface_map(struct wined3d_surface *surface, const RECT *rect, DWORD flags)
{
TRACE("surface %p, rect %s, flags %#x.\n",
surface, wine_dbgstr_rect(rect), flags);
if (!surface->resource.allocatedMemory)
{
/* This happens on gdi surfaces if the application set a user pointer
* and resets it. Recreate the DIB section. */
surface_create_dib_section(surface);
surface->resource.allocatedMemory = surface->dib.bitmap_data;
}
}
static void gdi_surface_unmap(struct wined3d_surface *surface)
{
TRACE("surface %p.\n", surface);
/* Tell the swapchain to update the screen. */
if (surface->container.type == WINED3D_CONTAINER_SWAPCHAIN)
{
struct wined3d_swapchain *swapchain = surface->container.u.swapchain;
if (surface == swapchain->front_buffer)
{
x11_copy_to_screen(swapchain, &surface->lockedRect);
}
}
memset(&surface->lockedRect, 0, sizeof(RECT));
}
static HRESULT gdi_surface_getdc(struct wined3d_surface *surface)
{
WINED3DLOCKED_RECT lock;
HRESULT hr;
TRACE("surface %p.\n", surface);
/* Should have a DIB section already. */
if (!(surface->flags & SFLAG_DIBSECTION))
{
WARN("DC not supported on this surface\n");
return WINED3DERR_INVALIDCALL;
}
/* Map the surface. */
hr = wined3d_surface_map(surface, &lock, NULL, 0);
if (FAILED(hr))
ERR("Map failed, hr %#x.\n", hr);
return hr;
}
static HRESULT gdi_surface_flip(struct wined3d_surface *surface, struct wined3d_surface *override)
{
TRACE("surface %p, override %p.\n", surface, override);
return WINED3D_OK;
}
static HRESULT gdi_surface_blt(struct wined3d_surface *dst_surface, const RECT *dst_rect,
struct wined3d_surface *src_surface, const RECT *src_rect, DWORD flags,
const WINEDDBLTFX *fx, WINED3DTEXTUREFILTERTYPE filter)
{
TRACE("dst_surface %p, dst_rect %s, src_surface %p, src_rect %s, flags %#x, fx %p, filter %s.\n",
dst_surface, wine_dbgstr_rect(dst_rect), src_surface, wine_dbgstr_rect(src_rect),
flags, fx, debug_d3dtexturefiltertype(filter));
return surface_cpu_blt(dst_surface, dst_rect, src_surface, src_rect, flags, fx, filter);
}
static HRESULT gdi_surface_bltfast(struct wined3d_surface *dst_surface, DWORD dst_x, DWORD dst_y,
struct wined3d_surface *src_surface, const RECT *src_rect, DWORD trans)
{
TRACE("dst_surface %p, dst_x %u, dst_y %u, src_surface %p, src_rect %s, flags %#x.\n",
dst_surface, dst_x, dst_y, src_surface, wine_dbgstr_rect(src_rect), trans);
return surface_cpu_bltfast(dst_surface, dst_x, dst_y, src_surface, src_rect, trans);
}
static HRESULT gdi_surface_set_mem(struct wined3d_surface *surface, void *mem)
{
TRACE("surface %p, mem %p.\n", surface, mem);
/* Render targets depend on their hdc, and we can't create an hdc on a user pointer. */
if (surface->resource.usage & WINED3DUSAGE_RENDERTARGET)
{
ERR("Not supported on render targets.\n");
return WINED3DERR_INVALIDCALL;
}
if (mem && mem != surface->resource.allocatedMemory)
{
void *release = NULL;
/* Do I have to copy the old surface content? */
if (surface->flags & SFLAG_DIBSECTION)
{
SelectObject(surface->hDC, surface->dib.holdbitmap);
DeleteDC(surface->hDC);
/* Release the DIB section. */
DeleteObject(surface->dib.DIBsection);
surface->dib.bitmap_data = NULL;
surface->resource.allocatedMemory = NULL;
surface->hDC = NULL;
surface->flags &= ~SFLAG_DIBSECTION;
}
else if (!(surface->flags & SFLAG_USERPTR))
{
release = surface->resource.allocatedMemory;
}
surface->resource.allocatedMemory = mem;
surface->flags |= SFLAG_USERPTR | SFLAG_INSYSMEM;
/* Now free the old memory, if any. */
HeapFree(GetProcessHeap(), 0, release);
}
else if (surface->flags & SFLAG_USERPTR)
{
/* Map() and GetDC() will re-create the dib section and allocated memory. */
surface->resource.allocatedMemory = NULL;
surface->flags &= ~SFLAG_USERPTR;
}
return WINED3D_OK;
}
static const struct wined3d_surface_ops gdi_surface_ops =
{
gdi_surface_private_setup,
surface_gdi_cleanup,
gdi_surface_realize_palette,
gdi_surface_draw_overlay,
gdi_surface_preload,
gdi_surface_map,
gdi_surface_unmap,
gdi_surface_getdc,
gdi_surface_flip,
gdi_surface_blt,
gdi_surface_bltfast,
gdi_surface_set_mem,
};
void surface_set_texture_name(struct wined3d_surface *surface, GLuint new_name, BOOL srgb)
{
GLuint *name;
DWORD flag;
TRACE("surface %p, new_name %u, srgb %#x.\n", surface, new_name, srgb);
if(srgb)
{
name = &surface->texture_name_srgb;
flag = SFLAG_INSRGBTEX;
}
else
{
name = &surface->texture_name;
flag = SFLAG_INTEXTURE;
}
if (!*name && new_name)
{
/* FIXME: We shouldn't need to remove SFLAG_INTEXTURE if the
* surface has no texture name yet. See if we can get rid of this. */
if (surface->flags & flag)
ERR("Surface has %s set, but no texture name.\n", debug_surflocation(flag));
surface_modify_location(surface, flag, FALSE);
}
*name = new_name;
surface_force_reload(surface);
}
void surface_set_texture_target(struct wined3d_surface *surface, GLenum target)
{
TRACE("surface %p, target %#x.\n", surface, target);
if (surface->texture_target != target)
{
if (target == GL_TEXTURE_RECTANGLE_ARB)
{
surface->flags &= ~SFLAG_NORMCOORD;
}
else if (surface->texture_target == GL_TEXTURE_RECTANGLE_ARB)
{
surface->flags |= SFLAG_NORMCOORD;
}
}
surface->texture_target = target;
surface_force_reload(surface);
}
/* Context activation is done by the caller. */
void surface_bind(struct wined3d_surface *surface, const struct wined3d_gl_info *gl_info, BOOL srgb)
{
TRACE("surface %p, gl_info %p, srgb %#x.\n", surface, gl_info, srgb);
if (surface->container.type == WINED3D_CONTAINER_TEXTURE)
{
struct wined3d_texture *texture = surface->container.u.texture;
TRACE("Passing to container (%p).\n", texture);
texture->texture_ops->texture_bind(texture, gl_info, srgb);
}
else
{
if (surface->texture_level)
{
ERR("Standalone surface %p is non-zero texture level %u.\n",
surface, surface->texture_level);
}
if (srgb)
ERR("Trying to bind standalone surface %p as sRGB.\n", surface);
ENTER_GL();
if (!surface->texture_name)
{
glGenTextures(1, &surface->texture_name);
checkGLcall("glGenTextures");
TRACE("Surface %p given name %u.\n", surface, surface->texture_name);
glBindTexture(surface->texture_target, surface->texture_name);
checkGLcall("glBindTexture");
glTexParameteri(surface->texture_target, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(surface->texture_target, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
glTexParameteri(surface->texture_target, GL_TEXTURE_WRAP_R, GL_CLAMP_TO_EDGE);
glTexParameteri(surface->texture_target, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
glTexParameteri(surface->texture_target, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
checkGLcall("glTexParameteri");
}
else
{
glBindTexture(surface->texture_target, surface->texture_name);
checkGLcall("glBindTexture");
}
LEAVE_GL();
}
}
/* This function checks if the primary render target uses the 8bit paletted format. */
static BOOL primary_render_target_is_p8(IWineD3DDeviceImpl *device)
{
if (device->render_targets && device->render_targets[0])
{
struct wined3d_surface *render_target = device->render_targets[0];
if ((render_target->resource.usage & WINED3DUSAGE_RENDERTARGET)
&& (render_target->resource.format->id == WINED3DFMT_P8_UINT))
return TRUE;
}
return FALSE;
}
/* This call just downloads data, the caller is responsible for binding the
* correct texture. */
/* Context activation is done by the caller. */
static void surface_download_data(struct wined3d_surface *surface, const struct wined3d_gl_info *gl_info)
{
const struct wined3d_format *format = surface->resource.format;
/* Only support read back of converted P8 surfaces. */
if (surface->flags & SFLAG_CONVERTED && format->id != WINED3DFMT_P8_UINT)
{
FIXME("Readback conversion not supported for format %s.\n", debug_d3dformat(format->id));
return;
}
ENTER_GL();
if (format->flags & WINED3DFMT_FLAG_COMPRESSED)
{
TRACE("(%p) : Calling glGetCompressedTexImageARB level %d, format %#x, type %#x, data %p.\n",
surface, surface->texture_level, format->glFormat, format->glType,
surface->resource.allocatedMemory);
if (surface->flags & SFLAG_PBO)
{
GL_EXTCALL(glBindBufferARB(GL_PIXEL_PACK_BUFFER_ARB, surface->pbo));
checkGLcall("glBindBufferARB");
GL_EXTCALL(glGetCompressedTexImageARB(surface->texture_target, surface->texture_level, NULL));
checkGLcall("glGetCompressedTexImageARB");
GL_EXTCALL(glBindBufferARB(GL_PIXEL_PACK_BUFFER_ARB, 0));
checkGLcall("glBindBufferARB");
}
else
{
GL_EXTCALL(glGetCompressedTexImageARB(surface->texture_target,
surface->texture_level, surface->resource.allocatedMemory));
checkGLcall("glGetCompressedTexImageARB");
}
LEAVE_GL();
}
else
{
void *mem;
GLenum gl_format = format->glFormat;
GLenum gl_type = format->glType;
int src_pitch = 0;
int dst_pitch = 0;
/* In case of P8 the index is stored in the alpha component if the primary render target uses P8. */
if (format->id == WINED3DFMT_P8_UINT && primary_render_target_is_p8(surface->resource.device))
{
gl_format = GL_ALPHA;
gl_type = GL_UNSIGNED_BYTE;
}
if (surface->flags & SFLAG_NONPOW2)
{
unsigned char alignment = surface->resource.device->surface_alignment;
src_pitch = format->byte_count * surface->pow2Width;
dst_pitch = wined3d_surface_get_pitch(surface);
src_pitch = (src_pitch + alignment - 1) & ~(alignment - 1);
mem = HeapAlloc(GetProcessHeap(), 0, src_pitch * surface->pow2Height);
}
else
{
mem = surface->resource.allocatedMemory;
}
TRACE("(%p) : Calling glGetTexImage level %d, format %#x, type %#x, data %p\n",
surface, surface->texture_level, gl_format, gl_type, mem);
if (surface->flags & SFLAG_PBO)
{
GL_EXTCALL(glBindBufferARB(GL_PIXEL_PACK_BUFFER_ARB, surface->pbo));
checkGLcall("glBindBufferARB");
glGetTexImage(surface->texture_target, surface->texture_level, gl_format, gl_type, NULL);
checkGLcall("glGetTexImage");
GL_EXTCALL(glBindBufferARB(GL_PIXEL_PACK_BUFFER_ARB, 0));
checkGLcall("glBindBufferARB");
}
else
{
glGetTexImage(surface->texture_target, surface->texture_level, gl_format, gl_type, mem);
checkGLcall("glGetTexImage");
}
LEAVE_GL();
if (surface->flags & SFLAG_NONPOW2)
{
const BYTE *src_data;
BYTE *dst_data;
UINT y;
/*
* Some games (e.g. warhammer 40k) don't work properly with the odd pitches, preventing
* the surface pitch from being used to box non-power2 textures. Instead we have to use a hack to
* repack the texture so that the bpp * width pitch can be used instead of bpp * pow2width.
*
* We're doing this...
*
* instead of boxing the texture :
* |<-texture width ->| -->pow2width| /\
* |111111111111111111| | |
* |222 Texture 222222| boxed empty | texture height
* |3333 Data 33333333| | |
* |444444444444444444| | \/
* ----------------------------------- |
* | boxed empty | boxed empty | pow2height
* | | | \/
* -----------------------------------
*
*
* we're repacking the data to the expected texture width
*
* |<-texture width ->| -->pow2width| /\
* |111111111111111111222222222222222| |
* |222333333333333333333444444444444| texture height
* |444444 | |
* | | \/
* | | |
* | empty | pow2height
* | | \/
* -----------------------------------
*
* == is the same as
*
* |<-texture width ->| /\
* |111111111111111111|
* |222222222222222222|texture height
* |333333333333333333|
* |444444444444444444| \/
* --------------------
*
* this also means that any references to allocatedMemory should work with the data as if were a
* standard texture with a non-power2 width instead of texture boxed up to be a power2 texture.
*
* internally the texture is still stored in a boxed format so any references to textureName will
* get a boxed texture with width pow2width and not a texture of width resource.width.
*
* Performance should not be an issue, because applications normally do not lock the surfaces when
* rendering. If an app does, the SFLAG_DYNLOCK flag will kick in and the memory copy won't be released,
* and doesn't have to be re-read. */
src_data = mem;
dst_data = surface->resource.allocatedMemory;
TRACE("(%p) : Repacking the surface data from pitch %d to pitch %d\n", surface, src_pitch, dst_pitch);
for (y = 1; y < surface->resource.height; ++y)
{
/* skip the first row */
src_data += src_pitch;
dst_data += dst_pitch;
memcpy(dst_data, src_data, dst_pitch);
}
HeapFree(GetProcessHeap(), 0, mem);
}
}
/* Surface has now been downloaded */
surface->flags |= SFLAG_INSYSMEM;
}
/* This call just uploads data, the caller is responsible for binding the
* correct texture. */
/* Context activation is done by the caller. */
static void surface_upload_data(struct wined3d_surface *surface, const struct wined3d_gl_info *gl_info,
const struct wined3d_format *format, BOOL srgb, const GLvoid *data)
{
GLsizei width = surface->resource.width;
GLsizei height = surface->resource.height;
GLenum internal;
if (srgb)
{
internal = format->glGammaInternal;
}
else if (surface->resource.usage & WINED3DUSAGE_RENDERTARGET && surface_is_offscreen(surface))
{
internal = format->rtInternal;
}
else
{
internal = format->glInternal;
}
TRACE("surface %p, internal %#x, width %d, height %d, format %#x, type %#x, data %p.\n",
surface, internal, width, height, format->glFormat, format->glType, data);
TRACE("target %#x, level %u, resource size %u.\n",
surface->texture_target, surface->texture_level, surface->resource.size);
if (format->heightscale != 1.0f && format->heightscale != 0.0f) height *= format->heightscale;
ENTER_GL();
if (surface->flags & SFLAG_PBO)
{
GL_EXTCALL(glBindBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB, surface->pbo));
checkGLcall("glBindBufferARB");
TRACE("(%p) pbo: %#x, data: %p.\n", surface, surface->pbo, data);
data = NULL;
}
if (format->flags & WINED3DFMT_FLAG_COMPRESSED)
{
TRACE("Calling glCompressedTexSubImage2DARB.\n");
GL_EXTCALL(glCompressedTexSubImage2DARB(surface->texture_target, surface->texture_level,
0, 0, width, height, internal, surface->resource.size, data));
checkGLcall("glCompressedTexSubImage2DARB");
}
else
{
TRACE("Calling glTexSubImage2D.\n");
glTexSubImage2D(surface->texture_target, surface->texture_level,
0, 0, width, height, format->glFormat, format->glType, data);
checkGLcall("glTexSubImage2D");
}
if (surface->flags & SFLAG_PBO)
{
GL_EXTCALL(glBindBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB, 0));
checkGLcall("glBindBufferARB");
}
LEAVE_GL();
if (gl_info->quirks & WINED3D_QUIRK_FBO_TEX_UPDATE)
{
IWineD3DDeviceImpl *device = surface->resource.device;
unsigned int i;
for (i = 0; i < device->context_count; ++i)
{
context_surface_update(device->contexts[i], surface);
}
}
}
/* This call just allocates the texture, the caller is responsible for binding
* the correct texture. */
/* Context activation is done by the caller. */
static void surface_allocate_surface(struct wined3d_surface *surface, const struct wined3d_gl_info *gl_info,
const struct wined3d_format *format, BOOL srgb)
{
BOOL enable_client_storage = FALSE;
GLsizei width = surface->pow2Width;
GLsizei height = surface->pow2Height;
const BYTE *mem = NULL;
GLenum internal;
if (srgb)
{
internal = format->glGammaInternal;
}
else if (surface->resource.usage & WINED3DUSAGE_RENDERTARGET && surface_is_offscreen(surface))
{
internal = format->rtInternal;
}
else
{
internal = format->glInternal;
}
if (format->heightscale != 1.0f && format->heightscale != 0.0f) height *= format->heightscale;
TRACE("(%p) : Creating surface (target %#x) level %d, d3d format %s, internal format %#x, width %d, height %d, gl format %#x, gl type=%#x\n",
surface, surface->texture_target, surface->texture_level, debug_d3dformat(format->id),
internal, width, height, format->glFormat, format->glType);
ENTER_GL();
if (gl_info->supported[APPLE_CLIENT_STORAGE])
{
if (surface->flags & (SFLAG_NONPOW2 | SFLAG_DIBSECTION | SFLAG_CONVERTED)
|| !surface->resource.allocatedMemory)
{
/* In some cases we want to disable client storage.
* SFLAG_NONPOW2 has a bigger opengl texture than the client memory, and different pitches
* SFLAG_DIBSECTION: Dibsections may have read / write protections on the memory. Avoid issues...
* SFLAG_CONVERTED: The conversion destination memory is freed after loading the surface
* allocatedMemory == NULL: Not defined in the extension. Seems to disable client storage effectively
*/
glPixelStorei(GL_UNPACK_CLIENT_STORAGE_APPLE, GL_FALSE);
checkGLcall("glPixelStorei(GL_UNPACK_CLIENT_STORAGE_APPLE, GL_FALSE)");
surface->flags &= ~SFLAG_CLIENT;
enable_client_storage = TRUE;
}
else
{
surface->flags |= SFLAG_CLIENT;
/* Point OpenGL to our allocated texture memory. Do not use
* resource.allocatedMemory here because it might point into a
* PBO. Instead use heapMemory, but get the alignment right. */
mem = (BYTE *)(((ULONG_PTR)surface->resource.heapMemory
+ (RESOURCE_ALIGNMENT - 1)) & ~(RESOURCE_ALIGNMENT - 1));
}
}
if (format->flags & WINED3DFMT_FLAG_COMPRESSED && mem)
{
GL_EXTCALL(glCompressedTexImage2DARB(surface->texture_target, surface->texture_level,
internal, width, height, 0, surface->resource.size, mem));
checkGLcall("glCompressedTexImage2DARB");
}
else
{
glTexImage2D(surface->texture_target, surface->texture_level,
internal, width, height, 0, format->glFormat, format->glType, mem);
checkGLcall("glTexImage2D");
}
if(enable_client_storage) {
glPixelStorei(GL_UNPACK_CLIENT_STORAGE_APPLE, GL_TRUE);
checkGLcall("glPixelStorei(GL_UNPACK_CLIENT_STORAGE_APPLE, GL_TRUE)");
}
LEAVE_GL();
}
/* In D3D the depth stencil dimensions have to be greater than or equal to the
* render target dimensions. With FBOs, the dimensions have to be an exact match. */
/* TODO: We should synchronize the renderbuffer's content with the texture's content. */
/* GL locking is done by the caller */
void surface_set_compatible_renderbuffer(struct wined3d_surface *surface, struct wined3d_surface *rt)
{
const struct wined3d_gl_info *gl_info = &surface->resource.device->adapter->gl_info;
renderbuffer_entry_t *entry;
GLuint renderbuffer = 0;
unsigned int src_width, src_height;
unsigned int width, height;
if (rt && rt->resource.format->id != WINED3DFMT_NULL)
{
width = rt->pow2Width;
height = rt->pow2Height;
}
else
{
width = surface->pow2Width;
height = surface->pow2Height;
}
src_width = surface->pow2Width;
src_height = surface->pow2Height;
/* A depth stencil smaller than the render target is not valid */
if (width > src_width || height > src_height) return;
/* Remove any renderbuffer set if the sizes match */
if (gl_info->supported[ARB_FRAMEBUFFER_OBJECT]
|| (width == src_width && height == src_height))
{
surface->current_renderbuffer = NULL;
return;
}
/* Look if we've already got a renderbuffer of the correct dimensions */
LIST_FOR_EACH_ENTRY(entry, &surface->renderbuffers, renderbuffer_entry_t, entry)
{
if (entry->width == width && entry->height == height)
{
renderbuffer = entry->id;
surface->current_renderbuffer = entry;
break;
}
}
if (!renderbuffer)
{
gl_info->fbo_ops.glGenRenderbuffers(1, &renderbuffer);
gl_info->fbo_ops.glBindRenderbuffer(GL_RENDERBUFFER, renderbuffer);
gl_info->fbo_ops.glRenderbufferStorage(GL_RENDERBUFFER,
surface->resource.format->glInternal, width, height);
entry = HeapAlloc(GetProcessHeap(), 0, sizeof(renderbuffer_entry_t));
entry->width = width;
entry->height = height;
entry->id = renderbuffer;
list_add_head(&surface->renderbuffers, &entry->entry);
surface->current_renderbuffer = entry;
}
checkGLcall("set_compatible_renderbuffer");
}
GLenum surface_get_gl_buffer(struct wined3d_surface *surface)
{
struct wined3d_swapchain *swapchain = surface->container.u.swapchain;
TRACE("surface %p.\n", surface);
if (surface->container.type != WINED3D_CONTAINER_SWAPCHAIN)
{
ERR("Surface %p is not on a swapchain.\n", surface);
return GL_NONE;
}
if (swapchain->back_buffers && swapchain->back_buffers[0] == surface)
{
if (swapchain->render_to_fbo)
{
TRACE("Returning GL_COLOR_ATTACHMENT0\n");
return GL_COLOR_ATTACHMENT0;
}
TRACE("Returning GL_BACK\n");
return GL_BACK;
}
else if (surface == swapchain->front_buffer)
{
TRACE("Returning GL_FRONT\n");
return GL_FRONT;
}
FIXME("Higher back buffer, returning GL_BACK\n");
return GL_BACK;
}
/* Slightly inefficient way to handle multiple dirty rects but it works :) */
void surface_add_dirty_rect(struct wined3d_surface *surface, const WINED3DBOX *dirty_rect)
{
TRACE("surface %p, dirty_rect %p.\n", surface, dirty_rect);
if (!(surface->flags & SFLAG_INSYSMEM) && (surface->flags & SFLAG_INTEXTURE))
/* No partial locking for textures yet. */
surface_load_location(surface, SFLAG_INSYSMEM, NULL);
surface_modify_location(surface, SFLAG_INSYSMEM, TRUE);
if (dirty_rect)
{
surface->dirtyRect.left = min(surface->dirtyRect.left, dirty_rect->Left);
surface->dirtyRect.top = min(surface->dirtyRect.top, dirty_rect->Top);
surface->dirtyRect.right = max(surface->dirtyRect.right, dirty_rect->Right);
surface->dirtyRect.bottom = max(surface->dirtyRect.bottom, dirty_rect->Bottom);
}
else
{
surface->dirtyRect.left = 0;
surface->dirtyRect.top = 0;
surface->dirtyRect.right = surface->resource.width;
surface->dirtyRect.bottom = surface->resource.height;
}
/* if the container is a texture then mark it dirty. */
if (surface->container.type == WINED3D_CONTAINER_TEXTURE)
{
TRACE("Passing to container.\n");
wined3d_texture_set_dirty(surface->container.u.texture, TRUE);
}
}
static BOOL surface_convert_color_to_float(struct wined3d_surface *surface,
DWORD color, WINED3DCOLORVALUE *float_color)
{
const struct wined3d_format *format = surface->resource.format;
IWineD3DDeviceImpl *device = surface->resource.device;
switch (format->id)
{
case WINED3DFMT_P8_UINT:
if (surface->palette)
{
float_color->r = surface->palette->palents[color].peRed / 255.0f;
float_color->g = surface->palette->palents[color].peGreen / 255.0f;
float_color->b = surface->palette->palents[color].peBlue / 255.0f;
}
else
{
float_color->r = 0.0f;
float_color->g = 0.0f;
float_color->b = 0.0f;
}
float_color->a = primary_render_target_is_p8(device) ? color / 255.0f : 1.0f;
break;
case WINED3DFMT_B5G6R5_UNORM:
float_color->r = ((color >> 11) & 0x1f) / 31.0f;
float_color->g = ((color >> 5) & 0x3f) / 63.0f;
float_color->b = (color & 0x1f) / 31.0f;
float_color->a = 1.0f;
break;
case WINED3DFMT_B8G8R8_UNORM:
case WINED3DFMT_B8G8R8X8_UNORM:
float_color->r = D3DCOLOR_R(color);
float_color->g = D3DCOLOR_G(color);
float_color->b = D3DCOLOR_B(color);
float_color->a = 1.0f;
break;
case WINED3DFMT_B8G8R8A8_UNORM:
float_color->r = D3DCOLOR_R(color);
float_color->g = D3DCOLOR_G(color);
float_color->b = D3DCOLOR_B(color);
float_color->a = D3DCOLOR_A(color);
break;
default:
ERR("Unhandled conversion from %s to floating point.\n", debug_d3dformat(format->id));
return FALSE;
}
return TRUE;
}
HRESULT surface_load(struct wined3d_surface *surface, BOOL srgb)
{
DWORD flag = srgb ? SFLAG_INSRGBTEX : SFLAG_INTEXTURE;
TRACE("surface %p, srgb %#x.\n", surface, srgb);
if (surface->resource.pool == WINED3DPOOL_SCRATCH)
{
ERR("Not supported on scratch surfaces.\n");
return WINED3DERR_INVALIDCALL;
}
if (!(surface->flags & flag))
{
TRACE("Reloading because surface is dirty\n");
}
/* Reload if either the texture and sysmem have different ideas about the
* color key, or the actual key values changed. */
else if (!(surface->flags & SFLAG_GLCKEY) != !(surface->CKeyFlags & WINEDDSD_CKSRCBLT)
|| ((surface->CKeyFlags & WINEDDSD_CKSRCBLT)
&& (surface->glCKey.dwColorSpaceLowValue != surface->SrcBltCKey.dwColorSpaceLowValue
|| surface->glCKey.dwColorSpaceHighValue != surface->SrcBltCKey.dwColorSpaceHighValue)))
{
TRACE("Reloading because of color keying\n");
/* To perform the color key conversion we need a sysmem copy of
* the surface. Make sure we have it. */
surface_load_location(surface, SFLAG_INSYSMEM, NULL);
/* Make sure the texture is reloaded because of the color key change,
* this kills performance though :( */
/* TODO: This is not necessarily needed with hw palettized texture support. */
surface_modify_location(surface, SFLAG_INSYSMEM, TRUE);
}
else
{
TRACE("surface is already in texture\n");
return WINED3D_OK;
}
/* No partial locking for textures yet. */
surface_load_location(surface, flag, NULL);
surface_evict_sysmem(surface);
return WINED3D_OK;
}
/* See also float_16_to_32() in wined3d_private.h */
static inline unsigned short float_32_to_16(const float *in)
{
int exp = 0;
float tmp = fabsf(*in);
unsigned int mantissa;
unsigned short ret;
/* Deal with special numbers */
if (*in == 0.0f)
return 0x0000;
if (isnan(*in))
return 0x7c01;
if (isinf(*in))
return (*in < 0.0f ? 0xfc00 : 0x7c00);
if (tmp < powf(2, 10))
{
do
{
tmp = tmp * 2.0f;
exp--;
} while (tmp < powf(2, 10));
}
else if (tmp >= powf(2, 11))
{
do
{
tmp /= 2.0f;
exp++;
} while (tmp >= powf(2, 11));
}
mantissa = (unsigned int)tmp;
if (tmp - mantissa >= 0.5f)
++mantissa; /* Round to nearest, away from zero. */
exp += 10; /* Normalize the mantissa. */
exp += 15; /* Exponent is encoded with excess 15. */
if (exp > 30) /* too big */
{
ret = 0x7c00; /* INF */
}
else if (exp <= 0)
{
/* exp == 0: Non-normalized mantissa. Returns 0x0000 (=0.0) for too small numbers. */
while (exp <= 0)
{
mantissa = mantissa >> 1;
++exp;
}
ret = mantissa & 0x3ff;
}
else
{
ret = (exp << 10) | (mantissa & 0x3ff);
}
ret |= ((*in < 0.0f ? 1 : 0) << 15); /* Add the sign */
return ret;
}
ULONG CDECL wined3d_surface_incref(struct wined3d_surface *surface)
{
ULONG refcount;
TRACE("Surface %p, container %p of type %#x.\n",
surface, surface->container.u.base, surface->container.type);
switch (surface->container.type)
{
case WINED3D_CONTAINER_TEXTURE:
return wined3d_texture_incref(surface->container.u.texture);
case WINED3D_CONTAINER_SWAPCHAIN:
return wined3d_swapchain_incref(surface->container.u.swapchain);
default:
ERR("Unhandled container type %#x.\n", surface->container.type);
case WINED3D_CONTAINER_NONE:
break;
}
refcount = InterlockedIncrement(&surface->resource.ref);
TRACE("%p increasing refcount to %u.\n", surface, refcount);
return refcount;
}
/* Do not call while under the GL lock. */
ULONG CDECL wined3d_surface_decref(struct wined3d_surface *surface)
{
ULONG refcount;
TRACE("Surface %p, container %p of type %#x.\n",
surface, surface->container.u.base, surface->container.type);
switch (surface->container.type)
{
case WINED3D_CONTAINER_TEXTURE:
return wined3d_texture_decref(surface->container.u.texture);
case WINED3D_CONTAINER_SWAPCHAIN:
return wined3d_swapchain_decref(surface->container.u.swapchain);
default:
ERR("Unhandled container type %#x.\n", surface->container.type);
case WINED3D_CONTAINER_NONE:
break;
}
refcount = InterlockedDecrement(&surface->resource.ref);
TRACE("%p decreasing refcount to %u.\n", surface, refcount);
if (!refcount)
{
surface->surface_ops->surface_cleanup(surface);
surface->resource.parent_ops->wined3d_object_destroyed(surface->resource.parent);
TRACE("Destroyed surface %p.\n", surface);
HeapFree(GetProcessHeap(), 0, surface);
}
return refcount;
}
HRESULT CDECL wined3d_surface_set_private_data(struct wined3d_surface *surface,
REFGUID riid, const void *data, DWORD data_size, DWORD flags)
{
return resource_set_private_data(&surface->resource, riid, data, data_size, flags);
}
HRESULT CDECL wined3d_surface_get_private_data(const struct wined3d_surface *surface,
REFGUID guid, void *data, DWORD *data_size)
{
return resource_get_private_data(&surface->resource, guid, data, data_size);
}
HRESULT CDECL wined3d_surface_free_private_data(struct wined3d_surface *surface, REFGUID refguid)
{
return resource_free_private_data(&surface->resource, refguid);
}
DWORD CDECL wined3d_surface_set_priority(struct wined3d_surface *surface, DWORD priority)
{
return resource_set_priority(&surface->resource, priority);
}
DWORD CDECL wined3d_surface_get_priority(const struct wined3d_surface *surface)
{
return resource_get_priority(&surface->resource);
}
void CDECL wined3d_surface_preload(struct wined3d_surface *surface)
{
TRACE("surface %p.\n", surface);
surface->surface_ops->surface_preload(surface);
}
void * CDECL wined3d_surface_get_parent(const struct wined3d_surface *surface)
{
TRACE("surface %p.\n", surface);
return surface->resource.parent;
}
struct wined3d_resource * CDECL wined3d_surface_get_resource(struct wined3d_surface *surface)
{
TRACE("surface %p.\n", surface);
return &surface->resource;
}
HRESULT CDECL wined3d_surface_get_blt_status(const struct wined3d_surface *surface, DWORD flags)
{
TRACE("surface %p, flags %#x.\n", surface, flags);
switch (flags)
{
case WINEDDGBS_CANBLT:
case WINEDDGBS_ISBLTDONE:
return WINED3D_OK;
default:
return WINED3DERR_INVALIDCALL;
}
}
HRESULT CDECL wined3d_surface_get_flip_status(const struct wined3d_surface *surface, DWORD flags)
{
TRACE("surface %p, flags %#x.\n", surface, flags);
/* XXX: DDERR_INVALIDSURFACETYPE */
switch (flags)
{
case WINEDDGFS_CANFLIP:
case WINEDDGFS_ISFLIPDONE:
return WINED3D_OK;
default:
return WINED3DERR_INVALIDCALL;
}
}
HRESULT CDECL wined3d_surface_is_lost(const struct wined3d_surface *surface)
{
TRACE("surface %p.\n", surface);
/* D3D8 and 9 loose full devices, ddraw only surfaces. */
return surface->flags & SFLAG_LOST ? WINED3DERR_DEVICELOST : WINED3D_OK;
}
HRESULT CDECL wined3d_surface_restore(struct wined3d_surface *surface)
{
TRACE("surface %p.\n", surface);
/* So far we don't lose anything :) */
surface->flags &= ~SFLAG_LOST;
return WINED3D_OK;
}
HRESULT CDECL wined3d_surface_set_palette(struct wined3d_surface *surface, struct wined3d_palette *palette)
{
TRACE("surface %p, palette %p.\n", surface, palette);
if (surface->palette == palette)
{
TRACE("Nop palette change.\n");
return WINED3D_OK;
}
if (surface->palette && (surface->resource.usage & WINED3DUSAGE_RENDERTARGET))
surface->palette->flags &= ~WINEDDPCAPS_PRIMARYSURFACE;
surface->palette = palette;
if (palette)
{
if (surface->resource.usage & WINED3DUSAGE_RENDERTARGET)
palette->flags |= WINEDDPCAPS_PRIMARYSURFACE;
surface->surface_ops->surface_realize_palette(surface);
}
return WINED3D_OK;
}
HRESULT CDECL wined3d_surface_set_color_key(struct wined3d_surface *surface,
DWORD flags, const WINEDDCOLORKEY *color_key)
{
TRACE("surface %p, flags %#x, color_key %p.\n", surface, flags, color_key);
if (flags & WINEDDCKEY_COLORSPACE)
{
FIXME(" colorkey value not supported (%08x) !\n", flags);
return WINED3DERR_INVALIDCALL;
}
/* Dirtify the surface, but only if a key was changed. */
if (color_key)
{
switch (flags & ~WINEDDCKEY_COLORSPACE)
{
case WINEDDCKEY_DESTBLT:
surface->DestBltCKey = *color_key;
surface->CKeyFlags |= WINEDDSD_CKDESTBLT;
break;
case WINEDDCKEY_DESTOVERLAY:
surface->DestOverlayCKey = *color_key;
surface->CKeyFlags |= WINEDDSD_CKDESTOVERLAY;
break;
case WINEDDCKEY_SRCOVERLAY:
surface->SrcOverlayCKey = *color_key;
surface->CKeyFlags |= WINEDDSD_CKSRCOVERLAY;
break;
case WINEDDCKEY_SRCBLT:
surface->SrcBltCKey = *color_key;
surface->CKeyFlags |= WINEDDSD_CKSRCBLT;
break;
}
}
else
{
switch (flags & ~WINEDDCKEY_COLORSPACE)
{
case WINEDDCKEY_DESTBLT:
surface->CKeyFlags &= ~WINEDDSD_CKDESTBLT;
break;
case WINEDDCKEY_DESTOVERLAY:
surface->CKeyFlags &= ~WINEDDSD_CKDESTOVERLAY;
break;
case WINEDDCKEY_SRCOVERLAY:
surface->CKeyFlags &= ~WINEDDSD_CKSRCOVERLAY;
break;
case WINEDDCKEY_SRCBLT:
surface->CKeyFlags &= ~WINEDDSD_CKSRCBLT;
break;
}
}
return WINED3D_OK;
}
struct wined3d_palette * CDECL wined3d_surface_get_palette(const struct wined3d_surface *surface)
{
TRACE("surface %p.\n", surface);
return surface->palette;
}
DWORD CDECL wined3d_surface_get_pitch(const struct wined3d_surface *surface)
{
const struct wined3d_format *format = surface->resource.format;
DWORD pitch;
TRACE("surface %p.\n", surface);
if ((format->flags & (WINED3DFMT_FLAG_COMPRESSED | WINED3DFMT_FLAG_BROKEN_PITCH)) == WINED3DFMT_FLAG_COMPRESSED)
{
/* Since compressed formats are block based, pitch means the amount of
* bytes to the next row of block rather than the next row of pixels. */
UINT row_block_count = (surface->resource.width + format->block_width - 1) / format->block_width;
pitch = row_block_count * format->block_byte_count;
}
else
{
unsigned char alignment = surface->resource.device->surface_alignment;
pitch = surface->resource.format->byte_count * surface->resource.width; /* Bytes / row */
pitch = (pitch + alignment - 1) & ~(alignment - 1);
}
TRACE("Returning %u.\n", pitch);
return pitch;
}
HRESULT CDECL wined3d_surface_set_mem(struct wined3d_surface *surface, void *mem)
{
TRACE("surface %p, mem %p.\n", surface, mem);
if (surface->flags & (SFLAG_LOCKED | SFLAG_DCINUSE))
{
WARN("Surface is locked or the DC is in use.\n");
return WINED3DERR_INVALIDCALL;
}
return surface->surface_ops->surface_set_mem(surface, mem);
}
HRESULT CDECL wined3d_surface_set_overlay_position(struct wined3d_surface *surface, LONG x, LONG y)
{
LONG w, h;
TRACE("surface %p, x %d, y %d.\n", surface, x, y);
if (!(surface->resource.usage & WINED3DUSAGE_OVERLAY))
{
WARN("Not an overlay surface.\n");
return WINEDDERR_NOTAOVERLAYSURFACE;
}
w = surface->overlay_destrect.right - surface->overlay_destrect.left;
h = surface->overlay_destrect.bottom - surface->overlay_destrect.top;
surface->overlay_destrect.left = x;
surface->overlay_destrect.top = y;
surface->overlay_destrect.right = x + w;
surface->overlay_destrect.bottom = y + h;
surface->surface_ops->surface_draw_overlay(surface);
return WINED3D_OK;
}
HRESULT CDECL wined3d_surface_get_overlay_position(const struct wined3d_surface *surface, LONG *x, LONG *y)
{
TRACE("surface %p, x %p, y %p.\n", surface, x, y);
if (!(surface->resource.usage & WINED3DUSAGE_OVERLAY))
{
TRACE("Not an overlay surface.\n");
return WINEDDERR_NOTAOVERLAYSURFACE;
}
if (!surface->overlay_dest)
{
TRACE("Overlay not visible.\n");
*x = 0;
*y = 0;
return WINEDDERR_OVERLAYNOTVISIBLE;
}
*x = surface->overlay_destrect.left;
*y = surface->overlay_destrect.top;
TRACE("Returning position %d, %d.\n", *x, *y);
return WINED3D_OK;
}
HRESULT CDECL wined3d_surface_update_overlay_z_order(struct wined3d_surface *surface,
DWORD flags, struct wined3d_surface *ref)
{
FIXME("surface %p, flags %#x, ref %p stub!\n", surface, flags, ref);
if (!(surface->resource.usage & WINED3DUSAGE_OVERLAY))
{
TRACE("Not an overlay surface.\n");
return WINEDDERR_NOTAOVERLAYSURFACE;
}
return WINED3D_OK;
}
HRESULT CDECL wined3d_surface_update_overlay(struct wined3d_surface *surface, const RECT *src_rect,
struct wined3d_surface *dst_surface, const RECT *dst_rect, DWORD flags, const WINEDDOVERLAYFX *fx)
{
TRACE("surface %p, src_rect %s, dst_surface %p, dst_rect %s, flags %#x, fx %p.\n",
surface, wine_dbgstr_rect(src_rect), dst_surface, wine_dbgstr_rect(dst_rect), flags, fx);
if (!(surface->resource.usage & WINED3DUSAGE_OVERLAY))
{
WARN("Not an overlay surface.\n");
return WINEDDERR_NOTAOVERLAYSURFACE;
}
else if (!dst_surface)
{
WARN("Dest surface is NULL.\n");
return WINED3DERR_INVALIDCALL;
}
if (src_rect)
{
surface->overlay_srcrect = *src_rect;
}
else
{
surface->overlay_srcrect.left = 0;
surface->overlay_srcrect.top = 0;
surface->overlay_srcrect.right = surface->resource.width;
surface->overlay_srcrect.bottom = surface->resource.height;
}
if (dst_rect)
{
surface->overlay_destrect = *dst_rect;
}
else
{
surface->overlay_destrect.left = 0;
surface->overlay_destrect.top = 0;
surface->overlay_destrect.right = dst_surface ? dst_surface->resource.width : 0;
surface->overlay_destrect.bottom = dst_surface ? dst_surface->resource.height : 0;
}
if (surface->overlay_dest && (surface->overlay_dest != dst_surface || flags & WINEDDOVER_HIDE))
{
list_remove(&surface->overlay_entry);
}
if (flags & WINEDDOVER_SHOW)
{
if (surface->overlay_dest != dst_surface)
{
surface->overlay_dest = dst_surface;
list_add_tail(&dst_surface->overlays, &surface->overlay_entry);
}
}
else if (flags & WINEDDOVER_HIDE)
{
/* tests show that the rectangles are erased on hide */
surface->overlay_srcrect.left = 0; surface->overlay_srcrect.top = 0;
surface->overlay_srcrect.right = 0; surface->overlay_srcrect.bottom = 0;
surface->overlay_destrect.left = 0; surface->overlay_destrect.top = 0;
surface->overlay_destrect.right = 0; surface->overlay_destrect.bottom = 0;
surface->overlay_dest = NULL;
}
surface->surface_ops->surface_draw_overlay(surface);
return WINED3D_OK;
}
HRESULT CDECL wined3d_surface_set_clipper(struct wined3d_surface *surface, struct wined3d_clipper *clipper)
{
TRACE("surface %p, clipper %p.\n", surface, clipper);
surface->clipper = clipper;
return WINED3D_OK;
}
struct wined3d_clipper * CDECL wined3d_surface_get_clipper(const struct wined3d_surface *surface)
{
TRACE("surface %p.\n", surface);
return surface->clipper;
}
HRESULT CDECL wined3d_surface_set_format(struct wined3d_surface *surface, enum wined3d_format_id format_id)
{
const struct wined3d_format *format = wined3d_get_format(&surface->resource.device->adapter->gl_info, format_id);
TRACE("surface %p, format %s.\n", surface, debug_d3dformat(format_id));
if (surface->resource.format->id != WINED3DFMT_UNKNOWN)
{
FIXME("The format of the surface must be WINED3DFORMAT_UNKNOWN.\n");
return WINED3DERR_INVALIDCALL;
}
surface->resource.size = wined3d_format_calculate_size(format, surface->resource.device->surface_alignment,
surface->pow2Width, surface->pow2Height);
surface->flags |= (WINED3DFMT_D16_LOCKABLE == format_id) ? SFLAG_LOCKABLE : 0;
surface->flags &= ~(SFLAG_ALLOCATED | SFLAG_SRGBALLOCATED);
surface->resource.format = format;
TRACE("size %u, byte_count %u\n", surface->resource.size, format->byte_count);
TRACE("glFormat %#x, glInternal %#x, glType %#x.\n",
format->glFormat, format->glInternal, format->glType);
return WINED3D_OK;
}
static void convert_r32_float_r16_float(const BYTE *src, BYTE *dst,
DWORD pitch_in, DWORD pitch_out, unsigned int w, unsigned int h)
{
unsigned short *dst_s;
const float *src_f;
unsigned int x, y;
TRACE("Converting %ux%u pixels, pitches %u %u.\n", w, h, pitch_in, pitch_out);
for (y = 0; y < h; ++y)
{
src_f = (const float *)(src + y * pitch_in);
dst_s = (unsigned short *) (dst + y * pitch_out);
for (x = 0; x < w; ++x)
{
dst_s[x] = float_32_to_16(src_f + x);
}
}
}
static void convert_r5g6b5_x8r8g8b8(const BYTE *src, BYTE *dst,
DWORD pitch_in, DWORD pitch_out, unsigned int w, unsigned int h)
{
static const unsigned char convert_5to8[] =
{
0x00, 0x08, 0x10, 0x19, 0x21, 0x29, 0x31, 0x3a,
0x42, 0x4a, 0x52, 0x5a, 0x63, 0x6b, 0x73, 0x7b,
0x84, 0x8c, 0x94, 0x9c, 0xa5, 0xad, 0xb5, 0xbd,
0xc5, 0xce, 0xd6, 0xde, 0xe6, 0xef, 0xf7, 0xff,
};
static const unsigned char convert_6to8[] =
{
0x00, 0x04, 0x08, 0x0c, 0x10, 0x14, 0x18, 0x1c,
0x20, 0x24, 0x28, 0x2d, 0x31, 0x35, 0x39, 0x3d,
0x41, 0x45, 0x49, 0x4d, 0x51, 0x55, 0x59, 0x5d,
0x61, 0x65, 0x69, 0x6d, 0x71, 0x75, 0x79, 0x7d,
0x82, 0x86, 0x8a, 0x8e, 0x92, 0x96, 0x9a, 0x9e,
0xa2, 0xa6, 0xaa, 0xae, 0xb2, 0xb6, 0xba, 0xbe,
0xc2, 0xc6, 0xca, 0xce, 0xd2, 0xd7, 0xdb, 0xdf,
0xe3, 0xe7, 0xeb, 0xef, 0xf3, 0xf7, 0xfb, 0xff,
};
unsigned int x, y;
TRACE("Converting %ux%u pixels, pitches %u %u.\n", w, h, pitch_in, pitch_out);
for (y = 0; y < h; ++y)
{
const WORD *src_line = (const WORD *)(src + y * pitch_in);
DWORD *dst_line = (DWORD *)(dst + y * pitch_out);
for (x = 0; x < w; ++x)
{
WORD pixel = src_line[x];
dst_line[x] = 0xff000000
| convert_5to8[(pixel & 0xf800) >> 11] << 16
| convert_6to8[(pixel & 0x07e0) >> 5] << 8
| convert_5to8[(pixel & 0x001f)];
}
}
}
static void convert_a8r8g8b8_x8r8g8b8(const BYTE *src, BYTE *dst,
DWORD pitch_in, DWORD pitch_out, unsigned int w, unsigned int h)
{
unsigned int x, y;
TRACE("Converting %ux%u pixels, pitches %u %u.\n", w, h, pitch_in, pitch_out);
for (y = 0; y < h; ++y)
{
const DWORD *src_line = (const DWORD *)(src + y * pitch_in);
DWORD *dst_line = (DWORD *)(dst + y * pitch_out);
for (x = 0; x < w; ++x)
{
dst_line[x] = 0xff000000 | (src_line[x] & 0xffffff);
}
}
}
static inline BYTE cliptobyte(int x)
{
return (BYTE)((x < 0) ? 0 : ((x > 255) ? 255 : x));
}
static void convert_yuy2_x8r8g8b8(const BYTE *src, BYTE *dst,
DWORD pitch_in, DWORD pitch_out, unsigned int w, unsigned int h)
{
int c2, d, e, r2 = 0, g2 = 0, b2 = 0;
unsigned int x, y;
TRACE("Converting %ux%u pixels, pitches %u %u.\n", w, h, pitch_in, pitch_out);
for (y = 0; y < h; ++y)
{
const BYTE *src_line = src + y * pitch_in;
DWORD *dst_line = (DWORD *)(dst + y * pitch_out);
for (x = 0; x < w; ++x)
{
/* YUV to RGB conversion formulas from http://en.wikipedia.org/wiki/YUV:
* C = Y - 16; D = U - 128; E = V - 128;
* R = cliptobyte((298 * C + 409 * E + 128) >> 8);
* G = cliptobyte((298 * C - 100 * D - 208 * E + 128) >> 8);
* B = cliptobyte((298 * C + 516 * D + 128) >> 8);
* Two adjacent YUY2 pixels are stored as four bytes: Y0 U Y1 V .
* U and V are shared between the pixels. */
if (!(x & 1)) /* For every even pixel, read new U and V. */
{
d = (int) src_line[1] - 128;
e = (int) src_line[3] - 128;
r2 = 409 * e + 128;
g2 = - 100 * d - 208 * e + 128;
b2 = 516 * d + 128;
}
c2 = 298 * ((int) src_line[0] - 16);
dst_line[x] = 0xff000000
| cliptobyte((c2 + r2) >> 8) << 16 /* red */
| cliptobyte((c2 + g2) >> 8) << 8 /* green */
| cliptobyte((c2 + b2) >> 8); /* blue */
/* Scale RGB values to 0..255 range,
* then clip them if still not in range (may be negative),
* then shift them within DWORD if necessary. */
src_line += 2;
}
}
}
static void convert_yuy2_r5g6b5(const BYTE *src, BYTE *dst,
DWORD pitch_in, DWORD pitch_out, unsigned int w, unsigned int h)
{
unsigned int x, y;
int c2, d, e, r2 = 0, g2 = 0, b2 = 0;
TRACE("Converting %ux%u pixels, pitches %u %u\n", w, h, pitch_in, pitch_out);
for (y = 0; y < h; ++y)
{
const BYTE *src_line = src + y * pitch_in;
WORD *dst_line = (WORD *)(dst + y * pitch_out);
for (x = 0; x < w; ++x)
{
/* YUV to RGB conversion formulas from http://en.wikipedia.org/wiki/YUV:
* C = Y - 16; D = U - 128; E = V - 128;
* R = cliptobyte((298 * C + 409 * E + 128) >> 8);
* G = cliptobyte((298 * C - 100 * D - 208 * E + 128) >> 8);
* B = cliptobyte((298 * C + 516 * D + 128) >> 8);
* Two adjacent YUY2 pixels are stored as four bytes: Y0 U Y1 V .
* U and V are shared between the pixels. */
if (!(x & 1)) /* For every even pixel, read new U and V. */
{
d = (int) src_line[1] - 128;
e = (int) src_line[3] - 128;
r2 = 409 * e + 128;
g2 = - 100 * d - 208 * e + 128;
b2 = 516 * d + 128;
}
c2 = 298 * ((int) src_line[0] - 16);
dst_line[x] = (cliptobyte((c2 + r2) >> 8) >> 3) << 11 /* red */
| (cliptobyte((c2 + g2) >> 8) >> 2) << 5 /* green */
| (cliptobyte((c2 + b2) >> 8) >> 3); /* blue */
/* Scale RGB values to 0..255 range,
* then clip them if still not in range (may be negative),
* then shift them within DWORD if necessary. */
src_line += 2;
}
}
}
struct d3dfmt_convertor_desc
{
enum wined3d_format_id from, to;
void (*convert)(const BYTE *src, BYTE *dst, DWORD pitch_in, DWORD pitch_out, unsigned int w, unsigned int h);
};
static const struct d3dfmt_convertor_desc convertors[] =
{
{WINED3DFMT_R32_FLOAT, WINED3DFMT_R16_FLOAT, convert_r32_float_r16_float},
{WINED3DFMT_B5G6R5_UNORM, WINED3DFMT_B8G8R8X8_UNORM, convert_r5g6b5_x8r8g8b8},
{WINED3DFMT_B8G8R8A8_UNORM, WINED3DFMT_B8G8R8X8_UNORM, convert_a8r8g8b8_x8r8g8b8},
{WINED3DFMT_YUY2, WINED3DFMT_B8G8R8X8_UNORM, convert_yuy2_x8r8g8b8},
{WINED3DFMT_YUY2, WINED3DFMT_B5G6R5_UNORM, convert_yuy2_r5g6b5},
};
static inline const struct d3dfmt_convertor_desc *find_convertor(enum wined3d_format_id from,
enum wined3d_format_id to)
{
unsigned int i;
for (i = 0; i < (sizeof(convertors) / sizeof(*convertors)); ++i)
{
if (convertors[i].from == from && convertors[i].to == to)
return &convertors[i];
}
return NULL;
}
/*****************************************************************************
* surface_convert_format
*
* Creates a duplicate of a surface in a different format. Is used by Blt to
* blit between surfaces with different formats.
*
* Parameters
* source: Source surface
* fmt: Requested destination format
*
*****************************************************************************/
static struct wined3d_surface *surface_convert_format(struct wined3d_surface *source, enum wined3d_format_id to_fmt)
{
const struct d3dfmt_convertor_desc *conv;
WINED3DLOCKED_RECT lock_src, lock_dst;
struct wined3d_surface *ret = NULL;
HRESULT hr;
conv = find_convertor(source->resource.format->id, to_fmt);
if (!conv)
{
FIXME("Cannot find a conversion function from format %s to %s.\n",
debug_d3dformat(source->resource.format->id), debug_d3dformat(to_fmt));
return NULL;
}
wined3d_surface_create((IWineD3DDevice *)source->resource.device, source->resource.width,
source->resource.height, to_fmt, TRUE /* lockable */, TRUE /* discard */, 0 /* level */,
0 /* usage */, WINED3DPOOL_SCRATCH, WINED3DMULTISAMPLE_NONE /* TODO: Multisampled conversion */,
0 /* MultiSampleQuality */, source->surface_type, NULL /* parent */, &wined3d_null_parent_ops, &ret);
if (!ret)
{
ERR("Failed to create a destination surface for conversion.\n");
return NULL;
}
memset(&lock_src, 0, sizeof(lock_src));
memset(&lock_dst, 0, sizeof(lock_dst));
hr = wined3d_surface_map(source, &lock_src, NULL, WINED3DLOCK_READONLY);
if (FAILED(hr))
{
ERR("Failed to lock the source surface.\n");
wined3d_surface_decref(ret);
return NULL;
}
hr = wined3d_surface_map(ret, &lock_dst, NULL, WINED3DLOCK_READONLY);
if (FAILED(hr))
{
ERR("Failed to lock the destination surface.\n");
wined3d_surface_unmap(source);
wined3d_surface_decref(ret);
return NULL;
}
conv->convert(lock_src.pBits, lock_dst.pBits, lock_src.Pitch, lock_dst.Pitch,
source->resource.width, source->resource.height);
wined3d_surface_unmap(ret);
wined3d_surface_unmap(source);
return ret;
}
static HRESULT _Blt_ColorFill(BYTE *buf, unsigned int width, unsigned int height,
unsigned int bpp, UINT pitch, DWORD color)
{
BYTE *first;
int x, y;
/* Do first row */
#define COLORFILL_ROW(type) \
do { \
type *d = (type *)buf; \
for (x = 0; x < width; ++x) \
d[x] = (type)color; \
} while(0)
switch (bpp)
{
case 1:
COLORFILL_ROW(BYTE);
break;
case 2:
COLORFILL_ROW(WORD);
break;
case 3:
{
BYTE *d = buf;
for (x = 0; x < width; ++x, d += 3)
{
d[0] = (color ) & 0xFF;
d[1] = (color >> 8) & 0xFF;
d[2] = (color >> 16) & 0xFF;
}
break;
}
case 4:
COLORFILL_ROW(DWORD);
break;
default:
FIXME("Color fill not implemented for bpp %u!\n", bpp * 8);
return WINED3DERR_NOTAVAILABLE;
}
#undef COLORFILL_ROW
/* Now copy first row. */
first = buf;
for (y = 1; y < height; ++y)
{
buf += pitch;
memcpy(buf, first, width * bpp);
}
return WINED3D_OK;
}
/* Do not call while under the GL lock. */
HRESULT CDECL wined3d_surface_blt(struct wined3d_surface *dst_surface, const RECT *dst_rect,
struct wined3d_surface *src_surface, const RECT *src_rect, DWORD flags,
const WINEDDBLTFX *fx, WINED3DTEXTUREFILTERTYPE filter)
{
TRACE("dst_surface %p, dst_rect %s, src_surface %p, src_rect %s, flags %#x, fx %p, filter %s.\n",
dst_surface, wine_dbgstr_rect(dst_rect), src_surface, wine_dbgstr_rect(src_rect),
flags, fx, debug_d3dtexturefiltertype(filter));
return dst_surface->surface_ops->surface_blt(dst_surface,
dst_rect, src_surface, src_rect, flags, fx, filter);
}
/* Do not call while under the GL lock. */
HRESULT CDECL wined3d_surface_bltfast(struct wined3d_surface *dst_surface, DWORD dst_x, DWORD dst_y,
struct wined3d_surface *src_surface, const RECT *src_rect, DWORD trans)
{
TRACE("dst_surface %p, dst_x %u, dst_y %u, src_surface %p, src_rect %s, trans %#x.\n",
dst_surface, dst_x, dst_y, src_surface, wine_dbgstr_rect(src_rect), trans);
return dst_surface->surface_ops->surface_bltfast(dst_surface,
dst_x, dst_y, src_surface, src_rect, trans);
}
HRESULT CDECL wined3d_surface_unmap(struct wined3d_surface *surface)
{
TRACE("surface %p.\n", surface);
if (!(surface->flags & SFLAG_LOCKED))
{
WARN("Trying to unmap unmapped surface.\n");
return WINEDDERR_NOTLOCKED;
}
surface->flags &= ~SFLAG_LOCKED;
surface->surface_ops->surface_unmap(surface);
return WINED3D_OK;
}
HRESULT CDECL wined3d_surface_map(struct wined3d_surface *surface,
WINED3DLOCKED_RECT *locked_rect, const RECT *rect, DWORD flags)
{
TRACE("surface %p, locked_rect %p, rect %s, flags %#x.\n",
surface, locked_rect, wine_dbgstr_rect(rect), flags);
if (surface->flags & SFLAG_LOCKED)
{
WARN("Surface is already mapped.\n");
return WINED3DERR_INVALIDCALL;
}
surface->flags |= SFLAG_LOCKED;
if (!(surface->flags & SFLAG_LOCKABLE))
WARN("Trying to lock unlockable surface.\n");
surface->surface_ops->surface_map(surface, rect, flags);
locked_rect->Pitch = wined3d_surface_get_pitch(surface);
if (!rect)
{
locked_rect->pBits = surface->resource.allocatedMemory;
surface->lockedRect.left = 0;
surface->lockedRect.top = 0;
surface->lockedRect.right = surface->resource.width;
surface->lockedRect.bottom = surface->resource.height;
}
else
{
const struct wined3d_format *format = surface->resource.format;
if ((format->flags & (WINED3DFMT_FLAG_COMPRESSED | WINED3DFMT_FLAG_BROKEN_PITCH)) == WINED3DFMT_FLAG_COMPRESSED)
{
/* Compressed textures are block based, so calculate the offset of
* the block that contains the top-left pixel of the locked rectangle. */
locked_rect->pBits = surface->resource.allocatedMemory
+ ((rect->top / format->block_height) * locked_rect->Pitch)
+ ((rect->left / format->block_width) * format->block_byte_count);
}
else
{
locked_rect->pBits = surface->resource.allocatedMemory
+ (locked_rect->Pitch * rect->top)
+ (rect->left * format->byte_count);
}
surface->lockedRect.left = rect->left;
surface->lockedRect.top = rect->top;
surface->lockedRect.right = rect->right;
surface->lockedRect.bottom = rect->bottom;
}
TRACE("Locked rect %s.\n", wine_dbgstr_rect(&surface->lockedRect));
TRACE("Returning memory %p, pitch %u.\n", locked_rect->pBits, locked_rect->Pitch);
return WINED3D_OK;
}
HRESULT CDECL wined3d_surface_getdc(struct wined3d_surface *surface, HDC *dc)
{
HRESULT hr;
TRACE("surface %p, dc %p.\n", surface, dc);
if (surface->flags & SFLAG_USERPTR)
{
ERR("Not supported on surfaces with application-provided memory.\n");
return WINEDDERR_NODC;
}
/* Give more detailed info for ddraw. */
if (surface->flags & SFLAG_DCINUSE)
return WINEDDERR_DCALREADYCREATED;
/* Can't GetDC if the surface is locked. */
if (surface->flags & SFLAG_LOCKED)
return WINED3DERR_INVALIDCALL;
hr = surface->surface_ops->surface_getdc(surface);
if (FAILED(hr))
return hr;
if (surface->resource.format->id == WINED3DFMT_P8_UINT
|| surface->resource.format->id == WINED3DFMT_P8_UINT_A8_UNORM)
{
/* GetDC on palettized formats is unsupported in D3D9, and the method
* is missing in D3D8, so this should only be used for DX <=7
* surfaces (with non-device palettes). */
const PALETTEENTRY *pal = NULL;
if (surface->palette)
{
pal = surface->palette->palents;
}
else
{
struct wined3d_swapchain *swapchain = surface->resource.device->swapchains[0];
struct wined3d_surface *dds_primary = swapchain->front_buffer;
if (dds_primary && dds_primary->palette)
pal = dds_primary->palette->palents;
}
if (pal)
{
RGBQUAD col[256];
unsigned int i;
for (i = 0; i < 256; ++i)
{
col[i].rgbRed = pal[i].peRed;
col[i].rgbGreen = pal[i].peGreen;
col[i].rgbBlue = pal[i].peBlue;
col[i].rgbReserved = 0;
}
SetDIBColorTable(surface->hDC, 0, 256, col);
}
}
surface->flags |= SFLAG_DCINUSE;
*dc = surface->hDC;
TRACE("Returning dc %p.\n", *dc);
return WINED3D_OK;
}
HRESULT CDECL wined3d_surface_releasedc(struct wined3d_surface *surface, HDC dc)
{
TRACE("surface %p, dc %p.\n", surface, dc);
if (!(surface->flags & SFLAG_DCINUSE))
return WINEDDERR_NODC;
if (surface->hDC != dc)
{
WARN("Application tries to release invalid DC %p, surface DC is %p.\n",
dc, surface->hDC);
return WINEDDERR_NODC;
}
if ((surface->flags & SFLAG_PBO) && surface->resource.allocatedMemory)
{
/* Copy the contents of the DIB over to the PBO. */
memcpy(surface->resource.allocatedMemory, surface->dib.bitmap_data, surface->dib.bitmap_size);
}
/* We locked first, so unlock now. */
wined3d_surface_unmap(surface);
surface->flags &= ~SFLAG_DCINUSE;
return WINED3D_OK;
}
HRESULT CDECL wined3d_surface_flip(struct wined3d_surface *surface, struct wined3d_surface *override, DWORD flags)
{
struct wined3d_swapchain *swapchain;
HRESULT hr;
TRACE("surface %p, override %p, flags %#x.\n", surface, override, flags);
if (surface->container.type != WINED3D_CONTAINER_SWAPCHAIN)
{
ERR("Flipped surface is not on a swapchain.\n");
return WINEDDERR_NOTFLIPPABLE;
}
swapchain = surface->container.u.swapchain;
hr = surface->surface_ops->surface_flip(surface, override);
if (FAILED(hr))
return hr;
/* Just overwrite the swapchain presentation interval. This is ok because
* only ddraw apps can call Flip, and only d3d8 and d3d9 applications
* specify the presentation interval. */
if (!(flags & (WINEDDFLIP_NOVSYNC | WINEDDFLIP_INTERVAL2 | WINEDDFLIP_INTERVAL3 | WINEDDFLIP_INTERVAL4)))
swapchain->presentParms.PresentationInterval = WINED3DPRESENT_INTERVAL_ONE;
else if (flags & WINEDDFLIP_NOVSYNC)
swapchain->presentParms.PresentationInterval = WINED3DPRESENT_INTERVAL_IMMEDIATE;
else if (flags & WINEDDFLIP_INTERVAL2)
swapchain->presentParms.PresentationInterval = WINED3DPRESENT_INTERVAL_TWO;
else if (flags & WINEDDFLIP_INTERVAL3)
swapchain->presentParms.PresentationInterval = WINED3DPRESENT_INTERVAL_THREE;
else
swapchain->presentParms.PresentationInterval = WINED3DPRESENT_INTERVAL_FOUR;
return wined3d_swapchain_present(swapchain, NULL, NULL, swapchain->win_handle, NULL, 0);
}
/* Do not call while under the GL lock. */
void surface_internal_preload(struct wined3d_surface *surface, enum WINED3DSRGB srgb)
{
IWineD3DDeviceImpl *device = surface->resource.device;
TRACE("iface %p, srgb %#x.\n", surface, srgb);
if (surface->container.type == WINED3D_CONTAINER_TEXTURE)
{
struct wined3d_texture *texture = surface->container.u.texture;
TRACE("Passing to container (%p).\n", texture);
texture->texture_ops->texture_preload(texture, srgb);
}
else
{
struct wined3d_context *context = NULL;
TRACE("(%p) : About to load surface\n", surface);
if (!device->isInDraw) context = context_acquire(device, NULL);
if (surface->resource.format->id == WINED3DFMT_P8_UINT
|| surface->resource.format->id == WINED3DFMT_P8_UINT_A8_UNORM)
{
if (palette9_changed(surface))
{
TRACE("Reloading surface because the d3d8/9 palette was changed\n");
/* TODO: This is not necessarily needed with hw palettized texture support */
surface_load_location(surface, SFLAG_INSYSMEM, NULL);
/* Make sure the texture is reloaded because of the palette change, this kills performance though :( */
surface_modify_location(surface, SFLAG_INTEXTURE, FALSE);
}
}
surface_load(surface, srgb == SRGB_SRGB ? TRUE : FALSE);
if (surface->resource.pool == WINED3DPOOL_DEFAULT)
{
/* Tell opengl to try and keep this texture in video ram (well mostly) */
GLclampf tmp;
tmp = 0.9f;
ENTER_GL();
glPrioritizeTextures(1, &surface->texture_name, &tmp);
LEAVE_GL();
}
if (context) context_release(context);
}
}
BOOL surface_init_sysmem(struct wined3d_surface *surface)
{
if (!surface->resource.allocatedMemory)
{
surface->resource.heapMemory = HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY,
surface->resource.size + RESOURCE_ALIGNMENT);
if (!surface->resource.heapMemory)
{
ERR("Out of memory\n");
return FALSE;
}
surface->resource.allocatedMemory =
(BYTE *)(((ULONG_PTR)surface->resource.heapMemory + (RESOURCE_ALIGNMENT - 1)) & ~(RESOURCE_ALIGNMENT - 1));
}
else
{
memset(surface->resource.allocatedMemory, 0, surface->resource.size);
}
surface_modify_location(surface, SFLAG_INSYSMEM, TRUE);
return TRUE;
}
/* Read the framebuffer back into the surface */
static void read_from_framebuffer(struct wined3d_surface *surface, const RECT *rect, void *dest, UINT pitch)
{
IWineD3DDeviceImpl *device = surface->resource.device;
const struct wined3d_gl_info *gl_info;
struct wined3d_context *context;
BYTE *mem;
GLint fmt;
GLint type;
BYTE *row, *top, *bottom;
int i;
BOOL bpp;
RECT local_rect;
BOOL srcIsUpsideDown;
GLint rowLen = 0;
GLint skipPix = 0;
GLint skipRow = 0;
if(wined3d_settings.rendertargetlock_mode == RTL_DISABLE) {
static BOOL warned = FALSE;
if(!warned) {
ERR("The application tries to lock the render target, but render target locking is disabled\n");
warned = TRUE;
}
return;
}
context = context_acquire(device, surface);
context_apply_blit_state(context, device);
gl_info = context->gl_info;
ENTER_GL();
/* Select the correct read buffer, and give some debug output.
* There is no need to keep track of the current read buffer or reset it, every part of the code
* that reads sets the read buffer as desired.
*/
if (surface_is_offscreen(surface))
{
/* Mapping the primary render target which is not on a swapchain.
* Read from the back buffer. */
TRACE("Mapping offscreen render target.\n");
glReadBuffer(device->offscreenBuffer);
srcIsUpsideDown = TRUE;
}
else
{
/* Onscreen surfaces are always part of a swapchain */
GLenum buffer = surface_get_gl_buffer(surface);
TRACE("Mapping %#x buffer.\n", buffer);
glReadBuffer(buffer);
checkGLcall("glReadBuffer");
srcIsUpsideDown = FALSE;
}
/* TODO: Get rid of the extra rectangle comparison and construction of a full surface rectangle */
if (!rect)
{
local_rect.left = 0;
local_rect.top = 0;
local_rect.right = surface->resource.width;
local_rect.bottom = surface->resource.height;
}
else
{
local_rect = *rect;
}
/* TODO: Get rid of the extra GetPitch call, LockRect does that too. Cache the pitch */
switch (surface->resource.format->id)
{
case WINED3DFMT_P8_UINT:
{
if (primary_render_target_is_p8(device))
{
/* In case of P8 render targets the index is stored in the alpha component */
fmt = GL_ALPHA;
type = GL_UNSIGNED_BYTE;
mem = dest;
bpp = surface->resource.format->byte_count;
}
else
{
/* GL can't return palettized data, so read ARGB pixels into a
* separate block of memory and convert them into palettized format
* in software. Slow, but if the app means to use palettized render
* targets and locks it...
*
* Use GL_RGB, GL_UNSIGNED_BYTE to read the surface for performance reasons
* Don't use GL_BGR as in the WINED3DFMT_R8G8B8 case, instead watch out
* for the color channels when palettizing the colors.
*/
fmt = GL_RGB;
type = GL_UNSIGNED_BYTE;
pitch *= 3;
mem = HeapAlloc(GetProcessHeap(), 0, surface->resource.size * 3);
if (!mem)
{
ERR("Out of memory\n");
LEAVE_GL();
return;
}
bpp = surface->resource.format->byte_count * 3;
}
}
break;
default:
mem = dest;
fmt = surface->resource.format->glFormat;
type = surface->resource.format->glType;
bpp = surface->resource.format->byte_count;
}
if (surface->flags & SFLAG_PBO)
{
GL_EXTCALL(glBindBufferARB(GL_PIXEL_PACK_BUFFER_ARB, surface->pbo));
checkGLcall("glBindBufferARB");
if (mem)
{
ERR("mem not null for pbo -- unexpected\n");
mem = NULL;
}
}
/* Save old pixel store pack state */
glGetIntegerv(GL_PACK_ROW_LENGTH, &rowLen);
checkGLcall("glGetIntegerv");
glGetIntegerv(GL_PACK_SKIP_PIXELS, &skipPix);
checkGLcall("glGetIntegerv");
glGetIntegerv(GL_PACK_SKIP_ROWS, &skipRow);
checkGLcall("glGetIntegerv");
/* Setup pixel store pack state -- to glReadPixels into the correct place */
glPixelStorei(GL_PACK_ROW_LENGTH, surface->resource.width);
checkGLcall("glPixelStorei");
glPixelStorei(GL_PACK_SKIP_PIXELS, local_rect.left);
checkGLcall("glPixelStorei");
glPixelStorei(GL_PACK_SKIP_ROWS, local_rect.top);
checkGLcall("glPixelStorei");
glReadPixels(local_rect.left, !srcIsUpsideDown ? (surface->resource.height - local_rect.bottom) : local_rect.top,
local_rect.right - local_rect.left,
local_rect.bottom - local_rect.top,
fmt, type, mem);
checkGLcall("glReadPixels");
/* Reset previous pixel store pack state */
glPixelStorei(GL_PACK_ROW_LENGTH, rowLen);
checkGLcall("glPixelStorei");
glPixelStorei(GL_PACK_SKIP_PIXELS, skipPix);
checkGLcall("glPixelStorei");
glPixelStorei(GL_PACK_SKIP_ROWS, skipRow);
checkGLcall("glPixelStorei");
if (surface->flags & SFLAG_PBO)
{
GL_EXTCALL(glBindBufferARB(GL_PIXEL_PACK_BUFFER_ARB, 0));
checkGLcall("glBindBufferARB");
/* Check if we need to flip the image. If we need to flip use glMapBufferARB
* to get a pointer to it and perform the flipping in software. This is a lot
* faster than calling glReadPixels for each line. In case we want more speed
* we should rerender it flipped in a FBO and read the data back from the FBO. */
if (!srcIsUpsideDown)
{
GL_EXTCALL(glBindBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB, surface->pbo));
checkGLcall("glBindBufferARB");
mem = GL_EXTCALL(glMapBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB, GL_READ_WRITE_ARB));
checkGLcall("glMapBufferARB");
}
}
/* TODO: Merge this with the palettization loop below for P8 targets */
if(!srcIsUpsideDown) {
UINT len, off;
/* glReadPixels returns the image upside down, and there is no way to prevent this.
Flip the lines in software */
len = (local_rect.right - local_rect.left) * bpp;
off = local_rect.left * bpp;
row = HeapAlloc(GetProcessHeap(), 0, len);
if(!row) {
ERR("Out of memory\n");
if (surface->resource.format->id == WINED3DFMT_P8_UINT)
HeapFree(GetProcessHeap(), 0, mem);
LEAVE_GL();
return;
}
top = mem + pitch * local_rect.top;
bottom = mem + pitch * (local_rect.bottom - 1);
for(i = 0; i < (local_rect.bottom - local_rect.top) / 2; i++) {
memcpy(row, top + off, len);
memcpy(top + off, bottom + off, len);
memcpy(bottom + off, row, len);
top += pitch;
bottom -= pitch;
}
HeapFree(GetProcessHeap(), 0, row);
/* Unmap the temp PBO buffer */
if (surface->flags & SFLAG_PBO)
{
GL_EXTCALL(glUnmapBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB));
GL_EXTCALL(glBindBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB, 0));
}
}
LEAVE_GL();
context_release(context);
/* For P8 textures we need to perform an inverse palette lookup. This is
* done by searching for a palette index which matches the RGB value.
* Note this isn't guaranteed to work when there are multiple entries for
* the same color but we have no choice. In case of P8 render targets,
* the index is stored in the alpha component so no conversion is needed. */
if (surface->resource.format->id == WINED3DFMT_P8_UINT && !primary_render_target_is_p8(device))
{
const PALETTEENTRY *pal = NULL;
DWORD width = pitch / 3;
int x, y, c;
if (surface->palette)
{
pal = surface->palette->palents;
}
else
{
ERR("Palette is missing, cannot perform inverse palette lookup\n");
HeapFree(GetProcessHeap(), 0, mem);
return;
}
for(y = local_rect.top; y < local_rect.bottom; y++) {
for(x = local_rect.left; x < local_rect.right; x++) {
/* start lines pixels */
const BYTE *blue = mem + y * pitch + x * (sizeof(BYTE) * 3);
const BYTE *green = blue + 1;
const BYTE *red = green + 1;
for(c = 0; c < 256; c++) {
if(*red == pal[c].peRed &&
*green == pal[c].peGreen &&
*blue == pal[c].peBlue)
{
*((BYTE *) dest + y * width + x) = c;
break;
}
}
}
}
HeapFree(GetProcessHeap(), 0, mem);
}
}
/* Read the framebuffer contents into a texture */
static void read_from_framebuffer_texture(struct wined3d_surface *surface, BOOL srgb)
{
IWineD3DDeviceImpl *device = surface->resource.device;
const struct wined3d_gl_info *gl_info;
struct wined3d_context *context;
if (!surface_is_offscreen(surface))
{
/* We would need to flip onscreen surfaces, but there's no efficient
* way to do that here. It makes more sense for the caller to
* explicitly go through sysmem. */
ERR("Not supported for onscreen targets.\n");
return;
}
/* Activate the surface to read from. In some situations it isn't the currently active target(e.g. backbuffer
* locking during offscreen rendering). RESOURCELOAD is ok because glCopyTexSubImage2D isn't affected by any
* states in the stateblock, and no driver was found yet that had bugs in that regard.
*/
context = context_acquire(device, surface);
gl_info = context->gl_info;
surface_prepare_texture(surface, gl_info, srgb);
surface_bind_and_dirtify(surface, gl_info, srgb);
TRACE("Reading back offscreen render target %p.\n", surface);
ENTER_GL();
glReadBuffer(device->offscreenBuffer);
checkGLcall("glReadBuffer");
glCopyTexSubImage2D(surface->texture_target, surface->texture_level,
0, 0, 0, 0, surface->resource.width, surface->resource.height);
checkGLcall("glCopyTexSubImage2D");
LEAVE_GL();
context_release(context);
}
/* Context activation is done by the caller. */
static void surface_prepare_texture_internal(struct wined3d_surface *surface,
const struct wined3d_gl_info *gl_info, BOOL srgb)
{
DWORD alloc_flag = srgb ? SFLAG_SRGBALLOCATED : SFLAG_ALLOCATED;
CONVERT_TYPES convert;
struct wined3d_format format;
if (surface->flags & alloc_flag) return;
d3dfmt_get_conv(surface, TRUE, TRUE, &format, &convert);
if (convert != NO_CONVERSION || format.convert) surface->flags |= SFLAG_CONVERTED;
else surface->flags &= ~SFLAG_CONVERTED;
surface_bind_and_dirtify(surface, gl_info, srgb);
surface_allocate_surface(surface, gl_info, &format, srgb);
surface->flags |= alloc_flag;
}
/* Context activation is done by the caller. */
void surface_prepare_texture(struct wined3d_surface *surface, const struct wined3d_gl_info *gl_info, BOOL srgb)
{
if (surface->container.type == WINED3D_CONTAINER_TEXTURE)
{
struct wined3d_texture *texture = surface->container.u.texture;
UINT sub_count = texture->level_count * texture->layer_count;
UINT i;
TRACE("surface %p is a subresource of texture %p.\n", surface, texture);
for (i = 0; i < sub_count; ++i)
{
struct wined3d_surface *s = surface_from_resource(texture->sub_resources[i]);
surface_prepare_texture_internal(s, gl_info, srgb);
}
return;
}
surface_prepare_texture_internal(surface, gl_info, srgb);
}
static void flush_to_framebuffer_drawpixels(struct wined3d_surface *surface,
const RECT *rect, GLenum fmt, GLenum type, UINT bpp, const BYTE *mem)
{
IWineD3DDeviceImpl *device = surface->resource.device;
UINT pitch = wined3d_surface_get_pitch(surface);
const struct wined3d_gl_info *gl_info;
struct wined3d_context *context;
RECT local_rect;
UINT w, h;
surface_get_rect(surface, rect, &local_rect);
mem += local_rect.top * pitch + local_rect.left * bpp;
w = local_rect.right - local_rect.left;
h = local_rect.bottom - local_rect.top;
/* Activate the correct context for the render target */
context = context_acquire(device, surface);
context_apply_blit_state(context, device);
gl_info = context->gl_info;
ENTER_GL();
if (!surface_is_offscreen(surface))
{
GLenum buffer = surface_get_gl_buffer(surface);
TRACE("Unlocking %#x buffer.\n", buffer);
context_set_draw_buffer(context, buffer);
surface_translate_drawable_coords(surface, context->win_handle, &local_rect);
glPixelZoom(1.0f, -1.0f);
}
else
{
/* Primary offscreen render target */
TRACE("Offscreen render target.\n");
context_set_draw_buffer(context, device->offscreenBuffer);
glPixelZoom(1.0f, 1.0f);
}
glRasterPos3i(local_rect.left, local_rect.top, 1);
checkGLcall("glRasterPos3i");
/* If not fullscreen, we need to skip a number of bytes to find the next row of data */
glPixelStorei(GL_UNPACK_ROW_LENGTH, surface->resource.width);
if (surface->flags & SFLAG_PBO)
{
GL_EXTCALL(glBindBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB, surface->pbo));
checkGLcall("glBindBufferARB");
}
glDrawPixels(w, h, fmt, type, mem);
checkGLcall("glDrawPixels");
if (surface->flags & SFLAG_PBO)
{
GL_EXTCALL(glBindBufferARB(GL_PIXEL_UNPACK_BUFFER_ARB, 0));
checkGLcall("glBindBufferARB");
}
glPixelStorei(GL_UNPACK_ROW_LENGTH, 0);
checkGLcall("glPixelStorei(GL_UNPACK_ROW_LENGTH, 0)");
LEAVE_GL();
if (wined3d_settings.strict_draw_ordering
|| (surface->container.type == WINED3D_CONTAINER_SWAPCHAIN
&& surface->container.u.swapchain->front_buffer == surface))
wglFlush();
context_release(context);
}
HRESULT d3dfmt_get_conv(struct wined3d_surface *surface, BOOL need_alpha_ck,
BOOL use_texturing, struct wined3d_format *format, CONVERT_TYPES *convert)
{
BOOL colorkey_active = need_alpha_ck && (surface->CKeyFlags & WINEDDSD_CKSRCBLT);
IWineD3DDeviceImpl *device = surface->resource.device;
const struct wined3d_gl_info *gl_info = &device->adapter->gl_info;
BOOL blit_supported = FALSE;
/* Copy the default values from the surface. Below we might perform fixups */
/* TODO: get rid of color keying desc fixups by using e.g. a table. */
*format = *surface->resource.format;
*convert = NO_CONVERSION;
/* Ok, now look if we have to do any conversion */
switch (surface->resource.format->id)
{
case WINED3DFMT_P8_UINT:
/* Below the call to blit_supported is disabled for Wine 1.2
* because the function isn't operating correctly yet. At the
* moment 8-bit blits are handled in software and if certain GL
* extensions are around, surface conversion is performed at
* upload time. The blit_supported call recognizes it as a
* destination fixup. This type of upload 'fixup' and 8-bit to
* 8-bit blits need to be handled by the blit_shader.
* TODO: get rid of this #if 0. */
#if 0
blit_supported = device->blitter->blit_supported(&device->adapter->gl_info, WINED3D_BLIT_OP_COLOR_BLIT,
&rect, surface->resource.usage, surface->resource.pool, surface->resource.format,
&rect, surface->resource.usage, surface->resource.pool, surface->resource.format);
#endif
blit_supported = gl_info->supported[EXT_PALETTED_TEXTURE] || gl_info->supported[ARB_FRAGMENT_PROGRAM];
/* Use conversion when the blit_shader backend supports it. It only supports this in case of
* texturing. Further also use conversion in case of color keying.
* Paletted textures can be emulated using shaders but only do that for 2D purposes e.g. situations
* in which the main render target uses p8. Some games like GTA Vice City use P8 for texturing which
* conflicts with this.
*/
if (!((blit_supported && device->render_targets && surface == device->render_targets[0]))
|| colorkey_active || !use_texturing)
{
format->glFormat = GL_RGBA;
format->glInternal = GL_RGBA;
format->glType = GL_UNSIGNED_BYTE;
format->conv_byte_count = 4;
if (colorkey_active)
*convert = CONVERT_PALETTED_CK;
else
*convert = CONVERT_PALETTED;
}
break;
case WINED3DFMT_B2G3R3_UNORM:
/* **********************
GL_UNSIGNED_BYTE_3_3_2
********************** */
if (colorkey_active) {
/* This texture format will never be used.. So do not care about color keying
up until the point in time it will be needed :-) */
FIXME(" ColorKeying not supported in the RGB 332 format !\n");
}
break;
case WINED3DFMT_B5G6R5_UNORM:
if (colorkey_active)
{
*convert = CONVERT_CK_565;
format->glFormat = GL_RGBA;
format->glInternal = GL_RGB5_A1;
format->glType = GL_UNSIGNED_SHORT_5_5_5_1;
format->conv_byte_count = 2;
}
break;
case WINED3DFMT_B5G5R5X1_UNORM:
if (colorkey_active)
{
*convert = CONVERT_CK_5551;
format->glFormat = GL_BGRA;
format->glInternal = GL_RGB5_A1;
format->glType = GL_UNSIGNED_SHORT_1_5_5_5_REV;
format->conv_byte_count = 2;
}
break;
case WINED3DFMT_B8G8R8_UNORM:
if (colorkey_active)
{
*convert = CONVERT_CK_RGB24;
format->glFormat = GL_RGBA;
format->glInternal = GL_RGBA8;
format->glType = GL_UNSIGNED_INT_8_8_8_8;
format->conv_byte_count = 4;
}
break;
case WINED3DFMT_B8G8R8X8_UNORM:
if (colorkey_active)
{
*convert = CONVERT_RGB32_888;
format->glFormat = GL_RGBA;
format->glInternal = GL_RGBA8;
format->glType = GL_UNSIGNED_INT_8_8_8_8;
format->conv_byte_count = 4;
}
break;
default:
break;
}
return WINED3D_OK;
}
void d3dfmt_p8_init_palette(struct wined3d_surface *surface, BYTE table[256][4], BOOL colorkey)
{
IWineD3DDeviceImpl *device = surface->resource.device;
struct wined3d_palette *pal = surface->palette;
BOOL index_in_alpha = FALSE;
unsigned int i;
/* Old games like StarCraft, C&C, Red Alert and others use P8 render targets.
* Reading back the RGB output each lockrect (each frame as they lock the whole screen)
* is slow. Further RGB->P8 conversion is not possible because palettes can have
* duplicate entries. Store the color key in the unused alpha component to speed the
* download up and to make conversion unneeded. */
index_in_alpha = primary_render_target_is_p8(device);
if (!pal)
{
UINT dxVersion = device->wined3d->dxVersion;
/* In DirectDraw the palette is a property of the surface, there are no such things as device palettes. */
if (dxVersion <= 7)
{
ERR("This code should never get entered for DirectDraw!, expect problems\n");
if (index_in_alpha)
{
/* Guarantees that memory representation remains correct after sysmem<->texture transfers even if
* there's no palette at this time. */
for (i = 0; i < 256; i++) table[i][3] = i;
}
}
else
{
/* Direct3D >= 8 palette usage style: P8 textures use device palettes, palette entry format is A8R8G8B8,
* alpha is stored in peFlags and may be used by the app if D3DPTEXTURECAPS_ALPHAPALETTE device
* capability flag is present (wine does advertise this capability) */
for (i = 0; i < 256; ++i)
{
table[i][0] = device->palettes[device->currentPalette][i].peRed;
table[i][1] = device->palettes[device->currentPalette][i].peGreen;
table[i][2] = device->palettes[device->currentPalette][i].peBlue;
table[i][3] = device->palettes[device->currentPalette][i].peFlags;
}
}
}
else
{
TRACE("Using surface palette %p\n", pal);
/* Get the surface's palette */
for (i = 0; i < 256; ++i)
{
table[i][0] = pal->palents[i].peRed;
table[i][1] = pal->palents[i].peGreen;
table[i][2] = pal->palents[i].peBlue;
/* When index_in_alpha is set the palette index is stored in the
* alpha component. In case of a readback we can then read
* GL_ALPHA. Color keying is handled in BltOverride using a
* GL_ALPHA_TEST using GL_NOT_EQUAL. In case of index_in_alpha the
* color key itself is passed to glAlphaFunc in other cases the
* alpha component of pixels that should be masked away is set to 0. */
if (index_in_alpha)
{
table[i][3] = i;
}
else if (colorkey && (i >= surface->SrcBltCKey.dwColorSpaceLowValue)
&& (i <= surface->SrcBltCKey.dwColorSpaceHighValue))
{
table[i][3] = 0x00;
}
else if (pal->flags & WINEDDPCAPS_ALPHA)
{
table[i][3] = pal->palents[i].peFlags;
}
else
{
table[i][3] = 0xFF;
}
}
}
}
static HRESULT d3dfmt_convert_surface(const BYTE *src, BYTE *dst, UINT pitch, UINT width,
UINT height, UINT outpitch, CONVERT_TYPES convert, struct wined3d_surface *surface)
{
const BYTE *source;
BYTE *dest;
TRACE("(%p)->(%p),(%d,%d,%d,%d,%p)\n", src, dst, pitch, height, outpitch, convert, surface);
switch (convert) {
case NO_CONVERSION:
{
memcpy(dst, src, pitch * height);
break;
}
case CONVERT_PALETTED:
case CONVERT_PALETTED_CK:
{
BYTE table[256][4];
unsigned int x, y;
d3dfmt_p8_init_palette(surface, table, (convert == CONVERT_PALETTED_CK));
for (y = 0; y < height; y++)
{
source = src + pitch * y;
dest = dst + outpitch * y;
/* This is an 1 bpp format, using the width here is fine */
for (x = 0; x < width; x++) {
BYTE color = *source++;
*dest++ = table[color][0];
*dest++ = table[color][1];
*dest++ = table[color][2];
*dest++ = table[color][3];
}
}
}
break;
case CONVERT_CK_565:
{
/* Converting the 565 format in 5551 packed to emulate color-keying.
Note : in all these conversion, it would be best to average the averaging
pixels to get the color of the pixel that will be color-keyed to
prevent 'color bleeding'. This will be done later on if ever it is
too visible.
Note2: Nvidia documents say that their driver does not support alpha + color keying
on the same surface and disables color keying in such a case
*/
unsigned int x, y;
const WORD *Source;
WORD *Dest;
TRACE("Color keyed 565\n");
for (y = 0; y < height; y++) {
Source = (const WORD *)(src + y * pitch);
Dest = (WORD *) (dst + y * outpitch);
for (x = 0; x < width; x++ ) {
WORD color = *Source++;
*Dest = ((color & 0xFFC0) | ((color & 0x1F) << 1));
if ((color < surface->SrcBltCKey.dwColorSpaceLowValue)
|| (color > surface->SrcBltCKey.dwColorSpaceHighValue))
*Dest |= 0x0001;
Dest++;
}
}
}
break;
case CONVERT_CK_5551:
{
/* Converting X1R5G5B5 format to R5G5B5A1 to emulate color-keying. */
unsigned int x, y;
const WORD *Source;
WORD *Dest;
TRACE("Color keyed 5551\n");
for (y = 0; y < height; y++) {
Source = (const WORD *)(src + y * pitch);
Dest = (WORD *) (dst + y * outpitch);
for (x = 0; x < width; x++ ) {
WORD color = *Source++;
*Dest = color;
if ((color < surface->SrcBltCKey.dwColorSpaceLowValue)
|| (color > surface->SrcBltCKey.dwColorSpaceHighValue))
*Dest |= (1 << 15);
else
*Dest &= ~(1 << 15);
Dest++;
}
}
}
break;
case CONVERT_CK_RGB24:
{
/* Converting R8G8B8 format to R8G8B8A8 with color-keying. */
unsigned int x, y;
for (y = 0; y < height; y++)
{
source = src + pitch * y;
dest = dst + outpitch * y;
for (x = 0; x < width; x++) {
DWORD color = ((DWORD)source[0] << 16) + ((DWORD)source[1] << 8) + (DWORD)source[2] ;
DWORD dstcolor = color << 8;
if ((color < surface->SrcBltCKey.dwColorSpaceLowValue)
|| (color > surface->SrcBltCKey.dwColorSpaceHighValue))
dstcolor |= 0xff;
*(DWORD*)dest = dstcolor;
source += 3;
dest += 4;
}
}
}
break;
case CONVERT_RGB32_888:
{
/* Converting X8R8G8B8 format to R8G8B8A8 with color-keying. */
unsigned int x, y;
for (y = 0; y < height; y++)
{
source = src + pitch * y;
dest = dst + outpitch * y;
for (x = 0; x < width; x++) {
DWORD color = 0xffffff & *(const DWORD*)source;
DWORD dstcolor = color << 8;
if ((color < surface->SrcBltCKey.dwColorSpaceLowValue)
|| (color > surface->SrcBltCKey.dwColorSpaceHighValue))
dstcolor |= 0xff;
*(DWORD*)dest = dstcolor;
source += 4;
dest += 4;
}
}
}
break;
default:
ERR("Unsupported conversion type %#x.\n", convert);
}
return WINED3D_OK;
}
BOOL palette9_changed(struct wined3d_surface *surface)
{
IWineD3DDeviceImpl *device = surface->resource.device;
if (surface->palette || (surface->resource.format->id != WINED3DFMT_P8_UINT
&& surface->resource.format->id != WINED3DFMT_P8_UINT_A8_UNORM))
{
/* If a ddraw-style palette is attached assume no d3d9 palette change.
* Also the palette isn't interesting if the surface format isn't P8 or A8P8
*/
return FALSE;
}
if (surface->palette9)
{
if (!memcmp(surface->palette9, device->palettes[device->currentPalette], sizeof(PALETTEENTRY) * 256))
{
return FALSE;
}
}
else
{
surface->palette9 = HeapAlloc(GetProcessHeap(), 0, sizeof(PALETTEENTRY) * 256);
}
memcpy(surface->palette9, device->palettes[device->currentPalette], sizeof(PALETTEENTRY) * 256);
return TRUE;
}
void flip_surface(struct wined3d_surface *front, struct wined3d_surface *back)
{
/* Flip the surface contents */
/* Flip the DC */
{
HDC tmp;
tmp = front->hDC;
front->hDC = back->hDC;
back->hDC = tmp;
}
/* Flip the DIBsection */
{
HBITMAP tmp;
BOOL hasDib = front->flags & SFLAG_DIBSECTION;
tmp = front->dib.DIBsection;
front->dib.DIBsection = back->dib.DIBsection;
back->dib.DIBsection = tmp;
if (back->flags & SFLAG_DIBSECTION) front->flags |= SFLAG_DIBSECTION;
else front->flags &= ~SFLAG_DIBSECTION;
if (hasDib) back->flags |= SFLAG_DIBSECTION;
else back->flags &= ~SFLAG_DIBSECTION;
}
/* Flip the surface data */
{
void* tmp;
tmp = front->dib.bitmap_data;
front->dib.bitmap_data = back->dib.bitmap_data;
back->dib.bitmap_data = tmp;
tmp = front->resource.allocatedMemory;
front->resource.allocatedMemory = back->resource.allocatedMemory;
back->resource.allocatedMemory = tmp;
tmp = front->resource.heapMemory;
front->resource.heapMemory = back->resource.heapMemory;
back->resource.heapMemory = tmp;
}
/* Flip the PBO */
{
GLuint tmp_pbo = front->pbo;
front->pbo = back->pbo;
back->pbo = tmp_pbo;
}
/* client_memory should not be different, but just in case */
{
BOOL tmp;
tmp = front->dib.client_memory;
front->dib.client_memory = back->dib.client_memory;
back->dib.client_memory = tmp;
}
/* Flip the opengl texture */
{
GLuint tmp;
tmp = back->texture_name;
back->texture_name = front->texture_name;
front->texture_name = tmp;
tmp = back->texture_name_srgb;
back->texture_name_srgb = front->texture_name_srgb;
front->texture_name_srgb = tmp;
resource_unload(&back->resource);
resource_unload(&front->resource);
}
{
DWORD tmp_flags = back->flags;
back->flags = front->flags;
front->flags = tmp_flags;
}
}
/* Does a direct frame buffer -> texture copy. Stretching is done with single
* pixel copy calls. */
static void fb_copy_to_texture_direct(struct wined3d_surface *dst_surface, struct wined3d_surface *src_surface,
const RECT *src_rect, const RECT *dst_rect_in, WINED3DTEXTUREFILTERTYPE Filter)
{
IWineD3DDeviceImpl *device = dst_surface->resource.device;
float xrel, yrel;
UINT row;
struct wined3d_context *context;
BOOL upsidedown = FALSE;
RECT dst_rect = *dst_rect_in;
/* Make sure that the top pixel is always above the bottom pixel, and keep a separate upside down flag
* glCopyTexSubImage is a bit picky about the parameters we pass to it
*/
if(dst_rect.top > dst_rect.bottom) {
UINT tmp = dst_rect.bottom;
dst_rect.bottom = dst_rect.top;
dst_rect.top = tmp;
upsidedown = TRUE;
}
context = context_acquire(device, src_surface);
context_apply_blit_state(context, device);
surface_internal_preload(dst_surface, SRGB_RGB);
ENTER_GL();
/* Bind the target texture */
glBindTexture(dst_surface->texture_target, dst_surface->texture_name);
checkGLcall("glBindTexture");
if (surface_is_offscreen(src_surface))
{
TRACE("Reading from an offscreen target\n");
upsidedown = !upsidedown;
glReadBuffer(device->offscreenBuffer);
}
else
{
glReadBuffer(surface_get_gl_buffer(src_surface));
}
checkGLcall("glReadBuffer");
xrel = (float) (src_rect->right - src_rect->left) / (float) (dst_rect.right - dst_rect.left);
yrel = (float) (src_rect->bottom - src_rect->top) / (float) (dst_rect.bottom - dst_rect.top);
if ((xrel - 1.0f < -eps) || (xrel - 1.0f > eps))
{
FIXME("Doing a pixel by pixel copy from the framebuffer to a texture, expect major performance issues\n");
if(Filter != WINED3DTEXF_NONE && Filter != WINED3DTEXF_POINT) {
ERR("Texture filtering not supported in direct blit\n");
}
}
else if ((Filter != WINED3DTEXF_NONE && Filter != WINED3DTEXF_POINT)
&& ((yrel - 1.0f < -eps) || (yrel - 1.0f > eps)))
{
ERR("Texture filtering not supported in direct blit\n");
}
if (upsidedown
&& !((xrel - 1.0f < -eps) || (xrel - 1.0f > eps))
&& !((yrel - 1.0f < -eps) || (yrel - 1.0f > eps)))
{
/* Upside down copy without stretching is nice, one glCopyTexSubImage call will do */
glCopyTexSubImage2D(dst_surface->texture_target, dst_surface->texture_level,
dst_rect.left /*xoffset */, dst_rect.top /* y offset */,
src_rect->left, src_surface->resource.height - src_rect->bottom,
dst_rect.right - dst_rect.left, dst_rect.bottom - dst_rect.top);
}
else
{
UINT yoffset = src_surface->resource.height - src_rect->top + dst_rect.top - 1;
/* I have to process this row by row to swap the image,
* otherwise it would be upside down, so stretching in y direction
* doesn't cost extra time
*
* However, stretching in x direction can be avoided if not necessary
*/
for(row = dst_rect.top; row < dst_rect.bottom; row++) {
if ((xrel - 1.0f < -eps) || (xrel - 1.0f > eps))
{
/* Well, that stuff works, but it's very slow.
* find a better way instead
*/
UINT col;
for (col = dst_rect.left; col < dst_rect.right; ++col)
{
glCopyTexSubImage2D(dst_surface->texture_target, dst_surface->texture_level,
dst_rect.left + col /* x offset */, row /* y offset */,
src_rect->left + col * xrel, yoffset - (int) (row * yrel), 1, 1);
}
}
else
{
glCopyTexSubImage2D(dst_surface->texture_target, dst_surface->texture_level,
dst_rect.left /* x offset */, row /* y offset */,
src_rect->left, yoffset - (int) (row * yrel), dst_rect.right - dst_rect.left, 1);
}
}
}
checkGLcall("glCopyTexSubImage2D");
LEAVE_GL();
context_release(context);
/* The texture is now most up to date - If the surface is a render target and has a drawable, this
* path is never entered
*/
surface_modify_location(dst_surface, SFLAG_INTEXTURE, TRUE);
}
/* Uses the hardware to stretch and flip the image */
static void fb_copy_to_texture_hwstretch(struct wined3d_surface *dst_surface, struct wined3d_surface *src_surface,
const RECT *src_rect, const RECT *dst_rect_in, WINED3DTEXTUREFILTERTYPE Filter)
{
IWineD3DDeviceImpl *device = dst_surface->resource.device;
struct wined3d_swapchain *src_swapchain = NULL;
GLuint src, backup = 0;
float left, right, top, bottom; /* Texture coordinates */
UINT fbwidth = src_surface->resource.width;
UINT fbheight = src_surface->resource.height;
struct wined3d_context *context;
GLenum drawBuffer = GL_BACK;
GLenum texture_target;
BOOL noBackBufferBackup;
BOOL src_offscreen;
BOOL upsidedown = FALSE;
RECT dst_rect = *dst_rect_in;
TRACE("Using hwstretch blit\n");
/* Activate the Proper context for reading from the source surface, set it up for blitting */
context = context_acquire(device, src_surface);
context_apply_blit_state(context, device);
surface_internal_preload(dst_surface, SRGB_RGB);
src_offscreen = surface_is_offscreen(src_surface);
noBackBufferBackup = src_offscreen && wined3d_settings.offscreen_rendering_mode == ORM_FBO;
if (!noBackBufferBackup && !src_surface->texture_name)
{
/* Get it a description */
surface_internal_preload(src_surface, SRGB_RGB);
}
ENTER_GL();
/* Try to use an aux buffer for drawing the rectangle. This way it doesn't need restoring.
* This way we don't have to wait for the 2nd readback to finish to leave this function.
*/
if (context->aux_buffers >= 2)
{
/* Got more than one aux buffer? Use the 2nd aux buffer */
drawBuffer = GL_AUX1;
}
else if ((!src_offscreen || device->offscreenBuffer == GL_BACK) && context->aux_buffers >= 1)
{
/* Only one aux buffer, but it isn't used (Onscreen rendering, or non-aux orm)? Use it! */
drawBuffer = GL_AUX0;
}
if(noBackBufferBackup) {
glGenTextures(1, &backup);
checkGLcall("glGenTextures");
glBindTexture(GL_TEXTURE_2D, backup);
checkGLcall("glBindTexture(GL_TEXTURE_2D, backup)");
texture_target = GL_TEXTURE_2D;
} else {
/* Backup the back buffer and copy the source buffer into a texture to draw an upside down stretched quad. If
* we are reading from the back buffer, the backup can be used as source texture
*/
texture_target = src_surface->texture_target;
glBindTexture(texture_target, src_surface->texture_name);
checkGLcall("glBindTexture(texture_target, src_surface->texture_name)");
glEnable(texture_target);
checkGLcall("glEnable(texture_target)");
/* For now invalidate the texture copy of the back buffer. Drawable and sysmem copy are untouched */
src_surface->flags &= ~SFLAG_INTEXTURE;
}
/* Make sure that the top pixel is always above the bottom pixel, and keep a separate upside down flag
* glCopyTexSubImage is a bit picky about the parameters we pass to it
*/
if(dst_rect.top > dst_rect.bottom) {
UINT tmp = dst_rect.bottom;
dst_rect.bottom = dst_rect.top;
dst_rect.top = tmp;
upsidedown = TRUE;
}
if (src_offscreen)
{
TRACE("Reading from an offscreen target\n");
upsidedown = !upsidedown;
glReadBuffer(device->offscreenBuffer);
}
else
{
glReadBuffer(surface_get_gl_buffer(src_surface));
}
/* TODO: Only back up the part that will be overwritten */
glCopyTexSubImage2D(texture_target, 0,
0, 0 /* read offsets */,
0, 0,
fbwidth,
fbheight);
checkGLcall("glCopyTexSubImage2D");
/* No issue with overriding these - the sampler is dirty due to blit usage */
glTexParameteri(texture_target, GL_TEXTURE_MAG_FILTER,
wined3d_gl_mag_filter(magLookup, Filter));
checkGLcall("glTexParameteri");
glTexParameteri(texture_target, GL_TEXTURE_MIN_FILTER,
wined3d_gl_min_mip_filter(minMipLookup, Filter, WINED3DTEXF_NONE));
checkGLcall("glTexParameteri");
if (src_surface->container.type == WINED3D_CONTAINER_SWAPCHAIN)
src_swapchain = src_surface->container.u.swapchain;
if (!src_swapchain || src_surface == src_swapchain->back_buffers[0])
{
src = backup ? backup : src_surface->texture_name;
}
else
{
glReadBuffer(GL_FRONT);
checkGLcall("glReadBuffer(GL_FRONT)");
glGenTextures(1, &src);
checkGLcall("glGenTextures(1, &src)");
glBindTexture(GL_TEXTURE_2D, src);
checkGLcall("glBindTexture(GL_TEXTURE_2D, src)");
/* TODO: Only copy the part that will be read. Use src_rect->left, src_rect->bottom as origin, but with the width watch
* out for power of 2 sizes
*/
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA, src_surface->pow2Width,
src_surface->pow2Height, 0, GL_RGBA, GL_UNSIGNED_BYTE, NULL);
checkGLcall("glTexImage2D");
glCopyTexSubImage2D(GL_TEXTURE_2D, 0,
0, 0 /* read offsets */,
0, 0,
fbwidth,
fbheight);
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
checkGLcall("glTexParameteri");
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
checkGLcall("glTexParameteri");
glReadBuffer(GL_BACK);
checkGLcall("glReadBuffer(GL_BACK)");
if(texture_target != GL_TEXTURE_2D) {
glDisable(texture_target);
glEnable(GL_TEXTURE_2D);
texture_target = GL_TEXTURE_2D;
}
}
checkGLcall("glEnd and previous");
left = src_rect->left;
right = src_rect->right;
if (!upsidedown)
{
top = src_surface->resource.height - src_rect->top;
bottom = src_surface->resource.height - src_rect->bottom;
}
else
{
top = src_surface->resource.height - src_rect->bottom;
bottom = src_surface->resource.height - src_rect->top;
}
if (src_surface->flags & SFLAG_NORMCOORD)
{
left /= src_surface->pow2Width;
right /= src_surface->pow2Width;
top /= src_surface->pow2Height;
bottom /= src_surface->pow2Height;
}
/* draw the source texture stretched and upside down. The correct surface is bound already */
glTexParameteri(texture_target, GL_TEXTURE_WRAP_S, GL_CLAMP);
glTexParameteri(texture_target, GL_TEXTURE_WRAP_T, GL_CLAMP);
context_set_draw_buffer(context, drawBuffer);
glReadBuffer(drawBuffer);
glBegin(GL_QUADS);
/* bottom left */
glTexCoord2f(left, bottom);
glVertex2i(0, 0);
/* top left */
glTexCoord2f(left, top);
glVertex2i(0, dst_rect.bottom - dst_rect.top);
/* top right */
glTexCoord2f(right, top);
glVertex2i(dst_rect.right - dst_rect.left, dst_rect.bottom - dst_rect.top);
/* bottom right */
glTexCoord2f(right, bottom);
glVertex2i(dst_rect.right - dst_rect.left, 0);
glEnd();
checkGLcall("glEnd and previous");
if (texture_target != dst_surface->texture_target)
{
glDisable(texture_target);
glEnable(dst_surface->texture_target);
texture_target = dst_surface->texture_target;
}
/* Now read the stretched and upside down image into the destination texture */
glBindTexture(texture_target, dst_surface->texture_name);
checkGLcall("glBindTexture");
glCopyTexSubImage2D(texture_target,
0,
dst_rect.left, dst_rect.top, /* xoffset, yoffset */
0, 0, /* We blitted the image to the origin */
dst_rect.right - dst_rect.left, dst_rect.bottom - dst_rect.top);
checkGLcall("glCopyTexSubImage2D");
if(drawBuffer == GL_BACK) {
/* Write the back buffer backup back */
if(backup) {
if(texture_target != GL_TEXTURE_2D) {
glDisable(texture_target);
glEnable(GL_TEXTURE_2D);
texture_target = GL_TEXTURE_2D;
}
glBindTexture(GL_TEXTURE_2D, backup);
checkGLcall("glBindTexture(GL_TEXTURE_2D, backup)");
}
else
{
if (texture_target != src_surface->texture_target)
{
glDisable(texture_target);
glEnable(src_surface->texture_target);
texture_target = src_surface->texture_target;
}
glBindTexture(src_surface->texture_target, src_surface->texture_name);
checkGLcall("glBindTexture(src_surface->texture_target, src_surface->texture_name)");
}
glBegin(GL_QUADS);
/* top left */
glTexCoord2f(0.0f, 0.0f);
glVertex2i(0, fbheight);
/* bottom left */
glTexCoord2f(0.0f, (float)fbheight / (float)src_surface->pow2Height);
glVertex2i(0, 0);
/* bottom right */
glTexCoord2f((float)fbwidth / (float)src_surface->pow2Width,
(float)fbheight / (float)src_surface->pow2Height);
glVertex2i(fbwidth, 0);
/* top right */
glTexCoord2f((float)fbwidth / (float)src_surface->pow2Width, 0.0f);
glVertex2i(fbwidth, fbheight);
glEnd();
}
glDisable(texture_target);
checkGLcall("glDisable(texture_target)");
/* Cleanup */
if (src != src_surface->texture_name && src != backup)
{
glDeleteTextures(1, &src);
checkGLcall("glDeleteTextures(1, &src)");
}
if(backup) {
glDeleteTextures(1, &backup);
checkGLcall("glDeleteTextures(1, &backup)");
}
LEAVE_GL();
if (wined3d_settings.strict_draw_ordering) wglFlush(); /* Flush to ensure ordering across contexts. */
context_release(context);
/* The texture is now most up to date - If the surface is a render target and has a drawable, this
* path is never entered
*/
surface_modify_location(dst_surface, SFLAG_INTEXTURE, TRUE);
}
/* Front buffer coordinates are always full screen coordinates, but our GL
* drawable is limited to the window's client area. The sysmem and texture
* copies do have the full screen size. Note that GL has a bottom-left
* origin, while D3D has a top-left origin. */
void surface_translate_drawable_coords(struct wined3d_surface *surface, HWND window, RECT *rect)
{
UINT drawable_height;
if (surface->container.type == WINED3D_CONTAINER_SWAPCHAIN
&& surface == surface->container.u.swapchain->front_buffer)
{
POINT offset = {0, 0};
RECT windowsize;
ScreenToClient(window, &offset);
OffsetRect(rect, offset.x, offset.y);
GetClientRect(window, &windowsize);
drawable_height = windowsize.bottom - windowsize.top;
}
else
{
drawable_height = surface->resource.height;
}
rect->top = drawable_height - rect->top;
rect->bottom = drawable_height - rect->bottom;
}
/* blit between surface locations. onscreen on different swapchains is not supported.
* depth / stencil is not supported. */
static void surface_blt_fbo(IWineD3DDeviceImpl *device, const WINED3DTEXTUREFILTERTYPE filter,
struct wined3d_surface *src_surface, DWORD src_location, const RECT *src_rect_in,
struct wined3d_surface *dst_surface, DWORD dst_location, const RECT *dst_rect_in)
{
const struct wined3d_gl_info *gl_info;
struct wined3d_context *context;
RECT src_rect, dst_rect;
GLenum gl_filter;
TRACE("device %p, filter %s,\n", device, debug_d3dtexturefiltertype(filter));
TRACE("src_surface %p, src_location %s, src_rect %s,\n",
src_surface, debug_surflocation(src_location), wine_dbgstr_rect(src_rect_in));
TRACE("dst_surface %p, dst_location %s, dst_rect %s.\n",
dst_surface, debug_surflocation(dst_location), wine_dbgstr_rect(dst_rect_in));
src_rect = *src_rect_in;
dst_rect = *dst_rect_in;
switch (filter)
{
case WINED3DTEXF_LINEAR:
gl_filter = GL_LINEAR;
break;
default:
FIXME("Unsupported filter mode %s (%#x).\n", debug_d3dtexturefiltertype(filter), filter);
case WINED3DTEXF_NONE:
case WINED3DTEXF_POINT:
gl_filter = GL_NEAREST;
break;
}
if (src_location == SFLAG_INDRAWABLE && surface_is_offscreen(src_surface))
src_location = SFLAG_INTEXTURE;
if (dst_location == SFLAG_INDRAWABLE && surface_is_offscreen(dst_surface))
dst_location = SFLAG_INTEXTURE;
/* Make sure the locations are up-to-date. Loading the destination
* surface isn't required if the entire surface is overwritten. (And is
* in fact harmful if we're being called by surface_load_location() with
* the purpose of loading the destination surface.) */
surface_load_location(src_surface, src_location, NULL);
if (!surface_is_full_rect(dst_surface, &dst_rect))
surface_load_location(dst_surface, dst_location, NULL);
if (src_location == SFLAG_INDRAWABLE) context = context_acquire(device, src_surface);
else if (dst_location == SFLAG_INDRAWABLE) context = context_acquire(device, dst_surface);
else context = context_acquire(device, NULL);
if (!context->valid)
{
context_release(context);
WARN("Invalid context, skipping blit.\n");
return;
}
gl_info = context->gl_info;
if (src_location == SFLAG_INDRAWABLE)
{
GLenum buffer = surface_get_gl_buffer(src_surface);
TRACE("Source surface %p is onscreen.\n", src_surface);
surface_translate_drawable_coords(src_surface, context->win_handle, &src_rect);
ENTER_GL();
context_bind_fbo(context, GL_READ_FRAMEBUFFER, NULL);
glReadBuffer(buffer);
checkGLcall("glReadBuffer()");
}
else
{
TRACE("Source surface %p is offscreen.\n", src_surface);
ENTER_GL();
context_apply_fbo_state_blit(context, GL_READ_FRAMEBUFFER, src_surface, NULL, src_location);
glReadBuffer(GL_COLOR_ATTACHMENT0);
checkGLcall("glReadBuffer()");
}
context_check_fbo_status(context, GL_READ_FRAMEBUFFER);
LEAVE_GL();
if (dst_location == SFLAG_INDRAWABLE)
{
GLenum buffer = surface_get_gl_buffer(dst_surface);
TRACE("Destination surface %p is onscreen.\n", dst_surface);
surface_translate_drawable_coords(dst_surface, context->win_handle, &dst_rect);
ENTER_GL();
context_bind_fbo(context, GL_DRAW_FRAMEBUFFER, NULL);
context_set_draw_buffer(context, buffer);
}
else
{
TRACE("Destination surface %p is offscreen.\n", dst_surface);
ENTER_GL();
context_apply_fbo_state_blit(context, GL_DRAW_FRAMEBUFFER, dst_surface, NULL, dst_location);
context_set_draw_buffer(context, GL_COLOR_ATTACHMENT0);
}
context_check_fbo_status(context, GL_DRAW_FRAMEBUFFER);
glColorMask(GL_TRUE, GL_TRUE, GL_TRUE, GL_TRUE);
IWineD3DDeviceImpl_MarkStateDirty(device, STATE_RENDER(WINED3DRS_COLORWRITEENABLE));
IWineD3DDeviceImpl_MarkStateDirty(device, STATE_RENDER(WINED3DRS_COLORWRITEENABLE1));
IWineD3DDeviceImpl_MarkStateDirty(device, STATE_RENDER(WINED3DRS_COLORWRITEENABLE2));
IWineD3DDeviceImpl_MarkStateDirty(device, STATE_RENDER(WINED3DRS_COLORWRITEENABLE3));
glDisable(GL_SCISSOR_TEST);
IWineD3DDeviceImpl_MarkStateDirty(device, STATE_RENDER(WINED3DRS_SCISSORTESTENABLE));
gl_info->fbo_ops.glBlitFramebuffer(src_rect.left, src_rect.top, src_rect.right, src_rect.bottom,
dst_rect.left, dst_rect.top, dst_rect.right, dst_rect.bottom, GL_COLOR_BUFFER_BIT, gl_filter);
checkGLcall("glBlitFramebuffer()");
LEAVE_GL();
if (wined3d_settings.strict_draw_ordering
|| (dst_location == SFLAG_INDRAWABLE
&& dst_surface->container.u.swapchain->front_buffer == dst_surface))
wglFlush();
context_release(context);
}
static void surface_blt_to_drawable(IWineD3DDeviceImpl *device,
WINED3DTEXTUREFILTERTYPE filter, BOOL color_key,
struct wined3d_surface *src_surface, const RECT *src_rect_in,
struct wined3d_surface *dst_surface, const RECT *dst_rect_in)
{
struct wined3d_context *context;
RECT src_rect, dst_rect;
src_rect = *src_rect_in;
dst_rect = *dst_rect_in;
/* Make sure the surface is up-to-date. This should probably use
* surface_load_location() and worry about the destination surface too,
* unless we're overwriting it completely. */
surface_internal_preload(src_surface, SRGB_RGB);
/* Activate the destination context, set it up for blitting */
context = context_acquire(device, dst_surface);
context_apply_blit_state(context, device);
if (!surface_is_offscreen(dst_surface))
surface_translate_drawable_coords(dst_surface, context->win_handle, &dst_rect);
device->blitter->set_shader(device->blit_priv, context->gl_info, src_surface);
ENTER_GL();
if (color_key)
{
glEnable(GL_ALPHA_TEST);
checkGLcall("glEnable(GL_ALPHA_TEST)");
/* When the primary render target uses P8, the alpha component
* contains the palette index. Which means that the colorkey is one of
* the palette entries. In other cases pixels that should be masked
* away have alpha set to 0. */
if (primary_render_target_is_p8(device))
glAlphaFunc(GL_NOTEQUAL, (float)src_surface->SrcBltCKey.dwColorSpaceLowValue / 256.0f);
else
glAlphaFunc(GL_NOTEQUAL, 0.0f);
checkGLcall("glAlphaFunc");
}
else
{
glDisable(GL_ALPHA_TEST);
checkGLcall("glDisable(GL_ALPHA_TEST)");
}
draw_textured_quad(src_surface, &src_rect, &dst_rect, filter);
if (color_key)
{
glDisable(GL_ALPHA_TEST);
checkGLcall("glDisable(GL_ALPHA_TEST)");
}
LEAVE_GL();
/* Leave the opengl state valid for blitting */
device->blitter->unset_shader(context->gl_info);
if (wined3d_settings.strict_draw_ordering
|| (dst_surface->container.type == WINED3D_CONTAINER_SWAPCHAIN
&& (dst_surface->container.u.swapchain->front_buffer == dst_surface)))
wglFlush(); /* Flush to ensure ordering across contexts. */
context_release(context);
}
/* Do not call while under the GL lock. */
HRESULT surface_color_fill(struct wined3d_surface *s, const RECT *rect, const WINED3DCOLORVALUE *color)
{
IWineD3DDeviceImpl *device = s->resource.device;
const struct blit_shader *blitter;
blitter = wined3d_select_blitter(&device->adapter->gl_info, WINED3D_BLIT_OP_COLOR_FILL,
NULL, 0, 0, NULL, rect, s->resource.usage, s->resource.pool, s->resource.format);
if (!blitter)
{
FIXME("No blitter is capable of performing the requested color fill operation.\n");
return WINED3DERR_INVALIDCALL;
}
return blitter->color_fill(device, s, rect, color);
}
/* Do not call while under the GL lock. */
static HRESULT IWineD3DSurfaceImpl_BltOverride(struct wined3d_surface *dst_surface, const RECT *DestRect,
struct wined3d_surface *src_surface, const RECT *SrcRect, DWORD flags, const WINEDDBLTFX *DDBltFx,
WINED3DTEXTUREFILTERTYPE Filter)
{
IWineD3DDeviceImpl *device = dst_surface->resource.device;
const struct wined3d_gl_info *gl_info = &device->adapter->gl_info;
struct wined3d_swapchain *srcSwapchain = NULL, *dstSwapchain = NULL;
RECT dst_rect, src_rect;
TRACE("dst_surface %p, dst_rect %s, src_surface %p, src_rect %s, flags %#x, blt_fx %p, filter %s.\n",
dst_surface, wine_dbgstr_rect(DestRect), src_surface, wine_dbgstr_rect(SrcRect),
flags, DDBltFx, debug_d3dtexturefiltertype(Filter));
/* Get the swapchain. One of the surfaces has to be a primary surface */
if (dst_surface->resource.pool == WINED3DPOOL_SYSTEMMEM)
{
WARN("Destination is in sysmem, rejecting gl blt\n");
return WINED3DERR_INVALIDCALL;
}
if (dst_surface->container.type == WINED3D_CONTAINER_SWAPCHAIN)
dstSwapchain = dst_surface->container.u.swapchain;
if (src_surface)
{
if (src_surface->resource.pool == WINED3DPOOL_SYSTEMMEM)
{
WARN("Src is in sysmem, rejecting gl blt\n");
return WINED3DERR_INVALIDCALL;
}
if (src_surface->container.type == WINED3D_CONTAINER_SWAPCHAIN)
srcSwapchain = src_surface->container.u.swapchain;
}
/* Early sort out of cases where no render target is used */
if (!dstSwapchain && !srcSwapchain
&& src_surface != device->render_targets[0]
&& dst_surface != device->render_targets[0])
{
TRACE("No surface is render target, not using hardware blit.\n");
return WINED3DERR_INVALIDCALL;
}
/* No destination color keying supported */
if (flags & (WINEDDBLT_KEYDEST | WINEDDBLT_KEYDESTOVERRIDE))
{
/* Can we support that with glBlendFunc if blitting to the frame buffer? */
TRACE("Destination color key not supported in accelerated Blit, falling back to software\n");
return WINED3DERR_INVALIDCALL;
}
surface_get_rect(dst_surface, DestRect, &dst_rect);
if (src_surface) surface_get_rect(src_surface, SrcRect, &src_rect);
/* The only case where both surfaces on a swapchain are supported is a back buffer -> front buffer blit on the same swapchain */
if (dstSwapchain && dstSwapchain == srcSwapchain && dstSwapchain->back_buffers
&& dst_surface == dstSwapchain->front_buffer
&& src_surface == dstSwapchain->back_buffers[0])
{
/* Half-Life does a Blt from the back buffer to the front buffer,
* Full surface size, no flags... Use present instead
*
* This path will only be entered for d3d7 and ddraw apps, because d3d8/9 offer no way to blit TO the front buffer
*/
/* Check rects - IWineD3DDevice_Present doesn't handle them */
while(1)
{
TRACE("Looking if a Present can be done...\n");
/* Source Rectangle must be full surface */
if (src_rect.left || src_rect.top
|| src_rect.right != src_surface->resource.width
|| src_rect.bottom != src_surface->resource.height)
{
TRACE("No, Source rectangle doesn't match\n");
break;
}
/* No stretching may occur */
if(src_rect.right != dst_rect.right - dst_rect.left ||
src_rect.bottom != dst_rect.bottom - dst_rect.top) {
TRACE("No, stretching is done\n");
break;
}
/* Destination must be full surface or match the clipping rectangle */
if (dst_surface->clipper && dst_surface->clipper->hWnd)
{
RECT cliprect;
POINT pos[2];
GetClientRect(dst_surface->clipper->hWnd, &cliprect);
pos[0].x = dst_rect.left;
pos[0].y = dst_rect.top;
pos[1].x = dst_rect.right;
pos[1].y = dst_rect.bottom;
MapWindowPoints(GetDesktopWindow(), dst_surface->clipper->hWnd, pos, 2);
if(pos[0].x != cliprect.left || pos[0].y != cliprect.top ||
pos[1].x != cliprect.right || pos[1].y != cliprect.bottom)
{
TRACE("No, dest rectangle doesn't match(clipper)\n");
TRACE("Clip rect at %s\n", wine_dbgstr_rect(&cliprect));
TRACE("Blt dest: %s\n", wine_dbgstr_rect(&dst_rect));
break;
}
}
else if (dst_rect.left || dst_rect.top
|| dst_rect.right != dst_surface->resource.width
|| dst_rect.bottom != dst_surface->resource.height)
{
TRACE("No, dest rectangle doesn't match(surface size)\n");
break;
}
TRACE("Yes\n");
/* These flags are unimportant for the flag check, remove them */
if (!(flags & ~(WINEDDBLT_DONOTWAIT | WINEDDBLT_WAIT)))
{
WINED3DSWAPEFFECT orig_swap = dstSwapchain->presentParms.SwapEffect;
/* The idea behind this is that a glReadPixels and a glDrawPixels call
* take very long, while a flip is fast.
* This applies to Half-Life, which does such Blts every time it finished
* a frame, and to Prince of Persia 3D, which uses this to draw at least the main
* menu. This is also used by all apps when they do windowed rendering
*
* The problem is that flipping is not really the same as copying. After a
* Blt the front buffer is a copy of the back buffer, and the back buffer is
* untouched. Therefore it's necessary to override the swap effect
* and to set it back after the flip.
*
* Windowed Direct3D < 7 apps do the same. The D3D7 sdk demos are nice
* testcases.
*/
dstSwapchain->presentParms.SwapEffect = WINED3DSWAPEFFECT_COPY;
dstSwapchain->presentParms.PresentationInterval = WINED3DPRESENT_INTERVAL_IMMEDIATE;
TRACE("Full screen back buffer -> front buffer blt, performing a flip instead.\n");
wined3d_swapchain_present(dstSwapchain, NULL, NULL, dstSwapchain->win_handle, NULL, 0);
dstSwapchain->presentParms.SwapEffect = orig_swap;
return WINED3D_OK;
}
break;
}
TRACE("Unsupported blit between buffers on the same swapchain\n");
return WINED3DERR_INVALIDCALL;
} else if(dstSwapchain && dstSwapchain == srcSwapchain) {
FIXME("Implement hardware blit between two surfaces on the same swapchain\n");
return WINED3DERR_INVALIDCALL;
} else if(dstSwapchain && srcSwapchain) {
FIXME("Implement hardware blit between two different swapchains\n");
return WINED3DERR_INVALIDCALL;
}
else if (dstSwapchain)
{
/* Handled with regular texture -> swapchain blit */
if (src_surface == device->render_targets[0])
TRACE("Blit from active render target to a swapchain\n");
}
else if (srcSwapchain && dst_surface == device->render_targets[0])
{
FIXME("Implement blit from a swapchain to the active render target\n");
return WINED3DERR_INVALIDCALL;
}
if ((srcSwapchain || src_surface == device->render_targets[0]) && !dstSwapchain)
{
/* Blit from render target to texture */
BOOL stretchx;
/* P8 read back is not implemented */
if (src_surface->resource.format->id == WINED3DFMT_P8_UINT
|| dst_surface->resource.format->id == WINED3DFMT_P8_UINT)
{
TRACE("P8 read back not supported by frame buffer to texture blit\n");
return WINED3DERR_INVALIDCALL;
}
if (flags & (WINEDDBLT_KEYSRC | WINEDDBLT_KEYSRCOVERRIDE))
{
TRACE("Color keying not supported by frame buffer to texture blit\n");
return WINED3DERR_INVALIDCALL;
/* Destination color key is checked above */
}
if(dst_rect.right - dst_rect.left != src_rect.right - src_rect.left) {
stretchx = TRUE;
} else {
stretchx = FALSE;
}
/* Blt is a pretty powerful call, while glCopyTexSubImage2D is not. glCopyTexSubImage cannot
* flip the image nor scale it.
*
* -> If the app asks for a unscaled, upside down copy, just perform one glCopyTexSubImage2D call
* -> If the app wants a image width an unscaled width, copy it line per line
* -> If the app wants a image that is scaled on the x axis, and the destination rectangle is smaller
* than the frame buffer, draw an upside down scaled image onto the fb, read it back and restore the
* back buffer. This is slower than reading line per line, thus not used for flipping
* -> If the app wants a scaled image with a dest rect that is bigger than the fb, it has to be copied
* pixel by pixel
*
* If EXT_framebuffer_blit is supported that can be used instead. Note that EXT_framebuffer_blit implies
* FBO support, so it doesn't really make sense to try and make it work with different offscreen rendering
* backends. */
if (fbo_blit_supported(gl_info, WINED3D_BLIT_OP_COLOR_BLIT,
&src_rect, src_surface->resource.usage, src_surface->resource.pool, src_surface->resource.format,
&dst_rect, dst_surface->resource.usage, dst_surface->resource.pool, dst_surface->resource.format))
{
surface_blt_fbo(device, Filter,
src_surface, SFLAG_INDRAWABLE, &src_rect,
dst_surface, SFLAG_INDRAWABLE, &dst_rect);
surface_modify_location(dst_surface, SFLAG_INDRAWABLE, TRUE);
}
else if (!stretchx || dst_rect.right - dst_rect.left > src_surface->resource.width
|| dst_rect.bottom - dst_rect.top > src_surface->resource.height)
{
TRACE("No stretching in x direction, using direct framebuffer -> texture copy\n");
fb_copy_to_texture_direct(dst_surface, src_surface, &src_rect, &dst_rect, Filter);
} else {
TRACE("Using hardware stretching to flip / stretch the texture\n");
fb_copy_to_texture_hwstretch(dst_surface, src_surface, &src_rect, &dst_rect, Filter);
}
if (!(dst_surface->flags & SFLAG_DONOTFREE))
{
HeapFree(GetProcessHeap(), 0, dst_surface->resource.heapMemory);
dst_surface->resource.allocatedMemory = NULL;
dst_surface->resource.heapMemory = NULL;
}
else
{
dst_surface->flags &= ~SFLAG_INSYSMEM;
}
return WINED3D_OK;
}
else if (src_surface)
{
/* Blit from offscreen surface to render target */
DWORD oldCKeyFlags = src_surface->CKeyFlags;
WINEDDCOLORKEY oldBltCKey = src_surface->SrcBltCKey;
TRACE("Blt from surface %p to rendertarget %p\n", src_surface, dst_surface);
if (!(flags & (WINEDDBLT_KEYSRC | WINEDDBLT_KEYSRCOVERRIDE))
&& fbo_blit_supported(gl_info, WINED3D_BLIT_OP_COLOR_BLIT,
&src_rect, src_surface->resource.usage, src_surface->resource.pool,
src_surface->resource.format,
&dst_rect, dst_surface->resource.usage, dst_surface->resource.pool,
dst_surface->resource.format))
{
TRACE("Using surface_blt_fbo.\n");
/* The source is always a texture, but never the currently active render target, and the texture
* contents are never upside down. */
surface_blt_fbo(device, Filter,
src_surface, SFLAG_INDRAWABLE, &src_rect,
dst_surface, SFLAG_INDRAWABLE, &dst_rect);
surface_modify_location(dst_surface, SFLAG_INDRAWABLE, TRUE);
return WINED3D_OK;
}
if (!(flags & (WINEDDBLT_KEYSRC | WINEDDBLT_KEYSRCOVERRIDE))
&& arbfp_blit.blit_supported(gl_info, WINED3D_BLIT_OP_COLOR_BLIT,
&src_rect, src_surface->resource.usage, src_surface->resource.pool,
src_surface->resource.format,
&dst_rect, dst_surface->resource.usage, dst_surface->resource.pool,
dst_surface->resource.format))
{
return arbfp_blit_surface(device, src_surface, &src_rect, dst_surface, &dst_rect,
WINED3D_BLIT_OP_COLOR_BLIT, Filter);
}
if (!device->blitter->blit_supported(gl_info, WINED3D_BLIT_OP_COLOR_BLIT,
&src_rect, src_surface->resource.usage, src_surface->resource.pool, src_surface->resource.format,
&dst_rect, dst_surface->resource.usage, dst_surface->resource.pool, dst_surface->resource.format))
{
FIXME("Unsupported blit operation falling back to software\n");
return WINED3DERR_INVALIDCALL;
}
/* Color keying: Check if we have to do a color keyed blt,
* and if not check if a color key is activated.
*
* Just modify the color keying parameters in the surface and restore them afterwards
* The surface keeps track of the color key last used to load the opengl surface.
* PreLoad will catch the change to the flags and color key and reload if necessary.
*/
if (flags & WINEDDBLT_KEYSRC)
{
/* Use color key from surface */
}
else if (flags & WINEDDBLT_KEYSRCOVERRIDE)
{
/* Use color key from DDBltFx */
src_surface->CKeyFlags |= WINEDDSD_CKSRCBLT;
src_surface->SrcBltCKey = DDBltFx->ddckSrcColorkey;
}
else
{
/* Do not use color key */
src_surface->CKeyFlags &= ~WINEDDSD_CKSRCBLT;
}
surface_blt_to_drawable(device, Filter, flags & (WINEDDBLT_KEYSRC | WINEDDBLT_KEYSRCOVERRIDE),
src_surface, &src_rect, dst_surface, &dst_rect);
/* Restore the color key parameters */
src_surface->CKeyFlags = oldCKeyFlags;
src_surface->SrcBltCKey = oldBltCKey;
surface_modify_location(dst_surface, SFLAG_INDRAWABLE, TRUE);
return WINED3D_OK;
}
else
{
/* Source-Less Blit to render target */
if (flags & WINEDDBLT_COLORFILL)
{
WINED3DCOLORVALUE color;
TRACE("Colorfill\n");
/* The color as given in the Blt function is in the surface format. */
if (!surface_convert_color_to_float(dst_surface, DDBltFx->u5.dwFillColor, &color))
return WINED3DERR_INVALIDCALL;
return surface_color_fill(dst_surface, &dst_rect, &color);
}
}
/* Default: Fall back to the generic blt. Not an error, a TRACE is enough */
TRACE("Didn't find any usable render target setup for hw blit, falling back to software\n");
return WINED3DERR_INVALIDCALL;
}
/* GL locking is done by the caller */
static void surface_depth_blt(struct wined3d_surface *surface, const struct wined3d_gl_info *gl_info,
GLuint texture, GLsizei w, GLsizei h, GLenum target)
{
IWineD3DDeviceImpl *device = surface->resource.device;
GLint compare_mode = GL_NONE;
struct blt_info info;
GLint old_binding = 0;
RECT rect;
glPushAttrib(GL_ENABLE_BIT | GL_DEPTH_BUFFER_BIT | GL_COLOR_BUFFER_BIT | GL_VIEWPORT_BIT);
glDisable(GL_CULL_FACE);
glDisable(GL_BLEND);
glDisable(GL_ALPHA_TEST);
glDisable(GL_SCISSOR_TEST);
glDisable(GL_STENCIL_TEST);
glEnable(GL_DEPTH_TEST);
glDepthFunc(GL_ALWAYS);
glDepthMask(GL_TRUE);
glColorMask(GL_FALSE, GL_FALSE, GL_FALSE, GL_FALSE);
glViewport(0, surface->pow2Height - h, w, h);
SetRect(&rect, 0, h, w, 0);
surface_get_blt_info(target, &rect, surface->pow2Width, surface->pow2Height, &info);
GL_EXTCALL(glActiveTextureARB(GL_TEXTURE0_ARB));
glGetIntegerv(info.binding, &old_binding);
glBindTexture(info.bind_target, texture);
if (gl_info->supported[ARB_SHADOW])
{
glGetTexParameteriv(info.bind_target, GL_TEXTURE_COMPARE_MODE_ARB, &compare_mode);
if (compare_mode != GL_NONE) glTexParameteri(info.bind_target, GL_TEXTURE_COMPARE_MODE_ARB, GL_NONE);
}
device->shader_backend->shader_select_depth_blt(device->shader_priv,
gl_info, info.tex_type, &surface->ds_current_size);
glBegin(GL_TRIANGLE_STRIP);
glTexCoord3fv(info.coords[0]);
glVertex2f(-1.0f, -1.0f);
glTexCoord3fv(info.coords[1]);
glVertex2f(1.0f, -1.0f);
glTexCoord3fv(info.coords[2]);
glVertex2f(-1.0f, 1.0f);
glTexCoord3fv(info.coords[3]);
glVertex2f(1.0f, 1.0f);
glEnd();
if (compare_mode != GL_NONE) glTexParameteri(info.bind_target, GL_TEXTURE_COMPARE_MODE_ARB, compare_mode);
glBindTexture(info.bind_target, old_binding);
glPopAttrib();
device->shader_backend->shader_deselect_depth_blt(device->shader_priv, gl_info);
}
void surface_modify_ds_location(struct wined3d_surface *surface,
DWORD location, UINT w, UINT h)
{
TRACE("surface %p, new location %#x, w %u, h %u.\n", surface, location, w, h);
if (location & ~SFLAG_DS_LOCATIONS)
FIXME("Invalid location (%#x) specified.\n", location);
surface->ds_current_size.cx = w;
surface->ds_current_size.cy = h;
surface->flags &= ~SFLAG_DS_LOCATIONS;
surface->flags |= location;
}
/* Context activation is done by the caller. */
void surface_load_ds_location(struct wined3d_surface *surface, struct wined3d_context *context, DWORD location)
{
IWineD3DDeviceImpl *device = surface->resource.device;
const struct wined3d_gl_info *gl_info = context->gl_info;
GLsizei w, h;
TRACE("surface %p, new location %#x.\n", surface, location);
/* TODO: Make this work for modes other than FBO */
if (wined3d_settings.offscreen_rendering_mode != ORM_FBO) return;
if (!(surface->flags & location))
{
w = surface->ds_current_size.cx;
h = surface->ds_current_size.cy;
surface->ds_current_size.cx = 0;
surface->ds_current_size.cy = 0;
}
else
{
w = surface->resource.width;
h = surface->resource.height;
}
if (surface->ds_current_size.cx == surface->resource.width
&& surface->ds_current_size.cy == surface->resource.height)
{
TRACE("Location (%#x) is already up to date.\n", location);
return;
}
if (surface->current_renderbuffer)
{
FIXME("Not supported with fixed up depth stencil.\n");
return;
}
if (!(surface->flags & SFLAG_DS_LOCATIONS))
{
/* This mostly happens when a depth / stencil is used without being
* cleared first. In principle we could upload from sysmem, or
* explicitly clear before first usage. For the moment there don't
* appear to be a lot of applications depending on this, so a FIXME
* should do. */
FIXME("No up to date depth stencil location.\n");
surface->flags |= location;
surface->ds_current_size.cx = surface->resource.width;
surface->ds_current_size.cy = surface->resource.height;
return;
}
if (location == SFLAG_DS_OFFSCREEN)
{
GLint old_binding = 0;
GLenum bind_target;
/* The render target is allowed to be smaller than the depth/stencil
* buffer, so the onscreen depth/stencil buffer is potentially smaller
* than the offscreen surface. Don't overwrite the offscreen surface
* with undefined data. */
w = min(w, context->swapchain->presentParms.BackBufferWidth);
h = min(h, context->swapchain->presentParms.BackBufferHeight);
TRACE("Copying onscreen depth buffer to depth texture.\n");
ENTER_GL();
if (!device->depth_blt_texture)
{
glGenTextures(1, &device->depth_blt_texture);
}
/* Note that we use depth_blt here as well, rather than glCopyTexImage2D
* directly on the FBO texture. That's because we need to flip. */
context_bind_fbo(context, GL_FRAMEBUFFER, NULL);
if (surface->texture_target == GL_TEXTURE_RECTANGLE_ARB)
{
glGetIntegerv(GL_TEXTURE_BINDING_RECTANGLE_ARB, &old_binding);
bind_target = GL_TEXTURE_RECTANGLE_ARB;
}
else
{
glGetIntegerv(GL_TEXTURE_BINDING_2D, &old_binding);
bind_target = GL_TEXTURE_2D;
}
glBindTexture(bind_target, device->depth_blt_texture);
glCopyTexImage2D(bind_target, surface->texture_level, surface->resource.format->glInternal, 0, 0, w, h, 0);
glTexParameteri(bind_target, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
glTexParameteri(bind_target, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
glTexParameteri(bind_target, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE);
glTexParameteri(bind_target, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE);
glTexParameteri(bind_target, GL_TEXTURE_WRAP_R, GL_CLAMP_TO_EDGE);
glTexParameteri(bind_target, GL_DEPTH_TEXTURE_MODE_ARB, GL_LUMINANCE);
glBindTexture(bind_target, old_binding);
/* Setup the destination */
if (!device->depth_blt_rb)
{
gl_info->fbo_ops.glGenRenderbuffers(1, &device->depth_blt_rb);
checkGLcall("glGenRenderbuffersEXT");
}
if (device->depth_blt_rb_w != w || device->depth_blt_rb_h != h)
{
gl_info->fbo_ops.glBindRenderbuffer(GL_RENDERBUFFER, device->depth_blt_rb);
checkGLcall("glBindRenderbufferEXT");
gl_info->fbo_ops.glRenderbufferStorage(GL_RENDERBUFFER, GL_RGBA8, w, h);
checkGLcall("glRenderbufferStorageEXT");
device->depth_blt_rb_w = w;
device->depth_blt_rb_h = h;
}
context_bind_fbo(context, GL_FRAMEBUFFER, &context->dst_fbo);
gl_info->fbo_ops.glFramebufferRenderbuffer(GL_FRAMEBUFFER,
GL_COLOR_ATTACHMENT0, GL_RENDERBUFFER, device->depth_blt_rb);
checkGLcall("glFramebufferRenderbufferEXT");
context_attach_depth_stencil_fbo(context, GL_FRAMEBUFFER, surface, FALSE);
/* Do the actual blit */
surface_depth_blt(surface, gl_info, device->depth_blt_texture, w, h, bind_target);
checkGLcall("depth_blt");
if (context->current_fbo) context_bind_fbo(context, GL_FRAMEBUFFER, &context->current_fbo->id);
else context_bind_fbo(context, GL_FRAMEBUFFER, NULL);
LEAVE_GL();
if (wined3d_settings.strict_draw_ordering) wglFlush(); /* Flush to ensure ordering across contexts. */
}
else if (location == SFLAG_DS_ONSCREEN)
{
TRACE("Copying depth texture to onscreen depth buffer.\n");
ENTER_GL();
context_bind_fbo(context, GL_FRAMEBUFFER, NULL);
surface_depth_blt(surface, gl_info, surface->texture_name,
w, h, surface->texture_target);
checkGLcall("depth_blt");
if (context->current_fbo) context_bind_fbo(context, GL_FRAMEBUFFER, &context->current_fbo->id);
LEAVE_GL();
if (wined3d_settings.strict_draw_ordering) wglFlush(); /* Flush to ensure ordering across contexts. */
}
else
{
ERR("Invalid location (%#x) specified.\n", location);
}
surface->flags |= location;
surface->ds_current_size.cx = surface->resource.width;
surface->ds_current_size.cy = surface->resource.height;
}
void surface_modify_location(struct wined3d_surface *surface, DWORD flag, BOOL persistent)
{
const struct wined3d_gl_info *gl_info = &surface->resource.device->adapter->gl_info;
struct wined3d_surface *overlay;
TRACE("surface %p, location %s, persistent %#x.\n",
surface, debug_surflocation(flag), persistent);
if (wined3d_settings.offscreen_rendering_mode == ORM_FBO)
{
if (surface_is_offscreen(surface))
{
/* With ORM_FBO, SFLAG_INTEXTURE and SFLAG_INDRAWABLE are the same for offscreen targets. */
if (flag & (SFLAG_INTEXTURE | SFLAG_INDRAWABLE)) flag |= (SFLAG_INTEXTURE | SFLAG_INDRAWABLE);
}
else
{
TRACE("Surface %p is an onscreen surface.\n", surface);
}
}
if (flag & (SFLAG_INTEXTURE | SFLAG_INSRGBTEX)
&& gl_info->supported[EXT_TEXTURE_SRGB_DECODE])
{
flag |= (SFLAG_INTEXTURE | SFLAG_INSRGBTEX);
}
if (persistent)
{
if (((surface->flags & SFLAG_INTEXTURE) && !(flag & SFLAG_INTEXTURE))
|| ((surface->flags & SFLAG_INSRGBTEX) && !(flag & SFLAG_INSRGBTEX)))
{
if (surface->container.type == WINED3D_CONTAINER_TEXTURE)
{
TRACE("Passing to container.\n");
wined3d_texture_set_dirty(surface->container.u.texture, TRUE);
}
}
surface->flags &= ~SFLAG_LOCATIONS;
surface->flags |= flag;
/* Redraw emulated overlays, if any */
if (flag & SFLAG_INDRAWABLE && !list_empty(&surface->overlays))
{
LIST_FOR_EACH_ENTRY(overlay, &surface->overlays, struct wined3d_surface, overlay_entry)
{
overlay->surface_ops->surface_draw_overlay(overlay);
}
}
}
else
{
if ((surface->flags & (SFLAG_INTEXTURE | SFLAG_INSRGBTEX)) && (flag & (SFLAG_INTEXTURE | SFLAG_INSRGBTEX)))
{
if (surface->container.type == WINED3D_CONTAINER_TEXTURE)
{
TRACE("Passing to container\n");
wined3d_texture_set_dirty(surface->container.u.texture, TRUE);
}
}
surface->flags &= ~flag;
}
if (!(surface->flags & SFLAG_LOCATIONS))
{
ERR("Surface %p does not have any up to date location.\n", surface);
}
}
static DWORD resource_access_from_location(DWORD location)
{
switch (location)
{
case SFLAG_INSYSMEM:
return WINED3D_RESOURCE_ACCESS_CPU;
case SFLAG_INDRAWABLE:
case SFLAG_INSRGBTEX:
case SFLAG_INTEXTURE:
return WINED3D_RESOURCE_ACCESS_GPU;
default:
FIXME("Unhandled location %#x.\n", location);
return 0;
}
}
HRESULT surface_load_location(struct wined3d_surface *surface, DWORD flag, const RECT *rect)
{
IWineD3DDeviceImpl *device = surface->resource.device;
const struct wined3d_gl_info *gl_info = &device->adapter->gl_info;
BOOL drawable_read_ok = surface_is_offscreen(surface);
struct wined3d_format format;
CONVERT_TYPES convert;
int width, pitch, outpitch;
BYTE *mem;
BOOL in_fbo = FALSE;
TRACE("surface %p, location %s, rect %s.\n", surface, debug_surflocation(flag), wine_dbgstr_rect(rect));
if (surface->resource.usage & WINED3DUSAGE_DEPTHSTENCIL)
{
if (flag == SFLAG_INTEXTURE)
{
struct wined3d_context *context = context_acquire(device, NULL);
surface_load_ds_location(surface, context, SFLAG_DS_OFFSCREEN);
context_release(context);
return WINED3D_OK;
}
else
{
FIXME("Unimplemented location %s for depth/stencil buffers.\n", debug_surflocation(flag));
return WINED3DERR_INVALIDCALL;
}
}
if (wined3d_settings.offscreen_rendering_mode == ORM_FBO)
{
if (surface_is_offscreen(surface))
{
/* With ORM_FBO, SFLAG_INTEXTURE and SFLAG_INDRAWABLE are the same for offscreen targets.
* Prefer SFLAG_INTEXTURE. */
if (flag == SFLAG_INDRAWABLE) flag = SFLAG_INTEXTURE;
drawable_read_ok = FALSE;
in_fbo = TRUE;
}
else
{
TRACE("Surface %p is an onscreen surface.\n", surface);
}
}
if (flag == SFLAG_INSRGBTEX && gl_info->supported[EXT_TEXTURE_SRGB_DECODE])
{
flag = SFLAG_INTEXTURE;
}
if (surface->flags & flag)
{
TRACE("Location already up to date\n");
return WINED3D_OK;
}
if (WARN_ON(d3d_surface))
{
DWORD required_access = resource_access_from_location(flag);
if ((surface->resource.access_flags & required_access) != required_access)
WARN("Operation requires %#x access, but surface only has %#x.\n",
required_access, surface->resource.access_flags);
}
if (!(surface->flags & SFLAG_LOCATIONS))
{
ERR("Surface %p does not have any up to date location.\n", surface);
surface->flags |= SFLAG_LOST;
return WINED3DERR_DEVICELOST;
}
if (flag == SFLAG_INSYSMEM)
{
surface_prepare_system_memory(surface);
/* Download the surface to system memory */
if (surface->flags & (SFLAG_INTEXTURE | SFLAG_INSRGBTEX))
{
struct wined3d_context *context = NULL;
if (!device->isInDraw) context = context_acquire(device, NULL);
surface_bind_and_dirtify(surface, gl_info, !(surface->flags & SFLAG_INTEXTURE));
surface_download_data(surface, gl_info);
if (context) context_release(context);
}
else
{
/* Note: It might be faster to download into a texture first. */
read_from_framebuffer(surface, rect, surface->resource.allocatedMemory,
wined3d_surface_get_pitch(surface));
}
}
else if (flag == SFLAG_INDRAWABLE)
{
if (wined3d_settings.rendertargetlock_mode == RTL_READTEX)
surface_load_location(surface, SFLAG_INTEXTURE, NULL);
if (surface->flags & SFLAG_INTEXTURE)
{
RECT r;
surface_get_rect(surface, rect, &r);
surface_blt_to_drawable(device, WINED3DTEXF_POINT, FALSE, surface, &r, surface, &r);
}
else
{
int byte_count;
if ((surface->flags & SFLAG_LOCATIONS) == SFLAG_INSRGBTEX)
{
/* This needs a shader to convert the srgb data sampled from the GL texture into RGB
* values, otherwise we get incorrect values in the target. For now go the slow way
* via a system memory copy
*/
surface_load_location(surface, SFLAG_INSYSMEM, rect);
}
d3dfmt_get_conv(surface, FALSE /* We need color keying */,
FALSE /* We won't use textures */, &format, &convert);
/* The width is in 'length' not in bytes */
width = surface->resource.width;
pitch = wined3d_surface_get_pitch(surface);
/* Don't use PBOs for converted surfaces. During PBO conversion we look at SFLAG_CONVERTED
* but it isn't set (yet) in all cases it is getting called. */
if ((convert != NO_CONVERSION) && (surface->flags & SFLAG_PBO))
{
struct wined3d_context *context = NULL;
TRACE("Removing the pbo attached to surface %p.\n", surface);
if (!device->isInDraw) context = context_acquire(device, NULL);
surface_remove_pbo(surface, gl_info);
if (context) context_release(context);
}
if ((convert != NO_CONVERSION) && surface->resource.allocatedMemory)
{
int height = surface->resource.height;
byte_count = format.conv_byte_count;
/* Stick to the alignment for the converted surface too, makes it easier to load the surface */
outpitch = width * byte_count;
outpitch = (outpitch + device->surface_alignment - 1) & ~(device->surface_alignment - 1);
mem = HeapAlloc(GetProcessHeap(), 0, outpitch * height);
if(!mem) {
ERR("Out of memory %d, %d!\n", outpitch, height);
return WINED3DERR_OUTOFVIDEOMEMORY;
}
d3dfmt_convert_surface(surface->resource.allocatedMemory, mem, pitch,
width, height, outpitch, convert, surface);
surface->flags |= SFLAG_CONVERTED;
}
else
{
surface->flags &= ~SFLAG_CONVERTED;
mem = surface->resource.allocatedMemory;
byte_count = format.byte_count;
}
flush_to_framebuffer_drawpixels(surface, rect, format.glFormat, format.glType, byte_count, mem);
/* Don't delete PBO memory */
if ((mem != surface->resource.allocatedMemory) && !(surface->flags & SFLAG_PBO))
HeapFree(GetProcessHeap(), 0, mem);
}
}
else /* if(flag & (SFLAG_INTEXTURE | SFLAG_INSRGBTEX)) */
{
const DWORD attach_flags = WINED3DFMT_FLAG_FBO_ATTACHABLE | WINED3DFMT_FLAG_FBO_ATTACHABLE_SRGB;
if (drawable_read_ok && (surface->flags & SFLAG_INDRAWABLE))
{
read_from_framebuffer_texture(surface, flag == SFLAG_INSRGBTEX);
}
else if (surface->flags & (SFLAG_INSRGBTEX | SFLAG_INTEXTURE)
&& (surface->resource.format->flags & attach_flags) == attach_flags
&& fbo_blit_supported(gl_info, WINED3D_BLIT_OP_COLOR_BLIT,
NULL, surface->resource.usage, surface->resource.pool, surface->resource.format,
NULL, surface->resource.usage, surface->resource.pool, surface->resource.format))
{
DWORD src_location = flag == SFLAG_INSRGBTEX ? SFLAG_INTEXTURE : SFLAG_INSRGBTEX;
RECT rect = {0, 0, surface->resource.width, surface->resource.height};
surface_blt_fbo(surface->resource.device, WINED3DTEXF_POINT,
surface, src_location, &rect, surface, flag, &rect);
}
else
{
/* Upload from system memory */
BOOL srgb = flag == SFLAG_INSRGBTEX;
struct wined3d_context *context = NULL;
d3dfmt_get_conv(surface, TRUE /* We need color keying */,
TRUE /* We will use textures */, &format, &convert);
if (srgb)
{
if ((surface->flags & (SFLAG_INTEXTURE | SFLAG_INSYSMEM)) == SFLAG_INTEXTURE)
{
/* Performance warning... */
FIXME("Downloading RGB surface %p to reload it as sRGB.\n", surface);
surface_load_location(surface, SFLAG_INSYSMEM, rect);
}
}
else
{
if ((surface->flags & (SFLAG_INSRGBTEX | SFLAG_INSYSMEM)) == SFLAG_INSRGBTEX)
{
/* Performance warning... */
FIXME("Downloading sRGB surface %p to reload it as RGB.\n", surface);
surface_load_location(surface, SFLAG_INSYSMEM, rect);
}
}
if (!(surface->flags & SFLAG_INSYSMEM))
{
WARN("Trying to load a texture from sysmem, but SFLAG_INSYSMEM is not set.\n");
/* Lets hope we get it from somewhere... */
surface_load_location(surface, SFLAG_INSYSMEM, rect);
}
if (!device->isInDraw) context = context_acquire(device, NULL);
surface_prepare_texture(surface, gl_info, srgb);
surface_bind_and_dirtify(surface, gl_info, srgb);
if (surface->CKeyFlags & WINEDDSD_CKSRCBLT)
{
surface->flags |= SFLAG_GLCKEY;
surface->glCKey = surface->SrcBltCKey;
}
else surface->flags &= ~SFLAG_GLCKEY;
/* The width is in 'length' not in bytes */
width = surface->resource.width;
pitch = wined3d_surface_get_pitch(surface);
/* Don't use PBOs for converted surfaces. During PBO conversion we look at SFLAG_CONVERTED
* but it isn't set (yet) in all cases it is getting called. */
if ((convert != NO_CONVERSION || format.convert) && (surface->flags & SFLAG_PBO))
{
TRACE("Removing the pbo attached to surface %p.\n", surface);
surface_remove_pbo(surface, gl_info);
}
if (format.convert)
{
/* This code is entered for texture formats which need a fixup. */
UINT height = surface->resource.height;
/* Stick to the alignment for the converted surface too, makes it easier to load the surface */
outpitch = width * format.conv_byte_count;
outpitch = (outpitch + device->surface_alignment - 1) & ~(device->surface_alignment - 1);
mem = HeapAlloc(GetProcessHeap(), 0, outpitch * height);
if(!mem) {
ERR("Out of memory %d, %d!\n", outpitch, height);
if (context) context_release(context);
return WINED3DERR_OUTOFVIDEOMEMORY;
}
format.convert(surface->resource.allocatedMemory, mem, pitch, width, height);
}
else if (convert != NO_CONVERSION && surface->resource.allocatedMemory)
{
/* This code is only entered for color keying fixups */
UINT height = surface->resource.height;
/* Stick to the alignment for the converted surface too, makes it easier to load the surface */
outpitch = width * format.conv_byte_count;
outpitch = (outpitch + device->surface_alignment - 1) & ~(device->surface_alignment - 1);
mem = HeapAlloc(GetProcessHeap(), 0, outpitch * height);
if(!mem) {
ERR("Out of memory %d, %d!\n", outpitch, height);
if (context) context_release(context);
return WINED3DERR_OUTOFVIDEOMEMORY;
}
d3dfmt_convert_surface(surface->resource.allocatedMemory, mem, pitch,
width, height, outpitch, convert, surface);
}
else
{
mem = surface->resource.allocatedMemory;
}
/* Make sure the correct pitch is used */
ENTER_GL();
glPixelStorei(GL_UNPACK_ROW_LENGTH, width);
LEAVE_GL();
if (mem || (surface->flags & SFLAG_PBO))
surface_upload_data(surface, gl_info, &format, srgb, mem);
/* Restore the default pitch */
ENTER_GL();
glPixelStorei(GL_UNPACK_ROW_LENGTH, 0);
LEAVE_GL();
if (context) context_release(context);
/* Don't delete PBO memory */
if ((mem != surface->resource.allocatedMemory) && !(surface->flags & SFLAG_PBO))
HeapFree(GetProcessHeap(), 0, mem);
}
}
if (!rect)
{
surface->flags |= flag;
if (flag != SFLAG_INSYSMEM && (surface->flags & SFLAG_INSYSMEM))
surface_evict_sysmem(surface);
}
if (in_fbo && (surface->flags & (SFLAG_INTEXTURE | SFLAG_INDRAWABLE)))
{
/* With ORM_FBO, SFLAG_INTEXTURE and SFLAG_INDRAWABLE are the same for offscreen targets. */
surface->flags |= (SFLAG_INTEXTURE | SFLAG_INDRAWABLE);
}
if (surface->flags & (SFLAG_INTEXTURE | SFLAG_INSRGBTEX)
&& gl_info->supported[EXT_TEXTURE_SRGB_DECODE])
{
surface->flags |= (SFLAG_INTEXTURE | SFLAG_INSRGBTEX);
}
return WINED3D_OK;
}
BOOL surface_is_offscreen(struct wined3d_surface *surface)
{
struct wined3d_swapchain *swapchain = surface->container.u.swapchain;
/* Not on a swapchain - must be offscreen */
if (surface->container.type != WINED3D_CONTAINER_SWAPCHAIN) return TRUE;
/* The front buffer is always onscreen */
if (surface == swapchain->front_buffer) return FALSE;
/* If the swapchain is rendered to an FBO, the backbuffer is
* offscreen, otherwise onscreen */
return swapchain->render_to_fbo;
}
static HRESULT ffp_blit_alloc(IWineD3DDeviceImpl *device) { return WINED3D_OK; }
/* Context activation is done by the caller. */
static void ffp_blit_free(IWineD3DDeviceImpl *device) { }
/* This function is used in case of 8bit paletted textures using GL_EXT_paletted_texture */
/* Context activation is done by the caller. */
static void ffp_blit_p8_upload_palette(struct wined3d_surface *surface, const struct wined3d_gl_info *gl_info)
{
BYTE table[256][4];
BOOL colorkey_active = (surface->CKeyFlags & WINEDDSD_CKSRCBLT) ? TRUE : FALSE;
d3dfmt_p8_init_palette(surface, table, colorkey_active);
TRACE("Using GL_EXT_PALETTED_TEXTURE for 8-bit paletted texture support\n");
ENTER_GL();
GL_EXTCALL(glColorTableEXT(surface->texture_target, GL_RGBA, 256, GL_RGBA, GL_UNSIGNED_BYTE, table));
LEAVE_GL();
}
/* Context activation is done by the caller. */
static HRESULT ffp_blit_set(void *blit_priv, const struct wined3d_gl_info *gl_info, struct wined3d_surface *surface)
{
enum complex_fixup fixup = get_complex_fixup(surface->resource.format->color_fixup);
/* When EXT_PALETTED_TEXTURE is around, palette conversion is done by the GPU
* else the surface is converted in software at upload time in LoadLocation.
*/
if(fixup == COMPLEX_FIXUP_P8 && gl_info->supported[EXT_PALETTED_TEXTURE])
ffp_blit_p8_upload_palette(surface, gl_info);
ENTER_GL();
glEnable(surface->texture_target);
checkGLcall("glEnable(surface->texture_target)");
LEAVE_GL();
return WINED3D_OK;
}
/* Context activation is done by the caller. */
static void ffp_blit_unset(const struct wined3d_gl_info *gl_info)
{
ENTER_GL();
glDisable(GL_TEXTURE_2D);
checkGLcall("glDisable(GL_TEXTURE_2D)");
if (gl_info->supported[ARB_TEXTURE_CUBE_MAP])
{
glDisable(GL_TEXTURE_CUBE_MAP_ARB);
checkGLcall("glDisable(GL_TEXTURE_CUBE_MAP_ARB)");
}
if (gl_info->supported[ARB_TEXTURE_RECTANGLE])
{
glDisable(GL_TEXTURE_RECTANGLE_ARB);
checkGLcall("glDisable(GL_TEXTURE_RECTANGLE_ARB)");
}
LEAVE_GL();
}
static BOOL ffp_blit_supported(const struct wined3d_gl_info *gl_info, enum wined3d_blit_op blit_op,
const RECT *src_rect, DWORD src_usage, WINED3DPOOL src_pool, const struct wined3d_format *src_format,
const RECT *dst_rect, DWORD dst_usage, WINED3DPOOL dst_pool, const struct wined3d_format *dst_format)
{
enum complex_fixup src_fixup;
switch (blit_op)
{
case WINED3D_BLIT_OP_COLOR_BLIT:
src_fixup = get_complex_fixup(src_format->color_fixup);
if (TRACE_ON(d3d_surface) && TRACE_ON(d3d))
{
TRACE("Checking support for fixup:\n");
dump_color_fixup_desc(src_format->color_fixup);
}
if (!is_identity_fixup(dst_format->color_fixup))
{
TRACE("Destination fixups are not supported\n");
return FALSE;
}
if (src_fixup == COMPLEX_FIXUP_P8 && gl_info->supported[EXT_PALETTED_TEXTURE])
{
TRACE("P8 fixup supported\n");
return TRUE;
}
/* We only support identity conversions. */
if (is_identity_fixup(src_format->color_fixup))
{
TRACE("[OK]\n");
return TRUE;
}
TRACE("[FAILED]\n");
return FALSE;
case WINED3D_BLIT_OP_COLOR_FILL:
if (!(dst_usage & WINED3DUSAGE_RENDERTARGET))
{
TRACE("Color fill not supported\n");
return FALSE;
}
return TRUE;
case WINED3D_BLIT_OP_DEPTH_FILL:
return TRUE;
default:
TRACE("Unsupported blit_op=%d\n", blit_op);
return FALSE;
}
}
/* Do not call while under the GL lock. */
static HRESULT ffp_blit_color_fill(IWineD3DDeviceImpl *device, struct wined3d_surface *dst_surface,
const RECT *dst_rect, const WINED3DCOLORVALUE *color)
{
const RECT draw_rect = {0, 0, dst_surface->resource.width, dst_surface->resource.height};
return device_clear_render_targets(device, 1, &dst_surface, NULL,
1, dst_rect, &draw_rect, WINED3DCLEAR_TARGET, color, 0.0f, 0);
}
/* Do not call while under the GL lock. */
static HRESULT ffp_blit_depth_fill(IWineD3DDeviceImpl *device,
struct wined3d_surface *surface, const RECT *rect, float depth)
{
const RECT draw_rect = {0, 0, surface->resource.width, surface->resource.height};
return device_clear_render_targets(device, 0, NULL, surface,
1, rect, &draw_rect, WINED3DCLEAR_ZBUFFER, 0, depth, 0);
}
const struct blit_shader ffp_blit = {
ffp_blit_alloc,
ffp_blit_free,
ffp_blit_set,
ffp_blit_unset,
ffp_blit_supported,
ffp_blit_color_fill,
ffp_blit_depth_fill,
};
static HRESULT cpu_blit_alloc(IWineD3DDeviceImpl *device)
{
return WINED3D_OK;
}
/* Context activation is done by the caller. */
static void cpu_blit_free(IWineD3DDeviceImpl *device)
{
}
/* Context activation is done by the caller. */
static HRESULT cpu_blit_set(void *blit_priv, const struct wined3d_gl_info *gl_info, struct wined3d_surface *surface)
{
return WINED3D_OK;
}
/* Context activation is done by the caller. */
static void cpu_blit_unset(const struct wined3d_gl_info *gl_info)
{
}
static BOOL cpu_blit_supported(const struct wined3d_gl_info *gl_info, enum wined3d_blit_op blit_op,
const RECT *src_rect, DWORD src_usage, WINED3DPOOL src_pool, const struct wined3d_format *src_format,
const RECT *dst_rect, DWORD dst_usage, WINED3DPOOL dst_pool, const struct wined3d_format *dst_format)
{
if (blit_op == WINED3D_BLIT_OP_COLOR_FILL)
{
return TRUE;
}
return FALSE;
}
static HRESULT surface_cpu_blt(struct wined3d_surface *dst_surface, const RECT *dst_rect,
struct wined3d_surface *src_surface, const RECT *src_rect, DWORD flags,
const WINEDDBLTFX *fx, WINED3DTEXTUREFILTERTYPE filter)
{
int bpp, srcheight, srcwidth, dstheight, dstwidth, width;
const struct wined3d_format *src_format, *dst_format;
struct wined3d_surface *orig_src = src_surface;
WINED3DLOCKED_RECT dlock, slock;
HRESULT hr = WINED3D_OK;
const BYTE *sbuf;
RECT xdst,xsrc;
BYTE *dbuf;
int x, y;
TRACE("dst_surface %p, dst_rect %s, src_surface %p, src_rect %s, flags %#x, fx %p, filter %s.\n",
dst_surface, wine_dbgstr_rect(dst_rect), src_surface, wine_dbgstr_rect(src_rect),
flags, fx, debug_d3dtexturefiltertype(filter));
if ((dst_surface->flags & SFLAG_LOCKED) || (src_surface && (src_surface->flags & SFLAG_LOCKED)))
{
WARN("Surface is busy, returning WINEDDERR_SURFACEBUSY\n");
return WINEDDERR_SURFACEBUSY;
}
/* First check for the validity of source / destination rectangles.
* This was verified using a test application and by MSDN. */
if (src_rect)
{
if (src_surface)
{
if (src_rect->right < src_rect->left || src_rect->bottom < src_rect->top
|| src_rect->left > src_surface->resource.width || src_rect->left < 0
|| src_rect->top > src_surface->resource.height || src_rect->top < 0
|| src_rect->right > src_surface->resource.width || src_rect->right < 0
|| src_rect->bottom > src_surface->resource.height || src_rect->bottom < 0)
{
WARN("Application gave us bad source rectangle for Blt.\n");
return WINEDDERR_INVALIDRECT;
}
if (!src_rect->right || !src_rect->bottom
|| src_rect->left == (int)src_surface->resource.width
|| src_rect->top == (int)src_surface->resource.height)
{
TRACE("Nothing to be done.\n");
return WINED3D_OK;
}
}
xsrc = *src_rect;
}
else if (src_surface)
{
xsrc.left = 0;
xsrc.top = 0;
xsrc.right = src_surface->resource.width;
xsrc.bottom = src_surface->resource.height;
}
else
{
memset(&xsrc, 0, sizeof(xsrc));
}
if (dst_rect)
{
/* For the Destination rect, it can be out of bounds on the condition
* that a clipper is set for the given surface. */
if (!dst_surface->clipper && (dst_rect->right < dst_rect->left || dst_rect->bottom < dst_rect->top
|| dst_rect->left > dst_surface->resource.width || dst_rect->left < 0
|| dst_rect->top > dst_surface->resource.height || dst_rect->top < 0
|| dst_rect->right > dst_surface->resource.width || dst_rect->right < 0
|| dst_rect->bottom > dst_surface->resource.height || dst_rect->bottom < 0))
{
WARN("Application gave us bad destination rectangle for Blt without a clipper set.\n");
return WINEDDERR_INVALIDRECT;
}
if (dst_rect->right <= 0 || dst_rect->bottom <= 0
|| dst_rect->left >= (int)dst_surface->resource.width
|| dst_rect->top >= (int)dst_surface->resource.height)
{
TRACE("Nothing to be done.\n");
return WINED3D_OK;
}
if (!src_surface)
{
RECT full_rect;
full_rect.left = 0;
full_rect.top = 0;
full_rect.right = dst_surface->resource.width;
full_rect.bottom = dst_surface->resource.height;
IntersectRect(&xdst, &full_rect, dst_rect);
}
else
{
BOOL clip_horiz, clip_vert;
xdst = *dst_rect;
clip_horiz = xdst.left < 0 || xdst.right > (int)dst_surface->resource.width;
clip_vert = xdst.top < 0 || xdst.bottom > (int)dst_surface->resource.height;
if (clip_vert || clip_horiz)
{
/* Now check if this is a special case or not... */
if ((flags & WINEDDBLT_DDFX)
|| (clip_horiz && xdst.right - xdst.left != xsrc.right - xsrc.left)
|| (clip_vert && xdst.bottom - xdst.top != xsrc.bottom - xsrc.top))
{
WARN("Out of screen rectangle in special case. Not handled right now.\n");
return WINED3D_OK;
}
if (clip_horiz)
{
if (xdst.left < 0)
{
xsrc.left -= xdst.left;
xdst.left = 0;
}
if (xdst.right > dst_surface->resource.width)
{
xsrc.right -= (xdst.right - (int)dst_surface->resource.width);
xdst.right = (int)dst_surface->resource.width;
}
}
if (clip_vert)
{
if (xdst.top < 0)
{
xsrc.top -= xdst.top;
xdst.top = 0;
}
if (xdst.bottom > dst_surface->resource.height)
{
xsrc.bottom -= (xdst.bottom - (int)dst_surface->resource.height);
xdst.bottom = (int)dst_surface->resource.height;
}
}
/* And check if after clipping something is still to be done... */
if ((xdst.right <= 0) || (xdst.bottom <= 0)
|| (xdst.left >= (int)dst_surface->resource.width)
|| (xdst.top >= (int)dst_surface->resource.height)
|| (xsrc.right <= 0) || (xsrc.bottom <= 0)
|| (xsrc.left >= (int)src_surface->resource.width)
|| (xsrc.top >= (int)src_surface->resource.height))
{
TRACE("Nothing to be done after clipping.\n");
return WINED3D_OK;
}
}
}
}
else
{
xdst.left = 0;
xdst.top = 0;
xdst.right = dst_surface->resource.width;
xdst.bottom = dst_surface->resource.height;
}
if (src_surface == dst_surface)
{
wined3d_surface_map(dst_surface, &dlock, NULL, 0);
slock = dlock;
src_format = dst_surface->resource.format;
dst_format = src_format;
}
else
{
dst_format = dst_surface->resource.format;
if (src_surface)
{
if (dst_surface->resource.format->id != src_surface->resource.format->id)
{
src_surface = surface_convert_format(src_surface, dst_format->id);
if (!src_surface)
{
/* The conv function writes a FIXME */
WARN("Cannot convert source surface format to dest format.\n");
goto release;
}
}
wined3d_surface_map(src_surface, &slock, NULL, WINED3DLOCK_READONLY);
src_format = src_surface->resource.format;
}
else
{
src_format = dst_format;
}
if (dst_rect)
wined3d_surface_map(dst_surface, &dlock, &xdst, 0);
else
wined3d_surface_map(dst_surface, &dlock, NULL, 0);
}
if (!fx || !(fx->dwDDFX)) flags &= ~WINEDDBLT_DDFX;
if (src_format->flags & dst_format->flags & WINED3DFMT_FLAG_FOURCC)
{
if (!dst_rect || src_surface == dst_surface)
{
memcpy(dlock.pBits, slock.pBits, dst_surface->resource.size);
goto release;
}
}
bpp = dst_surface->resource.format->byte_count;
srcheight = xsrc.bottom - xsrc.top;
srcwidth = xsrc.right - xsrc.left;
dstheight = xdst.bottom - xdst.top;
dstwidth = xdst.right - xdst.left;
width = (xdst.right - xdst.left) * bpp;
if (dst_rect && src_surface != dst_surface)
dbuf = dlock.pBits;
else
dbuf = (BYTE*)dlock.pBits+(xdst.top*dlock.Pitch)+(xdst.left*bpp);
if (flags & WINEDDBLT_WAIT)
{
flags &= ~WINEDDBLT_WAIT;
}
if (flags & WINEDDBLT_ASYNC)
{
static BOOL displayed = FALSE;
if (!displayed)
FIXME("Can't handle WINEDDBLT_ASYNC flag right now.\n");
displayed = TRUE;
flags &= ~WINEDDBLT_ASYNC;
}
if (flags & WINEDDBLT_DONOTWAIT)
{
/* WINEDDBLT_DONOTWAIT appeared in DX7 */
static BOOL displayed = FALSE;
if (!displayed)
FIXME("Can't handle WINEDDBLT_DONOTWAIT flag right now.\n");
displayed = TRUE;
flags &= ~WINEDDBLT_DONOTWAIT;
}
/* First, all the 'source-less' blits */
if (flags & WINEDDBLT_COLORFILL)
{
hr = _Blt_ColorFill(dbuf, dstwidth, dstheight, bpp, dlock.Pitch, fx->u5.dwFillColor);
flags &= ~WINEDDBLT_COLORFILL;
}
if (flags & WINEDDBLT_DEPTHFILL)
{
FIXME("DDBLT_DEPTHFILL needs to be implemented!\n");
}
if (flags & WINEDDBLT_ROP)
{
/* Catch some degenerate cases here. */
switch (fx->dwROP)
{
case BLACKNESS:
hr = _Blt_ColorFill(dbuf,dstwidth,dstheight,bpp,dlock.Pitch,0);
break;
case 0xAA0029: /* No-op */
break;
case WHITENESS:
hr = _Blt_ColorFill(dbuf,dstwidth,dstheight,bpp,dlock.Pitch,~0);
break;
case SRCCOPY: /* Well, we do that below? */
break;
default:
FIXME("Unsupported raster op: %08x Pattern: %p\n", fx->dwROP, fx->u5.lpDDSPattern);
goto error;
}
flags &= ~WINEDDBLT_ROP;
}
if (flags & WINEDDBLT_DDROPS)
{
FIXME("\tDdraw Raster Ops: %08x Pattern: %p\n", fx->dwDDROP, fx->u5.lpDDSPattern);
}
/* Now the 'with source' blits. */
if (src_surface)
{
const BYTE *sbase;
int sx, xinc, sy, yinc;
if (!dstwidth || !dstheight) /* Hmm... stupid program? */
goto release;
if (filter != WINED3DTEXF_NONE && filter != WINED3DTEXF_POINT
&& (srcwidth != dstwidth || srcheight != dstheight))
{
/* Can happen when d3d9 apps do a StretchRect() call which isn't handled in GL. */
FIXME("Filter %s not supported in software blit.\n", debug_d3dtexturefiltertype(filter));
}
sbase = (BYTE*)slock.pBits+(xsrc.top*slock.Pitch)+xsrc.left*bpp;
xinc = (srcwidth << 16) / dstwidth;
yinc = (srcheight << 16) / dstheight;
if (!flags)
{
/* No effects, we can cheat here. */
if (dstwidth == srcwidth)
{
if (dstheight == srcheight)
{
/* No stretching in either direction. This needs to be as
* fast as possible. */
sbuf = sbase;
/* Check for overlapping surfaces. */
if (src_surface != dst_surface || xdst.top < xsrc.top
|| xdst.right <= xsrc.left || xsrc.right <= xdst.left)
{
/* No overlap, or dst above src, so copy from top downwards. */
for (y = 0; y < dstheight; ++y)
{
memcpy(dbuf, sbuf, width);
sbuf += slock.Pitch;
dbuf += dlock.Pitch;
}
}
else if (xdst.top > xsrc.top)
{
/* Copy from bottom upwards. */
sbuf += (slock.Pitch*dstheight);
dbuf += (dlock.Pitch*dstheight);
for (y = 0; y < dstheight; ++y)
{
sbuf -= slock.Pitch;
dbuf -= dlock.Pitch;
memcpy(dbuf, sbuf, width);
}
}
else
{
/* Src and dst overlapping on the same line, use memmove. */
for (y = 0; y < dstheight; ++y)
{
memmove(dbuf, sbuf, width);
sbuf += slock.Pitch;
dbuf += dlock.Pitch;
}
}
}
else
{
/* Stretching in y direction only. */
for (y = sy = 0; y < dstheight; ++y, sy += yinc)
{
sbuf = sbase + (sy >> 16) * slock.Pitch;
memcpy(dbuf, sbuf, width);
dbuf += dlock.Pitch;
}
}
}
else
{
/* Stretching in X direction. */
int last_sy = -1;
for (y = sy = 0; y < dstheight; ++y, sy += yinc)
{
sbuf = sbase + (sy >> 16) * slock.Pitch;
if ((sy >> 16) == (last_sy >> 16))
{
/* This source row is the same as last source row -
* Copy the already stretched row. */
memcpy(dbuf, dbuf - dlock.Pitch, width);
}
else
{
#define STRETCH_ROW(type) \
do { \
const type *s = (const type *)sbuf; \
type *d = (type *)dbuf; \
for (x = sx = 0; x < dstwidth; ++x, sx += xinc) \
d[x] = s[sx >> 16]; \
} while(0)
switch(bpp)
{
case 1:
STRETCH_ROW(BYTE);
break;
case 2:
STRETCH_ROW(WORD);
break;
case 4:
STRETCH_ROW(DWORD);
break;
case 3:
{
const BYTE *s;
BYTE *d = dbuf;
for (x = sx = 0; x < dstwidth; x++, sx+= xinc)
{
DWORD pixel;
s = sbuf + 3 * (sx >> 16);
pixel = s[0] | (s[1] << 8) | (s[2] << 16);
d[0] = (pixel ) & 0xff;
d[1] = (pixel >> 8) & 0xff;
d[2] = (pixel >> 16) & 0xff;
d += 3;
}
break;
}
default:
FIXME("Stretched blit not implemented for bpp %u!\n", bpp * 8);
hr = WINED3DERR_NOTAVAILABLE;
goto error;
}
#undef STRETCH_ROW
}
dbuf += dlock.Pitch;
last_sy = sy;
}
}
}
else
{
LONG dstyinc = dlock.Pitch, dstxinc = bpp;
DWORD keylow = 0xFFFFFFFF, keyhigh = 0, keymask = 0xFFFFFFFF;
DWORD destkeylow = 0x0, destkeyhigh = 0xFFFFFFFF, destkeymask = 0xFFFFFFFF;
if (flags & (WINEDDBLT_KEYSRC | WINEDDBLT_KEYDEST | WINEDDBLT_KEYSRCOVERRIDE | WINEDDBLT_KEYDESTOVERRIDE))
{
/* The color keying flags are checked for correctness in ddraw */
if (flags & WINEDDBLT_KEYSRC)
{
keylow = src_surface->SrcBltCKey.dwColorSpaceLowValue;
keyhigh = src_surface->SrcBltCKey.dwColorSpaceHighValue;
}
else if (flags & WINEDDBLT_KEYSRCOVERRIDE)
{
keylow = fx->ddckSrcColorkey.dwColorSpaceLowValue;
keyhigh = fx->ddckSrcColorkey.dwColorSpaceHighValue;
}
if (flags & WINEDDBLT_KEYDEST)
{
/* Destination color keys are taken from the source surface! */
destkeylow = src_surface->DestBltCKey.dwColorSpaceLowValue;
destkeyhigh = src_surface->DestBltCKey.dwColorSpaceHighValue;
}
else if (flags & WINEDDBLT_KEYDESTOVERRIDE)
{
destkeylow = fx->ddckDestColorkey.dwColorSpaceLowValue;
destkeyhigh = fx->ddckDestColorkey.dwColorSpaceHighValue;
}
if (bpp == 1)
{
keymask = 0xff;
}
else
{
keymask = src_format->red_mask
| src_format->green_mask
| src_format->blue_mask;
}
flags &= ~(WINEDDBLT_KEYSRC | WINEDDBLT_KEYDEST | WINEDDBLT_KEYSRCOVERRIDE | WINEDDBLT_KEYDESTOVERRIDE);
}
if (flags & WINEDDBLT_DDFX)
{
BYTE *dTopLeft, *dTopRight, *dBottomLeft, *dBottomRight, *tmp;
LONG tmpxy;
dTopLeft = dbuf;
dTopRight = dbuf + ((dstwidth - 1) * bpp);
dBottomLeft = dTopLeft + ((dstheight - 1) * dlock.Pitch);
dBottomRight = dBottomLeft + ((dstwidth - 1) * bpp);
if (fx->dwDDFX & WINEDDBLTFX_ARITHSTRETCHY)
{
/* I don't think we need to do anything about this flag */
WARN("flags=DDBLT_DDFX nothing done for WINEDDBLTFX_ARITHSTRETCHY\n");
}
if (fx->dwDDFX & WINEDDBLTFX_MIRRORLEFTRIGHT)
{
tmp = dTopRight;
dTopRight = dTopLeft;
dTopLeft = tmp;
tmp = dBottomRight;
dBottomRight = dBottomLeft;
dBottomLeft = tmp;
dstxinc = dstxinc * -1;
}
if (fx->dwDDFX & WINEDDBLTFX_MIRRORUPDOWN)
{
tmp = dTopLeft;
dTopLeft = dBottomLeft;
dBottomLeft = tmp;
tmp = dTopRight;
dTopRight = dBottomRight;
dBottomRight = tmp;
dstyinc = dstyinc * -1;
}
if (fx->dwDDFX & WINEDDBLTFX_NOTEARING)
{
/* I don't think we need to do anything about this flag */
WARN("flags=DDBLT_DDFX nothing done for WINEDDBLTFX_NOTEARING\n");
}
if (fx->dwDDFX & WINEDDBLTFX_ROTATE180)
{
tmp = dBottomRight;
dBottomRight = dTopLeft;
dTopLeft = tmp;
tmp = dBottomLeft;
dBottomLeft = dTopRight;
dTopRight = tmp;
dstxinc = dstxinc * -1;
dstyinc = dstyinc * -1;
}
if (fx->dwDDFX & WINEDDBLTFX_ROTATE270)
{
tmp = dTopLeft;
dTopLeft = dBottomLeft;
dBottomLeft = dBottomRight;
dBottomRight = dTopRight;
dTopRight = tmp;
tmpxy = dstxinc;
dstxinc = dstyinc;
dstyinc = tmpxy;
dstxinc = dstxinc * -1;
}
if (fx->dwDDFX & WINEDDBLTFX_ROTATE90)
{
tmp = dTopLeft;
dTopLeft = dTopRight;
dTopRight = dBottomRight;
dBottomRight = dBottomLeft;
dBottomLeft = tmp;
tmpxy = dstxinc;
dstxinc = dstyinc;
dstyinc = tmpxy;
dstyinc = dstyinc * -1;
}
if (fx->dwDDFX & WINEDDBLTFX_ZBUFFERBASEDEST)
{
/* I don't think we need to do anything about this flag */
WARN("flags=WINEDDBLT_DDFX nothing done for WINEDDBLTFX_ZBUFFERBASEDEST\n");
}
dbuf = dTopLeft;
flags &= ~(WINEDDBLT_DDFX);
}
#define COPY_COLORKEY_FX(type) \
do { \
const type *s; \
type *d = (type *)dbuf, *dx, tmp; \
for (y = sy = 0; y < dstheight; ++y, sy += yinc) \
{ \
s = (const type *)(sbase + (sy >> 16) * slock.Pitch); \
dx = d; \
for (x = sx = 0; x < dstwidth; ++x, sx += xinc) \
{ \
tmp = s[sx >> 16]; \
if (((tmp & keymask) < keylow || (tmp & keymask) > keyhigh) \
&& ((dx[0] & destkeymask) >= destkeylow && (dx[0] & destkeymask) <= destkeyhigh)) \
{ \
dx[0] = tmp; \
} \
dx = (type *)(((BYTE *)dx) + dstxinc); \
} \
d = (type *)(((BYTE *)d) + dstyinc); \
} \
} while(0)
switch (bpp)
{
case 1:
COPY_COLORKEY_FX(BYTE);
break;
case 2:
COPY_COLORKEY_FX(WORD);
break;
case 4:
COPY_COLORKEY_FX(DWORD);
break;
case 3:
{
const BYTE *s;
BYTE *d = dbuf, *dx;
for (y = sy = 0; y < dstheight; ++y, sy += yinc)
{
sbuf = sbase + (sy >> 16) * slock.Pitch;
dx = d;
for (x = sx = 0; x < dstwidth; ++x, sx+= xinc)
{
DWORD pixel, dpixel = 0;
s = sbuf + 3 * (sx>>16);
pixel = s[0] | (s[1] << 8) | (s[2] << 16);
dpixel = dx[0] | (dx[1] << 8 ) | (dx[2] << 16);
if (((pixel & keymask) < keylow || (pixel & keymask) > keyhigh)
&& ((dpixel & keymask) >= destkeylow || (dpixel & keymask) <= keyhigh))
{
dx[0] = (pixel ) & 0xff;
dx[1] = (pixel >> 8) & 0xff;
dx[2] = (pixel >> 16) & 0xff;
}
dx += dstxinc;
}
d += dstyinc;
}
break;
}
default:
FIXME("%s color-keyed blit not implemented for bpp %u!\n",
(flags & WINEDDBLT_KEYSRC) ? "Source" : "Destination", bpp * 8);
hr = WINED3DERR_NOTAVAILABLE;
goto error;
#undef COPY_COLORKEY_FX
}
}
}
error:
if (flags && FIXME_ON(d3d_surface))
{
FIXME("\tUnsupported flags: %#x.\n", flags);
}
release:
wined3d_surface_unmap(dst_surface);
if (src_surface && src_surface != dst_surface)
wined3d_surface_unmap(src_surface);
/* Release the converted surface, if any. */
if (src_surface && src_surface != orig_src)
wined3d_surface_decref(src_surface);
return hr;
}
static HRESULT surface_cpu_bltfast(struct wined3d_surface *dst_surface, DWORD dst_x, DWORD dst_y,
struct wined3d_surface *src_surface, const RECT *src_rect, DWORD trans)
{
const struct wined3d_format *src_format, *dst_format;
RECT lock_src, lock_dst, lock_union;
WINED3DLOCKED_RECT dlock, slock;
HRESULT hr = WINED3D_OK;
int bpp, w, h, x, y;
const BYTE *sbuf;
BYTE *dbuf;
RECT rsrc2;
TRACE("dst_surface %p, dst_x %u, dst_y %u, src_surface %p, src_rect %s, flags %#x.\n",
dst_surface, dst_x, dst_y, src_surface, wine_dbgstr_rect(src_rect), trans);
if ((dst_surface->flags & SFLAG_LOCKED) || (src_surface->flags & SFLAG_LOCKED))
{
WARN(" Surface is busy, returning DDERR_SURFACEBUSY\n");
return WINEDDERR_SURFACEBUSY;
}
if (!src_rect)
{
WARN("src_rect is NULL!\n");
rsrc2.left = 0;
rsrc2.top = 0;
rsrc2.right = src_surface->resource.width;
rsrc2.bottom = src_surface->resource.height;
src_rect = &rsrc2;
}
/* Check source rect for validity. Copied from normal Blt. Fixes Baldur's Gate. */
if ((src_rect->bottom > src_surface->resource.height) || (src_rect->bottom < 0)
|| (src_rect->top > src_surface->resource.height) || (src_rect->top < 0)
|| (src_rect->left > src_surface->resource.width) || (src_rect->left < 0)
|| (src_rect->right > src_surface->resource.width) || (src_rect->right < 0)
|| (src_rect->right < src_rect->left) || (src_rect->bottom < src_rect->top))
{
WARN("Application gave us bad source rectangle for BltFast.\n");
return WINEDDERR_INVALIDRECT;
}
h = src_rect->bottom - src_rect->top;
if (h > dst_surface->resource.height - dst_y)
h = dst_surface->resource.height - dst_y;
if (h > src_surface->resource.height - src_rect->top)
h = src_surface->resource.height - src_rect->top;
if (h <= 0)
return WINEDDERR_INVALIDRECT;
w = src_rect->right - src_rect->left;
if (w > dst_surface->resource.width - dst_x)
w = dst_surface->resource.width - dst_x;
if (w > src_surface->resource.width - src_rect->left)
w = src_surface->resource.width - src_rect->left;
if (w <= 0)
return WINEDDERR_INVALIDRECT;
/* Now compute the locking rectangle... */
lock_src.left = src_rect->left;
lock_src.top = src_rect->top;
lock_src.right = lock_src.left + w;
lock_src.bottom = lock_src.top + h;
lock_dst.left = dst_x;
lock_dst.top = dst_y;
lock_dst.right = dst_x + w;
lock_dst.bottom = dst_y + h;
bpp = dst_surface->resource.format->byte_count;
/* We need to lock the surfaces, or we won't get refreshes when done. */
if (src_surface == dst_surface)
{
int pitch;
UnionRect(&lock_union, &lock_src, &lock_dst);
/* Lock the union of the two rectangles. */
hr = wined3d_surface_map(dst_surface, &dlock, &lock_union, 0);
if (FAILED(hr))
goto error;
pitch = dlock.Pitch;
slock.Pitch = dlock.Pitch;
/* Since slock was originally copied from this surface's description, we can just reuse it. */
sbuf = dst_surface->resource.allocatedMemory + lock_src.top * pitch + lock_src.left * bpp;
dbuf = dst_surface->resource.allocatedMemory + lock_dst.top * pitch + lock_dst.left * bpp;
src_format = src_surface->resource.format;
dst_format = src_format;
}
else
{
hr = wined3d_surface_map(src_surface, &slock, &lock_src, WINED3DLOCK_READONLY);
if (FAILED(hr))
goto error;
hr = wined3d_surface_map(dst_surface, &dlock, &lock_dst, 0);
if (FAILED(hr))
goto error;
sbuf = slock.pBits;
dbuf = dlock.pBits;
TRACE("Dst is at %p, Src is at %p.\n", dbuf, sbuf);
src_format = src_surface->resource.format;
dst_format = dst_surface->resource.format;
}
/* Handle compressed surfaces first... */
if (src_format->flags & dst_format->flags & WINED3DFMT_FLAG_COMPRESSED)
{
UINT row_block_count;
TRACE("compressed -> compressed copy\n");
if (trans)
FIXME("trans arg not supported when a compressed surface is involved\n");
if (dst_x || dst_y)
FIXME("offset for destination surface is not supported\n");
if (src_surface->resource.format->id != dst_surface->resource.format->id)
{
FIXME("compressed -> compressed copy only supported for the same type of surface\n");
hr = WINED3DERR_WRONGTEXTUREFORMAT;
goto error;
}
row_block_count = (w + dst_format->block_width - 1) / dst_format->block_width;
for (y = 0; y < h; y += dst_format->block_height)
{
memcpy(dbuf, sbuf, row_block_count * dst_format->block_byte_count);
dbuf += dlock.Pitch;
sbuf += slock.Pitch;
}
goto error;
}
if ((src_format->flags & WINED3DFMT_FLAG_COMPRESSED) && !(dst_format->flags & WINED3DFMT_FLAG_COMPRESSED))
{
/* TODO: Use the libtxc_dxtn.so shared library to do software
* decompression. */
ERR("Software decompression not supported.\n");
goto error;
}
if (trans & (WINEDDBLTFAST_SRCCOLORKEY | WINEDDBLTFAST_DESTCOLORKEY))
{
DWORD keylow, keyhigh;
DWORD mask = src_surface->resource.format->red_mask
| src_surface->resource.format->green_mask
| src_surface->resource.format->blue_mask;
/* For some 8-bit formats like L8 and P8 color masks don't make sense */
if (!mask && bpp == 1)
mask = 0xff;
TRACE("Color keyed copy.\n");
if (trans & WINEDDBLTFAST_SRCCOLORKEY)
{
keylow = src_surface->SrcBltCKey.dwColorSpaceLowValue;
keyhigh = src_surface->SrcBltCKey.dwColorSpaceHighValue;
}
else
{
/* I'm not sure if this is correct. */
FIXME("WINEDDBLTFAST_DESTCOLORKEY not fully supported yet.\n");
keylow = dst_surface->DestBltCKey.dwColorSpaceLowValue;
keyhigh = dst_surface->DestBltCKey.dwColorSpaceHighValue;
}
#define COPYBOX_COLORKEY(type) \
do { \
const type *s = (const type *)sbuf; \
type *d = (type *)dbuf; \
type tmp; \
for (y = 0; y < h; y++) \
{ \
for (x = 0; x < w; x++) \
{ \
tmp = s[x]; \
if ((tmp & mask) < keylow || (tmp & mask) > keyhigh) d[x] = tmp; \
} \
s = (const type *)((const BYTE *)s + slock.Pitch); \
d = (type *)((BYTE *)d + dlock.Pitch); \
} \
} while(0)
switch (bpp)
{
case 1:
COPYBOX_COLORKEY(BYTE);
break;
case 2:
COPYBOX_COLORKEY(WORD);
break;
case 4:
COPYBOX_COLORKEY(DWORD);
break;
case 3:
{
const BYTE *s;
DWORD tmp;
BYTE *d;
s = sbuf;
d = dbuf;
for (y = 0; y < h; ++y)
{
for (x = 0; x < w * 3; x += 3)
{
tmp = (DWORD)s[x] + ((DWORD)s[x + 1] << 8) + ((DWORD)s[x + 2] << 16);
if (tmp < keylow || tmp > keyhigh)
{
d[x + 0] = s[x + 0];
d[x + 1] = s[x + 1];
d[x + 2] = s[x + 2];
}
}
s += slock.Pitch;
d += dlock.Pitch;
}
break;
}
default:
FIXME("Source color key blitting not supported for bpp %u.\n", bpp * 8);
hr = WINED3DERR_NOTAVAILABLE;
goto error;
}
#undef COPYBOX_COLORKEY
TRACE("Copy done.\n");
}
else
{
int width = w * bpp;
INT sbufpitch, dbufpitch;
TRACE("No color key copy.\n");
/* Handle overlapping surfaces. */
if (sbuf < dbuf)
{
sbuf += (h - 1) * slock.Pitch;
dbuf += (h - 1) * dlock.Pitch;
sbufpitch = -slock.Pitch;
dbufpitch = -dlock.Pitch;
}
else
{
sbufpitch = slock.Pitch;
dbufpitch = dlock.Pitch;
}
for (y = 0; y < h; ++y)
{
/* This is pretty easy, a line for line memcpy. */
memmove(dbuf, sbuf, width);
sbuf += sbufpitch;
dbuf += dbufpitch;
}
TRACE("Copy done.\n");
}
error:
if (src_surface == dst_surface)
{
wined3d_surface_unmap(dst_surface);
}
else
{
wined3d_surface_unmap(dst_surface);
wined3d_surface_unmap(src_surface);
}
return hr;
}
/* Do not call while under the GL lock. */
static HRESULT cpu_blit_color_fill(IWineD3DDeviceImpl *device, struct wined3d_surface *dst_surface,
const RECT *dst_rect, const WINED3DCOLORVALUE *color)
{
WINEDDBLTFX BltFx;
memset(&BltFx, 0, sizeof(BltFx));
BltFx.dwSize = sizeof(BltFx);
BltFx.u5.dwFillColor = wined3d_format_convert_from_float(dst_surface->resource.format, color);
return wined3d_surface_blt(dst_surface, dst_rect, NULL, NULL,
WINEDDBLT_COLORFILL, &BltFx, WINED3DTEXF_POINT);
}
/* Do not call while under the GL lock. */
static HRESULT cpu_blit_depth_fill(IWineD3DDeviceImpl *device,
struct wined3d_surface *surface, const RECT *rect, float depth)
{
FIXME("Depth filling not implemented by cpu_blit.\n");
return WINED3DERR_INVALIDCALL;
}
const struct blit_shader cpu_blit = {
cpu_blit_alloc,
cpu_blit_free,
cpu_blit_set,
cpu_blit_unset,
cpu_blit_supported,
cpu_blit_color_fill,
cpu_blit_depth_fill,
};
static HRESULT surface_init(struct wined3d_surface *surface, WINED3DSURFTYPE surface_type, UINT alignment,
UINT width, UINT height, UINT level, BOOL lockable, BOOL discard, WINED3DMULTISAMPLE_TYPE multisample_type,
UINT multisample_quality, IWineD3DDeviceImpl *device, DWORD usage, enum wined3d_format_id format_id,
WINED3DPOOL pool, void *parent, const struct wined3d_parent_ops *parent_ops)
{
const struct wined3d_gl_info *gl_info = &device->adapter->gl_info;
const struct wined3d_format *format = wined3d_get_format(gl_info, format_id);
unsigned int resource_size;
HRESULT hr;
if (multisample_quality > 0)
{
FIXME("multisample_quality set to %u, substituting 0.\n", multisample_quality);
multisample_quality = 0;
}
/* Quick lockable sanity check.
* TODO: remove this after surfaces, usage and lockability have been debugged properly
* this function is too deep to need to care about things like this.
* Levels need to be checked too, since they all affect what can be done. */
switch (pool)
{
case WINED3DPOOL_SCRATCH:
if (!lockable)
{
FIXME("Called with a pool of SCRATCH and a lockable of FALSE "
"which are mutually exclusive, setting lockable to TRUE.\n");
lockable = TRUE;
}
break;
case WINED3DPOOL_SYSTEMMEM:
if (!lockable)
FIXME("Called with a pool of SYSTEMMEM and a lockable of FALSE, this is acceptable but unexpected.\n");
break;
case WINED3DPOOL_MANAGED:
if (usage & WINED3DUSAGE_DYNAMIC)
FIXME("Called with a pool of MANAGED and a usage of DYNAMIC which are mutually exclusive.\n");
break;
case WINED3DPOOL_DEFAULT:
if (lockable && !(usage & (WINED3DUSAGE_DYNAMIC | WINED3DUSAGE_RENDERTARGET | WINED3DUSAGE_DEPTHSTENCIL)))
WARN("Creating a lockable surface with a POOL of DEFAULT, that doesn't specify DYNAMIC usage.\n");
break;
default:
FIXME("Unknown pool %#x.\n", pool);
break;
};
if (usage & WINED3DUSAGE_RENDERTARGET && pool != WINED3DPOOL_DEFAULT)
FIXME("Trying to create a render target that isn't in the default pool.\n");
/* FIXME: Check that the format is supported by the device. */
resource_size = wined3d_format_calculate_size(format, alignment, width, height);
if (!resource_size)
return WINED3DERR_INVALIDCALL;
surface->surface_type = surface_type;
switch (surface_type)
{
case SURFACE_OPENGL:
surface->surface_ops = &surface_ops;
break;
case SURFACE_GDI:
surface->surface_ops = &gdi_surface_ops;
break;
default:
ERR("Requested unknown surface implementation %#x.\n", surface_type);
return WINED3DERR_INVALIDCALL;
}
hr = resource_init(&surface->resource, device, WINED3DRTYPE_SURFACE, format,
multisample_type, multisample_quality, usage, pool, width, height, 1,
resource_size, parent, parent_ops, &surface_resource_ops);
if (FAILED(hr))
{
WARN("Failed to initialize resource, returning %#x.\n", hr);
return hr;
}
/* "Standalone" surface. */
surface_set_container(surface, WINED3D_CONTAINER_NONE, NULL);
surface->texture_level = level;
list_init(&surface->overlays);
/* Flags */
surface->flags = SFLAG_NORMCOORD; /* Default to normalized coords. */
if (discard)
surface->flags |= SFLAG_DISCARD;
if (lockable || format_id == WINED3DFMT_D16_LOCKABLE)
surface->flags |= SFLAG_LOCKABLE;
/* I'm not sure if this qualifies as a hack or as an optimization. It
* seems reasonable to assume that lockable render targets will get
* locked, so we might as well set SFLAG_DYNLOCK right at surface
* creation. However, the other reason we want to do this is that several
* ddraw applications access surface memory while the surface isn't
* mapped. The SFLAG_DYNLOCK behaviour of keeping SYSMEM around for
* future locks prevents these from crashing. */
if (lockable && (usage & WINED3DUSAGE_RENDERTARGET))
surface->flags |= SFLAG_DYNLOCK;
/* Mark the texture as dirty so that it gets loaded first time around. */
surface_add_dirty_rect(surface, NULL);
list_init(&surface->renderbuffers);
TRACE("surface %p, memory %p, size %u\n",
surface, surface->resource.allocatedMemory, surface->resource.size);
/* Call the private setup routine */
hr = surface->surface_ops->surface_private_setup(surface);
if (FAILED(hr))
{
ERR("Private setup failed, returning %#x\n", hr);
surface->surface_ops->surface_cleanup(surface);
return hr;
}
return hr;
}
HRESULT CDECL wined3d_surface_create(IWineD3DDevice *iface, UINT width, UINT height,
enum wined3d_format_id format_id, BOOL lockable, BOOL discard, UINT level, DWORD usage, WINED3DPOOL pool,
WINED3DMULTISAMPLE_TYPE multisample_type, DWORD multisample_quality, WINED3DSURFTYPE surface_type,
void *parent, const struct wined3d_parent_ops *parent_ops, struct wined3d_surface **surface)
{
IWineD3DDeviceImpl *device = (IWineD3DDeviceImpl *)iface;
struct wined3d_surface *object;
HRESULT hr;
TRACE("device %p, width %u, height %u, format %s, lockable %#x, discard %#x, level %u\n",
device, width, height, debug_d3dformat(format_id), lockable, discard, level);
TRACE("surface %p, usage %s (%#x), pool %s, multisample_type %#x, multisample_quality %u\n",
surface, debug_d3dusage(usage), usage, debug_d3dpool(pool), multisample_type, multisample_quality);
TRACE("surface_type %#x, parent %p, parent_ops %p.\n", surface_type, parent, parent_ops);
if (surface_type == SURFACE_OPENGL && !device->adapter)
{
ERR("OpenGL surfaces are not available without OpenGL.\n");
return WINED3DERR_NOTAVAILABLE;
}
object = HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY, sizeof(*object));
if (!object)
{
ERR("Failed to allocate surface memory.\n");
return WINED3DERR_OUTOFVIDEOMEMORY;
}
hr = surface_init(object, surface_type, device->surface_alignment, width, height, level, lockable,
discard, multisample_type, multisample_quality, device, usage, format_id, pool, parent, parent_ops);
if (FAILED(hr))
{
WARN("Failed to initialize surface, returning %#x.\n", hr);
HeapFree(GetProcessHeap(), 0, object);
return hr;
}
TRACE("Created surface %p.\n", object);
*surface = object;
return hr;
}